diff --git "a/exp/log/log-train-2023-03-25-21-42-34-1" "b/exp/log/log-train-2023-03-25-21-42-34-1" new file mode 100644--- /dev/null +++ "b/exp/log/log-train-2023-03-25-21-42-34-1" @@ -0,0 +1,23596 @@ +2023-03-25 21:42:34,511 INFO [finetune.py:1046] (1/7) Training started +2023-03-25 21:42:34,511 INFO [finetune.py:1056] (1/7) Device: cuda:1 +2023-03-25 21:42:34,514 INFO [finetune.py:1065] (1/7) {'frame_shift_ms': 10.0, 'allowed_excess_duration_ratio': 0.1, 'best_train_loss': inf, 'best_valid_loss': inf, 'best_train_epoch': -1, 'best_valid_epoch': -1, 'batch_idx_train': 0, 'log_interval': 50, 'reset_interval': 200, 'valid_interval': 3000, 'feature_dim': 80, 'subsampling_factor': 4, 'warm_step': 2000, 'env_info': {'k2-version': '1.23.4', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': '62e404dd3f3a811d73e424199b3408e309c06e1a', 'k2-git-date': 'Mon Jan 30 02:26:16 2023', 'lhotse-version': '1.12.0.dev+git.3ccfeb7.clean', 'torch-version': '1.13.0', 'torch-cuda-available': True, 'torch-cuda-version': '11.7', 'python-version': '3.8', 'icefall-git-branch': 'master', 'icefall-git-sha1': 'd74822d-dirty', 'icefall-git-date': 'Tue Mar 21 21:35:32 2023', 'icefall-path': '/home/lishaojie/icefall', 'k2-path': '/home/lishaojie/.conda/envs/env_lishaojie/lib/python3.8/site-packages/k2/__init__.py', 'lhotse-path': '/home/lishaojie/.conda/envs/env_lishaojie/lib/python3.8/site-packages/lhotse/__init__.py', 'hostname': 'cnc533', 'IP address': '127.0.1.1'}, 'world_size': 7, 'master_port': 18181, 'tensorboard': True, 'num_epochs': 30, 'start_epoch': 1, 'start_batch': 0, 'exp_dir': PosixPath('pruned_transducer_stateless7_streaming/exp1'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'base_lr': 0.004, 'lr_batches': 100000.0, 'lr_epochs': 100.0, 'context_size': 2, 'prune_range': 5, 'lm_scale': 0.25, 'am_scale': 0.0, 'simple_loss_scale': 0.5, 'seed': 42, 'print_diagnostics': False, 'inf_check': False, 'save_every_n': 2000, 'keep_last_k': 30, 'average_period': 200, 'use_fp16': True, 'num_encoder_layers': '2,4,3,2,4', 'feedforward_dims': '1024,1024,2048,2048,1024', 'nhead': '8,8,8,8,8', 'encoder_dims': '384,384,384,384,384', 'attention_dims': '192,192,192,192,192', 'encoder_unmasked_dims': '256,256,256,256,256', 'zipformer_downsampling_factors': '1,2,4,8,2', 'cnn_module_kernels': '31,31,31,31,31', 'decoder_dim': 512, 'joiner_dim': 512, 'do_finetune': True, 'init_modules': 'encoder', 'finetune_ckpt': '/home/lishaojie/icefall/egs/commonvoice/ASR/pruned_transducer_stateless7_streaming/exp/english_pretrain/epoch-30.pt', 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 200, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'drop_last': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'input_strategy': 'PrecomputedFeatures', 'blank_id': 0, 'vocab_size': 500} +2023-03-25 21:42:34,514 INFO [finetune.py:1067] (1/7) About to create model +2023-03-25 21:42:34,900 INFO [zipformer.py:405] (1/7) At encoder stack 4, which has downsampling_factor=2, we will combine the outputs of layers 1 and 3, with downsampling_factors=2 and 8. +2023-03-25 21:42:34,909 INFO [finetune.py:1071] (1/7) Number of model parameters: 70369391 +2023-03-25 21:42:34,909 INFO [finetune.py:626] (1/7) Loading checkpoint from /home/lishaojie/icefall/egs/commonvoice/ASR/pruned_transducer_stateless7_streaming/exp/english_pretrain/epoch-30.pt +2023-03-25 21:42:35,456 INFO [finetune.py:647] (1/7) Loading parameters starting with prefix encoder +2023-03-25 21:42:36,935 INFO [finetune.py:1093] (1/7) Using DDP +2023-03-25 21:42:37,681 INFO [commonvoice_fr.py:392] (1/7) About to get train cuts +2023-03-25 21:42:37,687 INFO [commonvoice_fr.py:218] (1/7) Enable MUSAN +2023-03-25 21:42:37,687 INFO [commonvoice_fr.py:219] (1/7) About to get Musan cuts +2023-03-25 21:42:39,390 INFO [commonvoice_fr.py:243] (1/7) Enable SpecAugment +2023-03-25 21:42:39,390 INFO [commonvoice_fr.py:244] (1/7) Time warp factor: 80 +2023-03-25 21:42:39,390 INFO [commonvoice_fr.py:254] (1/7) Num frame mask: 10 +2023-03-25 21:42:39,390 INFO [commonvoice_fr.py:267] (1/7) About to create train dataset +2023-03-25 21:42:39,390 INFO [commonvoice_fr.py:294] (1/7) Using DynamicBucketingSampler. +2023-03-25 21:42:42,242 INFO [commonvoice_fr.py:309] (1/7) About to create train dataloader +2023-03-25 21:42:42,243 INFO [commonvoice_fr.py:399] (1/7) About to get dev cuts +2023-03-25 21:42:42,245 INFO [commonvoice_fr.py:340] (1/7) About to create dev dataset +2023-03-25 21:42:42,653 INFO [commonvoice_fr.py:357] (1/7) About to create dev dataloader +2023-03-25 21:42:42,653 INFO [finetune.py:1289] (1/7) Sanity check -- see if any of the batches in epoch 1 would cause OOM. +2023-03-25 21:46:46,136 INFO [finetune.py:1317] (1/7) Maximum memory allocated so far is 4472MB +2023-03-25 21:46:46,827 INFO [finetune.py:1317] (1/7) Maximum memory allocated so far is 5377MB +2023-03-25 21:46:48,914 INFO [finetune.py:1317] (1/7) Maximum memory allocated so far is 5377MB +2023-03-25 21:46:49,576 INFO [finetune.py:1317] (1/7) Maximum memory allocated so far is 5377MB +2023-03-25 21:46:50,267 INFO [finetune.py:1317] (1/7) Maximum memory allocated so far is 5377MB +2023-03-25 21:46:50,962 INFO [finetune.py:1317] (1/7) Maximum memory allocated so far is 5377MB +2023-03-25 21:46:59,842 INFO [finetune.py:976] (1/7) Epoch 1, batch 0, loss[loss=7.528, simple_loss=6.834, pruned_loss=6.929, over 4810.00 frames. ], tot_loss[loss=7.528, simple_loss=6.834, pruned_loss=6.929, over 4810.00 frames. ], batch size: 38, lr: 2.00e-03, grad_scale: 2.0 +2023-03-25 21:46:59,842 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-25 21:47:14,952 INFO [finetune.py:1010] (1/7) Epoch 1, validation: loss=7.294, simple_loss=6.606, pruned_loss=6.863, over 2265189.00 frames. +2023-03-25 21:47:14,952 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 5377MB +2023-03-25 21:47:19,869 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=5.0, num_to_drop=2, layers_to_drop={0, 3} +2023-03-25 21:47:30,297 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=23.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 21:47:49,822 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1297, 1.0040, 1.3283, 0.3497, 1.1835, 1.1759, 1.3974, 0.8719], + device='cuda:1'), covar=tensor([0.0350, 0.0202, 0.0120, 0.0269, 0.0148, 0.0181, 0.0155, 0.0286], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0176, 0.0133, 0.0143, 0.0149, 0.0145, 0.0169, 0.0183], + device='cuda:1'), out_proj_covar=tensor([1.1331e-04, 1.3185e-04, 9.7624e-05, 1.0458e-04, 1.0835e-04, 1.0894e-04, + 1.2687e-04, 1.3768e-04], device='cuda:1') +2023-03-25 21:48:00,211 INFO [finetune.py:976] (1/7) Epoch 1, batch 50, loss[loss=4.229, simple_loss=3.956, pruned_loss=2.626, over 4739.00 frames. ], tot_loss[loss=4.347, simple_loss=3.907, pruned_loss=4.221, over 215351.28 frames. ], batch size: 54, lr: 2.20e-03, grad_scale: 0.000244140625 +2023-03-25 21:48:01,887 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=50.24 vs. limit=5.0 +2023-03-25 21:48:24,078 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.31 vs. limit=2.0 +2023-03-25 21:48:31,037 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=10.44 vs. limit=5.0 +2023-03-25 21:48:31,585 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.39 vs. limit=2.0 +2023-03-25 21:48:33,004 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 21:48:51,469 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8827, 2.5190, 3.1312, 2.9673, 2.4335, 2.3534, 2.7835, 1.2942], + device='cuda:1'), covar=tensor([1.0926, 1.6611, 0.4756, 0.4855, 1.8799, 2.2015, 1.7866, 1.3529], + device='cuda:1'), in_proj_covar=tensor([0.0365, 0.0238, 0.0251, 0.0290, 0.0340, 0.0277, 0.0306, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 21:48:53,454 WARNING [finetune.py:966] (1/7) Grad scale is small: 0.000244140625 +2023-03-25 21:48:53,454 INFO [finetune.py:976] (1/7) Epoch 1, batch 100, loss[loss=2.179, simple_loss=2.063, pruned_loss=1.156, over 4871.00 frames. ], tot_loss[loss=3.427, simple_loss=3.155, pruned_loss=2.644, over 377845.48 frames. ], batch size: 34, lr: 2.40e-03, grad_scale: 0.00048828125 +2023-03-25 21:49:07,244 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4174, 1.9969, 1.6389, 3.1588, 2.3497, 1.9371, 3.4337, 2.2471], + device='cuda:1'), covar=tensor([0.0072, 0.0115, 0.0227, 0.0177, 0.0146, 0.0114, 0.0106, 0.0130], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0223, 0.0263, 0.0296, 0.0247, 0.0200, 0.0211, 0.0215], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 21:49:13,202 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 7.539e+02 2.791e+03 6.484e+03 1.700e+04 1.722e+07, threshold=1.297e+04, percent-clipped=0.0 +2023-03-25 21:49:28,871 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=144.0, num_to_drop=2, layers_to_drop={1, 2} +2023-03-25 21:49:37,443 INFO [finetune.py:976] (1/7) Epoch 1, batch 150, loss[loss=1.597, simple_loss=1.448, pruned_loss=1.213, over 4824.00 frames. ], tot_loss[loss=2.846, simple_loss=2.628, pruned_loss=2.075, over 508581.58 frames. ], batch size: 38, lr: 2.60e-03, grad_scale: 0.00048828125 +2023-03-25 21:49:45,923 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=35.81 vs. limit=5.0 +2023-03-25 21:49:47,967 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=29.54 vs. limit=5.0 +2023-03-25 21:50:02,581 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2547, 1.7818, 2.1883, 0.3649, 2.1227, 1.9635, 1.4442, 2.1225], + device='cuda:1'), covar=tensor([0.0256, 0.0203, 0.0265, 0.0433, 0.0228, 0.0227, 0.0292, 0.0215], + device='cuda:1'), in_proj_covar=tensor([0.0175, 0.0188, 0.0210, 0.0189, 0.0211, 0.0204, 0.0218, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 21:50:07,819 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=27.04 vs. limit=5.0 +2023-03-25 21:50:14,096 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=10.15 vs. limit=5.0 +2023-03-25 21:50:15,717 WARNING [finetune.py:966] (1/7) Grad scale is small: 0.00048828125 +2023-03-25 21:50:15,717 INFO [finetune.py:976] (1/7) Epoch 1, batch 200, loss[loss=1.25, simple_loss=1.081, pruned_loss=1.174, over 4772.00 frames. ], tot_loss[loss=2.349, simple_loss=2.149, pruned_loss=1.783, over 609045.99 frames. ], batch size: 26, lr: 2.80e-03, grad_scale: 0.0009765625 +2023-03-25 21:50:29,336 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 2.018e+02 7.406e+02 1.293e+03 3.197e+03 6.754e+04, threshold=2.586e+03, percent-clipped=12.0 +2023-03-25 21:50:54,579 INFO [finetune.py:976] (1/7) Epoch 1, batch 250, loss[loss=1.423, simple_loss=1.224, pruned_loss=1.296, over 4739.00 frames. ], tot_loss[loss=2.038, simple_loss=1.841, pruned_loss=1.621, over 687891.71 frames. ], batch size: 59, lr: 3.00e-03, grad_scale: 0.0009765625 +2023-03-25 21:51:43,795 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=296.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 21:51:44,352 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5438, 1.0497, 1.9519, 2.1261, 2.0772, 2.2700, 1.7787, 2.0305], + device='cuda:1'), covar=tensor([0.0481, 0.0754, 0.1371, 0.0426, 0.0673, 0.0733, 0.0754, 0.0501], + device='cuda:1'), in_proj_covar=tensor([0.0280, 0.0309, 0.0298, 0.0338, 0.0330, 0.0268, 0.0367, 0.0258], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 21:51:45,812 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=300.0, num_to_drop=2, layers_to_drop={2, 3} +2023-03-25 21:51:45,928 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=10.83 vs. limit=5.0 +2023-03-25 21:51:46,259 WARNING [finetune.py:966] (1/7) Grad scale is small: 0.0009765625 +2023-03-25 21:51:46,260 INFO [finetune.py:976] (1/7) Epoch 1, batch 300, loss[loss=1.248, simple_loss=1.049, pruned_loss=1.18, over 4866.00 frames. ], tot_loss[loss=1.85, simple_loss=1.649, pruned_loss=1.526, over 748059.87 frames. ], batch size: 31, lr: 3.20e-03, grad_scale: 0.001953125 +2023-03-25 21:51:58,579 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 2.075e+01 5.781e+01 1.827e+02 5.788e+02 1.230e+04, threshold=3.655e+02, percent-clipped=4.0 +2023-03-25 21:51:59,729 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3049, 1.7439, 2.0569, 2.9880, 2.6199, 2.8470, 1.2520, 1.8868], + device='cuda:1'), covar=tensor([0.1258, 0.1642, 0.1194, 0.1459, 0.0915, 0.1561, 0.1325, 0.1271], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0120, 0.0138, 0.0159, 0.0109, 0.0145, 0.0130, 0.0116], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-25 21:52:38,785 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=44.23 vs. limit=5.0 +2023-03-25 21:52:39,148 INFO [finetune.py:976] (1/7) Epoch 1, batch 350, loss[loss=1.254, simple_loss=1.047, pruned_loss=1.159, over 4720.00 frames. ], tot_loss[loss=1.702, simple_loss=1.498, pruned_loss=1.444, over 794782.32 frames. ], batch size: 59, lr: 3.40e-03, grad_scale: 0.001953125 +2023-03-25 21:52:47,106 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=357.0, num_to_drop=2, layers_to_drop={2, 3} +2023-03-25 21:53:08,661 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=12.24 vs. limit=5.0 +2023-03-25 21:53:10,165 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=387.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 21:53:27,702 WARNING [finetune.py:966] (1/7) Grad scale is small: 0.001953125 +2023-03-25 21:53:27,703 INFO [finetune.py:976] (1/7) Epoch 1, batch 400, loss[loss=1.197, simple_loss=0.9763, pruned_loss=1.135, over 4903.00 frames. ], tot_loss[loss=1.589, simple_loss=1.381, pruned_loss=1.378, over 831316.69 frames. ], batch size: 36, lr: 3.60e-03, grad_scale: 0.00390625 +2023-03-25 21:53:38,489 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=9.63 vs. limit=5.0 +2023-03-25 21:53:39,513 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=16.61 vs. limit=5.0 +2023-03-25 21:53:39,887 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.702e+01 2.277e+01 3.517e+01 1.113e+02 1.032e+03, threshold=7.035e+01, percent-clipped=3.0 +2023-03-25 21:53:49,829 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3639, 3.5681, 4.1052, 4.1476, 3.4264, 3.3986, 3.9188, 1.3744], + device='cuda:1'), covar=tensor([0.4626, 0.5114, 0.2906, 0.4080, 0.6944, 0.6738, 0.5702, 0.8870], + device='cuda:1'), in_proj_covar=tensor([0.0365, 0.0238, 0.0251, 0.0290, 0.0340, 0.0277, 0.0306, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 21:53:51,261 WARNING [optim.py:389] (1/7) Scaling gradients by 0.06621765345335007, model_norm_threshold=70.34587860107422 +2023-03-25 21:53:51,340 INFO [optim.py:451] (1/7) Parameter Dominanting tot_sumsq module.encoder.encoder_embed.conv.0.weight with proportion 0.67, where dominant_sumsq=(grad_sumsq*orig_rms_sq)=7.539e+05, grad_sumsq = 2.933e+06, orig_rms_sq=2.571e-01 +2023-03-25 21:54:00,686 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=439.0, num_to_drop=2, layers_to_drop={0, 2} +2023-03-25 21:54:05,301 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=448.0, num_to_drop=2, layers_to_drop={0, 2} +2023-03-25 21:54:05,862 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=3.92 vs. limit=2.0 +2023-03-25 21:54:06,751 INFO [finetune.py:976] (1/7) Epoch 1, batch 450, loss[loss=1.043, simple_loss=0.8455, pruned_loss=0.967, over 4903.00 frames. ], tot_loss[loss=1.485, simple_loss=1.273, pruned_loss=1.309, over 860333.91 frames. ], batch size: 43, lr: 3.80e-03, grad_scale: 0.00390625 +2023-03-25 21:54:13,032 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=4.23 vs. limit=2.0 +2023-03-25 21:54:28,355 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5427, 1.2481, 2.1215, 2.1860, 2.2792, 2.2251, 2.1344, 2.1298], + device='cuda:1'), covar=tensor([0.0201, 0.0550, 0.0679, 0.0295, 0.0494, 0.0436, 0.0326, 0.0333], + device='cuda:1'), in_proj_covar=tensor([0.0279, 0.0308, 0.0297, 0.0337, 0.0329, 0.0268, 0.0366, 0.0257], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 21:54:39,525 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=3.18 vs. limit=2.0 +2023-03-25 21:54:43,458 WARNING [finetune.py:966] (1/7) Grad scale is small: 0.00390625 +2023-03-25 21:54:43,459 INFO [finetune.py:976] (1/7) Epoch 1, batch 500, loss[loss=1.082, simple_loss=0.87, pruned_loss=0.9859, over 4851.00 frames. ], tot_loss[loss=1.387, simple_loss=1.176, pruned_loss=1.235, over 880509.27 frames. ], batch size: 49, lr: 4.00e-03, grad_scale: 0.0078125 +2023-03-25 21:54:57,600 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.430e+01 1.676e+01 1.950e+01 4.114e+01 1.062e+03, threshold=3.899e+01, percent-clipped=11.0 +2023-03-25 21:54:58,218 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0740, 1.2275, 1.4402, 1.0265, 1.2614, 2.5037, 0.9598, 1.7248], + device='cuda:1'), covar=tensor([0.0463, 0.0421, 0.0971, 0.0532, 0.0702, 0.0360, 0.0608, 0.0402], + device='cuda:1'), in_proj_covar=tensor([0.0072, 0.0081, 0.0071, 0.0073, 0.0090, 0.0076, 0.0084, 0.0076], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-25 21:55:28,672 INFO [finetune.py:976] (1/7) Epoch 1, batch 550, loss[loss=1.074, simple_loss=0.8469, pruned_loss=0.989, over 4871.00 frames. ], tot_loss[loss=1.299, simple_loss=1.088, pruned_loss=1.162, over 898230.18 frames. ], batch size: 34, lr: 4.00e-03, grad_scale: 0.0078125 +2023-03-25 21:55:39,559 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=562.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 21:55:42,636 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=568.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 21:55:58,101 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=590.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 21:56:09,541 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=600.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 21:56:09,981 WARNING [finetune.py:966] (1/7) Grad scale is small: 0.0078125 +2023-03-25 21:56:09,982 INFO [finetune.py:976] (1/7) Epoch 1, batch 600, loss[loss=1.176, simple_loss=0.9268, pruned_loss=1.048, over 4892.00 frames. ], tot_loss[loss=1.233, simple_loss=1.022, pruned_loss=1.106, over 910732.04 frames. ], batch size: 43, lr: 4.00e-03, grad_scale: 0.015625 +2023-03-25 21:56:22,616 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.472e+01 1.758e+01 2.024e+01 2.271e+01 8.528e+01, threshold=4.048e+01, percent-clipped=5.0 +2023-03-25 21:56:22,779 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=26.00 vs. limit=5.0 +2023-03-25 21:56:27,299 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=623.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 21:56:36,086 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=629.0, num_to_drop=2, layers_to_drop={0, 2} +2023-03-25 21:56:54,357 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=648.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 21:56:55,845 INFO [finetune.py:976] (1/7) Epoch 1, batch 650, loss[loss=0.8941, simple_loss=0.6935, pruned_loss=0.7978, over 4832.00 frames. ], tot_loss[loss=1.191, simple_loss=0.9758, pruned_loss=1.068, over 920248.32 frames. ], batch size: 25, lr: 4.00e-03, grad_scale: 0.015625 +2023-03-25 21:56:55,927 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=651.0, num_to_drop=2, layers_to_drop={1, 3} +2023-03-25 21:56:56,428 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=652.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 21:57:00,767 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.38 vs. limit=2.0 +2023-03-25 21:57:18,141 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=10.80 vs. limit=5.0 +2023-03-25 21:57:21,688 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=20.80 vs. limit=5.0 +2023-03-25 21:57:22,241 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-25 21:57:31,099 INFO [finetune.py:976] (1/7) Epoch 1, batch 700, loss[loss=1.091, simple_loss=0.8422, pruned_loss=0.9535, over 4914.00 frames. ], tot_loss[loss=1.156, simple_loss=0.936, pruned_loss=1.032, over 926274.30 frames. ], batch size: 36, lr: 4.00e-03, grad_scale: 0.03125 +2023-03-25 21:57:38,290 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.810e+01 2.037e+01 2.232e+01 2.628e+01 5.516e+01, threshold=4.463e+01, percent-clipped=4.0 +2023-03-25 21:57:59,225 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=739.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 21:58:01,230 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=743.0, num_to_drop=2, layers_to_drop={0, 2} +2023-03-25 21:58:05,816 INFO [finetune.py:976] (1/7) Epoch 1, batch 750, loss[loss=0.9275, simple_loss=0.7147, pruned_loss=0.7922, over 4804.00 frames. ], tot_loss[loss=1.126, simple_loss=0.902, pruned_loss=0.9996, over 934284.38 frames. ], batch size: 51, lr: 4.00e-03, grad_scale: 0.03125 +2023-03-25 21:58:29,192 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=787.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 21:58:36,511 INFO [finetune.py:976] (1/7) Epoch 1, batch 800, loss[loss=1.023, simple_loss=0.7731, pruned_loss=0.8779, over 4879.00 frames. ], tot_loss[loss=1.096, simple_loss=0.8694, pruned_loss=0.9673, over 938842.32 frames. ], batch size: 32, lr: 4.00e-03, grad_scale: 0.0625 +2023-03-25 21:58:41,750 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=2.29 vs. limit=2.0 +2023-03-25 21:58:45,191 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 2.059e+01 2.266e+01 2.508e+01 2.744e+01 4.199e+01, threshold=5.016e+01, percent-clipped=0.0 +2023-03-25 21:58:52,224 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2205, 2.4005, 1.8203, 1.1112, 2.9400, 1.8719, 2.0803, 1.9745], + device='cuda:1'), covar=tensor([0.0142, 0.0134, 0.0195, 0.0196, 0.0089, 0.0173, 0.0138, 0.0231], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0132, 0.0130, 0.0119, 0.0107, 0.0129, 0.0135, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 21:59:20,555 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=847.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 21:59:22,539 INFO [finetune.py:976] (1/7) Epoch 1, batch 850, loss[loss=0.9236, simple_loss=0.7031, pruned_loss=0.7664, over 4727.00 frames. ], tot_loss[loss=1.064, simple_loss=0.8355, pruned_loss=0.9311, over 943661.21 frames. ], batch size: 54, lr: 4.00e-03, grad_scale: 0.0625 +2023-03-25 22:00:05,324 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=11.59 vs. limit=5.0 +2023-03-25 22:00:12,166 INFO [finetune.py:976] (1/7) Epoch 1, batch 900, loss[loss=0.9972, simple_loss=0.7431, pruned_loss=0.8338, over 4828.00 frames. ], tot_loss[loss=1.031, simple_loss=0.8026, pruned_loss=0.8953, over 945192.72 frames. ], batch size: 33, lr: 4.00e-03, grad_scale: 0.125 +2023-03-25 22:00:15,807 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=908.0, num_to_drop=2, layers_to_drop={1, 2} +2023-03-25 22:00:25,565 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 2.101e+01 2.406e+01 2.575e+01 3.027e+01 5.726e+01, threshold=5.150e+01, percent-clipped=1.0 +2023-03-25 22:00:28,253 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=918.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 22:00:32,453 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=924.0, num_to_drop=2, layers_to_drop={0, 2} +2023-03-25 22:00:53,628 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=946.0, num_to_drop=2, layers_to_drop={2, 3} +2023-03-25 22:00:56,170 INFO [finetune.py:976] (1/7) Epoch 1, batch 950, loss[loss=0.9491, simple_loss=0.7072, pruned_loss=0.7766, over 4741.00 frames. ], tot_loss[loss=1.011, simple_loss=0.7797, pruned_loss=0.8694, over 948190.05 frames. ], batch size: 54, lr: 4.00e-03, grad_scale: 0.125 +2023-03-25 22:00:56,745 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=952.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 22:00:57,252 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6563, 1.2317, 1.4110, 1.4955, 1.4362, 4.2608, 1.7382, 1.9619], + device='cuda:1'), covar=tensor([0.2365, 0.1735, 0.1842, 0.1984, 0.1002, 0.0230, 0.1448, 0.0633], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0119, 0.0126, 0.0124, 0.0109, 0.0098, 0.0092, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-25 22:01:23,415 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2386, 3.8557, 3.9700, 4.1392, 3.8165, 3.7357, 4.2035, 2.1520], + device='cuda:1'), covar=tensor([0.1113, 0.1355, 0.1099, 0.1084, 0.1686, 0.1920, 0.1740, 0.3857], + device='cuda:1'), in_proj_covar=tensor([0.0366, 0.0238, 0.0251, 0.0290, 0.0340, 0.0277, 0.0306, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:01:43,865 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=1000.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:01:44,326 INFO [finetune.py:976] (1/7) Epoch 1, batch 1000, loss[loss=1.017, simple_loss=0.7421, pruned_loss=0.8363, over 4905.00 frames. ], tot_loss[loss=1.011, simple_loss=0.7723, pruned_loss=0.8604, over 951134.73 frames. ], batch size: 43, lr: 4.00e-03, grad_scale: 0.25 +2023-03-25 22:01:58,295 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 2.382e+01 2.890e+01 3.153e+01 3.664e+01 7.462e+01, threshold=6.306e+01, percent-clipped=2.0 +2023-03-25 22:02:01,587 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.56 vs. limit=2.0 +2023-03-25 22:02:21,753 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1043.0, num_to_drop=2, layers_to_drop={0, 2} +2023-03-25 22:02:29,308 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-25 22:02:31,288 INFO [finetune.py:976] (1/7) Epoch 1, batch 1050, loss[loss=1.101, simple_loss=0.806, pruned_loss=0.8841, over 4897.00 frames. ], tot_loss[loss=1.016, simple_loss=0.7688, pruned_loss=0.8555, over 953012.42 frames. ], batch size: 35, lr: 4.00e-03, grad_scale: 0.25 +2023-03-25 22:02:33,048 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=7.28 vs. limit=5.0 +2023-03-25 22:02:36,706 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4466, 1.7567, 1.4753, 1.8866, 2.1690, 1.6769, 1.3315, 1.3113], + device='cuda:1'), covar=tensor([0.2097, 0.1563, 0.1472, 0.1332, 0.1429, 0.1553, 0.1985, 0.1826], + device='cuda:1'), in_proj_covar=tensor([0.0257, 0.0238, 0.0229, 0.0209, 0.0271, 0.0204, 0.0235, 0.0209], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:03:07,624 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=1091.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:03:18,241 INFO [finetune.py:976] (1/7) Epoch 1, batch 1100, loss[loss=0.9603, simple_loss=0.7085, pruned_loss=0.7503, over 4117.00 frames. ], tot_loss[loss=1.009, simple_loss=0.7567, pruned_loss=0.8401, over 953372.90 frames. ], batch size: 65, lr: 4.00e-03, grad_scale: 0.5 +2023-03-25 22:03:30,770 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 2.698e+01 3.337e+01 3.640e+01 4.251e+01 7.174e+01, threshold=7.279e+01, percent-clipped=4.0 +2023-03-25 22:04:04,833 INFO [finetune.py:976] (1/7) Epoch 1, batch 1150, loss[loss=1.061, simple_loss=0.7758, pruned_loss=0.8236, over 4823.00 frames. ], tot_loss[loss=1.003, simple_loss=0.7472, pruned_loss=0.8251, over 955216.45 frames. ], batch size: 39, lr: 4.00e-03, grad_scale: 0.5 +2023-03-25 22:04:06,016 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1153.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:04:17,569 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7310, 3.3818, 3.3198, 3.3098, 3.3097, 3.1270, 3.8553, 1.7955], + device='cuda:1'), covar=tensor([0.1596, 0.2379, 0.1844, 0.2173, 0.2685, 0.3014, 0.1972, 0.7676], + device='cuda:1'), in_proj_covar=tensor([0.0366, 0.0238, 0.0251, 0.0290, 0.0340, 0.0277, 0.0306, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:04:24,274 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3164, 3.3470, 3.4100, 1.2837, 3.6343, 2.4819, 0.7573, 2.2994], + device='cuda:1'), covar=tensor([0.3577, 0.1870, 0.2107, 0.4270, 0.1307, 0.1456, 0.4997, 0.1608], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0150, 0.0160, 0.0126, 0.0152, 0.0116, 0.0144, 0.0118], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-25 22:04:46,447 INFO [finetune.py:976] (1/7) Epoch 1, batch 1200, loss[loss=0.9103, simple_loss=0.6674, pruned_loss=0.6921, over 4825.00 frames. ], tot_loss[loss=0.9886, simple_loss=0.7334, pruned_loss=0.8026, over 954083.75 frames. ], batch size: 39, lr: 4.00e-03, grad_scale: 1.0 +2023-03-25 22:04:47,539 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1203.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 22:04:59,185 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 3.248e+01 4.460e+01 5.563e+01 6.854e+01 1.013e+02, threshold=1.113e+02, percent-clipped=20.0 +2023-03-25 22:04:59,292 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1214.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 22:05:01,652 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1218.0, num_to_drop=2, layers_to_drop={1, 2} +2023-03-25 22:05:06,569 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1224.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 22:05:06,598 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3951, 1.4227, 1.3541, 1.7119, 1.6612, 1.4602, 2.2054, 1.5394], + device='cuda:1'), covar=tensor([0.3015, 0.4205, 0.4088, 0.2596, 0.3276, 0.2310, 0.1115, 0.4730], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0225, 0.0263, 0.0296, 0.0248, 0.0201, 0.0211, 0.0218], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:05:24,101 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1246.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:05:26,685 INFO [finetune.py:976] (1/7) Epoch 1, batch 1250, loss[loss=0.8686, simple_loss=0.6382, pruned_loss=0.6487, over 4753.00 frames. ], tot_loss[loss=0.969, simple_loss=0.7178, pruned_loss=0.7744, over 955859.73 frames. ], batch size: 23, lr: 4.00e-03, grad_scale: 1.0 +2023-03-25 22:05:46,031 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=1266.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:05:49,131 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=1272.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:06:08,472 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=1294.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:06:14,925 INFO [finetune.py:976] (1/7) Epoch 1, batch 1300, loss[loss=0.787, simple_loss=0.5847, pruned_loss=0.5726, over 4866.00 frames. ], tot_loss[loss=0.9417, simple_loss=0.6979, pruned_loss=0.7401, over 956568.44 frames. ], batch size: 31, lr: 4.00e-03, grad_scale: 1.0 +2023-03-25 22:06:19,441 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=17.09 vs. limit=5.0 +2023-03-25 22:06:23,502 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 5.599e+01 8.403e+01 9.999e+01 1.262e+02 2.600e+02, threshold=2.000e+02, percent-clipped=40.0 +2023-03-25 22:06:57,100 INFO [finetune.py:976] (1/7) Epoch 1, batch 1350, loss[loss=0.8518, simple_loss=0.6423, pruned_loss=0.6024, over 4096.00 frames. ], tot_loss[loss=0.9227, simple_loss=0.6854, pruned_loss=0.712, over 955526.42 frames. ], batch size: 65, lr: 4.00e-03, grad_scale: 1.0 +2023-03-25 22:07:50,278 INFO [finetune.py:976] (1/7) Epoch 1, batch 1400, loss[loss=0.8813, simple_loss=0.6726, pruned_loss=0.6083, over 4796.00 frames. ], tot_loss[loss=0.9053, simple_loss=0.6754, pruned_loss=0.6847, over 956628.50 frames. ], batch size: 29, lr: 4.00e-03, grad_scale: 1.0 +2023-03-25 22:07:58,274 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.277e+01 1.400e+02 1.610e+02 1.980e+02 2.974e+02, threshold=3.221e+02, percent-clipped=23.0 +2023-03-25 22:08:09,170 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1434.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:08:17,679 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6245, 1.6382, 1.0983, 1.1228, 1.6770, 2.0863, 2.0281, 1.5331], + device='cuda:1'), covar=tensor([0.0203, 0.0229, 0.0650, 0.0368, 0.0300, 0.0191, 0.0165, 0.0338], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0121, 0.0150, 0.0117, 0.0113, 0.0117, 0.0097, 0.0122], + device='cuda:1'), out_proj_covar=tensor([7.3493e-05, 9.5752e-05, 1.2252e-04, 9.2338e-05, 8.9580e-05, 8.8362e-05, + 7.4797e-05, 9.6059e-05], device='cuda:1') +2023-03-25 22:08:20,140 INFO [finetune.py:976] (1/7) Epoch 1, batch 1450, loss[loss=0.929, simple_loss=0.693, pruned_loss=0.6474, over 4092.00 frames. ], tot_loss[loss=0.8781, simple_loss=0.6591, pruned_loss=0.6512, over 954153.55 frames. ], batch size: 65, lr: 4.00e-03, grad_scale: 1.0 +2023-03-25 22:08:47,675 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1495.0, num_to_drop=2, layers_to_drop={2, 3} +2023-03-25 22:08:51,787 INFO [finetune.py:976] (1/7) Epoch 1, batch 1500, loss[loss=0.7316, simple_loss=0.5812, pruned_loss=0.4752, over 4742.00 frames. ], tot_loss[loss=0.8491, simple_loss=0.6425, pruned_loss=0.6167, over 953678.09 frames. ], batch size: 54, lr: 4.00e-03, grad_scale: 1.0 +2023-03-25 22:08:52,955 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7694, 3.2867, 3.3148, 3.6788, 3.5333, 3.4207, 3.8564, 1.2114], + device='cuda:1'), covar=tensor([0.0791, 0.0890, 0.0800, 0.0861, 0.1113, 0.1077, 0.0746, 0.5697], + device='cuda:1'), in_proj_covar=tensor([0.0371, 0.0244, 0.0258, 0.0294, 0.0346, 0.0285, 0.0310, 0.0306], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:08:52,977 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1503.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:09:02,815 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1509.0, num_to_drop=2, layers_to_drop={1, 2} +2023-03-25 22:09:03,465 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.68 vs. limit=5.0 +2023-03-25 22:09:05,945 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.968e+01 1.844e+02 2.293e+02 2.711e+02 4.587e+02, threshold=4.586e+02, percent-clipped=13.0 +2023-03-25 22:09:42,481 INFO [finetune.py:976] (1/7) Epoch 1, batch 1550, loss[loss=0.7347, simple_loss=0.585, pruned_loss=0.4718, over 4813.00 frames. ], tot_loss[loss=0.8195, simple_loss=0.6261, pruned_loss=0.5825, over 953976.77 frames. ], batch size: 45, lr: 4.00e-03, grad_scale: 1.0 +2023-03-25 22:09:42,539 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=1551.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:09:52,837 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1566.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:10:33,769 INFO [finetune.py:976] (1/7) Epoch 1, batch 1600, loss[loss=0.716, simple_loss=0.5752, pruned_loss=0.4523, over 4744.00 frames. ], tot_loss[loss=0.7848, simple_loss=0.6057, pruned_loss=0.5464, over 952921.38 frames. ], batch size: 27, lr: 4.00e-03, grad_scale: 2.0 +2023-03-25 22:10:40,879 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1611.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:10:42,943 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.367e+02 1.965e+02 2.441e+02 2.819e+02 5.041e+02, threshold=4.882e+02, percent-clipped=1.0 +2023-03-25 22:10:47,373 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.58 vs. limit=5.0 +2023-03-25 22:10:53,354 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3481, 1.4789, 1.1232, 1.8390, 1.4844, 1.2552, 2.4925, 1.2916], + device='cuda:1'), covar=tensor([0.6456, 0.8428, 0.7283, 0.6178, 0.4982, 0.4859, 0.1830, 1.0770], + device='cuda:1'), in_proj_covar=tensor([0.0182, 0.0211, 0.0249, 0.0276, 0.0231, 0.0193, 0.0197, 0.0207], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:10:55,970 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1627.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 22:11:18,720 INFO [finetune.py:976] (1/7) Epoch 1, batch 1650, loss[loss=0.6458, simple_loss=0.5318, pruned_loss=0.3957, over 4866.00 frames. ], tot_loss[loss=0.7546, simple_loss=0.5878, pruned_loss=0.515, over 953470.02 frames. ], batch size: 34, lr: 4.00e-03, grad_scale: 2.0 +2023-03-25 22:11:41,966 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1672.0, num_to_drop=2, layers_to_drop={1, 2} +2023-03-25 22:12:02,517 INFO [finetune.py:976] (1/7) Epoch 1, batch 1700, loss[loss=0.6364, simple_loss=0.5285, pruned_loss=0.3843, over 4745.00 frames. ], tot_loss[loss=0.7209, simple_loss=0.5678, pruned_loss=0.4822, over 954039.15 frames. ], batch size: 59, lr: 4.00e-03, grad_scale: 2.0 +2023-03-25 22:12:14,552 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.227e+02 2.187e+02 2.736e+02 3.197e+02 8.210e+02, threshold=5.471e+02, percent-clipped=2.0 +2023-03-25 22:12:53,681 INFO [finetune.py:976] (1/7) Epoch 1, batch 1750, loss[loss=0.6523, simple_loss=0.5543, pruned_loss=0.3836, over 4811.00 frames. ], tot_loss[loss=0.7029, simple_loss=0.5598, pruned_loss=0.4605, over 955638.77 frames. ], batch size: 41, lr: 4.00e-03, grad_scale: 2.0 +2023-03-25 22:13:29,993 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1790.0, num_to_drop=1, layers_to_drop={3} +2023-03-25 22:13:36,052 INFO [finetune.py:976] (1/7) Epoch 1, batch 1800, loss[loss=0.5791, simple_loss=0.4855, pruned_loss=0.3429, over 4778.00 frames. ], tot_loss[loss=0.6914, simple_loss=0.5566, pruned_loss=0.4443, over 954798.51 frames. ], batch size: 26, lr: 4.00e-03, grad_scale: 2.0 +2023-03-25 22:13:40,472 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1809.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:13:40,575 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.91 vs. limit=5.0 +2023-03-25 22:13:43,027 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.190e+02 2.215e+02 2.629e+02 3.291e+02 5.990e+02, threshold=5.258e+02, percent-clipped=1.0 +2023-03-25 22:13:59,214 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1838.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:14:01,959 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.46 vs. limit=5.0 +2023-03-25 22:14:06,687 INFO [finetune.py:976] (1/7) Epoch 1, batch 1850, loss[loss=0.5121, simple_loss=0.4336, pruned_loss=0.2992, over 4715.00 frames. ], tot_loss[loss=0.672, simple_loss=0.5469, pruned_loss=0.424, over 954917.18 frames. ], batch size: 23, lr: 4.00e-03, grad_scale: 2.0 +2023-03-25 22:14:10,079 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=1857.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:14:10,126 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1857.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:14:17,808 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1870.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:14:27,563 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1087, 0.7926, 1.0207, 0.9810, 0.7807, 0.8302, 1.0017, 0.9311], + device='cuda:1'), covar=tensor([16.5656, 32.2176, 15.9493, 25.8367, 33.8470, 17.5964, 39.8530, 14.8871], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0271, 0.0261, 0.0296, 0.0287, 0.0240, 0.0316, 0.0231], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:14:51,491 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1899.0, num_to_drop=2, layers_to_drop={0, 1} +2023-03-25 22:14:52,520 INFO [finetune.py:976] (1/7) Epoch 1, batch 1900, loss[loss=0.5318, simple_loss=0.4509, pruned_loss=0.3089, over 4143.00 frames. ], tot_loss[loss=0.6532, simple_loss=0.5374, pruned_loss=0.4049, over 953635.38 frames. ], batch size: 18, lr: 4.00e-03, grad_scale: 2.0 +2023-03-25 22:15:03,950 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.381e+02 2.208e+02 2.560e+02 3.227e+02 6.450e+02, threshold=5.121e+02, percent-clipped=1.0 +2023-03-25 22:15:11,546 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1918.0, num_to_drop=2, layers_to_drop={1, 3} +2023-03-25 22:15:14,227 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1922.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:15:20,182 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1931.0, num_to_drop=2, layers_to_drop={0, 2} +2023-03-25 22:15:37,066 INFO [finetune.py:976] (1/7) Epoch 1, batch 1950, loss[loss=0.4865, simple_loss=0.4256, pruned_loss=0.2745, over 4798.00 frames. ], tot_loss[loss=0.6303, simple_loss=0.5236, pruned_loss=0.3849, over 951722.10 frames. ], batch size: 29, lr: 4.00e-03, grad_scale: 2.0 +2023-03-25 22:15:41,600 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0033, 1.5792, 2.4764, 3.6220, 2.6487, 2.4670, 0.9835, 2.8107], + device='cuda:1'), covar=tensor([0.1686, 0.1674, 0.1156, 0.0452, 0.0826, 0.1466, 0.1915, 0.0700], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0115, 0.0128, 0.0150, 0.0103, 0.0138, 0.0121, 0.0107], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 22:15:46,008 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1967.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:15:52,091 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=12.22 vs. limit=5.0 +2023-03-25 22:16:06,021 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1449, 1.7976, 1.5615, 0.8269, 1.7567, 2.0768, 1.7197, 1.6431], + device='cuda:1'), covar=tensor([0.1166, 0.0392, 0.0525, 0.0912, 0.0413, 0.0224, 0.0406, 0.0468], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0157, 0.0120, 0.0130, 0.0133, 0.0126, 0.0152, 0.0160], + device='cuda:1'), out_proj_covar=tensor([1.0099e-04, 1.1699e-04, 8.7870e-05, 9.5522e-05, 9.6224e-05, 9.3799e-05, + 1.1381e-04, 1.1959e-04], device='cuda:1') +2023-03-25 22:16:12,899 INFO [finetune.py:976] (1/7) Epoch 1, batch 2000, loss[loss=0.4982, simple_loss=0.4476, pruned_loss=0.2744, over 4828.00 frames. ], tot_loss[loss=0.6077, simple_loss=0.5098, pruned_loss=0.3657, over 951201.81 frames. ], batch size: 40, lr: 4.00e-03, grad_scale: 4.0 +2023-03-25 22:16:22,846 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.403e+02 2.183e+02 2.758e+02 3.285e+02 7.843e+02, threshold=5.515e+02, percent-clipped=1.0 +2023-03-25 22:16:49,034 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4449, 3.5654, 3.6297, 1.4824, 3.8336, 2.7179, 1.0887, 2.6376], + device='cuda:1'), covar=tensor([0.2302, 0.1333, 0.1238, 0.3143, 0.0772, 0.0843, 0.3522, 0.1257], + device='cuda:1'), in_proj_covar=tensor([0.0145, 0.0146, 0.0154, 0.0121, 0.0144, 0.0109, 0.0133, 0.0112], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:16:57,263 INFO [finetune.py:976] (1/7) Epoch 1, batch 2050, loss[loss=0.4471, simple_loss=0.4122, pruned_loss=0.241, over 4804.00 frames. ], tot_loss[loss=0.5829, simple_loss=0.4953, pruned_loss=0.3452, over 953439.92 frames. ], batch size: 29, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:17:31,376 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2090.0, num_to_drop=2, layers_to_drop={1, 2} +2023-03-25 22:17:37,044 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-25 22:17:41,735 INFO [finetune.py:976] (1/7) Epoch 1, batch 2100, loss[loss=0.606, simple_loss=0.5355, pruned_loss=0.3382, over 4903.00 frames. ], tot_loss[loss=0.5671, simple_loss=0.4872, pruned_loss=0.3313, over 953370.60 frames. ], batch size: 36, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:17:55,063 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.336e+02 2.022e+02 2.484e+02 2.961e+02 6.695e+02, threshold=4.968e+02, percent-clipped=1.0 +2023-03-25 22:18:12,042 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=2138.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:18:29,607 INFO [finetune.py:976] (1/7) Epoch 1, batch 2150, loss[loss=0.6164, simple_loss=0.5552, pruned_loss=0.3388, over 4866.00 frames. ], tot_loss[loss=0.5602, simple_loss=0.4867, pruned_loss=0.3229, over 953104.69 frames. ], batch size: 34, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:18:54,468 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9526, 1.8716, 1.1482, 2.1507, 1.6526, 1.4358, 1.6440, 2.7139], + device='cuda:1'), covar=tensor([1.3561, 1.4955, 1.4157, 1.8341, 1.3061, 1.1138, 1.8912, 0.3720], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0233, 0.0206, 0.0269, 0.0225, 0.0192, 0.0234, 0.0174], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 22:19:03,493 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2194.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:19:08,508 INFO [finetune.py:976] (1/7) Epoch 1, batch 2200, loss[loss=0.5016, simple_loss=0.464, pruned_loss=0.2696, over 4921.00 frames. ], tot_loss[loss=0.5489, simple_loss=0.4827, pruned_loss=0.3122, over 954297.24 frames. ], batch size: 38, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:19:17,018 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2213.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:19:17,474 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.568e+02 2.355e+02 2.819e+02 3.325e+02 5.172e+02, threshold=5.637e+02, percent-clipped=1.0 +2023-03-25 22:19:22,617 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2222.0, num_to_drop=2, layers_to_drop={1, 2} +2023-03-25 22:19:28,116 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2226.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:19:57,057 INFO [finetune.py:976] (1/7) Epoch 1, batch 2250, loss[loss=0.5128, simple_loss=0.4616, pruned_loss=0.282, over 4804.00 frames. ], tot_loss[loss=0.5372, simple_loss=0.4769, pruned_loss=0.3024, over 953954.04 frames. ], batch size: 25, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:19:59,481 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0 +2023-03-25 22:20:18,388 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2267.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 22:20:18,929 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9001, 4.2969, 4.3814, 4.7053, 4.5790, 4.4139, 5.0258, 1.5157], + device='cuda:1'), covar=tensor([0.0785, 0.0750, 0.0619, 0.0766, 0.1358, 0.1148, 0.0505, 0.5237], + device='cuda:1'), in_proj_covar=tensor([0.0367, 0.0241, 0.0258, 0.0292, 0.0347, 0.0286, 0.0308, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:20:20,119 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=2270.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:21:00,800 INFO [finetune.py:976] (1/7) Epoch 1, batch 2300, loss[loss=0.4547, simple_loss=0.4186, pruned_loss=0.2454, over 4754.00 frames. ], tot_loss[loss=0.5262, simple_loss=0.4719, pruned_loss=0.2931, over 953866.73 frames. ], batch size: 28, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:21:15,881 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.417e+02 2.050e+02 2.425e+02 2.921e+02 4.362e+02, threshold=4.850e+02, percent-clipped=0.0 +2023-03-25 22:21:21,995 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=2315.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:21:35,737 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.84 vs. limit=5.0 +2023-03-25 22:21:37,385 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2340.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:21:54,802 INFO [finetune.py:976] (1/7) Epoch 1, batch 2350, loss[loss=0.4647, simple_loss=0.4301, pruned_loss=0.2496, over 4739.00 frames. ], tot_loss[loss=0.5112, simple_loss=0.4629, pruned_loss=0.2819, over 954496.75 frames. ], batch size: 54, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:22:57,438 INFO [finetune.py:976] (1/7) Epoch 1, batch 2400, loss[loss=0.454, simple_loss=0.4243, pruned_loss=0.2418, over 4911.00 frames. ], tot_loss[loss=0.4953, simple_loss=0.4527, pruned_loss=0.2707, over 956363.56 frames. ], batch size: 43, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:22:57,558 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2401.0, num_to_drop=2, layers_to_drop={2, 3} +2023-03-25 22:23:05,866 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.408e+02 1.953e+02 2.427e+02 2.971e+02 6.309e+02, threshold=4.853e+02, percent-clipped=1.0 +2023-03-25 22:23:32,011 INFO [finetune.py:976] (1/7) Epoch 1, batch 2450, loss[loss=0.5405, simple_loss=0.4863, pruned_loss=0.2973, over 4757.00 frames. ], tot_loss[loss=0.4802, simple_loss=0.443, pruned_loss=0.26, over 957318.58 frames. ], batch size: 59, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:24:21,668 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2494.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 22:24:25,633 INFO [finetune.py:976] (1/7) Epoch 1, batch 2500, loss[loss=0.4311, simple_loss=0.3935, pruned_loss=0.2344, over 4100.00 frames. ], tot_loss[loss=0.4749, simple_loss=0.4409, pruned_loss=0.2555, over 953523.73 frames. ], batch size: 18, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:24:34,883 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2513.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:24:35,371 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.353e+02 2.241e+02 2.593e+02 3.079e+02 4.323e+02, threshold=5.185e+02, percent-clipped=0.0 +2023-03-25 22:24:44,432 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2526.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:24:46,721 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5363, 3.4077, 3.4344, 1.5341, 3.6687, 2.6380, 1.0623, 2.3372], + device='cuda:1'), covar=tensor([0.2263, 0.1275, 0.1334, 0.3033, 0.0844, 0.0828, 0.3502, 0.1359], + device='cuda:1'), in_proj_covar=tensor([0.0146, 0.0147, 0.0154, 0.0121, 0.0144, 0.0110, 0.0135, 0.0113], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:24:54,542 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=2542.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:25:00,196 INFO [finetune.py:976] (1/7) Epoch 1, batch 2550, loss[loss=0.5238, simple_loss=0.4929, pruned_loss=0.2773, over 4826.00 frames. ], tot_loss[loss=0.4717, simple_loss=0.4412, pruned_loss=0.2519, over 953753.90 frames. ], batch size: 47, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:25:09,686 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=2561.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:25:18,673 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=2574.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:25:48,343 INFO [finetune.py:976] (1/7) Epoch 1, batch 2600, loss[loss=0.4601, simple_loss=0.432, pruned_loss=0.244, over 4064.00 frames. ], tot_loss[loss=0.4662, simple_loss=0.4393, pruned_loss=0.2472, over 953324.53 frames. ], batch size: 65, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:25:55,823 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.651e+02 2.205e+02 2.587e+02 2.996e+02 4.228e+02, threshold=5.174e+02, percent-clipped=0.0 +2023-03-25 22:26:19,959 INFO [finetune.py:976] (1/7) Epoch 1, batch 2650, loss[loss=0.3836, simple_loss=0.3773, pruned_loss=0.195, over 4731.00 frames. ], tot_loss[loss=0.4606, simple_loss=0.4374, pruned_loss=0.2424, over 955484.84 frames. ], batch size: 23, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:26:54,463 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2685.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:27:06,860 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2696.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:27:15,139 INFO [finetune.py:976] (1/7) Epoch 1, batch 2700, loss[loss=0.4588, simple_loss=0.4439, pruned_loss=0.2369, over 4897.00 frames. ], tot_loss[loss=0.454, simple_loss=0.4343, pruned_loss=0.2372, over 956896.96 frames. ], batch size: 36, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:27:17,970 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1108, 1.6846, 1.6651, 0.8508, 1.7789, 2.1803, 1.5872, 1.6371], + device='cuda:1'), covar=tensor([0.1001, 0.0447, 0.0436, 0.0769, 0.0371, 0.0212, 0.0403, 0.0432], + device='cuda:1'), in_proj_covar=tensor([0.0128, 0.0151, 0.0117, 0.0127, 0.0128, 0.0121, 0.0146, 0.0154], + device='cuda:1'), out_proj_covar=tensor([9.6693e-05, 1.1306e-04, 8.6100e-05, 9.3594e-05, 9.2791e-05, 8.9854e-05, + 1.0958e-04, 1.1463e-04], device='cuda:1') +2023-03-25 22:27:28,255 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.346e+02 2.127e+02 2.493e+02 3.058e+02 5.200e+02, threshold=4.985e+02, percent-clipped=1.0 +2023-03-25 22:28:14,535 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2746.0, num_to_drop=2, layers_to_drop={0, 3} +2023-03-25 22:28:17,274 INFO [finetune.py:976] (1/7) Epoch 1, batch 2750, loss[loss=0.4266, simple_loss=0.4112, pruned_loss=0.221, over 4743.00 frames. ], tot_loss[loss=0.4429, simple_loss=0.426, pruned_loss=0.2302, over 954938.84 frames. ], batch size: 26, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:28:58,716 INFO [finetune.py:976] (1/7) Epoch 1, batch 2800, loss[loss=0.3543, simple_loss=0.3603, pruned_loss=0.1741, over 4880.00 frames. ], tot_loss[loss=0.432, simple_loss=0.4182, pruned_loss=0.2232, over 956169.47 frames. ], batch size: 31, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:28:59,410 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7156, 1.8270, 1.6185, 1.2571, 2.2512, 2.0341, 1.8894, 1.6696], + device='cuda:1'), covar=tensor([0.0869, 0.0759, 0.1006, 0.1160, 0.0464, 0.0819, 0.0829, 0.1396], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0132, 0.0132, 0.0121, 0.0108, 0.0131, 0.0136, 0.0166], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:29:06,641 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.489e+02 2.264e+02 2.537e+02 3.001e+02 5.007e+02, threshold=5.073e+02, percent-clipped=1.0 +2023-03-25 22:29:09,064 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3079, 1.8959, 2.6231, 1.5441, 2.0556, 2.2310, 1.9644, 2.6344], + device='cuda:1'), covar=tensor([0.2263, 0.2343, 0.1815, 0.2859, 0.1562, 0.2362, 0.2717, 0.1285], + device='cuda:1'), in_proj_covar=tensor([0.0184, 0.0194, 0.0191, 0.0179, 0.0160, 0.0200, 0.0202, 0.0177], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:29:12,537 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2824.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:29:40,284 INFO [finetune.py:976] (1/7) Epoch 1, batch 2850, loss[loss=0.4527, simple_loss=0.4154, pruned_loss=0.245, over 4213.00 frames. ], tot_loss[loss=0.4256, simple_loss=0.4139, pruned_loss=0.2188, over 954899.69 frames. ], batch size: 65, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:29:40,395 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8941, 1.4119, 2.0096, 1.4656, 1.8055, 1.8961, 1.4900, 2.1371], + device='cuda:1'), covar=tensor([0.1829, 0.2557, 0.1742, 0.2043, 0.1234, 0.1800, 0.2975, 0.1191], + device='cuda:1'), in_proj_covar=tensor([0.0184, 0.0195, 0.0192, 0.0179, 0.0160, 0.0200, 0.0202, 0.0177], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:30:03,439 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2885.0, num_to_drop=2, layers_to_drop={0, 3} +2023-03-25 22:30:15,601 INFO [finetune.py:976] (1/7) Epoch 1, batch 2900, loss[loss=0.4419, simple_loss=0.4382, pruned_loss=0.2228, over 4907.00 frames. ], tot_loss[loss=0.4268, simple_loss=0.4154, pruned_loss=0.2192, over 951368.71 frames. ], batch size: 36, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:30:23,166 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.426e+02 2.100e+02 2.461e+02 2.914e+02 6.574e+02, threshold=4.923e+02, percent-clipped=3.0 +2023-03-25 22:30:23,368 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0 +2023-03-25 22:30:37,163 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2937.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:30:50,912 INFO [finetune.py:976] (1/7) Epoch 1, batch 2950, loss[loss=0.3811, simple_loss=0.3861, pruned_loss=0.188, over 4270.00 frames. ], tot_loss[loss=0.427, simple_loss=0.4177, pruned_loss=0.2182, over 950889.81 frames. ], batch size: 65, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:31:31,734 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2996.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:31:33,416 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2998.0, num_to_drop=2, layers_to_drop={1, 3} +2023-03-25 22:31:35,182 INFO [finetune.py:976] (1/7) Epoch 1, batch 3000, loss[loss=0.344, simple_loss=0.3368, pruned_loss=0.1756, over 4124.00 frames. ], tot_loss[loss=0.425, simple_loss=0.4171, pruned_loss=0.2166, over 950111.07 frames. ], batch size: 18, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:31:35,182 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-25 22:31:47,197 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6313, 1.3755, 1.7710, 1.2534, 1.4701, 1.7012, 1.3733, 1.9112], + device='cuda:1'), covar=tensor([0.1750, 0.2161, 0.1360, 0.1786, 0.1061, 0.1633, 0.2875, 0.1073], + device='cuda:1'), in_proj_covar=tensor([0.0186, 0.0195, 0.0192, 0.0180, 0.0161, 0.0202, 0.0202, 0.0179], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:31:56,383 INFO [finetune.py:1010] (1/7) Epoch 1, validation: loss=0.4228, simple_loss=0.4589, pruned_loss=0.1933, over 2265189.00 frames. +2023-03-25 22:31:56,384 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 5436MB +2023-03-25 22:32:17,068 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.364e+02 2.092e+02 2.490e+02 2.940e+02 5.162e+02, threshold=4.980e+02, percent-clipped=2.0 +2023-03-25 22:32:17,820 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.21 vs. limit=5.0 +2023-03-25 22:32:25,833 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3019.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:32:39,245 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3041.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:32:40,984 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=3044.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:32:44,999 INFO [finetune.py:976] (1/7) Epoch 1, batch 3050, loss[loss=0.3879, simple_loss=0.4007, pruned_loss=0.1876, over 4795.00 frames. ], tot_loss[loss=0.4226, simple_loss=0.4165, pruned_loss=0.2144, over 950420.23 frames. ], batch size: 51, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:33:09,739 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3080.0, num_to_drop=2, layers_to_drop={0, 3} +2023-03-25 22:33:10,468 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-03-25 22:33:38,269 INFO [finetune.py:976] (1/7) Epoch 1, batch 3100, loss[loss=0.456, simple_loss=0.4236, pruned_loss=0.2442, over 4775.00 frames. ], tot_loss[loss=0.4145, simple_loss=0.4105, pruned_loss=0.2092, over 951486.89 frames. ], batch size: 29, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:33:51,987 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.371e+02 2.009e+02 2.458e+02 3.052e+02 4.298e+02, threshold=4.916e+02, percent-clipped=0.0 +2023-03-25 22:34:39,799 INFO [finetune.py:976] (1/7) Epoch 1, batch 3150, loss[loss=0.3185, simple_loss=0.3251, pruned_loss=0.1559, over 3995.00 frames. ], tot_loss[loss=0.4062, simple_loss=0.4036, pruned_loss=0.2044, over 951002.75 frames. ], batch size: 17, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:35:08,902 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.99 vs. limit=2.0 +2023-03-25 22:35:11,095 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3180.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:35:39,617 INFO [finetune.py:976] (1/7) Epoch 1, batch 3200, loss[loss=0.4288, simple_loss=0.4218, pruned_loss=0.2179, over 4808.00 frames. ], tot_loss[loss=0.3978, simple_loss=0.3972, pruned_loss=0.1993, over 951481.11 frames. ], batch size: 41, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:35:52,730 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.131e+02 1.973e+02 2.320e+02 2.787e+02 5.091e+02, threshold=4.641e+02, percent-clipped=1.0 +2023-03-25 22:36:29,142 INFO [finetune.py:976] (1/7) Epoch 1, batch 3250, loss[loss=0.4311, simple_loss=0.4191, pruned_loss=0.2215, over 4075.00 frames. ], tot_loss[loss=0.3951, simple_loss=0.396, pruned_loss=0.1971, over 951284.09 frames. ], batch size: 65, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:37:12,259 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3288.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:37:20,001 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3293.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:37:24,558 INFO [finetune.py:976] (1/7) Epoch 1, batch 3300, loss[loss=0.3813, simple_loss=0.388, pruned_loss=0.1873, over 4794.00 frames. ], tot_loss[loss=0.3989, simple_loss=0.4004, pruned_loss=0.1987, over 953190.37 frames. ], batch size: 29, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:37:32,702 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.553e+02 2.210e+02 2.512e+02 3.057e+02 4.555e+02, threshold=5.024e+02, percent-clipped=0.0 +2023-03-25 22:37:36,225 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3806, 1.5253, 1.1071, 1.4711, 1.4637, 1.1815, 2.1379, 1.3930], + device='cuda:1'), covar=tensor([0.2732, 0.4454, 0.4926, 0.4906, 0.3132, 0.2668, 0.2665, 0.4048], + device='cuda:1'), in_proj_covar=tensor([0.0159, 0.0187, 0.0224, 0.0241, 0.0201, 0.0171, 0.0178, 0.0182], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:1') +2023-03-25 22:38:03,234 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3341.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:38:13,344 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3349.0, num_to_drop=2, layers_to_drop={2, 3} +2023-03-25 22:38:14,397 INFO [finetune.py:976] (1/7) Epoch 1, batch 3350, loss[loss=0.4357, simple_loss=0.4389, pruned_loss=0.2162, over 4898.00 frames. ], tot_loss[loss=0.3993, simple_loss=0.4024, pruned_loss=0.1982, over 953397.11 frames. ], batch size: 46, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:38:32,470 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-25 22:38:35,870 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7704, 2.0491, 1.7413, 2.0331, 1.2428, 4.6576, 1.7237, 2.2050], + device='cuda:1'), covar=tensor([0.3559, 0.2410, 0.2066, 0.2272, 0.2107, 0.0089, 0.2711, 0.1541], + device='cuda:1'), in_proj_covar=tensor([0.0118, 0.0101, 0.0109, 0.0107, 0.0099, 0.0086, 0.0085, 0.0083], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0004, 0.0005, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004], + device='cuda:1') +2023-03-25 22:38:42,736 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7965, 4.2886, 4.2179, 2.4959, 4.4947, 3.3450, 1.2552, 3.0162], + device='cuda:1'), covar=tensor([0.2356, 0.1112, 0.1055, 0.2487, 0.0622, 0.0735, 0.3642, 0.1136], + device='cuda:1'), in_proj_covar=tensor([0.0147, 0.0149, 0.0155, 0.0121, 0.0146, 0.0111, 0.0137, 0.0114], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:38:44,427 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3375.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:38:59,533 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=3389.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:39:06,747 INFO [finetune.py:976] (1/7) Epoch 1, batch 3400, loss[loss=0.3887, simple_loss=0.4057, pruned_loss=0.1859, over 4866.00 frames. ], tot_loss[loss=0.3985, simple_loss=0.4026, pruned_loss=0.1972, over 953890.57 frames. ], batch size: 31, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:39:17,912 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1633, 2.0923, 2.1922, 1.1064, 2.0984, 2.4996, 2.0349, 1.9666], + device='cuda:1'), covar=tensor([0.0585, 0.0319, 0.0248, 0.0456, 0.0243, 0.0172, 0.0193, 0.0259], + device='cuda:1'), in_proj_covar=tensor([0.0117, 0.0140, 0.0109, 0.0117, 0.0118, 0.0111, 0.0134, 0.0140], + device='cuda:1'), out_proj_covar=tensor([8.8226e-05, 1.0449e-04, 7.9891e-05, 8.6026e-05, 8.5284e-05, 8.2072e-05, + 1.0028e-04, 1.0440e-04], device='cuda:1') +2023-03-25 22:39:20,794 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.307e+02 1.988e+02 2.392e+02 2.720e+02 4.202e+02, threshold=4.784e+02, percent-clipped=0.0 +2023-03-25 22:40:08,846 INFO [finetune.py:976] (1/7) Epoch 1, batch 3450, loss[loss=0.3804, simple_loss=0.392, pruned_loss=0.1844, over 4922.00 frames. ], tot_loss[loss=0.3953, simple_loss=0.4006, pruned_loss=0.195, over 953797.61 frames. ], batch size: 42, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:40:40,369 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3480.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:40:42,160 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.32 vs. limit=5.0 +2023-03-25 22:40:56,106 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6184, 1.4469, 1.2249, 1.1202, 1.4721, 1.9380, 1.6031, 1.1241], + device='cuda:1'), covar=tensor([0.0232, 0.0297, 0.0478, 0.0399, 0.0254, 0.0156, 0.0236, 0.0454], + device='cuda:1'), in_proj_covar=tensor([0.0086, 0.0114, 0.0135, 0.0112, 0.0106, 0.0102, 0.0088, 0.0113], + device='cuda:1'), out_proj_covar=tensor([6.7197e-05, 9.0085e-05, 1.0905e-04, 8.9219e-05, 8.3936e-05, 7.6463e-05, + 6.8135e-05, 8.8765e-05], device='cuda:1') +2023-03-25 22:41:03,281 INFO [finetune.py:976] (1/7) Epoch 1, batch 3500, loss[loss=0.3847, simple_loss=0.3839, pruned_loss=0.1927, over 4801.00 frames. ], tot_loss[loss=0.3913, simple_loss=0.3968, pruned_loss=0.1929, over 954872.93 frames. ], batch size: 51, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:41:14,918 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.301e+02 2.271e+02 2.832e+02 3.824e+02 1.123e+03, threshold=5.664e+02, percent-clipped=12.0 +2023-03-25 22:41:30,603 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=3528.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:41:48,774 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3546.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:41:57,307 INFO [finetune.py:976] (1/7) Epoch 1, batch 3550, loss[loss=0.3269, simple_loss=0.3501, pruned_loss=0.1519, over 4818.00 frames. ], tot_loss[loss=0.3861, simple_loss=0.3922, pruned_loss=0.19, over 957106.93 frames. ], batch size: 41, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:42:07,646 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6638, 1.2936, 1.5152, 1.5350, 2.2534, 1.5018, 1.3808, 1.2125], + device='cuda:1'), covar=tensor([0.3889, 0.4534, 0.3450, 0.3794, 0.3664, 0.2595, 0.6101, 0.3138], + device='cuda:1'), in_proj_covar=tensor([0.0212, 0.0197, 0.0184, 0.0170, 0.0217, 0.0170, 0.0193, 0.0172], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:42:39,680 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3593.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:42:45,926 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8873, 1.2436, 1.0067, 1.6898, 2.0782, 1.4498, 1.3642, 1.9173], + device='cuda:1'), covar=tensor([0.1656, 0.2082, 0.2133, 0.1230, 0.2377, 0.2032, 0.1437, 0.1799], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0093, 0.0111, 0.0089, 0.0121, 0.0091, 0.0095, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 22:42:50,744 INFO [finetune.py:976] (1/7) Epoch 1, batch 3600, loss[loss=0.3418, simple_loss=0.3619, pruned_loss=0.1608, over 4863.00 frames. ], tot_loss[loss=0.3807, simple_loss=0.3878, pruned_loss=0.1869, over 957634.45 frames. ], batch size: 31, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:42:59,634 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3607.0, num_to_drop=2, layers_to_drop={2, 3} +2023-03-25 22:43:09,361 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.842e+02 2.557e+02 2.866e+02 3.769e+02 9.044e+02, threshold=5.732e+02, percent-clipped=5.0 +2023-03-25 22:43:44,100 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=3641.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:43:46,911 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3644.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 22:43:52,546 INFO [finetune.py:976] (1/7) Epoch 1, batch 3650, loss[loss=0.3649, simple_loss=0.383, pruned_loss=0.1735, over 4918.00 frames. ], tot_loss[loss=0.3829, simple_loss=0.3902, pruned_loss=0.1877, over 956383.97 frames. ], batch size: 38, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:44:20,322 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3675.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:44:21,457 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3676.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:44:40,055 INFO [finetune.py:976] (1/7) Epoch 1, batch 3700, loss[loss=0.4361, simple_loss=0.4351, pruned_loss=0.2185, over 4883.00 frames. ], tot_loss[loss=0.3869, simple_loss=0.3955, pruned_loss=0.1891, over 956334.77 frames. ], batch size: 32, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:44:51,249 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-25 22:44:52,812 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.548e+02 2.567e+02 2.980e+02 3.536e+02 5.905e+02, threshold=5.959e+02, percent-clipped=1.0 +2023-03-25 22:45:09,738 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=3723.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:45:23,631 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3737.0, num_to_drop=2, layers_to_drop={2, 3} +2023-03-25 22:45:32,455 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8938, 1.6027, 2.2136, 1.4520, 1.8122, 1.9948, 1.6306, 2.2928], + device='cuda:1'), covar=tensor([0.1805, 0.2114, 0.1320, 0.2166, 0.1253, 0.1674, 0.2273, 0.1060], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0197, 0.0193, 0.0182, 0.0165, 0.0207, 0.0203, 0.0182], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:45:43,733 INFO [finetune.py:976] (1/7) Epoch 1, batch 3750, loss[loss=0.3468, simple_loss=0.3722, pruned_loss=0.1607, over 4936.00 frames. ], tot_loss[loss=0.3848, simple_loss=0.3946, pruned_loss=0.1876, over 955940.25 frames. ], batch size: 29, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:46:33,630 INFO [finetune.py:976] (1/7) Epoch 1, batch 3800, loss[loss=0.4229, simple_loss=0.4211, pruned_loss=0.2123, over 4817.00 frames. ], tot_loss[loss=0.3835, simple_loss=0.3944, pruned_loss=0.1863, over 956536.31 frames. ], batch size: 25, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:46:34,883 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9069, 1.2585, 0.9343, 1.5736, 2.0018, 1.1523, 1.3894, 1.7544], + device='cuda:1'), covar=tensor([0.1683, 0.2173, 0.2286, 0.1331, 0.2555, 0.2071, 0.1440, 0.1827], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0094, 0.0112, 0.0090, 0.0122, 0.0092, 0.0095, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 22:46:47,079 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.580e+02 2.141e+02 2.900e+02 3.620e+02 1.043e+03, threshold=5.800e+02, percent-clipped=4.0 +2023-03-25 22:47:22,184 INFO [finetune.py:976] (1/7) Epoch 1, batch 3850, loss[loss=0.3482, simple_loss=0.3729, pruned_loss=0.1617, over 4902.00 frames. ], tot_loss[loss=0.3777, simple_loss=0.3903, pruned_loss=0.1825, over 956999.56 frames. ], batch size: 43, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:47:44,398 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.5862, 4.6713, 5.0723, 5.4648, 5.2169, 4.9591, 5.5832, 2.0194], + device='cuda:1'), covar=tensor([0.0603, 0.0939, 0.0648, 0.0665, 0.1153, 0.1244, 0.0582, 0.5040], + device='cuda:1'), in_proj_covar=tensor([0.0371, 0.0245, 0.0265, 0.0294, 0.0348, 0.0289, 0.0311, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:48:09,988 INFO [finetune.py:976] (1/7) Epoch 1, batch 3900, loss[loss=0.2845, simple_loss=0.2979, pruned_loss=0.1356, over 3987.00 frames. ], tot_loss[loss=0.3717, simple_loss=0.3848, pruned_loss=0.1793, over 956534.15 frames. ], batch size: 16, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:48:10,639 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3902.0, num_to_drop=1, layers_to_drop={3} +2023-03-25 22:48:20,244 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.580e+02 2.264e+02 2.673e+02 3.196e+02 5.181e+02, threshold=5.346e+02, percent-clipped=0.0 +2023-03-25 22:48:30,355 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3926.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:48:42,321 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7785, 1.1858, 0.8194, 1.5111, 1.9860, 0.9497, 1.3485, 1.5660], + device='cuda:1'), covar=tensor([0.1651, 0.2106, 0.2227, 0.1227, 0.2358, 0.2014, 0.1412, 0.1896], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0094, 0.0112, 0.0090, 0.0122, 0.0092, 0.0096, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 22:48:50,575 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3944.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:48:54,666 INFO [finetune.py:976] (1/7) Epoch 1, batch 3950, loss[loss=0.348, simple_loss=0.3661, pruned_loss=0.165, over 4917.00 frames. ], tot_loss[loss=0.3678, simple_loss=0.3806, pruned_loss=0.1775, over 953991.66 frames. ], batch size: 43, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:48:54,773 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3118, 1.6862, 1.3994, 1.0573, 1.8296, 2.8749, 2.3207, 1.6288], + device='cuda:1'), covar=tensor([0.0171, 0.0570, 0.0561, 0.0621, 0.0370, 0.0148, 0.0206, 0.0507], + device='cuda:1'), in_proj_covar=tensor([0.0085, 0.0113, 0.0133, 0.0111, 0.0105, 0.0101, 0.0088, 0.0112], + device='cuda:1'), out_proj_covar=tensor([6.6441e-05, 8.9649e-05, 1.0773e-04, 8.8397e-05, 8.3230e-05, 7.5504e-05, + 6.8054e-05, 8.7866e-05], device='cuda:1') +2023-03-25 22:49:45,839 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3987.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:49:53,754 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=3992.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:50:05,834 INFO [finetune.py:976] (1/7) Epoch 1, batch 4000, loss[loss=0.3359, simple_loss=0.3677, pruned_loss=0.1521, over 4825.00 frames. ], tot_loss[loss=0.363, simple_loss=0.3762, pruned_loss=0.1749, over 952957.78 frames. ], batch size: 33, lr: 4.00e-03, grad_scale: 8.0 +2023-03-25 22:50:06,025 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.73 vs. limit=2.0 +2023-03-25 22:50:18,328 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.331e+02 2.072e+02 2.562e+02 2.941e+02 5.028e+02, threshold=5.123e+02, percent-clipped=0.0 +2023-03-25 22:50:31,331 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4032.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 22:50:31,378 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3007, 1.3457, 1.5107, 1.1431, 1.2253, 1.4926, 1.3401, 1.6598], + device='cuda:1'), covar=tensor([0.1744, 0.2213, 0.1316, 0.1505, 0.1165, 0.1256, 0.2722, 0.0970], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0198, 0.0195, 0.0184, 0.0166, 0.0209, 0.0204, 0.0185], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:50:47,566 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-25 22:50:50,288 INFO [finetune.py:976] (1/7) Epoch 1, batch 4050, loss[loss=0.3379, simple_loss=0.3586, pruned_loss=0.1586, over 4769.00 frames. ], tot_loss[loss=0.3642, simple_loss=0.3779, pruned_loss=0.1752, over 951949.93 frames. ], batch size: 28, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:51:46,312 INFO [finetune.py:976] (1/7) Epoch 1, batch 4100, loss[loss=0.4091, simple_loss=0.4369, pruned_loss=0.1906, over 4916.00 frames. ], tot_loss[loss=0.3655, simple_loss=0.381, pruned_loss=0.175, over 953702.77 frames. ], batch size: 42, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:52:00,028 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.481e+02 2.010e+02 2.495e+02 2.957e+02 5.246e+02, threshold=4.990e+02, percent-clipped=1.0 +2023-03-25 22:52:34,826 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=4141.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:52:43,042 INFO [finetune.py:976] (1/7) Epoch 1, batch 4150, loss[loss=0.3917, simple_loss=0.4037, pruned_loss=0.1898, over 4896.00 frames. ], tot_loss[loss=0.3642, simple_loss=0.3807, pruned_loss=0.1739, over 954608.54 frames. ], batch size: 35, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:53:11,423 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4068, 1.1760, 1.1883, 1.2860, 1.7475, 1.2942, 0.9267, 1.0782], + device='cuda:1'), covar=tensor([0.2541, 0.2651, 0.2324, 0.2180, 0.2146, 0.1874, 0.3611, 0.2041], + device='cuda:1'), in_proj_covar=tensor([0.0206, 0.0192, 0.0178, 0.0165, 0.0213, 0.0164, 0.0189, 0.0168], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:53:43,097 INFO [finetune.py:976] (1/7) Epoch 1, batch 4200, loss[loss=0.3058, simple_loss=0.3485, pruned_loss=0.1315, over 4737.00 frames. ], tot_loss[loss=0.3627, simple_loss=0.3808, pruned_loss=0.1723, over 955591.76 frames. ], batch size: 54, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:53:43,841 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4202.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 22:53:43,862 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=4202.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:54:02,772 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.292e+02 2.089e+02 2.432e+02 2.936e+02 5.530e+02, threshold=4.864e+02, percent-clipped=1.0 +2023-03-25 22:54:44,475 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=4250.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:54:45,012 INFO [finetune.py:976] (1/7) Epoch 1, batch 4250, loss[loss=0.3738, simple_loss=0.3879, pruned_loss=0.1798, over 4897.00 frames. ], tot_loss[loss=0.3559, simple_loss=0.3752, pruned_loss=0.1683, over 956251.18 frames. ], batch size: 32, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:55:18,650 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-03-25 22:55:23,140 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4282.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:55:25,472 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.0015, 4.2001, 4.4647, 4.8198, 4.6731, 4.4454, 5.0979, 1.6545], + device='cuda:1'), covar=tensor([0.0675, 0.0934, 0.0674, 0.0771, 0.1261, 0.1169, 0.0533, 0.4934], + device='cuda:1'), in_proj_covar=tensor([0.0372, 0.0247, 0.0268, 0.0296, 0.0351, 0.0291, 0.0313, 0.0304], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 22:55:27,305 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=4288.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:55:35,148 INFO [finetune.py:976] (1/7) Epoch 1, batch 4300, loss[loss=0.3318, simple_loss=0.3508, pruned_loss=0.1564, over 4799.00 frames. ], tot_loss[loss=0.3496, simple_loss=0.3694, pruned_loss=0.1649, over 955178.87 frames. ], batch size: 25, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:55:45,448 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.425e+02 1.950e+02 2.267e+02 2.860e+02 4.056e+02, threshold=4.534e+02, percent-clipped=0.0 +2023-03-25 22:55:55,708 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-25 22:56:13,428 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4332.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 22:56:23,454 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.04 vs. limit=5.0 +2023-03-25 22:56:27,601 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4769, 1.3868, 1.1328, 1.1289, 1.4768, 1.7677, 1.4559, 1.0836], + device='cuda:1'), covar=tensor([0.0268, 0.0293, 0.0542, 0.0349, 0.0262, 0.0197, 0.0312, 0.0397], + device='cuda:1'), in_proj_covar=tensor([0.0084, 0.0113, 0.0133, 0.0111, 0.0105, 0.0100, 0.0088, 0.0111], + device='cuda:1'), out_proj_covar=tensor([6.6301e-05, 8.9902e-05, 1.0778e-04, 8.8375e-05, 8.3262e-05, 7.5056e-05, + 6.7793e-05, 8.7482e-05], device='cuda:1') +2023-03-25 22:56:35,733 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=4349.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:56:36,849 INFO [finetune.py:976] (1/7) Epoch 1, batch 4350, loss[loss=0.3467, simple_loss=0.3724, pruned_loss=0.1606, over 4912.00 frames. ], tot_loss[loss=0.3432, simple_loss=0.364, pruned_loss=0.1612, over 953922.60 frames. ], batch size: 37, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:57:17,014 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=4380.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 22:57:44,938 INFO [finetune.py:976] (1/7) Epoch 1, batch 4400, loss[loss=0.3837, simple_loss=0.403, pruned_loss=0.1822, over 4806.00 frames. ], tot_loss[loss=0.3432, simple_loss=0.364, pruned_loss=0.1612, over 954204.70 frames. ], batch size: 45, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:57:57,029 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.139e+02 1.978e+02 2.430e+02 2.895e+02 4.966e+02, threshold=4.860e+02, percent-clipped=1.0 +2023-03-25 22:58:13,127 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.93 vs. limit=5.0 +2023-03-25 22:58:28,114 INFO [finetune.py:976] (1/7) Epoch 1, batch 4450, loss[loss=0.3205, simple_loss=0.3611, pruned_loss=0.14, over 4921.00 frames. ], tot_loss[loss=0.3473, simple_loss=0.3681, pruned_loss=0.1632, over 950761.78 frames. ], batch size: 38, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:59:02,412 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0 +2023-03-25 22:59:09,843 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4497.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 22:59:12,226 INFO [finetune.py:976] (1/7) Epoch 1, batch 4500, loss[loss=0.3564, simple_loss=0.3832, pruned_loss=0.1649, over 4899.00 frames. ], tot_loss[loss=0.3479, simple_loss=0.3696, pruned_loss=0.1631, over 953242.60 frames. ], batch size: 36, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 22:59:29,321 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.227e+02 2.111e+02 2.516e+02 2.889e+02 5.762e+02, threshold=5.032e+02, percent-clipped=1.0 +2023-03-25 22:59:50,170 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.20 vs. limit=5.0 +2023-03-25 22:59:51,348 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.67 vs. limit=2.0 +2023-03-25 23:00:15,077 INFO [finetune.py:976] (1/7) Epoch 1, batch 4550, loss[loss=0.3611, simple_loss=0.3848, pruned_loss=0.1687, over 4881.00 frames. ], tot_loss[loss=0.349, simple_loss=0.3717, pruned_loss=0.1631, over 954092.81 frames. ], batch size: 35, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:00:44,376 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.75 vs. limit=5.0 +2023-03-25 23:00:55,785 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4582.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:01:05,362 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.87 vs. limit=5.0 +2023-03-25 23:01:25,594 INFO [finetune.py:976] (1/7) Epoch 1, batch 4600, loss[loss=0.3333, simple_loss=0.3555, pruned_loss=0.1556, over 4830.00 frames. ], tot_loss[loss=0.3471, simple_loss=0.3701, pruned_loss=0.162, over 955699.93 frames. ], batch size: 47, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:01:38,111 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.209e+02 2.082e+02 2.456e+02 3.064e+02 5.977e+02, threshold=4.911e+02, percent-clipped=1.0 +2023-03-25 23:01:59,354 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=4630.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:02:18,006 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4644.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:02:28,981 INFO [finetune.py:976] (1/7) Epoch 1, batch 4650, loss[loss=0.387, simple_loss=0.3716, pruned_loss=0.2012, over 4132.00 frames. ], tot_loss[loss=0.3427, simple_loss=0.3654, pruned_loss=0.16, over 954367.20 frames. ], batch size: 65, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:02:37,217 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2987, 3.7285, 3.8269, 4.1796, 4.0197, 3.8135, 4.4134, 1.4272], + device='cuda:1'), covar=tensor([0.0713, 0.0807, 0.0856, 0.0857, 0.1241, 0.1414, 0.0756, 0.5211], + device='cuda:1'), in_proj_covar=tensor([0.0372, 0.0246, 0.0269, 0.0297, 0.0350, 0.0291, 0.0313, 0.0304], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:03:06,337 INFO [finetune.py:976] (1/7) Epoch 1, batch 4700, loss[loss=0.2938, simple_loss=0.3184, pruned_loss=0.1346, over 4910.00 frames. ], tot_loss[loss=0.3357, simple_loss=0.3593, pruned_loss=0.1561, over 953803.45 frames. ], batch size: 32, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:03:20,653 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.265e+02 1.870e+02 2.224e+02 2.796e+02 5.273e+02, threshold=4.448e+02, percent-clipped=2.0 +2023-03-25 23:03:30,320 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5725, 1.4530, 1.8888, 2.7362, 2.0910, 1.9927, 0.8997, 2.2248], + device='cuda:1'), covar=tensor([0.1692, 0.1474, 0.1065, 0.0589, 0.0773, 0.1699, 0.1704, 0.0699], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0116, 0.0134, 0.0156, 0.0103, 0.0142, 0.0126, 0.0106], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 23:03:30,951 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0726, 0.6510, 0.9334, 0.8188, 0.7554, 0.7657, 0.8297, 0.9449], + device='cuda:1'), covar=tensor([4.2824, 8.5730, 5.3170, 6.7747, 7.2958, 4.8233, 9.2732, 4.8130], + device='cuda:1'), in_proj_covar=tensor([0.0204, 0.0232, 0.0217, 0.0245, 0.0228, 0.0200, 0.0255, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:03:59,441 INFO [finetune.py:976] (1/7) Epoch 1, batch 4750, loss[loss=0.284, simple_loss=0.323, pruned_loss=0.1224, over 4816.00 frames. ], tot_loss[loss=0.3321, simple_loss=0.3565, pruned_loss=0.1538, over 955011.04 frames. ], batch size: 51, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:04:36,374 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4797.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:04:39,215 INFO [finetune.py:976] (1/7) Epoch 1, batch 4800, loss[loss=0.4055, simple_loss=0.4065, pruned_loss=0.2023, over 4904.00 frames. ], tot_loss[loss=0.335, simple_loss=0.3594, pruned_loss=0.1553, over 953722.83 frames. ], batch size: 37, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:04:56,836 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.406e+02 2.096e+02 2.556e+02 3.186e+02 5.883e+02, threshold=5.111e+02, percent-clipped=4.0 +2023-03-25 23:05:32,153 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=4845.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:05:41,981 INFO [finetune.py:976] (1/7) Epoch 1, batch 4850, loss[loss=0.3049, simple_loss=0.3412, pruned_loss=0.1343, over 4868.00 frames. ], tot_loss[loss=0.3363, simple_loss=0.3621, pruned_loss=0.1552, over 953244.59 frames. ], batch size: 31, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:06:28,850 INFO [finetune.py:976] (1/7) Epoch 1, batch 4900, loss[loss=0.3146, simple_loss=0.3608, pruned_loss=0.1342, over 4733.00 frames. ], tot_loss[loss=0.3377, simple_loss=0.3642, pruned_loss=0.1556, over 955041.57 frames. ], batch size: 59, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:06:45,503 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.521e+02 2.056e+02 2.408e+02 2.893e+02 5.886e+02, threshold=4.817e+02, percent-clipped=2.0 +2023-03-25 23:06:49,263 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.76 vs. limit=5.0 +2023-03-25 23:07:15,535 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4944.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:07:19,810 INFO [finetune.py:976] (1/7) Epoch 1, batch 4950, loss[loss=0.3247, simple_loss=0.3491, pruned_loss=0.1502, over 4750.00 frames. ], tot_loss[loss=0.3358, simple_loss=0.3634, pruned_loss=0.1541, over 955072.35 frames. ], batch size: 28, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:07:49,157 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-25 23:08:11,782 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=4992.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:08:22,745 INFO [finetune.py:976] (1/7) Epoch 1, batch 5000, loss[loss=0.3562, simple_loss=0.3805, pruned_loss=0.166, over 4826.00 frames. ], tot_loss[loss=0.3293, simple_loss=0.3581, pruned_loss=0.1502, over 955465.18 frames. ], batch size: 40, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:08:32,727 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.012e+02 2.131e+02 2.461e+02 3.038e+02 5.796e+02, threshold=4.923e+02, percent-clipped=4.0 +2023-03-25 23:09:20,249 INFO [finetune.py:976] (1/7) Epoch 1, batch 5050, loss[loss=0.3189, simple_loss=0.3389, pruned_loss=0.1494, over 4927.00 frames. ], tot_loss[loss=0.3244, simple_loss=0.3533, pruned_loss=0.1477, over 956940.47 frames. ], batch size: 33, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:10:01,168 INFO [finetune.py:976] (1/7) Epoch 1, batch 5100, loss[loss=0.2755, simple_loss=0.3177, pruned_loss=0.1166, over 4790.00 frames. ], tot_loss[loss=0.3197, simple_loss=0.3489, pruned_loss=0.1453, over 957117.22 frames. ], batch size: 29, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:10:09,454 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.918e+02 2.379e+02 2.966e+02 8.444e+02, threshold=4.758e+02, percent-clipped=2.0 +2023-03-25 23:10:24,043 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3887, 1.3660, 1.3568, 1.3233, 0.8624, 2.2328, 0.6148, 1.3755], + device='cuda:1'), covar=tensor([0.3644, 0.2656, 0.2288, 0.2453, 0.2166, 0.0393, 0.3002, 0.1546], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0105, 0.0111, 0.0112, 0.0105, 0.0090, 0.0092, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0004, 0.0004], + device='cuda:1') +2023-03-25 23:10:34,835 INFO [finetune.py:976] (1/7) Epoch 1, batch 5150, loss[loss=0.3333, simple_loss=0.3641, pruned_loss=0.1512, over 4820.00 frames. ], tot_loss[loss=0.3201, simple_loss=0.3487, pruned_loss=0.1457, over 957164.64 frames. ], batch size: 38, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:11:14,837 INFO [finetune.py:976] (1/7) Epoch 1, batch 5200, loss[loss=0.3475, simple_loss=0.3959, pruned_loss=0.1496, over 4851.00 frames. ], tot_loss[loss=0.3263, simple_loss=0.3551, pruned_loss=0.1488, over 956229.83 frames. ], batch size: 44, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:11:24,772 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.601e+02 2.261e+02 2.570e+02 3.078e+02 5.221e+02, threshold=5.140e+02, percent-clipped=2.0 +2023-03-25 23:12:06,968 INFO [finetune.py:976] (1/7) Epoch 1, batch 5250, loss[loss=0.29, simple_loss=0.3305, pruned_loss=0.1248, over 4857.00 frames. ], tot_loss[loss=0.3264, simple_loss=0.3563, pruned_loss=0.1483, over 956092.16 frames. ], batch size: 31, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:12:55,129 INFO [finetune.py:976] (1/7) Epoch 1, batch 5300, loss[loss=0.3244, simple_loss=0.3525, pruned_loss=0.1481, over 4828.00 frames. ], tot_loss[loss=0.325, simple_loss=0.3559, pruned_loss=0.1471, over 956020.65 frames. ], batch size: 49, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:12:59,547 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.78 vs. limit=5.0 +2023-03-25 23:13:08,327 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.321e+02 2.032e+02 2.465e+02 2.907e+02 4.480e+02, threshold=4.930e+02, percent-clipped=0.0 +2023-03-25 23:13:49,134 INFO [finetune.py:976] (1/7) Epoch 1, batch 5350, loss[loss=0.3592, simple_loss=0.3863, pruned_loss=0.1661, over 4918.00 frames. ], tot_loss[loss=0.3236, simple_loss=0.355, pruned_loss=0.1461, over 953418.33 frames. ], batch size: 38, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:14:22,020 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7274, 2.4274, 2.1755, 1.0744, 2.2401, 2.1753, 1.8229, 2.2727], + device='cuda:1'), covar=tensor([0.0737, 0.1003, 0.1911, 0.2849, 0.1541, 0.1923, 0.2124, 0.1226], + device='cuda:1'), in_proj_covar=tensor([0.0158, 0.0176, 0.0189, 0.0173, 0.0196, 0.0194, 0.0199, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:14:48,105 INFO [finetune.py:976] (1/7) Epoch 1, batch 5400, loss[loss=0.2884, simple_loss=0.3225, pruned_loss=0.1272, over 4893.00 frames. ], tot_loss[loss=0.3185, simple_loss=0.35, pruned_loss=0.1435, over 954473.89 frames. ], batch size: 32, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:14:55,970 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.226e+02 1.939e+02 2.339e+02 2.729e+02 4.650e+02, threshold=4.678e+02, percent-clipped=0.0 +2023-03-25 23:15:02,322 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5488, 1.5662, 0.8072, 2.2393, 2.5255, 1.8742, 1.8599, 2.3702], + device='cuda:1'), covar=tensor([0.1519, 0.1967, 0.2478, 0.1102, 0.1954, 0.1844, 0.1287, 0.1792], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0096, 0.0115, 0.0092, 0.0123, 0.0095, 0.0098, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 23:15:39,350 INFO [finetune.py:976] (1/7) Epoch 1, batch 5450, loss[loss=0.3609, simple_loss=0.3874, pruned_loss=0.1672, over 4918.00 frames. ], tot_loss[loss=0.3124, simple_loss=0.3441, pruned_loss=0.1403, over 953360.30 frames. ], batch size: 37, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:16:31,588 INFO [finetune.py:976] (1/7) Epoch 1, batch 5500, loss[loss=0.307, simple_loss=0.3424, pruned_loss=0.1358, over 4772.00 frames. ], tot_loss[loss=0.3053, simple_loss=0.3379, pruned_loss=0.1363, over 952479.86 frames. ], batch size: 51, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:16:45,961 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.302e+02 2.026e+02 2.277e+02 2.875e+02 1.009e+03, threshold=4.553e+02, percent-clipped=5.0 +2023-03-25 23:17:20,559 INFO [finetune.py:976] (1/7) Epoch 1, batch 5550, loss[loss=0.2637, simple_loss=0.2957, pruned_loss=0.1158, over 4722.00 frames. ], tot_loss[loss=0.3085, simple_loss=0.3406, pruned_loss=0.1381, over 953441.15 frames. ], batch size: 23, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:18:01,090 INFO [finetune.py:976] (1/7) Epoch 1, batch 5600, loss[loss=0.3316, simple_loss=0.3692, pruned_loss=0.147, over 4845.00 frames. ], tot_loss[loss=0.3128, simple_loss=0.346, pruned_loss=0.1398, over 954391.03 frames. ], batch size: 49, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:18:19,452 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.468e+02 1.839e+02 2.287e+02 2.793e+02 4.099e+02, threshold=4.573e+02, percent-clipped=0.0 +2023-03-25 23:18:59,419 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=5650.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 23:18:59,874 INFO [finetune.py:976] (1/7) Epoch 1, batch 5650, loss[loss=0.3852, simple_loss=0.3892, pruned_loss=0.1906, over 4338.00 frames. ], tot_loss[loss=0.3186, simple_loss=0.3513, pruned_loss=0.143, over 952079.75 frames. ], batch size: 19, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:19:26,212 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-25 23:19:32,237 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-25 23:19:35,520 INFO [finetune.py:976] (1/7) Epoch 1, batch 5700, loss[loss=0.2501, simple_loss=0.2841, pruned_loss=0.108, over 4372.00 frames. ], tot_loss[loss=0.3139, simple_loss=0.3456, pruned_loss=0.141, over 934810.70 frames. ], batch size: 19, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:19:35,650 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7537, 1.4955, 1.3041, 1.3703, 1.5012, 1.4004, 1.3913, 2.3513], + device='cuda:1'), covar=tensor([3.2892, 3.0592, 2.6317, 4.1409, 2.4061, 2.0761, 3.3084, 0.8502], + device='cuda:1'), in_proj_covar=tensor([0.0225, 0.0214, 0.0196, 0.0249, 0.0209, 0.0178, 0.0213, 0.0159], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:19:41,538 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=5711.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 23:19:43,179 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.079e+02 1.818e+02 2.245e+02 2.685e+02 4.321e+02, threshold=4.489e+02, percent-clipped=0.0 +2023-03-25 23:20:08,411 INFO [finetune.py:976] (1/7) Epoch 2, batch 0, loss[loss=0.3361, simple_loss=0.3693, pruned_loss=0.1515, over 4809.00 frames. ], tot_loss[loss=0.3361, simple_loss=0.3693, pruned_loss=0.1515, over 4809.00 frames. ], batch size: 33, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:20:08,411 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-25 23:20:24,999 INFO [finetune.py:1010] (1/7) Epoch 2, validation: loss=0.2224, simple_loss=0.2847, pruned_loss=0.08, over 2265189.00 frames. +2023-03-25 23:20:25,000 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6048MB +2023-03-25 23:20:56,833 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=5755.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:20:59,302 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2730, 1.3843, 1.3040, 0.7954, 1.6493, 1.3289, 1.2845, 1.3484], + device='cuda:1'), covar=tensor([0.0837, 0.0816, 0.0913, 0.1190, 0.0619, 0.0989, 0.0932, 0.1306], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0132, 0.0138, 0.0126, 0.0107, 0.0136, 0.0142, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:21:22,834 INFO [finetune.py:976] (1/7) Epoch 2, batch 50, loss[loss=0.2965, simple_loss=0.3418, pruned_loss=0.1256, over 4913.00 frames. ], tot_loss[loss=0.3114, simple_loss=0.3452, pruned_loss=0.1388, over 214359.06 frames. ], batch size: 46, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:21:38,657 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.63 vs. limit=2.0 +2023-03-25 23:21:48,460 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0492, 1.8549, 1.8655, 0.8057, 1.9494, 2.3754, 1.8526, 1.9510], + device='cuda:1'), covar=tensor([0.0985, 0.0744, 0.0706, 0.0917, 0.0739, 0.0486, 0.0485, 0.0522], + device='cuda:1'), in_proj_covar=tensor([0.0120, 0.0144, 0.0112, 0.0122, 0.0122, 0.0111, 0.0137, 0.0139], + device='cuda:1'), out_proj_covar=tensor([9.0596e-05, 1.0742e-04, 8.1836e-05, 8.9549e-05, 8.8569e-05, 8.1705e-05, + 1.0228e-04, 1.0322e-04], device='cuda:1') +2023-03-25 23:21:54,356 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.238e+02 1.870e+02 2.317e+02 2.912e+02 7.564e+02, threshold=4.633e+02, percent-clipped=3.0 +2023-03-25 23:21:55,671 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=5816.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:22:11,076 INFO [finetune.py:976] (1/7) Epoch 2, batch 100, loss[loss=0.2565, simple_loss=0.3019, pruned_loss=0.1055, over 4759.00 frames. ], tot_loss[loss=0.3012, simple_loss=0.336, pruned_loss=0.1332, over 378971.60 frames. ], batch size: 28, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:22:23,814 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9516, 1.1954, 0.8658, 1.7670, 2.1805, 1.7829, 1.4108, 1.9659], + device='cuda:1'), covar=tensor([0.1740, 0.2452, 0.2649, 0.1440, 0.2334, 0.2011, 0.1649, 0.1932], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0096, 0.0115, 0.0092, 0.0124, 0.0095, 0.0098, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 23:22:36,192 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=5858.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:22:49,647 INFO [finetune.py:976] (1/7) Epoch 2, batch 150, loss[loss=0.2822, simple_loss=0.3272, pruned_loss=0.1186, over 4921.00 frames. ], tot_loss[loss=0.2945, simple_loss=0.3292, pruned_loss=0.1299, over 507198.37 frames. ], batch size: 33, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:23:18,195 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.293e+02 1.883e+02 2.329e+02 2.858e+02 5.160e+02, threshold=4.657e+02, percent-clipped=2.0 +2023-03-25 23:23:21,830 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=5919.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:23:28,108 INFO [finetune.py:976] (1/7) Epoch 2, batch 200, loss[loss=0.2525, simple_loss=0.2926, pruned_loss=0.1062, over 4718.00 frames. ], tot_loss[loss=0.2956, simple_loss=0.3292, pruned_loss=0.131, over 608413.57 frames. ], batch size: 23, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:23:30,701 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.74 vs. limit=5.0 +2023-03-25 23:24:01,218 INFO [finetune.py:976] (1/7) Epoch 2, batch 250, loss[loss=0.2821, simple_loss=0.3244, pruned_loss=0.1199, over 4768.00 frames. ], tot_loss[loss=0.3006, simple_loss=0.3341, pruned_loss=0.1335, over 682701.16 frames. ], batch size: 26, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:24:09,506 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-25 23:24:20,388 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=5990.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:24:42,067 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6006.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 23:24:48,586 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.078e+02 1.968e+02 2.365e+02 2.842e+02 7.361e+02, threshold=4.731e+02, percent-clipped=2.0 +2023-03-25 23:25:01,945 INFO [finetune.py:976] (1/7) Epoch 2, batch 300, loss[loss=0.2772, simple_loss=0.3115, pruned_loss=0.1215, over 4737.00 frames. ], tot_loss[loss=0.3034, simple_loss=0.3379, pruned_loss=0.1344, over 743590.51 frames. ], batch size: 23, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:25:08,373 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6101, 1.7227, 1.7108, 1.9259, 1.8529, 4.1592, 1.4452, 1.8736], + device='cuda:1'), covar=tensor([0.1012, 0.1605, 0.1386, 0.1064, 0.1613, 0.0190, 0.1538, 0.1759], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0078, 0.0075, 0.0077, 0.0090, 0.0078, 0.0083, 0.0077], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-25 23:25:23,431 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6043.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:25:33,855 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6051.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:25:40,142 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.27 vs. limit=5.0 +2023-03-25 23:25:53,965 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6204, 1.4664, 1.6711, 1.7680, 2.3647, 1.7032, 1.5378, 1.2722], + device='cuda:1'), covar=tensor([0.3586, 0.3412, 0.2647, 0.2618, 0.2917, 0.2044, 0.4095, 0.2774], + device='cuda:1'), in_proj_covar=tensor([0.0215, 0.0199, 0.0184, 0.0171, 0.0220, 0.0168, 0.0200, 0.0174], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:26:11,334 INFO [finetune.py:976] (1/7) Epoch 2, batch 350, loss[loss=0.3182, simple_loss=0.3614, pruned_loss=0.1375, over 4841.00 frames. ], tot_loss[loss=0.3061, simple_loss=0.3413, pruned_loss=0.1355, over 792056.90 frames. ], batch size: 47, lr: 4.00e-03, grad_scale: 32.0 +2023-03-25 23:26:35,010 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6104.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:26:39,146 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6111.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:26:41,480 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.275e+02 2.097e+02 2.536e+02 2.955e+02 5.135e+02, threshold=5.071e+02, percent-clipped=1.0 +2023-03-25 23:26:59,638 INFO [finetune.py:976] (1/7) Epoch 2, batch 400, loss[loss=0.3404, simple_loss=0.3719, pruned_loss=0.1545, over 4833.00 frames. ], tot_loss[loss=0.3063, simple_loss=0.342, pruned_loss=0.1353, over 829581.03 frames. ], batch size: 30, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:27:09,971 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6135.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:27:46,757 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0408, 1.6696, 2.6132, 1.4549, 1.9785, 2.0688, 1.7307, 2.3348], + device='cuda:1'), covar=tensor([0.1883, 0.2144, 0.1742, 0.2557, 0.1098, 0.1852, 0.2671, 0.1190], + device='cuda:1'), in_proj_covar=tensor([0.0200, 0.0202, 0.0200, 0.0191, 0.0172, 0.0218, 0.0209, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:27:49,726 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6170.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:27:59,429 INFO [finetune.py:976] (1/7) Epoch 2, batch 450, loss[loss=0.3068, simple_loss=0.3503, pruned_loss=0.1317, over 4833.00 frames. ], tot_loss[loss=0.3047, simple_loss=0.3406, pruned_loss=0.1344, over 859187.07 frames. ], batch size: 33, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:28:07,961 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7993, 1.6928, 1.3604, 2.0925, 2.1381, 1.5124, 2.4627, 1.7182], + device='cuda:1'), covar=tensor([0.3459, 0.7957, 0.7055, 0.6725, 0.4011, 0.3408, 0.5239, 0.4999], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0186, 0.0228, 0.0240, 0.0201, 0.0173, 0.0188, 0.0180], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:28:14,471 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6196.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:28:25,734 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6203.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:28:33,020 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6211.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:28:34,814 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6214.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:28:35,342 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.140e+02 1.932e+02 2.244e+02 2.718e+02 3.817e+02, threshold=4.487e+02, percent-clipped=0.0 +2023-03-25 23:28:40,959 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0350, 1.8203, 1.9932, 2.1111, 2.6742, 2.0382, 1.7672, 1.7745], + device='cuda:1'), covar=tensor([0.2495, 0.2497, 0.2031, 0.1971, 0.1922, 0.1482, 0.2817, 0.1977], + device='cuda:1'), in_proj_covar=tensor([0.0215, 0.0199, 0.0185, 0.0171, 0.0220, 0.0167, 0.0200, 0.0174], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:28:45,039 INFO [finetune.py:976] (1/7) Epoch 2, batch 500, loss[loss=0.2359, simple_loss=0.2849, pruned_loss=0.09341, over 4770.00 frames. ], tot_loss[loss=0.3009, simple_loss=0.3373, pruned_loss=0.1323, over 880831.55 frames. ], batch size: 28, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:28:52,711 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6231.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:29:19,934 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9533, 2.0740, 1.7285, 1.3051, 2.3606, 2.1163, 1.9670, 1.8576], + device='cuda:1'), covar=tensor([0.0728, 0.0597, 0.0941, 0.1053, 0.0351, 0.0765, 0.0878, 0.1022], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0134, 0.0139, 0.0128, 0.0109, 0.0138, 0.0144, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:29:25,576 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6264.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:29:26,785 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6266.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:29:35,113 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6272.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:29:38,740 INFO [finetune.py:976] (1/7) Epoch 2, batch 550, loss[loss=0.3535, simple_loss=0.3636, pruned_loss=0.1717, over 4898.00 frames. ], tot_loss[loss=0.2981, simple_loss=0.3339, pruned_loss=0.1312, over 896591.67 frames. ], batch size: 43, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:30:17,706 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6306.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 23:30:28,370 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5469, 1.2693, 1.3541, 1.5082, 1.6822, 1.4605, 0.8487, 1.2639], + device='cuda:1'), covar=tensor([0.2926, 0.2871, 0.2312, 0.2171, 0.2476, 0.1662, 0.3827, 0.2253], + device='cuda:1'), in_proj_covar=tensor([0.0215, 0.0199, 0.0185, 0.0171, 0.0220, 0.0168, 0.0200, 0.0174], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:30:28,974 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5791, 1.3048, 1.4298, 1.5417, 2.0797, 1.4953, 1.2323, 1.2714], + device='cuda:1'), covar=tensor([0.2968, 0.3128, 0.2556, 0.2307, 0.2546, 0.1848, 0.3866, 0.2290], + device='cuda:1'), in_proj_covar=tensor([0.0215, 0.0199, 0.0185, 0.0171, 0.0220, 0.0168, 0.0200, 0.0174], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:30:29,435 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.346e+02 1.947e+02 2.353e+02 2.718e+02 5.175e+02, threshold=4.705e+02, percent-clipped=1.0 +2023-03-25 23:30:30,145 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7554, 1.2629, 0.8665, 1.6968, 2.0717, 1.4065, 1.3797, 1.7751], + device='cuda:1'), covar=tensor([0.1639, 0.2151, 0.2259, 0.1180, 0.2207, 0.2162, 0.1497, 0.1849], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0096, 0.0115, 0.0092, 0.0123, 0.0096, 0.0098, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 23:30:30,314 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-03-25 23:30:37,371 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3826, 3.8334, 3.9448, 4.2724, 4.1431, 3.8535, 4.4807, 1.5427], + device='cuda:1'), covar=tensor([0.0746, 0.0794, 0.0774, 0.0926, 0.1156, 0.1401, 0.0617, 0.5033], + device='cuda:1'), in_proj_covar=tensor([0.0370, 0.0246, 0.0270, 0.0298, 0.0349, 0.0291, 0.0312, 0.0304], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:30:41,994 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6327.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:30:48,006 INFO [finetune.py:976] (1/7) Epoch 2, batch 600, loss[loss=0.2885, simple_loss=0.3191, pruned_loss=0.1289, over 4769.00 frames. ], tot_loss[loss=0.2982, simple_loss=0.334, pruned_loss=0.1312, over 906783.92 frames. ], batch size: 26, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:31:07,299 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6346.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:31:12,644 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6354.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 23:31:18,491 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.90 vs. limit=5.0 +2023-03-25 23:31:27,616 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.78 vs. limit=5.0 +2023-03-25 23:31:28,049 INFO [finetune.py:976] (1/7) Epoch 2, batch 650, loss[loss=0.3221, simple_loss=0.3724, pruned_loss=0.1359, over 4921.00 frames. ], tot_loss[loss=0.3027, simple_loss=0.3385, pruned_loss=0.1335, over 918024.41 frames. ], batch size: 42, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:31:42,390 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6399.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:31:42,422 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0085, 1.2635, 0.9205, 1.3384, 1.3280, 2.4431, 1.1004, 1.3649], + device='cuda:1'), covar=tensor([0.1147, 0.1731, 0.1307, 0.1056, 0.1756, 0.0392, 0.1633, 0.1854], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0078, 0.0075, 0.0077, 0.0090, 0.0079, 0.0083, 0.0077], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-25 23:31:43,187 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4326, 1.4629, 1.3984, 1.5281, 0.9559, 2.8859, 1.0360, 1.6528], + device='cuda:1'), covar=tensor([0.3694, 0.2576, 0.2310, 0.2411, 0.2312, 0.0270, 0.3431, 0.1584], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0107, 0.0113, 0.0114, 0.0108, 0.0092, 0.0095, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-25 23:31:51,328 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6411.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:31:53,613 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.435e+02 2.011e+02 2.373e+02 2.999e+02 4.783e+02, threshold=4.746e+02, percent-clipped=1.0 +2023-03-25 23:32:01,501 INFO [finetune.py:976] (1/7) Epoch 2, batch 700, loss[loss=0.3184, simple_loss=0.3557, pruned_loss=0.1406, over 4727.00 frames. ], tot_loss[loss=0.3057, simple_loss=0.3417, pruned_loss=0.1348, over 928328.13 frames. ], batch size: 59, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:32:04,775 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-25 23:32:22,201 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6459.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:32:24,527 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6462.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:32:24,564 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6462, 1.4270, 1.5931, 1.6594, 2.5446, 1.6634, 1.5496, 1.2911], + device='cuda:1'), covar=tensor([0.3099, 0.3166, 0.2535, 0.2351, 0.2428, 0.1779, 0.3513, 0.2401], + device='cuda:1'), in_proj_covar=tensor([0.0215, 0.0200, 0.0185, 0.0172, 0.0221, 0.0168, 0.0200, 0.0175], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:32:27,978 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6467.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:32:29,161 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3401, 1.2086, 1.5987, 2.2633, 1.6238, 2.0699, 0.8625, 1.8569], + device='cuda:1'), covar=tensor([0.2134, 0.1947, 0.1441, 0.0992, 0.1179, 0.1357, 0.1969, 0.1100], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0119, 0.0138, 0.0161, 0.0105, 0.0145, 0.0130, 0.0108], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-25 23:32:34,494 INFO [finetune.py:976] (1/7) Epoch 2, batch 750, loss[loss=0.3206, simple_loss=0.3475, pruned_loss=0.1468, over 4838.00 frames. ], tot_loss[loss=0.3075, simple_loss=0.3433, pruned_loss=0.1358, over 934075.91 frames. ], batch size: 47, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:32:37,633 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8479, 3.3763, 3.4983, 3.7846, 3.5863, 3.4244, 3.9430, 1.3426], + device='cuda:1'), covar=tensor([0.0780, 0.0846, 0.0750, 0.0843, 0.1269, 0.1339, 0.0789, 0.4556], + device='cuda:1'), in_proj_covar=tensor([0.0371, 0.0248, 0.0271, 0.0299, 0.0349, 0.0292, 0.0313, 0.0305], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:32:42,485 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6491.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:32:58,331 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6514.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:32:58,813 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.464e+02 1.998e+02 2.267e+02 2.688e+02 5.596e+02, threshold=4.534e+02, percent-clipped=2.0 +2023-03-25 23:33:04,284 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6523.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 23:33:06,049 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6526.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:33:07,176 INFO [finetune.py:976] (1/7) Epoch 2, batch 800, loss[loss=0.2456, simple_loss=0.287, pruned_loss=0.1021, over 4733.00 frames. ], tot_loss[loss=0.3066, simple_loss=0.343, pruned_loss=0.1351, over 939816.58 frames. ], batch size: 59, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:33:07,373 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6528.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:33:19,511 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-03-25 23:33:40,447 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6559.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:33:42,336 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6562.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:33:46,042 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6567.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:33:58,058 INFO [finetune.py:976] (1/7) Epoch 2, batch 850, loss[loss=0.3019, simple_loss=0.3282, pruned_loss=0.1378, over 4813.00 frames. ], tot_loss[loss=0.3024, simple_loss=0.3391, pruned_loss=0.1328, over 943075.26 frames. ], batch size: 39, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:34:37,095 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.216e+02 1.789e+02 2.218e+02 2.697e+02 5.451e+02, threshold=4.436e+02, percent-clipped=1.0 +2023-03-25 23:34:42,614 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6622.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:34:46,289 INFO [finetune.py:976] (1/7) Epoch 2, batch 900, loss[loss=0.2159, simple_loss=0.2729, pruned_loss=0.07942, over 4764.00 frames. ], tot_loss[loss=0.2984, simple_loss=0.3354, pruned_loss=0.1307, over 946403.07 frames. ], batch size: 27, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:34:47,097 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6629.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:34:55,686 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-25 23:35:02,804 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6646.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:35:10,462 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6864, 1.7398, 1.6707, 1.1560, 1.9646, 1.8083, 1.7595, 1.6402], + device='cuda:1'), covar=tensor([0.0680, 0.0663, 0.0748, 0.1013, 0.0517, 0.0749, 0.0687, 0.1127], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0133, 0.0139, 0.0128, 0.0109, 0.0138, 0.0144, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:35:25,036 INFO [finetune.py:976] (1/7) Epoch 2, batch 950, loss[loss=0.3321, simple_loss=0.3616, pruned_loss=0.1513, over 4932.00 frames. ], tot_loss[loss=0.295, simple_loss=0.3322, pruned_loss=0.1289, over 948765.51 frames. ], batch size: 38, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:35:32,463 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6690.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:35:34,865 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6694.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:35:37,897 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6699.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:35:48,888 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.136e+02 1.729e+02 2.077e+02 2.706e+02 4.933e+02, threshold=4.155e+02, percent-clipped=1.0 +2023-03-25 23:36:03,623 INFO [finetune.py:976] (1/7) Epoch 2, batch 1000, loss[loss=0.2419, simple_loss=0.2879, pruned_loss=0.09796, over 4912.00 frames. ], tot_loss[loss=0.2961, simple_loss=0.3331, pruned_loss=0.1296, over 952119.33 frames. ], batch size: 32, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:36:22,930 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6747.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:37:01,238 INFO [finetune.py:976] (1/7) Epoch 2, batch 1050, loss[loss=0.2891, simple_loss=0.3334, pruned_loss=0.1224, over 4758.00 frames. ], tot_loss[loss=0.2992, simple_loss=0.3365, pruned_loss=0.1309, over 952955.90 frames. ], batch size: 59, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:37:09,221 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6791.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:37:29,664 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.307e+02 2.050e+02 2.547e+02 2.958e+02 5.414e+02, threshold=5.095e+02, percent-clipped=8.0 +2023-03-25 23:37:37,311 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6818.0, num_to_drop=1, layers_to_drop={3} +2023-03-25 23:37:40,265 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6823.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:37:47,814 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6826.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:37:48,962 INFO [finetune.py:976] (1/7) Epoch 2, batch 1100, loss[loss=0.3285, simple_loss=0.3677, pruned_loss=0.1447, over 4723.00 frames. ], tot_loss[loss=0.2999, simple_loss=0.3379, pruned_loss=0.131, over 954321.02 frames. ], batch size: 59, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:37:56,677 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6839.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:38:14,090 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6859.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:38:23,164 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5591, 1.7578, 1.5415, 1.7474, 1.0462, 3.5912, 1.3308, 1.8933], + device='cuda:1'), covar=tensor([0.3953, 0.2519, 0.2305, 0.2490, 0.2220, 0.0197, 0.2999, 0.1555], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0106, 0.0112, 0.0113, 0.0108, 0.0092, 0.0095, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-25 23:38:23,786 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6867.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:38:28,919 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6874.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:38:31,316 INFO [finetune.py:976] (1/7) Epoch 2, batch 1150, loss[loss=0.2666, simple_loss=0.3168, pruned_loss=0.1082, over 4891.00 frames. ], tot_loss[loss=0.2984, simple_loss=0.337, pruned_loss=0.1299, over 954902.35 frames. ], batch size: 32, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:38:31,414 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6878.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 23:38:34,371 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5882, 1.3625, 1.2423, 1.2091, 1.6527, 1.7740, 1.4728, 1.1680], + device='cuda:1'), covar=tensor([0.0302, 0.0405, 0.0567, 0.0415, 0.0275, 0.0323, 0.0305, 0.0456], + device='cuda:1'), in_proj_covar=tensor([0.0083, 0.0112, 0.0131, 0.0112, 0.0103, 0.0098, 0.0088, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.4821e-05, 8.8561e-05, 1.0632e-04, 8.8825e-05, 8.2287e-05, 7.3265e-05, + 6.8239e-05, 8.5445e-05], device='cuda:1') +2023-03-25 23:38:52,177 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6907.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:38:53,403 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6909.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:38:56,974 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.223e+02 2.014e+02 2.348e+02 2.865e+02 5.036e+02, threshold=4.696e+02, percent-clipped=0.0 +2023-03-25 23:38:57,045 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6915.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:38:59,085 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4478, 1.2391, 1.0291, 1.0296, 1.1963, 1.1198, 1.1322, 1.9764], + device='cuda:1'), covar=tensor([2.6795, 2.6377, 2.1273, 3.1806, 2.0394, 1.5237, 2.6839, 0.7720], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0222, 0.0202, 0.0258, 0.0215, 0.0183, 0.0220, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:39:07,721 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6922.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:39:17,349 INFO [finetune.py:976] (1/7) Epoch 2, batch 1200, loss[loss=0.2755, simple_loss=0.3119, pruned_loss=0.1196, over 4741.00 frames. ], tot_loss[loss=0.2952, simple_loss=0.334, pruned_loss=0.1282, over 956013.38 frames. ], batch size: 23, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:39:21,050 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5557, 3.3578, 3.2756, 1.4368, 3.5126, 2.5870, 0.8686, 2.2423], + device='cuda:1'), covar=tensor([0.2230, 0.1897, 0.1617, 0.3444, 0.1121, 0.1039, 0.4170, 0.1535], + device='cuda:1'), in_proj_covar=tensor([0.0156, 0.0163, 0.0165, 0.0127, 0.0155, 0.0118, 0.0147, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-25 23:39:31,043 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6939.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 23:39:47,129 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-25 23:39:49,900 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=6970.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:39:49,969 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6970.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:39:56,177 INFO [finetune.py:976] (1/7) Epoch 2, batch 1250, loss[loss=0.2578, simple_loss=0.3101, pruned_loss=0.1028, over 4823.00 frames. ], tot_loss[loss=0.2915, simple_loss=0.3304, pruned_loss=0.1263, over 957171.28 frames. ], batch size: 41, lr: 4.00e-03, grad_scale: 16.0 +2023-03-25 23:40:00,602 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6985.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:40:13,851 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3043, 1.5800, 1.2731, 1.5029, 1.6004, 3.0044, 1.3364, 1.6963], + device='cuda:1'), covar=tensor([0.1182, 0.1669, 0.1255, 0.1181, 0.1681, 0.0293, 0.1600, 0.1760], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0078, 0.0075, 0.0078, 0.0090, 0.0079, 0.0084, 0.0077], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-25 23:40:23,990 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.288e+02 1.918e+02 2.412e+02 2.798e+02 4.765e+02, threshold=4.825e+02, percent-clipped=1.0 +2023-03-25 23:40:31,022 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-25 23:40:33,987 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5560, 1.3458, 1.4140, 1.5670, 1.8838, 1.5256, 0.9478, 1.3612], + device='cuda:1'), covar=tensor([0.3067, 0.3112, 0.2517, 0.2360, 0.2371, 0.1605, 0.3869, 0.2375], + device='cuda:1'), in_proj_covar=tensor([0.0216, 0.0200, 0.0186, 0.0172, 0.0221, 0.0166, 0.0201, 0.0175], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:40:39,101 INFO [finetune.py:976] (1/7) Epoch 2, batch 1300, loss[loss=0.3265, simple_loss=0.3455, pruned_loss=0.1538, over 4821.00 frames. ], tot_loss[loss=0.2879, simple_loss=0.3266, pruned_loss=0.1246, over 957692.07 frames. ], batch size: 38, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:40:46,516 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.58 vs. limit=2.0 +2023-03-25 23:40:52,827 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7048.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 23:41:17,561 INFO [finetune.py:976] (1/7) Epoch 2, batch 1350, loss[loss=0.2336, simple_loss=0.2864, pruned_loss=0.09043, over 4735.00 frames. ], tot_loss[loss=0.2875, simple_loss=0.326, pruned_loss=0.1245, over 957094.25 frames. ], batch size: 23, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:41:33,881 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0390, 0.7160, 0.8803, 0.7566, 1.1086, 1.1368, 0.9647, 0.8059], + device='cuda:1'), covar=tensor([0.0322, 0.0444, 0.0595, 0.0419, 0.0358, 0.0438, 0.0361, 0.0444], + device='cuda:1'), in_proj_covar=tensor([0.0082, 0.0111, 0.0131, 0.0112, 0.0103, 0.0098, 0.0088, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.4404e-05, 8.8034e-05, 1.0604e-04, 8.8636e-05, 8.1934e-05, 7.2982e-05, + 6.7828e-05, 8.4781e-05], device='cuda:1') +2023-03-25 23:41:35,824 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.57 vs. limit=5.0 +2023-03-25 23:41:41,157 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7109.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 23:41:47,302 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.456e+02 1.876e+02 2.208e+02 2.586e+02 5.614e+02, threshold=4.416e+02, percent-clipped=5.0 +2023-03-25 23:41:49,205 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7118.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 23:41:52,295 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7123.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:41:55,448 INFO [finetune.py:976] (1/7) Epoch 2, batch 1400, loss[loss=0.2676, simple_loss=0.3204, pruned_loss=0.1074, over 4899.00 frames. ], tot_loss[loss=0.2921, simple_loss=0.3309, pruned_loss=0.1267, over 958664.79 frames. ], batch size: 36, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:42:12,511 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7243, 1.5615, 1.1691, 1.5540, 1.4826, 1.3261, 1.3864, 2.3090], + device='cuda:1'), covar=tensor([2.6944, 2.8907, 2.2841, 3.9493, 2.5266, 1.7349, 3.0971, 0.8808], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0223, 0.0203, 0.0259, 0.0216, 0.0184, 0.0220, 0.0165], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:42:20,945 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0651, 1.8060, 2.0133, 0.9332, 2.0638, 2.4112, 1.8574, 2.0153], + device='cuda:1'), covar=tensor([0.1230, 0.0879, 0.0510, 0.0867, 0.0833, 0.0508, 0.0466, 0.0495], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0148, 0.0113, 0.0125, 0.0125, 0.0113, 0.0140, 0.0138], + device='cuda:1'), out_proj_covar=tensor([9.2731e-05, 1.1033e-04, 8.2885e-05, 9.2100e-05, 9.0651e-05, 8.3470e-05, + 1.0462e-04, 1.0256e-04], device='cuda:1') +2023-03-25 23:42:31,263 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7163.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:42:33,030 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=7166.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:42:36,030 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=7171.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:42:36,091 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7495, 1.4553, 2.0152, 1.3601, 1.7813, 1.8739, 1.4689, 2.0237], + device='cuda:1'), covar=tensor([0.1599, 0.2667, 0.1688, 0.2218, 0.1149, 0.1827, 0.2993, 0.1107], + device='cuda:1'), in_proj_covar=tensor([0.0204, 0.0205, 0.0205, 0.0195, 0.0177, 0.0223, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:42:40,186 INFO [finetune.py:976] (1/7) Epoch 2, batch 1450, loss[loss=0.3432, simple_loss=0.3751, pruned_loss=0.1556, over 4855.00 frames. ], tot_loss[loss=0.2948, simple_loss=0.334, pruned_loss=0.1278, over 959074.06 frames. ], batch size: 44, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:42:40,424 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.96 vs. limit=5.0 +2023-03-25 23:43:29,960 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.045e+02 1.933e+02 2.198e+02 2.731e+02 4.077e+02, threshold=4.395e+02, percent-clipped=0.0 +2023-03-25 23:43:40,432 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7224.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 23:43:42,758 INFO [finetune.py:976] (1/7) Epoch 2, batch 1500, loss[loss=0.2897, simple_loss=0.3316, pruned_loss=0.1239, over 4901.00 frames. ], tot_loss[loss=0.2949, simple_loss=0.335, pruned_loss=0.1275, over 959752.86 frames. ], batch size: 36, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:43:51,987 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7234.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 23:44:03,368 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5764, 1.4678, 1.7276, 1.0744, 1.5724, 1.9404, 1.7708, 1.5581], + device='cuda:1'), covar=tensor([0.1187, 0.0878, 0.0604, 0.0790, 0.0558, 0.0453, 0.0440, 0.0722], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0148, 0.0113, 0.0126, 0.0125, 0.0113, 0.0140, 0.0139], + device='cuda:1'), out_proj_covar=tensor([9.2689e-05, 1.1051e-04, 8.2957e-05, 9.2306e-05, 9.0581e-05, 8.3416e-05, + 1.0473e-04, 1.0279e-04], device='cuda:1') +2023-03-25 23:44:23,198 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0234, 2.5919, 1.8947, 1.6041, 2.7007, 2.6322, 2.2910, 2.3096], + device='cuda:1'), covar=tensor([0.1085, 0.0684, 0.1215, 0.1290, 0.0613, 0.0903, 0.0953, 0.1061], + device='cuda:1'), in_proj_covar=tensor([0.0137, 0.0133, 0.0140, 0.0128, 0.0109, 0.0138, 0.0144, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:44:36,526 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7265.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:44:47,907 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7275.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:44:49,672 INFO [finetune.py:976] (1/7) Epoch 2, batch 1550, loss[loss=0.2678, simple_loss=0.3187, pruned_loss=0.1084, over 4783.00 frames. ], tot_loss[loss=0.2947, simple_loss=0.3351, pruned_loss=0.1272, over 958394.28 frames. ], batch size: 26, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:44:57,984 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7285.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:45:06,993 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6043, 1.3301, 1.1497, 1.1593, 1.3347, 1.2431, 1.2647, 2.1130], + device='cuda:1'), covar=tensor([2.6599, 2.6487, 2.1241, 3.2419, 2.1047, 1.5338, 2.6090, 0.7767], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0224, 0.0204, 0.0260, 0.0217, 0.0184, 0.0221, 0.0166], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:45:16,405 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4613, 0.9000, 1.3123, 1.1457, 1.1379, 1.1285, 1.0627, 1.2581], + device='cuda:1'), covar=tensor([1.7851, 3.4102, 2.3310, 2.5783, 3.0529, 2.1145, 3.8372, 2.1230], + device='cuda:1'), in_proj_covar=tensor([0.0220, 0.0253, 0.0239, 0.0266, 0.0244, 0.0216, 0.0275, 0.0211], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:45:31,201 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.380e+02 1.989e+02 2.300e+02 2.720e+02 4.709e+02, threshold=4.600e+02, percent-clipped=4.0 +2023-03-25 23:45:39,140 INFO [finetune.py:976] (1/7) Epoch 2, batch 1600, loss[loss=0.2753, simple_loss=0.302, pruned_loss=0.1243, over 4277.00 frames. ], tot_loss[loss=0.2907, simple_loss=0.3308, pruned_loss=0.1253, over 956933.92 frames. ], batch size: 18, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:45:42,236 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=7333.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:45:44,167 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7336.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:45:57,980 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5265, 3.9349, 4.0285, 4.3506, 4.2491, 4.0274, 4.6277, 1.4155], + device='cuda:1'), covar=tensor([0.0720, 0.0765, 0.0725, 0.0878, 0.1172, 0.1256, 0.0568, 0.5027], + device='cuda:1'), in_proj_covar=tensor([0.0369, 0.0247, 0.0271, 0.0296, 0.0347, 0.0290, 0.0312, 0.0304], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:46:18,228 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0555, 1.2610, 1.0258, 1.7395, 2.2962, 1.3742, 1.4992, 1.6918], + device='cuda:1'), covar=tensor([0.1563, 0.2315, 0.2170, 0.1296, 0.1903, 0.1894, 0.1465, 0.2124], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0097, 0.0115, 0.0092, 0.0124, 0.0096, 0.0098, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 23:46:25,245 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5499, 1.3066, 1.3250, 1.5526, 1.8530, 1.4992, 0.9141, 1.2884], + device='cuda:1'), covar=tensor([0.2904, 0.2805, 0.2354, 0.2137, 0.2312, 0.1588, 0.3737, 0.2185], + device='cuda:1'), in_proj_covar=tensor([0.0219, 0.0201, 0.0187, 0.0173, 0.0222, 0.0168, 0.0203, 0.0176], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:46:35,385 INFO [finetune.py:976] (1/7) Epoch 2, batch 1650, loss[loss=0.2609, simple_loss=0.3103, pruned_loss=0.1057, over 4922.00 frames. ], tot_loss[loss=0.2883, simple_loss=0.3277, pruned_loss=0.1244, over 957856.54 frames. ], batch size: 43, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:46:47,892 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4132, 0.8877, 1.2167, 1.1044, 1.0130, 1.0494, 1.0221, 1.1428], + device='cuda:1'), covar=tensor([2.1909, 4.3344, 3.1000, 3.5400, 3.7739, 2.7145, 4.8125, 2.7604], + device='cuda:1'), in_proj_covar=tensor([0.0220, 0.0254, 0.0240, 0.0266, 0.0244, 0.0216, 0.0275, 0.0211], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:46:59,569 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7404.0, num_to_drop=1, layers_to_drop={1} +2023-03-25 23:47:01,419 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7407.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:47:07,155 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.344e+02 1.965e+02 2.306e+02 2.769e+02 7.650e+02, threshold=4.611e+02, percent-clipped=3.0 +2023-03-25 23:47:11,628 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-03-25 23:47:14,657 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6386, 1.2097, 1.3025, 1.1644, 1.7049, 1.8259, 1.5906, 1.1450], + device='cuda:1'), covar=tensor([0.0238, 0.0487, 0.0533, 0.0467, 0.0253, 0.0298, 0.0244, 0.0470], + device='cuda:1'), in_proj_covar=tensor([0.0082, 0.0112, 0.0131, 0.0112, 0.0103, 0.0097, 0.0087, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.4109e-05, 8.8377e-05, 1.0607e-04, 8.8759e-05, 8.2019e-05, 7.2871e-05, + 6.7484e-05, 8.4607e-05], device='cuda:1') +2023-03-25 23:47:15,137 INFO [finetune.py:976] (1/7) Epoch 2, batch 1700, loss[loss=0.3206, simple_loss=0.3444, pruned_loss=0.1484, over 4755.00 frames. ], tot_loss[loss=0.2852, simple_loss=0.3241, pruned_loss=0.1232, over 954579.98 frames. ], batch size: 27, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:47:34,986 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-03-25 23:47:48,741 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7468.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:48:00,184 INFO [finetune.py:976] (1/7) Epoch 2, batch 1750, loss[loss=0.3421, simple_loss=0.3726, pruned_loss=0.1558, over 4809.00 frames. ], tot_loss[loss=0.2874, simple_loss=0.3264, pruned_loss=0.1242, over 953327.55 frames. ], batch size: 45, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:48:35,399 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1535, 1.8546, 1.6678, 0.7241, 1.8002, 1.7359, 1.4104, 1.8489], + device='cuda:1'), covar=tensor([0.0905, 0.0930, 0.1719, 0.2323, 0.1357, 0.2215, 0.2371, 0.1051], + device='cuda:1'), in_proj_covar=tensor([0.0162, 0.0185, 0.0198, 0.0181, 0.0206, 0.0204, 0.0208, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:48:46,572 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.280e+02 2.019e+02 2.351e+02 2.794e+02 5.482e+02, threshold=4.701e+02, percent-clipped=1.0 +2023-03-25 23:48:53,181 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7519.0, num_to_drop=1, layers_to_drop={3} +2023-03-25 23:49:03,495 INFO [finetune.py:976] (1/7) Epoch 2, batch 1800, loss[loss=0.2625, simple_loss=0.3074, pruned_loss=0.1088, over 4767.00 frames. ], tot_loss[loss=0.288, simple_loss=0.3282, pruned_loss=0.1239, over 953022.48 frames. ], batch size: 28, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:49:12,358 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7534.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 23:49:44,258 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7565.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:49:57,758 INFO [finetune.py:976] (1/7) Epoch 2, batch 1850, loss[loss=0.262, simple_loss=0.3156, pruned_loss=0.1042, over 4921.00 frames. ], tot_loss[loss=0.2909, simple_loss=0.3312, pruned_loss=0.1253, over 952969.69 frames. ], batch size: 33, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:50:05,101 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=7582.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 23:50:16,302 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1709, 0.4333, 1.1144, 0.9170, 0.9310, 0.8898, 0.7754, 1.0634], + device='cuda:1'), covar=tensor([1.8433, 3.7605, 2.6185, 2.9739, 3.2020, 2.2182, 4.0720, 2.4058], + device='cuda:1'), in_proj_covar=tensor([0.0220, 0.0254, 0.0241, 0.0266, 0.0243, 0.0217, 0.0275, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:50:41,637 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=7613.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:50:48,710 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.368e+02 2.140e+02 2.552e+02 3.133e+02 4.516e+02, threshold=5.105e+02, percent-clipped=0.0 +2023-03-25 23:51:00,593 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7624.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:51:02,925 INFO [finetune.py:976] (1/7) Epoch 2, batch 1900, loss[loss=0.3547, simple_loss=0.387, pruned_loss=0.1611, over 4899.00 frames. ], tot_loss[loss=0.2908, simple_loss=0.3318, pruned_loss=0.1249, over 951495.80 frames. ], batch size: 37, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:51:09,863 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7631.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:51:41,823 INFO [finetune.py:976] (1/7) Epoch 2, batch 1950, loss[loss=0.311, simple_loss=0.3299, pruned_loss=0.146, over 4835.00 frames. ], tot_loss[loss=0.289, simple_loss=0.3304, pruned_loss=0.1238, over 954095.79 frames. ], batch size: 47, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:51:46,725 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-25 23:51:47,808 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7685.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:51:59,344 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7704.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 23:52:07,434 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.355e+02 1.951e+02 2.306e+02 2.786e+02 5.176e+02, threshold=4.611e+02, percent-clipped=1.0 +2023-03-25 23:52:11,057 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4314, 0.7125, 1.3275, 1.0615, 1.1212, 1.0918, 0.9291, 1.2516], + device='cuda:1'), covar=tensor([1.7305, 3.3672, 2.4154, 2.7930, 3.0183, 2.0593, 3.8523, 2.1136], + device='cuda:1'), in_proj_covar=tensor([0.0221, 0.0255, 0.0241, 0.0267, 0.0244, 0.0217, 0.0276, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:52:17,332 INFO [finetune.py:976] (1/7) Epoch 2, batch 2000, loss[loss=0.2752, simple_loss=0.3055, pruned_loss=0.1224, over 4774.00 frames. ], tot_loss[loss=0.2866, simple_loss=0.3275, pruned_loss=0.1228, over 953340.39 frames. ], batch size: 28, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:52:26,495 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7743.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:52:31,888 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=7752.0, num_to_drop=1, layers_to_drop={0} +2023-03-25 23:52:39,936 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7763.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:52:45,283 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6394, 1.6500, 2.0872, 3.3166, 2.3891, 2.3596, 1.1437, 2.5838], + device='cuda:1'), covar=tensor([0.1827, 0.1449, 0.1298, 0.0528, 0.0762, 0.1302, 0.1816, 0.0650], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0120, 0.0140, 0.0163, 0.0105, 0.0147, 0.0131, 0.0109], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-25 23:52:48,821 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4350, 3.8063, 3.9955, 4.3068, 4.1498, 3.9672, 4.5101, 1.5638], + device='cuda:1'), covar=tensor([0.0673, 0.0780, 0.0751, 0.0808, 0.1066, 0.1136, 0.0655, 0.4824], + device='cuda:1'), in_proj_covar=tensor([0.0370, 0.0248, 0.0273, 0.0297, 0.0347, 0.0289, 0.0315, 0.0305], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:52:49,965 INFO [finetune.py:976] (1/7) Epoch 2, batch 2050, loss[loss=0.2465, simple_loss=0.3019, pruned_loss=0.09556, over 4774.00 frames. ], tot_loss[loss=0.2829, simple_loss=0.3239, pruned_loss=0.1209, over 956019.02 frames. ], batch size: 28, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:53:05,252 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7802.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:53:06,504 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7804.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:53:14,628 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.310e+02 1.893e+02 2.255e+02 2.795e+02 6.508e+02, threshold=4.510e+02, percent-clipped=3.0 +2023-03-25 23:53:16,598 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5134, 0.8857, 1.2721, 1.1514, 1.0847, 1.1278, 1.0937, 1.2275], + device='cuda:1'), covar=tensor([1.9517, 3.5586, 2.7201, 3.0912, 3.3703, 2.2865, 4.0809, 2.3074], + device='cuda:1'), in_proj_covar=tensor([0.0221, 0.0255, 0.0242, 0.0268, 0.0245, 0.0218, 0.0277, 0.0213], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-25 23:53:17,176 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7819.0, num_to_drop=1, layers_to_drop={2} +2023-03-25 23:53:26,212 INFO [finetune.py:976] (1/7) Epoch 2, batch 2100, loss[loss=0.2634, simple_loss=0.2991, pruned_loss=0.1138, over 4828.00 frames. ], tot_loss[loss=0.283, simple_loss=0.323, pruned_loss=0.1215, over 955366.85 frames. ], batch size: 25, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:53:56,798 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7863.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:54:00,158 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=7867.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:54:08,357 INFO [finetune.py:976] (1/7) Epoch 2, batch 2150, loss[loss=0.2803, simple_loss=0.3382, pruned_loss=0.1111, over 4837.00 frames. ], tot_loss[loss=0.2852, simple_loss=0.326, pruned_loss=0.1222, over 953036.35 frames. ], batch size: 47, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:54:49,702 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.034e+02 1.983e+02 2.392e+02 2.856e+02 4.131e+02, threshold=4.785e+02, percent-clipped=0.0 +2023-03-25 23:55:10,156 INFO [finetune.py:976] (1/7) Epoch 2, batch 2200, loss[loss=0.249, simple_loss=0.2896, pruned_loss=0.1042, over 4777.00 frames. ], tot_loss[loss=0.2873, simple_loss=0.3284, pruned_loss=0.1231, over 952634.62 frames. ], batch size: 26, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:55:16,955 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7931.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:55:27,718 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-25 23:56:09,430 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3719, 1.2808, 1.3273, 1.3069, 0.7310, 2.1790, 0.6589, 1.2360], + device='cuda:1'), covar=tensor([0.3583, 0.2572, 0.2382, 0.2496, 0.2406, 0.0406, 0.3356, 0.1703], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0108, 0.0114, 0.0115, 0.0111, 0.0094, 0.0098, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-25 23:56:10,025 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9230, 4.2562, 4.4020, 4.7789, 4.5665, 4.3769, 5.0304, 1.6470], + device='cuda:1'), covar=tensor([0.0652, 0.0776, 0.0620, 0.0743, 0.1283, 0.1337, 0.0473, 0.5304], + device='cuda:1'), in_proj_covar=tensor([0.0375, 0.0251, 0.0278, 0.0301, 0.0352, 0.0294, 0.0318, 0.0310], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-25 23:56:12,554 INFO [finetune.py:976] (1/7) Epoch 2, batch 2250, loss[loss=0.3022, simple_loss=0.3421, pruned_loss=0.1312, over 4268.00 frames. ], tot_loss[loss=0.2891, simple_loss=0.3302, pruned_loss=0.124, over 949815.76 frames. ], batch size: 65, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:56:13,681 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=7979.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:56:14,295 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7980.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:57:05,129 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.321e+02 1.899e+02 2.269e+02 2.653e+02 4.132e+02, threshold=4.538e+02, percent-clipped=0.0 +2023-03-25 23:57:24,777 INFO [finetune.py:976] (1/7) Epoch 2, batch 2300, loss[loss=0.3088, simple_loss=0.3369, pruned_loss=0.1404, over 4198.00 frames. ], tot_loss[loss=0.288, simple_loss=0.33, pruned_loss=0.123, over 948298.35 frames. ], batch size: 65, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:57:35,942 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.73 vs. limit=5.0 +2023-03-25 23:57:53,415 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8063.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:58:05,649 INFO [finetune.py:976] (1/7) Epoch 2, batch 2350, loss[loss=0.2688, simple_loss=0.3094, pruned_loss=0.1142, over 4846.00 frames. ], tot_loss[loss=0.2865, simple_loss=0.3284, pruned_loss=0.1224, over 952191.20 frames. ], batch size: 49, lr: 3.99e-03, grad_scale: 16.0 +2023-03-25 23:58:20,621 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=8099.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:58:39,036 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=8111.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:58:40,301 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8227, 1.1696, 1.0319, 1.6749, 2.1772, 1.3202, 1.4190, 1.7093], + device='cuda:1'), covar=tensor([0.1699, 0.2391, 0.2225, 0.1406, 0.2115, 0.2117, 0.1541, 0.2175], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0098, 0.0117, 0.0093, 0.0125, 0.0097, 0.0099, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-25 23:58:41,409 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.369e+02 1.949e+02 2.211e+02 2.649e+02 5.737e+02, threshold=4.422e+02, percent-clipped=2.0 +2023-03-25 23:59:01,109 INFO [finetune.py:976] (1/7) Epoch 2, batch 2400, loss[loss=0.2515, simple_loss=0.3031, pruned_loss=0.09997, over 4866.00 frames. ], tot_loss[loss=0.2801, simple_loss=0.3218, pruned_loss=0.1192, over 950694.76 frames. ], batch size: 31, lr: 3.99e-03, grad_scale: 32.0 +2023-03-25 23:59:27,548 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=8158.0, num_to_drop=0, layers_to_drop=set() +2023-03-25 23:59:39,736 INFO [finetune.py:976] (1/7) Epoch 2, batch 2450, loss[loss=0.2471, simple_loss=0.2928, pruned_loss=0.1007, over 4814.00 frames. ], tot_loss[loss=0.2747, simple_loss=0.3168, pruned_loss=0.1163, over 951223.33 frames. ], batch size: 51, lr: 3.99e-03, grad_scale: 32.0 +2023-03-25 23:59:48,625 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7222, 1.0589, 0.9755, 1.6148, 2.0548, 1.3227, 1.3403, 1.6747], + device='cuda:1'), covar=tensor([0.1739, 0.2421, 0.2243, 0.1348, 0.2189, 0.2077, 0.1518, 0.1967], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0098, 0.0116, 0.0093, 0.0125, 0.0097, 0.0098, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 00:00:02,105 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 00:00:20,951 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.358e+02 1.915e+02 2.165e+02 2.652e+02 4.234e+02, threshold=4.330e+02, percent-clipped=0.0 +2023-03-26 00:00:32,954 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6342, 1.4645, 1.3001, 1.3447, 1.7830, 1.9832, 1.6470, 1.2930], + device='cuda:1'), covar=tensor([0.0269, 0.0438, 0.0563, 0.0429, 0.0273, 0.0306, 0.0308, 0.0437], + device='cuda:1'), in_proj_covar=tensor([0.0082, 0.0113, 0.0132, 0.0112, 0.0103, 0.0098, 0.0088, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.4191e-05, 8.9182e-05, 1.0705e-04, 8.9212e-05, 8.1815e-05, 7.3296e-05, + 6.7964e-05, 8.5363e-05], device='cuda:1') +2023-03-26 00:00:34,016 INFO [finetune.py:976] (1/7) Epoch 2, batch 2500, loss[loss=0.2902, simple_loss=0.3499, pruned_loss=0.1152, over 4896.00 frames. ], tot_loss[loss=0.2785, simple_loss=0.3198, pruned_loss=0.1186, over 952774.81 frames. ], batch size: 43, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:01:11,681 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8265.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:01:30,123 INFO [finetune.py:976] (1/7) Epoch 2, batch 2550, loss[loss=0.294, simple_loss=0.3089, pruned_loss=0.1395, over 4104.00 frames. ], tot_loss[loss=0.281, simple_loss=0.3232, pruned_loss=0.1194, over 951446.64 frames. ], batch size: 17, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:01:31,486 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8280.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:01:34,513 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1263, 2.8281, 1.8754, 1.5592, 3.0820, 2.7134, 2.1381, 2.3910], + device='cuda:1'), covar=tensor([0.0986, 0.0567, 0.1187, 0.1292, 0.0344, 0.0888, 0.1092, 0.1038], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0131, 0.0139, 0.0128, 0.0107, 0.0137, 0.0144, 0.0159], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:01:42,294 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6116, 1.7556, 1.8465, 1.9626, 1.9950, 4.1940, 1.6285, 2.0421], + device='cuda:1'), covar=tensor([0.0992, 0.1547, 0.1255, 0.1017, 0.1378, 0.0198, 0.1322, 0.1609], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0079, 0.0075, 0.0078, 0.0090, 0.0080, 0.0083, 0.0077], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 00:02:02,894 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.611e+02 2.134e+02 2.551e+02 3.097e+02 6.135e+02, threshold=5.102e+02, percent-clipped=4.0 +2023-03-26 00:02:09,731 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8326.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:02:10,867 INFO [finetune.py:976] (1/7) Epoch 2, batch 2600, loss[loss=0.3142, simple_loss=0.3611, pruned_loss=0.1336, over 4827.00 frames. ], tot_loss[loss=0.2835, simple_loss=0.3259, pruned_loss=0.1206, over 950605.13 frames. ], batch size: 39, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:02:10,926 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=8328.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:02:50,709 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1727, 1.2933, 1.4933, 0.6925, 1.1605, 1.5822, 1.6027, 1.4095], + device='cuda:1'), covar=tensor([0.0825, 0.0547, 0.0357, 0.0522, 0.0395, 0.0485, 0.0272, 0.0444], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0150, 0.0113, 0.0128, 0.0127, 0.0114, 0.0140, 0.0140], + device='cuda:1'), out_proj_covar=tensor([9.3721e-05, 1.1186e-04, 8.3151e-05, 9.3884e-05, 9.2029e-05, 8.4207e-05, + 1.0483e-04, 1.0352e-04], device='cuda:1') +2023-03-26 00:03:09,437 INFO [finetune.py:976] (1/7) Epoch 2, batch 2650, loss[loss=0.2748, simple_loss=0.3276, pruned_loss=0.111, over 4838.00 frames. ], tot_loss[loss=0.2848, simple_loss=0.3277, pruned_loss=0.1209, over 953198.65 frames. ], batch size: 49, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:03:28,507 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8399.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:03:45,680 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.253e+02 1.975e+02 2.270e+02 2.796e+02 5.066e+02, threshold=4.539e+02, percent-clipped=0.0 +2023-03-26 00:03:57,447 INFO [finetune.py:976] (1/7) Epoch 2, batch 2700, loss[loss=0.2572, simple_loss=0.3011, pruned_loss=0.1066, over 4175.00 frames. ], tot_loss[loss=0.282, simple_loss=0.3254, pruned_loss=0.1193, over 952028.81 frames. ], batch size: 65, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:04:13,877 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6353, 1.5234, 1.4715, 1.5882, 0.9888, 3.2419, 1.1937, 1.6637], + device='cuda:1'), covar=tensor([0.3621, 0.2621, 0.2257, 0.2377, 0.2241, 0.0205, 0.2804, 0.1601], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0107, 0.0113, 0.0115, 0.0111, 0.0093, 0.0097, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 00:04:17,923 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=8447.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:04:28,885 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8458.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:04:31,118 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1384, 0.8231, 0.9495, 1.0136, 1.2073, 1.2495, 1.0834, 0.9531], + device='cuda:1'), covar=tensor([0.0296, 0.0423, 0.0592, 0.0353, 0.0330, 0.0346, 0.0294, 0.0464], + device='cuda:1'), in_proj_covar=tensor([0.0082, 0.0113, 0.0133, 0.0113, 0.0104, 0.0098, 0.0088, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.4286e-05, 8.9359e-05, 1.0722e-04, 8.9406e-05, 8.2251e-05, 7.2988e-05, + 6.8055e-05, 8.5395e-05], device='cuda:1') +2023-03-26 00:04:47,924 INFO [finetune.py:976] (1/7) Epoch 2, batch 2750, loss[loss=0.2708, simple_loss=0.3089, pruned_loss=0.1164, over 4874.00 frames. ], tot_loss[loss=0.2795, simple_loss=0.322, pruned_loss=0.1184, over 951723.57 frames. ], batch size: 34, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:05:09,621 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=8506.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:05:25,225 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.170e+02 1.975e+02 2.311e+02 2.901e+02 5.278e+02, threshold=4.621e+02, percent-clipped=1.0 +2023-03-26 00:05:38,381 INFO [finetune.py:976] (1/7) Epoch 2, batch 2800, loss[loss=0.2347, simple_loss=0.2876, pruned_loss=0.09087, over 4826.00 frames. ], tot_loss[loss=0.2752, simple_loss=0.3178, pruned_loss=0.1163, over 951275.26 frames. ], batch size: 41, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:06:44,211 INFO [finetune.py:976] (1/7) Epoch 2, batch 2850, loss[loss=0.2857, simple_loss=0.3272, pruned_loss=0.1221, over 4912.00 frames. ], tot_loss[loss=0.2735, simple_loss=0.3159, pruned_loss=0.1155, over 952259.09 frames. ], batch size: 37, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:07:08,736 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.223e+02 1.882e+02 2.289e+02 2.777e+02 4.779e+02, threshold=4.578e+02, percent-clipped=1.0 +2023-03-26 00:07:10,130 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 00:07:13,354 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=8621.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:07:14,480 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6891, 1.3909, 1.9551, 1.2228, 1.6622, 1.6555, 1.1909, 1.8325], + device='cuda:1'), covar=tensor([0.1963, 0.2391, 0.1611, 0.2249, 0.1476, 0.1972, 0.3239, 0.1499], + device='cuda:1'), in_proj_covar=tensor([0.0205, 0.0207, 0.0205, 0.0195, 0.0178, 0.0223, 0.0214, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:07:15,123 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5826, 1.5366, 1.4695, 1.0415, 1.4684, 1.7792, 1.6588, 1.4353], + device='cuda:1'), covar=tensor([0.0889, 0.0539, 0.0451, 0.0618, 0.0427, 0.0408, 0.0350, 0.0627], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0149, 0.0114, 0.0127, 0.0126, 0.0114, 0.0140, 0.0139], + device='cuda:1'), out_proj_covar=tensor([9.3576e-05, 1.1141e-04, 8.3275e-05, 9.3542e-05, 9.1014e-05, 8.3809e-05, + 1.0493e-04, 1.0313e-04], device='cuda:1') +2023-03-26 00:07:18,068 INFO [finetune.py:976] (1/7) Epoch 2, batch 2900, loss[loss=0.2466, simple_loss=0.2796, pruned_loss=0.1069, over 3853.00 frames. ], tot_loss[loss=0.2758, simple_loss=0.3185, pruned_loss=0.1165, over 952256.87 frames. ], batch size: 16, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:07:50,889 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 00:07:51,241 INFO [finetune.py:976] (1/7) Epoch 2, batch 2950, loss[loss=0.3228, simple_loss=0.3727, pruned_loss=0.1364, over 4813.00 frames. ], tot_loss[loss=0.2811, simple_loss=0.3242, pruned_loss=0.119, over 953986.90 frames. ], batch size: 40, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:08:08,690 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.1902, 4.6595, 4.9197, 4.8597, 4.6198, 4.3824, 5.3218, 1.7957], + device='cuda:1'), covar=tensor([0.0869, 0.1144, 0.0830, 0.1345, 0.1907, 0.1869, 0.0749, 0.6161], + device='cuda:1'), in_proj_covar=tensor([0.0373, 0.0248, 0.0276, 0.0300, 0.0349, 0.0291, 0.0316, 0.0306], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:08:15,893 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-26 00:08:17,805 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.163e+02 1.871e+02 2.334e+02 2.881e+02 5.890e+02, threshold=4.669e+02, percent-clipped=2.0 +2023-03-26 00:08:27,818 INFO [finetune.py:976] (1/7) Epoch 2, batch 3000, loss[loss=0.3094, simple_loss=0.3364, pruned_loss=0.1412, over 4163.00 frames. ], tot_loss[loss=0.2814, simple_loss=0.3247, pruned_loss=0.1191, over 953409.47 frames. ], batch size: 65, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:08:27,818 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 00:08:31,341 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8019, 3.3771, 3.4439, 3.6755, 3.4779, 3.3387, 3.8860, 1.3630], + device='cuda:1'), covar=tensor([0.0945, 0.0879, 0.0976, 0.1156, 0.1613, 0.1571, 0.0757, 0.5266], + device='cuda:1'), in_proj_covar=tensor([0.0372, 0.0248, 0.0276, 0.0300, 0.0349, 0.0291, 0.0316, 0.0306], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:08:32,118 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7553, 1.5818, 2.0829, 2.8900, 2.1360, 2.2145, 1.1239, 2.3936], + device='cuda:1'), covar=tensor([0.1759, 0.1558, 0.1150, 0.0670, 0.0802, 0.1339, 0.1705, 0.0694], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0120, 0.0139, 0.0163, 0.0105, 0.0145, 0.0132, 0.0108], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 00:08:43,567 INFO [finetune.py:1010] (1/7) Epoch 2, validation: loss=0.1956, simple_loss=0.2636, pruned_loss=0.06384, over 2265189.00 frames. +2023-03-26 00:08:43,568 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6048MB +2023-03-26 00:08:44,826 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5612, 1.6429, 1.9033, 1.2960, 1.5859, 1.7461, 1.5997, 1.9274], + device='cuda:1'), covar=tensor([0.1635, 0.1996, 0.1402, 0.1793, 0.1122, 0.1543, 0.2466, 0.0984], + device='cuda:1'), in_proj_covar=tensor([0.0204, 0.0206, 0.0204, 0.0194, 0.0177, 0.0223, 0.0212, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:09:09,175 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8760.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:09:26,209 INFO [finetune.py:976] (1/7) Epoch 2, batch 3050, loss[loss=0.2645, simple_loss=0.3193, pruned_loss=0.1048, over 4825.00 frames. ], tot_loss[loss=0.2825, simple_loss=0.3259, pruned_loss=0.1196, over 951607.20 frames. ], batch size: 39, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:10:07,396 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.149e+02 1.935e+02 2.296e+02 2.584e+02 4.666e+02, threshold=4.592e+02, percent-clipped=0.0 +2023-03-26 00:10:11,732 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8821.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:10:16,913 INFO [finetune.py:976] (1/7) Epoch 2, batch 3100, loss[loss=0.2543, simple_loss=0.3039, pruned_loss=0.1023, over 4771.00 frames. ], tot_loss[loss=0.2791, simple_loss=0.3231, pruned_loss=0.1175, over 953597.02 frames. ], batch size: 29, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:10:22,848 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2041, 2.0452, 2.3402, 1.0381, 2.3742, 2.7940, 2.0725, 2.2051], + device='cuda:1'), covar=tensor([0.1030, 0.0866, 0.0567, 0.0905, 0.0635, 0.0478, 0.0594, 0.0550], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0151, 0.0114, 0.0129, 0.0126, 0.0115, 0.0141, 0.0140], + device='cuda:1'), out_proj_covar=tensor([9.4752e-05, 1.1237e-04, 8.3958e-05, 9.4474e-05, 9.1491e-05, 8.4566e-05, + 1.0574e-04, 1.0378e-04], device='cuda:1') +2023-03-26 00:10:22,870 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9622, 1.7719, 1.3627, 1.9112, 1.7062, 1.5819, 1.6096, 2.8746], + device='cuda:1'), covar=tensor([2.1080, 2.1943, 1.7232, 2.6680, 1.8378, 1.1909, 2.2742, 0.5075], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0230, 0.0209, 0.0266, 0.0222, 0.0188, 0.0227, 0.0170], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 00:10:59,133 INFO [finetune.py:976] (1/7) Epoch 2, batch 3150, loss[loss=0.2579, simple_loss=0.3101, pruned_loss=0.1028, over 4797.00 frames. ], tot_loss[loss=0.2768, simple_loss=0.32, pruned_loss=0.1169, over 954678.78 frames. ], batch size: 29, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:11:10,846 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6563, 3.3703, 3.2339, 1.4705, 3.4723, 2.6088, 0.7601, 2.3156], + device='cuda:1'), covar=tensor([0.2422, 0.1825, 0.1712, 0.3795, 0.1138, 0.1036, 0.5007, 0.1698], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0166, 0.0166, 0.0129, 0.0156, 0.0120, 0.0148, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 00:11:28,137 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.188e+02 1.838e+02 2.165e+02 2.835e+02 5.909e+02, threshold=4.329e+02, percent-clipped=1.0 +2023-03-26 00:11:33,407 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8921.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:11:43,256 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8927.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:11:43,733 INFO [finetune.py:976] (1/7) Epoch 2, batch 3200, loss[loss=0.3255, simple_loss=0.3465, pruned_loss=0.1523, over 4842.00 frames. ], tot_loss[loss=0.2729, simple_loss=0.3159, pruned_loss=0.1149, over 955371.89 frames. ], batch size: 47, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:12:25,338 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=8969.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:12:30,758 INFO [finetune.py:976] (1/7) Epoch 2, batch 3250, loss[loss=0.4236, simple_loss=0.4264, pruned_loss=0.2104, over 4828.00 frames. ], tot_loss[loss=0.272, simple_loss=0.3149, pruned_loss=0.1146, over 954307.88 frames. ], batch size: 40, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:12:36,281 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8986.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:12:37,509 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8988.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:12:37,539 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7159, 1.6316, 1.2895, 1.8299, 1.6955, 1.3801, 2.2156, 1.6393], + device='cuda:1'), covar=tensor([0.2941, 0.5849, 0.5925, 0.5336, 0.4252, 0.3091, 0.5118, 0.3910], + device='cuda:1'), in_proj_covar=tensor([0.0162, 0.0194, 0.0237, 0.0251, 0.0212, 0.0181, 0.0202, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:13:00,582 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9013.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:13:03,774 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.184e+02 1.894e+02 2.281e+02 2.922e+02 4.541e+02, threshold=4.561e+02, percent-clipped=3.0 +2023-03-26 00:13:06,353 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2996, 1.1438, 1.4978, 2.3553, 1.6119, 1.9765, 0.8126, 1.8788], + device='cuda:1'), covar=tensor([0.1944, 0.1665, 0.1256, 0.0693, 0.0953, 0.1418, 0.1736, 0.0883], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0120, 0.0139, 0.0162, 0.0104, 0.0145, 0.0132, 0.0108], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 00:13:11,706 INFO [finetune.py:976] (1/7) Epoch 2, batch 3300, loss[loss=0.2635, simple_loss=0.312, pruned_loss=0.1075, over 4815.00 frames. ], tot_loss[loss=0.2769, simple_loss=0.3204, pruned_loss=0.1167, over 955120.90 frames. ], batch size: 51, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:13:25,339 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9047.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:13:32,405 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.32 vs. limit=5.0 +2023-03-26 00:13:42,689 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9074.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:13:44,937 INFO [finetune.py:976] (1/7) Epoch 2, batch 3350, loss[loss=0.3066, simple_loss=0.3441, pruned_loss=0.1345, over 4890.00 frames. ], tot_loss[loss=0.2789, simple_loss=0.3229, pruned_loss=0.1174, over 956326.66 frames. ], batch size: 43, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:14:19,705 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7866, 1.8577, 1.6817, 1.8494, 1.1682, 4.2999, 1.5191, 2.2239], + device='cuda:1'), covar=tensor([0.3280, 0.2187, 0.1992, 0.2125, 0.1898, 0.0096, 0.2645, 0.1474], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0108, 0.0113, 0.0115, 0.0111, 0.0094, 0.0098, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 00:14:20,190 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.430e+02 1.942e+02 2.307e+02 2.996e+02 6.023e+02, threshold=4.614e+02, percent-clipped=2.0 +2023-03-26 00:14:20,871 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9116.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:14:28,092 INFO [finetune.py:976] (1/7) Epoch 2, batch 3400, loss[loss=0.305, simple_loss=0.3428, pruned_loss=0.1336, over 4738.00 frames. ], tot_loss[loss=0.2803, simple_loss=0.3247, pruned_loss=0.118, over 955897.36 frames. ], batch size: 54, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:15:12,814 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.68 vs. limit=5.0 +2023-03-26 00:15:23,464 INFO [finetune.py:976] (1/7) Epoch 2, batch 3450, loss[loss=0.2222, simple_loss=0.2758, pruned_loss=0.08428, over 4757.00 frames. ], tot_loss[loss=0.2789, simple_loss=0.3238, pruned_loss=0.117, over 955980.04 frames. ], batch size: 26, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:15:42,581 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5823, 1.6618, 1.8725, 1.9087, 1.7672, 3.1659, 1.4287, 1.7331], + device='cuda:1'), covar=tensor([0.0965, 0.1526, 0.1409, 0.0941, 0.1423, 0.0291, 0.1344, 0.1580], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0080, 0.0077, 0.0079, 0.0092, 0.0082, 0.0085, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 00:15:59,479 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.266e+02 1.993e+02 2.349e+02 2.850e+02 4.291e+02, threshold=4.698e+02, percent-clipped=0.0 +2023-03-26 00:16:12,592 INFO [finetune.py:976] (1/7) Epoch 2, batch 3500, loss[loss=0.2361, simple_loss=0.2837, pruned_loss=0.09421, over 4808.00 frames. ], tot_loss[loss=0.276, simple_loss=0.3203, pruned_loss=0.1159, over 954841.62 frames. ], batch size: 25, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:17:13,574 INFO [finetune.py:976] (1/7) Epoch 2, batch 3550, loss[loss=0.2501, simple_loss=0.2852, pruned_loss=0.1075, over 4769.00 frames. ], tot_loss[loss=0.2718, simple_loss=0.3158, pruned_loss=0.1139, over 954683.12 frames. ], batch size: 26, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:17:16,753 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9283.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:17:53,395 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.215e+02 1.769e+02 2.188e+02 2.766e+02 5.069e+02, threshold=4.376e+02, percent-clipped=2.0 +2023-03-26 00:18:09,363 INFO [finetune.py:976] (1/7) Epoch 2, batch 3600, loss[loss=0.2664, simple_loss=0.3106, pruned_loss=0.1111, over 4871.00 frames. ], tot_loss[loss=0.2683, simple_loss=0.3124, pruned_loss=0.1121, over 956074.54 frames. ], batch size: 34, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:18:22,988 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9342.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:18:45,311 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9369.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:18:51,741 INFO [finetune.py:976] (1/7) Epoch 2, batch 3650, loss[loss=0.2862, simple_loss=0.3245, pruned_loss=0.1239, over 4820.00 frames. ], tot_loss[loss=0.2709, simple_loss=0.315, pruned_loss=0.1134, over 954798.46 frames. ], batch size: 51, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:19:24,496 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9414.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:19:24,979 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.270e+02 1.902e+02 2.269e+02 2.850e+02 5.426e+02, threshold=4.539e+02, percent-clipped=4.0 +2023-03-26 00:19:31,401 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9416.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:19:45,282 INFO [finetune.py:976] (1/7) Epoch 2, batch 3700, loss[loss=0.2576, simple_loss=0.3138, pruned_loss=0.1007, over 4828.00 frames. ], tot_loss[loss=0.2741, simple_loss=0.3187, pruned_loss=0.1147, over 953878.07 frames. ], batch size: 30, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:20:15,849 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=9464.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:20:23,946 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9475.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:20:26,040 INFO [finetune.py:976] (1/7) Epoch 2, batch 3750, loss[loss=0.3133, simple_loss=0.3523, pruned_loss=0.1371, over 4811.00 frames. ], tot_loss[loss=0.2771, simple_loss=0.3221, pruned_loss=0.1161, over 956385.78 frames. ], batch size: 33, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:20:55,390 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.357e+02 1.893e+02 2.405e+02 2.686e+02 6.929e+02, threshold=4.810e+02, percent-clipped=1.0 +2023-03-26 00:21:10,674 INFO [finetune.py:976] (1/7) Epoch 2, batch 3800, loss[loss=0.249, simple_loss=0.2983, pruned_loss=0.09991, over 4901.00 frames. ], tot_loss[loss=0.2761, simple_loss=0.3216, pruned_loss=0.1154, over 956092.72 frames. ], batch size: 36, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:21:28,887 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6894, 1.6109, 1.4012, 1.3164, 1.9019, 2.0307, 1.6814, 1.3547], + device='cuda:1'), covar=tensor([0.0286, 0.0355, 0.0552, 0.0405, 0.0222, 0.0314, 0.0250, 0.0424], + device='cuda:1'), in_proj_covar=tensor([0.0082, 0.0112, 0.0132, 0.0112, 0.0102, 0.0097, 0.0086, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.3890e-05, 8.8572e-05, 1.0688e-04, 8.8619e-05, 8.1329e-05, 7.2202e-05, + 6.6605e-05, 8.4304e-05], device='cuda:1') +2023-03-26 00:21:53,455 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1787, 3.6530, 3.7848, 3.9842, 3.8887, 3.7470, 4.2778, 1.3988], + device='cuda:1'), covar=tensor([0.0816, 0.0961, 0.0800, 0.0955, 0.1391, 0.1351, 0.0789, 0.5085], + device='cuda:1'), in_proj_covar=tensor([0.0369, 0.0246, 0.0274, 0.0297, 0.0343, 0.0288, 0.0313, 0.0301], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:22:03,070 INFO [finetune.py:976] (1/7) Epoch 2, batch 3850, loss[loss=0.2378, simple_loss=0.2875, pruned_loss=0.09403, over 4930.00 frames. ], tot_loss[loss=0.274, simple_loss=0.3193, pruned_loss=0.1143, over 955681.97 frames. ], batch size: 33, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:22:07,222 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9583.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:22:07,842 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9584.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:22:08,543 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 00:22:12,135 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2491, 2.8580, 1.9704, 1.5822, 3.1190, 2.8186, 2.4785, 2.1851], + device='cuda:1'), covar=tensor([0.0877, 0.0545, 0.1066, 0.1224, 0.0331, 0.0776, 0.0844, 0.1146], + device='cuda:1'), in_proj_covar=tensor([0.0137, 0.0131, 0.0142, 0.0129, 0.0108, 0.0139, 0.0145, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:22:33,608 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.288e+02 1.890e+02 2.331e+02 2.867e+02 5.576e+02, threshold=4.662e+02, percent-clipped=3.0 +2023-03-26 00:22:42,843 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 00:22:48,506 INFO [finetune.py:976] (1/7) Epoch 2, batch 3900, loss[loss=0.3184, simple_loss=0.354, pruned_loss=0.1414, over 4919.00 frames. ], tot_loss[loss=0.2698, simple_loss=0.315, pruned_loss=0.1123, over 956709.25 frames. ], batch size: 38, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:22:55,986 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=9631.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:23:08,247 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9642.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:23:10,104 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9645.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:23:19,006 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7224, 1.4491, 1.0197, 0.2717, 1.2651, 1.5523, 1.4175, 1.5259], + device='cuda:1'), covar=tensor([0.0817, 0.0802, 0.1413, 0.2123, 0.1308, 0.2237, 0.2099, 0.0910], + device='cuda:1'), in_proj_covar=tensor([0.0162, 0.0189, 0.0199, 0.0183, 0.0209, 0.0204, 0.0209, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:23:28,457 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9658.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:23:29,682 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5802, 1.6089, 1.6044, 1.8320, 1.7380, 3.3639, 1.4361, 1.7840], + device='cuda:1'), covar=tensor([0.1055, 0.1667, 0.1102, 0.1022, 0.1537, 0.0273, 0.1468, 0.1624], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0080, 0.0077, 0.0079, 0.0092, 0.0082, 0.0085, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 00:23:40,519 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9669.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:23:46,463 INFO [finetune.py:976] (1/7) Epoch 2, batch 3950, loss[loss=0.2199, simple_loss=0.2731, pruned_loss=0.08332, over 4763.00 frames. ], tot_loss[loss=0.2647, simple_loss=0.3099, pruned_loss=0.1097, over 956213.89 frames. ], batch size: 28, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:24:00,115 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=9690.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:24:18,014 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3109, 2.1443, 1.8759, 2.3044, 2.4501, 2.0658, 2.7147, 2.2478], + device='cuda:1'), covar=tensor([0.2353, 0.4593, 0.5248, 0.4774, 0.2849, 0.2343, 0.3932, 0.3041], + device='cuda:1'), in_proj_covar=tensor([0.0163, 0.0196, 0.0239, 0.0253, 0.0215, 0.0182, 0.0204, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:24:28,618 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.008e+02 1.967e+02 2.302e+02 2.784e+02 7.100e+02, threshold=4.604e+02, percent-clipped=2.0 +2023-03-26 00:24:29,311 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=9717.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:24:30,626 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9719.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:24:33,555 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7381, 1.3775, 0.8793, 0.2135, 1.1130, 1.5457, 1.3367, 1.3884], + device='cuda:1'), covar=tensor([0.0759, 0.0893, 0.1291, 0.1913, 0.1348, 0.1927, 0.2037, 0.0825], + device='cuda:1'), in_proj_covar=tensor([0.0163, 0.0190, 0.0199, 0.0184, 0.0209, 0.0204, 0.0210, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:24:36,431 INFO [finetune.py:976] (1/7) Epoch 2, batch 4000, loss[loss=0.2929, simple_loss=0.3044, pruned_loss=0.1407, over 4726.00 frames. ], tot_loss[loss=0.2651, simple_loss=0.3095, pruned_loss=0.1103, over 954201.90 frames. ], batch size: 23, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:24:48,458 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9743.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:25:10,942 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9770.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:25:16,246 INFO [finetune.py:976] (1/7) Epoch 2, batch 4050, loss[loss=0.2492, simple_loss=0.2825, pruned_loss=0.1079, over 4086.00 frames. ], tot_loss[loss=0.269, simple_loss=0.3135, pruned_loss=0.1122, over 953059.80 frames. ], batch size: 17, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:25:37,104 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9804.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:25:44,425 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.258e+02 1.934e+02 2.184e+02 2.585e+02 5.351e+02, threshold=4.368e+02, percent-clipped=2.0 +2023-03-26 00:25:57,342 INFO [finetune.py:976] (1/7) Epoch 2, batch 4100, loss[loss=0.2103, simple_loss=0.2626, pruned_loss=0.07899, over 4734.00 frames. ], tot_loss[loss=0.2717, simple_loss=0.3161, pruned_loss=0.1137, over 948816.53 frames. ], batch size: 23, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:26:23,201 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.61 vs. limit=2.0 +2023-03-26 00:26:29,014 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-03-26 00:26:34,083 INFO [finetune.py:976] (1/7) Epoch 2, batch 4150, loss[loss=0.2956, simple_loss=0.3384, pruned_loss=0.1264, over 4757.00 frames. ], tot_loss[loss=0.2728, simple_loss=0.3181, pruned_loss=0.1137, over 952267.13 frames. ], batch size: 28, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:26:51,029 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8823, 1.9745, 1.8345, 1.2370, 2.2780, 2.1467, 1.9981, 1.7672], + device='cuda:1'), covar=tensor([0.0799, 0.0646, 0.0876, 0.1161, 0.0415, 0.0750, 0.0731, 0.1123], + device='cuda:1'), in_proj_covar=tensor([0.0138, 0.0131, 0.0143, 0.0129, 0.0108, 0.0140, 0.0146, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:27:03,955 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2667, 2.1306, 2.4227, 1.1513, 2.4746, 2.7693, 2.1931, 2.3072], + device='cuda:1'), covar=tensor([0.1004, 0.0868, 0.0594, 0.0827, 0.0723, 0.0821, 0.0512, 0.0636], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0154, 0.0116, 0.0132, 0.0130, 0.0117, 0.0143, 0.0143], + device='cuda:1'), out_proj_covar=tensor([9.5909e-05, 1.1453e-04, 8.4989e-05, 9.7049e-05, 9.4014e-05, 8.6014e-05, + 1.0713e-04, 1.0583e-04], device='cuda:1') +2023-03-26 00:27:05,051 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.284e+02 1.789e+02 2.151e+02 2.605e+02 5.306e+02, threshold=4.302e+02, percent-clipped=2.0 +2023-03-26 00:27:17,333 INFO [finetune.py:976] (1/7) Epoch 2, batch 4200, loss[loss=0.3206, simple_loss=0.3568, pruned_loss=0.1422, over 4820.00 frames. ], tot_loss[loss=0.2711, simple_loss=0.3176, pruned_loss=0.1123, over 955273.88 frames. ], batch size: 38, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:27:25,691 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9940.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:28:05,867 INFO [finetune.py:976] (1/7) Epoch 2, batch 4250, loss[loss=0.2302, simple_loss=0.2913, pruned_loss=0.08448, over 4890.00 frames. ], tot_loss[loss=0.2683, simple_loss=0.3151, pruned_loss=0.1108, over 956738.98 frames. ], batch size: 32, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:28:41,980 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10014.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:28:47,610 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.000e+02 1.798e+02 2.175e+02 2.768e+02 4.624e+02, threshold=4.351e+02, percent-clipped=3.0 +2023-03-26 00:29:00,124 INFO [finetune.py:976] (1/7) Epoch 2, batch 4300, loss[loss=0.242, simple_loss=0.2881, pruned_loss=0.09793, over 4735.00 frames. ], tot_loss[loss=0.2633, simple_loss=0.3102, pruned_loss=0.1082, over 958297.39 frames. ], batch size: 59, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:29:12,673 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10039.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:29:36,700 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10062.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 00:29:46,273 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10070.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:29:52,921 INFO [finetune.py:976] (1/7) Epoch 2, batch 4350, loss[loss=0.268, simple_loss=0.316, pruned_loss=0.11, over 4869.00 frames. ], tot_loss[loss=0.2596, simple_loss=0.3064, pruned_loss=0.1064, over 958347.43 frames. ], batch size: 34, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:29:58,430 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5693, 1.3629, 1.9021, 2.7950, 1.8341, 2.1561, 0.8981, 2.2950], + device='cuda:1'), covar=tensor([0.1958, 0.1885, 0.1422, 0.0929, 0.1065, 0.1529, 0.2123, 0.0834], + device='cuda:1'), in_proj_covar=tensor([0.0105, 0.0121, 0.0140, 0.0166, 0.0106, 0.0148, 0.0132, 0.0109], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 00:30:06,665 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10099.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:30:07,335 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10100.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 00:30:25,354 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.331e+02 1.930e+02 2.275e+02 2.717e+02 4.483e+02, threshold=4.550e+02, percent-clipped=1.0 +2023-03-26 00:30:27,182 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=10118.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:30:34,431 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10123.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 00:30:37,372 INFO [finetune.py:976] (1/7) Epoch 2, batch 4400, loss[loss=0.2765, simple_loss=0.3129, pruned_loss=0.12, over 4080.00 frames. ], tot_loss[loss=0.2621, simple_loss=0.3082, pruned_loss=0.108, over 955696.56 frames. ], batch size: 17, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:31:16,349 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3408, 2.8348, 2.7307, 1.3409, 2.9972, 2.1659, 0.6756, 1.7569], + device='cuda:1'), covar=tensor([0.2359, 0.2229, 0.1881, 0.3437, 0.1349, 0.1181, 0.4356, 0.1723], + device='cuda:1'), in_proj_covar=tensor([0.0158, 0.0169, 0.0169, 0.0130, 0.0158, 0.0122, 0.0149, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 00:31:35,876 INFO [finetune.py:976] (1/7) Epoch 2, batch 4450, loss[loss=0.3564, simple_loss=0.3853, pruned_loss=0.1637, over 4906.00 frames. ], tot_loss[loss=0.2676, simple_loss=0.3138, pruned_loss=0.1107, over 954622.10 frames. ], batch size: 43, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:31:40,448 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-26 00:32:18,689 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.270e+02 1.960e+02 2.424e+02 3.015e+02 7.276e+02, threshold=4.848e+02, percent-clipped=5.0 +2023-03-26 00:32:36,845 INFO [finetune.py:976] (1/7) Epoch 2, batch 4500, loss[loss=0.2829, simple_loss=0.3222, pruned_loss=0.1218, over 4898.00 frames. ], tot_loss[loss=0.271, simple_loss=0.3171, pruned_loss=0.1124, over 955588.95 frames. ], batch size: 37, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:32:38,196 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5123, 0.7253, 1.3243, 1.2080, 1.1179, 1.1256, 1.0222, 1.2582], + device='cuda:1'), covar=tensor([1.2472, 2.2520, 1.6802, 2.0371, 2.2653, 1.5135, 2.6175, 1.5343], + device='cuda:1'), in_proj_covar=tensor([0.0226, 0.0257, 0.0248, 0.0270, 0.0246, 0.0219, 0.0279, 0.0217], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 00:32:44,219 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10240.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:32:56,275 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-26 00:33:20,853 INFO [finetune.py:976] (1/7) Epoch 2, batch 4550, loss[loss=0.287, simple_loss=0.3291, pruned_loss=0.1224, over 4727.00 frames. ], tot_loss[loss=0.2704, simple_loss=0.3169, pruned_loss=0.112, over 955161.22 frames. ], batch size: 59, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:33:32,439 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=10288.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:33:32,521 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9747, 1.6561, 1.7803, 1.9513, 2.7040, 1.9415, 1.6860, 1.5560], + device='cuda:1'), covar=tensor([0.2813, 0.2889, 0.2292, 0.2181, 0.2380, 0.1463, 0.3160, 0.2299], + device='cuda:1'), in_proj_covar=tensor([0.0223, 0.0204, 0.0192, 0.0177, 0.0226, 0.0170, 0.0208, 0.0180], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:33:42,045 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7955, 3.9703, 3.8727, 1.8885, 4.1201, 3.0386, 0.7631, 2.7095], + device='cuda:1'), covar=tensor([0.2185, 0.1514, 0.1418, 0.3135, 0.1038, 0.0962, 0.4396, 0.1475], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0168, 0.0167, 0.0129, 0.0157, 0.0121, 0.0147, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 00:33:42,251 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.03 vs. limit=5.0 +2023-03-26 00:33:59,120 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10314.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:34:00,222 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.209e+02 1.803e+02 2.127e+02 2.576e+02 5.771e+02, threshold=4.255e+02, percent-clipped=1.0 +2023-03-26 00:34:14,434 INFO [finetune.py:976] (1/7) Epoch 2, batch 4600, loss[loss=0.2691, simple_loss=0.3121, pruned_loss=0.1131, over 4881.00 frames. ], tot_loss[loss=0.2678, simple_loss=0.3149, pruned_loss=0.1104, over 956515.77 frames. ], batch size: 35, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:34:45,544 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6063, 1.4062, 1.4808, 1.7209, 1.9297, 1.6614, 1.0114, 1.4044], + device='cuda:1'), covar=tensor([0.2734, 0.2835, 0.2339, 0.2072, 0.2153, 0.1380, 0.3327, 0.2133], + device='cuda:1'), in_proj_covar=tensor([0.0223, 0.0205, 0.0192, 0.0177, 0.0227, 0.0170, 0.0208, 0.0180], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:34:50,215 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=10362.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:35:01,344 INFO [finetune.py:976] (1/7) Epoch 2, batch 4650, loss[loss=0.2756, simple_loss=0.3237, pruned_loss=0.1138, over 4893.00 frames. ], tot_loss[loss=0.2667, simple_loss=0.3129, pruned_loss=0.1102, over 956131.89 frames. ], batch size: 35, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:35:12,306 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10395.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 00:35:13,537 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9764, 1.2473, 1.1746, 1.7238, 2.2275, 1.5310, 1.6242, 1.8300], + device='cuda:1'), covar=tensor([0.1374, 0.2150, 0.2027, 0.1254, 0.1957, 0.1919, 0.1340, 0.1797], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0098, 0.0117, 0.0094, 0.0125, 0.0097, 0.0099, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 00:35:14,745 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10399.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:35:14,776 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7923, 1.9272, 1.9128, 1.1468, 2.2240, 2.0801, 1.8553, 1.7713], + device='cuda:1'), covar=tensor([0.0930, 0.0705, 0.0880, 0.1277, 0.0442, 0.0861, 0.0954, 0.1213], + device='cuda:1'), in_proj_covar=tensor([0.0138, 0.0131, 0.0142, 0.0129, 0.0108, 0.0139, 0.0146, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:35:21,428 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10409.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:35:25,531 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.369e+02 1.834e+02 2.117e+02 2.478e+02 4.313e+02, threshold=4.233e+02, percent-clipped=1.0 +2023-03-26 00:35:27,258 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10418.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 00:35:34,604 INFO [finetune.py:976] (1/7) Epoch 2, batch 4700, loss[loss=0.2434, simple_loss=0.2789, pruned_loss=0.104, over 4435.00 frames. ], tot_loss[loss=0.2624, simple_loss=0.3085, pruned_loss=0.1081, over 955747.18 frames. ], batch size: 19, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:35:46,697 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=10447.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:35:46,878 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 00:35:51,005 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10454.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:36:01,571 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10470.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:36:07,253 INFO [finetune.py:976] (1/7) Epoch 2, batch 4750, loss[loss=0.2351, simple_loss=0.2882, pruned_loss=0.09102, over 4910.00 frames. ], tot_loss[loss=0.2596, simple_loss=0.306, pruned_loss=0.1066, over 956351.77 frames. ], batch size: 32, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:36:07,435 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 00:36:36,916 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10515.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 00:36:42,340 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.257e+02 1.733e+02 2.215e+02 2.656e+02 7.843e+02, threshold=4.429e+02, percent-clipped=2.0 +2023-03-26 00:36:48,431 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 00:36:51,173 INFO [finetune.py:976] (1/7) Epoch 2, batch 4800, loss[loss=0.3045, simple_loss=0.3467, pruned_loss=0.1311, over 4927.00 frames. ], tot_loss[loss=0.2633, simple_loss=0.31, pruned_loss=0.1083, over 956372.68 frames. ], batch size: 38, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:37:43,275 INFO [finetune.py:976] (1/7) Epoch 2, batch 4850, loss[loss=0.2186, simple_loss=0.2814, pruned_loss=0.07796, over 4888.00 frames. ], tot_loss[loss=0.2662, simple_loss=0.3135, pruned_loss=0.1094, over 956090.51 frames. ], batch size: 32, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:37:52,689 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1013, 0.9260, 1.0491, 0.3124, 0.7507, 1.1855, 1.2227, 1.0914], + device='cuda:1'), covar=tensor([0.0847, 0.0517, 0.0457, 0.0657, 0.0496, 0.0428, 0.0322, 0.0535], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0153, 0.0116, 0.0132, 0.0129, 0.0116, 0.0143, 0.0142], + device='cuda:1'), out_proj_covar=tensor([9.5665e-05, 1.1409e-04, 8.4624e-05, 9.7100e-05, 9.3302e-05, 8.5857e-05, + 1.0693e-04, 1.0527e-04], device='cuda:1') +2023-03-26 00:37:54,105 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 00:38:20,226 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3018, 1.3738, 1.4428, 0.9535, 1.3166, 1.6838, 1.7166, 1.4076], + device='cuda:1'), covar=tensor([0.1052, 0.0678, 0.0549, 0.0598, 0.0440, 0.0517, 0.0312, 0.0677], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0153, 0.0116, 0.0132, 0.0129, 0.0117, 0.0144, 0.0142], + device='cuda:1'), out_proj_covar=tensor([9.5895e-05, 1.1428e-04, 8.4753e-05, 9.7370e-05, 9.3406e-05, 8.6078e-05, + 1.0715e-04, 1.0529e-04], device='cuda:1') +2023-03-26 00:38:22,410 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.244e+02 1.945e+02 2.280e+02 2.839e+02 5.226e+02, threshold=4.560e+02, percent-clipped=1.0 +2023-03-26 00:38:33,265 INFO [finetune.py:976] (1/7) Epoch 2, batch 4900, loss[loss=0.2991, simple_loss=0.3397, pruned_loss=0.1293, over 4895.00 frames. ], tot_loss[loss=0.268, simple_loss=0.3153, pruned_loss=0.1104, over 957271.43 frames. ], batch size: 46, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:38:36,463 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-03-26 00:39:06,189 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 00:39:15,444 INFO [finetune.py:976] (1/7) Epoch 2, batch 4950, loss[loss=0.2478, simple_loss=0.3061, pruned_loss=0.09477, over 4933.00 frames. ], tot_loss[loss=0.2679, simple_loss=0.3156, pruned_loss=0.1101, over 956321.39 frames. ], batch size: 42, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:39:27,271 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10695.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:39:40,466 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.382e+02 1.832e+02 2.144e+02 2.563e+02 4.788e+02, threshold=4.289e+02, percent-clipped=1.0 +2023-03-26 00:39:42,226 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10718.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 00:39:48,191 INFO [finetune.py:976] (1/7) Epoch 2, batch 5000, loss[loss=0.28, simple_loss=0.3189, pruned_loss=0.1205, over 4896.00 frames. ], tot_loss[loss=0.265, simple_loss=0.3132, pruned_loss=0.1084, over 958368.94 frames. ], batch size: 35, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:39:59,860 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=10743.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:40:00,539 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10744.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:40:00,544 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0906, 0.8084, 0.9187, 1.0419, 1.1847, 1.1969, 1.0325, 0.9690], + device='cuda:1'), covar=tensor([0.0279, 0.0373, 0.0528, 0.0311, 0.0281, 0.0408, 0.0280, 0.0395], + device='cuda:1'), in_proj_covar=tensor([0.0080, 0.0111, 0.0132, 0.0111, 0.0101, 0.0097, 0.0087, 0.0106], + device='cuda:1'), out_proj_covar=tensor([6.2928e-05, 8.7768e-05, 1.0640e-04, 8.8257e-05, 8.0411e-05, 7.2115e-05, + 6.6735e-05, 8.2953e-05], device='cuda:1') +2023-03-26 00:40:13,892 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10765.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:40:14,499 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=10766.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 00:40:19,854 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10774.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:40:25,991 INFO [finetune.py:976] (1/7) Epoch 2, batch 5050, loss[loss=0.2632, simple_loss=0.3096, pruned_loss=0.1084, over 4861.00 frames. ], tot_loss[loss=0.262, simple_loss=0.3096, pruned_loss=0.1072, over 958613.57 frames. ], batch size: 49, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:40:48,615 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10805.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:40:52,129 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10810.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 00:40:54,763 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-03-26 00:40:55,640 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.117e+02 1.711e+02 2.025e+02 2.497e+02 4.623e+02, threshold=4.049e+02, percent-clipped=1.0 +2023-03-26 00:41:03,401 INFO [finetune.py:976] (1/7) Epoch 2, batch 5100, loss[loss=0.2997, simple_loss=0.3032, pruned_loss=0.1481, over 3759.00 frames. ], tot_loss[loss=0.2584, simple_loss=0.3058, pruned_loss=0.1054, over 955896.27 frames. ], batch size: 16, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:41:07,771 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10835.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:41:47,381 INFO [finetune.py:976] (1/7) Epoch 2, batch 5150, loss[loss=0.342, simple_loss=0.3703, pruned_loss=0.1569, over 4384.00 frames. ], tot_loss[loss=0.2594, simple_loss=0.3063, pruned_loss=0.1062, over 955740.57 frames. ], batch size: 65, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:41:54,265 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10885.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 00:41:54,349 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 00:42:25,067 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.204e+02 1.747e+02 2.089e+02 2.521e+02 5.475e+02, threshold=4.178e+02, percent-clipped=1.0 +2023-03-26 00:42:37,755 INFO [finetune.py:976] (1/7) Epoch 2, batch 5200, loss[loss=0.3095, simple_loss=0.3484, pruned_loss=0.1353, over 4897.00 frames. ], tot_loss[loss=0.263, simple_loss=0.31, pruned_loss=0.108, over 955119.54 frames. ], batch size: 37, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:43:00,783 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10946.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 00:43:29,722 INFO [finetune.py:976] (1/7) Epoch 2, batch 5250, loss[loss=0.2972, simple_loss=0.342, pruned_loss=0.1262, over 4865.00 frames. ], tot_loss[loss=0.2653, simple_loss=0.313, pruned_loss=0.1088, over 954282.90 frames. ], batch size: 34, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:44:03,160 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.206e+02 2.014e+02 2.376e+02 2.957e+02 8.531e+02, threshold=4.753e+02, percent-clipped=3.0 +2023-03-26 00:44:07,059 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-26 00:44:10,958 INFO [finetune.py:976] (1/7) Epoch 2, batch 5300, loss[loss=0.2096, simple_loss=0.2807, pruned_loss=0.06921, over 4814.00 frames. ], tot_loss[loss=0.2659, simple_loss=0.3139, pruned_loss=0.1089, over 954588.52 frames. ], batch size: 33, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:44:14,161 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7327, 0.5969, 1.4899, 1.2559, 1.3571, 1.3329, 1.1671, 1.5012], + device='cuda:1'), covar=tensor([1.3086, 2.6172, 2.1659, 2.0805, 2.3793, 1.5550, 2.9156, 1.8474], + device='cuda:1'), in_proj_covar=tensor([0.0227, 0.0257, 0.0251, 0.0270, 0.0245, 0.0219, 0.0280, 0.0218], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 00:44:40,300 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6717, 1.4622, 1.8965, 1.2921, 1.5901, 1.7800, 1.6105, 2.0035], + device='cuda:1'), covar=tensor([0.1250, 0.2147, 0.1263, 0.1494, 0.0973, 0.1120, 0.2286, 0.0758], + device='cuda:1'), in_proj_covar=tensor([0.0205, 0.0207, 0.0204, 0.0197, 0.0180, 0.0225, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:44:44,430 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11065.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:44:58,486 INFO [finetune.py:976] (1/7) Epoch 2, batch 5350, loss[loss=0.2652, simple_loss=0.3118, pruned_loss=0.1093, over 4836.00 frames. ], tot_loss[loss=0.2657, simple_loss=0.3142, pruned_loss=0.1086, over 954651.85 frames. ], batch size: 30, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:45:16,027 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11100.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:45:23,037 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11110.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 00:45:25,461 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=11113.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:45:27,189 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.217e+02 1.830e+02 2.178e+02 2.684e+02 7.389e+02, threshold=4.357e+02, percent-clipped=1.0 +2023-03-26 00:45:39,914 INFO [finetune.py:976] (1/7) Epoch 2, batch 5400, loss[loss=0.2389, simple_loss=0.2818, pruned_loss=0.09801, over 4850.00 frames. ], tot_loss[loss=0.2635, simple_loss=0.3113, pruned_loss=0.1079, over 951020.06 frames. ], batch size: 49, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:45:41,708 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11130.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:46:05,244 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=11158.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:46:22,020 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11174.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:46:24,381 INFO [finetune.py:976] (1/7) Epoch 2, batch 5450, loss[loss=0.2017, simple_loss=0.2626, pruned_loss=0.07043, over 4797.00 frames. ], tot_loss[loss=0.259, simple_loss=0.3067, pruned_loss=0.1056, over 952655.95 frames. ], batch size: 25, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:46:55,233 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.192e+02 1.719e+02 1.995e+02 2.396e+02 4.116e+02, threshold=3.991e+02, percent-clipped=0.0 +2023-03-26 00:47:11,560 INFO [finetune.py:976] (1/7) Epoch 2, batch 5500, loss[loss=0.258, simple_loss=0.3027, pruned_loss=0.1066, over 4825.00 frames. ], tot_loss[loss=0.2552, simple_loss=0.3027, pruned_loss=0.1038, over 952533.03 frames. ], batch size: 39, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:47:21,351 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11235.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:47:25,456 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11241.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 00:47:42,141 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6120, 1.1505, 0.8809, 1.4342, 1.9844, 0.6670, 1.2553, 1.5073], + device='cuda:1'), covar=tensor([0.1721, 0.2410, 0.2055, 0.1346, 0.2102, 0.2244, 0.1586, 0.2252], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0098, 0.0118, 0.0094, 0.0125, 0.0097, 0.0100, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 00:48:00,492 INFO [finetune.py:976] (1/7) Epoch 2, batch 5550, loss[loss=0.2309, simple_loss=0.2746, pruned_loss=0.0936, over 4778.00 frames. ], tot_loss[loss=0.2556, simple_loss=0.3032, pruned_loss=0.104, over 953206.45 frames. ], batch size: 26, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:48:06,204 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0274, 1.7736, 1.9442, 2.1667, 2.5064, 2.0752, 1.7115, 1.5917], + device='cuda:1'), covar=tensor([0.2909, 0.2884, 0.2215, 0.2127, 0.2737, 0.1451, 0.3365, 0.2439], + device='cuda:1'), in_proj_covar=tensor([0.0223, 0.0205, 0.0192, 0.0177, 0.0227, 0.0170, 0.0209, 0.0181], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:48:30,779 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.394e+02 1.988e+02 2.263e+02 2.636e+02 5.646e+02, threshold=4.525e+02, percent-clipped=4.0 +2023-03-26 00:48:37,263 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-26 00:48:38,279 INFO [finetune.py:976] (1/7) Epoch 2, batch 5600, loss[loss=0.2579, simple_loss=0.3, pruned_loss=0.1079, over 4725.00 frames. ], tot_loss[loss=0.2571, simple_loss=0.3058, pruned_loss=0.1042, over 953032.32 frames. ], batch size: 23, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:49:19,623 INFO [finetune.py:976] (1/7) Epoch 2, batch 5650, loss[loss=0.2399, simple_loss=0.2852, pruned_loss=0.09726, over 4695.00 frames. ], tot_loss[loss=0.2612, simple_loss=0.3101, pruned_loss=0.1061, over 953443.84 frames. ], batch size: 23, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:49:34,073 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.77 vs. limit=5.0 +2023-03-26 00:49:44,828 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11400.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:49:54,703 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.185e+02 1.700e+02 2.043e+02 2.333e+02 3.537e+02, threshold=4.085e+02, percent-clipped=0.0 +2023-03-26 00:50:01,882 INFO [finetune.py:976] (1/7) Epoch 2, batch 5700, loss[loss=0.1911, simple_loss=0.2343, pruned_loss=0.07396, over 4178.00 frames. ], tot_loss[loss=0.259, simple_loss=0.3064, pruned_loss=0.1058, over 936983.82 frames. ], batch size: 18, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:50:01,951 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.6482, 4.0777, 4.2279, 4.4721, 4.3609, 4.1364, 4.7386, 1.8073], + device='cuda:1'), covar=tensor([0.0712, 0.0796, 0.0795, 0.0894, 0.1072, 0.1367, 0.0643, 0.4504], + device='cuda:1'), in_proj_covar=tensor([0.0368, 0.0245, 0.0273, 0.0295, 0.0340, 0.0287, 0.0310, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 00:50:03,170 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11430.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:50:13,813 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=11448.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:50:33,843 INFO [finetune.py:976] (1/7) Epoch 3, batch 0, loss[loss=0.2147, simple_loss=0.2646, pruned_loss=0.0824, over 4811.00 frames. ], tot_loss[loss=0.2147, simple_loss=0.2646, pruned_loss=0.0824, over 4811.00 frames. ], batch size: 25, lr: 3.99e-03, grad_scale: 16.0 +2023-03-26 00:50:33,844 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 00:50:41,883 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6602, 0.8534, 1.5394, 1.4042, 1.3609, 1.3469, 1.2258, 1.4159], + device='cuda:1'), covar=tensor([1.2426, 2.0887, 1.5887, 1.7570, 1.9567, 1.3037, 2.2377, 1.4140], + device='cuda:1'), in_proj_covar=tensor([0.0227, 0.0257, 0.0250, 0.0269, 0.0244, 0.0218, 0.0279, 0.0217], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 00:50:46,663 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6196, 1.4924, 1.8723, 2.8629, 2.1919, 2.1036, 0.9762, 2.3204], + device='cuda:1'), covar=tensor([0.1876, 0.1769, 0.1405, 0.0629, 0.0831, 0.1318, 0.2047, 0.0732], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0120, 0.0139, 0.0164, 0.0105, 0.0146, 0.0130, 0.0108], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 00:50:55,328 INFO [finetune.py:1010] (1/7) Epoch 3, validation: loss=0.1864, simple_loss=0.2566, pruned_loss=0.05807, over 2265189.00 frames. +2023-03-26 00:50:55,329 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6289MB +2023-03-26 00:51:20,331 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=11478.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:51:22,965 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-03-26 00:51:25,255 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2667, 1.4119, 1.5228, 1.6760, 1.5181, 3.0934, 1.1951, 1.5201], + device='cuda:1'), covar=tensor([0.1249, 0.1803, 0.1550, 0.1276, 0.1746, 0.0316, 0.1768, 0.1970], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0080, 0.0077, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 00:51:29,403 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3541, 1.5323, 1.5866, 1.7860, 1.5247, 3.3508, 1.2517, 1.5717], + device='cuda:1'), covar=tensor([0.1174, 0.1733, 0.1280, 0.1135, 0.1766, 0.0240, 0.1613, 0.1846], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0080, 0.0078, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 00:51:37,281 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0723, 0.9739, 1.0892, 0.2535, 0.7394, 1.1676, 1.2327, 1.1298], + device='cuda:1'), covar=tensor([0.1016, 0.0568, 0.0428, 0.0703, 0.0509, 0.0562, 0.0402, 0.0617], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0155, 0.0117, 0.0134, 0.0131, 0.0117, 0.0144, 0.0143], + device='cuda:1'), out_proj_covar=tensor([9.6806e-05, 1.1525e-04, 8.5363e-05, 9.8409e-05, 9.4648e-05, 8.6350e-05, + 1.0786e-04, 1.0580e-04], device='cuda:1') +2023-03-26 00:51:38,815 INFO [finetune.py:976] (1/7) Epoch 3, batch 50, loss[loss=0.287, simple_loss=0.3363, pruned_loss=0.1188, over 4862.00 frames. ], tot_loss[loss=0.2634, simple_loss=0.3114, pruned_loss=0.1077, over 215570.01 frames. ], batch size: 34, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:51:45,831 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.161e+02 1.779e+02 2.075e+02 2.495e+02 4.593e+02, threshold=4.151e+02, percent-clipped=1.0 +2023-03-26 00:51:54,921 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11530.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:51:56,204 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11532.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:51:56,287 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.67 vs. limit=5.0 +2023-03-26 00:52:02,022 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11541.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 00:52:12,048 INFO [finetune.py:976] (1/7) Epoch 3, batch 100, loss[loss=0.1779, simple_loss=0.234, pruned_loss=0.06087, over 4828.00 frames. ], tot_loss[loss=0.2547, simple_loss=0.3026, pruned_loss=0.1034, over 381514.96 frames. ], batch size: 25, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:52:33,517 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=11589.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 00:52:36,431 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11593.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:52:47,407 INFO [finetune.py:976] (1/7) Epoch 3, batch 150, loss[loss=0.2446, simple_loss=0.2952, pruned_loss=0.09703, over 4820.00 frames. ], tot_loss[loss=0.2482, simple_loss=0.2962, pruned_loss=0.1001, over 508981.64 frames. ], batch size: 38, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:53:00,298 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.308e+02 1.853e+02 2.230e+02 2.582e+02 4.758e+02, threshold=4.459e+02, percent-clipped=3.0 +2023-03-26 00:53:24,123 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11635.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:53:48,111 INFO [finetune.py:976] (1/7) Epoch 3, batch 200, loss[loss=0.2765, simple_loss=0.3146, pruned_loss=0.1192, over 4108.00 frames. ], tot_loss[loss=0.2475, simple_loss=0.2948, pruned_loss=0.1001, over 608843.80 frames. ], batch size: 65, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:54:03,200 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.9233, 1.9295, 2.1317, 1.3259, 1.9207, 2.2383, 2.1182, 1.9327], + device='cuda:1'), covar=tensor([0.1063, 0.0677, 0.0416, 0.0796, 0.0365, 0.0463, 0.0327, 0.0570], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0155, 0.0117, 0.0135, 0.0132, 0.0118, 0.0145, 0.0144], + device='cuda:1'), out_proj_covar=tensor([9.7328e-05, 1.1563e-04, 8.5762e-05, 9.9095e-05, 9.5455e-05, 8.7137e-05, + 1.0854e-04, 1.0644e-04], device='cuda:1') +2023-03-26 00:54:15,071 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-03-26 00:54:23,864 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-26 00:54:35,492 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11696.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:54:37,310 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11699.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:54:44,848 INFO [finetune.py:976] (1/7) Epoch 3, batch 250, loss[loss=0.2468, simple_loss=0.3078, pruned_loss=0.09294, over 4867.00 frames. ], tot_loss[loss=0.2547, simple_loss=0.3015, pruned_loss=0.1039, over 685494.73 frames. ], batch size: 34, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:55:02,880 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.255e+02 1.783e+02 2.124e+02 2.629e+02 6.988e+02, threshold=4.248e+02, percent-clipped=1.0 +2023-03-26 00:55:15,655 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4602, 0.5375, 1.2532, 1.1548, 1.0996, 1.0908, 0.9928, 1.1818], + device='cuda:1'), covar=tensor([1.0209, 1.8214, 1.4436, 1.5270, 1.6552, 1.2012, 1.9511, 1.2521], + device='cuda:1'), in_proj_covar=tensor([0.0227, 0.0256, 0.0250, 0.0268, 0.0245, 0.0218, 0.0279, 0.0217], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 00:55:32,486 INFO [finetune.py:976] (1/7) Epoch 3, batch 300, loss[loss=0.3185, simple_loss=0.3499, pruned_loss=0.1436, over 4215.00 frames. ], tot_loss[loss=0.2571, simple_loss=0.3053, pruned_loss=0.1045, over 744437.43 frames. ], batch size: 65, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:55:36,034 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11760.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:56:13,665 INFO [finetune.py:976] (1/7) Epoch 3, batch 350, loss[loss=0.2816, simple_loss=0.34, pruned_loss=0.1116, over 4800.00 frames. ], tot_loss[loss=0.2595, simple_loss=0.3081, pruned_loss=0.1054, over 791170.11 frames. ], batch size: 51, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:56:20,323 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.015e+02 1.891e+02 2.265e+02 2.576e+02 3.939e+02, threshold=4.529e+02, percent-clipped=0.0 +2023-03-26 00:56:20,446 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11816.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:56:30,490 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11830.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:56:45,061 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11846.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:56:51,650 INFO [finetune.py:976] (1/7) Epoch 3, batch 400, loss[loss=0.26, simple_loss=0.2879, pruned_loss=0.116, over 4703.00 frames. ], tot_loss[loss=0.261, simple_loss=0.3098, pruned_loss=0.1061, over 826335.95 frames. ], batch size: 23, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:57:20,607 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11877.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:57:21,177 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=11878.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:57:33,123 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11888.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:57:55,601 INFO [finetune.py:976] (1/7) Epoch 3, batch 450, loss[loss=0.2635, simple_loss=0.3087, pruned_loss=0.1092, over 4817.00 frames. ], tot_loss[loss=0.2585, simple_loss=0.3075, pruned_loss=0.1047, over 855068.99 frames. ], batch size: 38, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:57:56,361 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11907.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:58:12,548 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.239e+02 1.814e+02 2.317e+02 2.793e+02 4.030e+02, threshold=4.633e+02, percent-clipped=0.0 +2023-03-26 00:58:15,056 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11920.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:58:39,240 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11948.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:58:50,438 INFO [finetune.py:976] (1/7) Epoch 3, batch 500, loss[loss=0.2476, simple_loss=0.3016, pruned_loss=0.09685, over 4819.00 frames. ], tot_loss[loss=0.2575, simple_loss=0.3055, pruned_loss=0.1047, over 876685.54 frames. ], batch size: 41, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:58:58,241 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8666, 1.6653, 1.3138, 1.6998, 1.6096, 1.5095, 1.4959, 2.5557], + device='cuda:1'), covar=tensor([1.3991, 1.4181, 1.1071, 1.4671, 1.1950, 0.7878, 1.4491, 0.3972], + device='cuda:1'), in_proj_covar=tensor([0.0262, 0.0241, 0.0216, 0.0277, 0.0231, 0.0193, 0.0235, 0.0179], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 00:59:19,995 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11981.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:59:27,098 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11984.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:59:31,811 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11991.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:59:46,859 INFO [finetune.py:976] (1/7) Epoch 3, batch 550, loss[loss=0.2456, simple_loss=0.2992, pruned_loss=0.09598, over 4897.00 frames. ], tot_loss[loss=0.2552, simple_loss=0.303, pruned_loss=0.1037, over 893947.53 frames. ], batch size: 35, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 00:59:49,285 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12009.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 00:59:53,392 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.083e+02 1.776e+02 2.066e+02 2.700e+02 4.009e+02, threshold=4.133e+02, percent-clipped=0.0 +2023-03-26 01:00:02,296 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4297, 1.4598, 1.4897, 1.7114, 1.6254, 3.2647, 1.3196, 1.6743], + device='cuda:1'), covar=tensor([0.1044, 0.1801, 0.1208, 0.1036, 0.1562, 0.0267, 0.1470, 0.1688], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0080, 0.0077, 0.0079, 0.0092, 0.0082, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 01:00:13,578 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12045.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:00:21,770 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12055.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:00:22,311 INFO [finetune.py:976] (1/7) Epoch 3, batch 600, loss[loss=0.2025, simple_loss=0.2535, pruned_loss=0.07575, over 4102.00 frames. ], tot_loss[loss=0.2541, simple_loss=0.3024, pruned_loss=0.1029, over 906692.91 frames. ], batch size: 18, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 01:00:41,416 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12076.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:00:42,047 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5115, 1.5094, 1.4825, 0.7821, 1.4194, 1.7254, 1.6829, 1.4512], + device='cuda:1'), covar=tensor([0.0876, 0.0524, 0.0533, 0.0670, 0.0419, 0.0429, 0.0304, 0.0521], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0156, 0.0117, 0.0135, 0.0132, 0.0118, 0.0146, 0.0143], + device='cuda:1'), out_proj_covar=tensor([9.7573e-05, 1.1579e-04, 8.5631e-05, 9.9222e-05, 9.5680e-05, 8.7229e-05, + 1.0896e-04, 1.0614e-04], device='cuda:1') +2023-03-26 01:00:48,037 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 01:01:07,907 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 01:01:09,780 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.79 vs. limit=5.0 +2023-03-26 01:01:16,812 INFO [finetune.py:976] (1/7) Epoch 3, batch 650, loss[loss=0.2369, simple_loss=0.3017, pruned_loss=0.08604, over 4850.00 frames. ], tot_loss[loss=0.2572, simple_loss=0.3063, pruned_loss=0.1041, over 918142.89 frames. ], batch size: 44, lr: 3.99e-03, grad_scale: 32.0 +2023-03-26 01:01:23,450 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.190e+02 1.901e+02 2.250e+02 2.651e+02 5.885e+02, threshold=4.501e+02, percent-clipped=2.0 +2023-03-26 01:01:50,149 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12137.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:02:02,047 INFO [finetune.py:976] (1/7) Epoch 3, batch 700, loss[loss=0.2772, simple_loss=0.325, pruned_loss=0.1147, over 4813.00 frames. ], tot_loss[loss=0.2588, simple_loss=0.3082, pruned_loss=0.1047, over 927046.52 frames. ], batch size: 39, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:02:02,767 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0488, 0.9247, 1.0298, 0.2654, 0.6854, 1.1421, 1.2034, 1.0575], + device='cuda:1'), covar=tensor([0.0981, 0.0554, 0.0480, 0.0741, 0.0523, 0.0461, 0.0355, 0.0551], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0157, 0.0118, 0.0136, 0.0133, 0.0119, 0.0147, 0.0144], + device='cuda:1'), out_proj_covar=tensor([9.8391e-05, 1.1709e-04, 8.6563e-05, 9.9923e-05, 9.6455e-05, 8.7789e-05, + 1.0979e-04, 1.0685e-04], device='cuda:1') +2023-03-26 01:02:12,356 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12172.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:02:24,036 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12188.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:02:39,282 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12202.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:02:43,052 INFO [finetune.py:976] (1/7) Epoch 3, batch 750, loss[loss=0.2674, simple_loss=0.3194, pruned_loss=0.1077, over 4909.00 frames. ], tot_loss[loss=0.2595, simple_loss=0.3091, pruned_loss=0.105, over 933749.58 frames. ], batch size: 36, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:02:54,949 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.207e+02 1.858e+02 2.308e+02 2.738e+02 5.308e+02, threshold=4.616e+02, percent-clipped=1.0 +2023-03-26 01:03:14,093 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=12236.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:03:26,043 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5212, 3.8778, 4.0560, 4.3621, 4.2040, 4.0042, 4.6404, 1.3969], + device='cuda:1'), covar=tensor([0.0699, 0.0752, 0.0738, 0.0898, 0.1259, 0.1370, 0.0542, 0.5149], + device='cuda:1'), in_proj_covar=tensor([0.0368, 0.0245, 0.0274, 0.0297, 0.0341, 0.0286, 0.0310, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:03:30,123 INFO [finetune.py:976] (1/7) Epoch 3, batch 800, loss[loss=0.2519, simple_loss=0.3065, pruned_loss=0.09859, over 4811.00 frames. ], tot_loss[loss=0.2574, simple_loss=0.3078, pruned_loss=0.1035, over 937459.88 frames. ], batch size: 39, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:03:42,862 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12276.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:03:44,702 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2466, 1.3708, 1.2455, 1.5632, 1.4255, 2.9805, 1.2439, 1.5024], + device='cuda:1'), covar=tensor([0.1135, 0.1851, 0.1225, 0.1017, 0.1686, 0.0310, 0.1531, 0.1713], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0080, 0.0077, 0.0079, 0.0092, 0.0083, 0.0085, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 01:04:03,327 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12291.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:04:11,113 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12304.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:04:12,239 INFO [finetune.py:976] (1/7) Epoch 3, batch 850, loss[loss=0.2373, simple_loss=0.2767, pruned_loss=0.09894, over 4222.00 frames. ], tot_loss[loss=0.2578, simple_loss=0.3069, pruned_loss=0.1043, over 942247.77 frames. ], batch size: 18, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:04:19,496 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.224e+02 1.828e+02 2.109e+02 2.576e+02 5.946e+02, threshold=4.217e+02, percent-clipped=1.0 +2023-03-26 01:04:46,215 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=12339.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:04:46,840 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12340.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:05:06,382 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12355.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:05:06,916 INFO [finetune.py:976] (1/7) Epoch 3, batch 900, loss[loss=0.2566, simple_loss=0.2901, pruned_loss=0.1115, over 4352.00 frames. ], tot_loss[loss=0.2528, simple_loss=0.3021, pruned_loss=0.1017, over 943636.44 frames. ], batch size: 65, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:05:30,276 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2554, 1.2168, 1.5344, 1.0988, 1.0650, 1.3741, 1.2133, 1.4049], + device='cuda:1'), covar=tensor([0.1536, 0.2241, 0.1459, 0.1484, 0.1476, 0.1394, 0.2734, 0.1222], + device='cuda:1'), in_proj_covar=tensor([0.0207, 0.0206, 0.0206, 0.0198, 0.0181, 0.0227, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:05:43,766 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=12403.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:05:45,596 INFO [finetune.py:976] (1/7) Epoch 3, batch 950, loss[loss=0.2361, simple_loss=0.2834, pruned_loss=0.09433, over 4721.00 frames. ], tot_loss[loss=0.2513, simple_loss=0.3001, pruned_loss=0.1012, over 947592.29 frames. ], batch size: 23, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:05:57,751 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.215e+02 1.779e+02 2.123e+02 2.532e+02 4.452e+02, threshold=4.246e+02, percent-clipped=1.0 +2023-03-26 01:06:12,418 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12432.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:06:14,282 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7936, 1.7694, 1.6914, 1.9581, 1.4268, 4.3341, 1.6055, 2.2971], + device='cuda:1'), covar=tensor([0.3419, 0.2259, 0.1926, 0.2033, 0.1690, 0.0144, 0.2441, 0.1292], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0110, 0.0115, 0.0119, 0.0115, 0.0097, 0.0100, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 01:06:22,002 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12446.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:06:28,911 INFO [finetune.py:976] (1/7) Epoch 3, batch 1000, loss[loss=0.2602, simple_loss=0.3071, pruned_loss=0.1067, over 4762.00 frames. ], tot_loss[loss=0.254, simple_loss=0.3031, pruned_loss=0.1025, over 950598.17 frames. ], batch size: 26, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:06:39,168 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12472.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:07:00,386 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1570, 1.9367, 1.5122, 2.1579, 2.2245, 1.7586, 2.7028, 2.0598], + device='cuda:1'), covar=tensor([0.2531, 0.5940, 0.6285, 0.6092, 0.4159, 0.2937, 0.5254, 0.3857], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0197, 0.0239, 0.0254, 0.0219, 0.0184, 0.0207, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:07:17,533 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12502.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:07:19,893 INFO [finetune.py:976] (1/7) Epoch 3, batch 1050, loss[loss=0.2532, simple_loss=0.3159, pruned_loss=0.09527, over 4904.00 frames. ], tot_loss[loss=0.2566, simple_loss=0.3061, pruned_loss=0.1035, over 950577.81 frames. ], batch size: 36, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:07:20,621 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12507.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:07:31,080 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.261e+02 2.008e+02 2.380e+02 2.733e+02 7.204e+02, threshold=4.759e+02, percent-clipped=3.0 +2023-03-26 01:07:38,120 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=12520.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:08:10,384 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=12550.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:08:17,959 INFO [finetune.py:976] (1/7) Epoch 3, batch 1100, loss[loss=0.267, simple_loss=0.3081, pruned_loss=0.1129, over 4136.00 frames. ], tot_loss[loss=0.2571, simple_loss=0.3066, pruned_loss=0.1038, over 950919.40 frames. ], batch size: 65, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:08:20,799 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-03-26 01:08:35,451 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12576.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:08:58,281 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12604.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:08:59,394 INFO [finetune.py:976] (1/7) Epoch 3, batch 1150, loss[loss=0.2873, simple_loss=0.3346, pruned_loss=0.12, over 4759.00 frames. ], tot_loss[loss=0.2579, simple_loss=0.3078, pruned_loss=0.104, over 951088.66 frames. ], batch size: 54, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:09:11,543 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.144e+02 1.712e+02 1.953e+02 2.432e+02 5.551e+02, threshold=3.906e+02, percent-clipped=1.0 +2023-03-26 01:09:21,030 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=12624.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:09:31,349 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.59 vs. limit=5.0 +2023-03-26 01:09:31,916 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.05 vs. limit=2.0 +2023-03-26 01:09:33,059 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8361, 1.2735, 0.8307, 1.6791, 2.0930, 1.1889, 1.5177, 1.7587], + device='cuda:1'), covar=tensor([0.1535, 0.2125, 0.2240, 0.1212, 0.2078, 0.2134, 0.1376, 0.2070], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0099, 0.0117, 0.0094, 0.0125, 0.0098, 0.0100, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 01:09:42,260 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12640.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:09:52,718 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0311, 1.5194, 1.6290, 1.7053, 1.4679, 1.5557, 1.5890, 1.6975], + device='cuda:1'), covar=tensor([1.4613, 2.3777, 1.7859, 2.1320, 2.3208, 1.5215, 2.8107, 1.5072], + device='cuda:1'), in_proj_covar=tensor([0.0228, 0.0256, 0.0253, 0.0269, 0.0245, 0.0219, 0.0279, 0.0219], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 01:09:55,578 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=12652.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:10:03,871 INFO [finetune.py:976] (1/7) Epoch 3, batch 1200, loss[loss=0.1773, simple_loss=0.2229, pruned_loss=0.06591, over 4213.00 frames. ], tot_loss[loss=0.2557, simple_loss=0.3053, pruned_loss=0.1031, over 951874.52 frames. ], batch size: 18, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:10:31,612 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=12688.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:10:43,574 INFO [finetune.py:976] (1/7) Epoch 3, batch 1250, loss[loss=0.2631, simple_loss=0.3027, pruned_loss=0.1117, over 4672.00 frames. ], tot_loss[loss=0.2536, simple_loss=0.303, pruned_loss=0.1022, over 952752.29 frames. ], batch size: 23, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:10:51,325 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.066e+02 1.791e+02 2.152e+02 2.550e+02 3.946e+02, threshold=4.304e+02, percent-clipped=1.0 +2023-03-26 01:11:06,480 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12732.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:11:11,415 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 01:11:25,781 INFO [finetune.py:976] (1/7) Epoch 3, batch 1300, loss[loss=0.2241, simple_loss=0.2716, pruned_loss=0.08827, over 4905.00 frames. ], tot_loss[loss=0.2497, simple_loss=0.2989, pruned_loss=0.1003, over 955236.56 frames. ], batch size: 46, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:11:28,850 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5334, 1.0510, 0.7980, 1.4311, 1.9201, 0.6943, 1.1918, 1.4249], + device='cuda:1'), covar=tensor([0.1548, 0.2350, 0.1998, 0.1253, 0.2230, 0.2260, 0.1553, 0.2025], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0099, 0.0117, 0.0094, 0.0125, 0.0098, 0.0100, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 01:11:31,223 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12763.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:11:46,017 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0213, 1.8546, 1.4078, 1.8754, 1.9338, 1.5904, 2.3330, 1.8691], + device='cuda:1'), covar=tensor([0.2142, 0.4389, 0.5041, 0.4632, 0.3462, 0.2381, 0.4415, 0.3052], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0198, 0.0239, 0.0255, 0.0220, 0.0184, 0.0208, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:11:48,953 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=12780.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:12:19,506 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12802.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:12:22,421 INFO [finetune.py:976] (1/7) Epoch 3, batch 1350, loss[loss=0.2351, simple_loss=0.2993, pruned_loss=0.08545, over 4816.00 frames. ], tot_loss[loss=0.2511, simple_loss=0.3001, pruned_loss=0.101, over 955377.74 frames. ], batch size: 38, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:12:40,691 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.079e+02 1.832e+02 2.182e+02 2.674e+02 3.468e+02, threshold=4.364e+02, percent-clipped=0.0 +2023-03-26 01:12:42,688 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 01:12:43,199 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6553, 1.4421, 1.3852, 1.6389, 1.7612, 1.6277, 0.7629, 1.4119], + device='cuda:1'), covar=tensor([0.2349, 0.2256, 0.1954, 0.1827, 0.1730, 0.1172, 0.3007, 0.1924], + device='cuda:1'), in_proj_covar=tensor([0.0225, 0.0206, 0.0193, 0.0179, 0.0228, 0.0170, 0.0210, 0.0183], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:12:50,666 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12824.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 01:12:53,882 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.15 vs. limit=5.0 +2023-03-26 01:12:55,666 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 01:12:57,885 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4176, 0.9373, 0.8119, 1.2926, 1.8334, 0.6689, 1.0807, 1.3097], + device='cuda:1'), covar=tensor([0.1703, 0.2527, 0.2001, 0.1428, 0.2144, 0.2298, 0.1658, 0.2113], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0098, 0.0117, 0.0094, 0.0124, 0.0098, 0.0100, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 01:13:21,545 INFO [finetune.py:976] (1/7) Epoch 3, batch 1400, loss[loss=0.2065, simple_loss=0.2639, pruned_loss=0.07457, over 4788.00 frames. ], tot_loss[loss=0.252, simple_loss=0.3019, pruned_loss=0.1011, over 955558.42 frames. ], batch size: 25, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:14:18,642 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8519, 1.6437, 1.3375, 1.7447, 1.8744, 1.5121, 2.1902, 1.6895], + device='cuda:1'), covar=tensor([0.2212, 0.4705, 0.5361, 0.4540, 0.3554, 0.2528, 0.4281, 0.3257], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0198, 0.0239, 0.0255, 0.0220, 0.0184, 0.0209, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:14:19,107 INFO [finetune.py:976] (1/7) Epoch 3, batch 1450, loss[loss=0.2907, simple_loss=0.3376, pruned_loss=0.1219, over 4821.00 frames. ], tot_loss[loss=0.2538, simple_loss=0.3043, pruned_loss=0.1016, over 954476.20 frames. ], batch size: 38, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:14:38,006 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.344e+02 1.849e+02 2.229e+02 2.789e+02 4.972e+02, threshold=4.459e+02, percent-clipped=1.0 +2023-03-26 01:15:00,772 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2444, 2.0092, 2.7202, 1.6906, 2.5570, 2.4899, 1.9445, 2.5389], + device='cuda:1'), covar=tensor([0.1897, 0.2169, 0.1888, 0.2725, 0.1075, 0.1909, 0.2668, 0.1329], + device='cuda:1'), in_proj_covar=tensor([0.0208, 0.0206, 0.0206, 0.0199, 0.0181, 0.0228, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:15:13,772 INFO [finetune.py:976] (1/7) Epoch 3, batch 1500, loss[loss=0.2523, simple_loss=0.3108, pruned_loss=0.09691, over 4717.00 frames. ], tot_loss[loss=0.2569, simple_loss=0.3073, pruned_loss=0.1033, over 956456.91 frames. ], batch size: 54, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:15:58,821 INFO [finetune.py:976] (1/7) Epoch 3, batch 1550, loss[loss=0.227, simple_loss=0.2834, pruned_loss=0.08531, over 4751.00 frames. ], tot_loss[loss=0.2563, simple_loss=0.3069, pruned_loss=0.1028, over 957893.52 frames. ], batch size: 27, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:16:09,621 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.917e+01 1.840e+02 2.219e+02 2.788e+02 8.539e+02, threshold=4.437e+02, percent-clipped=2.0 +2023-03-26 01:16:40,603 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8859, 1.1330, 0.9041, 1.7496, 2.1677, 1.2604, 1.4925, 1.7589], + device='cuda:1'), covar=tensor([0.1542, 0.2292, 0.1981, 0.1150, 0.1984, 0.1901, 0.1416, 0.1957], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0098, 0.0116, 0.0094, 0.0124, 0.0097, 0.0100, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 01:16:44,395 INFO [finetune.py:976] (1/7) Epoch 3, batch 1600, loss[loss=0.173, simple_loss=0.2383, pruned_loss=0.05383, over 4892.00 frames. ], tot_loss[loss=0.2524, simple_loss=0.3027, pruned_loss=0.1011, over 958913.71 frames. ], batch size: 35, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:17:41,241 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=13102.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:17:43,645 INFO [finetune.py:976] (1/7) Epoch 3, batch 1650, loss[loss=0.2121, simple_loss=0.2621, pruned_loss=0.081, over 4910.00 frames. ], tot_loss[loss=0.2508, simple_loss=0.3003, pruned_loss=0.1007, over 957072.73 frames. ], batch size: 43, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:17:55,152 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.264e+02 1.841e+02 2.109e+02 2.450e+02 4.226e+02, threshold=4.217e+02, percent-clipped=0.0 +2023-03-26 01:17:56,994 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=13119.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 01:17:57,017 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9833, 1.3197, 0.8589, 1.7440, 2.0780, 1.4699, 1.6547, 1.8114], + device='cuda:1'), covar=tensor([0.1791, 0.2707, 0.2635, 0.1522, 0.2282, 0.2434, 0.1653, 0.2519], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0098, 0.0116, 0.0093, 0.0124, 0.0097, 0.0100, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 01:18:32,133 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=13150.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:18:41,622 INFO [finetune.py:976] (1/7) Epoch 3, batch 1700, loss[loss=0.2738, simple_loss=0.3043, pruned_loss=0.1217, over 4860.00 frames. ], tot_loss[loss=0.2493, simple_loss=0.2987, pruned_loss=0.09996, over 959561.25 frames. ], batch size: 49, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:19:01,473 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-26 01:19:22,355 INFO [finetune.py:976] (1/7) Epoch 3, batch 1750, loss[loss=0.2878, simple_loss=0.337, pruned_loss=0.1193, over 4821.00 frames. ], tot_loss[loss=0.2508, simple_loss=0.3003, pruned_loss=0.1007, over 959476.05 frames. ], batch size: 33, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:19:31,217 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.198e+02 1.973e+02 2.314e+02 2.693e+02 6.749e+02, threshold=4.629e+02, percent-clipped=3.0 +2023-03-26 01:19:42,804 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8852, 1.6081, 1.3206, 1.5781, 1.5847, 1.5085, 1.5372, 2.3947], + device='cuda:1'), covar=tensor([1.2363, 1.1840, 1.0006, 1.2943, 1.0258, 0.7082, 1.1707, 0.3578], + device='cuda:1'), in_proj_covar=tensor([0.0266, 0.0243, 0.0216, 0.0279, 0.0232, 0.0194, 0.0236, 0.0181], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 01:20:02,794 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.6584, 3.1741, 3.3423, 3.5108, 3.4228, 3.2033, 3.7056, 1.5701], + device='cuda:1'), covar=tensor([0.0848, 0.0882, 0.0859, 0.0966, 0.1229, 0.1381, 0.0992, 0.4405], + device='cuda:1'), in_proj_covar=tensor([0.0372, 0.0247, 0.0278, 0.0299, 0.0345, 0.0290, 0.0314, 0.0305], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:20:08,787 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-26 01:20:18,400 INFO [finetune.py:976] (1/7) Epoch 3, batch 1800, loss[loss=0.2542, simple_loss=0.3121, pruned_loss=0.09813, over 4852.00 frames. ], tot_loss[loss=0.254, simple_loss=0.3038, pruned_loss=0.1021, over 956193.02 frames. ], batch size: 47, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:20:53,358 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8979, 1.8593, 1.4277, 1.9746, 2.0160, 1.5439, 2.5028, 1.9100], + device='cuda:1'), covar=tensor([0.2249, 0.4927, 0.5033, 0.4943, 0.3501, 0.2541, 0.4849, 0.3196], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0196, 0.0238, 0.0254, 0.0219, 0.0184, 0.0208, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:20:59,916 INFO [finetune.py:976] (1/7) Epoch 3, batch 1850, loss[loss=0.2734, simple_loss=0.3332, pruned_loss=0.1068, over 4882.00 frames. ], tot_loss[loss=0.2548, simple_loss=0.3049, pruned_loss=0.1024, over 953989.26 frames. ], batch size: 43, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:21:08,031 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.104e+02 1.691e+02 1.934e+02 2.488e+02 4.482e+02, threshold=3.868e+02, percent-clipped=0.0 +2023-03-26 01:21:50,069 INFO [finetune.py:976] (1/7) Epoch 3, batch 1900, loss[loss=0.2392, simple_loss=0.2884, pruned_loss=0.09504, over 4860.00 frames. ], tot_loss[loss=0.2547, simple_loss=0.3056, pruned_loss=0.1019, over 956581.80 frames. ], batch size: 44, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:22:40,358 INFO [finetune.py:976] (1/7) Epoch 3, batch 1950, loss[loss=0.2507, simple_loss=0.2955, pruned_loss=0.1029, over 4747.00 frames. ], tot_loss[loss=0.253, simple_loss=0.3035, pruned_loss=0.1013, over 956124.96 frames. ], batch size: 27, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:22:46,996 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.125e+02 1.849e+02 2.191e+02 2.472e+02 6.030e+02, threshold=4.381e+02, percent-clipped=4.0 +2023-03-26 01:22:48,831 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=13419.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:22:50,882 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 01:23:25,633 INFO [finetune.py:976] (1/7) Epoch 3, batch 2000, loss[loss=0.1847, simple_loss=0.2385, pruned_loss=0.06547, over 4671.00 frames. ], tot_loss[loss=0.249, simple_loss=0.2991, pruned_loss=0.09943, over 955132.46 frames. ], batch size: 23, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:23:36,602 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=13467.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:23:59,085 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1731, 3.6473, 3.8027, 4.0470, 3.9577, 3.6310, 4.2561, 1.3836], + device='cuda:1'), covar=tensor([0.0697, 0.0737, 0.0800, 0.0795, 0.1043, 0.1276, 0.0613, 0.4571], + device='cuda:1'), in_proj_covar=tensor([0.0370, 0.0246, 0.0277, 0.0296, 0.0343, 0.0289, 0.0312, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:24:12,058 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 01:24:18,862 INFO [finetune.py:976] (1/7) Epoch 3, batch 2050, loss[loss=0.2637, simple_loss=0.2988, pruned_loss=0.1142, over 4676.00 frames. ], tot_loss[loss=0.2447, simple_loss=0.295, pruned_loss=0.09724, over 956001.73 frames. ], batch size: 23, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:24:32,936 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.03 vs. limit=5.0 +2023-03-26 01:24:33,950 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.160e+02 1.806e+02 2.206e+02 2.674e+02 5.377e+02, threshold=4.412e+02, percent-clipped=2.0 +2023-03-26 01:25:00,751 INFO [finetune.py:976] (1/7) Epoch 3, batch 2100, loss[loss=0.2545, simple_loss=0.2929, pruned_loss=0.108, over 4766.00 frames. ], tot_loss[loss=0.2451, simple_loss=0.2953, pruned_loss=0.09747, over 957261.22 frames. ], batch size: 26, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:25:21,244 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.62 vs. limit=2.0 +2023-03-26 01:25:28,215 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6167, 1.1115, 1.4406, 1.3258, 1.2605, 1.2916, 1.2700, 1.3878], + device='cuda:1'), covar=tensor([0.9931, 1.5563, 1.1910, 1.4461, 1.5334, 1.0478, 1.9562, 1.1500], + device='cuda:1'), in_proj_covar=tensor([0.0228, 0.0257, 0.0253, 0.0269, 0.0245, 0.0218, 0.0280, 0.0219], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 01:25:45,311 INFO [finetune.py:976] (1/7) Epoch 3, batch 2150, loss[loss=0.3277, simple_loss=0.3616, pruned_loss=0.1469, over 4166.00 frames. ], tot_loss[loss=0.2497, simple_loss=0.3002, pruned_loss=0.09962, over 954855.20 frames. ], batch size: 66, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:26:01,365 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.135e+02 1.918e+02 2.256e+02 2.684e+02 5.304e+02, threshold=4.512e+02, percent-clipped=2.0 +2023-03-26 01:26:23,536 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8091, 1.5639, 1.5008, 1.5439, 1.9111, 1.5293, 2.0868, 1.7297], + device='cuda:1'), covar=tensor([0.2309, 0.4216, 0.4833, 0.4240, 0.3563, 0.2540, 0.3904, 0.3041], + device='cuda:1'), in_proj_covar=tensor([0.0163, 0.0195, 0.0238, 0.0253, 0.0219, 0.0184, 0.0207, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:26:27,573 INFO [finetune.py:976] (1/7) Epoch 3, batch 2200, loss[loss=0.2534, simple_loss=0.3202, pruned_loss=0.09323, over 4898.00 frames. ], tot_loss[loss=0.2515, simple_loss=0.3019, pruned_loss=0.1005, over 952128.98 frames. ], batch size: 35, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:27:00,297 INFO [finetune.py:976] (1/7) Epoch 3, batch 2250, loss[loss=0.2257, simple_loss=0.2662, pruned_loss=0.09255, over 3961.00 frames. ], tot_loss[loss=0.2536, simple_loss=0.3042, pruned_loss=0.1015, over 953768.87 frames. ], batch size: 17, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:27:08,388 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.215e+02 1.927e+02 2.166e+02 2.564e+02 5.587e+02, threshold=4.333e+02, percent-clipped=2.0 +2023-03-26 01:27:24,294 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=13737.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:27:40,497 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9182, 1.4108, 1.5831, 1.6556, 1.4076, 1.4962, 1.5469, 1.6144], + device='cuda:1'), covar=tensor([1.1123, 1.7260, 1.2525, 1.5553, 1.7375, 1.1502, 1.9895, 1.1523], + device='cuda:1'), in_proj_covar=tensor([0.0228, 0.0256, 0.0252, 0.0268, 0.0244, 0.0218, 0.0278, 0.0219], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 01:27:42,154 INFO [finetune.py:976] (1/7) Epoch 3, batch 2300, loss[loss=0.2157, simple_loss=0.2696, pruned_loss=0.08087, over 4755.00 frames. ], tot_loss[loss=0.2521, simple_loss=0.3034, pruned_loss=0.1005, over 951970.71 frames. ], batch size: 28, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:28:25,833 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=13798.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 01:28:27,211 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8553, 4.5553, 4.3910, 2.6135, 4.8067, 3.5229, 0.7860, 3.4007], + device='cuda:1'), covar=tensor([0.2410, 0.1688, 0.1322, 0.2976, 0.0735, 0.0836, 0.4890, 0.1398], + device='cuda:1'), in_proj_covar=tensor([0.0156, 0.0168, 0.0165, 0.0129, 0.0156, 0.0120, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 01:28:36,682 INFO [finetune.py:976] (1/7) Epoch 3, batch 2350, loss[loss=0.2539, simple_loss=0.2888, pruned_loss=0.1095, over 4871.00 frames. ], tot_loss[loss=0.25, simple_loss=0.3007, pruned_loss=0.09968, over 951744.90 frames. ], batch size: 34, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:28:50,255 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.589e+01 1.791e+02 2.182e+02 2.579e+02 6.380e+02, threshold=4.365e+02, percent-clipped=2.0 +2023-03-26 01:29:31,312 INFO [finetune.py:976] (1/7) Epoch 3, batch 2400, loss[loss=0.28, simple_loss=0.2979, pruned_loss=0.131, over 4247.00 frames. ], tot_loss[loss=0.2463, simple_loss=0.2972, pruned_loss=0.0977, over 951831.81 frames. ], batch size: 65, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:30:09,626 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6565, 1.5616, 1.5467, 1.6395, 1.2572, 3.5406, 1.3671, 2.0315], + device='cuda:1'), covar=tensor([0.3140, 0.2185, 0.1894, 0.2140, 0.1731, 0.0156, 0.2522, 0.1294], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0110, 0.0115, 0.0118, 0.0115, 0.0097, 0.0100, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 01:30:12,731 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7027, 1.6460, 1.5983, 1.7141, 1.0579, 3.6617, 1.4437, 2.1293], + device='cuda:1'), covar=tensor([0.3590, 0.2415, 0.2015, 0.2259, 0.2044, 0.0186, 0.2529, 0.1256], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0110, 0.0115, 0.0118, 0.0115, 0.0097, 0.0100, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 01:30:15,690 INFO [finetune.py:976] (1/7) Epoch 3, batch 2450, loss[loss=0.2236, simple_loss=0.2756, pruned_loss=0.08581, over 4832.00 frames. ], tot_loss[loss=0.2459, simple_loss=0.296, pruned_loss=0.09792, over 952842.69 frames. ], batch size: 47, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:30:18,062 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7250, 3.4102, 3.4264, 1.9546, 3.6613, 2.6029, 1.3530, 2.5457], + device='cuda:1'), covar=tensor([0.3284, 0.2074, 0.1492, 0.2871, 0.0975, 0.1110, 0.3627, 0.1535], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0168, 0.0166, 0.0129, 0.0156, 0.0121, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 01:30:29,103 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.953e+01 1.906e+02 2.150e+02 2.578e+02 4.181e+02, threshold=4.299e+02, percent-clipped=0.0 +2023-03-26 01:30:59,052 INFO [finetune.py:976] (1/7) Epoch 3, batch 2500, loss[loss=0.2246, simple_loss=0.2908, pruned_loss=0.07925, over 4827.00 frames. ], tot_loss[loss=0.2483, simple_loss=0.2981, pruned_loss=0.09922, over 954076.33 frames. ], batch size: 33, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:31:53,428 INFO [finetune.py:976] (1/7) Epoch 3, batch 2550, loss[loss=0.2347, simple_loss=0.3006, pruned_loss=0.08442, over 4826.00 frames. ], tot_loss[loss=0.2507, simple_loss=0.301, pruned_loss=0.1002, over 953562.59 frames. ], batch size: 51, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:32:00,756 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8916, 5.0186, 4.7450, 2.8559, 5.2081, 3.7766, 0.9285, 3.5116], + device='cuda:1'), covar=tensor([0.2427, 0.2058, 0.1297, 0.2999, 0.0728, 0.0841, 0.5147, 0.1486], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0168, 0.0166, 0.0129, 0.0157, 0.0121, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 01:32:02,446 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.924e+01 1.822e+02 2.075e+02 2.520e+02 4.375e+02, threshold=4.150e+02, percent-clipped=1.0 +2023-03-26 01:32:07,574 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1809, 1.3863, 1.0008, 1.3540, 1.5205, 2.4590, 1.2505, 1.5428], + device='cuda:1'), covar=tensor([0.0965, 0.1695, 0.1179, 0.0947, 0.1548, 0.0375, 0.1422, 0.1679], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0080, 0.0078, 0.0079, 0.0093, 0.0083, 0.0085, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 01:32:27,278 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-26 01:32:30,093 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1745, 1.8964, 2.0297, 1.0097, 2.2302, 2.4239, 1.8438, 1.9958], + device='cuda:1'), covar=tensor([0.1154, 0.0837, 0.0551, 0.0898, 0.0488, 0.0650, 0.0611, 0.0633], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0156, 0.0117, 0.0135, 0.0131, 0.0119, 0.0145, 0.0143], + device='cuda:1'), out_proj_covar=tensor([9.7503e-05, 1.1654e-04, 8.5434e-05, 9.9339e-05, 9.4989e-05, 8.7575e-05, + 1.0831e-04, 1.0568e-04], device='cuda:1') +2023-03-26 01:32:30,592 INFO [finetune.py:976] (1/7) Epoch 3, batch 2600, loss[loss=0.2262, simple_loss=0.2973, pruned_loss=0.0776, over 4806.00 frames. ], tot_loss[loss=0.2522, simple_loss=0.3029, pruned_loss=0.1007, over 954954.70 frames. ], batch size: 40, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:32:41,357 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14072.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:33:04,965 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14093.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 01:33:16,409 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14103.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:33:18,156 INFO [finetune.py:976] (1/7) Epoch 3, batch 2650, loss[loss=0.2363, simple_loss=0.2975, pruned_loss=0.0875, over 4810.00 frames. ], tot_loss[loss=0.2526, simple_loss=0.3043, pruned_loss=0.1004, over 956262.59 frames. ], batch size: 40, lr: 3.98e-03, grad_scale: 16.0 +2023-03-26 01:33:28,088 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8575, 1.4415, 1.5592, 1.6205, 1.4153, 1.4177, 1.5685, 1.5537], + device='cuda:1'), covar=tensor([1.0250, 1.5720, 1.2106, 1.4612, 1.5496, 1.1503, 1.9755, 1.1088], + device='cuda:1'), in_proj_covar=tensor([0.0227, 0.0255, 0.0253, 0.0267, 0.0244, 0.0218, 0.0279, 0.0218], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 01:33:34,825 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.159e+02 1.796e+02 2.195e+02 2.771e+02 4.502e+02, threshold=4.390e+02, percent-clipped=2.0 +2023-03-26 01:33:37,254 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14120.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 01:33:39,715 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1085, 1.9112, 1.4833, 2.2059, 2.1626, 1.6877, 2.7006, 2.0838], + device='cuda:1'), covar=tensor([0.2003, 0.4857, 0.5063, 0.4359, 0.3249, 0.2383, 0.4161, 0.2915], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0195, 0.0239, 0.0254, 0.0220, 0.0184, 0.0208, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:33:56,202 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14133.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:34:16,925 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9739, 1.6839, 2.4837, 3.7559, 2.6533, 2.6060, 0.7837, 2.9250], + device='cuda:1'), covar=tensor([0.1804, 0.1662, 0.1369, 0.0535, 0.0838, 0.1836, 0.2226, 0.0635], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0120, 0.0139, 0.0165, 0.0105, 0.0145, 0.0131, 0.0107], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 01:34:18,019 INFO [finetune.py:976] (1/7) Epoch 3, batch 2700, loss[loss=0.247, simple_loss=0.2964, pruned_loss=0.09877, over 4798.00 frames. ], tot_loss[loss=0.2503, simple_loss=0.3021, pruned_loss=0.09926, over 955757.49 frames. ], batch size: 29, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:34:28,742 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14164.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:34:50,967 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14181.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 01:35:13,612 INFO [finetune.py:976] (1/7) Epoch 3, batch 2750, loss[loss=0.1992, simple_loss=0.2608, pruned_loss=0.06881, over 4928.00 frames. ], tot_loss[loss=0.2482, simple_loss=0.2993, pruned_loss=0.09851, over 955067.11 frames. ], batch size: 33, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:35:20,820 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.010e+02 1.613e+02 1.949e+02 2.418e+02 3.837e+02, threshold=3.898e+02, percent-clipped=0.0 +2023-03-26 01:35:31,740 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 01:35:50,233 INFO [finetune.py:976] (1/7) Epoch 3, batch 2800, loss[loss=0.2366, simple_loss=0.2856, pruned_loss=0.09378, over 4807.00 frames. ], tot_loss[loss=0.2459, simple_loss=0.2963, pruned_loss=0.09778, over 954747.93 frames. ], batch size: 25, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:35:50,374 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3018, 2.1235, 1.5343, 2.0651, 2.2074, 1.7517, 2.7355, 2.2025], + device='cuda:1'), covar=tensor([0.1888, 0.4180, 0.4911, 0.4637, 0.3315, 0.2210, 0.5187, 0.2703], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0195, 0.0238, 0.0253, 0.0220, 0.0184, 0.0208, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:36:18,504 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8444, 1.7216, 1.3368, 1.8369, 1.8890, 1.5157, 2.2415, 1.8120], + device='cuda:1'), covar=tensor([0.2229, 0.4352, 0.5118, 0.4405, 0.3474, 0.2321, 0.4268, 0.3014], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0194, 0.0237, 0.0252, 0.0219, 0.0183, 0.0207, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:36:35,462 INFO [finetune.py:976] (1/7) Epoch 3, batch 2850, loss[loss=0.2578, simple_loss=0.2902, pruned_loss=0.1127, over 4240.00 frames. ], tot_loss[loss=0.2434, simple_loss=0.2943, pruned_loss=0.09625, over 955318.82 frames. ], batch size: 18, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:36:48,618 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.210e+02 1.714e+02 2.069e+02 2.397e+02 3.427e+02, threshold=4.138e+02, percent-clipped=0.0 +2023-03-26 01:36:51,952 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.73 vs. limit=5.0 +2023-03-26 01:37:26,619 INFO [finetune.py:976] (1/7) Epoch 3, batch 2900, loss[loss=0.2656, simple_loss=0.3388, pruned_loss=0.09625, over 4819.00 frames. ], tot_loss[loss=0.2479, simple_loss=0.2989, pruned_loss=0.0985, over 953703.36 frames. ], batch size: 38, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:38:03,913 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5536, 1.3737, 1.6578, 1.9920, 1.5426, 3.2182, 1.1845, 1.4996], + device='cuda:1'), covar=tensor([0.1038, 0.1899, 0.1483, 0.1077, 0.1760, 0.0271, 0.1796, 0.2007], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0080, 0.0078, 0.0079, 0.0093, 0.0083, 0.0085, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 01:38:04,601 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.30 vs. limit=5.0 +2023-03-26 01:38:06,366 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14393.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:38:25,748 INFO [finetune.py:976] (1/7) Epoch 3, batch 2950, loss[loss=0.2973, simple_loss=0.344, pruned_loss=0.1252, over 4820.00 frames. ], tot_loss[loss=0.2527, simple_loss=0.3038, pruned_loss=0.1008, over 954291.86 frames. ], batch size: 39, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:38:35,703 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14413.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:38:38,198 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.306e+02 1.859e+02 2.169e+02 2.721e+02 5.785e+02, threshold=4.339e+02, percent-clipped=3.0 +2023-03-26 01:38:50,260 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14428.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:38:53,856 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.88 vs. limit=5.0 +2023-03-26 01:39:02,087 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=14441.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:39:02,180 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6817, 0.5853, 1.5336, 1.3502, 1.4118, 1.3390, 1.1839, 1.4192], + device='cuda:1'), covar=tensor([0.7387, 1.3350, 1.0757, 1.0702, 1.2274, 0.8840, 1.3893, 0.9882], + device='cuda:1'), in_proj_covar=tensor([0.0226, 0.0254, 0.0252, 0.0266, 0.0242, 0.0217, 0.0277, 0.0218], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 01:39:02,801 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.32 vs. limit=5.0 +2023-03-26 01:39:20,295 INFO [finetune.py:976] (1/7) Epoch 3, batch 3000, loss[loss=0.2698, simple_loss=0.3242, pruned_loss=0.1077, over 4917.00 frames. ], tot_loss[loss=0.2515, simple_loss=0.3027, pruned_loss=0.1001, over 953937.11 frames. ], batch size: 38, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:39:20,296 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 01:39:22,658 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6666, 1.5719, 1.6186, 1.6364, 1.0849, 3.0509, 1.2852, 1.7697], + device='cuda:1'), covar=tensor([0.3364, 0.2385, 0.1927, 0.2214, 0.1965, 0.0249, 0.2573, 0.1382], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0111, 0.0116, 0.0119, 0.0116, 0.0097, 0.0101, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 01:39:37,114 INFO [finetune.py:1010] (1/7) Epoch 3, validation: loss=0.1777, simple_loss=0.2485, pruned_loss=0.05342, over 2265189.00 frames. +2023-03-26 01:39:37,114 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6289MB +2023-03-26 01:39:42,136 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14459.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:39:56,808 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14474.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:40:03,357 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14476.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 01:40:15,391 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14487.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:40:33,267 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 01:40:36,152 INFO [finetune.py:976] (1/7) Epoch 3, batch 3050, loss[loss=0.2214, simple_loss=0.2866, pruned_loss=0.07812, over 4773.00 frames. ], tot_loss[loss=0.2528, simple_loss=0.3043, pruned_loss=0.1007, over 953819.98 frames. ], batch size: 28, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:40:52,882 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.290e+02 1.934e+02 2.277e+02 2.724e+02 4.940e+02, threshold=4.554e+02, percent-clipped=2.0 +2023-03-26 01:41:02,911 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.36 vs. limit=5.0 +2023-03-26 01:41:17,994 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14548.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:41:22,798 INFO [finetune.py:976] (1/7) Epoch 3, batch 3100, loss[loss=0.1832, simple_loss=0.2532, pruned_loss=0.05664, over 4885.00 frames. ], tot_loss[loss=0.25, simple_loss=0.3016, pruned_loss=0.09917, over 954784.27 frames. ], batch size: 32, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:42:01,023 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-03-26 01:42:10,662 INFO [finetune.py:976] (1/7) Epoch 3, batch 3150, loss[loss=0.2476, simple_loss=0.3087, pruned_loss=0.09329, over 4768.00 frames. ], tot_loss[loss=0.247, simple_loss=0.2985, pruned_loss=0.09779, over 955396.71 frames. ], batch size: 26, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:42:18,343 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.181e+02 1.775e+02 2.189e+02 2.683e+02 4.981e+02, threshold=4.378e+02, percent-clipped=2.0 +2023-03-26 01:43:00,562 INFO [finetune.py:976] (1/7) Epoch 3, batch 3200, loss[loss=0.2445, simple_loss=0.2748, pruned_loss=0.1071, over 4826.00 frames. ], tot_loss[loss=0.2432, simple_loss=0.2943, pruned_loss=0.096, over 955431.33 frames. ], batch size: 30, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:43:41,546 INFO [finetune.py:976] (1/7) Epoch 3, batch 3250, loss[loss=0.1931, simple_loss=0.2522, pruned_loss=0.06699, over 4746.00 frames. ], tot_loss[loss=0.2433, simple_loss=0.2944, pruned_loss=0.09605, over 956518.78 frames. ], batch size: 26, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:43:54,504 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.171e+02 1.718e+02 2.102e+02 2.544e+02 5.358e+02, threshold=4.204e+02, percent-clipped=1.0 +2023-03-26 01:44:08,045 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14728.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:44:15,461 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14740.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:44:31,812 INFO [finetune.py:976] (1/7) Epoch 3, batch 3300, loss[loss=0.2349, simple_loss=0.2891, pruned_loss=0.0903, over 4832.00 frames. ], tot_loss[loss=0.2472, simple_loss=0.2986, pruned_loss=0.09789, over 953880.98 frames. ], batch size: 30, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:44:33,774 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14759.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:44:42,756 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14769.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:44:46,517 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.59 vs. limit=5.0 +2023-03-26 01:44:48,713 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=14776.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:44:48,744 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14776.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 01:45:08,529 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14801.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:45:16,879 INFO [finetune.py:976] (1/7) Epoch 3, batch 3350, loss[loss=0.2513, simple_loss=0.3069, pruned_loss=0.09791, over 4831.00 frames. ], tot_loss[loss=0.2491, simple_loss=0.3011, pruned_loss=0.09854, over 955370.54 frames. ], batch size: 47, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:45:17,533 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=14807.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:45:20,992 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4456, 1.3593, 1.3554, 1.4805, 0.9374, 2.8584, 1.0537, 1.5388], + device='cuda:1'), covar=tensor([0.3283, 0.2437, 0.2031, 0.2203, 0.2079, 0.0233, 0.2944, 0.1552], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0111, 0.0116, 0.0119, 0.0116, 0.0097, 0.0101, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 01:45:29,676 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.682e+02 2.084e+02 2.593e+02 4.183e+02, threshold=4.169e+02, percent-clipped=0.0 +2023-03-26 01:45:39,452 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=14824.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 01:45:52,840 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14843.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:46:01,259 INFO [finetune.py:976] (1/7) Epoch 3, batch 3400, loss[loss=0.2573, simple_loss=0.303, pruned_loss=0.1058, over 4808.00 frames. ], tot_loss[loss=0.2514, simple_loss=0.3029, pruned_loss=0.1, over 955468.50 frames. ], batch size: 40, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:46:16,658 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1030, 1.7578, 2.6942, 1.8501, 2.3197, 2.3331, 1.7565, 2.4607], + device='cuda:1'), covar=tensor([0.1504, 0.2143, 0.1544, 0.2189, 0.0952, 0.1687, 0.2664, 0.0947], + device='cuda:1'), in_proj_covar=tensor([0.0207, 0.0206, 0.0206, 0.0197, 0.0182, 0.0227, 0.0217, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:46:41,112 INFO [finetune.py:976] (1/7) Epoch 3, batch 3450, loss[loss=0.1941, simple_loss=0.2441, pruned_loss=0.07206, over 4673.00 frames. ], tot_loss[loss=0.2488, simple_loss=0.3009, pruned_loss=0.09835, over 954877.21 frames. ], batch size: 23, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:46:50,318 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7773, 1.7500, 2.0039, 2.0945, 1.9090, 3.6035, 1.6003, 1.8550], + device='cuda:1'), covar=tensor([0.1095, 0.1753, 0.1235, 0.1062, 0.1645, 0.0342, 0.1580, 0.1761], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0078, 0.0080, 0.0093, 0.0084, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 01:46:53,149 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.352e+02 1.935e+02 2.237e+02 2.692e+02 3.962e+02, threshold=4.475e+02, percent-clipped=0.0 +2023-03-26 01:47:05,781 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3172, 1.3047, 1.2673, 1.6484, 1.4913, 2.9209, 1.1847, 1.5135], + device='cuda:1'), covar=tensor([0.0981, 0.1682, 0.1243, 0.0923, 0.1533, 0.0288, 0.1452, 0.1583], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0078, 0.0079, 0.0093, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 01:47:27,611 INFO [finetune.py:976] (1/7) Epoch 3, batch 3500, loss[loss=0.2413, simple_loss=0.2952, pruned_loss=0.09369, over 4755.00 frames. ], tot_loss[loss=0.2461, simple_loss=0.2978, pruned_loss=0.09723, over 953985.79 frames. ], batch size: 59, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:47:59,682 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1740, 1.2973, 1.5422, 1.1370, 1.1427, 1.3403, 1.3030, 1.4852], + device='cuda:1'), covar=tensor([0.1340, 0.1904, 0.1183, 0.1333, 0.1051, 0.1205, 0.2546, 0.0937], + device='cuda:1'), in_proj_covar=tensor([0.0207, 0.0206, 0.0206, 0.0198, 0.0183, 0.0228, 0.0218, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 01:48:19,804 INFO [finetune.py:976] (1/7) Epoch 3, batch 3550, loss[loss=0.2277, simple_loss=0.2813, pruned_loss=0.08701, over 4815.00 frames. ], tot_loss[loss=0.2444, simple_loss=0.2954, pruned_loss=0.09673, over 954318.09 frames. ], batch size: 25, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:48:23,658 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 01:48:26,974 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.116e+02 1.780e+02 2.199e+02 2.800e+02 5.904e+02, threshold=4.398e+02, percent-clipped=2.0 +2023-03-26 01:49:03,540 INFO [finetune.py:976] (1/7) Epoch 3, batch 3600, loss[loss=0.2841, simple_loss=0.3267, pruned_loss=0.1207, over 4739.00 frames. ], tot_loss[loss=0.2431, simple_loss=0.2937, pruned_loss=0.09628, over 954986.34 frames. ], batch size: 59, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:49:04,441 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-03-26 01:49:12,103 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15069.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:49:42,970 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=15096.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:49:49,066 INFO [finetune.py:976] (1/7) Epoch 3, batch 3650, loss[loss=0.1982, simple_loss=0.2674, pruned_loss=0.06449, over 4867.00 frames. ], tot_loss[loss=0.2472, simple_loss=0.2977, pruned_loss=0.09838, over 953124.29 frames. ], batch size: 31, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:49:56,317 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.204e+02 1.927e+02 2.238e+02 2.686e+02 4.916e+02, threshold=4.476e+02, percent-clipped=1.0 +2023-03-26 01:49:56,389 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=15117.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:50:28,376 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15143.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:50:40,138 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.31 vs. limit=5.0 +2023-03-26 01:50:41,688 INFO [finetune.py:976] (1/7) Epoch 3, batch 3700, loss[loss=0.2485, simple_loss=0.298, pruned_loss=0.09952, over 4895.00 frames. ], tot_loss[loss=0.2507, simple_loss=0.3021, pruned_loss=0.09967, over 953612.59 frames. ], batch size: 35, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:51:00,621 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5535, 1.4278, 1.7068, 1.8466, 1.5768, 3.5853, 1.2306, 1.7050], + device='cuda:1'), covar=tensor([0.1054, 0.1761, 0.1219, 0.1041, 0.1662, 0.0230, 0.1616, 0.1778], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0078, 0.0079, 0.0093, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 01:51:16,177 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=15191.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:51:19,628 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=15195.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:51:29,140 INFO [finetune.py:976] (1/7) Epoch 3, batch 3750, loss[loss=0.2384, simple_loss=0.2956, pruned_loss=0.09066, over 4722.00 frames. ], tot_loss[loss=0.2488, simple_loss=0.3006, pruned_loss=0.09851, over 951015.69 frames. ], batch size: 54, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:51:40,515 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.151e+02 1.801e+02 2.153e+02 2.622e+02 6.720e+02, threshold=4.305e+02, percent-clipped=1.0 +2023-03-26 01:52:33,796 INFO [finetune.py:976] (1/7) Epoch 3, batch 3800, loss[loss=0.2088, simple_loss=0.2696, pruned_loss=0.07403, over 4807.00 frames. ], tot_loss[loss=0.249, simple_loss=0.3011, pruned_loss=0.09849, over 951134.30 frames. ], batch size: 25, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:52:33,936 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=15256.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 01:53:22,426 INFO [finetune.py:976] (1/7) Epoch 3, batch 3850, loss[loss=0.2734, simple_loss=0.3162, pruned_loss=0.1153, over 4808.00 frames. ], tot_loss[loss=0.2459, simple_loss=0.2986, pruned_loss=0.09665, over 952852.31 frames. ], batch size: 40, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:53:22,499 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7563, 3.6482, 3.5613, 1.6829, 3.7436, 2.8501, 0.7936, 2.6632], + device='cuda:1'), covar=tensor([0.2627, 0.1570, 0.1426, 0.3023, 0.0972, 0.0979, 0.4137, 0.1336], + device='cuda:1'), in_proj_covar=tensor([0.0155, 0.0167, 0.0164, 0.0128, 0.0154, 0.0120, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 01:53:38,224 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3256, 1.5036, 1.5181, 0.8885, 1.4740, 1.7642, 1.7590, 1.3663], + device='cuda:1'), covar=tensor([0.0999, 0.0542, 0.0398, 0.0624, 0.0400, 0.0472, 0.0289, 0.0585], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0158, 0.0119, 0.0137, 0.0133, 0.0121, 0.0148, 0.0144], + device='cuda:1'), out_proj_covar=tensor([9.8544e-05, 1.1795e-04, 8.6498e-05, 1.0102e-04, 9.6218e-05, 8.9182e-05, + 1.1053e-04, 1.0682e-04], device='cuda:1') +2023-03-26 01:53:39,317 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.276e+02 1.876e+02 2.253e+02 2.579e+02 5.032e+02, threshold=4.505e+02, percent-clipped=1.0 +2023-03-26 01:53:40,093 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1191, 1.9154, 1.5672, 2.0509, 1.9513, 1.8175, 1.6929, 2.9112], + device='cuda:1'), covar=tensor([1.1852, 1.2564, 0.9279, 1.2800, 0.9917, 0.6194, 1.3476, 0.3677], + device='cuda:1'), in_proj_covar=tensor([0.0272, 0.0249, 0.0219, 0.0284, 0.0235, 0.0196, 0.0239, 0.0185], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 01:53:42,532 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 01:54:26,485 INFO [finetune.py:976] (1/7) Epoch 3, batch 3900, loss[loss=0.2662, simple_loss=0.3066, pruned_loss=0.1129, over 4908.00 frames. ], tot_loss[loss=0.2442, simple_loss=0.2962, pruned_loss=0.0961, over 954862.05 frames. ], batch size: 37, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:55:10,141 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15396.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:55:16,819 INFO [finetune.py:976] (1/7) Epoch 3, batch 3950, loss[loss=0.1845, simple_loss=0.2409, pruned_loss=0.06409, over 4751.00 frames. ], tot_loss[loss=0.2393, simple_loss=0.2917, pruned_loss=0.09345, over 956573.24 frames. ], batch size: 27, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:55:25,255 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.128e+02 1.678e+02 2.164e+02 2.472e+02 4.231e+02, threshold=4.328e+02, percent-clipped=0.0 +2023-03-26 01:55:52,198 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=15444.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 01:55:59,826 INFO [finetune.py:976] (1/7) Epoch 3, batch 4000, loss[loss=0.235, simple_loss=0.293, pruned_loss=0.08847, over 4839.00 frames. ], tot_loss[loss=0.2405, simple_loss=0.2918, pruned_loss=0.09457, over 953774.66 frames. ], batch size: 47, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:57:05,458 INFO [finetune.py:976] (1/7) Epoch 3, batch 4050, loss[loss=0.2454, simple_loss=0.2943, pruned_loss=0.09824, over 4896.00 frames. ], tot_loss[loss=0.2434, simple_loss=0.2956, pruned_loss=0.09555, over 956328.51 frames. ], batch size: 35, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:57:20,270 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.142e+02 1.798e+02 2.110e+02 2.647e+02 5.396e+02, threshold=4.219e+02, percent-clipped=2.0 +2023-03-26 01:57:58,327 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=15551.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 01:58:06,352 INFO [finetune.py:976] (1/7) Epoch 3, batch 4100, loss[loss=0.2712, simple_loss=0.3209, pruned_loss=0.1108, over 4182.00 frames. ], tot_loss[loss=0.2468, simple_loss=0.2991, pruned_loss=0.09727, over 954156.99 frames. ], batch size: 66, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:59:02,792 INFO [finetune.py:976] (1/7) Epoch 3, batch 4150, loss[loss=0.2554, simple_loss=0.3042, pruned_loss=0.1033, over 4748.00 frames. ], tot_loss[loss=0.2473, simple_loss=0.2999, pruned_loss=0.09732, over 954989.64 frames. ], batch size: 27, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 01:59:10,587 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.238e+02 1.791e+02 2.157e+02 2.467e+02 4.537e+02, threshold=4.313e+02, percent-clipped=1.0 +2023-03-26 01:59:51,465 INFO [finetune.py:976] (1/7) Epoch 3, batch 4200, loss[loss=0.2384, simple_loss=0.2991, pruned_loss=0.08884, over 4788.00 frames. ], tot_loss[loss=0.2469, simple_loss=0.3002, pruned_loss=0.09682, over 954415.66 frames. ], batch size: 51, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 02:00:53,260 INFO [finetune.py:976] (1/7) Epoch 3, batch 4250, loss[loss=0.2509, simple_loss=0.3039, pruned_loss=0.09896, over 4748.00 frames. ], tot_loss[loss=0.2438, simple_loss=0.2971, pruned_loss=0.09524, over 955942.87 frames. ], batch size: 23, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 02:01:00,001 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.228e+02 1.731e+02 2.073e+02 2.469e+02 5.386e+02, threshold=4.147e+02, percent-clipped=2.0 +2023-03-26 02:01:14,005 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-26 02:01:43,527 INFO [finetune.py:976] (1/7) Epoch 3, batch 4300, loss[loss=0.2089, simple_loss=0.2615, pruned_loss=0.07818, over 4833.00 frames. ], tot_loss[loss=0.2406, simple_loss=0.2933, pruned_loss=0.09396, over 955944.40 frames. ], batch size: 33, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 02:02:02,406 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.77 vs. limit=2.0 +2023-03-26 02:02:43,560 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 02:02:43,902 INFO [finetune.py:976] (1/7) Epoch 3, batch 4350, loss[loss=0.2536, simple_loss=0.3066, pruned_loss=0.1003, over 4837.00 frames. ], tot_loss[loss=0.238, simple_loss=0.2905, pruned_loss=0.09278, over 955107.80 frames. ], batch size: 47, lr: 3.98e-03, grad_scale: 32.0 +2023-03-26 02:03:01,710 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.218e+02 1.712e+02 2.007e+02 2.460e+02 4.679e+02, threshold=4.015e+02, percent-clipped=1.0 +2023-03-26 02:03:33,437 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15851.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:03:36,425 INFO [finetune.py:976] (1/7) Epoch 3, batch 4400, loss[loss=0.2328, simple_loss=0.2899, pruned_loss=0.08785, over 4754.00 frames. ], tot_loss[loss=0.2395, simple_loss=0.2915, pruned_loss=0.09373, over 953782.63 frames. ], batch size: 27, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:04:05,493 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=15883.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:04:06,139 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8787, 1.0243, 1.6448, 1.5227, 1.3876, 1.4115, 1.4275, 1.5786], + device='cuda:1'), covar=tensor([0.9238, 1.4870, 1.1553, 1.3217, 1.4127, 0.9709, 1.6701, 1.0481], + device='cuda:1'), in_proj_covar=tensor([0.0226, 0.0253, 0.0252, 0.0265, 0.0242, 0.0216, 0.0277, 0.0218], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 02:04:20,879 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=15899.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:04:30,466 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.77 vs. limit=5.0 +2023-03-26 02:04:30,835 INFO [finetune.py:976] (1/7) Epoch 3, batch 4450, loss[loss=0.2646, simple_loss=0.3382, pruned_loss=0.0955, over 4822.00 frames. ], tot_loss[loss=0.2437, simple_loss=0.2961, pruned_loss=0.09567, over 951756.16 frames. ], batch size: 40, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:04:48,084 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.107e+02 1.872e+02 2.224e+02 2.526e+02 5.583e+02, threshold=4.448e+02, percent-clipped=1.0 +2023-03-26 02:04:52,504 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 02:05:15,993 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=15944.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:05:23,247 INFO [finetune.py:976] (1/7) Epoch 3, batch 4500, loss[loss=0.2886, simple_loss=0.3237, pruned_loss=0.1268, over 4822.00 frames. ], tot_loss[loss=0.2437, simple_loss=0.2965, pruned_loss=0.09548, over 952422.55 frames. ], batch size: 49, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:05:26,297 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0239, 1.8646, 1.8775, 2.2156, 1.4310, 4.5321, 1.7664, 2.3775], + device='cuda:1'), covar=tensor([0.3333, 0.2273, 0.1860, 0.1974, 0.1724, 0.0103, 0.2355, 0.1379], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0111, 0.0116, 0.0120, 0.0116, 0.0097, 0.0101, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 02:05:42,211 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6611, 1.5571, 1.6104, 0.9210, 1.6119, 1.8786, 1.7345, 1.5338], + device='cuda:1'), covar=tensor([0.1150, 0.0692, 0.0509, 0.0758, 0.0499, 0.0409, 0.0438, 0.0655], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0160, 0.0119, 0.0139, 0.0135, 0.0122, 0.0148, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.9781e-05, 1.1921e-04, 8.6884e-05, 1.0252e-04, 9.7455e-05, 8.9965e-05, + 1.1070e-04, 1.0826e-04], device='cuda:1') +2023-03-26 02:06:17,329 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16001.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:06:17,365 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8023, 1.2570, 1.7007, 1.5827, 1.4716, 1.4654, 1.4203, 1.6158], + device='cuda:1'), covar=tensor([0.7206, 1.0832, 0.8743, 1.0369, 1.0780, 0.8063, 1.3028, 0.8006], + device='cuda:1'), in_proj_covar=tensor([0.0227, 0.0254, 0.0253, 0.0266, 0.0242, 0.0217, 0.0278, 0.0219], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 02:06:24,310 INFO [finetune.py:976] (1/7) Epoch 3, batch 4550, loss[loss=0.2774, simple_loss=0.3312, pruned_loss=0.1118, over 4833.00 frames. ], tot_loss[loss=0.2459, simple_loss=0.299, pruned_loss=0.09643, over 953166.09 frames. ], batch size: 47, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:06:36,869 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.237e+02 1.760e+02 2.085e+02 2.487e+02 3.865e+02, threshold=4.170e+02, percent-clipped=0.0 +2023-03-26 02:06:59,581 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 02:07:05,907 INFO [finetune.py:976] (1/7) Epoch 3, batch 4600, loss[loss=0.2622, simple_loss=0.2988, pruned_loss=0.1129, over 4374.00 frames. ], tot_loss[loss=0.2435, simple_loss=0.2969, pruned_loss=0.095, over 952343.62 frames. ], batch size: 66, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:07:06,639 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2152, 2.0370, 2.3839, 1.0111, 2.3990, 2.6597, 2.1880, 2.2808], + device='cuda:1'), covar=tensor([0.0951, 0.0750, 0.0429, 0.0794, 0.0639, 0.0709, 0.0486, 0.0558], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0160, 0.0119, 0.0140, 0.0135, 0.0122, 0.0149, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.9996e-05, 1.1936e-04, 8.6938e-05, 1.0273e-04, 9.7884e-05, 9.0454e-05, + 1.1103e-04, 1.0849e-04], device='cuda:1') +2023-03-26 02:07:11,388 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16062.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:07:35,225 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16085.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:07:59,849 INFO [finetune.py:976] (1/7) Epoch 3, batch 4650, loss[loss=0.2571, simple_loss=0.2981, pruned_loss=0.108, over 4847.00 frames. ], tot_loss[loss=0.2423, simple_loss=0.2948, pruned_loss=0.09487, over 952640.04 frames. ], batch size: 49, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:08:07,260 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.007e+02 1.817e+02 2.201e+02 2.598e+02 3.850e+02, threshold=4.403e+02, percent-clipped=0.0 +2023-03-26 02:08:29,953 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3846, 1.4641, 1.4589, 0.8060, 1.6624, 1.4769, 1.5190, 1.4332], + device='cuda:1'), covar=tensor([0.0712, 0.0765, 0.0732, 0.1056, 0.0703, 0.0783, 0.0690, 0.1211], + device='cuda:1'), in_proj_covar=tensor([0.0141, 0.0134, 0.0146, 0.0131, 0.0111, 0.0144, 0.0148, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:08:30,556 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5488, 1.6206, 1.6695, 1.9580, 1.6693, 3.3616, 1.3704, 1.6920], + device='cuda:1'), covar=tensor([0.0969, 0.1614, 0.1194, 0.0951, 0.1514, 0.0212, 0.1422, 0.1663], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0077, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 02:08:39,866 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16146.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:08:42,113 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16149.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:08:52,381 INFO [finetune.py:976] (1/7) Epoch 3, batch 4700, loss[loss=0.1968, simple_loss=0.2478, pruned_loss=0.07296, over 4707.00 frames. ], tot_loss[loss=0.2381, simple_loss=0.2908, pruned_loss=0.0927, over 953574.78 frames. ], batch size: 23, lr: 3.97e-03, grad_scale: 64.0 +2023-03-26 02:09:03,343 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6418, 1.5542, 1.6367, 1.9187, 1.6380, 3.3369, 1.3549, 1.5955], + device='cuda:1'), covar=tensor([0.0981, 0.1811, 0.1122, 0.0954, 0.1686, 0.0234, 0.1544, 0.1781], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0077, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 02:09:41,914 INFO [finetune.py:976] (1/7) Epoch 3, batch 4750, loss[loss=0.2327, simple_loss=0.2859, pruned_loss=0.08982, over 4842.00 frames. ], tot_loss[loss=0.2355, simple_loss=0.2883, pruned_loss=0.09138, over 952648.98 frames. ], batch size: 44, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:09:44,971 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16210.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:09:49,753 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.299e+02 1.759e+02 2.065e+02 2.416e+02 4.123e+02, threshold=4.129e+02, percent-clipped=0.0 +2023-03-26 02:10:03,618 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16239.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:10:16,138 INFO [finetune.py:976] (1/7) Epoch 3, batch 4800, loss[loss=0.3029, simple_loss=0.34, pruned_loss=0.1329, over 4902.00 frames. ], tot_loss[loss=0.2381, simple_loss=0.2909, pruned_loss=0.09264, over 954320.23 frames. ], batch size: 37, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:10:20,266 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1021, 2.1256, 2.0593, 1.3921, 2.3319, 2.3133, 2.2358, 1.8849], + device='cuda:1'), covar=tensor([0.0621, 0.0563, 0.0721, 0.1040, 0.0397, 0.0673, 0.0642, 0.0955], + device='cuda:1'), in_proj_covar=tensor([0.0139, 0.0132, 0.0143, 0.0129, 0.0110, 0.0142, 0.0147, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:10:47,379 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-03-26 02:11:10,319 INFO [finetune.py:976] (1/7) Epoch 3, batch 4850, loss[loss=0.2425, simple_loss=0.3028, pruned_loss=0.09107, over 4824.00 frames. ], tot_loss[loss=0.2414, simple_loss=0.2945, pruned_loss=0.09414, over 952179.48 frames. ], batch size: 33, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:11:10,460 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7364, 1.6389, 1.3686, 1.7000, 1.8801, 1.4893, 2.1914, 1.7330], + device='cuda:1'), covar=tensor([0.2249, 0.3957, 0.4520, 0.3905, 0.3202, 0.2420, 0.3465, 0.3003], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0195, 0.0238, 0.0254, 0.0221, 0.0185, 0.0209, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:11:16,300 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4873, 2.4973, 2.1191, 1.8126, 2.5475, 2.8851, 2.8384, 2.3194], + device='cuda:1'), covar=tensor([0.0281, 0.0346, 0.0465, 0.0415, 0.0245, 0.0391, 0.0235, 0.0354], + device='cuda:1'), in_proj_covar=tensor([0.0085, 0.0114, 0.0138, 0.0118, 0.0105, 0.0100, 0.0091, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.6180e-05, 9.0517e-05, 1.1089e-04, 9.3336e-05, 8.3497e-05, 7.4228e-05, + 7.0179e-05, 8.5102e-05], device='cuda:1') +2023-03-26 02:11:18,710 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.278e+02 1.868e+02 2.289e+02 2.628e+02 4.977e+02, threshold=4.577e+02, percent-clipped=4.0 +2023-03-26 02:11:52,996 INFO [finetune.py:976] (1/7) Epoch 3, batch 4900, loss[loss=0.2507, simple_loss=0.3115, pruned_loss=0.09489, over 4916.00 frames. ], tot_loss[loss=0.2446, simple_loss=0.2972, pruned_loss=0.09598, over 953522.44 frames. ], batch size: 42, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:11:53,738 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16357.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:12:04,199 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16371.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:12:35,109 INFO [finetune.py:976] (1/7) Epoch 3, batch 4950, loss[loss=0.2071, simple_loss=0.2772, pruned_loss=0.0685, over 4135.00 frames. ], tot_loss[loss=0.2466, simple_loss=0.2999, pruned_loss=0.09666, over 955034.13 frames. ], batch size: 66, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:12:43,744 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.165e+02 1.793e+02 2.170e+02 2.564e+02 4.726e+02, threshold=4.340e+02, percent-clipped=1.0 +2023-03-26 02:12:53,401 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16432.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:12:59,269 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16441.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:13:03,666 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-03-26 02:13:11,532 INFO [finetune.py:976] (1/7) Epoch 3, batch 5000, loss[loss=0.2685, simple_loss=0.3185, pruned_loss=0.1092, over 4879.00 frames. ], tot_loss[loss=0.2452, simple_loss=0.2986, pruned_loss=0.09587, over 953616.87 frames. ], batch size: 32, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:13:25,355 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5215, 1.3937, 1.3985, 1.4230, 1.6552, 1.3403, 1.7328, 1.5059], + device='cuda:1'), covar=tensor([0.1916, 0.3102, 0.3678, 0.2851, 0.2714, 0.2003, 0.2923, 0.2411], + device='cuda:1'), in_proj_covar=tensor([0.0163, 0.0194, 0.0236, 0.0253, 0.0220, 0.0185, 0.0208, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:14:08,364 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16505.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:14:08,893 INFO [finetune.py:976] (1/7) Epoch 3, batch 5050, loss[loss=0.1804, simple_loss=0.2324, pruned_loss=0.06423, over 4752.00 frames. ], tot_loss[loss=0.242, simple_loss=0.2945, pruned_loss=0.09471, over 951687.31 frames. ], batch size: 23, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:14:27,727 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.054e+02 1.680e+02 2.024e+02 2.446e+02 4.498e+02, threshold=4.048e+02, percent-clipped=1.0 +2023-03-26 02:14:49,234 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16539.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:14:58,118 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5037, 1.3467, 1.2388, 1.5061, 1.6835, 1.5506, 0.7499, 1.2614], + device='cuda:1'), covar=tensor([0.2589, 0.2494, 0.2198, 0.1895, 0.1794, 0.1271, 0.3329, 0.2039], + device='cuda:1'), in_proj_covar=tensor([0.0229, 0.0207, 0.0196, 0.0181, 0.0231, 0.0171, 0.0212, 0.0185], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:15:09,829 INFO [finetune.py:976] (1/7) Epoch 3, batch 5100, loss[loss=0.2596, simple_loss=0.2901, pruned_loss=0.1145, over 4761.00 frames. ], tot_loss[loss=0.2382, simple_loss=0.2906, pruned_loss=0.09293, over 952584.90 frames. ], batch size: 27, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:15:36,550 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=16587.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:15:37,359 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 02:15:52,953 INFO [finetune.py:976] (1/7) Epoch 3, batch 5150, loss[loss=0.2337, simple_loss=0.3005, pruned_loss=0.08351, over 4902.00 frames. ], tot_loss[loss=0.2378, simple_loss=0.2902, pruned_loss=0.09269, over 953594.67 frames. ], batch size: 32, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:16:12,055 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.182e+02 1.762e+02 2.113e+02 2.590e+02 4.768e+02, threshold=4.226e+02, percent-clipped=2.0 +2023-03-26 02:16:47,967 INFO [finetune.py:976] (1/7) Epoch 3, batch 5200, loss[loss=0.1995, simple_loss=0.2613, pruned_loss=0.06887, over 4749.00 frames. ], tot_loss[loss=0.2401, simple_loss=0.2938, pruned_loss=0.09323, over 953573.09 frames. ], batch size: 27, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:16:48,671 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16657.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:17:40,756 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=16705.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:17:41,306 INFO [finetune.py:976] (1/7) Epoch 3, batch 5250, loss[loss=0.1936, simple_loss=0.2681, pruned_loss=0.05952, over 4794.00 frames. ], tot_loss[loss=0.2414, simple_loss=0.2954, pruned_loss=0.09369, over 954987.58 frames. ], batch size: 29, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:17:42,773 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 02:17:53,306 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.211e+02 1.833e+02 2.103e+02 2.647e+02 4.683e+02, threshold=4.205e+02, percent-clipped=1.0 +2023-03-26 02:17:58,302 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4732, 1.4608, 2.0887, 1.7558, 1.8266, 4.1823, 1.2967, 1.7305], + device='cuda:1'), covar=tensor([0.1177, 0.1975, 0.1257, 0.1229, 0.1780, 0.0207, 0.1781, 0.2031], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0078, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 02:18:00,082 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16727.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:18:12,254 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16741.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:18:21,889 INFO [finetune.py:976] (1/7) Epoch 3, batch 5300, loss[loss=0.2617, simple_loss=0.3109, pruned_loss=0.1062, over 4758.00 frames. ], tot_loss[loss=0.2426, simple_loss=0.2966, pruned_loss=0.09428, over 956184.15 frames. ], batch size: 54, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:18:46,985 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5423, 1.4243, 1.3739, 1.6296, 2.0040, 1.6057, 1.0505, 1.2737], + device='cuda:1'), covar=tensor([0.2730, 0.2685, 0.2388, 0.2053, 0.2180, 0.1394, 0.3499, 0.2215], + device='cuda:1'), in_proj_covar=tensor([0.0230, 0.0207, 0.0196, 0.0181, 0.0231, 0.0172, 0.0212, 0.0185], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:18:49,197 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=16789.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:19:07,726 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16805.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:19:07,955 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-26 02:19:08,237 INFO [finetune.py:976] (1/7) Epoch 3, batch 5350, loss[loss=0.2195, simple_loss=0.2758, pruned_loss=0.08163, over 4767.00 frames. ], tot_loss[loss=0.2419, simple_loss=0.2963, pruned_loss=0.09376, over 955073.27 frames. ], batch size: 26, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:19:21,973 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.184e+02 1.880e+02 2.227e+02 2.511e+02 3.677e+02, threshold=4.454e+02, percent-clipped=0.0 +2023-03-26 02:19:26,998 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9501, 1.8336, 1.8164, 1.9627, 1.3533, 4.5688, 1.7515, 2.3873], + device='cuda:1'), covar=tensor([0.3530, 0.2449, 0.2094, 0.2328, 0.1960, 0.0118, 0.2719, 0.1417], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0112, 0.0116, 0.0120, 0.0116, 0.0097, 0.0101, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 02:19:46,295 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=16853.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:19:48,084 INFO [finetune.py:976] (1/7) Epoch 3, batch 5400, loss[loss=0.2211, simple_loss=0.2733, pruned_loss=0.08442, over 4863.00 frames. ], tot_loss[loss=0.2395, simple_loss=0.2934, pruned_loss=0.09275, over 954750.41 frames. ], batch size: 31, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:19:53,079 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1365, 2.3320, 2.2700, 1.2065, 2.5497, 2.1988, 1.9701, 2.1676], + device='cuda:1'), covar=tensor([0.0865, 0.1212, 0.2138, 0.2773, 0.1938, 0.2310, 0.2367, 0.1487], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0199, 0.0204, 0.0191, 0.0217, 0.0211, 0.0218, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:19:55,564 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16868.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 02:20:30,322 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9797, 4.3599, 4.5137, 4.8155, 4.7075, 4.4646, 5.1029, 1.6748], + device='cuda:1'), covar=tensor([0.0765, 0.0805, 0.0637, 0.0906, 0.1164, 0.1350, 0.0524, 0.4943], + device='cuda:1'), in_proj_covar=tensor([0.0365, 0.0245, 0.0278, 0.0296, 0.0342, 0.0289, 0.0311, 0.0301], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:20:31,376 INFO [finetune.py:976] (1/7) Epoch 3, batch 5450, loss[loss=0.2275, simple_loss=0.2781, pruned_loss=0.0885, over 4829.00 frames. ], tot_loss[loss=0.2368, simple_loss=0.2901, pruned_loss=0.09175, over 954605.60 frames. ], batch size: 39, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:20:31,488 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16906.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:20:38,653 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.130e+02 1.665e+02 2.000e+02 2.450e+02 4.433e+02, threshold=4.000e+02, percent-clipped=0.0 +2023-03-26 02:20:46,017 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16929.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 02:21:01,714 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2259, 1.3242, 1.5727, 1.1667, 1.1785, 1.4107, 1.3143, 1.5855], + device='cuda:1'), covar=tensor([0.1347, 0.2158, 0.1281, 0.1442, 0.1128, 0.1361, 0.2696, 0.0983], + device='cuda:1'), in_proj_covar=tensor([0.0209, 0.0209, 0.0206, 0.0200, 0.0185, 0.0228, 0.0219, 0.0207], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:21:14,236 INFO [finetune.py:976] (1/7) Epoch 3, batch 5500, loss[loss=0.1727, simple_loss=0.2315, pruned_loss=0.05699, over 4695.00 frames. ], tot_loss[loss=0.2324, simple_loss=0.2855, pruned_loss=0.08964, over 952605.52 frames. ], batch size: 23, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:21:16,814 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7719, 1.5606, 1.5216, 1.7703, 2.0419, 1.7893, 1.0915, 1.4733], + device='cuda:1'), covar=tensor([0.2304, 0.2282, 0.2003, 0.1790, 0.1731, 0.1202, 0.2997, 0.1899], + device='cuda:1'), in_proj_covar=tensor([0.0230, 0.0208, 0.0197, 0.0182, 0.0232, 0.0172, 0.0212, 0.0186], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:21:26,195 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16967.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 02:22:07,776 INFO [finetune.py:976] (1/7) Epoch 3, batch 5550, loss[loss=0.2061, simple_loss=0.2782, pruned_loss=0.067, over 4746.00 frames. ], tot_loss[loss=0.2373, simple_loss=0.2901, pruned_loss=0.09227, over 953248.77 frames. ], batch size: 27, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:22:15,686 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.144e+02 1.712e+02 2.015e+02 2.380e+02 4.122e+02, threshold=4.030e+02, percent-clipped=1.0 +2023-03-26 02:22:21,768 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17027.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:22:42,646 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.7048, 1.5296, 1.5698, 0.9657, 1.7031, 1.9241, 1.8208, 1.4833], + device='cuda:1'), covar=tensor([0.1030, 0.0717, 0.0464, 0.0650, 0.0397, 0.0393, 0.0399, 0.0513], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0157, 0.0117, 0.0137, 0.0132, 0.0121, 0.0146, 0.0144], + device='cuda:1'), out_proj_covar=tensor([9.7934e-05, 1.1710e-04, 8.5148e-05, 1.0043e-04, 9.5735e-05, 8.9613e-05, + 1.0891e-04, 1.0696e-04], device='cuda:1') +2023-03-26 02:22:53,359 INFO [finetune.py:976] (1/7) Epoch 3, batch 5600, loss[loss=0.1971, simple_loss=0.2563, pruned_loss=0.06901, over 4785.00 frames. ], tot_loss[loss=0.2397, simple_loss=0.2938, pruned_loss=0.09275, over 954471.28 frames. ], batch size: 23, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:23:06,876 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 02:23:10,677 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=17075.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:23:26,998 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7397, 1.4962, 1.2906, 1.2172, 1.4351, 1.4585, 1.3946, 2.2232], + device='cuda:1'), covar=tensor([0.9307, 0.9786, 0.7600, 0.9720, 0.8132, 0.5146, 0.9091, 0.3209], + device='cuda:1'), in_proj_covar=tensor([0.0274, 0.0250, 0.0219, 0.0284, 0.0235, 0.0196, 0.0240, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 02:23:38,817 INFO [finetune.py:976] (1/7) Epoch 3, batch 5650, loss[loss=0.3307, simple_loss=0.3731, pruned_loss=0.1442, over 4262.00 frames. ], tot_loss[loss=0.2421, simple_loss=0.297, pruned_loss=0.09367, over 953722.93 frames. ], batch size: 65, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:23:45,803 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.238e+02 1.755e+02 2.152e+02 2.681e+02 4.789e+02, threshold=4.305e+02, percent-clipped=1.0 +2023-03-26 02:24:15,373 INFO [finetune.py:976] (1/7) Epoch 3, batch 5700, loss[loss=0.1977, simple_loss=0.2445, pruned_loss=0.07547, over 4175.00 frames. ], tot_loss[loss=0.238, simple_loss=0.2912, pruned_loss=0.09237, over 933475.09 frames. ], batch size: 18, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:24:57,704 INFO [finetune.py:976] (1/7) Epoch 4, batch 0, loss[loss=0.2481, simple_loss=0.3059, pruned_loss=0.09515, over 4819.00 frames. ], tot_loss[loss=0.2481, simple_loss=0.3059, pruned_loss=0.09515, over 4819.00 frames. ], batch size: 33, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:24:57,705 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 02:25:05,117 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.2164, 1.3864, 1.4757, 0.7978, 1.3073, 1.6003, 1.6533, 1.3995], + device='cuda:1'), covar=tensor([0.0974, 0.0590, 0.0435, 0.0533, 0.0415, 0.0468, 0.0324, 0.0565], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0157, 0.0117, 0.0136, 0.0132, 0.0121, 0.0146, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.8181e-05, 1.1719e-04, 8.5409e-05, 1.0017e-04, 9.5355e-05, 9.0035e-05, + 1.0877e-04, 1.0736e-04], device='cuda:1') +2023-03-26 02:25:18,215 INFO [finetune.py:1010] (1/7) Epoch 4, validation: loss=0.1768, simple_loss=0.2473, pruned_loss=0.0532, over 2265189.00 frames. +2023-03-26 02:25:18,216 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6289MB +2023-03-26 02:25:22,710 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17189.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:25:30,216 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0 +2023-03-26 02:25:55,810 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.188e+02 1.713e+02 2.128e+02 2.708e+02 4.853e+02, threshold=4.257e+02, percent-clipped=3.0 +2023-03-26 02:26:00,043 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17224.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 02:26:05,895 INFO [finetune.py:976] (1/7) Epoch 4, batch 50, loss[loss=0.2573, simple_loss=0.2933, pruned_loss=0.1107, over 4363.00 frames. ], tot_loss[loss=0.2446, simple_loss=0.298, pruned_loss=0.09564, over 217407.66 frames. ], batch size: 19, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:26:29,241 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17250.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:26:36,457 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17262.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 02:26:36,523 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6714, 1.4137, 1.3479, 1.5579, 2.0500, 1.5744, 1.0832, 1.4115], + device='cuda:1'), covar=tensor([0.2284, 0.2431, 0.2195, 0.1961, 0.1830, 0.1424, 0.3123, 0.1937], + device='cuda:1'), in_proj_covar=tensor([0.0232, 0.0209, 0.0198, 0.0183, 0.0233, 0.0173, 0.0214, 0.0186], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:26:55,374 INFO [finetune.py:976] (1/7) Epoch 4, batch 100, loss[loss=0.315, simple_loss=0.3151, pruned_loss=0.1574, over 4305.00 frames. ], tot_loss[loss=0.2399, simple_loss=0.2922, pruned_loss=0.09383, over 380214.67 frames. ], batch size: 65, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:27:20,977 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4932, 1.7252, 2.0915, 1.8128, 1.8793, 4.2916, 1.4879, 1.9943], + device='cuda:1'), covar=tensor([0.1039, 0.1614, 0.1089, 0.1102, 0.1501, 0.0183, 0.1432, 0.1609], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0077, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 02:27:26,862 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.228e+02 1.697e+02 1.982e+02 2.273e+02 3.827e+02, threshold=3.964e+02, percent-clipped=0.0 +2023-03-26 02:27:34,323 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8679, 1.3132, 1.1357, 1.8065, 2.1344, 1.5377, 1.4929, 1.8182], + device='cuda:1'), covar=tensor([0.1223, 0.1798, 0.1941, 0.0980, 0.1728, 0.2031, 0.1296, 0.1551], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0099, 0.0118, 0.0095, 0.0125, 0.0098, 0.0101, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 02:27:35,551 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2905, 2.7104, 2.4994, 1.4244, 2.7165, 2.3351, 2.2093, 2.3722], + device='cuda:1'), covar=tensor([0.1065, 0.1082, 0.2195, 0.2989, 0.2192, 0.2542, 0.2434, 0.1462], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0200, 0.0205, 0.0191, 0.0217, 0.0210, 0.0219, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:27:36,644 INFO [finetune.py:976] (1/7) Epoch 4, batch 150, loss[loss=0.2175, simple_loss=0.2769, pruned_loss=0.079, over 4846.00 frames. ], tot_loss[loss=0.2337, simple_loss=0.2856, pruned_loss=0.09089, over 505676.51 frames. ], batch size: 44, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:28:01,906 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17362.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:28:14,828 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6597, 1.4405, 1.3333, 1.3129, 1.7412, 1.3983, 1.7630, 1.5687], + device='cuda:1'), covar=tensor([0.1987, 0.3664, 0.4585, 0.3770, 0.3348, 0.2311, 0.3515, 0.2871], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0195, 0.0238, 0.0254, 0.0221, 0.0186, 0.0208, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:28:25,903 INFO [finetune.py:976] (1/7) Epoch 4, batch 200, loss[loss=0.2664, simple_loss=0.3109, pruned_loss=0.1109, over 4757.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.2857, pruned_loss=0.09194, over 605477.26 frames. ], batch size: 54, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:28:26,031 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17383.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 02:28:27,311 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6547, 1.6334, 1.4452, 1.6885, 2.2754, 1.7681, 1.2985, 1.3562], + device='cuda:1'), covar=tensor([0.2523, 0.2361, 0.2090, 0.2069, 0.1979, 0.1309, 0.3117, 0.2072], + device='cuda:1'), in_proj_covar=tensor([0.0232, 0.0209, 0.0198, 0.0183, 0.0233, 0.0173, 0.0214, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:28:55,784 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.314e+02 1.771e+02 2.098e+02 2.514e+02 4.657e+02, threshold=4.195e+02, percent-clipped=1.0 +2023-03-26 02:29:01,223 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17423.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:29:09,065 INFO [finetune.py:976] (1/7) Epoch 4, batch 250, loss[loss=0.2145, simple_loss=0.2785, pruned_loss=0.0752, over 4822.00 frames. ], tot_loss[loss=0.2369, simple_loss=0.2893, pruned_loss=0.09225, over 682639.04 frames. ], batch size: 33, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:29:17,834 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17444.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 02:29:49,004 INFO [finetune.py:976] (1/7) Epoch 4, batch 300, loss[loss=0.2198, simple_loss=0.2819, pruned_loss=0.07889, over 4921.00 frames. ], tot_loss[loss=0.2408, simple_loss=0.2933, pruned_loss=0.09418, over 740468.39 frames. ], batch size: 33, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:30:35,040 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.302e+02 1.964e+02 2.271e+02 2.699e+02 6.272e+02, threshold=4.542e+02, percent-clipped=2.0 +2023-03-26 02:30:39,312 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17524.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 02:30:44,607 INFO [finetune.py:976] (1/7) Epoch 4, batch 350, loss[loss=0.2873, simple_loss=0.3256, pruned_loss=0.1245, over 4885.00 frames. ], tot_loss[loss=0.2436, simple_loss=0.2961, pruned_loss=0.09554, over 787942.62 frames. ], batch size: 32, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:30:53,137 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17545.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:31:09,672 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5521, 1.3639, 1.1515, 1.0076, 1.3288, 1.3029, 1.2800, 1.9772], + device='cuda:1'), covar=tensor([0.9300, 0.8488, 0.6872, 0.8841, 0.7171, 0.4774, 0.7805, 0.3214], + device='cuda:1'), in_proj_covar=tensor([0.0274, 0.0250, 0.0218, 0.0283, 0.0235, 0.0196, 0.0239, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 02:31:10,218 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17562.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 02:31:16,312 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=17572.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 02:31:23,349 INFO [finetune.py:976] (1/7) Epoch 4, batch 400, loss[loss=0.2141, simple_loss=0.2765, pruned_loss=0.07585, over 4730.00 frames. ], tot_loss[loss=0.2435, simple_loss=0.2968, pruned_loss=0.09515, over 822844.98 frames. ], batch size: 54, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:31:34,145 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 02:31:45,719 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7076, 1.5122, 1.4411, 1.7191, 2.1522, 1.7601, 1.3199, 1.3626], + device='cuda:1'), covar=tensor([0.2344, 0.2365, 0.2047, 0.1846, 0.1945, 0.1266, 0.3007, 0.1918], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0208, 0.0198, 0.0183, 0.0232, 0.0173, 0.0214, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:31:46,282 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=17610.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:31:51,079 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.188e+02 1.790e+02 1.987e+02 2.567e+02 5.687e+02, threshold=3.975e+02, percent-clipped=1.0 +2023-03-26 02:32:10,368 INFO [finetune.py:976] (1/7) Epoch 4, batch 450, loss[loss=0.2462, simple_loss=0.2978, pruned_loss=0.09726, over 4813.00 frames. ], tot_loss[loss=0.2398, simple_loss=0.2933, pruned_loss=0.09317, over 852287.37 frames. ], batch size: 30, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:32:30,533 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4831, 1.3760, 1.9147, 2.8511, 2.0123, 2.0676, 1.1058, 2.2415], + device='cuda:1'), covar=tensor([0.1828, 0.1570, 0.1209, 0.0603, 0.0839, 0.1482, 0.1716, 0.0695], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0119, 0.0137, 0.0165, 0.0104, 0.0143, 0.0129, 0.0106], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 02:32:55,555 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5757, 1.5418, 2.1420, 3.3761, 2.4507, 2.3242, 1.0794, 2.6564], + device='cuda:1'), covar=tensor([0.1817, 0.1549, 0.1315, 0.0540, 0.0748, 0.1405, 0.1906, 0.0578], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0119, 0.0137, 0.0165, 0.0104, 0.0143, 0.0129, 0.0106], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 02:33:00,226 INFO [finetune.py:976] (1/7) Epoch 4, batch 500, loss[loss=0.2297, simple_loss=0.2812, pruned_loss=0.08912, over 4903.00 frames. ], tot_loss[loss=0.2383, simple_loss=0.2911, pruned_loss=0.09272, over 875983.89 frames. ], batch size: 43, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:33:11,003 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17700.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:33:20,382 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 02:33:24,343 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.264e+02 1.808e+02 2.091e+02 2.485e+02 4.480e+02, threshold=4.181e+02, percent-clipped=1.0 +2023-03-26 02:33:24,430 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17718.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:33:31,866 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17730.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:33:34,059 INFO [finetune.py:976] (1/7) Epoch 4, batch 550, loss[loss=0.3033, simple_loss=0.3205, pruned_loss=0.1431, over 4056.00 frames. ], tot_loss[loss=0.235, simple_loss=0.2875, pruned_loss=0.09126, over 893486.06 frames. ], batch size: 65, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:33:37,769 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17739.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 02:33:42,039 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-03-26 02:34:03,948 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17761.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:34:17,632 INFO [finetune.py:976] (1/7) Epoch 4, batch 600, loss[loss=0.2123, simple_loss=0.2644, pruned_loss=0.08005, over 4763.00 frames. ], tot_loss[loss=0.2349, simple_loss=0.2876, pruned_loss=0.09112, over 904788.09 frames. ], batch size: 26, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:34:22,569 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17791.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:34:41,399 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.036e+02 1.729e+02 2.101e+02 2.480e+02 7.519e+02, threshold=4.202e+02, percent-clipped=3.0 +2023-03-26 02:34:44,533 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5222, 1.4369, 1.4346, 1.4600, 1.1323, 2.9248, 1.0985, 1.5264], + device='cuda:1'), covar=tensor([0.3863, 0.2666, 0.2211, 0.2597, 0.2049, 0.0264, 0.2742, 0.1465], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0112, 0.0116, 0.0120, 0.0116, 0.0097, 0.0101, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 02:34:50,414 INFO [finetune.py:976] (1/7) Epoch 4, batch 650, loss[loss=0.2389, simple_loss=0.2866, pruned_loss=0.09561, over 4814.00 frames. ], tot_loss[loss=0.24, simple_loss=0.2931, pruned_loss=0.09342, over 917274.60 frames. ], batch size: 30, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:34:59,953 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17845.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:35:20,086 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6713, 1.5036, 2.1102, 1.2997, 1.7386, 1.8022, 1.3903, 2.0615], + device='cuda:1'), covar=tensor([0.1877, 0.2414, 0.1531, 0.2256, 0.1197, 0.1731, 0.2921, 0.1200], + device='cuda:1'), in_proj_covar=tensor([0.0208, 0.0209, 0.0206, 0.0200, 0.0184, 0.0228, 0.0218, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:35:31,912 INFO [finetune.py:976] (1/7) Epoch 4, batch 700, loss[loss=0.2471, simple_loss=0.3017, pruned_loss=0.09623, over 4737.00 frames. ], tot_loss[loss=0.2419, simple_loss=0.2952, pruned_loss=0.0943, over 926016.21 frames. ], batch size: 27, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:35:46,369 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=17893.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:36:18,034 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.240e+02 1.804e+02 2.027e+02 2.461e+02 4.855e+02, threshold=4.055e+02, percent-clipped=2.0 +2023-03-26 02:36:34,754 INFO [finetune.py:976] (1/7) Epoch 4, batch 750, loss[loss=0.2969, simple_loss=0.3323, pruned_loss=0.1307, over 4859.00 frames. ], tot_loss[loss=0.2414, simple_loss=0.2953, pruned_loss=0.09379, over 929257.26 frames. ], batch size: 31, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:36:59,898 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 02:37:08,574 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17960.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:37:30,395 INFO [finetune.py:976] (1/7) Epoch 4, batch 800, loss[loss=0.1892, simple_loss=0.2572, pruned_loss=0.06057, over 4831.00 frames. ], tot_loss[loss=0.2399, simple_loss=0.2945, pruned_loss=0.09263, over 932809.60 frames. ], batch size: 33, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:37:57,545 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-03-26 02:38:02,706 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5225, 1.0575, 0.7652, 1.3464, 1.9608, 0.6839, 1.1897, 1.4857], + device='cuda:1'), covar=tensor([0.1468, 0.2045, 0.1977, 0.1243, 0.1911, 0.2027, 0.1463, 0.1845], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0099, 0.0117, 0.0094, 0.0126, 0.0098, 0.0101, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 02:38:05,563 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.155e+02 1.784e+02 2.191e+02 2.808e+02 5.190e+02, threshold=4.382e+02, percent-clipped=3.0 +2023-03-26 02:38:05,658 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18018.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:38:08,533 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18021.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:38:20,833 INFO [finetune.py:976] (1/7) Epoch 4, batch 850, loss[loss=0.255, simple_loss=0.3046, pruned_loss=0.1027, over 4930.00 frames. ], tot_loss[loss=0.2372, simple_loss=0.2918, pruned_loss=0.09133, over 937620.66 frames. ], batch size: 38, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:38:26,578 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18039.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 02:38:39,164 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18056.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:38:45,698 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=18066.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:38:57,811 INFO [finetune.py:976] (1/7) Epoch 4, batch 900, loss[loss=0.166, simple_loss=0.2332, pruned_loss=0.04937, over 4844.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.2886, pruned_loss=0.09028, over 943103.37 frames. ], batch size: 33, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:38:59,681 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18086.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:39:00,271 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=18087.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 02:39:19,669 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.218e+02 1.720e+02 1.921e+02 2.312e+02 4.297e+02, threshold=3.842e+02, percent-clipped=0.0 +2023-03-26 02:39:35,995 INFO [finetune.py:976] (1/7) Epoch 4, batch 950, loss[loss=0.182, simple_loss=0.253, pruned_loss=0.05549, over 4787.00 frames. ], tot_loss[loss=0.2325, simple_loss=0.2867, pruned_loss=0.08916, over 948177.18 frames. ], batch size: 25, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:40:22,962 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18174.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:40:23,593 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18175.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:40:29,235 INFO [finetune.py:976] (1/7) Epoch 4, batch 1000, loss[loss=0.1954, simple_loss=0.2543, pruned_loss=0.06829, over 4764.00 frames. ], tot_loss[loss=0.2354, simple_loss=0.2897, pruned_loss=0.09059, over 949435.25 frames. ], batch size: 27, lr: 3.97e-03, grad_scale: 64.0 +2023-03-26 02:40:33,667 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9942, 2.3024, 2.2967, 1.3385, 2.5082, 2.0058, 1.7367, 2.1206], + device='cuda:1'), covar=tensor([0.1116, 0.1167, 0.2082, 0.2966, 0.2067, 0.2672, 0.2638, 0.1675], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0199, 0.0203, 0.0190, 0.0217, 0.0210, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:40:44,154 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8551, 1.7338, 1.3553, 1.5432, 1.6837, 1.6091, 1.6262, 2.4191], + device='cuda:1'), covar=tensor([0.9380, 1.0016, 0.7436, 1.0022, 0.7873, 0.5051, 0.8996, 0.3155], + device='cuda:1'), in_proj_covar=tensor([0.0275, 0.0250, 0.0218, 0.0283, 0.0235, 0.0196, 0.0239, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 02:40:54,755 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2744, 2.9183, 2.9911, 3.1897, 3.0275, 2.8593, 3.3269, 1.0654], + device='cuda:1'), covar=tensor([0.1091, 0.0960, 0.1030, 0.1277, 0.1726, 0.1670, 0.1102, 0.4930], + device='cuda:1'), in_proj_covar=tensor([0.0364, 0.0247, 0.0282, 0.0297, 0.0345, 0.0288, 0.0312, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:41:07,306 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.077e+02 1.690e+02 2.099e+02 2.471e+02 3.966e+02, threshold=4.198e+02, percent-clipped=1.0 +2023-03-26 02:41:28,636 INFO [finetune.py:976] (1/7) Epoch 4, batch 1050, loss[loss=0.2809, simple_loss=0.3417, pruned_loss=0.11, over 4837.00 frames. ], tot_loss[loss=0.2363, simple_loss=0.2915, pruned_loss=0.09054, over 952181.02 frames. ], batch size: 47, lr: 3.97e-03, grad_scale: 64.0 +2023-03-26 02:41:29,950 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18235.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:41:30,598 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18236.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:41:33,076 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9201, 1.8470, 1.4991, 1.9259, 1.7894, 1.7127, 1.6716, 2.7450], + device='cuda:1'), covar=tensor([1.0008, 1.1814, 0.8101, 1.2009, 0.9445, 0.5545, 1.0537, 0.3181], + device='cuda:1'), in_proj_covar=tensor([0.0274, 0.0250, 0.0218, 0.0282, 0.0234, 0.0196, 0.0238, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 02:42:18,623 INFO [finetune.py:976] (1/7) Epoch 4, batch 1100, loss[loss=0.2504, simple_loss=0.3111, pruned_loss=0.09488, over 4805.00 frames. ], tot_loss[loss=0.2386, simple_loss=0.2936, pruned_loss=0.09177, over 953622.13 frames. ], batch size: 41, lr: 3.97e-03, grad_scale: 64.0 +2023-03-26 02:42:39,972 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4690, 1.3335, 1.3644, 1.3784, 0.7860, 2.3613, 0.7444, 1.1435], + device='cuda:1'), covar=tensor([0.3421, 0.2499, 0.2097, 0.2269, 0.2221, 0.0345, 0.2679, 0.1441], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0111, 0.0116, 0.0119, 0.0115, 0.0096, 0.0100, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 02:42:49,579 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18316.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:42:50,749 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.170e+02 1.851e+02 2.220e+02 2.754e+02 4.687e+02, threshold=4.440e+02, percent-clipped=1.0 +2023-03-26 02:43:08,647 INFO [finetune.py:976] (1/7) Epoch 4, batch 1150, loss[loss=0.2301, simple_loss=0.2866, pruned_loss=0.08686, over 4816.00 frames. ], tot_loss[loss=0.2401, simple_loss=0.2949, pruned_loss=0.09271, over 954685.47 frames. ], batch size: 40, lr: 3.97e-03, grad_scale: 64.0 +2023-03-26 02:43:14,519 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-03-26 02:43:19,807 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1188, 3.5491, 3.7665, 3.9940, 3.8241, 3.5786, 4.1639, 1.2988], + device='cuda:1'), covar=tensor([0.0647, 0.0736, 0.0745, 0.0833, 0.1090, 0.1339, 0.0651, 0.4936], + device='cuda:1'), in_proj_covar=tensor([0.0362, 0.0245, 0.0279, 0.0295, 0.0342, 0.0286, 0.0310, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:43:21,093 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6681, 2.1575, 2.8742, 1.8973, 2.6932, 2.7334, 2.2121, 3.0199], + device='cuda:1'), covar=tensor([0.1506, 0.2265, 0.1573, 0.2545, 0.1132, 0.2068, 0.2479, 0.1085], + device='cuda:1'), in_proj_covar=tensor([0.0206, 0.0207, 0.0203, 0.0198, 0.0182, 0.0227, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:43:25,369 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18356.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:44:03,369 INFO [finetune.py:976] (1/7) Epoch 4, batch 1200, loss[loss=0.2343, simple_loss=0.2905, pruned_loss=0.08905, over 4812.00 frames. ], tot_loss[loss=0.2389, simple_loss=0.2934, pruned_loss=0.09216, over 955543.81 frames. ], batch size: 40, lr: 3.97e-03, grad_scale: 64.0 +2023-03-26 02:44:05,828 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18386.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:44:16,235 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.35 vs. limit=5.0 +2023-03-26 02:44:28,300 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=18404.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:44:36,919 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8897, 1.6829, 2.0568, 1.5955, 1.9080, 1.9887, 1.6314, 2.2885], + device='cuda:1'), covar=tensor([0.1408, 0.2088, 0.1421, 0.1751, 0.0912, 0.1574, 0.2556, 0.0858], + device='cuda:1'), in_proj_covar=tensor([0.0207, 0.0207, 0.0204, 0.0198, 0.0182, 0.0227, 0.0217, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:44:47,472 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.754e+02 2.082e+02 2.492e+02 3.668e+02, threshold=4.164e+02, percent-clipped=0.0 +2023-03-26 02:44:58,265 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-26 02:45:07,911 INFO [finetune.py:976] (1/7) Epoch 4, batch 1250, loss[loss=0.2164, simple_loss=0.2708, pruned_loss=0.08101, over 4899.00 frames. ], tot_loss[loss=0.2351, simple_loss=0.2896, pruned_loss=0.09033, over 956962.54 frames. ], batch size: 32, lr: 3.97e-03, grad_scale: 64.0 +2023-03-26 02:45:09,085 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=18434.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:45:19,783 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8565, 1.6035, 2.2323, 1.5414, 1.9292, 1.9918, 1.6847, 2.3558], + device='cuda:1'), covar=tensor([0.1677, 0.2331, 0.1486, 0.2087, 0.1062, 0.1800, 0.2686, 0.0951], + device='cuda:1'), in_proj_covar=tensor([0.0206, 0.0207, 0.0203, 0.0197, 0.0181, 0.0227, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:45:36,147 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6574, 0.5410, 1.5521, 1.3578, 1.3621, 1.3298, 1.1794, 1.5356], + device='cuda:1'), covar=tensor([0.7626, 1.1912, 0.9451, 1.0396, 1.1162, 0.8114, 1.3257, 0.8172], + device='cuda:1'), in_proj_covar=tensor([0.0229, 0.0255, 0.0256, 0.0265, 0.0243, 0.0219, 0.0278, 0.0222], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:45:39,472 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-26 02:45:42,955 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5335, 1.6128, 1.7229, 0.9283, 1.6039, 1.8305, 1.8452, 1.5802], + device='cuda:1'), covar=tensor([0.1162, 0.0733, 0.0472, 0.0647, 0.0433, 0.0600, 0.0338, 0.0659], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0159, 0.0117, 0.0138, 0.0134, 0.0122, 0.0147, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.8561e-05, 1.1789e-04, 8.5355e-05, 1.0134e-04, 9.6569e-05, 9.0827e-05, + 1.0928e-04, 1.0748e-04], device='cuda:1') +2023-03-26 02:45:59,525 INFO [finetune.py:976] (1/7) Epoch 4, batch 1300, loss[loss=0.2137, simple_loss=0.2596, pruned_loss=0.0839, over 3990.00 frames. ], tot_loss[loss=0.2315, simple_loss=0.2855, pruned_loss=0.08873, over 954963.15 frames. ], batch size: 17, lr: 3.97e-03, grad_scale: 64.0 +2023-03-26 02:46:00,922 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.76 vs. limit=5.0 +2023-03-26 02:46:10,683 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18490.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:46:21,375 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7829, 4.3063, 4.0114, 2.1226, 4.2852, 3.2194, 0.9817, 3.0443], + device='cuda:1'), covar=tensor([0.2831, 0.1618, 0.1509, 0.3256, 0.0858, 0.0983, 0.4500, 0.1381], + device='cuda:1'), in_proj_covar=tensor([0.0155, 0.0169, 0.0163, 0.0129, 0.0155, 0.0121, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 02:46:22,656 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18508.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 02:46:29,709 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.069e+02 1.658e+02 2.008e+02 2.632e+02 4.281e+02, threshold=4.017e+02, percent-clipped=1.0 +2023-03-26 02:46:36,933 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18530.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:46:37,494 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18531.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:46:41,958 INFO [finetune.py:976] (1/7) Epoch 4, batch 1350, loss[loss=0.2871, simple_loss=0.3398, pruned_loss=0.1172, over 4733.00 frames. ], tot_loss[loss=0.2331, simple_loss=0.2866, pruned_loss=0.08979, over 952836.14 frames. ], batch size: 59, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:46:51,612 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 02:46:55,204 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18551.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:47:01,305 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8195, 3.3162, 3.4551, 3.7179, 3.5394, 3.3750, 3.9091, 1.2583], + device='cuda:1'), covar=tensor([0.0931, 0.0887, 0.1011, 0.1101, 0.1413, 0.1647, 0.0896, 0.5333], + device='cuda:1'), in_proj_covar=tensor([0.0360, 0.0244, 0.0278, 0.0293, 0.0339, 0.0285, 0.0309, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:47:11,794 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18569.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 02:47:25,812 INFO [finetune.py:976] (1/7) Epoch 4, batch 1400, loss[loss=0.2729, simple_loss=0.3135, pruned_loss=0.1161, over 4831.00 frames. ], tot_loss[loss=0.2374, simple_loss=0.2915, pruned_loss=0.09167, over 952759.71 frames. ], batch size: 30, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:47:27,952 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 02:47:53,783 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18616.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:47:55,464 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.253e+02 1.863e+02 2.207e+02 2.712e+02 5.337e+02, threshold=4.415e+02, percent-clipped=2.0 +2023-03-26 02:48:10,033 INFO [finetune.py:976] (1/7) Epoch 4, batch 1450, loss[loss=0.2186, simple_loss=0.2733, pruned_loss=0.08198, over 4825.00 frames. ], tot_loss[loss=0.2379, simple_loss=0.2925, pruned_loss=0.09169, over 953298.54 frames. ], batch size: 25, lr: 3.97e-03, grad_scale: 32.0 +2023-03-26 02:48:42,793 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7051, 1.4777, 1.0416, 0.3608, 1.2601, 1.5038, 1.3472, 1.4086], + device='cuda:1'), covar=tensor([0.0908, 0.0811, 0.1441, 0.2057, 0.1376, 0.2242, 0.2208, 0.0827], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0200, 0.0203, 0.0191, 0.0218, 0.0211, 0.0219, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:48:43,321 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=18664.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:48:45,822 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0036, 1.8254, 1.6111, 1.9080, 2.0642, 1.7362, 2.4261, 2.0012], + device='cuda:1'), covar=tensor([0.1891, 0.3846, 0.4669, 0.3495, 0.3009, 0.2164, 0.3375, 0.2670], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0194, 0.0238, 0.0254, 0.0222, 0.0185, 0.0209, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:48:49,409 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18674.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:48:56,991 INFO [finetune.py:976] (1/7) Epoch 4, batch 1500, loss[loss=0.2123, simple_loss=0.2671, pruned_loss=0.07869, over 4744.00 frames. ], tot_loss[loss=0.2389, simple_loss=0.2938, pruned_loss=0.092, over 954443.29 frames. ], batch size: 28, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:49:36,303 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.377e+02 1.810e+02 2.122e+02 2.509e+02 6.153e+02, threshold=4.245e+02, percent-clipped=1.0 +2023-03-26 02:49:54,648 INFO [finetune.py:976] (1/7) Epoch 4, batch 1550, loss[loss=0.2764, simple_loss=0.3154, pruned_loss=0.1187, over 4793.00 frames. ], tot_loss[loss=0.2377, simple_loss=0.2926, pruned_loss=0.09137, over 952857.26 frames. ], batch size: 51, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:49:55,950 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18735.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:50:39,060 INFO [finetune.py:976] (1/7) Epoch 4, batch 1600, loss[loss=0.237, simple_loss=0.2815, pruned_loss=0.0962, over 4870.00 frames. ], tot_loss[loss=0.2349, simple_loss=0.2895, pruned_loss=0.09018, over 953507.51 frames. ], batch size: 31, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:51:19,525 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.176e+02 1.645e+02 2.011e+02 2.399e+02 5.772e+02, threshold=4.021e+02, percent-clipped=1.0 +2023-03-26 02:51:30,318 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18830.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:51:30,940 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18831.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:51:32,084 INFO [finetune.py:976] (1/7) Epoch 4, batch 1650, loss[loss=0.207, simple_loss=0.2621, pruned_loss=0.07597, over 4732.00 frames. ], tot_loss[loss=0.2325, simple_loss=0.2866, pruned_loss=0.08917, over 952958.37 frames. ], batch size: 59, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:51:39,804 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8297, 1.0609, 1.6217, 1.5528, 1.4758, 1.4606, 1.4607, 1.4965], + device='cuda:1'), covar=tensor([0.6811, 1.1130, 0.9092, 0.9844, 1.1087, 0.8253, 1.2346, 0.8577], + device='cuda:1'), in_proj_covar=tensor([0.0230, 0.0254, 0.0256, 0.0264, 0.0244, 0.0219, 0.0279, 0.0222], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:51:48,106 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18846.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:52:05,188 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18864.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 02:52:13,695 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2897, 2.9206, 3.0388, 3.1986, 3.0633, 2.8947, 3.3464, 0.9836], + device='cuda:1'), covar=tensor([0.1172, 0.1022, 0.1132, 0.1359, 0.1655, 0.1799, 0.1164, 0.5128], + device='cuda:1'), in_proj_covar=tensor([0.0361, 0.0244, 0.0277, 0.0293, 0.0337, 0.0286, 0.0309, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:52:23,039 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=18878.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:52:23,630 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=18879.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:52:25,971 INFO [finetune.py:976] (1/7) Epoch 4, batch 1700, loss[loss=0.2025, simple_loss=0.2567, pruned_loss=0.07416, over 4758.00 frames. ], tot_loss[loss=0.2288, simple_loss=0.283, pruned_loss=0.08732, over 952054.33 frames. ], batch size: 26, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:52:32,404 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8382, 1.2709, 1.6037, 1.5916, 1.4478, 1.4738, 1.5039, 1.5209], + device='cuda:1'), covar=tensor([0.7142, 1.0392, 0.8358, 0.9430, 1.0554, 0.7845, 1.2492, 0.7764], + device='cuda:1'), in_proj_covar=tensor([0.0230, 0.0254, 0.0256, 0.0264, 0.0243, 0.0219, 0.0278, 0.0222], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:53:00,726 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.073e+02 1.767e+02 2.149e+02 2.599e+02 5.673e+02, threshold=4.299e+02, percent-clipped=2.0 +2023-03-26 02:53:07,515 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18930.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:53:09,239 INFO [finetune.py:976] (1/7) Epoch 4, batch 1750, loss[loss=0.2909, simple_loss=0.3497, pruned_loss=0.116, over 4814.00 frames. ], tot_loss[loss=0.2322, simple_loss=0.2864, pruned_loss=0.08895, over 951224.51 frames. ], batch size: 45, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:53:43,500 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 02:53:52,365 INFO [finetune.py:976] (1/7) Epoch 4, batch 1800, loss[loss=0.2348, simple_loss=0.2857, pruned_loss=0.0919, over 4826.00 frames. ], tot_loss[loss=0.2336, simple_loss=0.2884, pruned_loss=0.08938, over 952754.18 frames. ], batch size: 30, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:53:54,978 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4373, 1.3883, 1.7060, 2.8313, 1.9073, 1.9536, 0.8236, 2.2083], + device='cuda:1'), covar=tensor([0.1966, 0.1663, 0.1377, 0.0691, 0.0889, 0.1794, 0.2013, 0.0759], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0119, 0.0137, 0.0167, 0.0104, 0.0143, 0.0129, 0.0106], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 02:53:55,120 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 02:53:57,473 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18991.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:53:57,639 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-03-26 02:54:32,253 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.317e+02 1.869e+02 2.115e+02 2.590e+02 5.981e+02, threshold=4.230e+02, percent-clipped=1.0 +2023-03-26 02:54:50,222 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19030.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:54:52,012 INFO [finetune.py:976] (1/7) Epoch 4, batch 1850, loss[loss=0.2265, simple_loss=0.2901, pruned_loss=0.08148, over 4898.00 frames. ], tot_loss[loss=0.2331, simple_loss=0.2885, pruned_loss=0.08888, over 952610.31 frames. ], batch size: 43, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:55:24,463 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3804, 1.3940, 1.9056, 1.8208, 1.7529, 3.7850, 1.3907, 1.6328], + device='cuda:1'), covar=tensor([0.1157, 0.1916, 0.1461, 0.1148, 0.1595, 0.0248, 0.1561, 0.1863], + device='cuda:1'), in_proj_covar=tensor([0.0079, 0.0082, 0.0078, 0.0080, 0.0093, 0.0083, 0.0086, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 02:55:39,892 INFO [finetune.py:976] (1/7) Epoch 4, batch 1900, loss[loss=0.2034, simple_loss=0.2672, pruned_loss=0.06985, over 4748.00 frames. ], tot_loss[loss=0.2353, simple_loss=0.291, pruned_loss=0.08986, over 953486.32 frames. ], batch size: 26, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:56:12,754 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.301e+02 1.752e+02 2.082e+02 2.658e+02 3.786e+02, threshold=4.164e+02, percent-clipped=0.0 +2023-03-26 02:56:25,547 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19129.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 02:56:33,401 INFO [finetune.py:976] (1/7) Epoch 4, batch 1950, loss[loss=0.1987, simple_loss=0.234, pruned_loss=0.08173, over 4284.00 frames. ], tot_loss[loss=0.2337, simple_loss=0.2894, pruned_loss=0.08902, over 953150.18 frames. ], batch size: 18, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:56:41,418 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19146.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:56:53,409 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19164.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 02:56:58,168 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8183, 1.6409, 1.5253, 1.7384, 2.2172, 1.7134, 1.3782, 1.4963], + device='cuda:1'), covar=tensor([0.2223, 0.2195, 0.1979, 0.1809, 0.1825, 0.1209, 0.2657, 0.1809], + device='cuda:1'), in_proj_covar=tensor([0.0231, 0.0208, 0.0196, 0.0183, 0.0232, 0.0173, 0.0213, 0.0186], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 02:57:09,562 INFO [finetune.py:976] (1/7) Epoch 4, batch 2000, loss[loss=0.2121, simple_loss=0.2679, pruned_loss=0.07817, over 4908.00 frames. ], tot_loss[loss=0.2314, simple_loss=0.2866, pruned_loss=0.08805, over 954523.18 frames. ], batch size: 37, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:57:12,073 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0209, 1.7503, 1.6363, 1.8607, 1.7363, 1.7662, 1.7510, 2.4474], + device='cuda:1'), covar=tensor([0.8026, 0.8845, 0.6170, 0.8080, 0.7393, 0.4206, 0.8430, 0.2618], + device='cuda:1'), in_proj_covar=tensor([0.0277, 0.0253, 0.0220, 0.0284, 0.0236, 0.0196, 0.0240, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 02:57:14,701 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19190.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 02:57:15,890 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7106, 1.5893, 1.6109, 1.7005, 1.0515, 3.7467, 1.4435, 1.8476], + device='cuda:1'), covar=tensor([0.3389, 0.2383, 0.2122, 0.2270, 0.2087, 0.0154, 0.2702, 0.1475], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0112, 0.0116, 0.0119, 0.0116, 0.0096, 0.0101, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 02:57:17,078 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=19194.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:57:35,167 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=19212.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 02:57:35,330 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.39 vs. limit=5.0 +2023-03-26 02:57:39,344 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.251e+02 1.681e+02 2.009e+02 2.396e+02 5.395e+02, threshold=4.017e+02, percent-clipped=3.0 +2023-03-26 02:57:49,440 INFO [finetune.py:976] (1/7) Epoch 4, batch 2050, loss[loss=0.2025, simple_loss=0.254, pruned_loss=0.07546, over 4810.00 frames. ], tot_loss[loss=0.2281, simple_loss=0.283, pruned_loss=0.08662, over 954026.05 frames. ], batch size: 25, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:58:31,395 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 02:58:31,806 INFO [finetune.py:976] (1/7) Epoch 4, batch 2100, loss[loss=0.2183, simple_loss=0.2742, pruned_loss=0.08125, over 4871.00 frames. ], tot_loss[loss=0.2283, simple_loss=0.2829, pruned_loss=0.08687, over 955507.67 frames. ], batch size: 34, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 02:58:34,245 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19286.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:59:09,861 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.136e+02 1.737e+02 2.057e+02 2.371e+02 3.601e+02, threshold=4.115e+02, percent-clipped=0.0 +2023-03-26 02:59:25,764 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19330.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 02:59:27,977 INFO [finetune.py:976] (1/7) Epoch 4, batch 2150, loss[loss=0.2206, simple_loss=0.2805, pruned_loss=0.0803, over 4795.00 frames. ], tot_loss[loss=0.2325, simple_loss=0.2875, pruned_loss=0.08871, over 955284.50 frames. ], batch size: 26, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:00:10,912 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=19378.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:00:13,945 INFO [finetune.py:976] (1/7) Epoch 4, batch 2200, loss[loss=0.2591, simple_loss=0.3056, pruned_loss=0.1064, over 4738.00 frames. ], tot_loss[loss=0.236, simple_loss=0.2906, pruned_loss=0.09067, over 954851.61 frames. ], batch size: 59, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:00:25,152 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6538, 1.3724, 1.2278, 1.0639, 1.3604, 1.3407, 1.3161, 2.0358], + device='cuda:1'), covar=tensor([0.8227, 0.8178, 0.6232, 0.7608, 0.6467, 0.4255, 0.7259, 0.2853], + device='cuda:1'), in_proj_covar=tensor([0.0276, 0.0252, 0.0219, 0.0283, 0.0234, 0.0196, 0.0239, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 03:00:33,371 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19394.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:01:05,355 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.065e+02 1.822e+02 2.092e+02 2.549e+02 4.918e+02, threshold=4.184e+02, percent-clipped=2.0 +2023-03-26 03:01:15,704 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8787, 1.7499, 1.5364, 1.7927, 1.9486, 1.5559, 2.2196, 1.8218], + device='cuda:1'), covar=tensor([0.1844, 0.3285, 0.4105, 0.3578, 0.2940, 0.2114, 0.3281, 0.2662], + device='cuda:1'), in_proj_covar=tensor([0.0166, 0.0194, 0.0239, 0.0255, 0.0224, 0.0186, 0.0211, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:01:19,369 INFO [finetune.py:976] (1/7) Epoch 4, batch 2250, loss[loss=0.2419, simple_loss=0.3001, pruned_loss=0.09181, over 4748.00 frames. ], tot_loss[loss=0.2377, simple_loss=0.2924, pruned_loss=0.09146, over 954959.47 frames. ], batch size: 28, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:01:46,301 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19455.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:01:49,381 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6475, 1.1384, 1.0317, 1.5714, 1.9518, 1.0826, 1.3969, 1.5838], + device='cuda:1'), covar=tensor([0.1449, 0.2162, 0.1834, 0.1202, 0.2032, 0.2234, 0.1388, 0.2020], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0099, 0.0117, 0.0094, 0.0125, 0.0097, 0.0101, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 03:02:09,423 INFO [finetune.py:976] (1/7) Epoch 4, batch 2300, loss[loss=0.2652, simple_loss=0.3051, pruned_loss=0.1126, over 4274.00 frames. ], tot_loss[loss=0.2353, simple_loss=0.2911, pruned_loss=0.08977, over 953858.55 frames. ], batch size: 66, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:02:12,900 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19485.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 03:02:37,770 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.125e+02 1.717e+02 2.025e+02 2.639e+02 4.089e+02, threshold=4.050e+02, percent-clipped=0.0 +2023-03-26 03:02:53,440 INFO [finetune.py:976] (1/7) Epoch 4, batch 2350, loss[loss=0.2329, simple_loss=0.2791, pruned_loss=0.09335, over 4801.00 frames. ], tot_loss[loss=0.233, simple_loss=0.2882, pruned_loss=0.08895, over 954199.93 frames. ], batch size: 25, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:03:37,787 INFO [finetune.py:976] (1/7) Epoch 4, batch 2400, loss[loss=0.1885, simple_loss=0.252, pruned_loss=0.06252, over 4757.00 frames. ], tot_loss[loss=0.2322, simple_loss=0.2865, pruned_loss=0.08891, over 954752.41 frames. ], batch size: 26, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:03:40,238 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19586.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:04:00,378 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9870, 1.8930, 1.7566, 1.9315, 2.5686, 2.0194, 1.7322, 1.5094], + device='cuda:1'), covar=tensor([0.2470, 0.2435, 0.2119, 0.2094, 0.2224, 0.1300, 0.2884, 0.2001], + device='cuda:1'), in_proj_covar=tensor([0.0231, 0.0208, 0.0197, 0.0183, 0.0232, 0.0173, 0.0213, 0.0186], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:04:10,033 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19618.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:04:10,537 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.076e+02 1.651e+02 1.936e+02 2.390e+02 3.810e+02, threshold=3.872e+02, percent-clipped=0.0 +2023-03-26 03:04:14,419 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6898, 1.4629, 1.4183, 1.3231, 1.7486, 1.4607, 1.6252, 1.6361], + device='cuda:1'), covar=tensor([0.1608, 0.2978, 0.3786, 0.2995, 0.2634, 0.1844, 0.2735, 0.2396], + device='cuda:1'), in_proj_covar=tensor([0.0165, 0.0194, 0.0237, 0.0254, 0.0223, 0.0186, 0.0210, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:04:19,620 INFO [finetune.py:976] (1/7) Epoch 4, batch 2450, loss[loss=0.2034, simple_loss=0.2577, pruned_loss=0.07453, over 4790.00 frames. ], tot_loss[loss=0.2296, simple_loss=0.2838, pruned_loss=0.08774, over 955912.06 frames. ], batch size: 26, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:04:20,302 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=19634.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:04:22,668 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7640, 1.1855, 1.6213, 1.5096, 1.4376, 1.3913, 1.4113, 1.4771], + device='cuda:1'), covar=tensor([0.5319, 0.8127, 0.7601, 0.8019, 0.9227, 0.7075, 0.9781, 0.6457], + device='cuda:1'), in_proj_covar=tensor([0.0230, 0.0253, 0.0258, 0.0265, 0.0243, 0.0219, 0.0279, 0.0222], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:04:25,745 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0257, 1.9425, 1.5712, 2.0395, 1.8659, 1.7849, 1.8333, 2.6249], + device='cuda:1'), covar=tensor([0.8610, 0.9716, 0.6729, 0.9812, 0.8055, 0.4835, 0.9503, 0.2960], + device='cuda:1'), in_proj_covar=tensor([0.0277, 0.0253, 0.0220, 0.0284, 0.0236, 0.0196, 0.0240, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 03:04:59,952 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19679.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:05:02,276 INFO [finetune.py:976] (1/7) Epoch 4, batch 2500, loss[loss=0.1986, simple_loss=0.261, pruned_loss=0.06817, over 4827.00 frames. ], tot_loss[loss=0.2304, simple_loss=0.2847, pruned_loss=0.08808, over 956230.79 frames. ], batch size: 25, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:05:06,112 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-03-26 03:05:30,653 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.296e+02 1.721e+02 2.075e+02 2.470e+02 4.533e+02, threshold=4.150e+02, percent-clipped=4.0 +2023-03-26 03:05:43,091 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 03:05:45,258 INFO [finetune.py:976] (1/7) Epoch 4, batch 2550, loss[loss=0.2153, simple_loss=0.2668, pruned_loss=0.08193, over 4699.00 frames. ], tot_loss[loss=0.2307, simple_loss=0.286, pruned_loss=0.08769, over 953845.67 frames. ], batch size: 23, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:05:57,958 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.76 vs. limit=2.0 +2023-03-26 03:05:58,475 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19750.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:06:14,020 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.51 vs. limit=5.0 +2023-03-26 03:06:31,491 INFO [finetune.py:976] (1/7) Epoch 4, batch 2600, loss[loss=0.2729, simple_loss=0.3123, pruned_loss=0.1167, over 4820.00 frames. ], tot_loss[loss=0.2351, simple_loss=0.2905, pruned_loss=0.08981, over 955526.55 frames. ], batch size: 30, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:06:33,328 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19785.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 03:07:15,919 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.335e+02 1.804e+02 2.222e+02 2.871e+02 4.406e+02, threshold=4.445e+02, percent-clipped=2.0 +2023-03-26 03:07:34,471 INFO [finetune.py:976] (1/7) Epoch 4, batch 2650, loss[loss=0.2342, simple_loss=0.2934, pruned_loss=0.08756, over 4783.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.2908, pruned_loss=0.08976, over 953004.18 frames. ], batch size: 51, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:07:34,540 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=19833.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 03:08:34,284 INFO [finetune.py:976] (1/7) Epoch 4, batch 2700, loss[loss=0.187, simple_loss=0.2441, pruned_loss=0.06496, over 4888.00 frames. ], tot_loss[loss=0.2343, simple_loss=0.2893, pruned_loss=0.0896, over 953235.16 frames. ], batch size: 32, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:09:16,013 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.181e+02 1.716e+02 2.005e+02 2.489e+02 3.950e+02, threshold=4.009e+02, percent-clipped=0.0 +2023-03-26 03:09:26,957 INFO [finetune.py:976] (1/7) Epoch 4, batch 2750, loss[loss=0.2397, simple_loss=0.2898, pruned_loss=0.09483, over 4772.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.2875, pruned_loss=0.08943, over 954006.04 frames. ], batch size: 28, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:09:37,444 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-26 03:09:54,786 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19974.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:10:00,205 INFO [finetune.py:976] (1/7) Epoch 4, batch 2800, loss[loss=0.1594, simple_loss=0.2205, pruned_loss=0.04909, over 4926.00 frames. ], tot_loss[loss=0.2282, simple_loss=0.2828, pruned_loss=0.08682, over 955088.15 frames. ], batch size: 37, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:10:09,668 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9060, 1.2644, 0.8386, 1.8093, 2.1330, 1.5847, 1.5333, 1.8281], + device='cuda:1'), covar=tensor([0.1516, 0.2437, 0.2473, 0.1246, 0.2186, 0.1935, 0.1680, 0.2193], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0100, 0.0118, 0.0094, 0.0126, 0.0098, 0.0101, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 03:10:24,993 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.030e+02 1.781e+02 2.091e+02 2.518e+02 3.954e+02, threshold=4.183e+02, percent-clipped=0.0 +2023-03-26 03:10:26,213 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20020.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:10:29,151 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6066, 1.5022, 1.4963, 1.7492, 1.9613, 1.7358, 1.1192, 1.3982], + device='cuda:1'), covar=tensor([0.2319, 0.2237, 0.1833, 0.1729, 0.2024, 0.1291, 0.3110, 0.1955], + device='cuda:1'), in_proj_covar=tensor([0.0232, 0.0210, 0.0198, 0.0184, 0.0233, 0.0174, 0.0214, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:10:34,540 INFO [finetune.py:976] (1/7) Epoch 4, batch 2850, loss[loss=0.2367, simple_loss=0.2858, pruned_loss=0.09381, over 4844.00 frames. ], tot_loss[loss=0.2258, simple_loss=0.2804, pruned_loss=0.08562, over 954915.08 frames. ], batch size: 25, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:10:45,844 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20050.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:11:10,074 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6003, 1.3327, 1.2461, 1.0609, 1.3040, 1.3424, 1.2995, 1.9599], + device='cuda:1'), covar=tensor([0.7792, 0.7405, 0.5766, 0.7300, 0.6065, 0.4023, 0.6954, 0.2689], + device='cuda:1'), in_proj_covar=tensor([0.0277, 0.0253, 0.0219, 0.0284, 0.0236, 0.0197, 0.0240, 0.0190], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 03:11:21,576 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20081.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:11:22,667 INFO [finetune.py:976] (1/7) Epoch 4, batch 2900, loss[loss=0.3355, simple_loss=0.3786, pruned_loss=0.1462, over 4738.00 frames. ], tot_loss[loss=0.2313, simple_loss=0.286, pruned_loss=0.08824, over 955012.32 frames. ], batch size: 59, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:11:31,812 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20089.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:11:42,681 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=20098.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:11:56,203 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20112.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:12:05,858 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.342e+02 1.862e+02 2.201e+02 2.748e+02 4.534e+02, threshold=4.402e+02, percent-clipped=1.0 +2023-03-26 03:12:27,412 INFO [finetune.py:976] (1/7) Epoch 4, batch 2950, loss[loss=0.2321, simple_loss=0.2883, pruned_loss=0.08799, over 4844.00 frames. ], tot_loss[loss=0.2336, simple_loss=0.2893, pruned_loss=0.08895, over 956041.34 frames. ], batch size: 47, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:12:45,224 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8575, 1.6922, 1.5000, 1.5229, 2.0014, 2.0612, 1.8529, 1.5072], + device='cuda:1'), covar=tensor([0.0227, 0.0381, 0.0538, 0.0382, 0.0218, 0.0453, 0.0276, 0.0426], + device='cuda:1'), in_proj_covar=tensor([0.0086, 0.0114, 0.0138, 0.0118, 0.0104, 0.0099, 0.0092, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.6974e-05, 8.9834e-05, 1.1134e-04, 9.3405e-05, 8.2798e-05, 7.3885e-05, + 7.0734e-05, 8.5570e-05], device='cuda:1') +2023-03-26 03:12:48,254 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20150.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:13:10,794 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20173.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:13:18,822 INFO [finetune.py:976] (1/7) Epoch 4, batch 3000, loss[loss=0.2208, simple_loss=0.2807, pruned_loss=0.08049, over 4753.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.2906, pruned_loss=0.08951, over 957535.95 frames. ], batch size: 27, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:13:18,822 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 03:13:27,503 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7506, 1.5736, 1.5392, 1.7315, 2.0839, 1.6916, 1.2486, 1.4836], + device='cuda:1'), covar=tensor([0.2381, 0.2467, 0.2032, 0.1962, 0.1832, 0.1314, 0.2869, 0.1971], + device='cuda:1'), in_proj_covar=tensor([0.0232, 0.0210, 0.0198, 0.0185, 0.0234, 0.0175, 0.0214, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:13:34,007 INFO [finetune.py:1010] (1/7) Epoch 4, validation: loss=0.169, simple_loss=0.2409, pruned_loss=0.04857, over 2265189.00 frames. +2023-03-26 03:13:34,008 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6290MB +2023-03-26 03:13:37,064 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4377, 1.3686, 1.4341, 0.7585, 1.5413, 1.5413, 1.4347, 1.2539], + device='cuda:1'), covar=tensor([0.0625, 0.0722, 0.0657, 0.0981, 0.0762, 0.0612, 0.0609, 0.1115], + device='cuda:1'), in_proj_covar=tensor([0.0142, 0.0135, 0.0148, 0.0131, 0.0113, 0.0146, 0.0150, 0.0165], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:13:58,690 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20204.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:14:18,151 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.180e+02 1.917e+02 2.134e+02 2.804e+02 4.274e+02, threshold=4.268e+02, percent-clipped=0.0 +2023-03-26 03:14:22,526 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1293, 2.6874, 2.5506, 1.3669, 2.6213, 2.2250, 2.0564, 2.3453], + device='cuda:1'), covar=tensor([0.0984, 0.1071, 0.1944, 0.2407, 0.2000, 0.2212, 0.2177, 0.1371], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0199, 0.0202, 0.0190, 0.0217, 0.0210, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:14:26,805 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 03:14:27,242 INFO [finetune.py:976] (1/7) Epoch 4, batch 3050, loss[loss=0.2095, simple_loss=0.2766, pruned_loss=0.0712, over 4745.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.2907, pruned_loss=0.08947, over 957777.35 frames. ], batch size: 54, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:14:38,743 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3936, 2.0736, 1.6626, 0.7289, 1.8561, 1.9053, 1.7397, 1.9165], + device='cuda:1'), covar=tensor([0.0991, 0.1017, 0.1739, 0.2564, 0.1559, 0.2401, 0.2246, 0.1061], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0200, 0.0203, 0.0191, 0.0218, 0.0211, 0.0219, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:14:59,710 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9643, 1.4947, 2.3897, 3.6650, 2.7054, 2.6157, 0.7867, 2.8596], + device='cuda:1'), covar=tensor([0.1852, 0.1678, 0.1365, 0.0556, 0.0835, 0.1802, 0.2112, 0.0623], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0119, 0.0137, 0.0167, 0.0105, 0.0144, 0.0130, 0.0105], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 03:15:01,625 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.33 vs. limit=5.0 +2023-03-26 03:15:06,976 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20265.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:15:16,631 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20274.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:15:24,345 INFO [finetune.py:976] (1/7) Epoch 4, batch 3100, loss[loss=0.2367, simple_loss=0.2842, pruned_loss=0.09462, over 4907.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.2888, pruned_loss=0.08878, over 956376.41 frames. ], batch size: 35, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:16:01,239 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.111e+02 1.549e+02 1.969e+02 2.570e+02 5.632e+02, threshold=3.937e+02, percent-clipped=1.0 +2023-03-26 03:16:03,107 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=20322.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:16:13,164 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.27 vs. limit=5.0 +2023-03-26 03:16:14,157 INFO [finetune.py:976] (1/7) Epoch 4, batch 3150, loss[loss=0.183, simple_loss=0.239, pruned_loss=0.06352, over 4822.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.2856, pruned_loss=0.08774, over 957757.07 frames. ], batch size: 40, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:16:56,930 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20376.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:17:01,160 INFO [finetune.py:976] (1/7) Epoch 4, batch 3200, loss[loss=0.2153, simple_loss=0.272, pruned_loss=0.07928, over 4722.00 frames. ], tot_loss[loss=0.2267, simple_loss=0.2819, pruned_loss=0.08573, over 958503.23 frames. ], batch size: 23, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:17:04,841 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20388.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:17:40,210 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20418.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:17:40,661 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.143e+02 1.625e+02 1.959e+02 2.342e+02 5.079e+02, threshold=3.919e+02, percent-clipped=3.0 +2023-03-26 03:17:58,484 INFO [finetune.py:976] (1/7) Epoch 4, batch 3250, loss[loss=0.1623, simple_loss=0.2359, pruned_loss=0.04432, over 4797.00 frames. ], tot_loss[loss=0.2277, simple_loss=0.2826, pruned_loss=0.08639, over 956970.46 frames. ], batch size: 29, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:18:06,353 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20445.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:18:08,883 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20449.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:18:21,888 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20468.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:18:29,628 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20479.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:18:30,837 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9595, 1.7797, 1.4099, 1.8033, 1.7079, 1.6439, 1.6067, 2.4875], + device='cuda:1'), covar=tensor([0.8315, 0.9051, 0.7184, 0.8999, 0.8033, 0.4676, 0.8813, 0.3150], + device='cuda:1'), in_proj_covar=tensor([0.0277, 0.0252, 0.0220, 0.0284, 0.0237, 0.0197, 0.0241, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 03:18:31,929 INFO [finetune.py:976] (1/7) Epoch 4, batch 3300, loss[loss=0.2031, simple_loss=0.2607, pruned_loss=0.07277, over 4741.00 frames. ], tot_loss[loss=0.2306, simple_loss=0.286, pruned_loss=0.08762, over 954333.30 frames. ], batch size: 27, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:18:35,045 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5023, 2.2007, 1.6825, 0.7820, 1.8622, 2.0412, 1.6898, 1.9312], + device='cuda:1'), covar=tensor([0.0833, 0.0907, 0.1474, 0.2255, 0.1463, 0.2080, 0.2299, 0.0948], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0199, 0.0202, 0.0190, 0.0217, 0.0210, 0.0219, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:19:02,407 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.47 vs. limit=5.0 +2023-03-26 03:19:08,330 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6494, 1.4425, 1.9953, 3.2401, 2.2675, 2.3169, 0.7107, 2.4395], + device='cuda:1'), covar=tensor([0.1754, 0.1523, 0.1352, 0.0534, 0.0763, 0.1506, 0.2090, 0.0639], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0119, 0.0136, 0.0167, 0.0104, 0.0142, 0.0130, 0.0105], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 03:19:09,458 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.164e+02 1.750e+02 2.039e+02 2.534e+02 4.074e+02, threshold=4.078e+02, percent-clipped=2.0 +2023-03-26 03:19:29,492 INFO [finetune.py:976] (1/7) Epoch 4, batch 3350, loss[loss=0.2408, simple_loss=0.3104, pruned_loss=0.08557, over 4891.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.2892, pruned_loss=0.08859, over 955558.96 frames. ], batch size: 43, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:19:59,048 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20560.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:20:17,266 INFO [finetune.py:976] (1/7) Epoch 4, batch 3400, loss[loss=0.2222, simple_loss=0.2873, pruned_loss=0.07857, over 4766.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.2906, pruned_loss=0.08927, over 957223.17 frames. ], batch size: 26, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:20:20,410 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20588.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:20:20,430 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3264, 1.1813, 1.1999, 1.1993, 1.5749, 1.5112, 1.4020, 1.2210], + device='cuda:1'), covar=tensor([0.0280, 0.0275, 0.0485, 0.0308, 0.0191, 0.0322, 0.0233, 0.0340], + device='cuda:1'), in_proj_covar=tensor([0.0085, 0.0112, 0.0136, 0.0117, 0.0104, 0.0098, 0.0091, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.6156e-05, 8.8878e-05, 1.0963e-04, 9.2306e-05, 8.2304e-05, 7.3071e-05, + 6.9907e-05, 8.4286e-05], device='cuda:1') +2023-03-26 03:20:46,762 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.150e+02 1.735e+02 2.048e+02 2.538e+02 3.974e+02, threshold=4.096e+02, percent-clipped=0.0 +2023-03-26 03:20:55,498 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8402, 3.3558, 3.5490, 3.7626, 3.5823, 3.4519, 3.9230, 1.1319], + device='cuda:1'), covar=tensor([0.0893, 0.0774, 0.0846, 0.0842, 0.1334, 0.1303, 0.0883, 0.5117], + device='cuda:1'), in_proj_covar=tensor([0.0356, 0.0242, 0.0276, 0.0291, 0.0337, 0.0283, 0.0305, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:20:56,723 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.6713, 4.8793, 5.2240, 5.5016, 5.3357, 5.1092, 5.7125, 2.4644], + device='cuda:1'), covar=tensor([0.0606, 0.0787, 0.0661, 0.0827, 0.1141, 0.1144, 0.0521, 0.4405], + device='cuda:1'), in_proj_covar=tensor([0.0356, 0.0242, 0.0276, 0.0291, 0.0337, 0.0283, 0.0305, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:20:57,513 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 03:21:05,097 INFO [finetune.py:976] (1/7) Epoch 4, batch 3450, loss[loss=0.233, simple_loss=0.2768, pruned_loss=0.09454, over 4001.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.2893, pruned_loss=0.08854, over 953591.82 frames. ], batch size: 17, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:21:15,658 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8543, 3.7257, 3.8387, 2.1191, 4.0218, 2.9848, 1.3779, 2.8602], + device='cuda:1'), covar=tensor([0.2908, 0.1824, 0.1231, 0.2603, 0.0788, 0.0820, 0.3493, 0.1191], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0170, 0.0162, 0.0128, 0.0155, 0.0121, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 03:21:22,504 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20649.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:21:45,691 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20676.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:21:51,861 INFO [finetune.py:976] (1/7) Epoch 4, batch 3500, loss[loss=0.2696, simple_loss=0.3108, pruned_loss=0.1142, over 4822.00 frames. ], tot_loss[loss=0.2319, simple_loss=0.2872, pruned_loss=0.08832, over 955067.07 frames. ], batch size: 33, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:22:24,581 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1842, 1.9853, 2.2462, 0.8493, 2.2791, 2.5358, 2.1002, 2.0205], + device='cuda:1'), covar=tensor([0.1043, 0.0776, 0.0451, 0.0943, 0.0609, 0.0767, 0.0509, 0.0710], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0156, 0.0118, 0.0136, 0.0131, 0.0121, 0.0146, 0.0144], + device='cuda:1'), out_proj_covar=tensor([9.7032e-05, 1.1578e-04, 8.5580e-05, 9.9445e-05, 9.5064e-05, 8.9681e-05, + 1.0916e-04, 1.0663e-04], device='cuda:1') +2023-03-26 03:22:31,509 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.236e+02 1.688e+02 2.022e+02 2.523e+02 5.341e+02, threshold=4.043e+02, percent-clipped=2.0 +2023-03-26 03:22:35,129 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=20724.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:22:43,665 INFO [finetune.py:976] (1/7) Epoch 4, batch 3550, loss[loss=0.2012, simple_loss=0.2554, pruned_loss=0.0735, over 4913.00 frames. ], tot_loss[loss=0.2296, simple_loss=0.2842, pruned_loss=0.08749, over 954895.07 frames. ], batch size: 46, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:22:56,119 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20744.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:22:56,746 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20745.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:23:24,104 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20768.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:23:25,293 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.90 vs. limit=5.0 +2023-03-26 03:23:33,431 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20774.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:23:41,427 INFO [finetune.py:976] (1/7) Epoch 4, batch 3600, loss[loss=0.2639, simple_loss=0.3119, pruned_loss=0.1079, over 4740.00 frames. ], tot_loss[loss=0.2258, simple_loss=0.2809, pruned_loss=0.08534, over 955127.77 frames. ], batch size: 54, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:23:52,914 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=20793.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:24:24,356 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=20816.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:24:31,500 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.828e+01 1.831e+02 2.152e+02 2.506e+02 5.159e+02, threshold=4.304e+02, percent-clipped=1.0 +2023-03-26 03:24:33,927 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3287, 1.8993, 2.6329, 4.0576, 2.9175, 2.6387, 0.6348, 3.1839], + device='cuda:1'), covar=tensor([0.1612, 0.1486, 0.1294, 0.0510, 0.0726, 0.1500, 0.2315, 0.0556], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0120, 0.0138, 0.0168, 0.0105, 0.0144, 0.0131, 0.0105], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 03:24:44,322 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0038, 1.9990, 1.9518, 1.3746, 2.1261, 2.2169, 1.9669, 1.6884], + device='cuda:1'), covar=tensor([0.0675, 0.0635, 0.0775, 0.0938, 0.0482, 0.0686, 0.0680, 0.1096], + device='cuda:1'), in_proj_covar=tensor([0.0141, 0.0135, 0.0147, 0.0129, 0.0112, 0.0146, 0.0149, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:24:50,039 INFO [finetune.py:976] (1/7) Epoch 4, batch 3650, loss[loss=0.2981, simple_loss=0.3397, pruned_loss=0.1283, over 4818.00 frames. ], tot_loss[loss=0.229, simple_loss=0.2844, pruned_loss=0.08681, over 953413.63 frames. ], batch size: 30, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:25:24,801 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20860.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:25:34,499 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3386, 3.7862, 3.9060, 4.1712, 4.0402, 3.8543, 4.4214, 1.3879], + device='cuda:1'), covar=tensor([0.0847, 0.0808, 0.0842, 0.1146, 0.1285, 0.1527, 0.0702, 0.5350], + device='cuda:1'), in_proj_covar=tensor([0.0359, 0.0242, 0.0277, 0.0295, 0.0338, 0.0285, 0.0307, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:25:52,756 INFO [finetune.py:976] (1/7) Epoch 4, batch 3700, loss[loss=0.2467, simple_loss=0.2913, pruned_loss=0.1011, over 4719.00 frames. ], tot_loss[loss=0.233, simple_loss=0.2893, pruned_loss=0.08838, over 953945.91 frames. ], batch size: 23, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:26:17,163 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=20908.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:26:24,267 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.096e+02 1.710e+02 2.022e+02 2.626e+02 5.956e+02, threshold=4.044e+02, percent-clipped=2.0 +2023-03-26 03:26:34,611 INFO [finetune.py:976] (1/7) Epoch 4, batch 3750, loss[loss=0.2383, simple_loss=0.2924, pruned_loss=0.0921, over 4803.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.2893, pruned_loss=0.08857, over 953778.03 frames. ], batch size: 51, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:26:46,550 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20944.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:27:03,849 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.6575, 2.9704, 2.5030, 1.4282, 2.8456, 2.6253, 2.0982, 2.1863], + device='cuda:1'), covar=tensor([0.0533, 0.1149, 0.1892, 0.2484, 0.1899, 0.1828, 0.2389, 0.1666], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0199, 0.0202, 0.0190, 0.0216, 0.0209, 0.0218, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:27:33,088 INFO [finetune.py:976] (1/7) Epoch 4, batch 3800, loss[loss=0.1891, simple_loss=0.2493, pruned_loss=0.06446, over 4812.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.29, pruned_loss=0.08818, over 955612.89 frames. ], batch size: 25, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:28:02,471 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.7921, 4.1675, 4.3224, 4.5534, 4.5357, 4.3495, 4.8308, 1.7726], + device='cuda:1'), covar=tensor([0.0591, 0.0636, 0.0731, 0.0799, 0.0928, 0.1078, 0.0466, 0.4090], + device='cuda:1'), in_proj_covar=tensor([0.0356, 0.0240, 0.0274, 0.0291, 0.0334, 0.0283, 0.0305, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:28:03,751 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-03-26 03:28:14,228 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.257e+02 1.718e+02 1.983e+02 2.372e+02 3.493e+02, threshold=3.966e+02, percent-clipped=0.0 +2023-03-26 03:28:29,876 INFO [finetune.py:976] (1/7) Epoch 4, batch 3850, loss[loss=0.1823, simple_loss=0.2333, pruned_loss=0.0656, over 4703.00 frames. ], tot_loss[loss=0.2313, simple_loss=0.2878, pruned_loss=0.08742, over 954934.71 frames. ], batch size: 23, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:28:37,703 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21044.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:28:45,306 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21052.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:29:08,443 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21074.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:29:19,707 INFO [finetune.py:976] (1/7) Epoch 4, batch 3900, loss[loss=0.2699, simple_loss=0.3042, pruned_loss=0.1178, over 4745.00 frames. ], tot_loss[loss=0.2299, simple_loss=0.2854, pruned_loss=0.08719, over 955402.92 frames. ], batch size: 27, lr: 3.96e-03, grad_scale: 64.0 +2023-03-26 03:29:21,062 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0562, 1.9164, 1.5192, 2.0108, 1.8675, 1.7360, 1.7386, 2.7492], + device='cuda:1'), covar=tensor([0.8407, 1.0165, 0.6973, 0.9910, 0.8674, 0.4701, 0.9896, 0.2869], + device='cuda:1'), in_proj_covar=tensor([0.0279, 0.0254, 0.0220, 0.0284, 0.0237, 0.0197, 0.0241, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:1') +2023-03-26 03:29:26,804 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=21092.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:29:43,709 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21113.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:29:46,782 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21118.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:29:47,847 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.038e+02 1.709e+02 1.984e+02 2.370e+02 6.134e+02, threshold=3.968e+02, percent-clipped=2.0 +2023-03-26 03:29:49,146 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=21122.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:29:57,028 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0820, 1.9169, 2.2132, 0.8845, 2.2935, 2.6445, 2.1396, 2.0377], + device='cuda:1'), covar=tensor([0.1118, 0.0897, 0.0441, 0.0885, 0.0594, 0.0382, 0.0516, 0.0542], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0158, 0.0119, 0.0137, 0.0133, 0.0121, 0.0148, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.8315e-05, 1.1755e-04, 8.6410e-05, 1.0050e-04, 9.6180e-05, 8.9934e-05, + 1.1014e-04, 1.0832e-04], device='cuda:1') +2023-03-26 03:29:59,343 INFO [finetune.py:976] (1/7) Epoch 4, batch 3950, loss[loss=0.2037, simple_loss=0.2708, pruned_loss=0.06836, over 4696.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.2807, pruned_loss=0.08485, over 954902.60 frames. ], batch size: 23, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:30:00,607 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.2896, 4.5613, 4.7365, 5.1148, 4.9714, 4.7061, 5.3771, 1.6843], + device='cuda:1'), covar=tensor([0.0627, 0.0815, 0.0674, 0.0698, 0.1076, 0.1329, 0.0512, 0.5046], + device='cuda:1'), in_proj_covar=tensor([0.0356, 0.0242, 0.0274, 0.0291, 0.0334, 0.0283, 0.0305, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:30:47,947 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21179.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:30:50,288 INFO [finetune.py:976] (1/7) Epoch 4, batch 4000, loss[loss=0.2359, simple_loss=0.3048, pruned_loss=0.08353, over 4915.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.2798, pruned_loss=0.08492, over 953532.59 frames. ], batch size: 36, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:30:56,472 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-03-26 03:31:04,149 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6149, 1.4736, 1.7288, 1.8039, 1.6144, 3.3700, 1.3208, 1.6701], + device='cuda:1'), covar=tensor([0.0913, 0.1671, 0.1306, 0.0949, 0.1525, 0.0246, 0.1494, 0.1648], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0082, 0.0078, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 03:31:13,446 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-26 03:31:19,051 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.151e+02 1.740e+02 2.148e+02 2.427e+02 4.357e+02, threshold=4.296e+02, percent-clipped=1.0 +2023-03-26 03:31:27,483 INFO [finetune.py:976] (1/7) Epoch 4, batch 4050, loss[loss=0.2152, simple_loss=0.2707, pruned_loss=0.07982, over 4714.00 frames. ], tot_loss[loss=0.23, simple_loss=0.2848, pruned_loss=0.08756, over 953006.28 frames. ], batch size: 23, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:31:35,274 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21244.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:32:12,118 INFO [finetune.py:976] (1/7) Epoch 4, batch 4100, loss[loss=0.2505, simple_loss=0.3032, pruned_loss=0.0989, over 4859.00 frames. ], tot_loss[loss=0.2334, simple_loss=0.2888, pruned_loss=0.08894, over 952770.29 frames. ], batch size: 44, lr: 3.96e-03, grad_scale: 32.0 +2023-03-26 03:32:18,162 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=21292.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:32:44,692 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21312.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:32:53,585 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.029e+02 1.825e+02 2.074e+02 2.571e+02 5.101e+02, threshold=4.147e+02, percent-clipped=2.0 +2023-03-26 03:33:04,188 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21328.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:33:07,022 INFO [finetune.py:976] (1/7) Epoch 4, batch 4150, loss[loss=0.2542, simple_loss=0.3112, pruned_loss=0.09856, over 4929.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.2889, pruned_loss=0.08872, over 953835.06 frames. ], batch size: 33, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:33:28,003 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-26 03:33:35,599 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-03-26 03:33:45,546 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21373.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:33:52,602 INFO [finetune.py:976] (1/7) Epoch 4, batch 4200, loss[loss=0.2586, simple_loss=0.3138, pruned_loss=0.1017, over 4894.00 frames. ], tot_loss[loss=0.232, simple_loss=0.2884, pruned_loss=0.08775, over 953449.19 frames. ], batch size: 35, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:34:02,011 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21389.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:34:21,942 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21408.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:34:30,331 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6135, 1.4900, 1.7768, 1.8741, 1.7549, 3.6367, 1.3755, 1.7349], + device='cuda:1'), covar=tensor([0.1084, 0.1922, 0.1246, 0.1033, 0.1648, 0.0233, 0.1620, 0.1698], + device='cuda:1'), in_proj_covar=tensor([0.0079, 0.0082, 0.0078, 0.0080, 0.0093, 0.0084, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 03:34:31,848 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-26 03:34:39,317 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.013e+02 1.725e+02 2.024e+02 2.533e+02 3.913e+02, threshold=4.049e+02, percent-clipped=0.0 +2023-03-26 03:34:50,755 INFO [finetune.py:976] (1/7) Epoch 4, batch 4250, loss[loss=0.2114, simple_loss=0.2674, pruned_loss=0.07767, over 4815.00 frames. ], tot_loss[loss=0.229, simple_loss=0.2852, pruned_loss=0.08643, over 953139.69 frames. ], batch size: 40, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:35:18,394 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21474.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:35:24,284 INFO [finetune.py:976] (1/7) Epoch 4, batch 4300, loss[loss=0.1872, simple_loss=0.2558, pruned_loss=0.05925, over 4717.00 frames. ], tot_loss[loss=0.2274, simple_loss=0.2832, pruned_loss=0.0858, over 953397.75 frames. ], batch size: 23, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:35:53,224 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.244e+02 1.671e+02 2.065e+02 2.560e+02 4.445e+02, threshold=4.130e+02, percent-clipped=1.0 +2023-03-26 03:36:01,103 INFO [finetune.py:976] (1/7) Epoch 4, batch 4350, loss[loss=0.1785, simple_loss=0.2257, pruned_loss=0.06565, over 4177.00 frames. ], tot_loss[loss=0.2246, simple_loss=0.28, pruned_loss=0.08462, over 952242.62 frames. ], batch size: 17, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:36:40,601 INFO [finetune.py:976] (1/7) Epoch 4, batch 4400, loss[loss=0.289, simple_loss=0.3499, pruned_loss=0.1141, over 4742.00 frames. ], tot_loss[loss=0.2265, simple_loss=0.2816, pruned_loss=0.0857, over 951743.14 frames. ], batch size: 54, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:36:49,012 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21595.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:37:05,925 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21613.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:37:15,011 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.167e+02 1.721e+02 2.036e+02 2.627e+02 4.967e+02, threshold=4.072e+02, percent-clipped=2.0 +2023-03-26 03:37:18,809 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0187, 1.9056, 1.5683, 2.0700, 2.0633, 1.6976, 2.4445, 2.0273], + device='cuda:1'), covar=tensor([0.1676, 0.3589, 0.3994, 0.3758, 0.2970, 0.2061, 0.3825, 0.2456], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0193, 0.0234, 0.0253, 0.0223, 0.0186, 0.0208, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:37:22,898 INFO [finetune.py:976] (1/7) Epoch 4, batch 4450, loss[loss=0.2195, simple_loss=0.2916, pruned_loss=0.07369, over 4922.00 frames. ], tot_loss[loss=0.2293, simple_loss=0.2849, pruned_loss=0.08683, over 951943.67 frames. ], batch size: 42, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:37:38,560 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21656.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:37:39,212 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 03:37:50,982 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21668.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:38:01,165 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21674.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:38:12,007 INFO [finetune.py:976] (1/7) Epoch 4, batch 4500, loss[loss=0.2223, simple_loss=0.286, pruned_loss=0.0793, over 4904.00 frames. ], tot_loss[loss=0.2307, simple_loss=0.2867, pruned_loss=0.08739, over 951648.32 frames. ], batch size: 37, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:38:12,670 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21684.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:38:19,921 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0266, 2.3574, 2.3010, 1.2709, 2.4079, 2.0777, 1.8951, 2.1499], + device='cuda:1'), covar=tensor([0.0873, 0.1036, 0.1827, 0.2414, 0.2071, 0.2671, 0.2383, 0.1432], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0201, 0.0203, 0.0191, 0.0219, 0.0209, 0.0219, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:38:41,406 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21708.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:38:49,076 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.156e+02 1.721e+02 2.032e+02 2.543e+02 5.339e+02, threshold=4.063e+02, percent-clipped=3.0 +2023-03-26 03:38:58,599 INFO [finetune.py:976] (1/7) Epoch 4, batch 4550, loss[loss=0.2532, simple_loss=0.3001, pruned_loss=0.1031, over 4120.00 frames. ], tot_loss[loss=0.2336, simple_loss=0.2894, pruned_loss=0.08886, over 951296.14 frames. ], batch size: 65, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:39:14,114 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=21756.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:39:18,462 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21763.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:39:30,508 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21774.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:39:40,862 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7957, 0.9025, 1.5939, 1.5316, 1.4407, 1.3967, 1.3708, 1.4771], + device='cuda:1'), covar=tensor([0.6088, 0.9217, 0.7631, 0.8113, 0.8945, 0.6920, 1.0435, 0.6790], + device='cuda:1'), in_proj_covar=tensor([0.0229, 0.0252, 0.0257, 0.0262, 0.0242, 0.0218, 0.0278, 0.0222], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:39:41,971 INFO [finetune.py:976] (1/7) Epoch 4, batch 4600, loss[loss=0.1657, simple_loss=0.2307, pruned_loss=0.05038, over 4775.00 frames. ], tot_loss[loss=0.2311, simple_loss=0.2875, pruned_loss=0.08732, over 952881.38 frames. ], batch size: 29, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:40:25,680 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.219e+02 1.789e+02 2.167e+02 2.687e+02 4.147e+02, threshold=4.334e+02, percent-clipped=1.0 +2023-03-26 03:40:32,332 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=21822.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:40:33,578 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21824.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:40:44,876 INFO [finetune.py:976] (1/7) Epoch 4, batch 4650, loss[loss=0.26, simple_loss=0.3109, pruned_loss=0.1045, over 4903.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.2849, pruned_loss=0.08647, over 954231.38 frames. ], batch size: 36, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:41:06,182 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21850.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:41:28,342 INFO [finetune.py:976] (1/7) Epoch 4, batch 4700, loss[loss=0.2066, simple_loss=0.2657, pruned_loss=0.0737, over 4770.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.2812, pruned_loss=0.08457, over 956159.93 frames. ], batch size: 28, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:41:29,699 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 03:41:44,566 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8907, 1.8955, 1.7919, 2.0827, 2.0716, 2.0346, 1.4558, 1.5446], + device='cuda:1'), covar=tensor([0.2018, 0.1935, 0.1718, 0.1568, 0.1987, 0.1083, 0.2681, 0.1778], + device='cuda:1'), in_proj_covar=tensor([0.0231, 0.0208, 0.0198, 0.0184, 0.0235, 0.0174, 0.0214, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:42:02,064 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21911.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:42:08,439 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.749e+02 2.081e+02 2.546e+02 7.973e+02, threshold=4.162e+02, percent-clipped=1.0 +2023-03-26 03:42:14,723 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3610, 2.2187, 2.8389, 1.8589, 2.5443, 2.9576, 2.2610, 2.8631], + device='cuda:1'), covar=tensor([0.1535, 0.1992, 0.1619, 0.2306, 0.1160, 0.1624, 0.2402, 0.1002], + device='cuda:1'), in_proj_covar=tensor([0.0205, 0.0205, 0.0202, 0.0195, 0.0183, 0.0223, 0.0214, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:42:16,914 INFO [finetune.py:976] (1/7) Epoch 4, batch 4750, loss[loss=0.2394, simple_loss=0.2894, pruned_loss=0.09468, over 4767.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2794, pruned_loss=0.08413, over 956376.28 frames. ], batch size: 26, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:42:29,496 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21951.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:42:43,175 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6889, 1.5350, 1.5632, 1.6114, 1.1383, 4.2823, 1.5270, 2.1535], + device='cuda:1'), covar=tensor([0.3576, 0.2457, 0.2119, 0.2391, 0.1981, 0.0109, 0.2591, 0.1381], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0112, 0.0116, 0.0120, 0.0116, 0.0097, 0.0100, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 03:42:46,678 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21968.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:42:53,026 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21969.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:42:53,729 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21970.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:42:57,545 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.76 vs. limit=5.0 +2023-03-26 03:43:02,053 INFO [finetune.py:976] (1/7) Epoch 4, batch 4800, loss[loss=0.2401, simple_loss=0.3079, pruned_loss=0.08617, over 4833.00 frames. ], tot_loss[loss=0.2276, simple_loss=0.2836, pruned_loss=0.08583, over 956534.56 frames. ], batch size: 47, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:43:02,774 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21984.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:43:29,806 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8024, 1.2282, 1.6178, 1.6111, 1.4128, 1.4317, 1.5316, 1.4956], + device='cuda:1'), covar=tensor([0.6887, 1.0023, 0.7690, 0.9064, 1.0102, 0.6982, 1.1341, 0.7893], + device='cuda:1'), in_proj_covar=tensor([0.0229, 0.0252, 0.0257, 0.0262, 0.0242, 0.0218, 0.0278, 0.0223], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:43:34,458 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22016.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:43:37,299 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.143e+02 1.769e+02 1.987e+02 2.631e+02 5.032e+02, threshold=3.974e+02, percent-clipped=2.0 +2023-03-26 03:43:44,131 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22031.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:43:44,661 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22032.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:43:45,188 INFO [finetune.py:976] (1/7) Epoch 4, batch 4850, loss[loss=0.2322, simple_loss=0.2936, pruned_loss=0.08541, over 4855.00 frames. ], tot_loss[loss=0.2293, simple_loss=0.2854, pruned_loss=0.08658, over 955290.87 frames. ], batch size: 31, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:44:19,577 INFO [finetune.py:976] (1/7) Epoch 4, batch 4900, loss[loss=0.2182, simple_loss=0.2904, pruned_loss=0.07298, over 4902.00 frames. ], tot_loss[loss=0.2312, simple_loss=0.2873, pruned_loss=0.08755, over 955173.69 frames. ], batch size: 35, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:44:58,884 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.54 vs. limit=5.0 +2023-03-26 03:45:00,545 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22119.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:45:01,042 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.093e+02 1.764e+02 2.232e+02 2.515e+02 4.523e+02, threshold=4.464e+02, percent-clipped=3.0 +2023-03-26 03:45:18,637 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22130.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:45:20,457 INFO [finetune.py:976] (1/7) Epoch 4, batch 4950, loss[loss=0.2308, simple_loss=0.2815, pruned_loss=0.09003, over 4850.00 frames. ], tot_loss[loss=0.2321, simple_loss=0.2887, pruned_loss=0.08776, over 955509.26 frames. ], batch size: 44, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:46:12,745 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22176.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:46:19,505 INFO [finetune.py:976] (1/7) Epoch 4, batch 5000, loss[loss=0.2798, simple_loss=0.3138, pruned_loss=0.1229, over 4830.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.2858, pruned_loss=0.08599, over 955742.79 frames. ], batch size: 47, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:46:24,492 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22191.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:46:34,468 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7502, 1.5434, 2.0560, 3.4623, 2.4328, 2.3905, 0.8875, 2.6724], + device='cuda:1'), covar=tensor([0.1773, 0.1536, 0.1424, 0.0582, 0.0788, 0.1421, 0.1959, 0.0600], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0120, 0.0138, 0.0167, 0.0104, 0.0144, 0.0130, 0.0105], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 03:46:35,062 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22206.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:46:40,550 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22215.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:46:43,477 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.178e+02 1.598e+02 2.028e+02 2.482e+02 4.524e+02, threshold=4.056e+02, percent-clipped=1.0 +2023-03-26 03:46:59,734 INFO [finetune.py:976] (1/7) Epoch 4, batch 5050, loss[loss=0.1857, simple_loss=0.2474, pruned_loss=0.06201, over 4769.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.2823, pruned_loss=0.08478, over 954757.33 frames. ], batch size: 54, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:47:02,253 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22237.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:47:02,285 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22237.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:47:12,655 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1229, 2.3769, 2.0215, 1.4218, 2.4833, 2.3393, 2.2279, 2.0044], + device='cuda:1'), covar=tensor([0.0709, 0.0636, 0.0979, 0.1022, 0.0444, 0.0789, 0.0751, 0.1021], + device='cuda:1'), in_proj_covar=tensor([0.0138, 0.0133, 0.0145, 0.0128, 0.0110, 0.0144, 0.0148, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:47:16,062 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22251.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:47:22,010 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5178, 3.8596, 4.0884, 4.3939, 4.2499, 3.9663, 4.5808, 1.5061], + device='cuda:1'), covar=tensor([0.0683, 0.0776, 0.0781, 0.0862, 0.1109, 0.1427, 0.0576, 0.5113], + device='cuda:1'), in_proj_covar=tensor([0.0359, 0.0243, 0.0275, 0.0292, 0.0336, 0.0283, 0.0305, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:47:28,157 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22269.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:47:32,929 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22276.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:47:40,354 INFO [finetune.py:976] (1/7) Epoch 4, batch 5100, loss[loss=0.2044, simple_loss=0.2654, pruned_loss=0.07165, over 4826.00 frames. ], tot_loss[loss=0.2237, simple_loss=0.2796, pruned_loss=0.08391, over 955259.89 frames. ], batch size: 40, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:47:46,368 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.88 vs. limit=5.0 +2023-03-26 03:47:49,918 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22298.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:47:50,452 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22299.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:47:55,163 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 03:48:03,275 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22317.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:48:05,001 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.872e+01 1.620e+02 1.853e+02 2.165e+02 3.345e+02, threshold=3.706e+02, percent-clipped=0.0 +2023-03-26 03:48:08,729 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22326.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:48:13,376 INFO [finetune.py:976] (1/7) Epoch 4, batch 5150, loss[loss=0.2544, simple_loss=0.3041, pruned_loss=0.1024, over 4777.00 frames. ], tot_loss[loss=0.2239, simple_loss=0.2794, pruned_loss=0.08417, over 954135.60 frames. ], batch size: 28, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:48:30,276 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4168, 2.1332, 1.9045, 1.6253, 2.4764, 2.7274, 2.5064, 2.0568], + device='cuda:1'), covar=tensor([0.0207, 0.0360, 0.0416, 0.0426, 0.0207, 0.0336, 0.0234, 0.0339], + device='cuda:1'), in_proj_covar=tensor([0.0086, 0.0115, 0.0139, 0.0119, 0.0105, 0.0100, 0.0092, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.7538e-05, 9.0270e-05, 1.1166e-04, 9.4112e-05, 8.3611e-05, 7.4798e-05, + 7.0495e-05, 8.5665e-05], device='cuda:1') +2023-03-26 03:48:49,768 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9394, 1.7782, 1.4847, 1.9560, 2.0238, 1.5648, 2.2183, 1.9061], + device='cuda:1'), covar=tensor([0.1883, 0.3520, 0.4210, 0.3599, 0.2837, 0.2185, 0.4072, 0.2555], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0196, 0.0239, 0.0256, 0.0227, 0.0189, 0.0212, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:48:51,914 INFO [finetune.py:976] (1/7) Epoch 4, batch 5200, loss[loss=0.2761, simple_loss=0.3278, pruned_loss=0.1122, over 4809.00 frames. ], tot_loss[loss=0.2292, simple_loss=0.2849, pruned_loss=0.08682, over 954348.19 frames. ], batch size: 41, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:48:53,240 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22385.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:49:26,382 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22419.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:49:26,882 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.279e+02 1.811e+02 2.155e+02 2.737e+02 4.498e+02, threshold=4.310e+02, percent-clipped=4.0 +2023-03-26 03:49:40,613 INFO [finetune.py:976] (1/7) Epoch 4, batch 5250, loss[loss=0.2374, simple_loss=0.2909, pruned_loss=0.09196, over 4845.00 frames. ], tot_loss[loss=0.2311, simple_loss=0.2872, pruned_loss=0.08751, over 953712.58 frames. ], batch size: 49, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:49:54,886 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22446.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:50:18,145 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22467.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:50:27,829 INFO [finetune.py:976] (1/7) Epoch 4, batch 5300, loss[loss=0.2302, simple_loss=0.2971, pruned_loss=0.08167, over 4751.00 frames. ], tot_loss[loss=0.2308, simple_loss=0.2878, pruned_loss=0.08687, over 954674.90 frames. ], batch size: 27, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:50:30,399 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22486.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:50:36,624 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.34 vs. limit=5.0 +2023-03-26 03:50:49,045 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22506.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:51:10,353 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.196e+02 1.644e+02 2.088e+02 2.525e+02 4.526e+02, threshold=4.176e+02, percent-clipped=1.0 +2023-03-26 03:51:29,035 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22532.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:51:29,580 INFO [finetune.py:976] (1/7) Epoch 4, batch 5350, loss[loss=0.2084, simple_loss=0.2748, pruned_loss=0.07098, over 4837.00 frames. ], tot_loss[loss=0.2292, simple_loss=0.2868, pruned_loss=0.08577, over 954413.21 frames. ], batch size: 47, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:51:40,397 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7967, 1.2776, 1.0746, 1.7850, 2.1251, 1.5672, 1.5673, 1.8201], + device='cuda:1'), covar=tensor([0.1374, 0.1964, 0.2043, 0.1055, 0.1800, 0.1933, 0.1310, 0.1646], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0099, 0.0117, 0.0093, 0.0125, 0.0097, 0.0101, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 03:51:43,430 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22554.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:51:50,037 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0616, 2.4173, 2.1072, 1.6025, 2.3784, 2.5643, 2.4552, 2.0850], + device='cuda:1'), covar=tensor([0.0792, 0.0533, 0.0967, 0.1081, 0.1095, 0.0654, 0.0605, 0.0967], + device='cuda:1'), in_proj_covar=tensor([0.0138, 0.0133, 0.0144, 0.0127, 0.0110, 0.0143, 0.0147, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:52:04,984 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22571.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:52:20,638 INFO [finetune.py:976] (1/7) Epoch 4, batch 5400, loss[loss=0.1955, simple_loss=0.2511, pruned_loss=0.06998, over 4800.00 frames. ], tot_loss[loss=0.2269, simple_loss=0.2844, pruned_loss=0.08466, over 954589.62 frames. ], batch size: 25, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:52:27,239 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22593.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:52:45,277 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.025e+02 1.598e+02 1.961e+02 2.261e+02 4.832e+02, threshold=3.922e+02, percent-clipped=2.0 +2023-03-26 03:52:50,434 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22626.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:52:54,548 INFO [finetune.py:976] (1/7) Epoch 4, batch 5450, loss[loss=0.1983, simple_loss=0.2592, pruned_loss=0.06872, over 4813.00 frames. ], tot_loss[loss=0.224, simple_loss=0.2809, pruned_loss=0.08354, over 952987.95 frames. ], batch size: 30, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:52:58,340 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22639.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:53:01,455 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.98 vs. limit=5.0 +2023-03-26 03:53:30,711 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22674.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:53:30,726 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8965, 3.9263, 3.7724, 1.8468, 4.0052, 3.1202, 1.1578, 2.9005], + device='cuda:1'), covar=tensor([0.2126, 0.1870, 0.1378, 0.3350, 0.0901, 0.0857, 0.4352, 0.1545], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0171, 0.0161, 0.0128, 0.0154, 0.0121, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 03:53:37,513 INFO [finetune.py:976] (1/7) Epoch 4, batch 5500, loss[loss=0.2294, simple_loss=0.2806, pruned_loss=0.08905, over 4862.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.278, pruned_loss=0.08252, over 954034.98 frames. ], batch size: 49, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:53:48,416 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22700.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 03:54:00,990 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.350e+02 1.787e+02 2.088e+02 2.580e+02 6.017e+02, threshold=4.176e+02, percent-clipped=5.0 +2023-03-26 03:54:16,194 INFO [finetune.py:976] (1/7) Epoch 4, batch 5550, loss[loss=0.2256, simple_loss=0.2832, pruned_loss=0.08396, over 4817.00 frames. ], tot_loss[loss=0.2254, simple_loss=0.2819, pruned_loss=0.08443, over 956112.27 frames. ], batch size: 38, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:54:23,407 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22741.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:54:32,500 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6964, 1.6097, 2.2365, 2.0571, 1.8913, 3.5273, 1.5381, 1.8849], + device='cuda:1'), covar=tensor([0.0896, 0.1544, 0.1375, 0.0928, 0.1348, 0.0257, 0.1332, 0.1495], + device='cuda:1'), in_proj_covar=tensor([0.0079, 0.0082, 0.0078, 0.0081, 0.0093, 0.0084, 0.0086, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 03:54:55,507 INFO [finetune.py:976] (1/7) Epoch 4, batch 5600, loss[loss=0.2822, simple_loss=0.34, pruned_loss=0.1122, over 4899.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.2869, pruned_loss=0.08704, over 954560.43 frames. ], batch size: 35, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:54:57,306 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22786.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:55:01,938 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22794.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:55:28,582 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.180e+02 1.750e+02 2.147e+02 2.459e+02 4.993e+02, threshold=4.295e+02, percent-clipped=1.0 +2023-03-26 03:55:31,584 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22825.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:55:40,654 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22832.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:55:41,210 INFO [finetune.py:976] (1/7) Epoch 4, batch 5650, loss[loss=0.218, simple_loss=0.2828, pruned_loss=0.07654, over 4762.00 frames. ], tot_loss[loss=0.2316, simple_loss=0.2886, pruned_loss=0.08725, over 953323.42 frames. ], batch size: 28, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:55:41,954 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22834.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:56:06,189 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22855.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:56:10,287 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0395, 4.2347, 3.8998, 2.2610, 4.3957, 3.2425, 1.2457, 3.1323], + device='cuda:1'), covar=tensor([0.2274, 0.1997, 0.1726, 0.3328, 0.0961, 0.0976, 0.4857, 0.1558], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0169, 0.0160, 0.0126, 0.0153, 0.0120, 0.0144, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 03:56:11,716 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-26 03:56:21,408 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22871.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:56:31,131 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22880.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:56:33,110 INFO [finetune.py:976] (1/7) Epoch 4, batch 5700, loss[loss=0.1995, simple_loss=0.2468, pruned_loss=0.07611, over 4151.00 frames. ], tot_loss[loss=0.2279, simple_loss=0.2833, pruned_loss=0.08625, over 932645.48 frames. ], batch size: 18, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:56:33,790 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2461, 1.8639, 1.4471, 0.6084, 1.6527, 1.9858, 1.6651, 1.8423], + device='cuda:1'), covar=tensor([0.0780, 0.0844, 0.1410, 0.1997, 0.1374, 0.1955, 0.2131, 0.0768], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0202, 0.0204, 0.0191, 0.0218, 0.0210, 0.0221, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:56:34,965 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22886.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:56:41,788 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22893.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:57:20,680 INFO [finetune.py:976] (1/7) Epoch 5, batch 0, loss[loss=0.2699, simple_loss=0.3228, pruned_loss=0.1085, over 4920.00 frames. ], tot_loss[loss=0.2699, simple_loss=0.3228, pruned_loss=0.1085, over 4920.00 frames. ], batch size: 42, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:57:20,680 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 03:57:32,757 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4664, 1.2017, 1.3638, 1.2821, 1.7000, 1.6029, 1.4591, 1.3126], + device='cuda:1'), covar=tensor([0.0337, 0.0395, 0.0568, 0.0369, 0.0261, 0.0506, 0.0354, 0.0427], + device='cuda:1'), in_proj_covar=tensor([0.0086, 0.0115, 0.0138, 0.0120, 0.0105, 0.0101, 0.0092, 0.0110], + device='cuda:1'), out_proj_covar=tensor([6.7570e-05, 9.0440e-05, 1.1130e-04, 9.4555e-05, 8.3660e-05, 7.5062e-05, + 7.0309e-05, 8.6112e-05], device='cuda:1') +2023-03-26 03:57:37,527 INFO [finetune.py:1010] (1/7) Epoch 5, validation: loss=0.1701, simple_loss=0.2413, pruned_loss=0.0494, over 2265189.00 frames. +2023-03-26 03:57:37,527 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6290MB +2023-03-26 03:57:47,698 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0057, 1.5368, 1.7419, 1.7659, 1.5415, 1.5931, 1.6952, 1.6658], + device='cuda:1'), covar=tensor([0.6512, 0.9707, 0.7635, 0.8985, 1.0220, 0.7841, 1.2379, 0.7253], + device='cuda:1'), in_proj_covar=tensor([0.0229, 0.0251, 0.0257, 0.0260, 0.0240, 0.0218, 0.0276, 0.0223], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 03:57:48,822 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22919.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:57:49,350 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.844e+01 1.616e+02 1.840e+02 2.309e+02 3.969e+02, threshold=3.680e+02, percent-clipped=0.0 +2023-03-26 03:58:02,581 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=22941.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:58:29,015 INFO [finetune.py:976] (1/7) Epoch 5, batch 50, loss[loss=0.2713, simple_loss=0.3099, pruned_loss=0.1164, over 4918.00 frames. ], tot_loss[loss=0.2327, simple_loss=0.2891, pruned_loss=0.08815, over 216906.53 frames. ], batch size: 33, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:59:05,721 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22995.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 03:59:17,857 INFO [finetune.py:976] (1/7) Epoch 5, batch 100, loss[loss=0.1983, simple_loss=0.2622, pruned_loss=0.06726, over 4791.00 frames. ], tot_loss[loss=0.2265, simple_loss=0.2828, pruned_loss=0.08514, over 380200.18 frames. ], batch size: 29, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 03:59:23,734 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.764e+02 2.029e+02 2.456e+02 6.922e+02, threshold=4.057e+02, percent-clipped=5.0 +2023-03-26 03:59:34,603 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 03:59:37,004 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23041.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 03:59:44,708 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4479, 1.4864, 1.6383, 1.7476, 1.5803, 3.2996, 1.3500, 1.5516], + device='cuda:1'), covar=tensor([0.0995, 0.1736, 0.1113, 0.1061, 0.1673, 0.0239, 0.1467, 0.1759], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0082, 0.0077, 0.0080, 0.0093, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 03:59:51,326 INFO [finetune.py:976] (1/7) Epoch 5, batch 150, loss[loss=0.2276, simple_loss=0.2803, pruned_loss=0.08749, over 4821.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2759, pruned_loss=0.08253, over 506857.62 frames. ], batch size: 39, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 04:00:01,627 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1024, 1.3263, 1.0886, 1.3394, 1.4346, 2.4791, 1.2621, 1.4653], + device='cuda:1'), covar=tensor([0.1032, 0.1784, 0.1261, 0.1101, 0.1695, 0.0371, 0.1515, 0.1650], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0081, 0.0077, 0.0079, 0.0092, 0.0083, 0.0085, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 04:00:08,772 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=23089.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:00:23,385 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23108.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:00:30,574 INFO [finetune.py:976] (1/7) Epoch 5, batch 200, loss[loss=0.2153, simple_loss=0.2865, pruned_loss=0.07205, over 4902.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2748, pruned_loss=0.08257, over 606799.80 frames. ], batch size: 43, lr: 3.95e-03, grad_scale: 64.0 +2023-03-26 04:00:42,593 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.047e+02 1.662e+02 1.994e+02 2.595e+02 4.858e+02, threshold=3.989e+02, percent-clipped=3.0 +2023-03-26 04:00:45,173 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9291, 1.7892, 1.4396, 1.7183, 1.6842, 1.6650, 1.6406, 2.4663], + device='cuda:1'), covar=tensor([0.7560, 0.7835, 0.6101, 0.7872, 0.6821, 0.4239, 0.7934, 0.2532], + device='cuda:1'), in_proj_covar=tensor([0.0280, 0.0254, 0.0220, 0.0284, 0.0237, 0.0199, 0.0242, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:01:01,383 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23150.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:01:09,513 INFO [finetune.py:976] (1/7) Epoch 5, batch 250, loss[loss=0.2406, simple_loss=0.3046, pruned_loss=0.08825, over 4909.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2795, pruned_loss=0.08409, over 682613.28 frames. ], batch size: 37, lr: 3.95e-03, grad_scale: 64.0 +2023-03-26 04:01:13,847 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6947, 1.6531, 1.4967, 1.8007, 2.2702, 1.7710, 1.3234, 1.3619], + device='cuda:1'), covar=tensor([0.2507, 0.2399, 0.2271, 0.2008, 0.2055, 0.1320, 0.3197, 0.2117], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0210, 0.0199, 0.0185, 0.0236, 0.0175, 0.0215, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:01:18,050 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23169.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:01:31,038 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23181.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:02:00,758 INFO [finetune.py:976] (1/7) Epoch 5, batch 300, loss[loss=0.2285, simple_loss=0.2944, pruned_loss=0.08132, over 4911.00 frames. ], tot_loss[loss=0.2256, simple_loss=0.2827, pruned_loss=0.08422, over 744116.40 frames. ], batch size: 37, lr: 3.95e-03, grad_scale: 64.0 +2023-03-26 04:02:11,910 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23217.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:02:20,056 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.235e+02 1.736e+02 2.142e+02 2.597e+02 5.294e+02, threshold=4.284e+02, percent-clipped=3.0 +2023-03-26 04:02:31,211 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23229.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:02:47,820 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1246, 2.5109, 2.4012, 1.2437, 2.7464, 2.1946, 1.9668, 2.3521], + device='cuda:1'), covar=tensor([0.0880, 0.1042, 0.1711, 0.2414, 0.1803, 0.2216, 0.2309, 0.1202], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0202, 0.0204, 0.0191, 0.0217, 0.0209, 0.0221, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:02:48,516 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3157, 1.5070, 1.6848, 0.8821, 1.4703, 1.8040, 1.7823, 1.4209], + device='cuda:1'), covar=tensor([0.1059, 0.0682, 0.0431, 0.0620, 0.0452, 0.0513, 0.0340, 0.0624], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0157, 0.0118, 0.0135, 0.0132, 0.0121, 0.0148, 0.0144], + device='cuda:1'), out_proj_covar=tensor([9.7305e-05, 1.1644e-04, 8.5437e-05, 9.8953e-05, 9.5027e-05, 8.9921e-05, + 1.0988e-04, 1.0692e-04], device='cuda:1') +2023-03-26 04:03:03,007 INFO [finetune.py:976] (1/7) Epoch 5, batch 350, loss[loss=0.2204, simple_loss=0.2858, pruned_loss=0.07753, over 4752.00 frames. ], tot_loss[loss=0.2274, simple_loss=0.2852, pruned_loss=0.08486, over 791718.68 frames. ], batch size: 26, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 04:03:20,897 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23278.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:03:31,461 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 04:03:34,241 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23290.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:03:41,711 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23295.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:03:51,622 INFO [finetune.py:976] (1/7) Epoch 5, batch 400, loss[loss=0.2027, simple_loss=0.267, pruned_loss=0.06917, over 4826.00 frames. ], tot_loss[loss=0.2296, simple_loss=0.2873, pruned_loss=0.08593, over 825397.82 frames. ], batch size: 49, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 04:03:58,186 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.903e+01 1.731e+02 2.116e+02 2.565e+02 5.981e+02, threshold=4.232e+02, percent-clipped=1.0 +2023-03-26 04:04:13,595 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=23343.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:04:24,904 INFO [finetune.py:976] (1/7) Epoch 5, batch 450, loss[loss=0.1833, simple_loss=0.246, pruned_loss=0.06027, over 4825.00 frames. ], tot_loss[loss=0.2271, simple_loss=0.2849, pruned_loss=0.08467, over 852549.50 frames. ], batch size: 25, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 04:05:10,552 INFO [finetune.py:976] (1/7) Epoch 5, batch 500, loss[loss=0.2767, simple_loss=0.3242, pruned_loss=0.1147, over 4871.00 frames. ], tot_loss[loss=0.2255, simple_loss=0.2823, pruned_loss=0.08435, over 871865.01 frames. ], batch size: 34, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 04:05:16,626 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.006e+02 1.775e+02 2.029e+02 2.615e+02 5.539e+02, threshold=4.057e+02, percent-clipped=1.0 +2023-03-26 04:05:33,325 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4365, 2.0266, 2.7118, 1.8152, 2.4625, 2.7598, 2.0198, 2.7647], + device='cuda:1'), covar=tensor([0.1669, 0.2337, 0.1618, 0.2555, 0.1177, 0.1810, 0.2679, 0.1111], + device='cuda:1'), in_proj_covar=tensor([0.0205, 0.0205, 0.0202, 0.0196, 0.0183, 0.0223, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:05:42,946 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23450.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:05:52,287 INFO [finetune.py:976] (1/7) Epoch 5, batch 550, loss[loss=0.1987, simple_loss=0.258, pruned_loss=0.06976, over 4880.00 frames. ], tot_loss[loss=0.2236, simple_loss=0.2797, pruned_loss=0.08369, over 888397.99 frames. ], batch size: 34, lr: 3.95e-03, grad_scale: 32.0 +2023-03-26 04:05:54,204 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23464.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:06:15,706 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23481.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:06:23,762 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23488.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:06:30,285 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=23498.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:06:38,163 INFO [finetune.py:976] (1/7) Epoch 5, batch 600, loss[loss=0.1698, simple_loss=0.2302, pruned_loss=0.0547, over 4893.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2798, pruned_loss=0.0839, over 903648.36 frames. ], batch size: 32, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:06:44,765 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.135e+02 1.759e+02 2.043e+02 2.434e+02 4.744e+02, threshold=4.086e+02, percent-clipped=2.0 +2023-03-26 04:06:51,178 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=23529.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:07:18,712 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23549.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 04:07:25,885 INFO [finetune.py:976] (1/7) Epoch 5, batch 650, loss[loss=0.3036, simple_loss=0.3612, pruned_loss=0.123, over 4766.00 frames. ], tot_loss[loss=0.228, simple_loss=0.2845, pruned_loss=0.08575, over 915398.25 frames. ], batch size: 54, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:07:33,748 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23573.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:07:42,969 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23585.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:08:01,907 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5791, 2.2441, 1.8157, 0.9006, 1.9898, 2.1336, 1.9410, 1.9925], + device='cuda:1'), covar=tensor([0.0750, 0.0871, 0.1583, 0.2186, 0.1588, 0.2057, 0.2091, 0.0910], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0200, 0.0202, 0.0190, 0.0216, 0.0209, 0.0220, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:08:14,419 INFO [finetune.py:976] (1/7) Epoch 5, batch 700, loss[loss=0.2463, simple_loss=0.3108, pruned_loss=0.0909, over 4821.00 frames. ], tot_loss[loss=0.231, simple_loss=0.2873, pruned_loss=0.08732, over 923829.87 frames. ], batch size: 40, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:08:30,907 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.701e+02 2.127e+02 2.576e+02 5.648e+02, threshold=4.253e+02, percent-clipped=2.0 +2023-03-26 04:09:04,042 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.2471, 1.3614, 1.5840, 0.7238, 1.2778, 1.6314, 1.6423, 1.4090], + device='cuda:1'), covar=tensor([0.0927, 0.0563, 0.0397, 0.0559, 0.0484, 0.0528, 0.0338, 0.0570], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0158, 0.0118, 0.0137, 0.0133, 0.0122, 0.0148, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.8256e-05, 1.1722e-04, 8.5898e-05, 9.9929e-05, 9.5671e-05, 9.0316e-05, + 1.1025e-04, 1.0767e-04], device='cuda:1') +2023-03-26 04:09:25,209 INFO [finetune.py:976] (1/7) Epoch 5, batch 750, loss[loss=0.2682, simple_loss=0.3145, pruned_loss=0.1109, over 4775.00 frames. ], tot_loss[loss=0.2307, simple_loss=0.2872, pruned_loss=0.08706, over 928148.34 frames. ], batch size: 59, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:09:46,283 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-26 04:10:02,097 INFO [finetune.py:976] (1/7) Epoch 5, batch 800, loss[loss=0.249, simple_loss=0.2988, pruned_loss=0.09965, over 4919.00 frames. ], tot_loss[loss=0.2295, simple_loss=0.2867, pruned_loss=0.08615, over 933957.77 frames. ], batch size: 33, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:10:07,717 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6739, 1.1182, 0.8638, 1.5910, 2.0888, 1.0016, 1.4349, 1.6766], + device='cuda:1'), covar=tensor([0.1588, 0.2143, 0.2260, 0.1212, 0.2055, 0.2212, 0.1417, 0.1834], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0100, 0.0118, 0.0094, 0.0125, 0.0098, 0.0102, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 04:10:08,708 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.149e+02 1.635e+02 1.954e+02 2.429e+02 4.773e+02, threshold=3.908e+02, percent-clipped=1.0 +2023-03-26 04:10:56,823 INFO [finetune.py:976] (1/7) Epoch 5, batch 850, loss[loss=0.2749, simple_loss=0.3211, pruned_loss=0.1143, over 4882.00 frames. ], tot_loss[loss=0.2283, simple_loss=0.2848, pruned_loss=0.08587, over 939431.15 frames. ], batch size: 32, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:11:04,267 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23764.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:11:54,660 INFO [finetune.py:976] (1/7) Epoch 5, batch 900, loss[loss=0.2105, simple_loss=0.2689, pruned_loss=0.07603, over 4816.00 frames. ], tot_loss[loss=0.2256, simple_loss=0.2822, pruned_loss=0.0845, over 945528.52 frames. ], batch size: 39, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:11:55,342 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=23812.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:12:00,783 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.157e+02 1.642e+02 1.957e+02 2.389e+02 4.840e+02, threshold=3.913e+02, percent-clipped=2.0 +2023-03-26 04:12:21,815 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23844.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 04:12:37,545 INFO [finetune.py:976] (1/7) Epoch 5, batch 950, loss[loss=0.1922, simple_loss=0.2584, pruned_loss=0.06299, over 4825.00 frames. ], tot_loss[loss=0.2244, simple_loss=0.2803, pruned_loss=0.08427, over 948028.31 frames. ], batch size: 51, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:12:40,713 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6544, 2.3378, 2.0391, 1.0894, 2.1300, 2.0280, 1.8002, 2.1140], + device='cuda:1'), covar=tensor([0.0824, 0.0857, 0.1761, 0.2307, 0.1636, 0.2176, 0.2298, 0.1226], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0199, 0.0202, 0.0189, 0.0215, 0.0208, 0.0219, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:12:42,508 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9702, 2.2418, 1.9414, 1.3683, 2.3012, 2.2938, 2.1623, 1.8937], + device='cuda:1'), covar=tensor([0.0792, 0.0590, 0.0939, 0.1080, 0.0561, 0.0791, 0.0701, 0.1026], + device='cuda:1'), in_proj_covar=tensor([0.0139, 0.0133, 0.0144, 0.0128, 0.0111, 0.0143, 0.0146, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:12:44,922 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23873.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:12:52,630 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23885.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:12:52,705 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-26 04:13:28,832 INFO [finetune.py:976] (1/7) Epoch 5, batch 1000, loss[loss=0.2367, simple_loss=0.3049, pruned_loss=0.08429, over 4934.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.2809, pruned_loss=0.08432, over 948387.39 frames. ], batch size: 33, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:13:38,585 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.160e+02 1.702e+02 2.066e+02 2.385e+02 5.722e+02, threshold=4.131e+02, percent-clipped=3.0 +2023-03-26 04:13:38,657 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=23921.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:13:47,711 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=23933.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:13:54,295 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8620, 1.2483, 2.1680, 3.5532, 2.2456, 2.5917, 0.9207, 2.9217], + device='cuda:1'), covar=tensor([0.2052, 0.2413, 0.1720, 0.0840, 0.1114, 0.1886, 0.2327, 0.0712], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0119, 0.0136, 0.0167, 0.0103, 0.0142, 0.0129, 0.0105], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 04:14:14,856 INFO [finetune.py:976] (1/7) Epoch 5, batch 1050, loss[loss=0.2236, simple_loss=0.2804, pruned_loss=0.08342, over 4865.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.2829, pruned_loss=0.08449, over 951341.42 frames. ], batch size: 31, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:15:10,635 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9205, 1.6609, 1.4132, 1.4421, 1.6599, 1.5538, 1.5768, 2.3058], + device='cuda:1'), covar=tensor([0.7686, 0.7524, 0.5852, 0.7202, 0.6430, 0.4133, 0.7252, 0.2928], + device='cuda:1'), in_proj_covar=tensor([0.0278, 0.0254, 0.0219, 0.0282, 0.0235, 0.0198, 0.0242, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:15:19,193 INFO [finetune.py:976] (1/7) Epoch 5, batch 1100, loss[loss=0.2203, simple_loss=0.2768, pruned_loss=0.08188, over 4746.00 frames. ], tot_loss[loss=0.2272, simple_loss=0.2846, pruned_loss=0.08489, over 952217.52 frames. ], batch size: 27, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:15:19,906 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.93 vs. limit=5.0 +2023-03-26 04:15:28,406 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.244e+02 1.825e+02 2.109e+02 2.589e+02 5.024e+02, threshold=4.219e+02, percent-clipped=4.0 +2023-03-26 04:15:38,241 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24037.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:15:52,169 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.9466, 1.9623, 2.1753, 1.3275, 1.9721, 2.2552, 2.1597, 1.8494], + device='cuda:1'), covar=tensor([0.1010, 0.0607, 0.0430, 0.0554, 0.0432, 0.0483, 0.0396, 0.0629], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0158, 0.0118, 0.0136, 0.0132, 0.0122, 0.0147, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.7555e-05, 1.1714e-04, 8.5647e-05, 9.9151e-05, 9.5531e-05, 9.0081e-05, + 1.0947e-04, 1.0733e-04], device='cuda:1') +2023-03-26 04:15:54,484 INFO [finetune.py:976] (1/7) Epoch 5, batch 1150, loss[loss=0.2145, simple_loss=0.2845, pruned_loss=0.07223, over 4887.00 frames. ], tot_loss[loss=0.2281, simple_loss=0.2856, pruned_loss=0.08534, over 953812.64 frames. ], batch size: 43, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:15:59,323 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.04 vs. limit=5.0 +2023-03-26 04:16:18,942 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24098.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:16:19,058 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 04:16:28,015 INFO [finetune.py:976] (1/7) Epoch 5, batch 1200, loss[loss=0.2648, simple_loss=0.3184, pruned_loss=0.1056, over 4747.00 frames. ], tot_loss[loss=0.2277, simple_loss=0.285, pruned_loss=0.0852, over 955027.54 frames. ], batch size: 54, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:16:37,229 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.193e+02 1.721e+02 2.129e+02 2.606e+02 7.150e+02, threshold=4.257e+02, percent-clipped=3.0 +2023-03-26 04:16:51,870 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=24144.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 04:16:54,322 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0608, 1.8573, 1.5567, 2.0210, 1.8377, 1.7382, 1.7486, 2.5946], + device='cuda:1'), covar=tensor([0.7196, 0.9414, 0.6206, 0.8629, 0.7608, 0.4305, 0.8584, 0.2621], + device='cuda:1'), in_proj_covar=tensor([0.0279, 0.0253, 0.0219, 0.0282, 0.0235, 0.0198, 0.0241, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:16:54,955 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 04:17:03,465 INFO [finetune.py:976] (1/7) Epoch 5, batch 1250, loss[loss=0.2412, simple_loss=0.2953, pruned_loss=0.09352, over 4191.00 frames. ], tot_loss[loss=0.2242, simple_loss=0.2812, pruned_loss=0.08363, over 953929.81 frames. ], batch size: 65, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:17:11,336 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5296, 1.5724, 1.9407, 1.1830, 1.6303, 1.8311, 1.4809, 2.0156], + device='cuda:1'), covar=tensor([0.1450, 0.1978, 0.1284, 0.1950, 0.0937, 0.1349, 0.2622, 0.0900], + device='cuda:1'), in_proj_covar=tensor([0.0205, 0.0204, 0.0201, 0.0195, 0.0183, 0.0222, 0.0214, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:17:28,509 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=24192.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:17:35,291 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7566, 1.1555, 0.9782, 1.7040, 2.1505, 1.4672, 1.6560, 1.7295], + device='cuda:1'), covar=tensor([0.1462, 0.2098, 0.2129, 0.1175, 0.1944, 0.2156, 0.1295, 0.1906], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0099, 0.0118, 0.0094, 0.0125, 0.0098, 0.0101, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 04:17:42,424 INFO [finetune.py:976] (1/7) Epoch 5, batch 1300, loss[loss=0.2247, simple_loss=0.2816, pruned_loss=0.08392, over 4924.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2773, pruned_loss=0.08165, over 955495.95 frames. ], batch size: 36, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:17:56,617 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.098e+02 1.607e+02 1.851e+02 2.364e+02 3.844e+02, threshold=3.702e+02, percent-clipped=0.0 +2023-03-26 04:17:56,733 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3178, 1.4326, 1.3062, 1.5254, 1.6056, 2.9940, 1.3229, 1.6051], + device='cuda:1'), covar=tensor([0.0990, 0.1709, 0.1172, 0.0977, 0.1508, 0.0286, 0.1387, 0.1617], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0077, 0.0080, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 04:18:34,749 INFO [finetune.py:976] (1/7) Epoch 5, batch 1350, loss[loss=0.2699, simple_loss=0.3355, pruned_loss=0.1022, over 4752.00 frames. ], tot_loss[loss=0.2202, simple_loss=0.2771, pruned_loss=0.08161, over 955550.52 frames. ], batch size: 54, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:18:48,383 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4066, 2.1529, 2.9634, 1.8172, 2.7946, 3.1451, 2.3188, 2.9392], + device='cuda:1'), covar=tensor([0.2187, 0.2452, 0.1611, 0.2883, 0.1143, 0.1568, 0.2496, 0.1115], + device='cuda:1'), in_proj_covar=tensor([0.0205, 0.0204, 0.0201, 0.0195, 0.0182, 0.0222, 0.0214, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:19:12,805 INFO [finetune.py:976] (1/7) Epoch 5, batch 1400, loss[loss=0.2044, simple_loss=0.2599, pruned_loss=0.0744, over 4771.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2804, pruned_loss=0.08287, over 956974.56 frames. ], batch size: 26, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:19:16,409 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5481, 1.4080, 1.4543, 1.5443, 1.1839, 3.5668, 1.4270, 1.8975], + device='cuda:1'), covar=tensor([0.3591, 0.2483, 0.2116, 0.2243, 0.1861, 0.0156, 0.2659, 0.1392], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0113, 0.0116, 0.0120, 0.0116, 0.0097, 0.0100, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 04:19:21,620 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.173e+02 1.714e+02 2.138e+02 2.571e+02 4.877e+02, threshold=4.276e+02, percent-clipped=6.0 +2023-03-26 04:19:56,902 INFO [finetune.py:976] (1/7) Epoch 5, batch 1450, loss[loss=0.1875, simple_loss=0.2589, pruned_loss=0.05801, over 4774.00 frames. ], tot_loss[loss=0.2253, simple_loss=0.2833, pruned_loss=0.08364, over 957163.74 frames. ], batch size: 29, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:20:20,202 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24393.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:20:20,844 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4529, 1.4576, 1.4715, 0.7258, 1.6349, 1.4768, 1.4261, 1.3873], + device='cuda:1'), covar=tensor([0.0658, 0.0752, 0.0737, 0.1086, 0.0699, 0.0821, 0.0771, 0.1251], + device='cuda:1'), in_proj_covar=tensor([0.0140, 0.0133, 0.0145, 0.0128, 0.0112, 0.0144, 0.0147, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:20:31,767 INFO [finetune.py:976] (1/7) Epoch 5, batch 1500, loss[loss=0.2297, simple_loss=0.2803, pruned_loss=0.08959, over 4803.00 frames. ], tot_loss[loss=0.2278, simple_loss=0.2854, pruned_loss=0.08509, over 956420.93 frames. ], batch size: 25, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:20:38,327 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.023e+02 1.776e+02 2.138e+02 2.564e+02 4.291e+02, threshold=4.276e+02, percent-clipped=1.0 +2023-03-26 04:20:39,661 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-03-26 04:21:13,451 INFO [finetune.py:976] (1/7) Epoch 5, batch 1550, loss[loss=0.1891, simple_loss=0.2569, pruned_loss=0.06064, over 4901.00 frames. ], tot_loss[loss=0.2276, simple_loss=0.2853, pruned_loss=0.08496, over 955256.44 frames. ], batch size: 43, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:21:47,121 INFO [finetune.py:976] (1/7) Epoch 5, batch 1600, loss[loss=0.2192, simple_loss=0.2809, pruned_loss=0.07879, over 4902.00 frames. ], tot_loss[loss=0.2267, simple_loss=0.2839, pruned_loss=0.08473, over 954696.97 frames. ], batch size: 36, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:21:58,799 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.205e+02 1.770e+02 2.018e+02 2.552e+02 5.194e+02, threshold=4.037e+02, percent-clipped=4.0 +2023-03-26 04:22:21,115 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24547.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:22:33,035 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.98 vs. limit=5.0 +2023-03-26 04:22:33,858 INFO [finetune.py:976] (1/7) Epoch 5, batch 1650, loss[loss=0.2307, simple_loss=0.2872, pruned_loss=0.08712, over 4931.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2801, pruned_loss=0.08299, over 955268.25 frames. ], batch size: 33, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:22:43,192 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24569.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:22:48,421 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7426, 1.6405, 1.5665, 1.8947, 2.1156, 1.8260, 1.3194, 1.5058], + device='cuda:1'), covar=tensor([0.2369, 0.2283, 0.1940, 0.1721, 0.1881, 0.1181, 0.2878, 0.1902], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0208, 0.0198, 0.0184, 0.0235, 0.0174, 0.0214, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:22:56,030 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4593, 1.5036, 1.6306, 1.7662, 1.6048, 3.2717, 1.3569, 1.6393], + device='cuda:1'), covar=tensor([0.0955, 0.1616, 0.1050, 0.0967, 0.1496, 0.0251, 0.1391, 0.1594], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0077, 0.0079, 0.0093, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 04:23:16,153 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8116, 1.3171, 0.7869, 1.6734, 2.0684, 1.4256, 1.7066, 1.7994], + device='cuda:1'), covar=tensor([0.1615, 0.2230, 0.2442, 0.1309, 0.2170, 0.2197, 0.1438, 0.2026], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0099, 0.0117, 0.0094, 0.0125, 0.0098, 0.0101, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 04:23:17,405 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4550, 1.4443, 1.8275, 1.8570, 1.5926, 3.3421, 1.2755, 1.6442], + device='cuda:1'), covar=tensor([0.0997, 0.1627, 0.1157, 0.0951, 0.1514, 0.0268, 0.1487, 0.1612], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0077, 0.0079, 0.0093, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 04:23:17,426 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24608.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:23:22,440 INFO [finetune.py:976] (1/7) Epoch 5, batch 1700, loss[loss=0.1904, simple_loss=0.2381, pruned_loss=0.0713, over 4258.00 frames. ], tot_loss[loss=0.2211, simple_loss=0.2775, pruned_loss=0.08235, over 953786.60 frames. ], batch size: 18, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:23:31,254 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.092e+02 1.673e+02 1.915e+02 2.251e+02 4.027e+02, threshold=3.830e+02, percent-clipped=0.0 +2023-03-26 04:23:42,119 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24630.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:24:06,425 INFO [finetune.py:976] (1/7) Epoch 5, batch 1750, loss[loss=0.2727, simple_loss=0.2913, pruned_loss=0.127, over 4053.00 frames. ], tot_loss[loss=0.2213, simple_loss=0.2779, pruned_loss=0.08232, over 952605.80 frames. ], batch size: 17, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:24:28,023 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=24693.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:24:39,305 INFO [finetune.py:976] (1/7) Epoch 5, batch 1800, loss[loss=0.1789, simple_loss=0.2534, pruned_loss=0.05223, over 4929.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2816, pruned_loss=0.08304, over 955435.51 frames. ], batch size: 33, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:24:43,521 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8525, 1.2834, 0.8454, 1.6072, 2.0932, 1.3751, 1.6188, 1.7131], + device='cuda:1'), covar=tensor([0.1496, 0.2114, 0.2261, 0.1247, 0.2038, 0.2102, 0.1380, 0.2000], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0099, 0.0117, 0.0094, 0.0125, 0.0097, 0.0101, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 04:24:45,828 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.243e+02 1.800e+02 2.166e+02 2.491e+02 4.201e+02, threshold=4.331e+02, percent-clipped=2.0 +2023-03-26 04:24:59,915 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=24741.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:25:12,942 INFO [finetune.py:976] (1/7) Epoch 5, batch 1850, loss[loss=0.2495, simple_loss=0.3123, pruned_loss=0.0934, over 4890.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.2832, pruned_loss=0.08333, over 955823.87 frames. ], batch size: 35, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:25:15,482 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24765.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:25:46,395 INFO [finetune.py:976] (1/7) Epoch 5, batch 1900, loss[loss=0.1994, simple_loss=0.2654, pruned_loss=0.06668, over 4816.00 frames. ], tot_loss[loss=0.2256, simple_loss=0.2841, pruned_loss=0.08353, over 954681.95 frames. ], batch size: 38, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:25:51,958 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9171, 1.7295, 2.3083, 1.3938, 2.2464, 2.1738, 1.6678, 2.3021], + device='cuda:1'), covar=tensor([0.1576, 0.2201, 0.1489, 0.2358, 0.0950, 0.1652, 0.2751, 0.1032], + device='cuda:1'), in_proj_covar=tensor([0.0205, 0.0205, 0.0201, 0.0195, 0.0182, 0.0222, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:25:52,457 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.150e+02 1.802e+02 2.061e+02 2.489e+02 6.200e+02, threshold=4.122e+02, percent-clipped=1.0 +2023-03-26 04:25:57,962 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24826.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:26:20,827 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24848.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:26:29,631 INFO [finetune.py:976] (1/7) Epoch 5, batch 1950, loss[loss=0.2309, simple_loss=0.2913, pruned_loss=0.08523, over 4816.00 frames. ], tot_loss[loss=0.2254, simple_loss=0.2838, pruned_loss=0.08355, over 957751.49 frames. ], batch size: 39, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:26:34,007 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.1750, 4.4866, 4.6681, 5.0361, 4.8779, 4.6696, 5.2946, 1.7599], + device='cuda:1'), covar=tensor([0.0743, 0.0781, 0.0763, 0.0906, 0.1139, 0.1420, 0.0559, 0.5311], + device='cuda:1'), in_proj_covar=tensor([0.0362, 0.0246, 0.0277, 0.0295, 0.0341, 0.0287, 0.0308, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:26:57,653 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24903.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:27:01,810 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24909.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:27:02,943 INFO [finetune.py:976] (1/7) Epoch 5, batch 2000, loss[loss=0.2198, simple_loss=0.2786, pruned_loss=0.08048, over 4818.00 frames. ], tot_loss[loss=0.2226, simple_loss=0.2806, pruned_loss=0.08232, over 958028.59 frames. ], batch size: 38, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:27:13,005 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.094e+02 1.611e+02 2.012e+02 2.424e+02 3.709e+02, threshold=4.024e+02, percent-clipped=0.0 +2023-03-26 04:27:15,538 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24925.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:27:31,947 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6071, 1.3922, 1.9221, 3.1383, 2.1160, 2.3358, 1.1938, 2.5465], + device='cuda:1'), covar=tensor([0.1910, 0.1672, 0.1430, 0.0542, 0.0855, 0.1235, 0.1829, 0.0654], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0119, 0.0137, 0.0167, 0.0103, 0.0143, 0.0129, 0.0105], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 04:27:45,096 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4757, 1.3576, 1.3912, 1.3991, 0.8072, 2.0101, 0.7612, 1.3098], + device='cuda:1'), covar=tensor([0.2919, 0.2112, 0.1742, 0.2054, 0.1715, 0.0356, 0.2178, 0.1050], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0114, 0.0118, 0.0121, 0.0117, 0.0098, 0.0101, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 04:27:46,361 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9439, 1.2041, 1.7342, 1.6496, 1.4982, 1.5147, 1.5781, 1.6130], + device='cuda:1'), covar=tensor([0.5468, 0.7971, 0.6176, 0.7448, 0.8354, 0.6079, 0.9287, 0.5828], + device='cuda:1'), in_proj_covar=tensor([0.0230, 0.0251, 0.0257, 0.0261, 0.0243, 0.0220, 0.0277, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:27:50,325 INFO [finetune.py:976] (1/7) Epoch 5, batch 2050, loss[loss=0.1808, simple_loss=0.2425, pruned_loss=0.05957, over 4765.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2764, pruned_loss=0.08088, over 957054.92 frames. ], batch size: 28, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:28:05,582 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5771, 2.3864, 2.0628, 1.1029, 2.1956, 1.9762, 1.7357, 2.1537], + device='cuda:1'), covar=tensor([0.0964, 0.0811, 0.1753, 0.2297, 0.1672, 0.2178, 0.2154, 0.1141], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0200, 0.0203, 0.0190, 0.0216, 0.0209, 0.0221, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:28:12,176 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24995.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:28:23,337 INFO [finetune.py:976] (1/7) Epoch 5, batch 2100, loss[loss=0.2139, simple_loss=0.2788, pruned_loss=0.07454, over 4778.00 frames. ], tot_loss[loss=0.2193, simple_loss=0.276, pruned_loss=0.08129, over 955728.99 frames. ], batch size: 27, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:28:32,336 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0464, 1.3369, 0.8025, 1.8342, 2.2855, 1.6950, 1.5961, 1.9897], + device='cuda:1'), covar=tensor([0.1561, 0.2151, 0.2332, 0.1260, 0.2140, 0.2156, 0.1525, 0.1922], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0116, 0.0093, 0.0124, 0.0096, 0.0100, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 04:28:39,037 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.247e+01 1.663e+02 1.989e+02 2.457e+02 4.446e+02, threshold=3.978e+02, percent-clipped=1.0 +2023-03-26 04:28:52,266 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9852, 1.3159, 1.8468, 1.6929, 1.6467, 1.6070, 1.5749, 1.7017], + device='cuda:1'), covar=tensor([0.5434, 0.7864, 0.6178, 0.6955, 0.7937, 0.5936, 0.8740, 0.5831], + device='cuda:1'), in_proj_covar=tensor([0.0230, 0.0251, 0.0257, 0.0261, 0.0243, 0.0220, 0.0277, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:29:08,273 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25056.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 04:29:11,160 INFO [finetune.py:976] (1/7) Epoch 5, batch 2150, loss[loss=0.2246, simple_loss=0.2896, pruned_loss=0.07978, over 4770.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.2785, pruned_loss=0.08151, over 954800.03 frames. ], batch size: 28, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:29:28,105 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1327, 1.9016, 2.6977, 1.6480, 2.2795, 2.4383, 1.8253, 2.5333], + device='cuda:1'), covar=tensor([0.1415, 0.2016, 0.1483, 0.2315, 0.0949, 0.1559, 0.2375, 0.0857], + device='cuda:1'), in_proj_covar=tensor([0.0207, 0.0208, 0.0203, 0.0197, 0.0185, 0.0225, 0.0217, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:29:45,149 INFO [finetune.py:976] (1/7) Epoch 5, batch 2200, loss[loss=0.295, simple_loss=0.346, pruned_loss=0.1219, over 4842.00 frames. ], tot_loss[loss=0.2217, simple_loss=0.2805, pruned_loss=0.08143, over 956645.60 frames. ], batch size: 44, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:29:52,263 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.691e+02 1.983e+02 2.301e+02 4.176e+02, threshold=3.967e+02, percent-clipped=1.0 +2023-03-26 04:29:52,339 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25121.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:30:18,168 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7140, 0.6895, 1.6145, 1.4628, 1.4669, 1.3971, 1.2896, 1.5185], + device='cuda:1'), covar=tensor([0.5318, 0.7270, 0.6690, 0.6619, 0.7663, 0.5594, 0.7902, 0.5736], + device='cuda:1'), in_proj_covar=tensor([0.0228, 0.0249, 0.0255, 0.0259, 0.0241, 0.0218, 0.0274, 0.0222], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:30:18,606 INFO [finetune.py:976] (1/7) Epoch 5, batch 2250, loss[loss=0.1956, simple_loss=0.2446, pruned_loss=0.0733, over 4436.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2819, pruned_loss=0.08172, over 957783.64 frames. ], batch size: 19, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:30:26,304 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4968, 3.3219, 3.1686, 1.6836, 3.4691, 2.4784, 0.6850, 2.2608], + device='cuda:1'), covar=tensor([0.2719, 0.1879, 0.1665, 0.3161, 0.1110, 0.1079, 0.4607, 0.1563], + device='cuda:1'), in_proj_covar=tensor([0.0156, 0.0171, 0.0163, 0.0128, 0.0155, 0.0122, 0.0146, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 04:30:45,632 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3543, 1.4785, 1.5656, 1.7446, 1.6279, 3.3543, 1.4400, 1.5622], + device='cuda:1'), covar=tensor([0.1114, 0.1857, 0.1162, 0.1071, 0.1688, 0.0263, 0.1441, 0.1809], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0082, 0.0077, 0.0080, 0.0093, 0.0083, 0.0086, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 04:30:46,279 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25203.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:30:47,453 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25204.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:30:53,042 INFO [finetune.py:976] (1/7) Epoch 5, batch 2300, loss[loss=0.2466, simple_loss=0.296, pruned_loss=0.09862, over 4784.00 frames. ], tot_loss[loss=0.2237, simple_loss=0.2832, pruned_loss=0.08207, over 958419.37 frames. ], batch size: 26, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:31:05,175 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.234e+02 1.840e+02 2.117e+02 2.638e+02 5.911e+02, threshold=4.234e+02, percent-clipped=5.0 +2023-03-26 04:31:14,023 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25225.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:31:35,989 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=25251.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:31:40,665 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7900, 1.2213, 0.8890, 1.6960, 2.0883, 1.4344, 1.6054, 1.7689], + device='cuda:1'), covar=tensor([0.1440, 0.2063, 0.2181, 0.1175, 0.2039, 0.2109, 0.1448, 0.1891], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0116, 0.0093, 0.0124, 0.0097, 0.0101, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 04:31:42,890 INFO [finetune.py:976] (1/7) Epoch 5, batch 2350, loss[loss=0.2282, simple_loss=0.278, pruned_loss=0.08919, over 4800.00 frames. ], tot_loss[loss=0.2234, simple_loss=0.282, pruned_loss=0.08241, over 955397.99 frames. ], batch size: 45, lr: 3.94e-03, grad_scale: 64.0 +2023-03-26 04:31:51,256 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=25273.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:32:16,768 INFO [finetune.py:976] (1/7) Epoch 5, batch 2400, loss[loss=0.1809, simple_loss=0.2419, pruned_loss=0.06, over 4801.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2782, pruned_loss=0.08091, over 955693.55 frames. ], batch size: 25, lr: 3.94e-03, grad_scale: 64.0 +2023-03-26 04:32:23,861 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.140e+02 1.631e+02 1.900e+02 2.318e+02 5.058e+02, threshold=3.799e+02, percent-clipped=1.0 +2023-03-26 04:32:58,151 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25351.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 04:33:04,183 INFO [finetune.py:976] (1/7) Epoch 5, batch 2450, loss[loss=0.3838, simple_loss=0.3784, pruned_loss=0.1946, over 4207.00 frames. ], tot_loss[loss=0.2174, simple_loss=0.2754, pruned_loss=0.07972, over 956735.77 frames. ], batch size: 65, lr: 3.94e-03, grad_scale: 64.0 +2023-03-26 04:33:51,186 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-26 04:34:02,098 INFO [finetune.py:976] (1/7) Epoch 5, batch 2500, loss[loss=0.1932, simple_loss=0.2674, pruned_loss=0.05946, over 4819.00 frames. ], tot_loss[loss=0.219, simple_loss=0.2768, pruned_loss=0.08063, over 955171.32 frames. ], batch size: 39, lr: 3.94e-03, grad_scale: 64.0 +2023-03-26 04:34:18,830 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.080e+02 1.793e+02 2.115e+02 2.620e+02 5.379e+02, threshold=4.229e+02, percent-clipped=6.0 +2023-03-26 04:34:18,925 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25421.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:34:18,969 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6478, 1.5635, 1.4270, 1.6739, 1.9338, 1.6662, 1.1020, 1.3706], + device='cuda:1'), covar=tensor([0.2435, 0.2282, 0.2056, 0.1895, 0.1900, 0.1219, 0.3038, 0.1953], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0209, 0.0198, 0.0184, 0.0235, 0.0174, 0.0214, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:34:24,845 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5769, 3.4949, 3.2875, 1.5881, 3.5671, 2.7543, 0.9141, 2.4671], + device='cuda:1'), covar=tensor([0.2950, 0.2088, 0.1580, 0.3536, 0.1091, 0.0969, 0.4405, 0.1568], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0172, 0.0164, 0.0129, 0.0156, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 04:34:26,151 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=25428.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 04:34:47,668 INFO [finetune.py:976] (1/7) Epoch 5, batch 2550, loss[loss=0.2084, simple_loss=0.2805, pruned_loss=0.06817, over 4834.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2814, pruned_loss=0.08234, over 955848.49 frames. ], batch size: 49, lr: 3.94e-03, grad_scale: 64.0 +2023-03-26 04:34:51,887 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7667, 1.5725, 1.5079, 1.6452, 1.4113, 4.1346, 1.8710, 2.4435], + device='cuda:1'), covar=tensor([0.4279, 0.3258, 0.2386, 0.2811, 0.1766, 0.0191, 0.2342, 0.1189], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0117, 0.0122, 0.0117, 0.0098, 0.0101, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 04:34:53,567 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=25469.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:35:07,243 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25489.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 04:35:16,169 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25504.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:35:20,834 INFO [finetune.py:976] (1/7) Epoch 5, batch 2600, loss[loss=0.2379, simple_loss=0.3041, pruned_loss=0.0859, over 4820.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2826, pruned_loss=0.08248, over 953589.49 frames. ], batch size: 33, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:35:28,043 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.057e+02 1.693e+02 2.088e+02 2.425e+02 4.415e+02, threshold=4.177e+02, percent-clipped=1.0 +2023-03-26 04:35:48,649 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=25552.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:35:52,439 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9744, 1.5440, 1.7562, 1.7820, 1.5983, 1.5741, 1.7123, 1.6061], + device='cuda:1'), covar=tensor([0.6593, 0.9499, 0.7485, 0.8970, 0.9552, 0.7180, 1.1927, 0.6983], + device='cuda:1'), in_proj_covar=tensor([0.0228, 0.0249, 0.0255, 0.0259, 0.0242, 0.0219, 0.0275, 0.0223], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:35:54,565 INFO [finetune.py:976] (1/7) Epoch 5, batch 2650, loss[loss=0.2421, simple_loss=0.2966, pruned_loss=0.09379, over 4838.00 frames. ], tot_loss[loss=0.2242, simple_loss=0.2833, pruned_loss=0.08255, over 953644.43 frames. ], batch size: 44, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:36:13,436 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7501, 2.5378, 3.1353, 4.5739, 3.3394, 3.3562, 1.6317, 3.7234], + device='cuda:1'), covar=tensor([0.1517, 0.1373, 0.1245, 0.0410, 0.0684, 0.1150, 0.1807, 0.0503], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0119, 0.0136, 0.0166, 0.0102, 0.0142, 0.0128, 0.0104], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 04:36:31,006 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2769, 2.0845, 1.6978, 0.7426, 1.8738, 1.8366, 1.5689, 1.8491], + device='cuda:1'), covar=tensor([0.0808, 0.0864, 0.1509, 0.2109, 0.1258, 0.2213, 0.2262, 0.0965], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0199, 0.0202, 0.0188, 0.0215, 0.0207, 0.0218, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:36:33,947 INFO [finetune.py:976] (1/7) Epoch 5, batch 2700, loss[loss=0.1968, simple_loss=0.259, pruned_loss=0.06734, over 4813.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.283, pruned_loss=0.08228, over 953094.69 frames. ], batch size: 39, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:36:50,928 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.205e+02 1.711e+02 2.002e+02 2.331e+02 3.948e+02, threshold=4.004e+02, percent-clipped=0.0 +2023-03-26 04:37:26,215 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25651.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 04:37:32,345 INFO [finetune.py:976] (1/7) Epoch 5, batch 2750, loss[loss=0.2446, simple_loss=0.2854, pruned_loss=0.1018, over 4833.00 frames. ], tot_loss[loss=0.2225, simple_loss=0.2807, pruned_loss=0.08214, over 954493.97 frames. ], batch size: 47, lr: 3.94e-03, grad_scale: 32.0 +2023-03-26 04:37:37,833 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0127, 1.7677, 1.5338, 1.7106, 1.7242, 1.6298, 1.6242, 2.5365], + device='cuda:1'), covar=tensor([0.7097, 0.8226, 0.5852, 0.7334, 0.6548, 0.4055, 0.7472, 0.2428], + device='cuda:1'), in_proj_covar=tensor([0.0280, 0.0255, 0.0220, 0.0283, 0.0237, 0.0199, 0.0242, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:37:46,693 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1839, 1.3263, 1.6248, 1.0617, 1.1337, 1.4151, 1.3659, 1.5263], + device='cuda:1'), covar=tensor([0.1566, 0.2450, 0.1384, 0.1812, 0.1156, 0.1507, 0.3176, 0.1135], + device='cuda:1'), in_proj_covar=tensor([0.0208, 0.0208, 0.0204, 0.0198, 0.0186, 0.0225, 0.0218, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:37:58,589 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=25699.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:38:07,099 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8033, 3.6674, 3.4463, 1.6885, 3.7949, 2.8772, 0.8077, 2.4871], + device='cuda:1'), covar=tensor([0.2278, 0.1938, 0.1568, 0.3620, 0.1053, 0.0968, 0.4683, 0.1809], + device='cuda:1'), in_proj_covar=tensor([0.0156, 0.0173, 0.0164, 0.0129, 0.0156, 0.0123, 0.0146, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 04:38:07,649 INFO [finetune.py:976] (1/7) Epoch 5, batch 2800, loss[loss=0.1774, simple_loss=0.2456, pruned_loss=0.05463, over 4887.00 frames. ], tot_loss[loss=0.2184, simple_loss=0.2767, pruned_loss=0.07999, over 956505.79 frames. ], batch size: 32, lr: 3.93e-03, grad_scale: 32.0 +2023-03-26 04:38:23,888 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.133e+02 1.611e+02 1.888e+02 2.301e+02 3.388e+02, threshold=3.776e+02, percent-clipped=0.0 +2023-03-26 04:39:03,401 INFO [finetune.py:976] (1/7) Epoch 5, batch 2850, loss[loss=0.221, simple_loss=0.283, pruned_loss=0.07955, over 4900.00 frames. ], tot_loss[loss=0.2168, simple_loss=0.2747, pruned_loss=0.07942, over 957334.54 frames. ], batch size: 32, lr: 3.93e-03, grad_scale: 32.0 +2023-03-26 04:39:03,576 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 04:39:18,029 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25784.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 04:39:29,158 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2528, 1.2078, 1.1422, 1.2911, 1.5061, 1.4173, 1.3063, 1.1477], + device='cuda:1'), covar=tensor([0.0311, 0.0253, 0.0515, 0.0248, 0.0190, 0.0397, 0.0261, 0.0346], + device='cuda:1'), in_proj_covar=tensor([0.0088, 0.0114, 0.0140, 0.0119, 0.0106, 0.0102, 0.0092, 0.0110], + device='cuda:1'), out_proj_covar=tensor([6.8535e-05, 8.9402e-05, 1.1239e-04, 9.3971e-05, 8.3894e-05, 7.5969e-05, + 7.0109e-05, 8.5833e-05], device='cuda:1') +2023-03-26 04:39:37,534 INFO [finetune.py:976] (1/7) Epoch 5, batch 2900, loss[loss=0.2293, simple_loss=0.289, pruned_loss=0.08478, over 4247.00 frames. ], tot_loss[loss=0.2207, simple_loss=0.2785, pruned_loss=0.08146, over 955448.41 frames. ], batch size: 65, lr: 3.93e-03, grad_scale: 32.0 +2023-03-26 04:39:38,887 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1498, 1.5415, 0.9660, 2.0339, 2.1896, 1.8557, 1.8030, 1.9843], + device='cuda:1'), covar=tensor([0.1190, 0.1814, 0.2245, 0.0982, 0.1870, 0.1945, 0.1270, 0.1589], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0116, 0.0093, 0.0124, 0.0097, 0.0101, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 04:39:44,762 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.812e+02 2.065e+02 2.463e+02 5.082e+02, threshold=4.130e+02, percent-clipped=4.0 +2023-03-26 04:40:09,019 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=25858.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 04:40:10,734 INFO [finetune.py:976] (1/7) Epoch 5, batch 2950, loss[loss=0.2661, simple_loss=0.3264, pruned_loss=0.1029, over 4806.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2818, pruned_loss=0.08287, over 956253.37 frames. ], batch size: 41, lr: 3.93e-03, grad_scale: 32.0 +2023-03-26 04:40:35,002 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4899, 3.4580, 3.2556, 1.3761, 3.5824, 2.5645, 0.7765, 2.1511], + device='cuda:1'), covar=tensor([0.2384, 0.1862, 0.1498, 0.3653, 0.1069, 0.1087, 0.4359, 0.1677], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0174, 0.0165, 0.0130, 0.0157, 0.0124, 0.0147, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 04:40:43,942 INFO [finetune.py:976] (1/7) Epoch 5, batch 3000, loss[loss=0.2492, simple_loss=0.3133, pruned_loss=0.09257, over 4815.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.2835, pruned_loss=0.083, over 956217.62 frames. ], batch size: 38, lr: 3.93e-03, grad_scale: 32.0 +2023-03-26 04:40:43,942 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 04:40:53,589 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8661, 1.9184, 1.9683, 1.2996, 2.0958, 1.9957, 1.9144, 1.7350], + device='cuda:1'), covar=tensor([0.0642, 0.0615, 0.0642, 0.0896, 0.0600, 0.0768, 0.0657, 0.1042], + device='cuda:1'), in_proj_covar=tensor([0.0138, 0.0133, 0.0143, 0.0127, 0.0110, 0.0142, 0.0145, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:40:54,557 INFO [finetune.py:1010] (1/7) Epoch 5, validation: loss=0.1652, simple_loss=0.2371, pruned_loss=0.04667, over 2265189.00 frames. +2023-03-26 04:40:54,557 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6290MB +2023-03-26 04:40:55,260 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6720, 1.5841, 1.5257, 1.6446, 1.0136, 3.6351, 1.4739, 1.9054], + device='cuda:1'), covar=tensor([0.3389, 0.2396, 0.2068, 0.2262, 0.1942, 0.0176, 0.2678, 0.1368], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0117, 0.0122, 0.0118, 0.0098, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 04:40:58,812 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5405, 1.4877, 1.9146, 2.8431, 2.0598, 2.0824, 0.9434, 2.2838], + device='cuda:1'), covar=tensor([0.1715, 0.1500, 0.1188, 0.0538, 0.0755, 0.1346, 0.1891, 0.0672], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0119, 0.0137, 0.0167, 0.0103, 0.0143, 0.0129, 0.0105], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 04:41:00,059 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6508, 1.6031, 2.2950, 3.4276, 2.3874, 2.5117, 1.0763, 2.7522], + device='cuda:1'), covar=tensor([0.1919, 0.1649, 0.1311, 0.0566, 0.0799, 0.1316, 0.2112, 0.0604], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0119, 0.0137, 0.0167, 0.0103, 0.0143, 0.0129, 0.0105], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 04:41:00,097 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25919.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 04:41:01,808 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.188e+02 1.739e+02 2.096e+02 2.435e+02 4.160e+02, threshold=4.193e+02, percent-clipped=2.0 +2023-03-26 04:41:27,994 INFO [finetune.py:976] (1/7) Epoch 5, batch 3050, loss[loss=0.2271, simple_loss=0.2855, pruned_loss=0.08433, over 4927.00 frames. ], tot_loss[loss=0.2253, simple_loss=0.2845, pruned_loss=0.08301, over 956607.79 frames. ], batch size: 41, lr: 3.93e-03, grad_scale: 32.0 +2023-03-26 04:41:33,446 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8909, 1.6210, 1.5651, 1.8287, 2.2283, 1.9018, 1.3403, 1.5261], + device='cuda:1'), covar=tensor([0.2323, 0.2296, 0.1999, 0.1842, 0.1913, 0.1202, 0.2853, 0.1991], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0208, 0.0199, 0.0183, 0.0235, 0.0173, 0.0214, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:41:47,137 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9927, 2.2105, 1.9840, 1.5902, 2.3343, 2.2385, 2.0910, 1.8487], + device='cuda:1'), covar=tensor([0.0574, 0.0481, 0.0703, 0.0825, 0.0410, 0.0683, 0.0575, 0.0897], + device='cuda:1'), in_proj_covar=tensor([0.0138, 0.0133, 0.0144, 0.0128, 0.0110, 0.0143, 0.0146, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:41:47,755 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1593, 1.3026, 1.4394, 0.7538, 1.2452, 1.5453, 1.5984, 1.3463], + device='cuda:1'), covar=tensor([0.1211, 0.0672, 0.0547, 0.0612, 0.0521, 0.0728, 0.0380, 0.0760], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0160, 0.0120, 0.0138, 0.0134, 0.0124, 0.0149, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.9395e-05, 1.1846e-04, 8.6931e-05, 1.0096e-04, 9.6901e-05, 9.1632e-05, + 1.1107e-04, 1.0877e-04], device='cuda:1') +2023-03-26 04:42:08,229 INFO [finetune.py:976] (1/7) Epoch 5, batch 3100, loss[loss=0.2176, simple_loss=0.2723, pruned_loss=0.08139, over 4739.00 frames. ], tot_loss[loss=0.2228, simple_loss=0.2821, pruned_loss=0.08171, over 956086.03 frames. ], batch size: 54, lr: 3.93e-03, grad_scale: 32.0 +2023-03-26 04:42:17,132 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.08 vs. limit=5.0 +2023-03-26 04:42:25,414 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.176e+02 1.603e+02 1.879e+02 2.413e+02 4.411e+02, threshold=3.758e+02, percent-clipped=2.0 +2023-03-26 04:42:58,351 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.89 vs. limit=5.0 +2023-03-26 04:43:10,313 INFO [finetune.py:976] (1/7) Epoch 5, batch 3150, loss[loss=0.2569, simple_loss=0.2948, pruned_loss=0.1095, over 4890.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2787, pruned_loss=0.08046, over 957456.71 frames. ], batch size: 35, lr: 3.93e-03, grad_scale: 32.0 +2023-03-26 04:43:25,553 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-26 04:43:27,800 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0086, 4.7250, 4.4982, 2.5157, 4.7637, 3.7021, 1.3874, 3.3423], + device='cuda:1'), covar=tensor([0.2107, 0.1781, 0.1143, 0.2778, 0.0702, 0.0838, 0.4068, 0.1306], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0174, 0.0166, 0.0130, 0.0157, 0.0123, 0.0147, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 04:43:30,937 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26084.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 04:43:42,327 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26102.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:43:49,556 INFO [finetune.py:976] (1/7) Epoch 5, batch 3200, loss[loss=0.2176, simple_loss=0.2745, pruned_loss=0.08038, over 4823.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2755, pruned_loss=0.07958, over 956060.74 frames. ], batch size: 39, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:43:58,348 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.040e+02 1.599e+02 1.999e+02 2.453e+02 4.323e+02, threshold=3.997e+02, percent-clipped=1.0 +2023-03-26 04:44:01,974 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8896, 4.3515, 4.1291, 2.2603, 4.4483, 3.4651, 1.1658, 2.9849], + device='cuda:1'), covar=tensor([0.2411, 0.1814, 0.1276, 0.3114, 0.0820, 0.0814, 0.4282, 0.1346], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0173, 0.0165, 0.0130, 0.0156, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 04:44:04,856 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=26132.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 04:44:05,435 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0713, 4.5996, 4.3351, 2.4886, 4.6696, 3.5599, 0.9814, 3.1923], + device='cuda:1'), covar=tensor([0.2351, 0.1754, 0.1304, 0.2964, 0.0808, 0.0903, 0.4780, 0.1375], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0174, 0.0165, 0.0130, 0.0156, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 04:44:34,957 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2685, 2.7841, 2.5603, 1.3717, 2.7123, 2.2938, 2.0131, 2.2783], + device='cuda:1'), covar=tensor([0.0758, 0.1076, 0.1864, 0.2544, 0.1979, 0.2197, 0.2229, 0.1420], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0202, 0.0205, 0.0192, 0.0219, 0.0211, 0.0223, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:44:36,707 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8983, 4.2022, 3.9701, 2.0744, 4.2143, 3.1817, 1.2961, 2.8964], + device='cuda:1'), covar=tensor([0.2289, 0.1752, 0.1584, 0.3403, 0.0900, 0.0885, 0.4487, 0.1454], + device='cuda:1'), in_proj_covar=tensor([0.0157, 0.0174, 0.0166, 0.0130, 0.0157, 0.0124, 0.0148, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 04:44:37,841 INFO [finetune.py:976] (1/7) Epoch 5, batch 3250, loss[loss=0.2193, simple_loss=0.2853, pruned_loss=0.07661, over 4873.00 frames. ], tot_loss[loss=0.2181, simple_loss=0.2757, pruned_loss=0.08024, over 955050.09 frames. ], batch size: 34, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:44:43,924 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26163.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:44:46,752 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5711, 1.5993, 1.8804, 1.2132, 1.5901, 1.7792, 1.5747, 1.9776], + device='cuda:1'), covar=tensor([0.1419, 0.2049, 0.1224, 0.1722, 0.1043, 0.1329, 0.2528, 0.0799], + device='cuda:1'), in_proj_covar=tensor([0.0207, 0.0207, 0.0204, 0.0198, 0.0186, 0.0226, 0.0218, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:44:47,962 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26168.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:44:48,578 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5205, 1.6606, 2.0640, 1.7534, 1.8633, 4.4391, 1.4550, 1.9607], + device='cuda:1'), covar=tensor([0.1114, 0.1789, 0.1196, 0.1174, 0.1634, 0.0175, 0.1555, 0.1717], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0081, 0.0077, 0.0080, 0.0092, 0.0082, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 04:44:59,076 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7703, 3.2546, 3.3989, 3.6599, 3.5067, 3.2553, 3.8516, 1.2827], + device='cuda:1'), covar=tensor([0.0930, 0.0895, 0.0966, 0.1030, 0.1443, 0.1662, 0.0817, 0.4953], + device='cuda:1'), in_proj_covar=tensor([0.0354, 0.0243, 0.0273, 0.0289, 0.0335, 0.0281, 0.0301, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:45:40,774 INFO [finetune.py:976] (1/7) Epoch 5, batch 3300, loss[loss=0.2404, simple_loss=0.2924, pruned_loss=0.09424, over 4865.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2798, pruned_loss=0.08158, over 953804.01 frames. ], batch size: 44, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:45:47,647 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26214.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 04:46:00,068 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.165e+02 1.721e+02 2.149e+02 2.490e+02 4.939e+02, threshold=4.298e+02, percent-clipped=1.0 +2023-03-26 04:46:06,002 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26229.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:46:15,012 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 04:46:19,855 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26250.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:46:28,378 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1924, 1.4545, 0.9356, 2.0585, 2.4473, 1.9103, 1.7497, 2.0136], + device='cuda:1'), covar=tensor([0.1368, 0.2003, 0.2144, 0.1082, 0.1818, 0.1853, 0.1330, 0.1772], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0116, 0.0093, 0.0123, 0.0097, 0.0100, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 04:46:29,545 INFO [finetune.py:976] (1/7) Epoch 5, batch 3350, loss[loss=0.1927, simple_loss=0.2715, pruned_loss=0.05697, over 4852.00 frames. ], tot_loss[loss=0.2221, simple_loss=0.2809, pruned_loss=0.08158, over 952677.61 frames. ], batch size: 44, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:46:38,855 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-26 04:46:49,796 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3243, 2.8342, 2.1502, 1.7250, 2.7527, 2.8228, 2.5567, 2.3653], + device='cuda:1'), covar=tensor([0.0658, 0.0423, 0.0847, 0.0938, 0.0608, 0.0638, 0.0604, 0.0826], + device='cuda:1'), in_proj_covar=tensor([0.0140, 0.0134, 0.0146, 0.0129, 0.0112, 0.0145, 0.0147, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:47:12,172 INFO [finetune.py:976] (1/7) Epoch 5, batch 3400, loss[loss=0.2647, simple_loss=0.3167, pruned_loss=0.1063, over 4859.00 frames. ], tot_loss[loss=0.223, simple_loss=0.282, pruned_loss=0.08197, over 952435.52 frames. ], batch size: 44, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:47:12,292 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26311.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:47:19,914 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.097e+02 1.624e+02 1.904e+02 2.361e+02 4.543e+02, threshold=3.807e+02, percent-clipped=1.0 +2023-03-26 04:47:42,288 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.00 vs. limit=5.0 +2023-03-26 04:47:53,149 INFO [finetune.py:976] (1/7) Epoch 5, batch 3450, loss[loss=0.2138, simple_loss=0.276, pruned_loss=0.07582, over 4841.00 frames. ], tot_loss[loss=0.2233, simple_loss=0.2826, pruned_loss=0.08201, over 953208.27 frames. ], batch size: 49, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:47:55,116 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6988, 1.6437, 1.5484, 1.7939, 2.1960, 1.6924, 1.4418, 1.4046], + device='cuda:1'), covar=tensor([0.2194, 0.2110, 0.1771, 0.1673, 0.1758, 0.1207, 0.2605, 0.1747], + device='cuda:1'), in_proj_covar=tensor([0.0234, 0.0208, 0.0198, 0.0183, 0.0235, 0.0174, 0.0214, 0.0187], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:48:37,958 INFO [finetune.py:976] (1/7) Epoch 5, batch 3500, loss[loss=0.1654, simple_loss=0.2281, pruned_loss=0.0513, over 4752.00 frames. ], tot_loss[loss=0.2211, simple_loss=0.28, pruned_loss=0.08105, over 953739.41 frames. ], batch size: 27, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:48:39,217 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9877, 1.8697, 1.5179, 1.7148, 1.8881, 1.6117, 2.2239, 1.9225], + device='cuda:1'), covar=tensor([0.1743, 0.2986, 0.3891, 0.3528, 0.3249, 0.2069, 0.4956, 0.2168], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0193, 0.0236, 0.0254, 0.0230, 0.0189, 0.0210, 0.0190], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:48:43,893 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26414.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:48:51,309 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.234e+02 1.586e+02 1.962e+02 2.229e+02 4.326e+02, threshold=3.925e+02, percent-clipped=1.0 +2023-03-26 04:49:23,758 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26458.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:49:25,518 INFO [finetune.py:976] (1/7) Epoch 5, batch 3550, loss[loss=0.1917, simple_loss=0.2521, pruned_loss=0.06566, over 4891.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2768, pruned_loss=0.07964, over 954490.87 frames. ], batch size: 35, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:49:34,678 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26475.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:49:47,633 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8805, 1.8087, 1.8254, 1.2874, 1.9030, 1.9780, 1.8936, 1.6031], + device='cuda:1'), covar=tensor([0.0438, 0.0468, 0.0527, 0.0756, 0.0592, 0.0476, 0.0461, 0.0897], + device='cuda:1'), in_proj_covar=tensor([0.0140, 0.0134, 0.0145, 0.0129, 0.0112, 0.0145, 0.0147, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:50:11,928 INFO [finetune.py:976] (1/7) Epoch 5, batch 3600, loss[loss=0.2598, simple_loss=0.3051, pruned_loss=0.1072, over 4820.00 frames. ], tot_loss[loss=0.217, simple_loss=0.2749, pruned_loss=0.07954, over 955663.31 frames. ], batch size: 38, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:50:13,371 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-26 04:50:13,868 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26514.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 04:50:19,778 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.118e+02 1.588e+02 1.920e+02 2.290e+02 3.397e+02, threshold=3.841e+02, percent-clipped=0.0 +2023-03-26 04:50:20,510 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26524.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:50:33,087 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6523, 1.5139, 1.4925, 1.5543, 1.1000, 3.5898, 1.3540, 1.9181], + device='cuda:1'), covar=tensor([0.3343, 0.2432, 0.2108, 0.2270, 0.1863, 0.0167, 0.2690, 0.1336], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0114, 0.0116, 0.0121, 0.0117, 0.0097, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 04:50:55,027 INFO [finetune.py:976] (1/7) Epoch 5, batch 3650, loss[loss=0.2432, simple_loss=0.3043, pruned_loss=0.091, over 4899.00 frames. ], tot_loss[loss=0.2192, simple_loss=0.2769, pruned_loss=0.08077, over 954137.88 frames. ], batch size: 35, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:50:55,700 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=26562.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 04:51:31,365 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26606.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:51:34,854 INFO [finetune.py:976] (1/7) Epoch 5, batch 3700, loss[loss=0.2294, simple_loss=0.2807, pruned_loss=0.08904, over 4823.00 frames. ], tot_loss[loss=0.2239, simple_loss=0.2823, pruned_loss=0.08273, over 955640.52 frames. ], batch size: 33, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:51:42,594 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.101e+02 1.866e+02 2.222e+02 2.784e+02 4.852e+02, threshold=4.444e+02, percent-clipped=6.0 +2023-03-26 04:52:07,673 INFO [finetune.py:976] (1/7) Epoch 5, batch 3750, loss[loss=0.2815, simple_loss=0.3289, pruned_loss=0.1171, over 4814.00 frames. ], tot_loss[loss=0.2265, simple_loss=0.2846, pruned_loss=0.0842, over 955286.19 frames. ], batch size: 39, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:52:42,414 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0361, 1.7204, 2.5808, 3.8288, 2.6523, 2.5374, 0.9484, 2.9564], + device='cuda:1'), covar=tensor([0.1814, 0.1711, 0.1396, 0.0679, 0.0843, 0.1819, 0.2274, 0.0628], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0119, 0.0137, 0.0166, 0.0102, 0.0142, 0.0129, 0.0104], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 04:53:00,962 INFO [finetune.py:976] (1/7) Epoch 5, batch 3800, loss[loss=0.2393, simple_loss=0.2957, pruned_loss=0.09147, over 4884.00 frames. ], tot_loss[loss=0.2279, simple_loss=0.2859, pruned_loss=0.08494, over 957101.68 frames. ], batch size: 32, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:53:08,702 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.110e+02 1.710e+02 2.076e+02 2.649e+02 5.488e+02, threshold=4.152e+02, percent-clipped=2.0 +2023-03-26 04:53:30,532 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6518, 1.5975, 1.9344, 1.1830, 1.6456, 1.8616, 1.5725, 2.0275], + device='cuda:1'), covar=tensor([0.1265, 0.2230, 0.1333, 0.1934, 0.1016, 0.1454, 0.2717, 0.0978], + device='cuda:1'), in_proj_covar=tensor([0.0204, 0.0204, 0.0201, 0.0196, 0.0184, 0.0223, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:53:57,418 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26758.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:53:59,695 INFO [finetune.py:976] (1/7) Epoch 5, batch 3850, loss[loss=0.275, simple_loss=0.3194, pruned_loss=0.1153, over 4131.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.2844, pruned_loss=0.08371, over 955302.73 frames. ], batch size: 65, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:54:11,276 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26770.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:55:00,183 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=26806.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:55:03,123 INFO [finetune.py:976] (1/7) Epoch 5, batch 3900, loss[loss=0.2017, simple_loss=0.258, pruned_loss=0.07276, over 4822.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2809, pruned_loss=0.08258, over 952704.65 frames. ], batch size: 40, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:55:21,751 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.223e+02 1.797e+02 2.146e+02 2.516e+02 4.177e+02, threshold=4.292e+02, percent-clipped=1.0 +2023-03-26 04:55:22,490 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26824.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:55:25,537 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1937, 3.5946, 3.8135, 4.0223, 3.9511, 3.6975, 4.3046, 1.3743], + device='cuda:1'), covar=tensor([0.0832, 0.0901, 0.0826, 0.1022, 0.1337, 0.1687, 0.0661, 0.4918], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0241, 0.0271, 0.0290, 0.0333, 0.0281, 0.0300, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:55:57,245 INFO [finetune.py:976] (1/7) Epoch 5, batch 3950, loss[loss=0.1907, simple_loss=0.2492, pruned_loss=0.06609, over 4775.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2759, pruned_loss=0.08008, over 954940.49 frames. ], batch size: 28, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:55:59,912 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.19 vs. limit=5.0 +2023-03-26 04:56:06,702 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=26872.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:56:13,462 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8983, 1.3863, 0.8234, 1.7469, 2.1501, 1.2976, 1.5575, 1.7463], + device='cuda:1'), covar=tensor([0.1480, 0.2220, 0.2232, 0.1191, 0.2001, 0.2135, 0.1561, 0.2124], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0115, 0.0094, 0.0123, 0.0096, 0.0101, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 04:56:23,182 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6681, 1.6106, 1.5508, 1.8757, 2.0873, 1.8380, 1.2951, 1.4757], + device='cuda:1'), covar=tensor([0.2548, 0.2298, 0.2046, 0.1746, 0.1766, 0.1231, 0.2927, 0.2011], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0208, 0.0199, 0.0183, 0.0234, 0.0174, 0.0213, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:56:29,752 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26904.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 04:56:30,923 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26906.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:56:34,935 INFO [finetune.py:976] (1/7) Epoch 5, batch 4000, loss[loss=0.2458, simple_loss=0.3044, pruned_loss=0.09359, over 4922.00 frames. ], tot_loss[loss=0.2194, simple_loss=0.2763, pruned_loss=0.08126, over 954985.32 frames. ], batch size: 43, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:56:43,169 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.227e+02 1.662e+02 2.009e+02 2.453e+02 3.802e+02, threshold=4.018e+02, percent-clipped=0.0 +2023-03-26 04:57:05,886 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26950.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:57:08,737 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=26954.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:57:18,757 INFO [finetune.py:976] (1/7) Epoch 5, batch 4050, loss[loss=0.1785, simple_loss=0.2439, pruned_loss=0.0565, over 4724.00 frames. ], tot_loss[loss=0.2207, simple_loss=0.2783, pruned_loss=0.08157, over 954137.89 frames. ], batch size: 23, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:57:26,826 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26965.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 04:57:30,425 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6703, 1.5626, 1.4557, 1.7621, 2.2952, 1.7461, 1.5571, 1.3223], + device='cuda:1'), covar=tensor([0.2425, 0.2328, 0.2144, 0.1988, 0.1877, 0.1290, 0.2756, 0.2091], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0208, 0.0199, 0.0183, 0.0235, 0.0174, 0.0214, 0.0188], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:58:01,907 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-26 04:58:10,819 INFO [finetune.py:976] (1/7) Epoch 5, batch 4100, loss[loss=0.1887, simple_loss=0.2594, pruned_loss=0.05901, over 4780.00 frames. ], tot_loss[loss=0.2253, simple_loss=0.2835, pruned_loss=0.08355, over 956715.91 frames. ], batch size: 26, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:58:11,439 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27011.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:58:19,120 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9603, 1.2034, 1.5643, 1.6341, 1.3945, 1.4920, 1.5210, 1.5467], + device='cuda:1'), covar=tensor([0.7202, 0.9444, 0.8829, 0.8938, 1.1248, 0.8223, 1.1964, 0.8718], + device='cuda:1'), in_proj_covar=tensor([0.0228, 0.0247, 0.0254, 0.0258, 0.0241, 0.0218, 0.0273, 0.0223], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 04:58:19,549 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.223e+02 1.740e+02 2.111e+02 2.577e+02 5.326e+02, threshold=4.223e+02, percent-clipped=3.0 +2023-03-26 04:58:58,625 INFO [finetune.py:976] (1/7) Epoch 5, batch 4150, loss[loss=0.2572, simple_loss=0.3132, pruned_loss=0.1006, over 4803.00 frames. ], tot_loss[loss=0.2279, simple_loss=0.2859, pruned_loss=0.08494, over 956357.02 frames. ], batch size: 51, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:59:05,642 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27070.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:59:32,487 INFO [finetune.py:976] (1/7) Epoch 5, batch 4200, loss[loss=0.2264, simple_loss=0.2926, pruned_loss=0.08007, over 4894.00 frames. ], tot_loss[loss=0.2264, simple_loss=0.2852, pruned_loss=0.08377, over 956771.88 frames. ], batch size: 37, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 04:59:37,724 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=27118.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 04:59:41,614 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.001e+02 1.682e+02 1.955e+02 2.295e+02 5.538e+02, threshold=3.911e+02, percent-clipped=3.0 +2023-03-26 04:59:52,341 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7986, 1.2262, 0.8066, 1.7230, 2.2032, 1.4530, 1.4736, 1.6967], + device='cuda:1'), covar=tensor([0.1559, 0.2231, 0.2295, 0.1286, 0.1882, 0.2012, 0.1502, 0.1996], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0115, 0.0094, 0.0123, 0.0096, 0.0100, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 04:59:57,810 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=27148.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:00:10,425 INFO [finetune.py:976] (1/7) Epoch 5, batch 4250, loss[loss=0.2543, simple_loss=0.3056, pruned_loss=0.1015, over 4816.00 frames. ], tot_loss[loss=0.2232, simple_loss=0.2819, pruned_loss=0.08223, over 956162.38 frames. ], batch size: 40, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:01:08,498 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27209.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:01:09,574 INFO [finetune.py:976] (1/7) Epoch 5, batch 4300, loss[loss=0.2063, simple_loss=0.2655, pruned_loss=0.07355, over 4824.00 frames. ], tot_loss[loss=0.2209, simple_loss=0.2788, pruned_loss=0.0815, over 956095.39 frames. ], batch size: 40, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:01:26,945 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.146e+02 1.762e+02 2.023e+02 2.453e+02 1.035e+03, threshold=4.046e+02, percent-clipped=2.0 +2023-03-26 05:01:27,083 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2383, 1.8602, 1.4329, 0.5385, 1.7274, 1.8680, 1.6447, 1.7575], + device='cuda:1'), covar=tensor([0.0819, 0.0903, 0.1605, 0.2191, 0.1451, 0.2465, 0.2409, 0.1015], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0203, 0.0204, 0.0191, 0.0219, 0.0211, 0.0223, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:01:59,784 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=27260.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 05:02:00,300 INFO [finetune.py:976] (1/7) Epoch 5, batch 4350, loss[loss=0.2561, simple_loss=0.3123, pruned_loss=0.09995, over 4843.00 frames. ], tot_loss[loss=0.2172, simple_loss=0.275, pruned_loss=0.07972, over 956006.75 frames. ], batch size: 49, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:02:30,556 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=27306.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:02:33,998 INFO [finetune.py:976] (1/7) Epoch 5, batch 4400, loss[loss=0.2119, simple_loss=0.2914, pruned_loss=0.06618, over 4903.00 frames. ], tot_loss[loss=0.217, simple_loss=0.2749, pruned_loss=0.07953, over 952919.94 frames. ], batch size: 36, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:02:41,212 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.108e+02 1.601e+02 1.888e+02 2.389e+02 3.644e+02, threshold=3.775e+02, percent-clipped=0.0 +2023-03-26 05:03:07,514 INFO [finetune.py:976] (1/7) Epoch 5, batch 4450, loss[loss=0.256, simple_loss=0.3211, pruned_loss=0.09551, over 4920.00 frames. ], tot_loss[loss=0.2218, simple_loss=0.2802, pruned_loss=0.08171, over 954988.85 frames. ], batch size: 38, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:03:25,200 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4186, 1.2959, 1.3177, 1.3219, 0.7552, 2.2344, 0.8351, 1.1894], + device='cuda:1'), covar=tensor([0.3760, 0.2707, 0.2292, 0.2586, 0.2376, 0.0425, 0.2859, 0.1602], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0117, 0.0122, 0.0117, 0.0098, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 05:03:33,768 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5965, 1.5135, 1.3664, 1.6164, 2.0281, 1.6501, 1.2171, 1.3169], + device='cuda:1'), covar=tensor([0.2341, 0.2230, 0.2106, 0.1955, 0.1744, 0.1264, 0.2822, 0.1978], + device='cuda:1'), in_proj_covar=tensor([0.0235, 0.0210, 0.0202, 0.0185, 0.0236, 0.0176, 0.0215, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:03:40,731 INFO [finetune.py:976] (1/7) Epoch 5, batch 4500, loss[loss=0.1741, simple_loss=0.2325, pruned_loss=0.05781, over 4777.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2815, pruned_loss=0.08197, over 954903.20 frames. ], batch size: 26, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:03:48,439 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.291e+02 1.723e+02 2.077e+02 2.543e+02 6.449e+02, threshold=4.154e+02, percent-clipped=4.0 +2023-03-26 05:04:14,224 INFO [finetune.py:976] (1/7) Epoch 5, batch 4550, loss[loss=0.1953, simple_loss=0.2562, pruned_loss=0.06718, over 4921.00 frames. ], tot_loss[loss=0.2226, simple_loss=0.2818, pruned_loss=0.08169, over 955216.90 frames. ], batch size: 33, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:04:14,451 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-03-26 05:04:42,679 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=27504.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:04:47,381 INFO [finetune.py:976] (1/7) Epoch 5, batch 4600, loss[loss=0.1903, simple_loss=0.2554, pruned_loss=0.06262, over 4868.00 frames. ], tot_loss[loss=0.222, simple_loss=0.2815, pruned_loss=0.08127, over 956618.95 frames. ], batch size: 31, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:04:55,105 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.119e+02 1.596e+02 2.009e+02 2.524e+02 8.514e+02, threshold=4.018e+02, percent-clipped=5.0 +2023-03-26 05:05:18,257 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7267, 1.6400, 1.3230, 1.3665, 1.9259, 1.8749, 1.7681, 1.4243], + device='cuda:1'), covar=tensor([0.0310, 0.0379, 0.0623, 0.0427, 0.0244, 0.0587, 0.0365, 0.0511], + device='cuda:1'), in_proj_covar=tensor([0.0086, 0.0111, 0.0137, 0.0116, 0.0104, 0.0100, 0.0090, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.7021e-05, 8.7268e-05, 1.0983e-04, 9.2123e-05, 8.1861e-05, 7.4333e-05, + 6.8927e-05, 8.4541e-05], device='cuda:1') +2023-03-26 05:05:20,037 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27560.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 05:05:20,541 INFO [finetune.py:976] (1/7) Epoch 5, batch 4650, loss[loss=0.2313, simple_loss=0.2838, pruned_loss=0.08945, over 4873.00 frames. ], tot_loss[loss=0.2193, simple_loss=0.2783, pruned_loss=0.08015, over 955572.03 frames. ], batch size: 34, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:05:36,632 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3653, 1.4961, 1.7910, 1.7276, 1.5643, 3.3974, 1.2820, 1.6467], + device='cuda:1'), covar=tensor([0.1044, 0.1738, 0.1182, 0.1084, 0.1650, 0.0252, 0.1512, 0.1724], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0082, 0.0078, 0.0080, 0.0093, 0.0083, 0.0086, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 05:06:07,095 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27606.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:06:07,183 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0 +2023-03-26 05:06:13,640 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=27608.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 05:06:15,418 INFO [finetune.py:976] (1/7) Epoch 5, batch 4700, loss[loss=0.1584, simple_loss=0.2254, pruned_loss=0.04565, over 4905.00 frames. ], tot_loss[loss=0.2155, simple_loss=0.274, pruned_loss=0.07852, over 954755.69 frames. ], batch size: 35, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:06:27,905 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.549e+02 1.903e+02 2.293e+02 3.137e+02, threshold=3.806e+02, percent-clipped=0.0 +2023-03-26 05:07:05,703 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=27654.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:07:11,571 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.23 vs. limit=5.0 +2023-03-26 05:07:14,262 INFO [finetune.py:976] (1/7) Epoch 5, batch 4750, loss[loss=0.2652, simple_loss=0.3011, pruned_loss=0.1146, over 4781.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.272, pruned_loss=0.07787, over 957263.73 frames. ], batch size: 27, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:07:22,776 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6530, 1.8097, 1.7866, 1.0929, 1.8202, 2.0359, 2.0080, 1.5650], + device='cuda:1'), covar=tensor([0.0904, 0.0540, 0.0461, 0.0570, 0.0379, 0.0458, 0.0313, 0.0648], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0159, 0.0121, 0.0137, 0.0134, 0.0124, 0.0149, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.8772e-05, 1.1753e-04, 8.7586e-05, 9.9959e-05, 9.6265e-05, 9.1542e-05, + 1.1030e-04, 1.0833e-04], device='cuda:1') +2023-03-26 05:07:42,690 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3417, 1.4627, 1.1698, 1.2896, 1.7032, 1.5408, 1.4331, 1.2977], + device='cuda:1'), covar=tensor([0.0351, 0.0335, 0.0564, 0.0327, 0.0230, 0.0499, 0.0329, 0.0408], + device='cuda:1'), in_proj_covar=tensor([0.0087, 0.0112, 0.0138, 0.0117, 0.0104, 0.0101, 0.0091, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.7830e-05, 8.8116e-05, 1.1079e-04, 9.3015e-05, 8.2308e-05, 7.5025e-05, + 6.9456e-05, 8.5263e-05], device='cuda:1') +2023-03-26 05:07:54,134 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 05:07:56,929 INFO [finetune.py:976] (1/7) Epoch 5, batch 4800, loss[loss=0.2709, simple_loss=0.3261, pruned_loss=0.1079, over 4914.00 frames. ], tot_loss[loss=0.2182, simple_loss=0.2762, pruned_loss=0.08012, over 957727.79 frames. ], batch size: 42, lr: 3.93e-03, grad_scale: 16.0 +2023-03-26 05:07:57,110 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.80 vs. limit=5.0 +2023-03-26 05:07:59,317 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=27714.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 05:08:04,701 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.165e+02 1.708e+02 2.066e+02 2.363e+02 4.852e+02, threshold=4.133e+02, percent-clipped=3.0 +2023-03-26 05:08:11,500 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.53 vs. limit=5.0 +2023-03-26 05:08:30,291 INFO [finetune.py:976] (1/7) Epoch 5, batch 4850, loss[loss=0.2302, simple_loss=0.2883, pruned_loss=0.08601, over 4821.00 frames. ], tot_loss[loss=0.2224, simple_loss=0.2809, pruned_loss=0.08194, over 955192.00 frames. ], batch size: 38, lr: 3.92e-03, grad_scale: 16.0 +2023-03-26 05:08:39,490 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27775.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 05:08:59,125 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27804.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:09:04,311 INFO [finetune.py:976] (1/7) Epoch 5, batch 4900, loss[loss=0.2319, simple_loss=0.2736, pruned_loss=0.09508, over 4248.00 frames. ], tot_loss[loss=0.2226, simple_loss=0.281, pruned_loss=0.08214, over 951815.08 frames. ], batch size: 18, lr: 3.92e-03, grad_scale: 16.0 +2023-03-26 05:09:09,112 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6920, 1.4436, 2.0735, 3.3514, 2.3499, 2.3660, 0.8946, 2.6647], + device='cuda:1'), covar=tensor([0.1837, 0.1684, 0.1428, 0.0558, 0.0844, 0.1600, 0.2128, 0.0619], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0118, 0.0136, 0.0166, 0.0103, 0.0143, 0.0128, 0.0103], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 05:09:12,047 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.126e+02 1.628e+02 1.864e+02 2.335e+02 3.818e+02, threshold=3.728e+02, percent-clipped=0.0 +2023-03-26 05:09:30,670 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=27852.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:09:37,157 INFO [finetune.py:976] (1/7) Epoch 5, batch 4950, loss[loss=0.2177, simple_loss=0.2765, pruned_loss=0.07948, over 4787.00 frames. ], tot_loss[loss=0.2234, simple_loss=0.2825, pruned_loss=0.08216, over 953435.64 frames. ], batch size: 29, lr: 3.92e-03, grad_scale: 16.0 +2023-03-26 05:10:08,688 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5951, 1.5380, 2.0082, 1.9151, 1.7493, 4.0343, 1.3412, 1.7590], + device='cuda:1'), covar=tensor([0.0987, 0.1778, 0.1246, 0.1020, 0.1593, 0.0218, 0.1560, 0.1742], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0081, 0.0077, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 05:10:10,462 INFO [finetune.py:976] (1/7) Epoch 5, batch 5000, loss[loss=0.2164, simple_loss=0.2802, pruned_loss=0.07626, over 4790.00 frames. ], tot_loss[loss=0.2222, simple_loss=0.281, pruned_loss=0.08168, over 953396.77 frames. ], batch size: 51, lr: 3.92e-03, grad_scale: 16.0 +2023-03-26 05:10:19,080 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.230e+02 1.618e+02 1.837e+02 2.301e+02 4.829e+02, threshold=3.674e+02, percent-clipped=1.0 +2023-03-26 05:10:33,686 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-03-26 05:10:43,561 INFO [finetune.py:976] (1/7) Epoch 5, batch 5050, loss[loss=0.2185, simple_loss=0.269, pruned_loss=0.08407, over 4760.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2781, pruned_loss=0.08095, over 954021.92 frames. ], batch size: 28, lr: 3.92e-03, grad_scale: 16.0 +2023-03-26 05:11:48,643 INFO [finetune.py:976] (1/7) Epoch 5, batch 5100, loss[loss=0.2351, simple_loss=0.2797, pruned_loss=0.09528, over 4856.00 frames. ], tot_loss[loss=0.2174, simple_loss=0.2747, pruned_loss=0.08002, over 953730.78 frames. ], batch size: 44, lr: 3.92e-03, grad_scale: 16.0 +2023-03-26 05:12:02,949 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.177e+02 1.639e+02 1.875e+02 2.408e+02 3.954e+02, threshold=3.749e+02, percent-clipped=2.0 +2023-03-26 05:12:28,071 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7106, 3.2207, 3.3432, 3.5263, 3.4335, 3.2568, 3.7928, 1.3399], + device='cuda:1'), covar=tensor([0.0918, 0.0906, 0.1008, 0.1201, 0.1465, 0.1681, 0.0886, 0.5065], + device='cuda:1'), in_proj_covar=tensor([0.0355, 0.0244, 0.0276, 0.0293, 0.0337, 0.0286, 0.0304, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:12:30,149 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 05:12:32,817 INFO [finetune.py:976] (1/7) Epoch 5, batch 5150, loss[loss=0.2128, simple_loss=0.2715, pruned_loss=0.07711, over 4883.00 frames. ], tot_loss[loss=0.2182, simple_loss=0.2753, pruned_loss=0.08054, over 956243.33 frames. ], batch size: 32, lr: 3.92e-03, grad_scale: 16.0 +2023-03-26 05:12:38,914 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28070.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 05:13:06,306 INFO [finetune.py:976] (1/7) Epoch 5, batch 5200, loss[loss=0.2374, simple_loss=0.3057, pruned_loss=0.08458, over 4931.00 frames. ], tot_loss[loss=0.2226, simple_loss=0.28, pruned_loss=0.08258, over 955136.47 frames. ], batch size: 33, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:13:12,472 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 05:13:14,537 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.308e+02 1.748e+02 1.996e+02 2.342e+02 5.311e+02, threshold=3.992e+02, percent-clipped=1.0 +2023-03-26 05:13:39,349 INFO [finetune.py:976] (1/7) Epoch 5, batch 5250, loss[loss=0.1975, simple_loss=0.2653, pruned_loss=0.06481, over 4861.00 frames. ], tot_loss[loss=0.2246, simple_loss=0.2827, pruned_loss=0.08326, over 954184.41 frames. ], batch size: 34, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:14:12,314 INFO [finetune.py:976] (1/7) Epoch 5, batch 5300, loss[loss=0.1892, simple_loss=0.2522, pruned_loss=0.06307, over 4859.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.2833, pruned_loss=0.08353, over 953905.98 frames. ], batch size: 31, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:14:19,226 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-26 05:14:19,562 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.181e+02 1.725e+02 1.957e+02 2.435e+02 6.444e+02, threshold=3.915e+02, percent-clipped=2.0 +2023-03-26 05:14:24,324 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28229.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:14:36,304 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3320, 1.4715, 1.6114, 0.9166, 1.4349, 1.7941, 1.8579, 1.4264], + device='cuda:1'), covar=tensor([0.1017, 0.0652, 0.0454, 0.0581, 0.0443, 0.0507, 0.0329, 0.0660], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0159, 0.0121, 0.0138, 0.0133, 0.0124, 0.0148, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.8065e-05, 1.1765e-04, 8.7583e-05, 1.0040e-04, 9.5965e-05, 9.1572e-05, + 1.0984e-04, 1.0835e-04], device='cuda:1') +2023-03-26 05:14:49,535 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.01 vs. limit=2.0 +2023-03-26 05:14:51,798 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28255.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 05:14:56,275 INFO [finetune.py:976] (1/7) Epoch 5, batch 5350, loss[loss=0.2171, simple_loss=0.2774, pruned_loss=0.07843, over 4868.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.2834, pruned_loss=0.08306, over 954083.75 frames. ], batch size: 34, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:15:13,541 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28287.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:15:15,758 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28290.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:15:28,416 INFO [finetune.py:976] (1/7) Epoch 5, batch 5400, loss[loss=0.2409, simple_loss=0.2952, pruned_loss=0.0933, over 4804.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.2794, pruned_loss=0.08113, over 952563.08 frames. ], batch size: 45, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:15:31,991 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28316.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 05:15:36,016 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.900e+01 1.568e+02 1.878e+02 2.260e+02 3.573e+02, threshold=3.756e+02, percent-clipped=0.0 +2023-03-26 05:15:36,887 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.75 vs. limit=5.0 +2023-03-26 05:15:53,750 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28348.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:16:01,446 INFO [finetune.py:976] (1/7) Epoch 5, batch 5450, loss[loss=0.2115, simple_loss=0.2558, pruned_loss=0.08362, over 3965.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2774, pruned_loss=0.08102, over 953778.81 frames. ], batch size: 17, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:16:18,305 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28370.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 05:17:02,776 INFO [finetune.py:976] (1/7) Epoch 5, batch 5500, loss[loss=0.2448, simple_loss=0.2946, pruned_loss=0.0975, over 4816.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.2741, pruned_loss=0.07928, over 955902.72 frames. ], batch size: 38, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:17:12,775 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=28418.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 05:17:21,150 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.059e+02 1.651e+02 2.036e+02 2.478e+02 5.642e+02, threshold=4.072e+02, percent-clipped=3.0 +2023-03-26 05:18:07,652 INFO [finetune.py:976] (1/7) Epoch 5, batch 5550, loss[loss=0.2509, simple_loss=0.3083, pruned_loss=0.09677, over 4901.00 frames. ], tot_loss[loss=0.2182, simple_loss=0.2762, pruned_loss=0.08009, over 956489.87 frames. ], batch size: 43, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:18:36,699 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28486.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:18:45,447 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28492.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:19:02,063 INFO [finetune.py:976] (1/7) Epoch 5, batch 5600, loss[loss=0.1913, simple_loss=0.2641, pruned_loss=0.05929, over 4855.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2804, pruned_loss=0.08133, over 954413.91 frames. ], batch size: 49, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:19:14,571 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.120e+02 1.753e+02 2.098e+02 2.591e+02 4.684e+02, threshold=4.196e+02, percent-clipped=2.0 +2023-03-26 05:19:20,396 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9583, 5.1146, 4.7182, 2.8658, 5.1596, 3.8449, 1.0286, 3.6741], + device='cuda:1'), covar=tensor([0.2601, 0.1630, 0.1709, 0.2973, 0.0871, 0.0795, 0.5256, 0.1265], + device='cuda:1'), in_proj_covar=tensor([0.0156, 0.0173, 0.0164, 0.0129, 0.0157, 0.0124, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 05:19:38,090 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.71 vs. limit=5.0 +2023-03-26 05:19:40,182 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28547.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:19:46,401 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5930, 1.4965, 1.4822, 1.5609, 1.1203, 3.4180, 1.3803, 1.8889], + device='cuda:1'), covar=tensor([0.3194, 0.2295, 0.1969, 0.2182, 0.1883, 0.0179, 0.2725, 0.1271], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0114, 0.0117, 0.0121, 0.0117, 0.0098, 0.0101, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 05:19:47,594 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28553.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:19:52,217 INFO [finetune.py:976] (1/7) Epoch 5, batch 5650, loss[loss=0.218, simple_loss=0.2822, pruned_loss=0.07695, over 4813.00 frames. ], tot_loss[loss=0.2216, simple_loss=0.2813, pruned_loss=0.081, over 954330.75 frames. ], batch size: 45, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:20:06,330 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28585.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:20:21,892 INFO [finetune.py:976] (1/7) Epoch 5, batch 5700, loss[loss=0.1935, simple_loss=0.2375, pruned_loss=0.07469, over 4206.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2783, pruned_loss=0.08051, over 938429.18 frames. ], batch size: 18, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:20:21,951 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28611.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 05:20:29,013 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.136e+02 1.580e+02 1.894e+02 2.210e+02 5.665e+02, threshold=3.789e+02, percent-clipped=1.0 +2023-03-26 05:20:30,879 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0082, 1.8773, 1.9545, 1.9294, 1.6969, 3.9397, 1.7974, 2.4391], + device='cuda:1'), covar=tensor([0.3235, 0.2265, 0.1885, 0.2158, 0.1591, 0.0185, 0.2326, 0.1097], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0113, 0.0117, 0.0121, 0.0117, 0.0097, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 05:20:35,588 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6427, 1.9343, 1.2827, 2.5819, 3.0413, 2.4531, 2.2987, 2.6920], + device='cuda:1'), covar=tensor([0.1363, 0.1894, 0.1931, 0.1038, 0.1533, 0.1502, 0.1319, 0.1796], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0098, 0.0116, 0.0093, 0.0124, 0.0097, 0.0100, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 05:20:36,892 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.55 vs. limit=2.0 +2023-03-26 05:20:53,188 INFO [finetune.py:976] (1/7) Epoch 6, batch 0, loss[loss=0.2329, simple_loss=0.2862, pruned_loss=0.08982, over 4879.00 frames. ], tot_loss[loss=0.2329, simple_loss=0.2862, pruned_loss=0.08982, over 4879.00 frames. ], batch size: 43, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:20:53,188 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 05:20:56,570 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4583, 1.5653, 1.5105, 1.6335, 1.6926, 2.9283, 1.4424, 1.6754], + device='cuda:1'), covar=tensor([0.0998, 0.1670, 0.1045, 0.0988, 0.1461, 0.0332, 0.1398, 0.1560], + device='cuda:1'), in_proj_covar=tensor([0.0079, 0.0083, 0.0078, 0.0081, 0.0094, 0.0084, 0.0087, 0.0081], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 05:21:12,694 INFO [finetune.py:1010] (1/7) Epoch 6, validation: loss=0.1659, simple_loss=0.2379, pruned_loss=0.04693, over 2265189.00 frames. +2023-03-26 05:21:12,694 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 05:21:15,237 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28643.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:21:15,291 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3852, 2.4298, 2.2477, 1.6224, 2.5781, 2.5614, 2.3914, 2.0358], + device='cuda:1'), covar=tensor([0.0744, 0.0659, 0.0810, 0.1029, 0.0588, 0.0759, 0.0736, 0.1085], + device='cuda:1'), in_proj_covar=tensor([0.0141, 0.0136, 0.0146, 0.0131, 0.0114, 0.0145, 0.0149, 0.0166], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:21:18,013 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9841, 2.0490, 1.9744, 1.2391, 2.2250, 2.1409, 1.9861, 1.7381], + device='cuda:1'), covar=tensor([0.0692, 0.0657, 0.0703, 0.1033, 0.0483, 0.0711, 0.0686, 0.1147], + device='cuda:1'), in_proj_covar=tensor([0.0141, 0.0136, 0.0146, 0.0131, 0.0114, 0.0145, 0.0148, 0.0166], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:21:59,715 INFO [finetune.py:976] (1/7) Epoch 6, batch 50, loss[loss=0.2032, simple_loss=0.2774, pruned_loss=0.06451, over 4886.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2781, pruned_loss=0.0794, over 215628.73 frames. ], batch size: 43, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:22:08,382 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 05:22:17,198 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6161, 1.4667, 1.8475, 2.6935, 1.8976, 2.0331, 1.1363, 2.0876], + device='cuda:1'), covar=tensor([0.1667, 0.1516, 0.1303, 0.0784, 0.0828, 0.2284, 0.1714, 0.0790], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0118, 0.0136, 0.0165, 0.0102, 0.0141, 0.0127, 0.0103], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 05:22:22,663 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9447, 1.7551, 1.7007, 1.8495, 1.3851, 4.4185, 1.6732, 2.3805], + device='cuda:1'), covar=tensor([0.3143, 0.2343, 0.2067, 0.2212, 0.1747, 0.0092, 0.2459, 0.1230], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0113, 0.0117, 0.0121, 0.0117, 0.0097, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 05:22:30,501 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.152e+02 1.667e+02 1.978e+02 2.446e+02 6.098e+02, threshold=3.955e+02, percent-clipped=3.0 +2023-03-26 05:22:41,772 INFO [finetune.py:976] (1/7) Epoch 6, batch 100, loss[loss=0.1736, simple_loss=0.2343, pruned_loss=0.05651, over 4829.00 frames. ], tot_loss[loss=0.2132, simple_loss=0.2712, pruned_loss=0.07766, over 379093.04 frames. ], batch size: 33, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:22:52,787 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4420, 1.5627, 1.8887, 1.8166, 1.6303, 3.3066, 1.2699, 1.6757], + device='cuda:1'), covar=tensor([0.0931, 0.1589, 0.1109, 0.0958, 0.1421, 0.0249, 0.1448, 0.1703], + device='cuda:1'), in_proj_covar=tensor([0.0078, 0.0082, 0.0078, 0.0080, 0.0093, 0.0084, 0.0087, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 05:23:07,007 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-26 05:23:15,380 INFO [finetune.py:976] (1/7) Epoch 6, batch 150, loss[loss=0.2248, simple_loss=0.2683, pruned_loss=0.09065, over 4853.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2687, pruned_loss=0.07771, over 507519.25 frames. ], batch size: 44, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:23:15,561 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.66 vs. limit=5.0 +2023-03-26 05:23:37,612 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.084e+02 1.593e+02 1.877e+02 2.260e+02 4.734e+02, threshold=3.755e+02, percent-clipped=1.0 +2023-03-26 05:23:42,572 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 05:23:48,136 INFO [finetune.py:976] (1/7) Epoch 6, batch 200, loss[loss=0.1886, simple_loss=0.2517, pruned_loss=0.06271, over 4769.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2678, pruned_loss=0.07702, over 608534.10 frames. ], batch size: 26, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:23:50,545 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28842.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:23:55,122 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28848.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:24:03,972 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28861.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:24:19,102 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28885.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:24:26,389 INFO [finetune.py:976] (1/7) Epoch 6, batch 250, loss[loss=0.2144, simple_loss=0.2912, pruned_loss=0.06879, over 4811.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2751, pruned_loss=0.08007, over 683482.71 frames. ], batch size: 51, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:24:37,371 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8824, 1.6559, 1.4520, 1.4981, 1.5862, 1.5720, 1.6020, 2.3877], + device='cuda:1'), covar=tensor([0.6029, 0.6017, 0.4677, 0.6203, 0.5551, 0.3423, 0.6163, 0.2225], + device='cuda:1'), in_proj_covar=tensor([0.0282, 0.0258, 0.0220, 0.0284, 0.0238, 0.0202, 0.0245, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:24:37,500 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-26 05:24:47,333 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28911.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 05:25:02,677 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28922.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:25:03,152 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.192e+02 1.635e+02 1.950e+02 2.422e+02 4.878e+02, threshold=3.900e+02, percent-clipped=5.0 +2023-03-26 05:25:09,282 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=28933.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:25:13,899 INFO [finetune.py:976] (1/7) Epoch 6, batch 300, loss[loss=0.235, simple_loss=0.3017, pruned_loss=0.08412, over 4725.00 frames. ], tot_loss[loss=0.2201, simple_loss=0.2786, pruned_loss=0.08085, over 743463.33 frames. ], batch size: 54, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:25:16,455 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28943.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:25:27,537 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9653, 1.8528, 1.5227, 1.9464, 1.9813, 1.6188, 2.3852, 2.0270], + device='cuda:1'), covar=tensor([0.1835, 0.3596, 0.4065, 0.3956, 0.3219, 0.2099, 0.3735, 0.2339], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0193, 0.0236, 0.0255, 0.0231, 0.0190, 0.0211, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:25:28,661 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=28959.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 05:25:47,463 INFO [finetune.py:976] (1/7) Epoch 6, batch 350, loss[loss=0.2752, simple_loss=0.3196, pruned_loss=0.1154, over 4816.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.2801, pruned_loss=0.08078, over 791864.66 frames. ], batch size: 39, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:25:49,425 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=28991.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:26:02,726 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29002.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:26:28,280 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.301e+02 1.836e+02 2.201e+02 2.620e+02 4.241e+02, threshold=4.402e+02, percent-clipped=2.0 +2023-03-26 05:26:38,005 INFO [finetune.py:976] (1/7) Epoch 6, batch 400, loss[loss=0.2031, simple_loss=0.2748, pruned_loss=0.06565, over 4831.00 frames. ], tot_loss[loss=0.2199, simple_loss=0.28, pruned_loss=0.07986, over 827558.56 frames. ], batch size: 30, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:26:46,090 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.35 vs. limit=5.0 +2023-03-26 05:27:01,631 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29063.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:27:17,312 INFO [finetune.py:976] (1/7) Epoch 6, batch 450, loss[loss=0.1967, simple_loss=0.2614, pruned_loss=0.06602, over 4867.00 frames. ], tot_loss[loss=0.2189, simple_loss=0.2787, pruned_loss=0.07952, over 856369.12 frames. ], batch size: 34, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:27:45,351 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.158e+02 1.708e+02 1.996e+02 2.284e+02 5.200e+02, threshold=3.993e+02, percent-clipped=1.0 +2023-03-26 05:27:55,104 INFO [finetune.py:976] (1/7) Epoch 6, batch 500, loss[loss=0.214, simple_loss=0.2737, pruned_loss=0.07718, over 4881.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.2759, pruned_loss=0.07834, over 880019.30 frames. ], batch size: 31, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:27:57,042 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29142.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:27:59,367 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0130, 1.3573, 0.7616, 1.8972, 2.4230, 1.8549, 1.6368, 1.9360], + device='cuda:1'), covar=tensor([0.1393, 0.1955, 0.2210, 0.1100, 0.1765, 0.1959, 0.1296, 0.1920], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0116, 0.0093, 0.0124, 0.0097, 0.0101, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 05:28:01,679 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29148.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:28:02,857 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7780, 3.9310, 3.6537, 2.1340, 3.9847, 2.9109, 1.0782, 2.7343], + device='cuda:1'), covar=tensor([0.2522, 0.1817, 0.1705, 0.3213, 0.1081, 0.0968, 0.4494, 0.1607], + device='cuda:1'), in_proj_covar=tensor([0.0156, 0.0174, 0.0165, 0.0129, 0.0158, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 05:28:28,351 INFO [finetune.py:976] (1/7) Epoch 6, batch 550, loss[loss=0.2172, simple_loss=0.2622, pruned_loss=0.08617, over 4828.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2734, pruned_loss=0.07778, over 895960.39 frames. ], batch size: 30, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:28:28,996 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=29190.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:28:34,377 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=29196.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:29:00,095 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29217.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:29:04,678 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.109e+02 1.638e+02 2.076e+02 2.525e+02 4.090e+02, threshold=4.153e+02, percent-clipped=1.0 +2023-03-26 05:29:18,506 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29236.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:29:20,235 INFO [finetune.py:976] (1/7) Epoch 6, batch 600, loss[loss=0.1852, simple_loss=0.2456, pruned_loss=0.06243, over 4774.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.275, pruned_loss=0.0792, over 908789.95 frames. ], batch size: 26, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:30:20,519 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6117, 2.4130, 2.0122, 2.5726, 2.5903, 2.1097, 3.0177, 2.4961], + device='cuda:1'), covar=tensor([0.1529, 0.3118, 0.4055, 0.3676, 0.2942, 0.1891, 0.3560, 0.2159], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0193, 0.0236, 0.0254, 0.0231, 0.0190, 0.0211, 0.0190], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:30:24,096 INFO [finetune.py:976] (1/7) Epoch 6, batch 650, loss[loss=0.1776, simple_loss=0.2473, pruned_loss=0.05397, over 4811.00 frames. ], tot_loss[loss=0.2181, simple_loss=0.277, pruned_loss=0.07957, over 919631.54 frames. ], batch size: 45, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:30:34,096 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29297.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:31:13,223 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.225e+02 1.735e+02 1.982e+02 2.438e+02 4.583e+02, threshold=3.965e+02, percent-clipped=2.0 +2023-03-26 05:31:33,679 INFO [finetune.py:976] (1/7) Epoch 6, batch 700, loss[loss=0.2446, simple_loss=0.3082, pruned_loss=0.09054, over 4800.00 frames. ], tot_loss[loss=0.2189, simple_loss=0.2786, pruned_loss=0.07958, over 927830.58 frames. ], batch size: 51, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:31:46,289 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29358.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:31:49,405 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3427, 2.1536, 1.7263, 2.2631, 2.1472, 1.8662, 2.7225, 2.3614], + device='cuda:1'), covar=tensor([0.1507, 0.3523, 0.3824, 0.3688, 0.3044, 0.1954, 0.3791, 0.2113], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0194, 0.0237, 0.0255, 0.0232, 0.0191, 0.0212, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:32:13,550 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9865, 1.9062, 1.5483, 2.0837, 2.0178, 1.6301, 2.3654, 1.9448], + device='cuda:1'), covar=tensor([0.1529, 0.2900, 0.3408, 0.2730, 0.2448, 0.1798, 0.3180, 0.2022], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0194, 0.0238, 0.0256, 0.0232, 0.0191, 0.0212, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:32:17,027 INFO [finetune.py:976] (1/7) Epoch 6, batch 750, loss[loss=0.2186, simple_loss=0.2813, pruned_loss=0.07797, over 4827.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2796, pruned_loss=0.07976, over 934038.38 frames. ], batch size: 30, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:32:59,102 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.656e+01 1.814e+02 2.109e+02 2.501e+02 5.044e+02, threshold=4.217e+02, percent-clipped=3.0 +2023-03-26 05:33:08,702 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.75 vs. limit=5.0 +2023-03-26 05:33:26,483 INFO [finetune.py:976] (1/7) Epoch 6, batch 800, loss[loss=0.2176, simple_loss=0.2661, pruned_loss=0.08458, over 4803.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2793, pruned_loss=0.07942, over 939368.16 frames. ], batch size: 25, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:33:57,201 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-03-26 05:34:02,335 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 05:34:10,554 INFO [finetune.py:976] (1/7) Epoch 6, batch 850, loss[loss=0.222, simple_loss=0.2804, pruned_loss=0.08183, over 4885.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2772, pruned_loss=0.07897, over 941497.83 frames. ], batch size: 35, lr: 3.92e-03, grad_scale: 32.0 +2023-03-26 05:34:20,929 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4827, 1.3628, 1.4390, 1.3415, 0.8150, 2.2596, 0.7157, 1.2829], + device='cuda:1'), covar=tensor([0.3386, 0.2571, 0.1996, 0.2449, 0.2141, 0.0353, 0.2863, 0.1367], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0117, 0.0121, 0.0117, 0.0098, 0.0101, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 05:34:43,538 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29517.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:34:52,273 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.241e+02 1.650e+02 2.021e+02 2.394e+02 5.702e+02, threshold=4.042e+02, percent-clipped=1.0 +2023-03-26 05:35:14,735 INFO [finetune.py:976] (1/7) Epoch 6, batch 900, loss[loss=0.2132, simple_loss=0.2764, pruned_loss=0.07503, over 4860.00 frames. ], tot_loss[loss=0.215, simple_loss=0.2739, pruned_loss=0.0781, over 944853.51 frames. ], batch size: 31, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:35:16,221 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-03-26 05:35:21,839 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-03-26 05:35:32,159 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29551.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:35:46,187 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=29565.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:36:14,402 INFO [finetune.py:976] (1/7) Epoch 6, batch 950, loss[loss=0.2179, simple_loss=0.2736, pruned_loss=0.08109, over 4913.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2718, pruned_loss=0.07709, over 947020.51 frames. ], batch size: 43, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:36:21,368 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29592.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:36:23,288 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5347, 2.2528, 2.9335, 1.8228, 2.7732, 2.8451, 2.0783, 2.9005], + device='cuda:1'), covar=tensor([0.1854, 0.2083, 0.1368, 0.2557, 0.1015, 0.1725, 0.2615, 0.1028], + device='cuda:1'), in_proj_covar=tensor([0.0204, 0.0203, 0.0198, 0.0194, 0.0184, 0.0220, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:36:43,480 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29612.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:36:53,896 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7871, 1.6025, 1.4049, 1.4016, 1.5049, 1.5028, 1.5097, 2.2432], + device='cuda:1'), covar=tensor([0.6760, 0.6467, 0.5238, 0.6383, 0.5741, 0.3801, 0.6339, 0.2499], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0259, 0.0220, 0.0285, 0.0240, 0.0203, 0.0246, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:36:56,223 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.131e+02 1.661e+02 1.977e+02 2.359e+02 4.237e+02, threshold=3.954e+02, percent-clipped=2.0 +2023-03-26 05:37:17,831 INFO [finetune.py:976] (1/7) Epoch 6, batch 1000, loss[loss=0.2212, simple_loss=0.2863, pruned_loss=0.07802, over 4794.00 frames. ], tot_loss[loss=0.2182, simple_loss=0.277, pruned_loss=0.07968, over 948848.70 frames. ], batch size: 29, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:37:40,148 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29658.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:37:47,522 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29662.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:38:20,623 INFO [finetune.py:976] (1/7) Epoch 6, batch 1050, loss[loss=0.2449, simple_loss=0.305, pruned_loss=0.09238, over 4811.00 frames. ], tot_loss[loss=0.2201, simple_loss=0.2799, pruned_loss=0.08016, over 948703.12 frames. ], batch size: 38, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:38:41,607 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=29706.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:38:52,628 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-03-26 05:38:59,761 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0 +2023-03-26 05:39:02,905 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.157e+02 1.795e+02 2.095e+02 2.533e+02 7.754e+02, threshold=4.191e+02, percent-clipped=4.0 +2023-03-26 05:39:03,058 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29723.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 05:39:10,871 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3570, 1.2594, 1.5163, 2.4224, 1.7452, 1.9926, 0.8109, 1.9801], + device='cuda:1'), covar=tensor([0.1703, 0.1465, 0.1239, 0.0709, 0.0888, 0.1397, 0.1725, 0.0699], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0117, 0.0135, 0.0164, 0.0101, 0.0141, 0.0127, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 05:39:19,630 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 05:39:23,373 INFO [finetune.py:976] (1/7) Epoch 6, batch 1100, loss[loss=0.1679, simple_loss=0.2408, pruned_loss=0.04747, over 4771.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2813, pruned_loss=0.08082, over 951369.91 frames. ], batch size: 28, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:40:24,911 INFO [finetune.py:976] (1/7) Epoch 6, batch 1150, loss[loss=0.1692, simple_loss=0.2291, pruned_loss=0.05468, over 4782.00 frames. ], tot_loss[loss=0.2214, simple_loss=0.2812, pruned_loss=0.08083, over 951520.36 frames. ], batch size: 26, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:40:35,352 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3835, 3.8004, 3.9945, 4.2134, 4.0897, 3.8027, 4.4455, 1.2847], + device='cuda:1'), covar=tensor([0.0716, 0.0784, 0.0823, 0.0909, 0.1137, 0.1538, 0.0640, 0.5234], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0241, 0.0273, 0.0290, 0.0331, 0.0282, 0.0302, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:40:37,340 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3753, 3.7958, 3.9859, 4.2078, 4.0960, 3.8211, 4.4391, 1.3335], + device='cuda:1'), covar=tensor([0.0670, 0.0727, 0.0769, 0.0777, 0.1084, 0.1430, 0.0579, 0.5007], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0241, 0.0273, 0.0290, 0.0331, 0.0282, 0.0302, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:40:37,974 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29802.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:41:01,069 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.060e+02 1.742e+02 2.057e+02 2.377e+02 6.600e+02, threshold=4.115e+02, percent-clipped=1.0 +2023-03-26 05:41:11,693 INFO [finetune.py:976] (1/7) Epoch 6, batch 1200, loss[loss=0.2453, simple_loss=0.295, pruned_loss=0.09777, over 4798.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2793, pruned_loss=0.08003, over 952351.51 frames. ], batch size: 45, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:41:28,776 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29863.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:41:29,997 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 05:41:41,149 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.9352, 3.3995, 3.6124, 3.7164, 3.7102, 3.5244, 3.9652, 1.5389], + device='cuda:1'), covar=tensor([0.0764, 0.0828, 0.0756, 0.0939, 0.1088, 0.1131, 0.0705, 0.4331], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0240, 0.0273, 0.0290, 0.0331, 0.0282, 0.0302, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:41:45,292 INFO [finetune.py:976] (1/7) Epoch 6, batch 1250, loss[loss=0.2028, simple_loss=0.2671, pruned_loss=0.06921, over 4914.00 frames. ], tot_loss[loss=0.2175, simple_loss=0.2769, pruned_loss=0.07908, over 953110.55 frames. ], batch size: 36, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:41:47,165 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29892.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:41:57,704 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29907.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:42:07,774 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.170e+02 1.638e+02 1.953e+02 2.201e+02 4.150e+02, threshold=3.906e+02, percent-clipped=1.0 +2023-03-26 05:42:18,518 INFO [finetune.py:976] (1/7) Epoch 6, batch 1300, loss[loss=0.1927, simple_loss=0.2385, pruned_loss=0.07347, over 4866.00 frames. ], tot_loss[loss=0.2136, simple_loss=0.2727, pruned_loss=0.07725, over 953109.96 frames. ], batch size: 31, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:42:19,141 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=29940.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:42:53,738 INFO [finetune.py:976] (1/7) Epoch 6, batch 1350, loss[loss=0.2032, simple_loss=0.2571, pruned_loss=0.07466, over 4730.00 frames. ], tot_loss[loss=0.2154, simple_loss=0.2743, pruned_loss=0.07829, over 952269.95 frames. ], batch size: 54, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:43:22,452 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30018.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 05:43:25,359 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.107e+02 1.660e+02 2.040e+02 2.486e+02 4.804e+02, threshold=4.081e+02, percent-clipped=2.0 +2023-03-26 05:43:35,490 INFO [finetune.py:976] (1/7) Epoch 6, batch 1400, loss[loss=0.2223, simple_loss=0.2886, pruned_loss=0.078, over 4896.00 frames. ], tot_loss[loss=0.218, simple_loss=0.277, pruned_loss=0.07946, over 952024.91 frames. ], batch size: 35, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:43:39,914 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3013, 1.3203, 1.5507, 1.1017, 1.2950, 1.4701, 1.2866, 1.6267], + device='cuda:1'), covar=tensor([0.1263, 0.2001, 0.1225, 0.1571, 0.1000, 0.1181, 0.2890, 0.0823], + device='cuda:1'), in_proj_covar=tensor([0.0206, 0.0205, 0.0199, 0.0195, 0.0184, 0.0220, 0.0216, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:44:14,308 INFO [finetune.py:976] (1/7) Epoch 6, batch 1450, loss[loss=0.2324, simple_loss=0.293, pruned_loss=0.08589, over 4814.00 frames. ], tot_loss[loss=0.2178, simple_loss=0.2774, pruned_loss=0.07913, over 953792.05 frames. ], batch size: 51, lr: 3.91e-03, grad_scale: 64.0 +2023-03-26 05:44:34,785 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1225, 1.9031, 1.6534, 1.8687, 2.1392, 1.7885, 2.3560, 2.0686], + device='cuda:1'), covar=tensor([0.1656, 0.3119, 0.4124, 0.3428, 0.3095, 0.1966, 0.3841, 0.2352], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0193, 0.0237, 0.0254, 0.0231, 0.0191, 0.0211, 0.0192], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:44:58,650 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.184e+02 1.825e+02 2.216e+02 2.642e+02 7.386e+02, threshold=4.431e+02, percent-clipped=2.0 +2023-03-26 05:44:59,433 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30125.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:45:16,732 INFO [finetune.py:976] (1/7) Epoch 6, batch 1500, loss[loss=0.243, simple_loss=0.3166, pruned_loss=0.08469, over 4796.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2798, pruned_loss=0.08034, over 953558.47 frames. ], batch size: 51, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:45:36,553 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-26 05:45:38,718 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30158.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:45:59,674 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30186.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:46:01,366 INFO [finetune.py:976] (1/7) Epoch 6, batch 1550, loss[loss=0.1766, simple_loss=0.2486, pruned_loss=0.05228, over 4805.00 frames. ], tot_loss[loss=0.2194, simple_loss=0.2792, pruned_loss=0.07979, over 952863.00 frames. ], batch size: 39, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:46:12,978 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30207.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:46:14,235 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30209.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:46:25,109 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.142e+02 1.698e+02 2.038e+02 2.458e+02 3.858e+02, threshold=4.076e+02, percent-clipped=0.0 +2023-03-26 05:46:34,720 INFO [finetune.py:976] (1/7) Epoch 6, batch 1600, loss[loss=0.2342, simple_loss=0.2959, pruned_loss=0.08622, over 4738.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2767, pruned_loss=0.07898, over 954181.68 frames. ], batch size: 54, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:46:44,947 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=30255.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:46:47,393 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6456, 1.5909, 1.4951, 0.9795, 1.5716, 1.8348, 1.7690, 1.4642], + device='cuda:1'), covar=tensor([0.0858, 0.0606, 0.0479, 0.0561, 0.0400, 0.0368, 0.0292, 0.0548], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0159, 0.0122, 0.0138, 0.0133, 0.0124, 0.0148, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.8907e-05, 1.1787e-04, 8.8477e-05, 1.0100e-04, 9.5649e-05, 9.1662e-05, + 1.0952e-04, 1.0859e-04], device='cuda:1') +2023-03-26 05:46:56,001 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30270.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:47:08,006 INFO [finetune.py:976] (1/7) Epoch 6, batch 1650, loss[loss=0.2152, simple_loss=0.2755, pruned_loss=0.07746, over 4920.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2733, pruned_loss=0.07713, over 956652.17 frames. ], batch size: 36, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:47:27,147 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30318.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 05:47:31,616 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.138e+02 1.710e+02 1.999e+02 2.408e+02 3.997e+02, threshold=3.998e+02, percent-clipped=0.0 +2023-03-26 05:47:41,358 INFO [finetune.py:976] (1/7) Epoch 6, batch 1700, loss[loss=0.2662, simple_loss=0.3022, pruned_loss=0.1151, over 4864.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2724, pruned_loss=0.07729, over 956535.23 frames. ], batch size: 31, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:47:52,758 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3099, 1.5643, 1.6302, 0.9224, 1.3980, 1.8243, 1.8512, 1.4956], + device='cuda:1'), covar=tensor([0.0909, 0.0480, 0.0411, 0.0485, 0.0429, 0.0433, 0.0279, 0.0614], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0159, 0.0122, 0.0138, 0.0133, 0.0124, 0.0148, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.8844e-05, 1.1765e-04, 8.8353e-05, 1.0111e-04, 9.5827e-05, 9.1637e-05, + 1.0947e-04, 1.0842e-04], device='cuda:1') +2023-03-26 05:47:55,194 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30360.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:47:58,773 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=30366.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:48:27,033 INFO [finetune.py:976] (1/7) Epoch 6, batch 1750, loss[loss=0.1939, simple_loss=0.2586, pruned_loss=0.06459, over 4905.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2744, pruned_loss=0.07797, over 955771.04 frames. ], batch size: 32, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:48:31,402 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6664, 1.5828, 1.3730, 1.4647, 1.8881, 1.8535, 1.6004, 1.3705], + device='cuda:1'), covar=tensor([0.0286, 0.0322, 0.0556, 0.0337, 0.0196, 0.0426, 0.0293, 0.0401], + device='cuda:1'), in_proj_covar=tensor([0.0087, 0.0111, 0.0137, 0.0117, 0.0103, 0.0100, 0.0091, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.8149e-05, 8.7466e-05, 1.1001e-04, 9.2598e-05, 8.1438e-05, 7.4003e-05, + 6.9276e-05, 8.4854e-05], device='cuda:1') +2023-03-26 05:48:48,753 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30421.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:48:50,962 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.220e+01 1.748e+02 2.265e+02 2.778e+02 4.150e+02, threshold=4.530e+02, percent-clipped=1.0 +2023-03-26 05:49:00,662 INFO [finetune.py:976] (1/7) Epoch 6, batch 1800, loss[loss=0.1887, simple_loss=0.258, pruned_loss=0.05968, over 4792.00 frames. ], tot_loss[loss=0.2178, simple_loss=0.2779, pruned_loss=0.07882, over 955213.85 frames. ], batch size: 29, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:49:08,198 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-03-26 05:49:13,231 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30458.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:49:28,993 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30481.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:49:31,506 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6539, 1.4884, 1.3537, 1.3378, 1.7155, 1.4579, 1.5488, 1.6016], + device='cuda:1'), covar=tensor([0.1539, 0.2683, 0.3779, 0.2686, 0.2715, 0.1826, 0.2641, 0.2108], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0192, 0.0235, 0.0253, 0.0230, 0.0190, 0.0211, 0.0190], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:49:33,833 INFO [finetune.py:976] (1/7) Epoch 6, batch 1850, loss[loss=0.2117, simple_loss=0.2864, pruned_loss=0.06855, over 4815.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2786, pruned_loss=0.07903, over 954112.54 frames. ], batch size: 39, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:49:44,725 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=30506.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:50:03,760 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.078e+02 1.671e+02 2.010e+02 2.577e+02 4.739e+02, threshold=4.020e+02, percent-clipped=1.0 +2023-03-26 05:50:22,792 INFO [finetune.py:976] (1/7) Epoch 6, batch 1900, loss[loss=0.2368, simple_loss=0.3021, pruned_loss=0.08578, over 4814.00 frames. ], tot_loss[loss=0.2193, simple_loss=0.2797, pruned_loss=0.07948, over 954067.89 frames. ], batch size: 51, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:50:39,121 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.12 vs. limit=5.0 +2023-03-26 05:50:43,955 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4812, 1.1867, 1.2622, 1.2791, 1.6202, 1.5411, 1.4327, 1.2020], + device='cuda:1'), covar=tensor([0.0262, 0.0338, 0.0537, 0.0313, 0.0224, 0.0460, 0.0276, 0.0436], + device='cuda:1'), in_proj_covar=tensor([0.0086, 0.0111, 0.0136, 0.0116, 0.0103, 0.0099, 0.0090, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.7393e-05, 8.7049e-05, 1.0921e-04, 9.1833e-05, 8.1163e-05, 7.3527e-05, + 6.8700e-05, 8.4255e-05], device='cuda:1') +2023-03-26 05:50:52,224 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30565.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:51:02,765 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6478, 2.5801, 2.3165, 2.6919, 3.2170, 2.6488, 2.3034, 2.1715], + device='cuda:1'), covar=tensor([0.2060, 0.1825, 0.1711, 0.1537, 0.1690, 0.0948, 0.2203, 0.1740], + device='cuda:1'), in_proj_covar=tensor([0.0235, 0.0210, 0.0203, 0.0187, 0.0238, 0.0177, 0.0215, 0.0190], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:51:07,460 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-26 05:51:20,333 INFO [finetune.py:976] (1/7) Epoch 6, batch 1950, loss[loss=0.183, simple_loss=0.238, pruned_loss=0.06395, over 4888.00 frames. ], tot_loss[loss=0.2177, simple_loss=0.2779, pruned_loss=0.07877, over 954986.58 frames. ], batch size: 32, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:51:50,154 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-03-26 05:51:58,194 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.083e+02 1.556e+02 1.883e+02 2.215e+02 4.222e+02, threshold=3.767e+02, percent-clipped=2.0 +2023-03-26 05:52:09,310 INFO [finetune.py:976] (1/7) Epoch 6, batch 2000, loss[loss=0.243, simple_loss=0.3032, pruned_loss=0.09136, over 4924.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.2763, pruned_loss=0.07856, over 955372.09 frames. ], batch size: 37, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:52:19,640 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30655.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:52:56,172 INFO [finetune.py:976] (1/7) Epoch 6, batch 2050, loss[loss=0.1824, simple_loss=0.2468, pruned_loss=0.05894, over 4902.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2727, pruned_loss=0.07695, over 957137.00 frames. ], batch size: 32, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:53:14,332 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30716.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:53:14,393 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30716.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 05:53:20,053 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.195e+02 1.687e+02 1.889e+02 2.318e+02 4.112e+02, threshold=3.779e+02, percent-clipped=3.0 +2023-03-26 05:53:40,714 INFO [finetune.py:976] (1/7) Epoch 6, batch 2100, loss[loss=0.2822, simple_loss=0.3275, pruned_loss=0.1185, over 4842.00 frames. ], tot_loss[loss=0.2142, simple_loss=0.2737, pruned_loss=0.07739, over 959358.42 frames. ], batch size: 49, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:54:13,621 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30781.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:54:18,840 INFO [finetune.py:976] (1/7) Epoch 6, batch 2150, loss[loss=0.24, simple_loss=0.2758, pruned_loss=0.1021, over 3958.00 frames. ], tot_loss[loss=0.2178, simple_loss=0.2776, pruned_loss=0.07896, over 957420.36 frames. ], batch size: 17, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:54:27,775 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4024, 1.2878, 1.4648, 1.6366, 1.4125, 2.9922, 1.1538, 1.4051], + device='cuda:1'), covar=tensor([0.1295, 0.2500, 0.1312, 0.1285, 0.2144, 0.0329, 0.2184, 0.2662], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0082, 0.0077, 0.0080, 0.0092, 0.0083, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 05:54:42,039 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.093e+02 1.708e+02 2.011e+02 2.625e+02 4.679e+02, threshold=4.022e+02, percent-clipped=7.0 +2023-03-26 05:54:46,134 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=30829.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:54:52,228 INFO [finetune.py:976] (1/7) Epoch 6, batch 2200, loss[loss=0.2274, simple_loss=0.2812, pruned_loss=0.08679, over 4885.00 frames. ], tot_loss[loss=0.2193, simple_loss=0.2791, pruned_loss=0.0797, over 955044.56 frames. ], batch size: 34, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:54:55,885 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9927, 4.3162, 4.5304, 4.8486, 4.6942, 4.4904, 5.1102, 1.5525], + device='cuda:1'), covar=tensor([0.0703, 0.0780, 0.0629, 0.0820, 0.1161, 0.1368, 0.0492, 0.5369], + device='cuda:1'), in_proj_covar=tensor([0.0354, 0.0242, 0.0276, 0.0292, 0.0330, 0.0282, 0.0303, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:54:57,735 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.77 vs. limit=5.0 +2023-03-26 05:55:16,375 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30865.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:55:37,465 INFO [finetune.py:976] (1/7) Epoch 6, batch 2250, loss[loss=0.2187, simple_loss=0.2881, pruned_loss=0.07464, over 4834.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2782, pruned_loss=0.07849, over 955163.69 frames. ], batch size: 49, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:55:49,186 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6140, 0.6350, 1.4710, 1.3170, 1.2729, 1.2594, 1.2092, 1.3458], + device='cuda:1'), covar=tensor([0.4720, 0.6517, 0.5398, 0.5759, 0.6438, 0.5102, 0.6975, 0.5145], + device='cuda:1'), in_proj_covar=tensor([0.0229, 0.0246, 0.0254, 0.0256, 0.0242, 0.0219, 0.0273, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:55:53,168 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30903.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:56:10,384 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=30913.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:56:14,010 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6145, 3.7341, 3.4257, 1.6586, 3.8125, 2.8953, 0.8227, 2.5231], + device='cuda:1'), covar=tensor([0.2240, 0.1413, 0.1584, 0.3019, 0.1007, 0.0947, 0.4047, 0.1354], + device='cuda:1'), in_proj_covar=tensor([0.0155, 0.0170, 0.0162, 0.0127, 0.0155, 0.0122, 0.0144, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 05:56:22,247 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.065e+02 1.611e+02 1.958e+02 2.302e+02 5.232e+02, threshold=3.915e+02, percent-clipped=2.0 +2023-03-26 05:56:22,996 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30925.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:56:34,297 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30934.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:56:42,903 INFO [finetune.py:976] (1/7) Epoch 6, batch 2300, loss[loss=0.1945, simple_loss=0.2558, pruned_loss=0.06657, over 4862.00 frames. ], tot_loss[loss=0.2177, simple_loss=0.2783, pruned_loss=0.07859, over 952635.04 frames. ], batch size: 31, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:56:42,983 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2197, 2.9500, 2.7825, 1.3168, 3.0343, 2.2702, 0.7871, 1.9042], + device='cuda:1'), covar=tensor([0.2507, 0.2287, 0.2010, 0.3716, 0.1392, 0.1157, 0.4256, 0.1806], + device='cuda:1'), in_proj_covar=tensor([0.0155, 0.0170, 0.0163, 0.0128, 0.0156, 0.0123, 0.0145, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 05:57:16,178 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30964.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:57:46,775 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30986.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:57:49,007 INFO [finetune.py:976] (1/7) Epoch 6, batch 2350, loss[loss=0.2067, simple_loss=0.2711, pruned_loss=0.07117, over 4930.00 frames. ], tot_loss[loss=0.2159, simple_loss=0.2765, pruned_loss=0.07771, over 953981.14 frames. ], batch size: 38, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:57:57,496 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30995.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:58:08,480 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 05:58:19,344 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31011.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 05:58:19,369 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31011.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 05:58:28,250 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31016.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:58:32,529 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4908, 1.0315, 0.7796, 1.3692, 1.8232, 0.8000, 1.2978, 1.3839], + device='cuda:1'), covar=tensor([0.1249, 0.1884, 0.1663, 0.0964, 0.1743, 0.2210, 0.1239, 0.1649], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0115, 0.0093, 0.0124, 0.0097, 0.0100, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 05:58:33,048 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.055e+02 1.621e+02 1.904e+02 2.250e+02 3.149e+02, threshold=3.808e+02, percent-clipped=0.0 +2023-03-26 05:58:40,639 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3663, 2.6991, 2.1040, 1.6608, 2.5442, 2.7224, 2.6383, 2.4036], + device='cuda:1'), covar=tensor([0.0681, 0.0565, 0.0928, 0.1068, 0.0775, 0.0738, 0.0670, 0.0879], + device='cuda:1'), in_proj_covar=tensor([0.0139, 0.0135, 0.0144, 0.0128, 0.0113, 0.0144, 0.0146, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 05:58:53,277 INFO [finetune.py:976] (1/7) Epoch 6, batch 2400, loss[loss=0.2281, simple_loss=0.2832, pruned_loss=0.08653, over 4831.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2743, pruned_loss=0.07735, over 953286.89 frames. ], batch size: 33, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 05:59:16,908 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=31064.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 05:59:24,050 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31072.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 05:59:44,204 INFO [finetune.py:976] (1/7) Epoch 6, batch 2450, loss[loss=0.2528, simple_loss=0.2851, pruned_loss=0.1103, over 3930.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2724, pruned_loss=0.07731, over 953061.86 frames. ], batch size: 17, lr: 3.91e-03, grad_scale: 32.0 +2023-03-26 06:00:11,373 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2551, 1.4353, 1.4053, 1.5990, 1.4951, 2.9079, 1.3419, 1.5503], + device='cuda:1'), covar=tensor([0.1018, 0.1618, 0.1145, 0.0943, 0.1508, 0.0315, 0.1359, 0.1567], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0081, 0.0077, 0.0079, 0.0092, 0.0082, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 06:00:21,949 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.316e+02 1.762e+02 2.221e+02 2.552e+02 5.044e+02, threshold=4.442e+02, percent-clipped=4.0 +2023-03-26 06:00:30,922 INFO [finetune.py:976] (1/7) Epoch 6, batch 2500, loss[loss=0.248, simple_loss=0.3006, pruned_loss=0.09771, over 4819.00 frames. ], tot_loss[loss=0.215, simple_loss=0.2738, pruned_loss=0.07807, over 954056.99 frames. ], batch size: 51, lr: 3.91e-03, grad_scale: 16.0 +2023-03-26 06:01:00,109 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7515, 3.8702, 3.7105, 1.8409, 3.8823, 2.9379, 0.7494, 2.7004], + device='cuda:1'), covar=tensor([0.2352, 0.1769, 0.1506, 0.3110, 0.0969, 0.0969, 0.4396, 0.1370], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0167, 0.0160, 0.0126, 0.0153, 0.0120, 0.0142, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 06:01:06,650 INFO [finetune.py:976] (1/7) Epoch 6, batch 2550, loss[loss=0.2633, simple_loss=0.3253, pruned_loss=0.1006, over 4871.00 frames. ], tot_loss[loss=0.2195, simple_loss=0.2787, pruned_loss=0.08021, over 953214.20 frames. ], batch size: 34, lr: 3.91e-03, grad_scale: 16.0 +2023-03-26 06:01:26,693 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.62 vs. limit=5.0 +2023-03-26 06:01:50,622 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.038e+02 1.680e+02 2.034e+02 2.315e+02 3.655e+02, threshold=4.067e+02, percent-clipped=0.0 +2023-03-26 06:02:04,704 INFO [finetune.py:976] (1/7) Epoch 6, batch 2600, loss[loss=0.2312, simple_loss=0.2972, pruned_loss=0.08258, over 4824.00 frames. ], tot_loss[loss=0.2181, simple_loss=0.2781, pruned_loss=0.07909, over 954868.28 frames. ], batch size: 51, lr: 3.91e-03, grad_scale: 16.0 +2023-03-26 06:02:33,427 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31259.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:02:58,100 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7204, 1.7186, 1.7491, 1.0441, 1.9538, 1.8972, 1.7640, 1.5100], + device='cuda:1'), covar=tensor([0.0606, 0.0744, 0.0687, 0.0947, 0.0484, 0.0648, 0.0625, 0.1143], + device='cuda:1'), in_proj_covar=tensor([0.0140, 0.0135, 0.0145, 0.0129, 0.0114, 0.0145, 0.0146, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:02:59,297 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31281.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:03:09,261 INFO [finetune.py:976] (1/7) Epoch 6, batch 2650, loss[loss=0.2554, simple_loss=0.3202, pruned_loss=0.09528, over 4740.00 frames. ], tot_loss[loss=0.2181, simple_loss=0.2785, pruned_loss=0.0789, over 955044.88 frames. ], batch size: 54, lr: 3.91e-03, grad_scale: 16.0 +2023-03-26 06:03:09,918 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31290.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:03:38,122 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31311.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:03:47,924 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.100e+02 1.761e+02 2.106e+02 2.462e+02 3.966e+02, threshold=4.213e+02, percent-clipped=0.0 +2023-03-26 06:03:59,088 INFO [finetune.py:976] (1/7) Epoch 6, batch 2700, loss[loss=0.1842, simple_loss=0.2557, pruned_loss=0.05632, over 4893.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2773, pruned_loss=0.07797, over 956102.81 frames. ], batch size: 43, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:04:20,301 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0888, 1.9489, 1.7403, 2.1569, 1.5617, 4.6605, 1.8972, 2.4518], + device='cuda:1'), covar=tensor([0.3231, 0.2321, 0.2001, 0.2035, 0.1654, 0.0100, 0.2134, 0.1195], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0118, 0.0122, 0.0117, 0.0099, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 06:04:23,326 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=31359.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:04:33,887 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31367.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 06:04:52,795 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3741, 1.5994, 1.9036, 0.9608, 1.7033, 1.9797, 1.9534, 1.5893], + device='cuda:1'), covar=tensor([0.0816, 0.0583, 0.0399, 0.0527, 0.0419, 0.0503, 0.0360, 0.0552], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0158, 0.0120, 0.0138, 0.0132, 0.0125, 0.0147, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.8214e-05, 1.1714e-04, 8.6826e-05, 1.0036e-04, 9.5008e-05, 9.2058e-05, + 1.0872e-04, 1.0690e-04], device='cuda:1') +2023-03-26 06:05:04,057 INFO [finetune.py:976] (1/7) Epoch 6, batch 2750, loss[loss=0.2178, simple_loss=0.2737, pruned_loss=0.0809, over 4796.00 frames. ], tot_loss[loss=0.2144, simple_loss=0.2743, pruned_loss=0.07721, over 956783.49 frames. ], batch size: 51, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:05:28,033 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31409.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:05:46,888 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.163e+02 1.636e+02 1.989e+02 2.265e+02 3.886e+02, threshold=3.977e+02, percent-clipped=0.0 +2023-03-26 06:05:54,617 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6783, 2.3168, 2.0258, 1.0601, 2.2432, 2.0183, 1.7239, 2.0560], + device='cuda:1'), covar=tensor([0.0849, 0.1075, 0.1656, 0.2494, 0.1597, 0.2311, 0.2462, 0.1262], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0202, 0.0201, 0.0190, 0.0217, 0.0209, 0.0221, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:05:56,330 INFO [finetune.py:976] (1/7) Epoch 6, batch 2800, loss[loss=0.1643, simple_loss=0.2268, pruned_loss=0.05091, over 4805.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2708, pruned_loss=0.07553, over 956822.69 frames. ], batch size: 25, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:06:16,534 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31470.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:06:21,212 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4997, 1.3202, 1.2066, 1.4797, 1.6254, 1.4619, 0.8272, 1.2219], + device='cuda:1'), covar=tensor([0.2270, 0.2345, 0.2041, 0.1815, 0.1738, 0.1299, 0.2972, 0.1897], + device='cuda:1'), in_proj_covar=tensor([0.0235, 0.0209, 0.0203, 0.0186, 0.0237, 0.0176, 0.0213, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:06:42,527 INFO [finetune.py:976] (1/7) Epoch 6, batch 2850, loss[loss=0.1511, simple_loss=0.2196, pruned_loss=0.0413, over 4761.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2696, pruned_loss=0.0754, over 956368.37 frames. ], batch size: 26, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:06:43,858 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31491.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 06:07:26,132 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.153e+02 1.576e+02 1.974e+02 2.520e+02 5.037e+02, threshold=3.948e+02, percent-clipped=2.0 +2023-03-26 06:07:41,440 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0606, 2.0780, 1.9685, 1.3674, 2.2414, 2.0999, 2.0664, 1.7877], + device='cuda:1'), covar=tensor([0.0681, 0.0650, 0.0806, 0.0966, 0.0464, 0.0828, 0.0720, 0.1050], + device='cuda:1'), in_proj_covar=tensor([0.0140, 0.0135, 0.0146, 0.0129, 0.0115, 0.0145, 0.0147, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:07:42,542 INFO [finetune.py:976] (1/7) Epoch 6, batch 2900, loss[loss=0.1719, simple_loss=0.2233, pruned_loss=0.06029, over 4801.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2727, pruned_loss=0.07643, over 957838.07 frames. ], batch size: 25, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:07:43,333 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.89 vs. limit=5.0 +2023-03-26 06:07:55,763 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31552.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 06:08:04,326 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31559.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:08:15,056 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5033, 1.3341, 1.3804, 1.4379, 1.1064, 2.9826, 1.1543, 1.6633], + device='cuda:1'), covar=tensor([0.3448, 0.2462, 0.2154, 0.2458, 0.1939, 0.0226, 0.2711, 0.1314], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0115, 0.0119, 0.0123, 0.0118, 0.0099, 0.0102, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 06:08:19,196 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31581.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:08:27,343 INFO [finetune.py:976] (1/7) Epoch 6, batch 2950, loss[loss=0.2924, simple_loss=0.3446, pruned_loss=0.1201, over 4836.00 frames. ], tot_loss[loss=0.2177, simple_loss=0.2779, pruned_loss=0.07878, over 956879.62 frames. ], batch size: 49, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:08:33,422 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31590.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:08:47,761 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=31607.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:08:55,272 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6998, 1.5239, 1.4996, 1.4029, 1.7429, 1.4655, 1.8225, 1.6495], + device='cuda:1'), covar=tensor([0.1486, 0.2574, 0.3151, 0.2630, 0.2634, 0.1762, 0.2839, 0.2044], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0193, 0.0237, 0.0254, 0.0232, 0.0192, 0.0212, 0.0192], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:08:59,951 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.136e+02 1.780e+02 2.105e+02 2.565e+02 4.082e+02, threshold=4.210e+02, percent-clipped=1.0 +2023-03-26 06:09:02,945 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=31629.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:09:09,344 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=31638.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:09:09,893 INFO [finetune.py:976] (1/7) Epoch 6, batch 3000, loss[loss=0.2295, simple_loss=0.2901, pruned_loss=0.08443, over 4802.00 frames. ], tot_loss[loss=0.2202, simple_loss=0.2803, pruned_loss=0.08009, over 955115.99 frames. ], batch size: 51, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:09:09,893 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 06:09:13,434 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6022, 1.4833, 2.0210, 2.9533, 2.1679, 2.2424, 0.8841, 2.3485], + device='cuda:1'), covar=tensor([0.1802, 0.1591, 0.1195, 0.0621, 0.0814, 0.1263, 0.1915, 0.0660], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0119, 0.0138, 0.0168, 0.0103, 0.0143, 0.0130, 0.0104], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 06:09:16,633 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6786, 1.5220, 1.5257, 1.6369, 1.0266, 3.0653, 1.1866, 1.7649], + device='cuda:1'), covar=tensor([0.3439, 0.2441, 0.2203, 0.2405, 0.2123, 0.0280, 0.2686, 0.1359], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0115, 0.0119, 0.0123, 0.0118, 0.0099, 0.0102, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 06:09:19,925 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4615, 1.2784, 1.2829, 1.3371, 1.6558, 1.5782, 1.4265, 1.2442], + device='cuda:1'), covar=tensor([0.0301, 0.0328, 0.0618, 0.0324, 0.0279, 0.0435, 0.0333, 0.0441], + device='cuda:1'), in_proj_covar=tensor([0.0086, 0.0111, 0.0137, 0.0116, 0.0103, 0.0100, 0.0090, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.7290e-05, 8.7190e-05, 1.0945e-04, 9.1587e-05, 8.1257e-05, 7.4104e-05, + 6.8327e-05, 8.4034e-05], device='cuda:1') +2023-03-26 06:09:23,489 INFO [finetune.py:1010] (1/7) Epoch 6, validation: loss=0.1625, simple_loss=0.2344, pruned_loss=0.04534, over 2265189.00 frames. +2023-03-26 06:09:23,489 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 06:09:55,814 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31667.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 06:10:13,512 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0386, 1.6729, 2.3277, 3.9150, 2.7242, 2.6378, 0.8200, 3.1232], + device='cuda:1'), covar=tensor([0.1830, 0.1604, 0.1414, 0.0487, 0.0799, 0.1719, 0.2067, 0.0560], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0120, 0.0138, 0.0169, 0.0104, 0.0144, 0.0131, 0.0104], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 06:10:19,292 INFO [finetune.py:976] (1/7) Epoch 6, batch 3050, loss[loss=0.215, simple_loss=0.2723, pruned_loss=0.0789, over 4835.00 frames. ], tot_loss[loss=0.2214, simple_loss=0.2817, pruned_loss=0.08048, over 955382.91 frames. ], batch size: 47, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:10:20,014 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1391, 1.3436, 0.9744, 1.2940, 1.4394, 2.5098, 1.2503, 1.5072], + device='cuda:1'), covar=tensor([0.0929, 0.1816, 0.1209, 0.0956, 0.1623, 0.0377, 0.1454, 0.1650], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0082, 0.0077, 0.0079, 0.0092, 0.0084, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 06:10:27,505 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4032, 2.3881, 2.0038, 1.0324, 2.1570, 1.8622, 1.6152, 2.0882], + device='cuda:1'), covar=tensor([0.0839, 0.0698, 0.1520, 0.2166, 0.1397, 0.2414, 0.2336, 0.1084], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0202, 0.0202, 0.0190, 0.0217, 0.0209, 0.0221, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:10:29,958 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31704.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:10:34,225 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0990, 2.1372, 2.0920, 1.4998, 2.2765, 2.3025, 2.1705, 1.8377], + device='cuda:1'), covar=tensor([0.0620, 0.0545, 0.0687, 0.0895, 0.0460, 0.0636, 0.0600, 0.0971], + device='cuda:1'), in_proj_covar=tensor([0.0139, 0.0135, 0.0144, 0.0129, 0.0113, 0.0145, 0.0146, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:10:36,649 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=31715.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 06:10:43,058 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.029e+02 1.708e+02 2.072e+02 2.420e+02 4.871e+02, threshold=4.144e+02, percent-clipped=1.0 +2023-03-26 06:10:44,414 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6646, 1.4255, 1.0910, 0.5296, 1.3066, 1.4763, 1.3010, 1.3480], + device='cuda:1'), covar=tensor([0.0574, 0.0537, 0.0862, 0.1328, 0.0793, 0.1220, 0.1219, 0.0597], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0202, 0.0201, 0.0190, 0.0217, 0.0209, 0.0221, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:10:59,685 INFO [finetune.py:976] (1/7) Epoch 6, batch 3100, loss[loss=0.1569, simple_loss=0.2078, pruned_loss=0.053, over 4344.00 frames. ], tot_loss[loss=0.2192, simple_loss=0.2792, pruned_loss=0.07958, over 952990.97 frames. ], batch size: 19, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:11:17,251 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5632, 3.6902, 3.4515, 1.7897, 3.7760, 2.7648, 0.8830, 2.6519], + device='cuda:1'), covar=tensor([0.2580, 0.1589, 0.1538, 0.3145, 0.0979, 0.1000, 0.4320, 0.1410], + device='cuda:1'), in_proj_covar=tensor([0.0156, 0.0172, 0.0163, 0.0128, 0.0156, 0.0123, 0.0145, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 06:11:20,230 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31765.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:11:20,275 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31765.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:11:35,801 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31788.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:11:36,281 INFO [finetune.py:976] (1/7) Epoch 6, batch 3150, loss[loss=0.2445, simple_loss=0.2933, pruned_loss=0.09786, over 4824.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2757, pruned_loss=0.07832, over 953967.37 frames. ], batch size: 40, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:12:00,894 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31823.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:12:01,992 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.748e+01 1.682e+02 2.020e+02 2.535e+02 4.605e+02, threshold=4.040e+02, percent-clipped=1.0 +2023-03-26 06:12:05,038 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4505, 1.6437, 1.7214, 1.0731, 1.6636, 1.9377, 1.9904, 1.4731], + device='cuda:1'), covar=tensor([0.1082, 0.0554, 0.0430, 0.0575, 0.0447, 0.0544, 0.0291, 0.0674], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0158, 0.0121, 0.0138, 0.0133, 0.0124, 0.0147, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.8235e-05, 1.1680e-04, 8.7390e-05, 1.0036e-04, 9.5885e-05, 9.1986e-05, + 1.0894e-04, 1.0685e-04], device='cuda:1') +2023-03-26 06:12:15,952 INFO [finetune.py:976] (1/7) Epoch 6, batch 3200, loss[loss=0.1993, simple_loss=0.2551, pruned_loss=0.07174, over 4829.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2707, pruned_loss=0.07598, over 953670.69 frames. ], batch size: 40, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:12:25,649 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31847.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 06:12:31,916 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31849.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:12:44,573 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8679, 1.6778, 2.2600, 1.5624, 1.9760, 2.0874, 1.6863, 2.2750], + device='cuda:1'), covar=tensor([0.1537, 0.2056, 0.1487, 0.2105, 0.1096, 0.1672, 0.2827, 0.1005], + device='cuda:1'), in_proj_covar=tensor([0.0206, 0.0205, 0.0199, 0.0196, 0.0184, 0.0222, 0.0219, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:12:53,305 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 06:13:06,134 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31878.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:13:11,180 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31884.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:13:14,090 INFO [finetune.py:976] (1/7) Epoch 6, batch 3250, loss[loss=0.258, simple_loss=0.3165, pruned_loss=0.09975, over 4810.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2715, pruned_loss=0.0763, over 953881.76 frames. ], batch size: 39, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:13:51,961 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31915.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:14:01,796 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.169e+02 1.695e+02 2.048e+02 2.376e+02 4.231e+02, threshold=4.096e+02, percent-clipped=1.0 +2023-03-26 06:14:01,983 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-03-26 06:14:17,633 INFO [finetune.py:976] (1/7) Epoch 6, batch 3300, loss[loss=0.2129, simple_loss=0.2795, pruned_loss=0.07315, over 4913.00 frames. ], tot_loss[loss=0.216, simple_loss=0.2761, pruned_loss=0.07799, over 954795.22 frames. ], batch size: 36, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:14:17,758 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31939.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:14:29,141 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31950.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:14:46,585 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31976.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:14:47,736 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7378, 3.7553, 3.5890, 1.7569, 3.8562, 2.7817, 0.6986, 2.5643], + device='cuda:1'), covar=tensor([0.2512, 0.1736, 0.1547, 0.3385, 0.1000, 0.1032, 0.4494, 0.1397], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0169, 0.0161, 0.0127, 0.0154, 0.0121, 0.0143, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 06:14:49,086 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.04 vs. limit=5.0 +2023-03-26 06:14:54,838 INFO [finetune.py:976] (1/7) Epoch 6, batch 3350, loss[loss=0.2358, simple_loss=0.2935, pruned_loss=0.08908, over 4816.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2777, pruned_loss=0.07843, over 953770.32 frames. ], batch size: 38, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:14:59,186 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1103, 1.8371, 2.6096, 3.9267, 2.7738, 2.7225, 0.7533, 3.1138], + device='cuda:1'), covar=tensor([0.1790, 0.1515, 0.1275, 0.0511, 0.0772, 0.1684, 0.2246, 0.0546], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0119, 0.0137, 0.0168, 0.0103, 0.0143, 0.0130, 0.0104], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 06:15:11,495 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32011.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:15:22,687 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.186e+02 1.743e+02 2.097e+02 2.540e+02 4.089e+02, threshold=4.194e+02, percent-clipped=0.0 +2023-03-26 06:15:31,411 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9494, 1.4532, 0.9225, 1.7759, 2.1131, 1.5472, 1.7455, 1.7763], + device='cuda:1'), covar=tensor([0.1445, 0.2033, 0.2157, 0.1150, 0.1987, 0.1919, 0.1346, 0.1914], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0115, 0.0092, 0.0124, 0.0096, 0.0100, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 06:15:32,661 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2194, 2.1297, 1.6032, 2.3478, 2.1499, 1.7937, 2.8379, 2.2248], + device='cuda:1'), covar=tensor([0.1723, 0.3664, 0.4214, 0.3548, 0.3312, 0.2091, 0.4106, 0.2463], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0192, 0.0236, 0.0253, 0.0231, 0.0191, 0.0211, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:15:41,803 INFO [finetune.py:976] (1/7) Epoch 6, batch 3400, loss[loss=0.2213, simple_loss=0.2895, pruned_loss=0.07649, over 4812.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2787, pruned_loss=0.07869, over 955264.79 frames. ], batch size: 33, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:15:43,135 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7633, 1.7700, 1.5376, 1.8741, 2.5546, 1.8930, 1.6887, 1.4099], + device='cuda:1'), covar=tensor([0.2192, 0.2001, 0.1896, 0.1691, 0.1608, 0.1175, 0.2433, 0.1904], + device='cuda:1'), in_proj_covar=tensor([0.0234, 0.0208, 0.0202, 0.0185, 0.0237, 0.0175, 0.0213, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:16:04,433 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32060.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:16:13,825 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32065.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:16:41,401 INFO [finetune.py:976] (1/7) Epoch 6, batch 3450, loss[loss=0.2272, simple_loss=0.2787, pruned_loss=0.08782, over 4885.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2772, pruned_loss=0.07803, over 954271.82 frames. ], batch size: 35, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:17:01,308 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9974, 1.2826, 1.7136, 1.7359, 1.5552, 1.5871, 1.6615, 1.7293], + device='cuda:1'), covar=tensor([0.7935, 0.9512, 0.8309, 0.8529, 1.0338, 0.7690, 1.1026, 0.7827], + device='cuda:1'), in_proj_covar=tensor([0.0231, 0.0247, 0.0256, 0.0258, 0.0243, 0.0221, 0.0275, 0.0226], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:17:07,733 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=32113.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:17:16,937 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.122e+02 1.573e+02 1.926e+02 2.516e+02 4.351e+02, threshold=3.853e+02, percent-clipped=2.0 +2023-03-26 06:17:25,471 INFO [finetune.py:976] (1/7) Epoch 6, batch 3500, loss[loss=0.2615, simple_loss=0.3057, pruned_loss=0.1086, over 4709.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.275, pruned_loss=0.07765, over 953472.95 frames. ], batch size: 23, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:17:29,064 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32144.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:17:30,888 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32147.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 06:18:07,557 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3364, 2.1598, 1.8051, 0.9374, 2.0171, 1.8458, 1.6736, 1.8828], + device='cuda:1'), covar=tensor([0.0845, 0.0748, 0.1248, 0.1824, 0.1260, 0.1824, 0.1826, 0.0947], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0202, 0.0200, 0.0189, 0.0216, 0.0208, 0.0220, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:18:09,961 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32179.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:18:15,990 INFO [finetune.py:976] (1/7) Epoch 6, batch 3550, loss[loss=0.1811, simple_loss=0.2455, pruned_loss=0.05831, over 4754.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2721, pruned_loss=0.07638, over 955483.97 frames. ], batch size: 28, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:18:19,691 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=32195.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 06:18:21,523 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.93 vs. limit=5.0 +2023-03-26 06:18:40,463 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.015e+02 1.617e+02 1.889e+02 2.318e+02 4.823e+02, threshold=3.777e+02, percent-clipped=2.0 +2023-03-26 06:18:52,022 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32234.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:18:52,121 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1608, 1.5247, 1.9623, 1.9501, 1.7707, 1.7770, 1.8541, 1.8212], + device='cuda:1'), covar=tensor([0.4694, 0.6555, 0.5977, 0.5877, 0.7497, 0.5445, 0.7797, 0.5220], + device='cuda:1'), in_proj_covar=tensor([0.0229, 0.0245, 0.0254, 0.0255, 0.0241, 0.0219, 0.0273, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:19:00,027 INFO [finetune.py:976] (1/7) Epoch 6, batch 3600, loss[loss=0.1726, simple_loss=0.2421, pruned_loss=0.05155, over 4783.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2693, pruned_loss=0.07533, over 956120.70 frames. ], batch size: 29, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:19:36,027 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32271.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:19:57,774 INFO [finetune.py:976] (1/7) Epoch 6, batch 3650, loss[loss=0.2083, simple_loss=0.2703, pruned_loss=0.07311, over 4746.00 frames. ], tot_loss[loss=0.214, simple_loss=0.2736, pruned_loss=0.07725, over 956641.61 frames. ], batch size: 27, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:20:14,262 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32306.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:20:17,997 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0598, 0.8271, 0.9119, 1.0193, 1.1699, 1.1215, 1.0218, 0.9691], + device='cuda:1'), covar=tensor([0.0297, 0.0286, 0.0589, 0.0257, 0.0271, 0.0412, 0.0274, 0.0396], + device='cuda:1'), in_proj_covar=tensor([0.0086, 0.0110, 0.0136, 0.0115, 0.0102, 0.0098, 0.0089, 0.0107], + device='cuda:1'), out_proj_covar=tensor([6.7039e-05, 8.6961e-05, 1.0903e-04, 9.0758e-05, 8.0731e-05, 7.3125e-05, + 6.7995e-05, 8.3578e-05], device='cuda:1') +2023-03-26 06:20:26,743 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.020e+02 1.776e+02 2.129e+02 2.587e+02 5.126e+02, threshold=4.259e+02, percent-clipped=5.0 +2023-03-26 06:20:34,999 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 06:20:43,050 INFO [finetune.py:976] (1/7) Epoch 6, batch 3700, loss[loss=0.2061, simple_loss=0.2709, pruned_loss=0.07066, over 4866.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.2767, pruned_loss=0.07842, over 955465.68 frames. ], batch size: 31, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:20:56,543 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32360.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:21:16,591 INFO [finetune.py:976] (1/7) Epoch 6, batch 3750, loss[loss=0.1811, simple_loss=0.249, pruned_loss=0.05661, over 4847.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.2773, pruned_loss=0.07804, over 954912.68 frames. ], batch size: 31, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:21:23,858 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5874, 3.9609, 4.1404, 4.4890, 4.3309, 4.1090, 4.7080, 1.5059], + device='cuda:1'), covar=tensor([0.0767, 0.0788, 0.0790, 0.0864, 0.1251, 0.1376, 0.0566, 0.5370], + device='cuda:1'), in_proj_covar=tensor([0.0355, 0.0242, 0.0275, 0.0293, 0.0333, 0.0283, 0.0302, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:21:37,052 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=32408.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:21:48,323 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.045e+02 1.626e+02 1.929e+02 2.294e+02 3.909e+02, threshold=3.857e+02, percent-clipped=0.0 +2023-03-26 06:21:58,657 INFO [finetune.py:976] (1/7) Epoch 6, batch 3800, loss[loss=0.1819, simple_loss=0.2635, pruned_loss=0.05012, over 4815.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2776, pruned_loss=0.07765, over 954911.48 frames. ], batch size: 41, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:22:01,196 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8934, 2.6851, 3.2095, 2.3045, 2.8209, 3.2439, 2.7101, 3.4093], + device='cuda:1'), covar=tensor([0.1516, 0.2024, 0.1606, 0.2458, 0.1102, 0.1678, 0.2491, 0.0906], + device='cuda:1'), in_proj_covar=tensor([0.0206, 0.0206, 0.0200, 0.0197, 0.0185, 0.0222, 0.0218, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:22:01,799 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32444.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:22:01,837 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0591, 1.9636, 1.5847, 2.0110, 1.8992, 1.9037, 1.8055, 2.8036], + device='cuda:1'), covar=tensor([0.6659, 0.7782, 0.5568, 0.7386, 0.6787, 0.3853, 0.7123, 0.2514], + device='cuda:1'), in_proj_covar=tensor([0.0281, 0.0257, 0.0219, 0.0282, 0.0238, 0.0202, 0.0244, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:22:09,071 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4989, 2.6851, 2.2618, 1.6571, 2.9077, 2.8550, 2.7352, 2.3819], + device='cuda:1'), covar=tensor([0.0622, 0.0527, 0.0841, 0.1001, 0.0481, 0.0615, 0.0638, 0.0851], + device='cuda:1'), in_proj_covar=tensor([0.0141, 0.0136, 0.0146, 0.0130, 0.0114, 0.0146, 0.0147, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:22:10,239 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0 +2023-03-26 06:22:23,237 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9062, 1.5744, 2.2644, 1.5780, 1.9711, 2.1646, 1.6470, 2.2611], + device='cuda:1'), covar=tensor([0.1669, 0.2136, 0.1475, 0.2275, 0.1092, 0.1660, 0.2832, 0.0996], + device='cuda:1'), in_proj_covar=tensor([0.0206, 0.0206, 0.0200, 0.0196, 0.0185, 0.0221, 0.0218, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:22:24,408 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32479.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:22:31,248 INFO [finetune.py:976] (1/7) Epoch 6, batch 3850, loss[loss=0.1887, simple_loss=0.2554, pruned_loss=0.061, over 4891.00 frames. ], tot_loss[loss=0.2148, simple_loss=0.2759, pruned_loss=0.07687, over 955666.75 frames. ], batch size: 32, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:22:33,633 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=32492.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:22:53,999 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.617e+01 1.620e+02 2.056e+02 2.694e+02 5.558e+02, threshold=4.113e+02, percent-clipped=4.0 +2023-03-26 06:22:55,288 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=32527.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:23:00,103 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32534.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:23:04,461 INFO [finetune.py:976] (1/7) Epoch 6, batch 3900, loss[loss=0.1868, simple_loss=0.2432, pruned_loss=0.06523, over 4168.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2733, pruned_loss=0.0768, over 954748.41 frames. ], batch size: 65, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:23:24,756 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32571.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:23:27,201 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7507, 1.6339, 2.2358, 1.5269, 1.8786, 2.0343, 1.6431, 2.2543], + device='cuda:1'), covar=tensor([0.1598, 0.2110, 0.1455, 0.2145, 0.1027, 0.1498, 0.2565, 0.0981], + device='cuda:1'), in_proj_covar=tensor([0.0206, 0.0206, 0.0199, 0.0196, 0.0184, 0.0220, 0.0218, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:23:31,329 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=32582.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:23:34,494 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.59 vs. limit=5.0 +2023-03-26 06:23:36,051 INFO [finetune.py:976] (1/7) Epoch 6, batch 3950, loss[loss=0.2285, simple_loss=0.2673, pruned_loss=0.0949, over 4336.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2696, pruned_loss=0.07523, over 954003.56 frames. ], batch size: 19, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:23:53,885 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32606.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:24:00,063 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.20 vs. limit=5.0 +2023-03-26 06:24:07,658 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=32619.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:24:11,251 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.084e+02 1.672e+02 2.141e+02 2.509e+02 4.478e+02, threshold=4.281e+02, percent-clipped=2.0 +2023-03-26 06:24:20,841 INFO [finetune.py:976] (1/7) Epoch 6, batch 4000, loss[loss=0.2006, simple_loss=0.2729, pruned_loss=0.06417, over 4911.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2701, pruned_loss=0.07624, over 954692.32 frames. ], batch size: 37, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:24:30,954 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=32654.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:24:55,641 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0771, 0.7927, 0.8718, 1.0881, 1.2019, 1.1866, 1.0121, 0.9430], + device='cuda:1'), covar=tensor([0.0298, 0.0349, 0.0643, 0.0291, 0.0276, 0.0435, 0.0282, 0.0430], + device='cuda:1'), in_proj_covar=tensor([0.0086, 0.0110, 0.0135, 0.0114, 0.0102, 0.0098, 0.0089, 0.0106], + device='cuda:1'), out_proj_covar=tensor([6.6982e-05, 8.6403e-05, 1.0828e-04, 9.0139e-05, 8.0253e-05, 7.2895e-05, + 6.7934e-05, 8.2956e-05], device='cuda:1') +2023-03-26 06:25:04,480 INFO [finetune.py:976] (1/7) Epoch 6, batch 4050, loss[loss=0.2772, simple_loss=0.3222, pruned_loss=0.1161, over 4147.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.272, pruned_loss=0.07676, over 949359.28 frames. ], batch size: 65, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:25:28,436 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.148e+02 1.831e+02 2.133e+02 2.637e+02 5.226e+02, threshold=4.267e+02, percent-clipped=1.0 +2023-03-26 06:25:40,996 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8148, 1.6964, 1.5686, 1.8484, 2.2995, 1.8694, 1.4645, 1.4456], + device='cuda:1'), covar=tensor([0.2317, 0.2175, 0.2009, 0.1801, 0.1969, 0.1240, 0.2691, 0.1957], + device='cuda:1'), in_proj_covar=tensor([0.0235, 0.0209, 0.0202, 0.0185, 0.0237, 0.0175, 0.0211, 0.0190], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:25:42,718 INFO [finetune.py:976] (1/7) Epoch 6, batch 4100, loss[loss=0.1948, simple_loss=0.2596, pruned_loss=0.06499, over 4813.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2756, pruned_loss=0.07782, over 950507.91 frames. ], batch size: 40, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:26:34,200 INFO [finetune.py:976] (1/7) Epoch 6, batch 4150, loss[loss=0.2486, simple_loss=0.3096, pruned_loss=0.09378, over 4807.00 frames. ], tot_loss[loss=0.2172, simple_loss=0.2772, pruned_loss=0.07859, over 950611.86 frames. ], batch size: 40, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:27:18,094 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.203e+02 1.784e+02 2.149e+02 2.523e+02 4.029e+02, threshold=4.299e+02, percent-clipped=0.0 +2023-03-26 06:27:28,919 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8944, 1.6899, 1.4449, 1.6008, 1.8465, 1.5630, 2.1132, 1.8031], + device='cuda:1'), covar=tensor([0.1522, 0.2876, 0.3649, 0.3089, 0.2873, 0.1947, 0.3104, 0.2128], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0190, 0.0233, 0.0251, 0.0229, 0.0189, 0.0210, 0.0189], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:27:37,294 INFO [finetune.py:976] (1/7) Epoch 6, batch 4200, loss[loss=0.1852, simple_loss=0.2416, pruned_loss=0.06439, over 4872.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2772, pruned_loss=0.07796, over 953364.66 frames. ], batch size: 34, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:28:01,957 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3414, 1.2587, 1.5800, 1.1180, 1.2152, 1.4220, 1.2912, 1.5467], + device='cuda:1'), covar=tensor([0.1151, 0.2019, 0.1088, 0.1444, 0.0946, 0.1167, 0.2782, 0.0768], + device='cuda:1'), in_proj_covar=tensor([0.0203, 0.0204, 0.0198, 0.0195, 0.0183, 0.0219, 0.0216, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:28:34,220 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5616, 1.5026, 1.8567, 1.8111, 1.5780, 3.4207, 1.2892, 1.7058], + device='cuda:1'), covar=tensor([0.0922, 0.1706, 0.1183, 0.0968, 0.1524, 0.0247, 0.1420, 0.1667], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0081, 0.0077, 0.0079, 0.0093, 0.0083, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 06:28:34,720 INFO [finetune.py:976] (1/7) Epoch 6, batch 4250, loss[loss=0.1749, simple_loss=0.232, pruned_loss=0.0589, over 4844.00 frames. ], tot_loss[loss=0.2153, simple_loss=0.2753, pruned_loss=0.07764, over 952665.09 frames. ], batch size: 44, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:28:42,472 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-03-26 06:29:25,480 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.113e+02 1.595e+02 1.920e+02 2.303e+02 3.727e+02, threshold=3.841e+02, percent-clipped=0.0 +2023-03-26 06:29:44,594 INFO [finetune.py:976] (1/7) Epoch 6, batch 4300, loss[loss=0.2008, simple_loss=0.2599, pruned_loss=0.07086, over 4917.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2725, pruned_loss=0.07653, over 953688.51 frames. ], batch size: 37, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:30:06,326 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9213, 0.9929, 1.7826, 1.6728, 1.5837, 1.5746, 1.5507, 1.6255], + device='cuda:1'), covar=tensor([0.5295, 0.6991, 0.5688, 0.6149, 0.7363, 0.5367, 0.7313, 0.5369], + device='cuda:1'), in_proj_covar=tensor([0.0229, 0.0244, 0.0253, 0.0256, 0.0242, 0.0219, 0.0272, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:30:17,417 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32966.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:30:47,043 INFO [finetune.py:976] (1/7) Epoch 6, batch 4350, loss[loss=0.199, simple_loss=0.2683, pruned_loss=0.06483, over 4919.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2693, pruned_loss=0.07529, over 954907.23 frames. ], batch size: 37, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:31:32,016 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.139e+02 1.730e+02 2.041e+02 2.589e+02 3.941e+02, threshold=4.082e+02, percent-clipped=1.0 +2023-03-26 06:31:33,338 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33027.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:31:50,659 INFO [finetune.py:976] (1/7) Epoch 6, batch 4400, loss[loss=0.2213, simple_loss=0.2796, pruned_loss=0.08151, over 4912.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.2729, pruned_loss=0.07744, over 955987.25 frames. ], batch size: 37, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:32:31,754 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33069.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:32:54,583 INFO [finetune.py:976] (1/7) Epoch 6, batch 4450, loss[loss=0.1981, simple_loss=0.2658, pruned_loss=0.0652, over 4906.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2759, pruned_loss=0.07784, over 957072.00 frames. ], batch size: 37, lr: 3.90e-03, grad_scale: 16.0 +2023-03-26 06:33:28,743 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8165, 1.4500, 1.1142, 1.7857, 2.0547, 1.5292, 1.6856, 1.8788], + device='cuda:1'), covar=tensor([0.1220, 0.1718, 0.1913, 0.0982, 0.1777, 0.1910, 0.1130, 0.1485], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0114, 0.0092, 0.0123, 0.0096, 0.0100, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 06:33:39,235 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.174e+02 1.714e+02 2.125e+02 2.690e+02 4.211e+02, threshold=4.250e+02, percent-clipped=1.0 +2023-03-26 06:33:47,519 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33130.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:33:58,507 INFO [finetune.py:976] (1/7) Epoch 6, batch 4500, loss[loss=0.2238, simple_loss=0.2723, pruned_loss=0.08762, over 4233.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2771, pruned_loss=0.07852, over 955072.25 frames. ], batch size: 65, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:34:19,040 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33155.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:34:53,891 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9960, 1.4536, 1.8465, 1.8479, 1.6642, 1.6749, 1.7526, 1.6736], + device='cuda:1'), covar=tensor([0.5044, 0.6546, 0.5843, 0.5957, 0.7614, 0.5388, 0.8071, 0.5356], + device='cuda:1'), in_proj_covar=tensor([0.0229, 0.0244, 0.0253, 0.0256, 0.0242, 0.0219, 0.0272, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:34:59,387 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1668, 1.0603, 1.0865, 0.5083, 0.8417, 1.2070, 1.2812, 1.0803], + device='cuda:1'), covar=tensor([0.0807, 0.0450, 0.0373, 0.0445, 0.0455, 0.0448, 0.0328, 0.0516], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0159, 0.0122, 0.0139, 0.0134, 0.0125, 0.0148, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.7774e-05, 1.1747e-04, 8.8282e-05, 1.0141e-04, 9.5959e-05, 9.2594e-05, + 1.0998e-04, 1.0853e-04], device='cuda:1') +2023-03-26 06:35:01,058 INFO [finetune.py:976] (1/7) Epoch 6, batch 4550, loss[loss=0.2261, simple_loss=0.294, pruned_loss=0.07906, over 4892.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2794, pruned_loss=0.07938, over 956851.73 frames. ], batch size: 43, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:35:33,145 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33216.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:35:44,645 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.755e+02 2.084e+02 2.335e+02 4.621e+02, threshold=4.168e+02, percent-clipped=2.0 +2023-03-26 06:36:05,046 INFO [finetune.py:976] (1/7) Epoch 6, batch 4600, loss[loss=0.1836, simple_loss=0.2574, pruned_loss=0.05488, over 4849.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2781, pruned_loss=0.07802, over 958524.09 frames. ], batch size: 44, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:36:36,140 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1966, 2.0365, 2.1896, 0.8628, 2.2925, 2.6423, 2.2122, 2.0380], + device='cuda:1'), covar=tensor([0.0806, 0.0647, 0.0362, 0.0695, 0.0470, 0.0333, 0.0395, 0.0540], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0159, 0.0122, 0.0139, 0.0134, 0.0125, 0.0149, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.7659e-05, 1.1764e-04, 8.8526e-05, 1.0122e-04, 9.6054e-05, 9.2490e-05, + 1.1001e-04, 1.0833e-04], device='cuda:1') +2023-03-26 06:37:07,545 INFO [finetune.py:976] (1/7) Epoch 6, batch 4650, loss[loss=0.2139, simple_loss=0.2691, pruned_loss=0.07936, over 4742.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2738, pruned_loss=0.07589, over 959361.57 frames. ], batch size: 59, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:37:48,450 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33322.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:37:50,591 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.141e+02 1.572e+02 1.961e+02 2.434e+02 3.752e+02, threshold=3.921e+02, percent-clipped=0.0 +2023-03-26 06:38:10,951 INFO [finetune.py:976] (1/7) Epoch 6, batch 4700, loss[loss=0.1711, simple_loss=0.2473, pruned_loss=0.04744, over 4897.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.2699, pruned_loss=0.07444, over 957733.22 frames. ], batch size: 32, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:39:20,540 INFO [finetune.py:976] (1/7) Epoch 6, batch 4750, loss[loss=0.1812, simple_loss=0.2324, pruned_loss=0.065, over 4824.00 frames. ], tot_loss[loss=0.207, simple_loss=0.2668, pruned_loss=0.07361, over 956581.14 frames. ], batch size: 25, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:39:24,280 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0816, 2.0157, 1.7839, 2.1290, 2.4796, 2.0257, 1.7751, 1.5250], + device='cuda:1'), covar=tensor([0.2435, 0.2130, 0.1952, 0.1756, 0.2241, 0.1191, 0.2588, 0.1954], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0210, 0.0203, 0.0186, 0.0238, 0.0176, 0.0213, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:40:04,157 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.123e+02 1.633e+02 1.917e+02 2.350e+02 3.562e+02, threshold=3.835e+02, percent-clipped=0.0 +2023-03-26 06:40:04,237 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33425.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:40:24,233 INFO [finetune.py:976] (1/7) Epoch 6, batch 4800, loss[loss=0.2433, simple_loss=0.2916, pruned_loss=0.09746, over 4748.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2714, pruned_loss=0.07603, over 955901.86 frames. ], batch size: 27, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:40:26,732 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 06:41:07,475 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33472.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:41:28,614 INFO [finetune.py:976] (1/7) Epoch 6, batch 4850, loss[loss=0.2747, simple_loss=0.3355, pruned_loss=0.107, over 4900.00 frames. ], tot_loss[loss=0.2148, simple_loss=0.2757, pruned_loss=0.07701, over 956542.54 frames. ], batch size: 43, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:41:41,809 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 06:41:59,889 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33511.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:42:14,017 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.198e+02 1.841e+02 2.130e+02 2.477e+02 4.983e+02, threshold=4.260e+02, percent-clipped=3.0 +2023-03-26 06:42:24,106 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33533.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:42:32,810 INFO [finetune.py:976] (1/7) Epoch 6, batch 4900, loss[loss=0.2057, simple_loss=0.2629, pruned_loss=0.07428, over 4814.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2772, pruned_loss=0.07787, over 955468.96 frames. ], batch size: 39, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:43:04,828 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3866, 1.2285, 1.4009, 0.8110, 1.5493, 1.4350, 1.3954, 1.1058], + device='cuda:1'), covar=tensor([0.0733, 0.0879, 0.0783, 0.1100, 0.0723, 0.0787, 0.0798, 0.1648], + device='cuda:1'), in_proj_covar=tensor([0.0140, 0.0135, 0.0145, 0.0129, 0.0113, 0.0145, 0.0146, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:43:36,026 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 06:43:36,449 INFO [finetune.py:976] (1/7) Epoch 6, batch 4950, loss[loss=0.1962, simple_loss=0.2721, pruned_loss=0.06013, over 4905.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2776, pruned_loss=0.07757, over 953921.25 frames. ], batch size: 37, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:44:20,475 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33622.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:44:22,686 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.562e+01 1.670e+02 2.087e+02 2.382e+02 5.310e+02, threshold=4.173e+02, percent-clipped=3.0 +2023-03-26 06:44:41,927 INFO [finetune.py:976] (1/7) Epoch 6, batch 5000, loss[loss=0.1529, simple_loss=0.2136, pruned_loss=0.0461, over 4718.00 frames. ], tot_loss[loss=0.2144, simple_loss=0.2758, pruned_loss=0.07653, over 954707.21 frames. ], batch size: 23, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:45:14,312 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=33670.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:45:19,067 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33677.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:45:23,166 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2972, 3.6868, 3.8891, 4.1129, 4.0441, 3.7821, 4.3787, 1.3435], + device='cuda:1'), covar=tensor([0.0742, 0.0802, 0.0887, 0.0940, 0.1134, 0.1489, 0.0653, 0.5502], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0241, 0.0272, 0.0292, 0.0332, 0.0282, 0.0300, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:45:26,109 INFO [finetune.py:976] (1/7) Epoch 6, batch 5050, loss[loss=0.1744, simple_loss=0.2385, pruned_loss=0.05517, over 4752.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2743, pruned_loss=0.07668, over 954564.75 frames. ], batch size: 27, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:45:46,736 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7795, 1.6674, 1.5721, 1.8805, 2.3109, 1.8827, 1.5040, 1.4554], + device='cuda:1'), covar=tensor([0.2471, 0.2411, 0.2139, 0.1832, 0.2084, 0.1333, 0.2901, 0.2164], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0211, 0.0204, 0.0187, 0.0239, 0.0177, 0.0214, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:45:50,301 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.151e+02 1.621e+02 1.877e+02 2.380e+02 3.773e+02, threshold=3.754e+02, percent-clipped=0.0 +2023-03-26 06:45:50,413 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33725.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:45:56,423 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 06:45:58,840 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33738.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 06:45:59,299 INFO [finetune.py:976] (1/7) Epoch 6, batch 5100, loss[loss=0.1658, simple_loss=0.2368, pruned_loss=0.04738, over 4809.00 frames. ], tot_loss[loss=0.2097, simple_loss=0.2697, pruned_loss=0.07485, over 953578.71 frames. ], batch size: 25, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:46:00,802 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 06:46:22,562 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=33773.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:46:31,691 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.65 vs. limit=5.0 +2023-03-26 06:46:32,755 INFO [finetune.py:976] (1/7) Epoch 6, batch 5150, loss[loss=0.2493, simple_loss=0.3106, pruned_loss=0.09402, over 4827.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2708, pruned_loss=0.07586, over 953171.40 frames. ], batch size: 47, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:46:48,264 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33811.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:46:51,324 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.53 vs. limit=5.0 +2023-03-26 06:46:57,549 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.284e+02 1.768e+02 2.107e+02 2.614e+02 3.782e+02, threshold=4.214e+02, percent-clipped=1.0 +2023-03-26 06:46:59,466 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33828.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:47:02,567 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6217, 1.5203, 1.3797, 1.3385, 1.7053, 1.4072, 1.7801, 1.6469], + device='cuda:1'), covar=tensor([0.1597, 0.2814, 0.3679, 0.3077, 0.2984, 0.1973, 0.3334, 0.2162], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0191, 0.0235, 0.0255, 0.0232, 0.0191, 0.0212, 0.0192], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:47:06,634 INFO [finetune.py:976] (1/7) Epoch 6, batch 5200, loss[loss=0.2476, simple_loss=0.2981, pruned_loss=0.09851, over 4903.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2728, pruned_loss=0.07638, over 953894.39 frames. ], batch size: 36, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:47:19,509 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=33859.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:47:44,978 INFO [finetune.py:976] (1/7) Epoch 6, batch 5250, loss[loss=0.2241, simple_loss=0.292, pruned_loss=0.07812, over 4927.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2741, pruned_loss=0.07593, over 953334.07 frames. ], batch size: 33, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:48:10,027 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.099e+02 1.728e+02 2.112e+02 2.559e+02 4.196e+02, threshold=4.224e+02, percent-clipped=0.0 +2023-03-26 06:48:15,724 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4630, 3.3530, 3.1763, 1.3721, 3.5264, 2.5687, 0.6908, 2.2742], + device='cuda:1'), covar=tensor([0.2469, 0.1796, 0.1741, 0.3352, 0.1077, 0.1017, 0.4195, 0.1498], + device='cuda:1'), in_proj_covar=tensor([0.0155, 0.0172, 0.0163, 0.0128, 0.0156, 0.0122, 0.0146, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 06:48:18,714 INFO [finetune.py:976] (1/7) Epoch 6, batch 5300, loss[loss=0.2259, simple_loss=0.2879, pruned_loss=0.08199, over 4909.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2769, pruned_loss=0.07763, over 953726.13 frames. ], batch size: 37, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:48:18,808 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33939.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:48:53,985 INFO [finetune.py:976] (1/7) Epoch 6, batch 5350, loss[loss=0.1928, simple_loss=0.2545, pruned_loss=0.06552, over 4772.00 frames. ], tot_loss[loss=0.2154, simple_loss=0.2765, pruned_loss=0.07709, over 954492.81 frames. ], batch size: 26, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:49:07,553 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34000.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:49:37,157 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-26 06:49:40,310 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.730e+02 2.013e+02 2.437e+02 5.230e+02, threshold=4.026e+02, percent-clipped=3.0 +2023-03-26 06:49:50,500 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34033.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 06:49:59,342 INFO [finetune.py:976] (1/7) Epoch 6, batch 5400, loss[loss=0.2328, simple_loss=0.2827, pruned_loss=0.09148, over 4845.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2728, pruned_loss=0.07562, over 953242.49 frames. ], batch size: 47, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:50:02,485 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34044.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:50:09,249 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8088, 1.7673, 1.6434, 1.9206, 2.3845, 1.9112, 1.5229, 1.4928], + device='cuda:1'), covar=tensor([0.2062, 0.2050, 0.1803, 0.1733, 0.1755, 0.1182, 0.2551, 0.1838], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0211, 0.0204, 0.0187, 0.0239, 0.0177, 0.0214, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:50:32,529 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6314, 0.6325, 1.5677, 1.4188, 1.3506, 1.3331, 1.3324, 1.4124], + device='cuda:1'), covar=tensor([0.4619, 0.5600, 0.4978, 0.5335, 0.5941, 0.4880, 0.6192, 0.4699], + device='cuda:1'), in_proj_covar=tensor([0.0230, 0.0244, 0.0253, 0.0256, 0.0241, 0.0219, 0.0272, 0.0225], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:50:51,575 INFO [finetune.py:976] (1/7) Epoch 6, batch 5450, loss[loss=0.2298, simple_loss=0.2508, pruned_loss=0.1044, over 3726.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2692, pruned_loss=0.07453, over 952067.11 frames. ], batch size: 15, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:51:04,656 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34105.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:51:17,059 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.064e+02 1.588e+02 1.777e+02 2.163e+02 4.002e+02, threshold=3.553e+02, percent-clipped=0.0 +2023-03-26 06:51:19,936 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34128.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:51:27,296 INFO [finetune.py:976] (1/7) Epoch 6, batch 5500, loss[loss=0.3123, simple_loss=0.336, pruned_loss=0.1443, over 4749.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2669, pruned_loss=0.07443, over 949972.24 frames. ], batch size: 54, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:51:31,026 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34145.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:51:36,925 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0228, 2.1814, 1.8464, 1.6082, 2.3052, 2.4282, 2.1075, 1.9099], + device='cuda:1'), covar=tensor([0.0372, 0.0321, 0.0543, 0.0390, 0.0306, 0.0539, 0.0400, 0.0410], + device='cuda:1'), in_proj_covar=tensor([0.0087, 0.0111, 0.0138, 0.0116, 0.0104, 0.0099, 0.0090, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.7933e-05, 8.6961e-05, 1.1027e-04, 9.1623e-05, 8.1992e-05, 7.3770e-05, + 6.8746e-05, 8.4399e-05], device='cuda:1') +2023-03-26 06:51:50,702 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=34176.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:51:59,801 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3577, 3.7735, 3.9277, 4.1993, 4.1118, 3.9316, 4.5005, 1.3438], + device='cuda:1'), covar=tensor([0.0804, 0.0893, 0.0731, 0.1018, 0.1138, 0.1476, 0.0573, 0.5438], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0241, 0.0273, 0.0293, 0.0333, 0.0284, 0.0302, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:52:00,938 INFO [finetune.py:976] (1/7) Epoch 6, batch 5550, loss[loss=0.2461, simple_loss=0.3147, pruned_loss=0.08877, over 4819.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.2683, pruned_loss=0.0752, over 947932.96 frames. ], batch size: 38, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:52:15,689 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34206.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:52:37,389 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.199e+02 1.703e+02 1.979e+02 2.376e+02 4.570e+02, threshold=3.959e+02, percent-clipped=2.0 +2023-03-26 06:52:55,410 INFO [finetune.py:976] (1/7) Epoch 6, batch 5600, loss[loss=0.2231, simple_loss=0.2887, pruned_loss=0.07871, over 4812.00 frames. ], tot_loss[loss=0.2123, simple_loss=0.2723, pruned_loss=0.07619, over 948120.07 frames. ], batch size: 40, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:53:54,498 INFO [finetune.py:976] (1/7) Epoch 6, batch 5650, loss[loss=0.1803, simple_loss=0.2468, pruned_loss=0.05685, over 4761.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2754, pruned_loss=0.07736, over 947425.13 frames. ], batch size: 28, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:53:58,426 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34295.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:54:09,121 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.7410, 4.1094, 4.3385, 4.6211, 4.4883, 4.1799, 4.8259, 1.8624], + device='cuda:1'), covar=tensor([0.0639, 0.0803, 0.0696, 0.0701, 0.0977, 0.1339, 0.0504, 0.4726], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0242, 0.0273, 0.0293, 0.0333, 0.0284, 0.0302, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 06:54:35,324 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.805e+01 1.675e+02 2.029e+02 2.481e+02 4.265e+02, threshold=4.057e+02, percent-clipped=3.0 +2023-03-26 06:54:40,132 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34333.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:54:48,432 INFO [finetune.py:976] (1/7) Epoch 6, batch 5700, loss[loss=0.2242, simple_loss=0.2565, pruned_loss=0.09593, over 4368.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2694, pruned_loss=0.07572, over 928416.42 frames. ], batch size: 19, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:54:58,678 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9022, 1.7205, 1.7426, 1.8406, 1.3526, 3.2543, 1.5844, 2.0556], + device='cuda:1'), covar=tensor([0.2738, 0.2129, 0.1797, 0.2084, 0.1650, 0.0225, 0.2186, 0.1064], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0118, 0.0122, 0.0117, 0.0098, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 06:54:59,940 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 06:55:38,388 INFO [finetune.py:976] (1/7) Epoch 7, batch 0, loss[loss=0.2296, simple_loss=0.2838, pruned_loss=0.08768, over 4813.00 frames. ], tot_loss[loss=0.2296, simple_loss=0.2838, pruned_loss=0.08768, over 4813.00 frames. ], batch size: 33, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:55:38,389 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 06:55:55,927 INFO [finetune.py:1010] (1/7) Epoch 7, validation: loss=0.165, simple_loss=0.2365, pruned_loss=0.04677, over 2265189.00 frames. +2023-03-26 06:55:55,928 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 06:56:14,684 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=34381.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:56:36,607 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34400.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:56:59,207 INFO [finetune.py:976] (1/7) Epoch 7, batch 50, loss[loss=0.2099, simple_loss=0.2706, pruned_loss=0.07464, over 4827.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2785, pruned_loss=0.0805, over 215012.79 frames. ], batch size: 30, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:56:59,309 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0551, 1.3412, 1.1736, 1.3133, 1.3683, 2.4553, 1.2668, 1.4281], + device='cuda:1'), covar=tensor([0.1085, 0.1817, 0.1168, 0.1029, 0.1718, 0.0366, 0.1476, 0.1723], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0081, 0.0076, 0.0079, 0.0092, 0.0083, 0.0084, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 06:57:09,543 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.105e+02 1.600e+02 2.022e+02 2.565e+02 5.766e+02, threshold=4.045e+02, percent-clipped=4.0 +2023-03-26 06:57:19,403 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 06:58:05,579 INFO [finetune.py:976] (1/7) Epoch 7, batch 100, loss[loss=0.1991, simple_loss=0.2651, pruned_loss=0.06659, over 4746.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2711, pruned_loss=0.07512, over 379965.88 frames. ], batch size: 26, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:58:29,314 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6393, 1.6268, 1.8632, 1.9611, 1.5772, 3.5149, 1.4517, 1.7064], + device='cuda:1'), covar=tensor([0.0959, 0.1743, 0.1093, 0.0922, 0.1553, 0.0240, 0.1421, 0.1630], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0080, 0.0076, 0.0079, 0.0092, 0.0083, 0.0084, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 06:58:45,250 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34501.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 06:59:06,468 INFO [finetune.py:976] (1/7) Epoch 7, batch 150, loss[loss=0.1863, simple_loss=0.2483, pruned_loss=0.06218, over 4821.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2658, pruned_loss=0.07267, over 507210.26 frames. ], batch size: 38, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 06:59:07,691 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8051, 1.9216, 1.5538, 1.4042, 2.0145, 2.1939, 1.9179, 1.7659], + device='cuda:1'), covar=tensor([0.0383, 0.0351, 0.0586, 0.0392, 0.0310, 0.0503, 0.0312, 0.0439], + device='cuda:1'), in_proj_covar=tensor([0.0088, 0.0111, 0.0139, 0.0116, 0.0104, 0.0100, 0.0091, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.8405e-05, 8.7596e-05, 1.1116e-04, 9.1828e-05, 8.2120e-05, 7.4255e-05, + 6.8972e-05, 8.4730e-05], device='cuda:1') +2023-03-26 06:59:17,659 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.007e+02 1.586e+02 1.902e+02 2.328e+02 6.438e+02, threshold=3.804e+02, percent-clipped=3.0 +2023-03-26 06:59:21,510 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9706, 1.8181, 1.4885, 1.7066, 1.6506, 1.6161, 1.6592, 2.4835], + device='cuda:1'), covar=tensor([0.6040, 0.6369, 0.4942, 0.5648, 0.5457, 0.3626, 0.6012, 0.2215], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0258, 0.0220, 0.0282, 0.0240, 0.0204, 0.0245, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:00:10,777 INFO [finetune.py:976] (1/7) Epoch 7, batch 200, loss[loss=0.2122, simple_loss=0.272, pruned_loss=0.07622, over 4873.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2657, pruned_loss=0.07336, over 607413.44 frames. ], batch size: 31, lr: 3.89e-03, grad_scale: 32.0 +2023-03-26 07:00:42,250 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34593.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:00:43,469 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34595.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:01:13,422 INFO [finetune.py:976] (1/7) Epoch 7, batch 250, loss[loss=0.1689, simple_loss=0.2377, pruned_loss=0.05002, over 4762.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2712, pruned_loss=0.07532, over 684456.70 frames. ], batch size: 27, lr: 3.88e-03, grad_scale: 32.0 +2023-03-26 07:01:22,724 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34622.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:01:24,450 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.284e+02 1.737e+02 2.023e+02 2.550e+02 3.958e+02, threshold=4.047e+02, percent-clipped=1.0 +2023-03-26 07:01:25,174 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34626.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:01:45,023 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=34643.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:01:57,385 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34654.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:02:13,097 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 07:02:14,154 INFO [finetune.py:976] (1/7) Epoch 7, batch 300, loss[loss=0.1697, simple_loss=0.2382, pruned_loss=0.05064, over 4774.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.2733, pruned_loss=0.07623, over 742065.63 frames. ], batch size: 28, lr: 3.88e-03, grad_scale: 32.0 +2023-03-26 07:02:34,476 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34683.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 07:02:36,957 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34687.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:02:55,189 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34700.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:03:06,626 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34710.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 07:03:15,678 INFO [finetune.py:976] (1/7) Epoch 7, batch 350, loss[loss=0.247, simple_loss=0.2977, pruned_loss=0.09816, over 4762.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2754, pruned_loss=0.0766, over 788994.68 frames. ], batch size: 28, lr: 3.88e-03, grad_scale: 32.0 +2023-03-26 07:03:27,161 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.146e+02 1.664e+02 2.043e+02 2.496e+02 5.690e+02, threshold=4.087e+02, percent-clipped=3.0 +2023-03-26 07:03:54,736 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=34748.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:04:16,438 INFO [finetune.py:976] (1/7) Epoch 7, batch 400, loss[loss=0.2822, simple_loss=0.3292, pruned_loss=0.1176, over 4931.00 frames. ], tot_loss[loss=0.2147, simple_loss=0.2762, pruned_loss=0.07657, over 825901.35 frames. ], batch size: 42, lr: 3.88e-03, grad_scale: 32.0 +2023-03-26 07:04:24,035 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34771.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 07:04:36,396 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2401, 2.1076, 1.7137, 2.1735, 2.0972, 1.9038, 1.9790, 3.0107], + device='cuda:1'), covar=tensor([0.6195, 0.7196, 0.5177, 0.6844, 0.6275, 0.3939, 0.6900, 0.2220], + device='cuda:1'), in_proj_covar=tensor([0.0284, 0.0259, 0.0221, 0.0284, 0.0241, 0.0205, 0.0245, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:04:47,121 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4174, 2.2047, 1.8771, 2.2711, 2.2700, 1.9991, 2.7206, 2.3921], + device='cuda:1'), covar=tensor([0.1577, 0.3358, 0.3756, 0.3554, 0.3017, 0.1946, 0.4354, 0.2214], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0192, 0.0236, 0.0256, 0.0234, 0.0192, 0.0213, 0.0192], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:04:55,077 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34801.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:04:55,247 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 07:05:14,336 INFO [finetune.py:976] (1/7) Epoch 7, batch 450, loss[loss=0.196, simple_loss=0.2576, pruned_loss=0.06715, over 4818.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2748, pruned_loss=0.07599, over 854244.86 frames. ], batch size: 38, lr: 3.88e-03, grad_scale: 32.0 +2023-03-26 07:05:25,982 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.133e+01 1.634e+02 1.827e+02 2.211e+02 3.915e+02, threshold=3.654e+02, percent-clipped=0.0 +2023-03-26 07:05:41,216 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0903, 2.7227, 2.4980, 1.4062, 2.6113, 2.2574, 1.9866, 2.2555], + device='cuda:1'), covar=tensor([0.0956, 0.0900, 0.1742, 0.2184, 0.1851, 0.1903, 0.1937, 0.1258], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0203, 0.0202, 0.0190, 0.0218, 0.0208, 0.0221, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:05:50,913 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=34849.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:06:11,594 INFO [finetune.py:976] (1/7) Epoch 7, batch 500, loss[loss=0.2206, simple_loss=0.2777, pruned_loss=0.08172, over 4858.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2725, pruned_loss=0.07518, over 877742.53 frames. ], batch size: 31, lr: 3.88e-03, grad_scale: 32.0 +2023-03-26 07:07:15,210 INFO [finetune.py:976] (1/7) Epoch 7, batch 550, loss[loss=0.2225, simple_loss=0.2979, pruned_loss=0.07355, over 4801.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2688, pruned_loss=0.07386, over 897256.10 frames. ], batch size: 45, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:07:26,484 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.074e+02 1.634e+02 2.013e+02 2.383e+02 4.182e+02, threshold=4.026e+02, percent-clipped=3.0 +2023-03-26 07:07:49,977 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1305, 1.9988, 1.7468, 2.1172, 2.1444, 1.8838, 2.5023, 2.1144], + device='cuda:1'), covar=tensor([0.1605, 0.2960, 0.3669, 0.2933, 0.2916, 0.1856, 0.3444, 0.2177], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0191, 0.0235, 0.0254, 0.0232, 0.0191, 0.0212, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:07:53,592 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34949.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:08:13,240 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6435, 1.5121, 1.9961, 1.3023, 1.7121, 1.7719, 1.4913, 2.0369], + device='cuda:1'), covar=tensor([0.1442, 0.2352, 0.1295, 0.1958, 0.1043, 0.1571, 0.3018, 0.0929], + device='cuda:1'), in_proj_covar=tensor([0.0205, 0.0206, 0.0199, 0.0197, 0.0183, 0.0222, 0.0219, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:08:14,352 INFO [finetune.py:976] (1/7) Epoch 7, batch 600, loss[loss=0.1798, simple_loss=0.2513, pruned_loss=0.0542, over 4898.00 frames. ], tot_loss[loss=0.2093, simple_loss=0.2698, pruned_loss=0.07446, over 910202.87 frames. ], batch size: 32, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:08:31,408 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34978.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 07:08:34,783 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34982.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:08:41,758 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4444, 3.7960, 4.0020, 4.2758, 4.1069, 3.9218, 4.5010, 1.3570], + device='cuda:1'), covar=tensor([0.0726, 0.0950, 0.0721, 0.0885, 0.1230, 0.1387, 0.0620, 0.5488], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0242, 0.0273, 0.0293, 0.0334, 0.0283, 0.0303, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:09:16,190 INFO [finetune.py:976] (1/7) Epoch 7, batch 650, loss[loss=0.2475, simple_loss=0.3157, pruned_loss=0.08967, over 4819.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2728, pruned_loss=0.07514, over 921417.07 frames. ], batch size: 40, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:09:23,911 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7204, 1.5564, 1.4314, 1.3156, 1.7896, 1.4955, 1.7565, 1.7072], + device='cuda:1'), covar=tensor([0.1591, 0.2715, 0.3558, 0.2890, 0.2877, 0.1917, 0.3319, 0.2084], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0191, 0.0235, 0.0254, 0.0233, 0.0191, 0.0212, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:09:27,411 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.303e+02 1.723e+02 2.031e+02 2.472e+02 3.902e+02, threshold=4.061e+02, percent-clipped=0.0 +2023-03-26 07:10:05,473 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.17 vs. limit=5.0 +2023-03-26 07:10:17,350 INFO [finetune.py:976] (1/7) Epoch 7, batch 700, loss[loss=0.2431, simple_loss=0.2977, pruned_loss=0.09424, over 4924.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2739, pruned_loss=0.0752, over 928805.20 frames. ], batch size: 42, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:10:17,430 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35066.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 07:10:46,480 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35092.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:11:11,733 INFO [finetune.py:976] (1/7) Epoch 7, batch 750, loss[loss=0.2151, simple_loss=0.2875, pruned_loss=0.07131, over 4831.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2744, pruned_loss=0.07525, over 934533.67 frames. ], batch size: 49, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:11:23,577 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.067e+02 1.641e+02 2.027e+02 2.429e+02 3.682e+02, threshold=4.054e+02, percent-clipped=0.0 +2023-03-26 07:11:31,180 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5163, 3.3016, 3.1304, 1.4773, 3.4591, 2.5511, 0.7458, 2.2760], + device='cuda:1'), covar=tensor([0.2663, 0.1768, 0.1777, 0.3451, 0.1097, 0.1050, 0.4538, 0.1522], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0170, 0.0161, 0.0127, 0.0154, 0.0121, 0.0144, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 07:11:31,844 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7108, 1.5001, 1.5454, 1.5731, 1.1392, 3.3677, 1.2893, 1.8748], + device='cuda:1'), covar=tensor([0.3198, 0.2568, 0.2075, 0.2334, 0.1966, 0.0211, 0.2564, 0.1315], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0118, 0.0122, 0.0117, 0.0098, 0.0101, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 07:12:01,757 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35153.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:12:14,609 INFO [finetune.py:976] (1/7) Epoch 7, batch 800, loss[loss=0.2265, simple_loss=0.287, pruned_loss=0.08297, over 4866.00 frames. ], tot_loss[loss=0.2123, simple_loss=0.2743, pruned_loss=0.07517, over 938641.26 frames. ], batch size: 31, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:13:17,535 INFO [finetune.py:976] (1/7) Epoch 7, batch 850, loss[loss=0.1927, simple_loss=0.2601, pruned_loss=0.06259, over 4702.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2723, pruned_loss=0.07482, over 943134.09 frames. ], batch size: 23, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:13:23,117 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5411, 1.5059, 1.6858, 1.7516, 1.6472, 3.1642, 1.3060, 1.6351], + device='cuda:1'), covar=tensor([0.0909, 0.1623, 0.1234, 0.0957, 0.1433, 0.0258, 0.1366, 0.1511], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0081, 0.0077, 0.0079, 0.0092, 0.0083, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 07:13:27,728 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.064e+02 1.645e+02 1.948e+02 2.227e+02 3.525e+02, threshold=3.897e+02, percent-clipped=0.0 +2023-03-26 07:13:42,087 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.60 vs. limit=2.0 +2023-03-26 07:13:55,180 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35249.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:14:14,655 INFO [finetune.py:976] (1/7) Epoch 7, batch 900, loss[loss=0.2072, simple_loss=0.2712, pruned_loss=0.07164, over 4841.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2692, pruned_loss=0.0733, over 945620.55 frames. ], batch size: 49, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:14:25,402 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5401, 1.5101, 1.7373, 1.7487, 1.6041, 3.4835, 1.3482, 1.5843], + device='cuda:1'), covar=tensor([0.1046, 0.1693, 0.1018, 0.1035, 0.1591, 0.0210, 0.1434, 0.1686], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0081, 0.0077, 0.0079, 0.0092, 0.0083, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 07:14:32,323 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35278.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 07:14:35,289 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35282.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:14:55,278 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=35297.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:15:14,831 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.6315, 3.9771, 4.1866, 4.4096, 4.3585, 4.1293, 4.6854, 1.4774], + device='cuda:1'), covar=tensor([0.0674, 0.0843, 0.0792, 0.0961, 0.1035, 0.1350, 0.0605, 0.5182], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0240, 0.0272, 0.0291, 0.0331, 0.0281, 0.0301, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:15:16,597 INFO [finetune.py:976] (1/7) Epoch 7, batch 950, loss[loss=0.175, simple_loss=0.2401, pruned_loss=0.05495, over 4775.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2673, pruned_loss=0.07272, over 947618.63 frames. ], batch size: 28, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:15:31,370 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.193e+02 1.515e+02 1.811e+02 2.306e+02 3.628e+02, threshold=3.621e+02, percent-clipped=0.0 +2023-03-26 07:15:31,437 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=35326.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:15:33,887 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=35330.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:15:50,923 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35345.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 07:16:18,532 INFO [finetune.py:976] (1/7) Epoch 7, batch 1000, loss[loss=0.2229, simple_loss=0.2854, pruned_loss=0.08018, over 4894.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.2695, pruned_loss=0.07364, over 948756.24 frames. ], batch size: 35, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:16:18,630 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35366.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 07:16:27,460 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35374.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:16:41,726 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-03-26 07:16:51,636 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1146, 1.8678, 1.8064, 2.0169, 2.5365, 2.0504, 1.9074, 1.5784], + device='cuda:1'), covar=tensor([0.2293, 0.2250, 0.1977, 0.1742, 0.2030, 0.1219, 0.2424, 0.1876], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0209, 0.0204, 0.0186, 0.0238, 0.0177, 0.0213, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:17:08,112 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35406.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 07:17:17,932 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=35414.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 07:17:18,773 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0 +2023-03-26 07:17:19,085 INFO [finetune.py:976] (1/7) Epoch 7, batch 1050, loss[loss=0.2361, simple_loss=0.2997, pruned_loss=0.08631, over 4860.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2731, pruned_loss=0.07474, over 948687.80 frames. ], batch size: 31, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:17:28,543 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 07:17:30,142 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.274e+02 1.696e+02 1.924e+02 2.371e+02 5.787e+02, threshold=3.848e+02, percent-clipped=4.0 +2023-03-26 07:17:41,387 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35435.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:17:59,820 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35448.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:18:03,468 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0145, 1.9315, 1.6130, 2.0021, 1.7911, 1.8228, 1.7295, 2.6801], + device='cuda:1'), covar=tensor([0.6210, 0.8305, 0.5191, 0.7178, 0.6780, 0.3900, 0.7576, 0.2473], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0258, 0.0220, 0.0282, 0.0240, 0.0205, 0.0244, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:18:20,379 INFO [finetune.py:976] (1/7) Epoch 7, batch 1100, loss[loss=0.234, simple_loss=0.2931, pruned_loss=0.0875, over 4884.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2734, pruned_loss=0.07482, over 949023.60 frames. ], batch size: 35, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:18:30,499 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4663, 1.5616, 1.4846, 0.8319, 1.5364, 1.7140, 1.7368, 1.3656], + device='cuda:1'), covar=tensor([0.0750, 0.0527, 0.0482, 0.0530, 0.0383, 0.0533, 0.0275, 0.0584], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0158, 0.0121, 0.0138, 0.0133, 0.0125, 0.0147, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.7368e-05, 1.1651e-04, 8.7325e-05, 1.0030e-04, 9.5365e-05, 9.2613e-05, + 1.0831e-04, 1.0778e-04], device='cuda:1') +2023-03-26 07:18:32,373 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9301, 1.7348, 1.4900, 1.7092, 1.6069, 1.6202, 1.6329, 2.3982], + device='cuda:1'), covar=tensor([0.5719, 0.6563, 0.4796, 0.5811, 0.5844, 0.3363, 0.6072, 0.2280], + device='cuda:1'), in_proj_covar=tensor([0.0282, 0.0258, 0.0220, 0.0281, 0.0239, 0.0204, 0.0244, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:18:51,245 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1116, 1.9615, 1.6913, 1.9750, 1.8284, 1.7563, 1.8663, 2.6456], + device='cuda:1'), covar=tensor([0.5872, 0.6312, 0.4830, 0.5886, 0.5618, 0.3552, 0.5860, 0.2333], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0258, 0.0220, 0.0282, 0.0240, 0.0205, 0.0244, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:19:22,492 INFO [finetune.py:976] (1/7) Epoch 7, batch 1150, loss[loss=0.1436, simple_loss=0.2244, pruned_loss=0.03143, over 4738.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2739, pruned_loss=0.07515, over 950284.85 frames. ], batch size: 27, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:19:33,759 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.207e+02 1.764e+02 2.068e+02 2.432e+02 4.937e+02, threshold=4.137e+02, percent-clipped=2.0 +2023-03-26 07:19:41,304 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35530.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:20:11,144 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5256, 1.4969, 1.9976, 1.2616, 1.6503, 1.8205, 1.4402, 1.9840], + device='cuda:1'), covar=tensor([0.1585, 0.2361, 0.1394, 0.1927, 0.1018, 0.1481, 0.3141, 0.0962], + device='cuda:1'), in_proj_covar=tensor([0.0204, 0.0205, 0.0199, 0.0197, 0.0182, 0.0221, 0.0218, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:20:24,974 INFO [finetune.py:976] (1/7) Epoch 7, batch 1200, loss[loss=0.1712, simple_loss=0.2531, pruned_loss=0.04467, over 4791.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2722, pruned_loss=0.0748, over 949878.65 frames. ], batch size: 51, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:20:31,931 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 07:20:51,597 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35591.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:21:20,130 INFO [finetune.py:976] (1/7) Epoch 7, batch 1250, loss[loss=0.2705, simple_loss=0.3091, pruned_loss=0.116, over 4193.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.27, pruned_loss=0.07478, over 950257.04 frames. ], batch size: 65, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:21:31,425 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.265e+02 1.688e+02 2.018e+02 2.672e+02 1.298e+03, threshold=4.035e+02, percent-clipped=4.0 +2023-03-26 07:21:38,449 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.56 vs. limit=5.0 +2023-03-26 07:22:18,864 INFO [finetune.py:976] (1/7) Epoch 7, batch 1300, loss[loss=0.2052, simple_loss=0.2673, pruned_loss=0.07159, over 4894.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.2662, pruned_loss=0.07255, over 952355.95 frames. ], batch size: 32, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:23:05,605 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35701.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 07:23:06,761 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9756, 1.7861, 1.7025, 1.8673, 1.3785, 4.4819, 1.6156, 2.4141], + device='cuda:1'), covar=tensor([0.3328, 0.2364, 0.2176, 0.2445, 0.1799, 0.0113, 0.2563, 0.1300], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0115, 0.0119, 0.0123, 0.0117, 0.0099, 0.0102, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 07:23:25,710 INFO [finetune.py:976] (1/7) Epoch 7, batch 1350, loss[loss=0.1813, simple_loss=0.2544, pruned_loss=0.05412, over 4926.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2672, pruned_loss=0.07299, over 954679.02 frames. ], batch size: 38, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:23:38,291 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.109e+01 1.659e+02 1.871e+02 2.249e+02 4.421e+02, threshold=3.743e+02, percent-clipped=1.0 +2023-03-26 07:23:40,797 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35730.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:24:06,301 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35748.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:24:25,443 INFO [finetune.py:976] (1/7) Epoch 7, batch 1400, loss[loss=0.2681, simple_loss=0.3251, pruned_loss=0.1056, over 4762.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2701, pruned_loss=0.07396, over 952580.96 frames. ], batch size: 54, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:24:33,478 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35771.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:24:43,640 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 07:24:44,708 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4284, 1.2581, 1.3058, 1.2905, 0.8720, 2.3144, 0.7660, 1.3171], + device='cuda:1'), covar=tensor([0.3479, 0.2530, 0.2259, 0.2440, 0.2031, 0.0345, 0.2765, 0.1441], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0114, 0.0118, 0.0122, 0.0117, 0.0099, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 07:25:01,856 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=35796.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:25:06,173 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7953, 1.6886, 1.6751, 1.7609, 1.4160, 3.8406, 1.6575, 2.4239], + device='cuda:1'), covar=tensor([0.3379, 0.2417, 0.2058, 0.2258, 0.1796, 0.0184, 0.2392, 0.1120], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0114, 0.0118, 0.0122, 0.0117, 0.0099, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 07:25:20,904 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35811.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:25:23,858 INFO [finetune.py:976] (1/7) Epoch 7, batch 1450, loss[loss=0.1809, simple_loss=0.2389, pruned_loss=0.06143, over 4414.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.2709, pruned_loss=0.07392, over 951561.66 frames. ], batch size: 19, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:25:33,652 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.204e+02 1.737e+02 2.011e+02 2.560e+02 4.083e+02, threshold=4.021e+02, percent-clipped=3.0 +2023-03-26 07:25:43,679 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35832.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:26:15,552 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9044, 2.2232, 2.0465, 1.5779, 2.2776, 2.2826, 2.2586, 1.8494], + device='cuda:1'), covar=tensor([0.0732, 0.0570, 0.0811, 0.0860, 0.0474, 0.0739, 0.0611, 0.1018], + device='cuda:1'), in_proj_covar=tensor([0.0140, 0.0134, 0.0144, 0.0127, 0.0114, 0.0145, 0.0146, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:26:24,785 INFO [finetune.py:976] (1/7) Epoch 7, batch 1500, loss[loss=0.2429, simple_loss=0.3061, pruned_loss=0.08982, over 4813.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.273, pruned_loss=0.07475, over 953031.95 frames. ], batch size: 47, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:26:32,635 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35872.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:26:47,872 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35886.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:27:11,196 INFO [finetune.py:976] (1/7) Epoch 7, batch 1550, loss[loss=0.1911, simple_loss=0.2536, pruned_loss=0.06435, over 4742.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2728, pruned_loss=0.0744, over 951161.86 frames. ], batch size: 26, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:27:17,686 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.940e+01 1.555e+02 1.902e+02 2.352e+02 4.828e+02, threshold=3.804e+02, percent-clipped=1.0 +2023-03-26 07:27:29,641 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4312, 3.8544, 4.0152, 4.2185, 4.1347, 3.9177, 4.5172, 1.3790], + device='cuda:1'), covar=tensor([0.0740, 0.0723, 0.0729, 0.1059, 0.1246, 0.1329, 0.0630, 0.5155], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0239, 0.0269, 0.0289, 0.0328, 0.0280, 0.0299, 0.0291], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:27:45,066 INFO [finetune.py:976] (1/7) Epoch 7, batch 1600, loss[loss=0.2317, simple_loss=0.2928, pruned_loss=0.08534, over 4898.00 frames. ], tot_loss[loss=0.21, simple_loss=0.271, pruned_loss=0.07447, over 951055.47 frames. ], batch size: 32, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:27:45,883 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.77 vs. limit=2.0 +2023-03-26 07:28:25,875 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36001.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 07:28:35,456 INFO [finetune.py:976] (1/7) Epoch 7, batch 1650, loss[loss=0.2221, simple_loss=0.2705, pruned_loss=0.08688, over 4931.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2676, pruned_loss=0.07292, over 952842.90 frames. ], batch size: 38, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:28:41,919 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.242e+02 1.569e+02 1.867e+02 2.342e+02 3.778e+02, threshold=3.734e+02, percent-clipped=0.0 +2023-03-26 07:28:50,685 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36030.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:29:07,541 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-03-26 07:29:09,233 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=36049.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 07:29:25,210 INFO [finetune.py:976] (1/7) Epoch 7, batch 1700, loss[loss=0.2273, simple_loss=0.2708, pruned_loss=0.09191, over 4820.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2667, pruned_loss=0.07334, over 951990.47 frames. ], batch size: 25, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:29:39,786 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=36078.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:30:15,625 INFO [finetune.py:976] (1/7) Epoch 7, batch 1750, loss[loss=0.2329, simple_loss=0.3004, pruned_loss=0.08269, over 4752.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2704, pruned_loss=0.07506, over 952829.32 frames. ], batch size: 59, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:30:27,792 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.647e+02 1.960e+02 2.452e+02 4.962e+02, threshold=3.920e+02, percent-clipped=3.0 +2023-03-26 07:30:28,487 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36127.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:31:18,546 INFO [finetune.py:976] (1/7) Epoch 7, batch 1800, loss[loss=0.2364, simple_loss=0.3007, pruned_loss=0.08607, over 4758.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2721, pruned_loss=0.0753, over 953525.37 frames. ], batch size: 59, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:31:19,225 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36167.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:31:38,143 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36181.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:31:47,171 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36186.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:32:21,146 INFO [finetune.py:976] (1/7) Epoch 7, batch 1850, loss[loss=0.2241, simple_loss=0.2826, pruned_loss=0.08284, over 4840.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2738, pruned_loss=0.0761, over 952148.56 frames. ], batch size: 49, lr: 3.88e-03, grad_scale: 16.0 +2023-03-26 07:32:33,455 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.247e+02 1.736e+02 2.131e+02 2.651e+02 6.216e+02, threshold=4.263e+02, percent-clipped=3.0 +2023-03-26 07:32:40,443 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=36234.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:32:50,348 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36242.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:32:51,046 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 07:33:21,417 INFO [finetune.py:976] (1/7) Epoch 7, batch 1900, loss[loss=0.271, simple_loss=0.3164, pruned_loss=0.1128, over 4262.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2746, pruned_loss=0.07581, over 952222.42 frames. ], batch size: 65, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:33:22,783 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6188, 1.8580, 1.3595, 1.7251, 2.0386, 1.9175, 1.6696, 1.4428], + device='cuda:1'), covar=tensor([0.0327, 0.0261, 0.0579, 0.0306, 0.0220, 0.0482, 0.0332, 0.0455], + device='cuda:1'), in_proj_covar=tensor([0.0088, 0.0111, 0.0138, 0.0116, 0.0104, 0.0099, 0.0091, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.8894e-05, 8.6979e-05, 1.1031e-04, 9.1087e-05, 8.1760e-05, 7.3793e-05, + 6.8968e-05, 8.4831e-05], device='cuda:1') +2023-03-26 07:34:25,473 INFO [finetune.py:976] (1/7) Epoch 7, batch 1950, loss[loss=0.1654, simple_loss=0.2204, pruned_loss=0.05518, over 4724.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2719, pruned_loss=0.07442, over 950970.80 frames. ], batch size: 23, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:34:36,926 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.969e+01 1.685e+02 2.051e+02 2.475e+02 4.640e+02, threshold=4.103e+02, percent-clipped=3.0 +2023-03-26 07:35:18,617 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-26 07:35:28,761 INFO [finetune.py:976] (1/7) Epoch 7, batch 2000, loss[loss=0.2023, simple_loss=0.2606, pruned_loss=0.07198, over 4850.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2697, pruned_loss=0.07375, over 952105.64 frames. ], batch size: 44, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:36:30,439 INFO [finetune.py:976] (1/7) Epoch 7, batch 2050, loss[loss=0.1918, simple_loss=0.2481, pruned_loss=0.06781, over 4938.00 frames. ], tot_loss[loss=0.207, simple_loss=0.2675, pruned_loss=0.07328, over 953790.80 frames. ], batch size: 38, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:36:43,670 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 7.910e+01 1.532e+02 1.893e+02 2.218e+02 7.941e+02, threshold=3.786e+02, percent-clipped=2.0 +2023-03-26 07:36:44,403 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36427.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:36:45,606 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0826, 3.5435, 3.7793, 3.7454, 3.6333, 3.4632, 4.1377, 1.3618], + device='cuda:1'), covar=tensor([0.1399, 0.1783, 0.1535, 0.2253, 0.2258, 0.2465, 0.1400, 0.7113], + device='cuda:1'), in_proj_covar=tensor([0.0345, 0.0241, 0.0272, 0.0289, 0.0329, 0.0280, 0.0300, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:36:45,641 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0880, 2.0274, 2.0700, 1.3613, 2.2400, 2.2664, 2.1261, 1.7668], + device='cuda:1'), covar=tensor([0.0575, 0.0639, 0.0696, 0.0941, 0.0507, 0.0660, 0.0562, 0.1059], + device='cuda:1'), in_proj_covar=tensor([0.0140, 0.0135, 0.0145, 0.0128, 0.0115, 0.0146, 0.0147, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:37:34,745 INFO [finetune.py:976] (1/7) Epoch 7, batch 2100, loss[loss=0.1646, simple_loss=0.2455, pruned_loss=0.04188, over 4907.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2665, pruned_loss=0.07254, over 954473.84 frames. ], batch size: 35, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:37:35,446 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36467.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:37:39,002 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4818, 3.8295, 4.0499, 4.3106, 4.2209, 3.9803, 4.5868, 1.5003], + device='cuda:1'), covar=tensor([0.0727, 0.0889, 0.0869, 0.0990, 0.1126, 0.1425, 0.0653, 0.5241], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0243, 0.0273, 0.0291, 0.0330, 0.0282, 0.0301, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:37:45,796 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=36475.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:38:36,450 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=36515.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:38:36,994 INFO [finetune.py:976] (1/7) Epoch 7, batch 2150, loss[loss=0.2711, simple_loss=0.3323, pruned_loss=0.1049, over 4817.00 frames. ], tot_loss[loss=0.2095, simple_loss=0.2707, pruned_loss=0.07413, over 953193.39 frames. ], batch size: 40, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:38:48,002 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.129e+02 1.787e+02 2.211e+02 2.590e+02 5.595e+02, threshold=4.423e+02, percent-clipped=4.0 +2023-03-26 07:39:00,122 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36537.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:39:35,926 INFO [finetune.py:976] (1/7) Epoch 7, batch 2200, loss[loss=0.2275, simple_loss=0.2894, pruned_loss=0.08283, over 4757.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2721, pruned_loss=0.07465, over 952935.68 frames. ], batch size: 28, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:40:25,399 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36605.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:40:27,741 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36608.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:40:30,241 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36612.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:40:38,177 INFO [finetune.py:976] (1/7) Epoch 7, batch 2250, loss[loss=0.2188, simple_loss=0.2783, pruned_loss=0.07964, over 4814.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2737, pruned_loss=0.07558, over 952896.19 frames. ], batch size: 39, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:40:49,911 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.198e+02 1.735e+02 1.950e+02 2.446e+02 5.153e+02, threshold=3.899e+02, percent-clipped=1.0 +2023-03-26 07:41:19,706 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8328, 3.3836, 3.0261, 1.8233, 3.1870, 2.8203, 2.6091, 2.8552], + device='cuda:1'), covar=tensor([0.0837, 0.0753, 0.1389, 0.2234, 0.1418, 0.1586, 0.1953, 0.1097], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0203, 0.0200, 0.0189, 0.0219, 0.0208, 0.0222, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:41:29,029 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 07:41:41,215 INFO [finetune.py:976] (1/7) Epoch 7, batch 2300, loss[loss=0.2006, simple_loss=0.2705, pruned_loss=0.06538, over 4784.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2748, pruned_loss=0.07569, over 950884.02 frames. ], batch size: 51, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:41:41,345 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36666.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:41:43,620 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36669.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:41:51,486 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36673.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:42:18,139 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36700.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 07:42:36,646 INFO [finetune.py:976] (1/7) Epoch 7, batch 2350, loss[loss=0.2136, simple_loss=0.2693, pruned_loss=0.07892, over 4751.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2728, pruned_loss=0.07512, over 951274.88 frames. ], batch size: 26, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:42:37,368 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3627, 1.3865, 1.4788, 1.6166, 1.4619, 2.9988, 1.2887, 1.5163], + device='cuda:1'), covar=tensor([0.1040, 0.1828, 0.1122, 0.1011, 0.1707, 0.0302, 0.1535, 0.1692], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0076, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 07:42:49,192 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.103e+02 1.509e+02 1.866e+02 2.321e+02 4.735e+02, threshold=3.732e+02, percent-clipped=2.0 +2023-03-26 07:42:56,046 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8510, 2.5613, 2.4279, 1.3226, 2.6036, 2.1725, 1.9205, 2.3225], + device='cuda:1'), covar=tensor([0.1059, 0.0884, 0.1463, 0.2266, 0.1664, 0.2188, 0.2038, 0.1138], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0202, 0.0200, 0.0188, 0.0217, 0.0207, 0.0221, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:43:27,593 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-26 07:43:29,350 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36761.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 07:43:37,712 INFO [finetune.py:976] (1/7) Epoch 7, batch 2400, loss[loss=0.1754, simple_loss=0.2418, pruned_loss=0.05445, over 4815.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2697, pruned_loss=0.07387, over 949748.24 frames. ], batch size: 41, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:43:45,009 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8421, 1.7398, 1.8829, 1.1650, 2.0034, 1.8991, 1.8188, 1.6154], + device='cuda:1'), covar=tensor([0.0568, 0.0741, 0.0694, 0.0966, 0.0562, 0.0785, 0.0669, 0.1094], + device='cuda:1'), in_proj_covar=tensor([0.0138, 0.0133, 0.0143, 0.0126, 0.0114, 0.0144, 0.0146, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:44:28,031 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36805.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 07:44:40,195 INFO [finetune.py:976] (1/7) Epoch 7, batch 2450, loss[loss=0.1695, simple_loss=0.2307, pruned_loss=0.05419, over 4805.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2678, pruned_loss=0.07299, over 950858.92 frames. ], batch size: 25, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:44:51,697 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.798e+02 2.140e+02 2.594e+02 4.660e+02, threshold=4.281e+02, percent-clipped=3.0 +2023-03-26 07:45:10,239 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36837.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:45:49,131 INFO [finetune.py:976] (1/7) Epoch 7, batch 2500, loss[loss=0.207, simple_loss=0.2847, pruned_loss=0.06462, over 4796.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2685, pruned_loss=0.07302, over 950604.64 frames. ], batch size: 45, lr: 3.87e-03, grad_scale: 16.0 +2023-03-26 07:45:49,276 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36866.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 07:46:12,540 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=36885.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:46:44,097 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36911.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:46:52,299 INFO [finetune.py:976] (1/7) Epoch 7, batch 2550, loss[loss=0.2302, simple_loss=0.2937, pruned_loss=0.08333, over 4747.00 frames. ], tot_loss[loss=0.209, simple_loss=0.271, pruned_loss=0.07348, over 952462.77 frames. ], batch size: 27, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:46:53,634 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.7111, 1.8062, 1.9655, 1.1618, 1.9931, 2.2325, 2.0712, 1.6189], + device='cuda:1'), covar=tensor([0.0815, 0.0559, 0.0403, 0.0536, 0.0397, 0.0456, 0.0311, 0.0606], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0156, 0.0120, 0.0137, 0.0131, 0.0124, 0.0145, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.6552e-05, 1.1494e-04, 8.6493e-05, 9.9556e-05, 9.4365e-05, 9.1059e-05, + 1.0708e-04, 1.0693e-04], device='cuda:1') +2023-03-26 07:47:03,264 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.261e+02 1.620e+02 1.912e+02 2.307e+02 6.491e+02, threshold=3.825e+02, percent-clipped=1.0 +2023-03-26 07:47:37,569 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 07:47:45,968 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36958.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:47:47,751 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36961.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:47:55,199 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36964.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:47:56,361 INFO [finetune.py:976] (1/7) Epoch 7, batch 2600, loss[loss=0.2325, simple_loss=0.2809, pruned_loss=0.09206, over 4774.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2732, pruned_loss=0.07458, over 953399.81 frames. ], batch size: 27, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:47:57,665 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36968.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:48:05,153 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36972.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:48:06,988 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6185, 1.5966, 2.1873, 1.9383, 1.8167, 3.5395, 1.4883, 1.8599], + device='cuda:1'), covar=tensor([0.0944, 0.1535, 0.1363, 0.0956, 0.1417, 0.0290, 0.1414, 0.1492], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0076, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 07:48:41,465 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37004.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:48:53,701 INFO [finetune.py:976] (1/7) Epoch 7, batch 2650, loss[loss=0.2275, simple_loss=0.2938, pruned_loss=0.08064, over 4788.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2744, pruned_loss=0.07493, over 949766.86 frames. ], batch size: 29, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:48:56,099 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37019.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:49:05,451 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.083e+02 1.627e+02 1.954e+02 2.393e+02 3.704e+02, threshold=3.907e+02, percent-clipped=0.0 +2023-03-26 07:49:41,688 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37056.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 07:49:47,738 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37065.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:49:48,223 INFO [finetune.py:976] (1/7) Epoch 7, batch 2700, loss[loss=0.2062, simple_loss=0.2569, pruned_loss=0.07776, over 4826.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2725, pruned_loss=0.07415, over 949902.88 frames. ], batch size: 30, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:49:55,709 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.08 vs. limit=5.0 +2023-03-26 07:49:57,392 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7354, 1.7125, 1.4088, 1.5373, 1.9341, 1.9650, 1.7475, 1.4229], + device='cuda:1'), covar=tensor([0.0455, 0.0303, 0.0558, 0.0314, 0.0238, 0.0501, 0.0311, 0.0478], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0112, 0.0140, 0.0117, 0.0105, 0.0101, 0.0091, 0.0110], + device='cuda:1'), out_proj_covar=tensor([6.9611e-05, 8.7689e-05, 1.1184e-04, 9.1938e-05, 8.2783e-05, 7.4827e-05, + 6.9049e-05, 8.5675e-05], device='cuda:1') +2023-03-26 07:50:22,105 INFO [finetune.py:976] (1/7) Epoch 7, batch 2750, loss[loss=0.1879, simple_loss=0.2498, pruned_loss=0.06303, over 4929.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2691, pruned_loss=0.07291, over 950444.67 frames. ], batch size: 33, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:50:28,698 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.127e+02 1.629e+02 1.991e+02 2.307e+02 4.303e+02, threshold=3.983e+02, percent-clipped=1.0 +2023-03-26 07:50:38,480 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.72 vs. limit=5.0 +2023-03-26 07:50:58,190 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37161.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 07:51:03,500 INFO [finetune.py:976] (1/7) Epoch 7, batch 2800, loss[loss=0.1616, simple_loss=0.2291, pruned_loss=0.04703, over 4772.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2655, pruned_loss=0.07105, over 953685.19 frames. ], batch size: 28, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:52:09,300 INFO [finetune.py:976] (1/7) Epoch 7, batch 2850, loss[loss=0.178, simple_loss=0.2439, pruned_loss=0.05603, over 4933.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2642, pruned_loss=0.07068, over 955175.20 frames. ], batch size: 38, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:52:19,149 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4670, 1.5731, 1.7743, 0.8896, 1.5935, 1.8436, 1.8838, 1.5054], + device='cuda:1'), covar=tensor([0.0832, 0.0588, 0.0393, 0.0552, 0.0450, 0.0502, 0.0299, 0.0636], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0156, 0.0120, 0.0136, 0.0131, 0.0124, 0.0145, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.6173e-05, 1.1482e-04, 8.6826e-05, 9.9271e-05, 9.4123e-05, 9.0943e-05, + 1.0716e-04, 1.0681e-04], device='cuda:1') +2023-03-26 07:52:20,862 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.100e+02 1.582e+02 1.929e+02 2.327e+02 4.539e+02, threshold=3.857e+02, percent-clipped=3.0 +2023-03-26 07:52:49,840 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2883, 1.3418, 1.3614, 1.5537, 1.4604, 2.9385, 1.1643, 1.4102], + device='cuda:1'), covar=tensor([0.1251, 0.2336, 0.1365, 0.1181, 0.1876, 0.0353, 0.1984, 0.2322], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0076, 0.0080, 0.0092, 0.0083, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 07:53:00,971 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37261.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:53:02,798 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37264.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:53:03,915 INFO [finetune.py:976] (1/7) Epoch 7, batch 2900, loss[loss=0.2403, simple_loss=0.3119, pruned_loss=0.08434, over 4921.00 frames. ], tot_loss[loss=0.206, simple_loss=0.2673, pruned_loss=0.0724, over 955003.48 frames. ], batch size: 42, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:53:09,379 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37267.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:53:09,980 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37268.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:53:10,000 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37268.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:53:11,098 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0903, 1.7792, 2.6090, 4.2228, 2.9457, 2.7621, 0.9089, 3.2026], + device='cuda:1'), covar=tensor([0.1872, 0.1676, 0.1448, 0.0423, 0.0740, 0.1561, 0.2302, 0.0568], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0118, 0.0135, 0.0166, 0.0102, 0.0141, 0.0129, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 07:53:41,654 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9791, 1.9320, 1.9744, 1.2312, 2.0984, 2.0755, 1.9816, 1.6458], + device='cuda:1'), covar=tensor([0.0624, 0.0645, 0.0753, 0.0985, 0.0544, 0.0680, 0.0637, 0.1149], + device='cuda:1'), in_proj_covar=tensor([0.0138, 0.0134, 0.0145, 0.0127, 0.0114, 0.0145, 0.0147, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:54:04,930 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=37309.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:54:06,753 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=37312.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:54:07,993 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37314.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:54:13,844 INFO [finetune.py:976] (1/7) Epoch 7, batch 2950, loss[loss=0.2187, simple_loss=0.2877, pruned_loss=0.07487, over 4788.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2705, pruned_loss=0.07269, over 955968.78 frames. ], batch size: 51, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:54:13,904 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=37316.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:54:25,448 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.193e+02 1.702e+02 2.045e+02 2.514e+02 5.908e+02, threshold=4.090e+02, percent-clipped=3.0 +2023-03-26 07:54:27,400 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37329.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:55:00,467 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37356.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 07:55:05,684 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37360.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:55:08,106 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37364.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:55:09,239 INFO [finetune.py:976] (1/7) Epoch 7, batch 3000, loss[loss=0.1609, simple_loss=0.2049, pruned_loss=0.05845, over 4286.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.272, pruned_loss=0.07408, over 954939.85 frames. ], batch size: 18, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:55:09,239 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 07:55:11,241 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1995, 1.3965, 1.4074, 0.6951, 1.3125, 1.5930, 1.6781, 1.3091], + device='cuda:1'), covar=tensor([0.0918, 0.0560, 0.0571, 0.0576, 0.0497, 0.0568, 0.0313, 0.0723], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0156, 0.0120, 0.0137, 0.0131, 0.0124, 0.0145, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.6111e-05, 1.1481e-04, 8.6984e-05, 9.9453e-05, 9.3983e-05, 9.1066e-05, + 1.0714e-04, 1.0661e-04], device='cuda:1') +2023-03-26 07:55:12,115 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9473, 1.7666, 1.6012, 1.7554, 2.0177, 1.6959, 2.1700, 1.9485], + device='cuda:1'), covar=tensor([0.1882, 0.3158, 0.3994, 0.3245, 0.3294, 0.2099, 0.4379, 0.2407], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0191, 0.0235, 0.0255, 0.0234, 0.0192, 0.0212, 0.0192], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:55:18,386 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9652, 1.7702, 1.6051, 1.6264, 1.6882, 1.6786, 1.6615, 2.3837], + device='cuda:1'), covar=tensor([0.6229, 0.6406, 0.4805, 0.5855, 0.5481, 0.3691, 0.5805, 0.2457], + device='cuda:1'), in_proj_covar=tensor([0.0284, 0.0258, 0.0220, 0.0281, 0.0240, 0.0204, 0.0244, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:55:19,045 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2484, 2.0579, 1.4871, 0.6150, 1.7381, 1.8964, 1.7717, 1.9250], + device='cuda:1'), covar=tensor([0.0867, 0.0746, 0.1366, 0.2049, 0.1335, 0.2481, 0.2122, 0.0844], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0203, 0.0201, 0.0190, 0.0220, 0.0208, 0.0223, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:55:25,742 INFO [finetune.py:1010] (1/7) Epoch 7, validation: loss=0.161, simple_loss=0.2327, pruned_loss=0.04464, over 2265189.00 frames. +2023-03-26 07:55:25,742 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 07:55:33,745 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5109, 1.3404, 1.4563, 1.4077, 0.9334, 2.2157, 0.7803, 1.3105], + device='cuda:1'), covar=tensor([0.3478, 0.2567, 0.2182, 0.2504, 0.1894, 0.0384, 0.2640, 0.1420], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0118, 0.0122, 0.0116, 0.0098, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 07:55:35,740 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0 +2023-03-26 07:55:53,945 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4307, 1.5496, 1.8737, 1.7170, 1.5765, 3.3254, 1.3445, 1.6877], + device='cuda:1'), covar=tensor([0.0997, 0.1725, 0.1237, 0.1041, 0.1555, 0.0241, 0.1412, 0.1663], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0076, 0.0079, 0.0092, 0.0083, 0.0084, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 07:55:54,503 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=37404.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 07:56:02,292 INFO [finetune.py:976] (1/7) Epoch 7, batch 3050, loss[loss=0.1916, simple_loss=0.2625, pruned_loss=0.06038, over 4731.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2738, pruned_loss=0.07469, over 956118.04 frames. ], batch size: 59, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:56:11,536 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37425.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 07:56:12,025 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.173e+02 1.581e+02 1.871e+02 2.387e+02 4.591e+02, threshold=3.742e+02, percent-clipped=1.0 +2023-03-26 07:56:12,174 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3573, 2.4435, 2.0581, 1.7530, 2.6600, 2.6672, 2.4784, 2.1349], + device='cuda:1'), covar=tensor([0.0364, 0.0321, 0.0453, 0.0388, 0.0300, 0.0475, 0.0284, 0.0354], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0111, 0.0138, 0.0116, 0.0104, 0.0100, 0.0090, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.9179e-05, 8.6950e-05, 1.1061e-04, 9.1180e-05, 8.1920e-05, 7.4164e-05, + 6.8597e-05, 8.5004e-05], device='cuda:1') +2023-03-26 07:56:34,654 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.71 vs. limit=2.0 +2023-03-26 07:56:48,966 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37461.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 07:56:57,059 INFO [finetune.py:976] (1/7) Epoch 7, batch 3100, loss[loss=0.2001, simple_loss=0.2592, pruned_loss=0.07052, over 4832.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.2727, pruned_loss=0.07411, over 956481.90 frames. ], batch size: 25, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:57:52,634 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=37509.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 07:58:01,949 INFO [finetune.py:976] (1/7) Epoch 7, batch 3150, loss[loss=0.2185, simple_loss=0.2669, pruned_loss=0.08507, over 4810.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2707, pruned_loss=0.07357, over 956330.96 frames. ], batch size: 41, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:58:12,058 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 07:58:13,088 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.225e+02 1.704e+02 2.041e+02 2.515e+02 5.799e+02, threshold=4.081e+02, percent-clipped=3.0 +2023-03-26 07:58:52,867 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-03-26 07:58:56,499 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9388, 4.3167, 4.5820, 4.7644, 4.6873, 4.4511, 5.0498, 1.4934], + device='cuda:1'), covar=tensor([0.0613, 0.0758, 0.0630, 0.0825, 0.0974, 0.1272, 0.0505, 0.5259], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0241, 0.0273, 0.0292, 0.0330, 0.0282, 0.0302, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 07:59:05,775 INFO [finetune.py:976] (1/7) Epoch 7, batch 3200, loss[loss=0.2263, simple_loss=0.2798, pruned_loss=0.08641, over 4875.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2666, pruned_loss=0.07161, over 957751.97 frames. ], batch size: 31, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 07:59:06,468 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37567.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:00:08,888 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37614.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:00:09,462 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=37615.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:00:14,948 INFO [finetune.py:976] (1/7) Epoch 7, batch 3250, loss[loss=0.1773, simple_loss=0.2437, pruned_loss=0.05549, over 4825.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.266, pruned_loss=0.07162, over 956793.30 frames. ], batch size: 33, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 08:00:19,906 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37624.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:00:26,116 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.216e+02 1.664e+02 1.918e+02 2.274e+02 4.430e+02, threshold=3.836e+02, percent-clipped=1.0 +2023-03-26 08:00:51,961 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 08:01:09,565 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37660.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:01:10,705 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=37662.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:01:18,270 INFO [finetune.py:976] (1/7) Epoch 7, batch 3300, loss[loss=0.2373, simple_loss=0.2925, pruned_loss=0.09105, over 4910.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2697, pruned_loss=0.07311, over 956755.97 frames. ], batch size: 43, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 08:01:23,258 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1235, 0.9355, 0.9531, 0.4712, 0.8078, 1.1095, 1.1739, 0.9838], + device='cuda:1'), covar=tensor([0.0925, 0.0624, 0.0498, 0.0547, 0.0564, 0.0661, 0.0412, 0.0696], + device='cuda:1'), in_proj_covar=tensor([0.0128, 0.0154, 0.0119, 0.0136, 0.0130, 0.0123, 0.0144, 0.0144], + device='cuda:1'), out_proj_covar=tensor([9.5602e-05, 1.1347e-04, 8.6378e-05, 9.8803e-05, 9.3204e-05, 9.0614e-05, + 1.0593e-04, 1.0602e-04], device='cuda:1') +2023-03-26 08:01:40,130 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9087, 4.5964, 4.3194, 2.3387, 4.7962, 3.5183, 0.9416, 3.1033], + device='cuda:1'), covar=tensor([0.2566, 0.2095, 0.1393, 0.3278, 0.0802, 0.0915, 0.4784, 0.1562], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0172, 0.0161, 0.0128, 0.0153, 0.0122, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 08:02:01,363 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1860, 2.1858, 1.6449, 2.2696, 2.2104, 1.7902, 2.7056, 2.2216], + device='cuda:1'), covar=tensor([0.1569, 0.2969, 0.3654, 0.3267, 0.2781, 0.1853, 0.3646, 0.2054], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0189, 0.0233, 0.0253, 0.0232, 0.0191, 0.0210, 0.0191], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:02:12,454 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=37708.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:02:22,969 INFO [finetune.py:976] (1/7) Epoch 7, batch 3350, loss[loss=0.2522, simple_loss=0.2952, pruned_loss=0.1046, over 4838.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2711, pruned_loss=0.07331, over 955612.51 frames. ], batch size: 49, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 08:02:25,465 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37720.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:02:26,298 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.66 vs. limit=5.0 +2023-03-26 08:02:34,111 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.166e+02 1.767e+02 2.019e+02 2.457e+02 5.992e+02, threshold=4.038e+02, percent-clipped=4.0 +2023-03-26 08:02:56,976 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8826, 1.6490, 1.7478, 1.9458, 1.6459, 4.6808, 1.8878, 2.6472], + device='cuda:1'), covar=tensor([0.3362, 0.2517, 0.2106, 0.2289, 0.1732, 0.0085, 0.2353, 0.1147], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0115, 0.0119, 0.0123, 0.0117, 0.0098, 0.0102, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 08:03:28,096 INFO [finetune.py:976] (1/7) Epoch 7, batch 3400, loss[loss=0.1952, simple_loss=0.2722, pruned_loss=0.05912, over 4810.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2717, pruned_loss=0.07316, over 955410.37 frames. ], batch size: 25, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 08:04:32,124 INFO [finetune.py:976] (1/7) Epoch 7, batch 3450, loss[loss=0.1922, simple_loss=0.2583, pruned_loss=0.06301, over 4867.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2723, pruned_loss=0.0729, over 955247.87 frames. ], batch size: 31, lr: 3.87e-03, grad_scale: 32.0 +2023-03-26 08:04:43,342 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.130e+02 1.715e+02 1.991e+02 2.496e+02 6.747e+02, threshold=3.982e+02, percent-clipped=3.0 +2023-03-26 08:05:36,356 INFO [finetune.py:976] (1/7) Epoch 7, batch 3500, loss[loss=0.1559, simple_loss=0.2202, pruned_loss=0.04579, over 4720.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2698, pruned_loss=0.07277, over 953685.72 frames. ], batch size: 54, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:06:07,313 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7900, 1.5942, 1.5103, 1.3554, 1.8684, 1.5411, 1.7839, 1.7623], + device='cuda:1'), covar=tensor([0.1666, 0.2806, 0.3748, 0.3097, 0.2889, 0.1945, 0.3104, 0.2356], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0190, 0.0235, 0.0255, 0.0234, 0.0193, 0.0211, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:06:41,165 INFO [finetune.py:976] (1/7) Epoch 7, batch 3550, loss[loss=0.2133, simple_loss=0.272, pruned_loss=0.0773, over 4695.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2659, pruned_loss=0.07079, over 955218.71 frames. ], batch size: 23, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:06:51,853 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37924.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:06:58,533 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.177e+02 1.559e+02 1.846e+02 2.185e+02 4.242e+02, threshold=3.693e+02, percent-clipped=1.0 +2023-03-26 08:06:59,875 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5967, 1.5027, 1.3181, 1.4056, 1.7809, 1.7464, 1.5386, 1.3481], + device='cuda:1'), covar=tensor([0.0308, 0.0259, 0.0611, 0.0279, 0.0212, 0.0440, 0.0293, 0.0377], + device='cuda:1'), in_proj_covar=tensor([0.0087, 0.0110, 0.0137, 0.0115, 0.0103, 0.0099, 0.0090, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.8337e-05, 8.6527e-05, 1.0991e-04, 9.0176e-05, 8.1432e-05, 7.3726e-05, + 6.8284e-05, 8.4336e-05], device='cuda:1') +2023-03-26 08:07:09,426 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7368, 1.5886, 1.9528, 2.0704, 1.7843, 3.7103, 1.6051, 1.8323], + device='cuda:1'), covar=tensor([0.0961, 0.1779, 0.1086, 0.0965, 0.1561, 0.0229, 0.1413, 0.1723], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0076, 0.0079, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 08:07:52,117 INFO [finetune.py:976] (1/7) Epoch 7, batch 3600, loss[loss=0.2082, simple_loss=0.2791, pruned_loss=0.06865, over 4932.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.264, pruned_loss=0.07027, over 956576.14 frames. ], batch size: 33, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:07:56,328 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=37972.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:08:06,502 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37979.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:08:37,180 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-26 08:08:59,068 INFO [finetune.py:976] (1/7) Epoch 7, batch 3650, loss[loss=0.279, simple_loss=0.3455, pruned_loss=0.1063, over 4841.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2668, pruned_loss=0.07194, over 956892.58 frames. ], batch size: 49, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:09:07,222 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=38020.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:09:10,798 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.158e+02 1.699e+02 2.068e+02 2.418e+02 4.148e+02, threshold=4.136e+02, percent-clipped=4.0 +2023-03-26 08:09:27,925 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=38040.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:09:46,796 INFO [finetune.py:976] (1/7) Epoch 7, batch 3700, loss[loss=0.2615, simple_loss=0.3184, pruned_loss=0.1023, over 4133.00 frames. ], tot_loss[loss=0.2095, simple_loss=0.271, pruned_loss=0.07407, over 954382.67 frames. ], batch size: 65, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:09:48,562 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=38068.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:10:19,922 INFO [finetune.py:976] (1/7) Epoch 7, batch 3750, loss[loss=0.1515, simple_loss=0.2227, pruned_loss=0.04012, over 4782.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2731, pruned_loss=0.07491, over 956970.44 frames. ], batch size: 26, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:10:26,924 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.737e+01 1.627e+02 1.982e+02 2.503e+02 4.763e+02, threshold=3.965e+02, percent-clipped=1.0 +2023-03-26 08:10:57,173 INFO [finetune.py:976] (1/7) Epoch 7, batch 3800, loss[loss=0.143, simple_loss=0.2062, pruned_loss=0.03989, over 4734.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2739, pruned_loss=0.07518, over 956272.29 frames. ], batch size: 23, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:10:57,301 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2934, 2.1435, 1.7829, 0.9040, 1.9949, 1.7720, 1.6274, 1.9497], + device='cuda:1'), covar=tensor([0.0856, 0.0905, 0.1700, 0.2131, 0.1367, 0.2426, 0.2214, 0.1033], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0200, 0.0199, 0.0188, 0.0217, 0.0205, 0.0221, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:11:05,225 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6948, 3.6365, 3.5448, 1.6610, 3.7611, 2.6934, 0.8198, 2.5560], + device='cuda:1'), covar=tensor([0.2783, 0.1715, 0.1614, 0.3427, 0.1006, 0.1099, 0.4511, 0.1451], + device='cuda:1'), in_proj_covar=tensor([0.0155, 0.0175, 0.0164, 0.0131, 0.0155, 0.0124, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 08:11:30,369 INFO [finetune.py:976] (1/7) Epoch 7, batch 3850, loss[loss=0.2348, simple_loss=0.2852, pruned_loss=0.09223, over 4915.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2723, pruned_loss=0.07447, over 955120.49 frames. ], batch size: 46, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:11:43,043 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.105e+02 1.610e+02 2.090e+02 2.406e+02 4.877e+02, threshold=4.181e+02, percent-clipped=2.0 +2023-03-26 08:12:25,294 INFO [finetune.py:976] (1/7) Epoch 7, batch 3900, loss[loss=0.2002, simple_loss=0.2637, pruned_loss=0.06829, over 4749.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2692, pruned_loss=0.07336, over 954756.94 frames. ], batch size: 27, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:13:28,060 INFO [finetune.py:976] (1/7) Epoch 7, batch 3950, loss[loss=0.2349, simple_loss=0.2849, pruned_loss=0.09241, over 4927.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2658, pruned_loss=0.07166, over 956344.65 frames. ], batch size: 38, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:13:45,513 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.011e+02 1.693e+02 1.988e+02 2.374e+02 4.679e+02, threshold=3.976e+02, percent-clipped=1.0 +2023-03-26 08:13:56,364 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=38335.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:14:21,683 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=38360.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:14:25,157 INFO [finetune.py:976] (1/7) Epoch 7, batch 4000, loss[loss=0.1897, simple_loss=0.2604, pruned_loss=0.05952, over 4747.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2652, pruned_loss=0.07172, over 957103.29 frames. ], batch size: 54, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:14:46,788 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.54 vs. limit=5.0 +2023-03-26 08:14:54,125 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8504, 1.7816, 1.8841, 1.1879, 1.9816, 1.9784, 1.9378, 1.5505], + device='cuda:1'), covar=tensor([0.0589, 0.0606, 0.0594, 0.0909, 0.0489, 0.0617, 0.0557, 0.1105], + device='cuda:1'), in_proj_covar=tensor([0.0137, 0.0133, 0.0143, 0.0126, 0.0113, 0.0146, 0.0146, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:15:01,732 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.81 vs. limit=5.0 +2023-03-26 08:15:29,792 INFO [finetune.py:976] (1/7) Epoch 7, batch 4050, loss[loss=0.1584, simple_loss=0.2313, pruned_loss=0.0427, over 4779.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.27, pruned_loss=0.07364, over 957568.07 frames. ], batch size: 26, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:15:33,988 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=38421.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:15:42,068 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.248e+02 1.778e+02 2.129e+02 2.625e+02 5.238e+02, threshold=4.258e+02, percent-clipped=5.0 +2023-03-26 08:16:32,427 INFO [finetune.py:976] (1/7) Epoch 7, batch 4100, loss[loss=0.2268, simple_loss=0.2917, pruned_loss=0.08097, over 4888.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2739, pruned_loss=0.07502, over 958731.04 frames. ], batch size: 43, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:17:12,863 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-26 08:17:31,591 INFO [finetune.py:976] (1/7) Epoch 7, batch 4150, loss[loss=0.2392, simple_loss=0.2957, pruned_loss=0.09132, over 4822.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2754, pruned_loss=0.07556, over 958601.09 frames. ], batch size: 47, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:17:43,680 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.188e+02 1.723e+02 2.145e+02 2.598e+02 6.605e+02, threshold=4.291e+02, percent-clipped=2.0 +2023-03-26 08:18:25,178 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9857, 1.7481, 1.6075, 1.7943, 1.6822, 1.7092, 1.6523, 2.4263], + device='cuda:1'), covar=tensor([0.5047, 0.5957, 0.4273, 0.5485, 0.5462, 0.3337, 0.5828, 0.2078], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0221, 0.0282, 0.0241, 0.0206, 0.0246, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:18:34,215 INFO [finetune.py:976] (1/7) Epoch 7, batch 4200, loss[loss=0.1588, simple_loss=0.2129, pruned_loss=0.05233, over 3979.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2756, pruned_loss=0.07562, over 956048.56 frames. ], batch size: 17, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:19:33,020 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8672, 1.7557, 1.7138, 1.9977, 2.2282, 2.0475, 1.4194, 1.5121], + device='cuda:1'), covar=tensor([0.2589, 0.2421, 0.2112, 0.1924, 0.2055, 0.1233, 0.3003, 0.2196], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0209, 0.0204, 0.0186, 0.0238, 0.0177, 0.0214, 0.0192], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:19:34,111 INFO [finetune.py:976] (1/7) Epoch 7, batch 4250, loss[loss=0.2131, simple_loss=0.2778, pruned_loss=0.07425, over 4791.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2719, pruned_loss=0.07418, over 954845.45 frames. ], batch size: 51, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:19:44,840 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.083e+02 1.606e+02 1.980e+02 2.259e+02 5.740e+02, threshold=3.960e+02, percent-clipped=2.0 +2023-03-26 08:19:55,309 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7527, 1.6033, 2.0450, 1.3215, 1.8471, 1.9071, 1.5190, 2.1205], + device='cuda:1'), covar=tensor([0.1454, 0.2164, 0.1417, 0.2085, 0.0915, 0.1538, 0.2786, 0.0946], + device='cuda:1'), in_proj_covar=tensor([0.0203, 0.0205, 0.0199, 0.0196, 0.0182, 0.0221, 0.0217, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:20:02,302 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=38635.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:20:31,963 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1294, 1.8815, 1.4239, 0.5467, 1.6793, 1.8076, 1.6384, 1.7339], + device='cuda:1'), covar=tensor([0.0926, 0.0902, 0.1439, 0.2265, 0.1340, 0.2222, 0.2404, 0.0918], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0199, 0.0199, 0.0187, 0.0217, 0.0204, 0.0221, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:20:38,695 INFO [finetune.py:976] (1/7) Epoch 7, batch 4300, loss[loss=0.2016, simple_loss=0.2607, pruned_loss=0.07128, over 4872.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2697, pruned_loss=0.07377, over 956735.76 frames. ], batch size: 34, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:20:40,648 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8105, 1.7228, 1.6110, 1.7877, 1.6689, 4.4943, 1.9263, 2.3516], + device='cuda:1'), covar=tensor([0.4255, 0.2982, 0.2426, 0.2907, 0.1755, 0.0163, 0.2255, 0.1262], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0119, 0.0122, 0.0117, 0.0098, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 08:20:59,380 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=38683.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:21:31,978 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9621, 1.2103, 0.8139, 1.8838, 2.3152, 1.7576, 1.4784, 1.8533], + device='cuda:1'), covar=tensor([0.1620, 0.2438, 0.2254, 0.1347, 0.2016, 0.2062, 0.1594, 0.2106], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0099, 0.0115, 0.0094, 0.0125, 0.0097, 0.0102, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 08:21:41,238 INFO [finetune.py:976] (1/7) Epoch 7, batch 4350, loss[loss=0.2478, simple_loss=0.2961, pruned_loss=0.09974, over 4918.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.266, pruned_loss=0.07209, over 957693.96 frames. ], batch size: 43, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:21:41,305 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=38716.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:21:52,343 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.223e+02 1.679e+02 1.871e+02 2.197e+02 5.866e+02, threshold=3.741e+02, percent-clipped=4.0 +2023-03-26 08:22:04,765 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.30 vs. limit=5.0 +2023-03-26 08:22:33,943 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 08:22:43,659 INFO [finetune.py:976] (1/7) Epoch 7, batch 4400, loss[loss=0.1971, simple_loss=0.271, pruned_loss=0.06164, over 4910.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2676, pruned_loss=0.07291, over 957646.80 frames. ], batch size: 37, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:23:14,690 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7183, 3.4177, 3.2519, 1.4464, 3.5513, 2.5840, 0.7755, 2.3448], + device='cuda:1'), covar=tensor([0.2361, 0.2669, 0.1789, 0.3973, 0.1291, 0.1207, 0.4847, 0.1939], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0174, 0.0163, 0.0130, 0.0155, 0.0123, 0.0146, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 08:23:21,189 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4730, 1.3661, 2.2353, 3.4748, 2.2314, 2.4354, 0.9400, 2.6969], + device='cuda:1'), covar=tensor([0.1986, 0.1677, 0.1374, 0.0488, 0.0926, 0.1470, 0.2089, 0.0574], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0134, 0.0165, 0.0102, 0.0140, 0.0127, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 08:23:33,714 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9864, 1.9061, 1.4556, 1.9476, 1.9825, 1.6635, 2.2734, 1.9666], + device='cuda:1'), covar=tensor([0.1625, 0.3011, 0.3907, 0.3257, 0.2739, 0.1977, 0.3878, 0.2221], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0190, 0.0234, 0.0253, 0.0233, 0.0192, 0.0211, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:23:42,201 INFO [finetune.py:976] (1/7) Epoch 7, batch 4450, loss[loss=0.2032, simple_loss=0.2682, pruned_loss=0.06909, over 4759.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2705, pruned_loss=0.0736, over 956697.61 frames. ], batch size: 27, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:23:51,912 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=38823.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:23:53,601 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.165e+02 1.712e+02 1.965e+02 2.330e+02 4.727e+02, threshold=3.929e+02, percent-clipped=4.0 +2023-03-26 08:24:44,783 INFO [finetune.py:976] (1/7) Epoch 7, batch 4500, loss[loss=0.2096, simple_loss=0.2826, pruned_loss=0.06831, over 4899.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.273, pruned_loss=0.07473, over 957215.94 frames. ], batch size: 37, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:25:06,333 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=38884.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:25:17,592 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6777, 3.7051, 3.5655, 1.8876, 3.7663, 2.8389, 0.6559, 2.5453], + device='cuda:1'), covar=tensor([0.2689, 0.1728, 0.1496, 0.3326, 0.0991, 0.0937, 0.4668, 0.1543], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0174, 0.0163, 0.0130, 0.0155, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 08:25:49,102 INFO [finetune.py:976] (1/7) Epoch 7, batch 4550, loss[loss=0.2316, simple_loss=0.3035, pruned_loss=0.07989, over 4770.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.2725, pruned_loss=0.07427, over 953867.26 frames. ], batch size: 29, lr: 3.86e-03, grad_scale: 64.0 +2023-03-26 08:25:59,505 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.041e+02 1.671e+02 2.000e+02 2.524e+02 3.434e+02, threshold=4.000e+02, percent-clipped=0.0 +2023-03-26 08:26:05,509 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4372, 1.4027, 1.4979, 1.7745, 1.5411, 3.1092, 1.2709, 1.5831], + device='cuda:1'), covar=tensor([0.0935, 0.1828, 0.1227, 0.0916, 0.1551, 0.0247, 0.1503, 0.1620], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0091, 0.0082, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 08:26:13,063 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5493, 1.3843, 1.4700, 1.5315, 0.9327, 3.0331, 1.0069, 1.6407], + device='cuda:1'), covar=tensor([0.3462, 0.2610, 0.2172, 0.2361, 0.2169, 0.0212, 0.2827, 0.1406], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0115, 0.0119, 0.0123, 0.0117, 0.0098, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 08:26:47,278 INFO [finetune.py:976] (1/7) Epoch 7, batch 4600, loss[loss=0.1801, simple_loss=0.2502, pruned_loss=0.05501, over 4892.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2715, pruned_loss=0.07309, over 955504.21 frames. ], batch size: 32, lr: 3.86e-03, grad_scale: 64.0 +2023-03-26 08:27:15,242 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.58 vs. limit=5.0 +2023-03-26 08:27:54,984 INFO [finetune.py:976] (1/7) Epoch 7, batch 4650, loss[loss=0.2072, simple_loss=0.2531, pruned_loss=0.08068, over 4249.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2696, pruned_loss=0.07291, over 956209.49 frames. ], batch size: 65, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:27:55,086 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39016.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:28:06,174 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.761e+01 1.505e+02 1.924e+02 2.345e+02 4.238e+02, threshold=3.847e+02, percent-clipped=2.0 +2023-03-26 08:28:50,939 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=39064.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:28:57,994 INFO [finetune.py:976] (1/7) Epoch 7, batch 4700, loss[loss=0.1877, simple_loss=0.2453, pruned_loss=0.06508, over 4830.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2658, pruned_loss=0.07167, over 954427.11 frames. ], batch size: 30, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:29:09,886 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0865, 3.5117, 3.7081, 3.9558, 3.7964, 3.5868, 4.1474, 1.2368], + device='cuda:1'), covar=tensor([0.0807, 0.0853, 0.0893, 0.0877, 0.1288, 0.1499, 0.0762, 0.5268], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0243, 0.0274, 0.0292, 0.0331, 0.0282, 0.0302, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:29:56,018 INFO [finetune.py:976] (1/7) Epoch 7, batch 4750, loss[loss=0.2157, simple_loss=0.2763, pruned_loss=0.0776, over 4905.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.264, pruned_loss=0.07115, over 955269.78 frames. ], batch size: 37, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:30:08,811 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.221e+02 1.609e+02 1.819e+02 2.206e+02 4.512e+02, threshold=3.638e+02, percent-clipped=2.0 +2023-03-26 08:30:17,945 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8287, 1.6203, 1.6342, 1.7977, 1.4118, 3.8239, 1.5447, 2.1226], + device='cuda:1'), covar=tensor([0.3289, 0.2444, 0.2088, 0.2304, 0.1711, 0.0154, 0.2460, 0.1294], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0119, 0.0122, 0.0116, 0.0098, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 08:30:38,547 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9833, 1.7377, 1.7292, 1.8555, 1.7111, 4.4774, 1.7938, 2.2738], + device='cuda:1'), covar=tensor([0.3322, 0.2378, 0.2030, 0.2136, 0.1543, 0.0137, 0.2305, 0.1287], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0114, 0.0118, 0.0122, 0.0116, 0.0098, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 08:30:58,729 INFO [finetune.py:976] (1/7) Epoch 7, batch 4800, loss[loss=0.2487, simple_loss=0.3139, pruned_loss=0.09174, over 4928.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2657, pruned_loss=0.07151, over 955370.35 frames. ], batch size: 38, lr: 3.86e-03, grad_scale: 32.0 +2023-03-26 08:31:12,881 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39179.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:31:22,013 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39185.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:31:40,475 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8177, 3.3492, 3.5120, 3.7555, 3.5682, 3.3321, 3.8763, 1.1716], + device='cuda:1'), covar=tensor([0.0887, 0.0873, 0.0900, 0.0892, 0.1340, 0.1672, 0.0845, 0.5369], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0242, 0.0274, 0.0292, 0.0329, 0.0280, 0.0301, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:31:50,656 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6553, 3.7017, 3.6825, 1.6431, 3.8410, 2.8511, 0.7536, 2.6519], + device='cuda:1'), covar=tensor([0.2968, 0.2021, 0.1445, 0.3536, 0.0952, 0.1032, 0.4522, 0.1459], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0174, 0.0163, 0.0130, 0.0155, 0.0123, 0.0148, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 08:31:54,686 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9970, 4.8401, 4.6083, 2.4867, 4.9645, 3.7145, 0.7881, 3.4656], + device='cuda:1'), covar=tensor([0.2495, 0.1808, 0.1225, 0.3344, 0.0674, 0.0915, 0.5007, 0.1259], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0174, 0.0163, 0.0130, 0.0155, 0.0123, 0.0148, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 08:31:57,082 INFO [finetune.py:976] (1/7) Epoch 7, batch 4850, loss[loss=0.1837, simple_loss=0.2441, pruned_loss=0.06161, over 4874.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2696, pruned_loss=0.07307, over 955963.19 frames. ], batch size: 31, lr: 3.86e-03, grad_scale: 16.0 +2023-03-26 08:32:00,255 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-26 08:32:06,049 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.092e+02 1.691e+02 2.004e+02 2.499e+02 4.240e+02, threshold=4.008e+02, percent-clipped=2.0 +2023-03-26 08:32:18,155 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39246.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:32:25,356 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1526, 2.0974, 1.6520, 2.1268, 2.2101, 1.7580, 2.4752, 2.1372], + device='cuda:1'), covar=tensor([0.1476, 0.2682, 0.3471, 0.3008, 0.2630, 0.1879, 0.3600, 0.2010], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0190, 0.0234, 0.0253, 0.0233, 0.0192, 0.0211, 0.0192], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:32:30,567 INFO [finetune.py:976] (1/7) Epoch 7, batch 4900, loss[loss=0.2385, simple_loss=0.3031, pruned_loss=0.0869, over 4816.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2728, pruned_loss=0.0749, over 956071.10 frames. ], batch size: 40, lr: 3.86e-03, grad_scale: 16.0 +2023-03-26 08:32:30,709 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1260, 2.2245, 2.4335, 0.9296, 2.7257, 2.7339, 2.3734, 2.1752], + device='cuda:1'), covar=tensor([0.1149, 0.0840, 0.0543, 0.0851, 0.0374, 0.1133, 0.0435, 0.0760], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0156, 0.0121, 0.0137, 0.0132, 0.0125, 0.0145, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.6270e-05, 1.1493e-04, 8.7302e-05, 9.9753e-05, 9.4466e-05, 9.1906e-05, + 1.0655e-04, 1.0734e-04], device='cuda:1') +2023-03-26 08:32:43,207 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39284.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:33:03,518 INFO [finetune.py:976] (1/7) Epoch 7, batch 4950, loss[loss=0.2053, simple_loss=0.2691, pruned_loss=0.07079, over 4858.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2741, pruned_loss=0.07492, over 957087.17 frames. ], batch size: 31, lr: 3.86e-03, grad_scale: 16.0 +2023-03-26 08:33:12,728 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.619e+02 1.980e+02 2.423e+02 3.796e+02, threshold=3.961e+02, percent-clipped=0.0 +2023-03-26 08:33:24,238 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39345.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:33:35,540 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4606, 1.2397, 1.2774, 1.3863, 1.5863, 1.5577, 1.3774, 1.2265], + device='cuda:1'), covar=tensor([0.0268, 0.0324, 0.0609, 0.0304, 0.0311, 0.0490, 0.0338, 0.0432], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0112, 0.0141, 0.0116, 0.0105, 0.0101, 0.0092, 0.0110], + device='cuda:1'), out_proj_covar=tensor([6.9666e-05, 8.8007e-05, 1.1253e-04, 9.1714e-05, 8.2279e-05, 7.5506e-05, + 6.9378e-05, 8.5814e-05], device='cuda:1') +2023-03-26 08:33:37,224 INFO [finetune.py:976] (1/7) Epoch 7, batch 5000, loss[loss=0.2323, simple_loss=0.2879, pruned_loss=0.08837, over 4713.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2729, pruned_loss=0.07438, over 957472.66 frames. ], batch size: 23, lr: 3.86e-03, grad_scale: 16.0 +2023-03-26 08:34:10,936 INFO [finetune.py:976] (1/7) Epoch 7, batch 5050, loss[loss=0.2149, simple_loss=0.2749, pruned_loss=0.07748, over 4898.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2701, pruned_loss=0.07365, over 958687.38 frames. ], batch size: 32, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:34:19,591 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.096e+02 1.623e+02 1.955e+02 2.404e+02 3.498e+02, threshold=3.910e+02, percent-clipped=0.0 +2023-03-26 08:34:52,720 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39463.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:34:54,406 INFO [finetune.py:976] (1/7) Epoch 7, batch 5100, loss[loss=0.1776, simple_loss=0.244, pruned_loss=0.05561, over 4828.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2671, pruned_loss=0.07234, over 958702.40 frames. ], batch size: 33, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:34:55,737 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8126, 0.9262, 1.6840, 1.5737, 1.4551, 1.4528, 1.4557, 1.5480], + device='cuda:1'), covar=tensor([0.3996, 0.5399, 0.4450, 0.4455, 0.5703, 0.4125, 0.5889, 0.4138], + device='cuda:1'), in_proj_covar=tensor([0.0230, 0.0242, 0.0255, 0.0254, 0.0245, 0.0220, 0.0273, 0.0226], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:35:05,361 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39479.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:35:32,600 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 08:35:33,722 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0470, 2.0530, 2.1157, 1.4155, 2.1531, 2.2764, 2.0985, 1.6853], + device='cuda:1'), covar=tensor([0.0564, 0.0594, 0.0627, 0.0857, 0.0573, 0.0550, 0.0569, 0.1110], + device='cuda:1'), in_proj_covar=tensor([0.0137, 0.0134, 0.0144, 0.0126, 0.0114, 0.0145, 0.0147, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:35:39,560 INFO [finetune.py:976] (1/7) Epoch 7, batch 5150, loss[loss=0.183, simple_loss=0.2472, pruned_loss=0.05935, over 4882.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.2673, pruned_loss=0.07287, over 957122.23 frames. ], batch size: 32, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:35:40,913 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5458, 1.3777, 1.4120, 1.4500, 0.9402, 2.9289, 1.0974, 1.5525], + device='cuda:1'), covar=tensor([0.3245, 0.2500, 0.2139, 0.2376, 0.2084, 0.0247, 0.2684, 0.1393], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0119, 0.0123, 0.0117, 0.0099, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 08:35:47,017 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39524.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:35:48,749 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=39527.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:35:49,814 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.405e+01 1.710e+02 2.016e+02 2.412e+02 5.054e+02, threshold=4.032e+02, percent-clipped=2.0 +2023-03-26 08:36:08,538 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39541.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:36:24,638 INFO [finetune.py:976] (1/7) Epoch 7, batch 5200, loss[loss=0.2151, simple_loss=0.2911, pruned_loss=0.06959, over 4745.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.2695, pruned_loss=0.07336, over 955839.64 frames. ], batch size: 54, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:36:33,235 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-26 08:36:59,639 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3210, 2.2310, 1.7740, 0.9102, 1.9609, 1.7613, 1.7384, 1.9731], + device='cuda:1'), covar=tensor([0.0953, 0.0867, 0.1747, 0.2230, 0.1480, 0.2375, 0.2172, 0.1135], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0200, 0.0201, 0.0187, 0.0217, 0.0206, 0.0222, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:37:07,820 INFO [finetune.py:976] (1/7) Epoch 7, batch 5250, loss[loss=0.219, simple_loss=0.2792, pruned_loss=0.07936, over 4812.00 frames. ], tot_loss[loss=0.2093, simple_loss=0.2712, pruned_loss=0.07366, over 955865.26 frames. ], batch size: 33, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:37:11,763 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 08:37:15,021 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.077e+02 1.709e+02 2.070e+02 2.577e+02 5.953e+02, threshold=4.140e+02, percent-clipped=1.0 +2023-03-26 08:37:25,990 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39639.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 08:37:26,589 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39640.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:37:34,920 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-03-26 08:37:43,684 INFO [finetune.py:976] (1/7) Epoch 7, batch 5300, loss[loss=0.2351, simple_loss=0.3038, pruned_loss=0.08324, over 4825.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2724, pruned_loss=0.07402, over 954538.66 frames. ], batch size: 40, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:37:54,623 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0848, 1.7469, 2.4231, 1.6215, 2.2351, 2.4498, 1.7974, 2.4877], + device='cuda:1'), covar=tensor([0.1591, 0.2549, 0.1591, 0.2553, 0.1148, 0.1677, 0.3090, 0.1127], + device='cuda:1'), in_proj_covar=tensor([0.0201, 0.0203, 0.0197, 0.0195, 0.0180, 0.0220, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:38:17,440 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39700.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 08:38:32,613 INFO [finetune.py:976] (1/7) Epoch 7, batch 5350, loss[loss=0.1931, simple_loss=0.2654, pruned_loss=0.06038, over 4794.00 frames. ], tot_loss[loss=0.2096, simple_loss=0.2718, pruned_loss=0.07371, over 950466.09 frames. ], batch size: 45, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:38:40,829 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.014e+02 1.565e+02 1.855e+02 2.323e+02 5.491e+02, threshold=3.710e+02, percent-clipped=1.0 +2023-03-26 08:38:50,733 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.99 vs. limit=2.0 +2023-03-26 08:39:03,527 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7317, 1.5612, 1.5191, 1.6363, 1.0991, 3.4245, 1.4388, 2.0136], + device='cuda:1'), covar=tensor([0.3004, 0.2304, 0.2068, 0.2183, 0.1802, 0.0204, 0.2618, 0.1280], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0115, 0.0119, 0.0123, 0.0117, 0.0099, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 08:39:15,939 INFO [finetune.py:976] (1/7) Epoch 7, batch 5400, loss[loss=0.1544, simple_loss=0.2235, pruned_loss=0.0426, over 4765.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.2679, pruned_loss=0.07174, over 949486.31 frames. ], batch size: 27, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:39:16,661 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39767.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:39:32,051 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 08:39:50,813 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39815.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:39:51,311 INFO [finetune.py:976] (1/7) Epoch 7, batch 5450, loss[loss=0.2049, simple_loss=0.271, pruned_loss=0.06945, over 4819.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2656, pruned_loss=0.07084, over 951480.00 frames. ], batch size: 38, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:39:53,199 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39819.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:40:03,634 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.339e+01 1.599e+02 1.928e+02 2.299e+02 3.698e+02, threshold=3.856e+02, percent-clipped=0.0 +2023-03-26 08:40:03,741 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39828.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:40:16,808 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39841.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:40:24,648 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2941, 2.8408, 2.6855, 1.5217, 2.8085, 2.3609, 2.2382, 2.5755], + device='cuda:1'), covar=tensor([0.0872, 0.1042, 0.1939, 0.2228, 0.1752, 0.2115, 0.2201, 0.1236], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0201, 0.0203, 0.0189, 0.0219, 0.0207, 0.0224, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:40:54,098 INFO [finetune.py:976] (1/7) Epoch 7, batch 5500, loss[loss=0.1597, simple_loss=0.235, pruned_loss=0.04218, over 4791.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2632, pruned_loss=0.07039, over 952096.93 frames. ], batch size: 29, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:41:05,330 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39876.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:41:18,431 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=39889.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:41:26,284 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4877, 1.4664, 1.7814, 1.8585, 1.6391, 3.2317, 1.3287, 1.6545], + device='cuda:1'), covar=tensor([0.0875, 0.1711, 0.1143, 0.0937, 0.1439, 0.0279, 0.1450, 0.1557], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 08:41:57,091 INFO [finetune.py:976] (1/7) Epoch 7, batch 5550, loss[loss=0.2062, simple_loss=0.2889, pruned_loss=0.06173, over 4801.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2644, pruned_loss=0.07066, over 953048.97 frames. ], batch size: 45, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:42:09,838 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.056e+02 1.647e+02 1.995e+02 2.278e+02 3.177e+02, threshold=3.991e+02, percent-clipped=0.0 +2023-03-26 08:42:09,948 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5444, 1.2930, 1.7606, 1.9389, 1.5713, 3.3425, 1.1346, 1.4774], + device='cuda:1'), covar=tensor([0.1169, 0.2533, 0.1444, 0.1161, 0.1934, 0.0316, 0.2233, 0.2406], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 08:42:27,985 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39940.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:42:41,271 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1131, 1.6506, 2.7118, 4.0053, 2.7122, 2.6540, 1.3433, 3.2070], + device='cuda:1'), covar=tensor([0.1658, 0.1606, 0.1199, 0.0525, 0.0783, 0.1523, 0.1679, 0.0491], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0134, 0.0165, 0.0101, 0.0139, 0.0127, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 08:42:58,493 INFO [finetune.py:976] (1/7) Epoch 7, batch 5600, loss[loss=0.2235, simple_loss=0.2846, pruned_loss=0.08123, over 4909.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.2688, pruned_loss=0.07189, over 954229.07 frames. ], batch size: 37, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:43:08,676 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 08:43:20,784 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=39988.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:43:30,216 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39995.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 08:43:32,590 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3230, 1.5425, 1.6637, 0.8595, 1.4272, 1.8017, 1.8562, 1.4631], + device='cuda:1'), covar=tensor([0.1065, 0.0731, 0.0470, 0.0637, 0.0481, 0.0689, 0.0362, 0.0712], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0157, 0.0121, 0.0138, 0.0133, 0.0126, 0.0146, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.7039e-05, 1.1610e-04, 8.7247e-05, 1.0027e-04, 9.5243e-05, 9.2192e-05, + 1.0727e-04, 1.0810e-04], device='cuda:1') +2023-03-26 08:43:52,202 INFO [finetune.py:976] (1/7) Epoch 7, batch 5650, loss[loss=0.1923, simple_loss=0.2597, pruned_loss=0.06241, over 4934.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2725, pruned_loss=0.07282, over 954722.44 frames. ], batch size: 33, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:44:09,060 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.182e+02 1.647e+02 1.995e+02 2.469e+02 4.643e+02, threshold=3.989e+02, percent-clipped=3.0 +2023-03-26 08:44:50,690 INFO [finetune.py:976] (1/7) Epoch 7, batch 5700, loss[loss=0.1468, simple_loss=0.1962, pruned_loss=0.04869, over 4032.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2689, pruned_loss=0.0728, over 938256.91 frames. ], batch size: 17, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:45:03,165 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.28 vs. limit=5.0 +2023-03-26 08:45:42,071 INFO [finetune.py:976] (1/7) Epoch 8, batch 0, loss[loss=0.2331, simple_loss=0.2997, pruned_loss=0.08324, over 4839.00 frames. ], tot_loss[loss=0.2331, simple_loss=0.2997, pruned_loss=0.08324, over 4839.00 frames. ], batch size: 49, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:45:42,072 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 08:45:59,526 INFO [finetune.py:1010] (1/7) Epoch 8, validation: loss=0.1624, simple_loss=0.234, pruned_loss=0.04544, over 2265189.00 frames. +2023-03-26 08:45:59,527 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 08:46:20,336 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40119.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:46:26,527 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40123.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:46:29,510 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.031e+02 1.580e+02 2.018e+02 2.508e+02 5.130e+02, threshold=4.036e+02, percent-clipped=1.0 +2023-03-26 08:46:30,863 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5409, 1.5197, 1.2779, 1.5724, 1.7777, 1.6702, 1.5559, 1.3026], + device='cuda:1'), covar=tensor([0.0285, 0.0316, 0.0575, 0.0292, 0.0237, 0.0552, 0.0330, 0.0431], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0113, 0.0140, 0.0116, 0.0105, 0.0101, 0.0092, 0.0110], + device='cuda:1'), out_proj_covar=tensor([6.9914e-05, 8.8522e-05, 1.1225e-04, 9.1261e-05, 8.2521e-05, 7.5197e-05, + 6.9415e-05, 8.5638e-05], device='cuda:1') +2023-03-26 08:46:41,220 INFO [finetune.py:976] (1/7) Epoch 8, batch 50, loss[loss=0.2, simple_loss=0.2684, pruned_loss=0.06577, over 4855.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2766, pruned_loss=0.07485, over 217687.51 frames. ], batch size: 44, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:47:04,281 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=40167.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:47:10,679 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40171.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:47:26,487 INFO [finetune.py:976] (1/7) Epoch 8, batch 100, loss[loss=0.1932, simple_loss=0.2605, pruned_loss=0.06291, over 4927.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.266, pruned_loss=0.07064, over 380928.06 frames. ], batch size: 37, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:47:27,218 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40195.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:47:42,319 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40218.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:47:48,307 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.570e+02 1.832e+02 2.394e+02 3.868e+02, threshold=3.663e+02, percent-clipped=0.0 +2023-03-26 08:47:58,789 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9077, 1.3917, 0.8505, 1.8374, 2.1541, 1.4966, 1.6421, 1.8593], + device='cuda:1'), covar=tensor([0.1344, 0.1991, 0.2164, 0.1134, 0.1952, 0.1979, 0.1414, 0.1834], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0114, 0.0092, 0.0123, 0.0095, 0.0099, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 08:47:59,304 INFO [finetune.py:976] (1/7) Epoch 8, batch 150, loss[loss=0.1836, simple_loss=0.2449, pruned_loss=0.06118, over 4931.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2612, pruned_loss=0.0693, over 510110.71 frames. ], batch size: 38, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:48:07,719 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40256.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:48:18,388 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40272.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:48:19,030 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2090, 1.9635, 2.6428, 1.6309, 2.4339, 2.3990, 1.7894, 2.5761], + device='cuda:1'), covar=tensor([0.1560, 0.1889, 0.1541, 0.2491, 0.0926, 0.1568, 0.2931, 0.1001], + device='cuda:1'), in_proj_covar=tensor([0.0204, 0.0206, 0.0200, 0.0198, 0.0183, 0.0223, 0.0220, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:48:22,637 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40279.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:48:33,049 INFO [finetune.py:976] (1/7) Epoch 8, batch 200, loss[loss=0.2109, simple_loss=0.2753, pruned_loss=0.07329, over 4908.00 frames. ], tot_loss[loss=0.2, simple_loss=0.2611, pruned_loss=0.06939, over 610396.36 frames. ], batch size: 36, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:48:33,737 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40295.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 08:48:55,741 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.120e+02 1.652e+02 1.957e+02 2.371e+02 3.958e+02, threshold=3.914e+02, percent-clipped=3.0 +2023-03-26 08:48:57,102 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40330.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:48:58,970 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40333.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:49:05,954 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=40343.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 08:49:06,468 INFO [finetune.py:976] (1/7) Epoch 8, batch 250, loss[loss=0.2486, simple_loss=0.3098, pruned_loss=0.09376, over 4801.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2668, pruned_loss=0.07195, over 686592.47 frames. ], batch size: 41, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:49:18,427 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 08:49:37,960 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40391.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:49:40,040 INFO [finetune.py:976] (1/7) Epoch 8, batch 300, loss[loss=0.215, simple_loss=0.2485, pruned_loss=0.09075, over 4127.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2696, pruned_loss=0.07212, over 745877.38 frames. ], batch size: 17, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:49:59,882 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40423.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:50:07,992 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.190e+02 1.680e+02 2.022e+02 2.440e+02 4.521e+02, threshold=4.043e+02, percent-clipped=1.0 +2023-03-26 08:50:16,624 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5779, 1.6490, 1.2995, 1.6443, 1.8520, 1.7865, 1.6115, 1.3529], + device='cuda:1'), covar=tensor([0.0347, 0.0294, 0.0583, 0.0250, 0.0213, 0.0450, 0.0312, 0.0375], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0112, 0.0141, 0.0116, 0.0105, 0.0102, 0.0091, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.0365e-05, 8.8071e-05, 1.1249e-04, 9.1651e-05, 8.2591e-05, 7.5484e-05, + 6.9195e-05, 8.5411e-05], device='cuda:1') +2023-03-26 08:50:27,979 INFO [finetune.py:976] (1/7) Epoch 8, batch 350, loss[loss=0.1897, simple_loss=0.2644, pruned_loss=0.0575, over 4842.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2704, pruned_loss=0.07204, over 790793.84 frames. ], batch size: 49, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:51:01,401 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=40471.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:51:01,427 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40471.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:51:26,232 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40492.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:51:27,353 INFO [finetune.py:976] (1/7) Epoch 8, batch 400, loss[loss=0.1781, simple_loss=0.2509, pruned_loss=0.05261, over 4928.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.2728, pruned_loss=0.07337, over 828157.06 frames. ], batch size: 42, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:51:52,934 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=40519.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:51:58,841 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.015e+02 1.662e+02 2.008e+02 2.590e+02 4.107e+02, threshold=4.016e+02, percent-clipped=2.0 +2023-03-26 08:52:11,107 INFO [finetune.py:976] (1/7) Epoch 8, batch 450, loss[loss=0.2119, simple_loss=0.2685, pruned_loss=0.07768, over 4820.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2707, pruned_loss=0.07295, over 854764.47 frames. ], batch size: 38, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:52:21,158 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40551.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:52:22,460 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40553.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:52:37,433 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6538, 1.1287, 0.8459, 1.4996, 2.0392, 1.0397, 1.3763, 1.5387], + device='cuda:1'), covar=tensor([0.1523, 0.2244, 0.2058, 0.1262, 0.1987, 0.2086, 0.1629, 0.2048], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0096, 0.0113, 0.0092, 0.0122, 0.0095, 0.0099, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 08:52:41,186 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40574.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:52:54,277 INFO [finetune.py:976] (1/7) Epoch 8, batch 500, loss[loss=0.1701, simple_loss=0.2241, pruned_loss=0.05808, over 4751.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.269, pruned_loss=0.07241, over 878398.40 frames. ], batch size: 26, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:53:17,849 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.064e+02 1.648e+02 1.946e+02 2.379e+02 4.476e+02, threshold=3.892e+02, percent-clipped=1.0 +2023-03-26 08:53:17,928 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40628.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:53:28,115 INFO [finetune.py:976] (1/7) Epoch 8, batch 550, loss[loss=0.1919, simple_loss=0.2619, pruned_loss=0.06099, over 4903.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.266, pruned_loss=0.07155, over 896362.86 frames. ], batch size: 43, lr: 3.85e-03, grad_scale: 16.0 +2023-03-26 08:53:32,521 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40651.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:53:56,752 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40686.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:54:01,541 INFO [finetune.py:976] (1/7) Epoch 8, batch 600, loss[loss=0.2072, simple_loss=0.2673, pruned_loss=0.07356, over 4784.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2673, pruned_loss=0.07242, over 910996.53 frames. ], batch size: 25, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 08:54:14,592 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40712.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:54:24,591 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.141e+02 1.757e+02 2.080e+02 2.524e+02 4.426e+02, threshold=4.160e+02, percent-clipped=1.0 +2023-03-26 08:54:27,611 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7616, 2.5273, 2.0519, 1.0861, 2.2945, 2.0862, 1.9294, 2.3708], + device='cuda:1'), covar=tensor([0.0970, 0.0802, 0.1823, 0.2301, 0.1490, 0.2137, 0.2249, 0.1023], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0200, 0.0202, 0.0189, 0.0218, 0.0206, 0.0223, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:54:34,712 INFO [finetune.py:976] (1/7) Epoch 8, batch 650, loss[loss=0.2017, simple_loss=0.2797, pruned_loss=0.06182, over 4807.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.2703, pruned_loss=0.07297, over 921087.25 frames. ], batch size: 51, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 08:54:59,708 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 08:55:08,422 INFO [finetune.py:976] (1/7) Epoch 8, batch 700, loss[loss=0.182, simple_loss=0.2194, pruned_loss=0.07232, over 4168.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2703, pruned_loss=0.07262, over 926176.30 frames. ], batch size: 18, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 08:55:12,809 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9537, 1.7915, 1.5376, 1.7799, 1.6914, 1.6878, 1.7175, 2.4251], + device='cuda:1'), covar=tensor([0.4761, 0.5373, 0.4005, 0.5029, 0.4947, 0.2979, 0.5050, 0.1830], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0257, 0.0219, 0.0279, 0.0241, 0.0205, 0.0244, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:55:31,876 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.288e+02 1.702e+02 1.948e+02 2.422e+02 4.930e+02, threshold=3.896e+02, percent-clipped=3.0 +2023-03-26 08:55:51,722 INFO [finetune.py:976] (1/7) Epoch 8, batch 750, loss[loss=0.2386, simple_loss=0.3059, pruned_loss=0.08562, over 4847.00 frames. ], tot_loss[loss=0.2093, simple_loss=0.2721, pruned_loss=0.07331, over 931873.91 frames. ], batch size: 44, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 08:55:54,207 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40848.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:56:00,047 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-03-26 08:56:01,140 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40851.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:56:20,354 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40865.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:56:32,133 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40874.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:56:55,830 INFO [finetune.py:976] (1/7) Epoch 8, batch 800, loss[loss=0.1596, simple_loss=0.2338, pruned_loss=0.04274, over 4827.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2729, pruned_loss=0.07354, over 937357.42 frames. ], batch size: 47, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 08:57:03,992 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=40899.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:57:04,654 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40900.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:57:23,553 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=40922.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:57:26,525 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40926.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:57:27,577 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.008e+02 1.606e+02 1.985e+02 2.397e+02 9.945e+02, threshold=3.971e+02, percent-clipped=3.0 +2023-03-26 08:57:27,688 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40928.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:57:45,609 INFO [finetune.py:976] (1/7) Epoch 8, batch 850, loss[loss=0.1876, simple_loss=0.2585, pruned_loss=0.05836, over 4928.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.271, pruned_loss=0.07293, over 941802.22 frames. ], batch size: 38, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 08:58:00,389 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40961.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 08:58:06,567 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 08:58:10,963 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=40976.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:58:18,102 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40986.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:58:22,825 INFO [finetune.py:976] (1/7) Epoch 8, batch 900, loss[loss=0.1873, simple_loss=0.244, pruned_loss=0.06532, over 4802.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2679, pruned_loss=0.07162, over 945402.28 frames. ], batch size: 29, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 08:58:25,184 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40997.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:58:31,501 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41007.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:58:46,165 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.012e+02 1.525e+02 1.868e+02 2.283e+02 3.598e+02, threshold=3.736e+02, percent-clipped=0.0 +2023-03-26 08:58:50,347 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=41034.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:58:56,854 INFO [finetune.py:976] (1/7) Epoch 8, batch 950, loss[loss=0.1791, simple_loss=0.2458, pruned_loss=0.0562, over 4908.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2671, pruned_loss=0.07181, over 945976.73 frames. ], batch size: 37, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 08:58:57,584 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41045.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:59:06,033 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41058.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:59:30,583 INFO [finetune.py:976] (1/7) Epoch 8, batch 1000, loss[loss=0.1873, simple_loss=0.2446, pruned_loss=0.06497, over 4903.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2698, pruned_loss=0.07269, over 950009.94 frames. ], batch size: 32, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 08:59:36,007 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41102.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:59:38,361 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41106.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 08:59:43,847 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2208, 2.2949, 2.0906, 1.5382, 2.3310, 2.2792, 2.1632, 1.9239], + device='cuda:1'), covar=tensor([0.0606, 0.0616, 0.0808, 0.0944, 0.0477, 0.0745, 0.0716, 0.1082], + device='cuda:1'), in_proj_covar=tensor([0.0137, 0.0134, 0.0145, 0.0126, 0.0114, 0.0145, 0.0146, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 08:59:52,970 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.084e+02 1.652e+02 2.000e+02 2.359e+02 4.809e+02, threshold=4.000e+02, percent-clipped=2.0 +2023-03-26 08:59:59,032 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0703, 1.5337, 2.8706, 1.7289, 2.3627, 2.3182, 1.5690, 2.4766], + device='cuda:1'), covar=tensor([0.1601, 0.2438, 0.1154, 0.1935, 0.1004, 0.1535, 0.3016, 0.1080], + device='cuda:1'), in_proj_covar=tensor([0.0203, 0.0205, 0.0198, 0.0196, 0.0181, 0.0220, 0.0220, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:00:04,080 INFO [finetune.py:976] (1/7) Epoch 8, batch 1050, loss[loss=0.2727, simple_loss=0.3328, pruned_loss=0.1063, over 4823.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2736, pruned_loss=0.07355, over 951920.85 frames. ], batch size: 47, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:00:06,622 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41148.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:00:16,284 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41163.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:00:37,444 INFO [finetune.py:976] (1/7) Epoch 8, batch 1100, loss[loss=0.2216, simple_loss=0.2946, pruned_loss=0.07426, over 4810.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2739, pruned_loss=0.07401, over 953191.31 frames. ], batch size: 39, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:00:38,742 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=41196.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:00:54,958 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41221.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:00:59,684 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.124e+02 1.750e+02 2.155e+02 2.664e+02 4.791e+02, threshold=4.309e+02, percent-clipped=2.0 +2023-03-26 09:01:17,459 INFO [finetune.py:976] (1/7) Epoch 8, batch 1150, loss[loss=0.219, simple_loss=0.2868, pruned_loss=0.0756, over 4853.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2737, pruned_loss=0.07343, over 953746.10 frames. ], batch size: 49, lr: 3.84e-03, grad_scale: 32.0 +2023-03-26 09:01:31,994 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41256.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 09:02:15,112 INFO [finetune.py:976] (1/7) Epoch 8, batch 1200, loss[loss=0.2241, simple_loss=0.284, pruned_loss=0.08213, over 4814.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2713, pruned_loss=0.07254, over 954559.56 frames. ], batch size: 33, lr: 3.84e-03, grad_scale: 32.0 +2023-03-26 09:02:17,526 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6768, 1.6161, 2.1265, 1.9956, 1.8967, 4.4332, 1.6393, 2.0448], + device='cuda:1'), covar=tensor([0.0967, 0.1750, 0.1066, 0.0970, 0.1523, 0.0144, 0.1398, 0.1542], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 09:02:24,085 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41307.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:02:37,270 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.297e+01 1.635e+02 1.914e+02 2.289e+02 4.123e+02, threshold=3.829e+02, percent-clipped=0.0 +2023-03-26 09:02:51,370 INFO [finetune.py:976] (1/7) Epoch 8, batch 1250, loss[loss=0.2034, simple_loss=0.2521, pruned_loss=0.07737, over 4744.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2676, pruned_loss=0.07148, over 954039.69 frames. ], batch size: 23, lr: 3.84e-03, grad_scale: 32.0 +2023-03-26 09:02:51,646 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-03-26 09:03:02,678 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41353.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:03:04,420 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=41355.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:03:21,184 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3175, 1.4368, 1.2836, 1.4451, 1.6157, 1.5777, 1.3478, 1.2372], + device='cuda:1'), covar=tensor([0.0396, 0.0245, 0.0489, 0.0261, 0.0223, 0.0443, 0.0332, 0.0387], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0112, 0.0140, 0.0117, 0.0105, 0.0101, 0.0091, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.0432e-05, 8.7906e-05, 1.1226e-04, 9.1909e-05, 8.2755e-05, 7.4953e-05, + 6.9185e-05, 8.5591e-05], device='cuda:1') +2023-03-26 09:03:32,985 INFO [finetune.py:976] (1/7) Epoch 8, batch 1300, loss[loss=0.1883, simple_loss=0.2418, pruned_loss=0.06743, over 4809.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2643, pruned_loss=0.0704, over 953986.22 frames. ], batch size: 45, lr: 3.84e-03, grad_scale: 32.0 +2023-03-26 09:03:37,902 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41401.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:03:41,658 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.57 vs. limit=5.0 +2023-03-26 09:03:46,177 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7826, 1.6880, 1.4848, 1.3885, 1.8191, 1.5399, 1.8909, 1.7473], + device='cuda:1'), covar=tensor([0.1639, 0.2366, 0.3353, 0.2831, 0.2816, 0.1946, 0.2915, 0.2130], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0189, 0.0233, 0.0253, 0.0234, 0.0192, 0.0210, 0.0192], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:03:46,778 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1318, 2.7783, 2.5647, 1.3501, 2.7530, 2.2567, 2.1420, 2.4137], + device='cuda:1'), covar=tensor([0.1122, 0.0913, 0.1545, 0.2321, 0.1687, 0.2194, 0.2033, 0.1222], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0201, 0.0202, 0.0189, 0.0219, 0.0207, 0.0224, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:03:56,247 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.077e+02 1.674e+02 1.900e+02 2.309e+02 4.379e+02, threshold=3.799e+02, percent-clipped=1.0 +2023-03-26 09:04:06,253 INFO [finetune.py:976] (1/7) Epoch 8, batch 1350, loss[loss=0.206, simple_loss=0.2773, pruned_loss=0.06736, over 4863.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2652, pruned_loss=0.07117, over 953498.85 frames. ], batch size: 34, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:04:16,291 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41458.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:04:38,804 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4115, 2.3073, 1.9749, 1.0565, 2.1565, 1.8162, 1.6125, 2.0471], + device='cuda:1'), covar=tensor([0.0809, 0.0763, 0.1484, 0.1969, 0.1485, 0.2029, 0.2143, 0.0991], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0200, 0.0203, 0.0189, 0.0218, 0.0206, 0.0223, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:04:39,877 INFO [finetune.py:976] (1/7) Epoch 8, batch 1400, loss[loss=0.2097, simple_loss=0.2539, pruned_loss=0.0827, over 4062.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2683, pruned_loss=0.07192, over 952215.13 frames. ], batch size: 17, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:04:42,844 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8570, 3.3520, 3.5068, 3.7352, 3.6226, 3.3114, 3.9342, 1.1418], + device='cuda:1'), covar=tensor([0.0883, 0.0856, 0.0851, 0.1034, 0.1306, 0.1588, 0.0771, 0.5179], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0242, 0.0274, 0.0293, 0.0332, 0.0281, 0.0301, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:04:58,583 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41521.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:05:03,328 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.146e+02 1.704e+02 2.004e+02 2.444e+02 3.700e+02, threshold=4.008e+02, percent-clipped=0.0 +2023-03-26 09:05:12,591 INFO [finetune.py:976] (1/7) Epoch 8, batch 1450, loss[loss=0.258, simple_loss=0.3138, pruned_loss=0.1011, over 4799.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2688, pruned_loss=0.07125, over 953124.27 frames. ], batch size: 51, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:05:21,944 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41556.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 09:05:30,811 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=41569.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:05:30,857 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4730, 1.3183, 1.3524, 1.3652, 0.9077, 2.9494, 1.0445, 1.5386], + device='cuda:1'), covar=tensor([0.3589, 0.2582, 0.2236, 0.2612, 0.2150, 0.0260, 0.2817, 0.1403], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0119, 0.0123, 0.0116, 0.0098, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 09:05:46,315 INFO [finetune.py:976] (1/7) Epoch 8, batch 1500, loss[loss=0.2314, simple_loss=0.2885, pruned_loss=0.08711, over 4924.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2707, pruned_loss=0.07205, over 951784.09 frames. ], batch size: 33, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:05:52,970 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8424, 1.7587, 2.1130, 2.1869, 1.9940, 3.7895, 1.7183, 1.9350], + device='cuda:1'), covar=tensor([0.0914, 0.1631, 0.0947, 0.0838, 0.1385, 0.0308, 0.1333, 0.1439], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0082, 0.0076, 0.0079, 0.0093, 0.0083, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 09:05:53,540 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=41604.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:06:10,823 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.231e+02 1.635e+02 1.924e+02 2.365e+02 3.634e+02, threshold=3.848e+02, percent-clipped=0.0 +2023-03-26 09:06:22,436 INFO [finetune.py:976] (1/7) Epoch 8, batch 1550, loss[loss=0.2126, simple_loss=0.2688, pruned_loss=0.07819, over 4747.00 frames. ], tot_loss[loss=0.207, simple_loss=0.27, pruned_loss=0.07198, over 951345.68 frames. ], batch size: 27, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:06:34,441 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41653.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:06:35,067 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41654.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:06:38,624 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1263, 1.6799, 2.0625, 1.9563, 1.7291, 1.7662, 1.8545, 1.8843], + device='cuda:1'), covar=tensor([0.4556, 0.5992, 0.4272, 0.5831, 0.6403, 0.4947, 0.7456, 0.4251], + device='cuda:1'), in_proj_covar=tensor([0.0231, 0.0241, 0.0253, 0.0253, 0.0244, 0.0221, 0.0272, 0.0227], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:07:19,721 INFO [finetune.py:976] (1/7) Epoch 8, batch 1600, loss[loss=0.2106, simple_loss=0.2626, pruned_loss=0.07928, over 4824.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2679, pruned_loss=0.07145, over 952781.50 frames. ], batch size: 33, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:07:25,552 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=41701.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:07:25,603 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41701.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:07:37,769 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0469, 2.0770, 2.0507, 1.2933, 2.1673, 2.1298, 1.9902, 1.7532], + device='cuda:1'), covar=tensor([0.0654, 0.0662, 0.0737, 0.1027, 0.0529, 0.0783, 0.0745, 0.1169], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0130, 0.0140, 0.0124, 0.0112, 0.0141, 0.0142, 0.0158], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:07:39,607 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41715.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 09:07:48,462 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1411, 1.6676, 2.0320, 1.8729, 1.6583, 1.6966, 1.8135, 1.8913], + device='cuda:1'), covar=tensor([0.4871, 0.6491, 0.4654, 0.6116, 0.7076, 0.5607, 0.8037, 0.4507], + device='cuda:1'), in_proj_covar=tensor([0.0231, 0.0241, 0.0254, 0.0254, 0.0245, 0.0221, 0.0272, 0.0227], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:07:48,911 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.136e+02 1.581e+02 1.949e+02 2.490e+02 4.755e+02, threshold=3.899e+02, percent-clipped=2.0 +2023-03-26 09:07:58,438 INFO [finetune.py:976] (1/7) Epoch 8, batch 1650, loss[loss=0.17, simple_loss=0.2276, pruned_loss=0.05616, over 4902.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2644, pruned_loss=0.07001, over 951867.01 frames. ], batch size: 32, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:08:01,528 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=41749.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:08:09,848 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41758.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:08:42,690 INFO [finetune.py:976] (1/7) Epoch 8, batch 1700, loss[loss=0.1357, simple_loss=0.2086, pruned_loss=0.03143, over 4799.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2644, pruned_loss=0.07033, over 954676.28 frames. ], batch size: 26, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:08:50,311 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=41806.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:09:02,716 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1369, 1.4240, 2.0202, 1.9142, 1.7745, 1.7274, 1.8396, 1.8366], + device='cuda:1'), covar=tensor([0.4095, 0.5308, 0.4576, 0.4677, 0.6147, 0.4562, 0.6317, 0.4251], + device='cuda:1'), in_proj_covar=tensor([0.0232, 0.0242, 0.0255, 0.0255, 0.0245, 0.0222, 0.0273, 0.0227], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:09:06,694 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.178e+02 1.763e+02 2.034e+02 2.335e+02 4.675e+02, threshold=4.069e+02, percent-clipped=2.0 +2023-03-26 09:09:10,998 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 09:09:16,758 INFO [finetune.py:976] (1/7) Epoch 8, batch 1750, loss[loss=0.2498, simple_loss=0.3048, pruned_loss=0.09742, over 4876.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2649, pruned_loss=0.07059, over 954871.06 frames. ], batch size: 34, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:09:33,013 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6493, 1.5123, 1.4262, 1.5983, 0.9687, 3.6414, 1.3504, 2.1087], + device='cuda:1'), covar=tensor([0.3417, 0.2523, 0.2188, 0.2303, 0.2022, 0.0150, 0.2856, 0.1273], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0115, 0.0119, 0.0123, 0.0117, 0.0098, 0.0101, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 09:09:38,948 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0443, 2.0214, 1.4777, 1.9621, 1.9773, 1.6888, 2.3980, 2.0040], + device='cuda:1'), covar=tensor([0.1522, 0.2608, 0.3827, 0.3203, 0.3005, 0.1949, 0.3484, 0.2151], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0189, 0.0234, 0.0254, 0.0235, 0.0193, 0.0211, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:09:50,589 INFO [finetune.py:976] (1/7) Epoch 8, batch 1800, loss[loss=0.2044, simple_loss=0.2767, pruned_loss=0.06609, over 4896.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2695, pruned_loss=0.07246, over 955738.62 frames. ], batch size: 35, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:09:57,995 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41906.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:10:07,251 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0048, 1.9621, 1.9731, 1.5390, 2.0728, 2.1866, 2.0796, 1.6658], + device='cuda:1'), covar=tensor([0.0529, 0.0587, 0.0734, 0.0823, 0.0669, 0.0524, 0.0534, 0.1001], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0132, 0.0142, 0.0125, 0.0113, 0.0142, 0.0144, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:10:13,609 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.102e+02 1.792e+02 2.103e+02 2.633e+02 4.479e+02, threshold=4.207e+02, percent-clipped=2.0 +2023-03-26 09:10:23,653 INFO [finetune.py:976] (1/7) Epoch 8, batch 1850, loss[loss=0.177, simple_loss=0.2509, pruned_loss=0.05156, over 4776.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2713, pruned_loss=0.07317, over 955823.66 frames. ], batch size: 28, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:10:26,667 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41948.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 09:10:30,555 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-26 09:10:38,712 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41967.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:10:57,329 INFO [finetune.py:976] (1/7) Epoch 8, batch 1900, loss[loss=0.1868, simple_loss=0.2613, pruned_loss=0.05616, over 4730.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2731, pruned_loss=0.07361, over 955465.23 frames. ], batch size: 59, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:11:02,717 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6133, 1.0546, 0.7983, 1.5236, 1.9964, 0.9718, 1.2429, 1.4251], + device='cuda:1'), covar=tensor([0.1623, 0.2483, 0.2200, 0.1430, 0.2183, 0.2167, 0.1785, 0.2228], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0098, 0.0115, 0.0093, 0.0125, 0.0097, 0.0102, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 09:11:08,254 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42009.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 09:11:08,792 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42010.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 09:11:22,120 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.207e+02 1.556e+02 1.920e+02 2.218e+02 3.872e+02, threshold=3.841e+02, percent-clipped=0.0 +2023-03-26 09:11:32,121 INFO [finetune.py:976] (1/7) Epoch 8, batch 1950, loss[loss=0.1453, simple_loss=0.2129, pruned_loss=0.03887, over 4217.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2706, pruned_loss=0.07222, over 955648.31 frames. ], batch size: 65, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:11:34,017 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8235, 1.7263, 1.7974, 1.0882, 1.9054, 1.9027, 1.7860, 1.5093], + device='cuda:1'), covar=tensor([0.0541, 0.0648, 0.0661, 0.0939, 0.0529, 0.0648, 0.0606, 0.1158], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0131, 0.0141, 0.0124, 0.0112, 0.0141, 0.0143, 0.0159], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:12:30,959 INFO [finetune.py:976] (1/7) Epoch 8, batch 2000, loss[loss=0.2145, simple_loss=0.2677, pruned_loss=0.08065, over 4249.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.269, pruned_loss=0.07182, over 956533.12 frames. ], batch size: 66, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:12:48,576 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-26 09:12:56,437 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.048e+02 1.506e+02 1.840e+02 2.176e+02 3.856e+02, threshold=3.679e+02, percent-clipped=1.0 +2023-03-26 09:13:06,567 INFO [finetune.py:976] (1/7) Epoch 8, batch 2050, loss[loss=0.2526, simple_loss=0.3145, pruned_loss=0.09532, over 4853.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2653, pruned_loss=0.0701, over 956794.34 frames. ], batch size: 47, lr: 3.84e-03, grad_scale: 16.0 +2023-03-26 09:13:53,279 INFO [finetune.py:976] (1/7) Epoch 8, batch 2100, loss[loss=0.171, simple_loss=0.2458, pruned_loss=0.04812, over 4828.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2658, pruned_loss=0.07072, over 957407.35 frames. ], batch size: 33, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:14:02,724 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-03-26 09:14:16,279 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.091e+02 1.710e+02 1.945e+02 2.376e+02 4.149e+02, threshold=3.889e+02, percent-clipped=2.0 +2023-03-26 09:14:26,997 INFO [finetune.py:976] (1/7) Epoch 8, batch 2150, loss[loss=0.2148, simple_loss=0.2778, pruned_loss=0.07594, over 4896.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2685, pruned_loss=0.07158, over 957945.28 frames. ], batch size: 32, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:14:38,878 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42262.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:14:49,292 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.7027, 1.7756, 1.7762, 1.1590, 1.8252, 1.9532, 1.9223, 1.5444], + device='cuda:1'), covar=tensor([0.1000, 0.0556, 0.0397, 0.0544, 0.0384, 0.0614, 0.0320, 0.0702], + device='cuda:1'), in_proj_covar=tensor([0.0128, 0.0155, 0.0120, 0.0135, 0.0131, 0.0125, 0.0144, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.5403e-05, 1.1412e-04, 8.6606e-05, 9.8280e-05, 9.3733e-05, 9.1471e-05, + 1.0598e-04, 1.0755e-04], device='cuda:1') +2023-03-26 09:15:18,053 INFO [finetune.py:976] (1/7) Epoch 8, batch 2200, loss[loss=0.1934, simple_loss=0.2619, pruned_loss=0.06247, over 4818.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2698, pruned_loss=0.0718, over 956761.46 frames. ], batch size: 38, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:15:25,344 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42304.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 09:15:29,023 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42310.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 09:15:32,573 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3848, 1.2349, 1.6130, 2.4541, 1.6902, 2.1340, 0.9020, 1.9783], + device='cuda:1'), covar=tensor([0.1848, 0.1768, 0.1322, 0.0728, 0.1006, 0.1252, 0.1766, 0.0824], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0135, 0.0165, 0.0102, 0.0139, 0.0127, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 09:15:45,962 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.121e+02 1.543e+02 1.921e+02 2.479e+02 5.347e+02, threshold=3.843e+02, percent-clipped=1.0 +2023-03-26 09:15:53,219 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1369, 2.0452, 1.6113, 2.2151, 2.0668, 1.7242, 2.4806, 2.0602], + device='cuda:1'), covar=tensor([0.1551, 0.2585, 0.3753, 0.3078, 0.2951, 0.2017, 0.3986, 0.2311], + device='cuda:1'), in_proj_covar=tensor([0.0173, 0.0189, 0.0235, 0.0254, 0.0237, 0.0194, 0.0212, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:16:07,395 INFO [finetune.py:976] (1/7) Epoch 8, batch 2250, loss[loss=0.1744, simple_loss=0.2513, pruned_loss=0.04875, over 4882.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2715, pruned_loss=0.07271, over 956106.06 frames. ], batch size: 43, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:16:27,423 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=42358.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:16:47,910 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42379.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:17:07,512 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-03-26 09:17:09,027 INFO [finetune.py:976] (1/7) Epoch 8, batch 2300, loss[loss=0.2106, simple_loss=0.273, pruned_loss=0.0741, over 4852.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2724, pruned_loss=0.07277, over 955479.76 frames. ], batch size: 49, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:17:15,807 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6436, 3.8044, 3.5452, 1.8372, 3.9883, 2.9415, 0.8080, 2.7572], + device='cuda:1'), covar=tensor([0.2579, 0.1780, 0.1518, 0.3158, 0.0903, 0.1024, 0.4220, 0.1412], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0171, 0.0160, 0.0128, 0.0155, 0.0122, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 09:17:51,921 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6579, 1.6257, 2.0019, 1.3202, 1.8252, 1.9372, 1.5414, 2.1603], + device='cuda:1'), covar=tensor([0.1360, 0.1748, 0.1419, 0.1884, 0.0883, 0.1348, 0.2540, 0.0775], + device='cuda:1'), in_proj_covar=tensor([0.0201, 0.0203, 0.0197, 0.0195, 0.0180, 0.0219, 0.0219, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:17:57,226 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.935e+01 1.473e+02 1.816e+02 2.175e+02 3.275e+02, threshold=3.633e+02, percent-clipped=0.0 +2023-03-26 09:18:06,500 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42440.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:18:09,264 INFO [finetune.py:976] (1/7) Epoch 8, batch 2350, loss[loss=0.2162, simple_loss=0.2794, pruned_loss=0.07647, over 4771.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2687, pruned_loss=0.07123, over 955550.88 frames. ], batch size: 26, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:18:40,031 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1302, 1.9102, 1.4819, 0.6259, 1.6512, 1.7746, 1.5664, 1.8699], + device='cuda:1'), covar=tensor([0.0836, 0.0656, 0.1273, 0.1797, 0.1227, 0.1969, 0.2000, 0.0661], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0201, 0.0203, 0.0188, 0.0219, 0.0207, 0.0223, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:18:49,073 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9968, 2.0512, 1.7014, 1.5574, 2.2335, 2.2095, 1.9820, 1.9044], + device='cuda:1'), covar=tensor([0.0276, 0.0295, 0.0501, 0.0345, 0.0310, 0.0517, 0.0332, 0.0344], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0111, 0.0139, 0.0116, 0.0104, 0.0101, 0.0091, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.0068e-05, 8.7079e-05, 1.1133e-04, 9.1347e-05, 8.1824e-05, 7.4771e-05, + 6.8501e-05, 8.4775e-05], device='cuda:1') +2023-03-26 09:18:51,872 INFO [finetune.py:976] (1/7) Epoch 8, batch 2400, loss[loss=0.2478, simple_loss=0.2861, pruned_loss=0.1047, over 4295.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2664, pruned_loss=0.07101, over 955936.44 frames. ], batch size: 65, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:19:02,409 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42506.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:19:25,681 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.076e+02 1.517e+02 1.798e+02 2.223e+02 5.682e+02, threshold=3.597e+02, percent-clipped=2.0 +2023-03-26 09:19:35,402 INFO [finetune.py:976] (1/7) Epoch 8, batch 2450, loss[loss=0.1941, simple_loss=0.272, pruned_loss=0.05808, over 4898.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2628, pruned_loss=0.06912, over 956230.35 frames. ], batch size: 32, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:19:47,983 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42562.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:19:51,990 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42567.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:20:04,304 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.74 vs. limit=2.0 +2023-03-26 09:20:08,915 INFO [finetune.py:976] (1/7) Epoch 8, batch 2500, loss[loss=0.2126, simple_loss=0.2792, pruned_loss=0.07294, over 4911.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2669, pruned_loss=0.07172, over 956076.80 frames. ], batch size: 37, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:20:16,176 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0618, 1.2761, 1.1284, 1.3507, 1.3298, 2.4099, 1.1806, 1.4272], + device='cuda:1'), covar=tensor([0.0980, 0.1818, 0.1105, 0.0946, 0.1706, 0.0372, 0.1495, 0.1673], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0093, 0.0082, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 09:20:16,183 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42604.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 09:20:20,726 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=42610.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:20:33,719 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.056e+02 1.709e+02 1.979e+02 2.316e+02 5.134e+02, threshold=3.959e+02, percent-clipped=4.0 +2023-03-26 09:20:42,882 INFO [finetune.py:976] (1/7) Epoch 8, batch 2550, loss[loss=0.1906, simple_loss=0.2586, pruned_loss=0.06131, over 4901.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2694, pruned_loss=0.07214, over 956888.44 frames. ], batch size: 32, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:20:46,541 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42649.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:20:53,410 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=42652.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 09:21:12,819 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5985, 3.4758, 3.3253, 1.5002, 3.5330, 2.5451, 1.1099, 2.3935], + device='cuda:1'), covar=tensor([0.2172, 0.2141, 0.1396, 0.3400, 0.1008, 0.1167, 0.3920, 0.1481], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0169, 0.0158, 0.0128, 0.0154, 0.0121, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 09:21:25,355 INFO [finetune.py:976] (1/7) Epoch 8, batch 2600, loss[loss=0.2177, simple_loss=0.2796, pruned_loss=0.07794, over 4831.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2703, pruned_loss=0.07247, over 954103.83 frames. ], batch size: 49, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:21:36,676 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42710.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:21:49,602 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.153e+02 1.771e+02 2.168e+02 2.787e+02 4.495e+02, threshold=4.337e+02, percent-clipped=4.0 +2023-03-26 09:21:53,824 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42735.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:21:55,774 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.98 vs. limit=5.0 +2023-03-26 09:21:57,451 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3090, 1.4510, 1.3527, 1.5976, 1.5019, 2.9711, 1.3412, 1.5769], + device='cuda:1'), covar=tensor([0.0964, 0.1718, 0.1118, 0.0923, 0.1592, 0.0252, 0.1424, 0.1675], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0082, 0.0084, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 09:21:59,151 INFO [finetune.py:976] (1/7) Epoch 8, batch 2650, loss[loss=0.1928, simple_loss=0.2659, pruned_loss=0.05986, over 4772.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2714, pruned_loss=0.07245, over 956790.53 frames. ], batch size: 26, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:22:40,399 INFO [finetune.py:976] (1/7) Epoch 8, batch 2700, loss[loss=0.1962, simple_loss=0.2597, pruned_loss=0.0663, over 4828.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2692, pruned_loss=0.07071, over 957067.64 frames. ], batch size: 47, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:23:27,701 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.159e+02 1.604e+02 1.897e+02 2.218e+02 3.599e+02, threshold=3.793e+02, percent-clipped=0.0 +2023-03-26 09:23:37,241 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-26 09:23:46,885 INFO [finetune.py:976] (1/7) Epoch 8, batch 2750, loss[loss=0.1759, simple_loss=0.2405, pruned_loss=0.05567, over 4909.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2652, pruned_loss=0.06867, over 956658.78 frames. ], batch size: 35, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:23:55,959 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42859.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:24:02,855 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42862.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:24:14,757 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42879.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:24:31,685 INFO [finetune.py:976] (1/7) Epoch 8, batch 2800, loss[loss=0.1731, simple_loss=0.231, pruned_loss=0.05762, over 4827.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.262, pruned_loss=0.06765, over 958157.63 frames. ], batch size: 30, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:24:34,886 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 09:24:37,810 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0692, 2.6175, 2.5693, 1.1792, 2.7437, 2.0625, 0.8584, 1.8347], + device='cuda:1'), covar=tensor([0.2279, 0.2186, 0.1635, 0.3596, 0.1376, 0.1194, 0.3792, 0.1751], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0170, 0.0158, 0.0128, 0.0154, 0.0122, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 09:24:48,560 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42920.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:24:54,864 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.137e+02 1.632e+02 1.941e+02 2.379e+02 3.960e+02, threshold=3.882e+02, percent-clipped=2.0 +2023-03-26 09:25:02,618 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42940.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:25:04,920 INFO [finetune.py:976] (1/7) Epoch 8, batch 2850, loss[loss=0.2148, simple_loss=0.2822, pruned_loss=0.07374, over 4887.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2621, pruned_loss=0.06849, over 957980.58 frames. ], batch size: 32, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:25:13,460 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5277, 1.6808, 1.9904, 1.8709, 1.7293, 4.4082, 1.5049, 1.9554], + device='cuda:1'), covar=tensor([0.1312, 0.2161, 0.1505, 0.1324, 0.1959, 0.0244, 0.1978, 0.2163], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0076, 0.0078, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 09:25:38,295 INFO [finetune.py:976] (1/7) Epoch 8, batch 2900, loss[loss=0.2034, simple_loss=0.2753, pruned_loss=0.06577, over 4151.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2653, pruned_loss=0.06965, over 958465.53 frames. ], batch size: 65, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:25:38,473 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 09:25:45,628 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43005.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:25:52,312 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4891, 1.0681, 0.7567, 1.3769, 1.8266, 0.6806, 1.2659, 1.2574], + device='cuda:1'), covar=tensor([0.1558, 0.2248, 0.1937, 0.1300, 0.2224, 0.2182, 0.1616, 0.2174], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0096, 0.0114, 0.0092, 0.0123, 0.0095, 0.0100, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 09:26:03,397 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.926e+01 1.731e+02 1.970e+02 2.373e+02 5.777e+02, threshold=3.941e+02, percent-clipped=2.0 +2023-03-26 09:26:13,024 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43035.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:26:13,079 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7828, 0.6608, 1.6222, 1.4919, 1.4386, 1.3892, 1.4041, 1.5401], + device='cuda:1'), covar=tensor([0.3342, 0.4807, 0.4056, 0.4157, 0.5361, 0.4024, 0.5193, 0.3838], + device='cuda:1'), in_proj_covar=tensor([0.0232, 0.0241, 0.0254, 0.0255, 0.0245, 0.0223, 0.0273, 0.0227], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:26:18,879 INFO [finetune.py:976] (1/7) Epoch 8, batch 2950, loss[loss=0.2411, simple_loss=0.2945, pruned_loss=0.0938, over 4806.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2704, pruned_loss=0.07188, over 958317.48 frames. ], batch size: 45, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:26:18,986 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8765, 1.7153, 1.6819, 1.7210, 1.3940, 4.1531, 1.5612, 2.2376], + device='cuda:1'), covar=tensor([0.3226, 0.2303, 0.2009, 0.2131, 0.1728, 0.0141, 0.2455, 0.1281], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0118, 0.0122, 0.0116, 0.0098, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 09:26:44,522 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=43083.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:26:44,623 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7976, 0.9938, 1.7240, 1.6410, 1.4986, 1.4885, 1.5404, 1.5509], + device='cuda:1'), covar=tensor([0.4357, 0.5797, 0.4809, 0.4837, 0.6109, 0.4699, 0.6294, 0.4525], + device='cuda:1'), in_proj_covar=tensor([0.0232, 0.0241, 0.0254, 0.0254, 0.0245, 0.0222, 0.0273, 0.0227], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:26:52,571 INFO [finetune.py:976] (1/7) Epoch 8, batch 3000, loss[loss=0.2588, simple_loss=0.3053, pruned_loss=0.1062, over 4879.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2722, pruned_loss=0.07267, over 957742.29 frames. ], batch size: 35, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:26:52,571 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 09:26:58,158 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7148, 3.3296, 3.4292, 3.5741, 3.5004, 3.3113, 3.7386, 1.6002], + device='cuda:1'), covar=tensor([0.0780, 0.0748, 0.0754, 0.0825, 0.1069, 0.1266, 0.0602, 0.4050], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0242, 0.0277, 0.0295, 0.0331, 0.0284, 0.0302, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:26:59,181 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7353, 1.5275, 2.0650, 1.3917, 1.7041, 1.9373, 1.5391, 2.0363], + device='cuda:1'), covar=tensor([0.1280, 0.2125, 0.1286, 0.1826, 0.0967, 0.1262, 0.2995, 0.0852], + device='cuda:1'), in_proj_covar=tensor([0.0202, 0.0205, 0.0198, 0.0197, 0.0183, 0.0221, 0.0220, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:27:10,877 INFO [finetune.py:1010] (1/7) Epoch 8, validation: loss=0.16, simple_loss=0.2311, pruned_loss=0.04446, over 2265189.00 frames. +2023-03-26 09:27:10,878 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 09:27:49,849 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.022e+02 1.686e+02 2.049e+02 2.426e+02 3.920e+02, threshold=4.099e+02, percent-clipped=0.0 +2023-03-26 09:28:00,453 INFO [finetune.py:976] (1/7) Epoch 8, batch 3050, loss[loss=0.2006, simple_loss=0.2605, pruned_loss=0.07035, over 4819.00 frames. ], tot_loss[loss=0.2095, simple_loss=0.2734, pruned_loss=0.0728, over 958659.15 frames. ], batch size: 33, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:28:01,720 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5061, 1.5861, 1.7296, 1.8747, 1.6130, 3.5600, 1.3959, 1.6551], + device='cuda:1'), covar=tensor([0.0991, 0.1758, 0.1160, 0.0941, 0.1575, 0.0222, 0.1466, 0.1746], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0080, 0.0075, 0.0078, 0.0091, 0.0082, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:1') +2023-03-26 09:28:09,344 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.67 vs. limit=5.0 +2023-03-26 09:28:13,436 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43162.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:28:34,347 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9356, 1.7784, 1.4697, 1.7003, 1.6765, 1.6526, 1.7057, 2.4226], + device='cuda:1'), covar=tensor([0.4946, 0.5811, 0.4267, 0.5156, 0.5075, 0.2980, 0.5155, 0.1915], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0257, 0.0220, 0.0278, 0.0239, 0.0204, 0.0243, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:28:36,035 INFO [finetune.py:976] (1/7) Epoch 8, batch 3100, loss[loss=0.208, simple_loss=0.2651, pruned_loss=0.07545, over 4833.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2692, pruned_loss=0.07076, over 956163.68 frames. ], batch size: 30, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:28:52,821 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=43210.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:29:01,146 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43215.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:29:03,607 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6396, 1.5058, 2.2519, 3.4681, 2.3874, 2.4931, 1.1136, 2.6007], + device='cuda:1'), covar=tensor([0.1886, 0.1600, 0.1340, 0.0661, 0.0814, 0.1367, 0.1983, 0.0653], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0134, 0.0165, 0.0102, 0.0138, 0.0127, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 09:29:14,666 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.165e+02 1.652e+02 1.930e+02 2.337e+02 4.149e+02, threshold=3.860e+02, percent-clipped=1.0 +2023-03-26 09:29:23,815 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43235.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:29:34,856 INFO [finetune.py:976] (1/7) Epoch 8, batch 3150, loss[loss=0.2041, simple_loss=0.2614, pruned_loss=0.07338, over 4891.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2661, pruned_loss=0.07007, over 957226.91 frames. ], batch size: 35, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:30:07,267 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 09:30:24,889 INFO [finetune.py:976] (1/7) Epoch 8, batch 3200, loss[loss=0.198, simple_loss=0.2619, pruned_loss=0.06701, over 4905.00 frames. ], tot_loss[loss=0.199, simple_loss=0.2614, pruned_loss=0.06831, over 954343.20 frames. ], batch size: 32, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:30:32,644 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43305.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:30:49,474 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.070e+02 1.668e+02 2.087e+02 2.541e+02 1.424e+03, threshold=4.174e+02, percent-clipped=3.0 +2023-03-26 09:31:03,657 INFO [finetune.py:976] (1/7) Epoch 8, batch 3250, loss[loss=0.1864, simple_loss=0.2481, pruned_loss=0.06232, over 4797.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.263, pruned_loss=0.06923, over 954098.73 frames. ], batch size: 25, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:31:15,390 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=43353.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:31:59,777 INFO [finetune.py:976] (1/7) Epoch 8, batch 3300, loss[loss=0.2429, simple_loss=0.3154, pruned_loss=0.08514, over 4851.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2676, pruned_loss=0.07097, over 955112.20 frames. ], batch size: 44, lr: 3.83e-03, grad_scale: 16.0 +2023-03-26 09:32:45,623 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.223e+02 1.691e+02 2.029e+02 2.462e+02 4.055e+02, threshold=4.059e+02, percent-clipped=0.0 +2023-03-26 09:33:04,670 INFO [finetune.py:976] (1/7) Epoch 8, batch 3350, loss[loss=0.179, simple_loss=0.2405, pruned_loss=0.05878, over 4808.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2689, pruned_loss=0.07109, over 955017.46 frames. ], batch size: 25, lr: 3.83e-03, grad_scale: 32.0 +2023-03-26 09:33:37,126 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.30 vs. limit=5.0 +2023-03-26 09:33:51,119 INFO [finetune.py:976] (1/7) Epoch 8, batch 3400, loss[loss=0.2112, simple_loss=0.2782, pruned_loss=0.07206, over 4863.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.27, pruned_loss=0.07144, over 954705.11 frames. ], batch size: 34, lr: 3.83e-03, grad_scale: 32.0 +2023-03-26 09:34:04,478 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43514.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:34:05,112 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43515.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:34:15,392 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.089e+02 1.653e+02 1.866e+02 2.233e+02 4.638e+02, threshold=3.733e+02, percent-clipped=1.0 +2023-03-26 09:34:19,630 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43535.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:34:22,093 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5599, 2.2414, 1.8742, 0.9528, 2.0149, 1.9321, 1.7578, 2.1533], + device='cuda:1'), covar=tensor([0.0783, 0.0777, 0.1536, 0.2095, 0.1556, 0.2222, 0.2188, 0.0932], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0201, 0.0203, 0.0188, 0.0218, 0.0208, 0.0223, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:34:25,007 INFO [finetune.py:976] (1/7) Epoch 8, batch 3450, loss[loss=0.2226, simple_loss=0.2765, pruned_loss=0.08437, over 4815.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2699, pruned_loss=0.07092, over 954823.67 frames. ], batch size: 38, lr: 3.83e-03, grad_scale: 32.0 +2023-03-26 09:34:43,301 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=43563.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:34:55,895 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43575.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:35:02,198 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=43583.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:35:08,909 INFO [finetune.py:976] (1/7) Epoch 8, batch 3500, loss[loss=0.2292, simple_loss=0.2911, pruned_loss=0.0837, over 4796.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2673, pruned_loss=0.07016, over 954977.94 frames. ], batch size: 29, lr: 3.83e-03, grad_scale: 32.0 +2023-03-26 09:35:34,550 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.199e+02 1.655e+02 2.028e+02 2.395e+02 4.370e+02, threshold=4.057e+02, percent-clipped=4.0 +2023-03-26 09:35:44,695 INFO [finetune.py:976] (1/7) Epoch 8, batch 3550, loss[loss=0.2007, simple_loss=0.2547, pruned_loss=0.07339, over 4898.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2642, pruned_loss=0.06929, over 953934.98 frames. ], batch size: 36, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:36:12,273 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.90 vs. limit=5.0 +2023-03-26 09:36:13,463 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 09:36:18,004 INFO [finetune.py:976] (1/7) Epoch 8, batch 3600, loss[loss=0.1935, simple_loss=0.2701, pruned_loss=0.05847, over 4918.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2624, pruned_loss=0.06869, over 954853.65 frames. ], batch size: 43, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:36:18,670 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6016, 3.1058, 3.0188, 1.5903, 3.2623, 2.4735, 1.3750, 2.2998], + device='cuda:1'), covar=tensor([0.2920, 0.2066, 0.1613, 0.2999, 0.1203, 0.1014, 0.3167, 0.1423], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0171, 0.0160, 0.0129, 0.0155, 0.0122, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 09:36:38,843 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 09:36:40,279 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.156e+02 1.641e+02 1.836e+02 2.142e+02 3.900e+02, threshold=3.673e+02, percent-clipped=0.0 +2023-03-26 09:36:55,748 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4103, 1.3667, 0.9190, 2.0938, 2.5695, 1.8492, 1.9168, 2.2051], + device='cuda:1'), covar=tensor([0.1397, 0.2228, 0.2137, 0.1203, 0.1883, 0.1838, 0.1450, 0.1809], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0096, 0.0112, 0.0091, 0.0122, 0.0094, 0.0099, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 09:36:56,356 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8207, 1.6641, 2.0539, 2.0531, 1.8775, 4.3863, 1.6047, 1.9675], + device='cuda:1'), covar=tensor([0.0944, 0.1786, 0.1135, 0.0940, 0.1520, 0.0228, 0.1452, 0.1694], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0076, 0.0078, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 09:37:03,251 INFO [finetune.py:976] (1/7) Epoch 8, batch 3650, loss[loss=0.203, simple_loss=0.2739, pruned_loss=0.06602, over 4919.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2646, pruned_loss=0.06988, over 954393.03 frames. ], batch size: 36, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:37:41,722 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43787.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:37:53,053 INFO [finetune.py:976] (1/7) Epoch 8, batch 3700, loss[loss=0.1778, simple_loss=0.2498, pruned_loss=0.05292, over 4891.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2693, pruned_loss=0.07143, over 955877.75 frames. ], batch size: 35, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:38:37,141 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.066e+01 1.637e+02 2.055e+02 2.501e+02 4.825e+02, threshold=4.110e+02, percent-clipped=4.0 +2023-03-26 09:38:56,856 INFO [finetune.py:976] (1/7) Epoch 8, batch 3750, loss[loss=0.2602, simple_loss=0.3229, pruned_loss=0.09875, over 4889.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2712, pruned_loss=0.0717, over 956716.40 frames. ], batch size: 43, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:39:00,432 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43848.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:39:13,805 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43870.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:39:30,691 INFO [finetune.py:976] (1/7) Epoch 8, batch 3800, loss[loss=0.2192, simple_loss=0.2812, pruned_loss=0.07863, over 4896.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.2721, pruned_loss=0.07232, over 953829.89 frames. ], batch size: 36, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:39:40,897 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43909.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:40:01,519 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.167e+02 1.597e+02 1.980e+02 2.453e+02 5.062e+02, threshold=3.959e+02, percent-clipped=3.0 +2023-03-26 09:40:16,148 INFO [finetune.py:976] (1/7) Epoch 8, batch 3850, loss[loss=0.1887, simple_loss=0.2672, pruned_loss=0.05511, over 4825.00 frames. ], tot_loss[loss=0.206, simple_loss=0.2703, pruned_loss=0.07083, over 955040.07 frames. ], batch size: 39, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:40:26,725 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43959.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:40:33,711 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43970.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:40:54,207 INFO [finetune.py:976] (1/7) Epoch 8, batch 3900, loss[loss=0.1978, simple_loss=0.2477, pruned_loss=0.07391, over 4790.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2675, pruned_loss=0.07003, over 955062.31 frames. ], batch size: 45, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:41:17,057 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44020.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:41:22,369 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.038e+02 1.530e+02 1.807e+02 2.225e+02 3.627e+02, threshold=3.614e+02, percent-clipped=0.0 +2023-03-26 09:41:32,526 INFO [finetune.py:976] (1/7) Epoch 8, batch 3950, loss[loss=0.1663, simple_loss=0.2407, pruned_loss=0.04592, over 4759.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2623, pruned_loss=0.06764, over 955463.22 frames. ], batch size: 26, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:41:38,507 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6878, 1.5651, 1.4473, 1.7266, 2.2604, 1.8252, 1.3553, 1.3701], + device='cuda:1'), covar=tensor([0.2384, 0.2141, 0.2060, 0.1867, 0.1742, 0.1202, 0.2644, 0.2012], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0209, 0.0205, 0.0188, 0.0240, 0.0178, 0.0214, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:41:53,970 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9743, 4.3301, 4.4990, 4.7847, 4.6813, 4.5156, 5.1108, 1.7212], + device='cuda:1'), covar=tensor([0.0777, 0.0812, 0.0757, 0.0970, 0.1249, 0.1433, 0.0582, 0.5193], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0241, 0.0276, 0.0294, 0.0332, 0.0284, 0.0301, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:42:01,904 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.78 vs. limit=5.0 +2023-03-26 09:42:06,485 INFO [finetune.py:976] (1/7) Epoch 8, batch 4000, loss[loss=0.2002, simple_loss=0.2746, pruned_loss=0.06285, over 4934.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2626, pruned_loss=0.06882, over 954892.91 frames. ], batch size: 38, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:42:37,513 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.105e+02 1.793e+02 2.145e+02 2.588e+02 4.712e+02, threshold=4.291e+02, percent-clipped=10.0 +2023-03-26 09:42:56,487 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44143.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:42:57,046 INFO [finetune.py:976] (1/7) Epoch 8, batch 4050, loss[loss=0.2913, simple_loss=0.3497, pruned_loss=0.1164, over 4858.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2662, pruned_loss=0.07031, over 954673.79 frames. ], batch size: 44, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:43:31,060 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44170.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:43:49,921 INFO [finetune.py:976] (1/7) Epoch 8, batch 4100, loss[loss=0.2698, simple_loss=0.3143, pruned_loss=0.1126, over 4866.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.2696, pruned_loss=0.07145, over 954049.49 frames. ], batch size: 34, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:43:55,535 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5332, 1.4708, 1.9632, 1.8403, 1.7649, 3.9112, 1.3527, 1.7323], + device='cuda:1'), covar=tensor([0.0976, 0.1806, 0.1348, 0.0972, 0.1578, 0.0215, 0.1513, 0.1718], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0082, 0.0076, 0.0079, 0.0093, 0.0083, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 09:44:15,789 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=44218.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:44:16,415 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4302, 1.3354, 1.9381, 2.9011, 1.8971, 2.1409, 1.0931, 2.2476], + device='cuda:1'), covar=tensor([0.1878, 0.1609, 0.1287, 0.0574, 0.0951, 0.1247, 0.1812, 0.0727], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0118, 0.0136, 0.0167, 0.0103, 0.0140, 0.0128, 0.0103], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 09:44:22,431 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.086e+02 1.696e+02 1.895e+02 2.360e+02 4.949e+02, threshold=3.791e+02, percent-clipped=1.0 +2023-03-26 09:44:31,522 INFO [finetune.py:976] (1/7) Epoch 8, batch 4150, loss[loss=0.2441, simple_loss=0.3189, pruned_loss=0.08463, over 4927.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2734, pruned_loss=0.07324, over 952942.47 frames. ], batch size: 42, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:44:36,197 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7903, 3.3508, 3.4446, 3.6989, 3.5412, 3.3753, 3.8630, 1.1831], + device='cuda:1'), covar=tensor([0.0925, 0.0865, 0.0932, 0.1096, 0.1375, 0.1566, 0.0871, 0.5115], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0240, 0.0273, 0.0292, 0.0330, 0.0282, 0.0299, 0.0291], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:44:46,738 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44265.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:45:06,135 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7066, 1.3042, 1.0458, 1.6292, 1.9994, 1.2798, 1.4630, 1.6585], + device='cuda:1'), covar=tensor([0.1244, 0.1821, 0.1809, 0.1067, 0.1798, 0.2038, 0.1269, 0.1613], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0096, 0.0113, 0.0091, 0.0122, 0.0095, 0.0099, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 09:45:07,272 INFO [finetune.py:976] (1/7) Epoch 8, batch 4200, loss[loss=0.2002, simple_loss=0.2563, pruned_loss=0.07205, over 4696.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2729, pruned_loss=0.07277, over 952299.45 frames. ], batch size: 59, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:45:30,593 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44315.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:45:40,452 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.242e+02 1.746e+02 1.975e+02 2.337e+02 5.106e+02, threshold=3.951e+02, percent-clipped=1.0 +2023-03-26 09:45:54,906 INFO [finetune.py:976] (1/7) Epoch 8, batch 4250, loss[loss=0.1885, simple_loss=0.2463, pruned_loss=0.06538, over 4771.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2691, pruned_loss=0.07128, over 952031.17 frames. ], batch size: 26, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:46:20,077 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4185, 2.0568, 1.6909, 0.7455, 1.8424, 1.9467, 1.6928, 1.8557], + device='cuda:1'), covar=tensor([0.1116, 0.0777, 0.1658, 0.2057, 0.1557, 0.2313, 0.2209, 0.1039], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0200, 0.0201, 0.0188, 0.0216, 0.0205, 0.0221, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:46:26,567 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0802, 2.2092, 1.9258, 1.8007, 2.4756, 2.5178, 2.2430, 2.0309], + device='cuda:1'), covar=tensor([0.0340, 0.0334, 0.0522, 0.0349, 0.0264, 0.0373, 0.0275, 0.0389], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0110, 0.0139, 0.0115, 0.0103, 0.0101, 0.0091, 0.0109], + device='cuda:1'), out_proj_covar=tensor([6.9860e-05, 8.6409e-05, 1.1131e-04, 9.0217e-05, 8.0750e-05, 7.4646e-05, + 6.8925e-05, 8.4129e-05], device='cuda:1') +2023-03-26 09:46:32,565 INFO [finetune.py:976] (1/7) Epoch 8, batch 4300, loss[loss=0.1575, simple_loss=0.2213, pruned_loss=0.04687, over 4814.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2665, pruned_loss=0.07056, over 953736.12 frames. ], batch size: 25, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:46:52,976 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.57 vs. limit=5.0 +2023-03-26 09:46:56,824 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.088e+02 1.550e+02 1.851e+02 2.365e+02 4.860e+02, threshold=3.701e+02, percent-clipped=2.0 +2023-03-26 09:47:04,685 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5033, 1.3375, 1.2681, 1.5045, 1.6457, 1.5229, 0.8357, 1.3175], + device='cuda:1'), covar=tensor([0.2168, 0.2146, 0.1883, 0.1737, 0.1565, 0.1217, 0.2817, 0.1843], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0209, 0.0206, 0.0187, 0.0240, 0.0179, 0.0214, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:47:05,846 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44443.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:47:06,351 INFO [finetune.py:976] (1/7) Epoch 8, batch 4350, loss[loss=0.1721, simple_loss=0.2414, pruned_loss=0.0514, over 4765.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2629, pruned_loss=0.06947, over 955064.80 frames. ], batch size: 27, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:47:37,686 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=44491.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:47:39,432 INFO [finetune.py:976] (1/7) Epoch 8, batch 4400, loss[loss=0.2404, simple_loss=0.3019, pruned_loss=0.08952, over 4814.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2633, pruned_loss=0.0696, over 954717.22 frames. ], batch size: 40, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:48:17,051 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=44527.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:48:18,613 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.261e+02 1.746e+02 1.995e+02 2.515e+02 6.158e+02, threshold=3.991e+02, percent-clipped=2.0 +2023-03-26 09:48:28,720 INFO [finetune.py:976] (1/7) Epoch 8, batch 4450, loss[loss=0.1867, simple_loss=0.2504, pruned_loss=0.06152, over 4825.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2676, pruned_loss=0.07097, over 955765.08 frames. ], batch size: 30, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:48:37,003 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=44550.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:48:51,040 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44565.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:49:19,356 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44588.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:49:28,021 INFO [finetune.py:976] (1/7) Epoch 8, batch 4500, loss[loss=0.2377, simple_loss=0.2919, pruned_loss=0.0918, over 4793.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2689, pruned_loss=0.07137, over 955714.93 frames. ], batch size: 51, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:49:47,688 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44611.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:49:48,881 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=44613.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:49:50,663 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44615.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:50:00,000 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.305e+02 1.720e+02 2.026e+02 2.465e+02 5.780e+02, threshold=4.053e+02, percent-clipped=2.0 +2023-03-26 09:50:10,541 INFO [finetune.py:976] (1/7) Epoch 8, batch 4550, loss[loss=0.2077, simple_loss=0.2826, pruned_loss=0.06638, over 4919.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2692, pruned_loss=0.07095, over 956195.92 frames. ], batch size: 41, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:50:27,737 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=44663.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:50:31,998 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.7487, 1.5907, 1.7267, 0.8968, 1.7779, 2.0823, 1.8709, 1.5198], + device='cuda:1'), covar=tensor([0.0955, 0.0808, 0.0501, 0.0717, 0.0511, 0.0495, 0.0435, 0.0733], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0155, 0.0121, 0.0136, 0.0132, 0.0125, 0.0146, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.5748e-05, 1.1429e-04, 8.7295e-05, 9.8590e-05, 9.4723e-05, 9.1648e-05, + 1.0698e-04, 1.0851e-04], device='cuda:1') +2023-03-26 09:50:52,807 INFO [finetune.py:976] (1/7) Epoch 8, batch 4600, loss[loss=0.1759, simple_loss=0.2399, pruned_loss=0.05593, over 4659.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2681, pruned_loss=0.07013, over 955671.68 frames. ], batch size: 23, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:51:15,457 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.778e+01 1.566e+02 1.839e+02 2.114e+02 3.234e+02, threshold=3.678e+02, percent-clipped=0.0 +2023-03-26 09:51:25,985 INFO [finetune.py:976] (1/7) Epoch 8, batch 4650, loss[loss=0.2002, simple_loss=0.2623, pruned_loss=0.06905, over 4889.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2658, pruned_loss=0.06982, over 956593.15 frames. ], batch size: 35, lr: 3.82e-03, grad_scale: 32.0 +2023-03-26 09:51:38,113 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 09:51:59,449 INFO [finetune.py:976] (1/7) Epoch 8, batch 4700, loss[loss=0.1987, simple_loss=0.2596, pruned_loss=0.06886, over 4819.00 frames. ], tot_loss[loss=0.1984, simple_loss=0.2614, pruned_loss=0.06769, over 956408.19 frames. ], batch size: 38, lr: 3.82e-03, grad_scale: 16.0 +2023-03-26 09:52:22,745 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.499e+02 1.877e+02 2.319e+02 4.193e+02, threshold=3.754e+02, percent-clipped=1.0 +2023-03-26 09:52:32,228 INFO [finetune.py:976] (1/7) Epoch 8, batch 4750, loss[loss=0.1915, simple_loss=0.248, pruned_loss=0.06747, over 4809.00 frames. ], tot_loss[loss=0.197, simple_loss=0.2596, pruned_loss=0.06722, over 956525.08 frames. ], batch size: 25, lr: 3.82e-03, grad_scale: 16.0 +2023-03-26 09:52:39,350 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2369, 1.8843, 2.1194, 1.9997, 1.7777, 1.8551, 2.0381, 2.0491], + device='cuda:1'), covar=tensor([0.3824, 0.4593, 0.3849, 0.4981, 0.5702, 0.4408, 0.5739, 0.3527], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0241, 0.0253, 0.0255, 0.0247, 0.0223, 0.0273, 0.0228], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:52:58,205 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44883.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:52:58,321 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 09:53:05,273 INFO [finetune.py:976] (1/7) Epoch 8, batch 4800, loss[loss=0.1277, simple_loss=0.1906, pruned_loss=0.03235, over 4257.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2629, pruned_loss=0.06923, over 954395.99 frames. ], batch size: 18, lr: 3.82e-03, grad_scale: 16.0 +2023-03-26 09:53:15,865 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44906.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:53:40,829 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9103, 1.8391, 1.6338, 2.0399, 2.4467, 1.9056, 1.9195, 1.4987], + device='cuda:1'), covar=tensor([0.2284, 0.2223, 0.1972, 0.1706, 0.2091, 0.1252, 0.2372, 0.1943], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0210, 0.0206, 0.0188, 0.0241, 0.0178, 0.0215, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:53:41,286 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.083e+02 1.589e+02 1.932e+02 2.383e+02 4.430e+02, threshold=3.864e+02, percent-clipped=2.0 +2023-03-26 09:53:55,401 INFO [finetune.py:976] (1/7) Epoch 8, batch 4850, loss[loss=0.2314, simple_loss=0.2932, pruned_loss=0.08474, over 4773.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2663, pruned_loss=0.07005, over 952301.00 frames. ], batch size: 26, lr: 3.82e-03, grad_scale: 16.0 +2023-03-26 09:54:57,803 INFO [finetune.py:976] (1/7) Epoch 8, batch 4900, loss[loss=0.2116, simple_loss=0.283, pruned_loss=0.07016, over 4895.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2694, pruned_loss=0.07114, over 950580.07 frames. ], batch size: 43, lr: 3.82e-03, grad_scale: 16.0 +2023-03-26 09:55:25,592 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.182e+02 1.694e+02 2.008e+02 2.325e+02 4.035e+02, threshold=4.016e+02, percent-clipped=1.0 +2023-03-26 09:55:44,370 INFO [finetune.py:976] (1/7) Epoch 8, batch 4950, loss[loss=0.2037, simple_loss=0.2727, pruned_loss=0.06735, over 4748.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2708, pruned_loss=0.07149, over 953133.28 frames. ], batch size: 54, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 09:55:57,027 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45056.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:56:13,247 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45081.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:56:21,541 INFO [finetune.py:976] (1/7) Epoch 8, batch 5000, loss[loss=0.2131, simple_loss=0.2653, pruned_loss=0.08047, over 4883.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2694, pruned_loss=0.07083, over 953854.43 frames. ], batch size: 32, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 09:56:34,174 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7948, 1.7779, 1.5830, 1.8944, 2.3045, 1.9045, 1.4883, 1.4918], + device='cuda:1'), covar=tensor([0.2138, 0.1969, 0.1826, 0.1520, 0.1925, 0.1176, 0.2597, 0.1862], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0210, 0.0205, 0.0188, 0.0241, 0.0178, 0.0215, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:56:37,085 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45117.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 09:56:45,212 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.621e+02 1.919e+02 2.483e+02 3.797e+02, threshold=3.837e+02, percent-clipped=0.0 +2023-03-26 09:56:52,662 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45142.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:56:53,771 INFO [finetune.py:976] (1/7) Epoch 8, batch 5050, loss[loss=0.1908, simple_loss=0.2585, pruned_loss=0.06158, over 4911.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2659, pruned_loss=0.06965, over 956671.91 frames. ], batch size: 37, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 09:57:19,980 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45183.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:57:26,575 INFO [finetune.py:976] (1/7) Epoch 8, batch 5100, loss[loss=0.1766, simple_loss=0.2419, pruned_loss=0.0557, over 4753.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.263, pruned_loss=0.06896, over 957247.80 frames. ], batch size: 27, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 09:57:35,005 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45206.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:57:50,991 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0294, 1.9546, 1.4157, 1.8469, 1.8948, 1.5694, 2.6484, 1.9292], + device='cuda:1'), covar=tensor([0.1530, 0.2452, 0.3868, 0.3383, 0.3113, 0.1991, 0.2661, 0.2216], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0189, 0.0233, 0.0253, 0.0235, 0.0193, 0.0211, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:57:54,676 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9124, 1.3587, 1.8498, 1.7992, 1.5929, 1.5513, 1.6957, 1.6592], + device='cuda:1'), covar=tensor([0.4124, 0.4893, 0.4056, 0.4523, 0.5331, 0.4308, 0.5864, 0.3943], + device='cuda:1'), in_proj_covar=tensor([0.0231, 0.0241, 0.0252, 0.0255, 0.0246, 0.0223, 0.0272, 0.0228], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:57:55,113 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.141e+02 1.630e+02 1.903e+02 2.262e+02 3.588e+02, threshold=3.806e+02, percent-clipped=0.0 +2023-03-26 09:57:55,794 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=45231.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:58:04,088 INFO [finetune.py:976] (1/7) Epoch 8, batch 5150, loss[loss=0.2806, simple_loss=0.3292, pruned_loss=0.116, over 4810.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2649, pruned_loss=0.0707, over 956715.29 frames. ], batch size: 41, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 09:58:10,695 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2045, 1.7931, 2.5998, 3.9325, 2.9099, 2.7189, 0.7922, 3.1414], + device='cuda:1'), covar=tensor([0.1713, 0.1558, 0.1415, 0.0539, 0.0694, 0.1600, 0.2236, 0.0489], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0118, 0.0136, 0.0167, 0.0104, 0.0140, 0.0128, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 09:58:11,257 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=45254.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 09:58:40,359 INFO [finetune.py:976] (1/7) Epoch 8, batch 5200, loss[loss=0.183, simple_loss=0.2564, pruned_loss=0.05483, over 4831.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2685, pruned_loss=0.07187, over 956983.35 frames. ], batch size: 30, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 09:59:08,001 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0598, 1.8771, 1.5766, 1.8690, 1.8503, 1.8141, 1.7809, 2.6254], + device='cuda:1'), covar=tensor([0.5168, 0.5746, 0.4325, 0.5349, 0.5008, 0.3109, 0.5318, 0.2056], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0258, 0.0220, 0.0278, 0.0240, 0.0206, 0.0244, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 09:59:09,668 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.225e+02 1.659e+02 1.911e+02 2.326e+02 4.760e+02, threshold=3.822e+02, percent-clipped=1.0 +2023-03-26 09:59:18,770 INFO [finetune.py:976] (1/7) Epoch 8, batch 5250, loss[loss=0.2088, simple_loss=0.2485, pruned_loss=0.08454, over 4239.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2696, pruned_loss=0.07159, over 956973.47 frames. ], batch size: 18, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 09:59:27,650 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45348.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 09:59:28,964 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 10:00:03,246 INFO [finetune.py:976] (1/7) Epoch 8, batch 5300, loss[loss=0.2097, simple_loss=0.2714, pruned_loss=0.07397, over 4816.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2694, pruned_loss=0.07136, over 952066.28 frames. ], batch size: 25, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:00:10,788 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8970, 1.2544, 1.8145, 1.8435, 1.6095, 1.5499, 1.6966, 1.6468], + device='cuda:1'), covar=tensor([0.4552, 0.5859, 0.4709, 0.4843, 0.6441, 0.4801, 0.6286, 0.4269], + device='cuda:1'), in_proj_covar=tensor([0.0231, 0.0240, 0.0253, 0.0254, 0.0246, 0.0222, 0.0272, 0.0227], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:00:11,352 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6157, 1.5118, 2.0268, 3.1810, 2.1316, 2.2706, 1.0876, 2.4938], + device='cuda:1'), covar=tensor([0.1677, 0.1439, 0.1272, 0.0588, 0.0860, 0.1443, 0.1775, 0.0606], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0118, 0.0135, 0.0166, 0.0103, 0.0140, 0.0127, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:00:16,202 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45409.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 10:00:18,467 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45412.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 10:00:29,036 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6415, 1.1496, 0.9521, 1.5389, 1.9585, 1.2610, 1.3905, 1.5617], + device='cuda:1'), covar=tensor([0.1653, 0.2245, 0.2133, 0.1345, 0.2256, 0.2193, 0.1580, 0.2017], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0098, 0.0115, 0.0093, 0.0124, 0.0097, 0.0102, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 10:00:30,690 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.820e+01 1.516e+02 1.858e+02 2.399e+02 4.469e+02, threshold=3.716e+02, percent-clipped=2.0 +2023-03-26 10:00:39,967 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45437.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:00:49,408 INFO [finetune.py:976] (1/7) Epoch 8, batch 5350, loss[loss=0.1898, simple_loss=0.2492, pruned_loss=0.06524, over 4714.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2687, pruned_loss=0.0704, over 953289.79 frames. ], batch size: 23, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:00:53,730 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-03-26 10:01:01,777 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6351, 1.4954, 1.5129, 0.8157, 1.6387, 1.7476, 1.8534, 1.4199], + device='cuda:1'), covar=tensor([0.0749, 0.0622, 0.0402, 0.0556, 0.0397, 0.0423, 0.0279, 0.0567], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0158, 0.0122, 0.0138, 0.0133, 0.0126, 0.0148, 0.0150], + device='cuda:1'), out_proj_covar=tensor([9.7243e-05, 1.1664e-04, 8.8054e-05, 1.0014e-04, 9.5490e-05, 9.2366e-05, + 1.0842e-04, 1.1025e-04], device='cuda:1') +2023-03-26 10:01:54,085 INFO [finetune.py:976] (1/7) Epoch 8, batch 5400, loss[loss=0.1486, simple_loss=0.2226, pruned_loss=0.03726, over 4767.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.267, pruned_loss=0.06984, over 954319.36 frames. ], batch size: 28, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:02:34,313 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.106e+02 1.640e+02 2.019e+02 2.439e+02 4.086e+02, threshold=4.037e+02, percent-clipped=1.0 +2023-03-26 10:02:42,964 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1695, 1.9871, 2.7325, 1.4903, 2.2414, 2.3991, 1.9125, 2.5382], + device='cuda:1'), covar=tensor([0.1605, 0.2161, 0.1494, 0.2651, 0.1006, 0.1838, 0.2786, 0.1056], + device='cuda:1'), in_proj_covar=tensor([0.0202, 0.0206, 0.0197, 0.0197, 0.0183, 0.0221, 0.0220, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:02:54,208 INFO [finetune.py:976] (1/7) Epoch 8, batch 5450, loss[loss=0.1947, simple_loss=0.2557, pruned_loss=0.06684, over 4816.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.264, pruned_loss=0.06885, over 953677.07 frames. ], batch size: 41, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:03:54,237 INFO [finetune.py:976] (1/7) Epoch 8, batch 5500, loss[loss=0.1854, simple_loss=0.2502, pruned_loss=0.06035, over 4764.00 frames. ], tot_loss[loss=0.1977, simple_loss=0.2605, pruned_loss=0.06751, over 952578.07 frames. ], batch size: 26, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:04:15,408 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1912, 2.0051, 1.8561, 2.1617, 2.6874, 2.1253, 1.9861, 1.6792], + device='cuda:1'), covar=tensor([0.2126, 0.2125, 0.1782, 0.1611, 0.1814, 0.1111, 0.2178, 0.1778], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0209, 0.0205, 0.0187, 0.0240, 0.0178, 0.0213, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:04:18,373 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.004e+02 1.550e+02 1.829e+02 2.210e+02 3.729e+02, threshold=3.658e+02, percent-clipped=0.0 +2023-03-26 10:04:28,410 INFO [finetune.py:976] (1/7) Epoch 8, batch 5550, loss[loss=0.1738, simple_loss=0.2438, pruned_loss=0.05191, over 4716.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2619, pruned_loss=0.06785, over 954948.81 frames. ], batch size: 23, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:04:45,258 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-03-26 10:04:47,615 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1765, 1.4110, 1.5119, 0.8135, 1.3737, 1.6312, 1.7298, 1.3307], + device='cuda:1'), covar=tensor([0.0810, 0.0505, 0.0478, 0.0480, 0.0514, 0.0623, 0.0318, 0.0672], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0158, 0.0122, 0.0137, 0.0134, 0.0126, 0.0148, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.7082e-05, 1.1661e-04, 8.8060e-05, 9.9755e-05, 9.5651e-05, 9.2519e-05, + 1.0867e-04, 1.0990e-04], device='cuda:1') +2023-03-26 10:05:19,426 INFO [finetune.py:976] (1/7) Epoch 8, batch 5600, loss[loss=0.1851, simple_loss=0.2512, pruned_loss=0.0595, over 4765.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2668, pruned_loss=0.07006, over 956348.09 frames. ], batch size: 26, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:05:24,125 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45702.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:05:25,255 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45704.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 10:05:29,907 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45712.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 10:05:40,353 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.219e+02 1.739e+02 1.993e+02 2.505e+02 5.014e+02, threshold=3.987e+02, percent-clipped=3.0 +2023-03-26 10:05:41,731 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 10:05:44,490 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45737.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:05:48,901 INFO [finetune.py:976] (1/7) Epoch 8, batch 5650, loss[loss=0.2596, simple_loss=0.3287, pruned_loss=0.09524, over 4845.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2692, pruned_loss=0.0696, over 956335.55 frames. ], batch size: 47, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:05:58,590 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=45760.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:05:59,240 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2939, 2.1520, 1.7132, 2.2527, 2.1050, 1.9048, 2.6216, 2.2788], + device='cuda:1'), covar=tensor([0.1337, 0.2611, 0.3184, 0.3060, 0.2724, 0.1784, 0.3684, 0.1951], + device='cuda:1'), in_proj_covar=tensor([0.0173, 0.0189, 0.0234, 0.0255, 0.0237, 0.0194, 0.0212, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:06:00,371 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45763.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:06:06,759 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45774.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:06:13,161 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=45785.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:06:18,811 INFO [finetune.py:976] (1/7) Epoch 8, batch 5700, loss[loss=0.1421, simple_loss=0.1968, pruned_loss=0.04374, over 4409.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2669, pruned_loss=0.06975, over 939619.64 frames. ], batch size: 19, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:06:54,870 INFO [finetune.py:976] (1/7) Epoch 9, batch 0, loss[loss=0.2191, simple_loss=0.2793, pruned_loss=0.07945, over 4820.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2793, pruned_loss=0.07945, over 4820.00 frames. ], batch size: 39, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:06:54,870 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 10:07:00,327 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8111, 1.6028, 2.1076, 2.7410, 2.0118, 2.2870, 1.1913, 2.2217], + device='cuda:1'), covar=tensor([0.1450, 0.1294, 0.0973, 0.0622, 0.0842, 0.1251, 0.1488, 0.0663], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0118, 0.0135, 0.0166, 0.0103, 0.0140, 0.0127, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:07:11,027 INFO [finetune.py:1010] (1/7) Epoch 9, validation: loss=0.1616, simple_loss=0.233, pruned_loss=0.04515, over 2265189.00 frames. +2023-03-26 10:07:11,028 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 10:07:17,602 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.132e+01 1.600e+02 1.914e+02 2.307e+02 4.538e+02, threshold=3.829e+02, percent-clipped=2.0 +2023-03-26 10:07:22,767 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45835.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:07:33,763 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0297, 1.3239, 0.8666, 1.9463, 2.3373, 1.8027, 1.5575, 1.8628], + device='cuda:1'), covar=tensor([0.1326, 0.1957, 0.2059, 0.1074, 0.1834, 0.2066, 0.1330, 0.1831], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0097, 0.0114, 0.0092, 0.0123, 0.0096, 0.0100, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:07:46,513 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45859.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:07:55,654 INFO [finetune.py:976] (1/7) Epoch 9, batch 50, loss[loss=0.1966, simple_loss=0.2639, pruned_loss=0.0647, over 4865.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2672, pruned_loss=0.07007, over 215657.34 frames. ], batch size: 31, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:08:13,655 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6578, 1.7623, 1.8481, 1.0224, 1.9432, 2.0693, 2.0451, 1.5486], + device='cuda:1'), covar=tensor([0.0941, 0.0641, 0.0469, 0.0688, 0.0388, 0.0514, 0.0348, 0.0673], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0157, 0.0121, 0.0136, 0.0132, 0.0125, 0.0147, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.6434e-05, 1.1582e-04, 8.7317e-05, 9.9096e-05, 9.4817e-05, 9.1873e-05, + 1.0774e-04, 1.0883e-04], device='cuda:1') +2023-03-26 10:08:35,698 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45920.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:08:36,813 INFO [finetune.py:976] (1/7) Epoch 9, batch 100, loss[loss=0.1755, simple_loss=0.2419, pruned_loss=0.0545, over 4860.00 frames. ], tot_loss[loss=0.1971, simple_loss=0.2604, pruned_loss=0.06688, over 380536.87 frames. ], batch size: 34, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:08:42,572 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.088e+02 1.760e+02 2.008e+02 2.423e+02 3.807e+02, threshold=4.016e+02, percent-clipped=0.0 +2023-03-26 10:08:51,773 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.78 vs. limit=5.0 +2023-03-26 10:09:10,374 INFO [finetune.py:976] (1/7) Epoch 9, batch 150, loss[loss=0.1798, simple_loss=0.2511, pruned_loss=0.05428, over 4781.00 frames. ], tot_loss[loss=0.1934, simple_loss=0.2564, pruned_loss=0.06521, over 508804.06 frames. ], batch size: 29, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:09:17,776 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.57 vs. limit=5.0 +2023-03-26 10:09:32,053 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46004.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 10:09:32,098 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9473, 1.5157, 1.9816, 1.8559, 1.6674, 1.6301, 1.7953, 1.7743], + device='cuda:1'), covar=tensor([0.3738, 0.4071, 0.3336, 0.3830, 0.4867, 0.3944, 0.4620, 0.3274], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0242, 0.0254, 0.0255, 0.0248, 0.0224, 0.0275, 0.0230], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:09:49,161 INFO [finetune.py:976] (1/7) Epoch 9, batch 200, loss[loss=0.1929, simple_loss=0.2571, pruned_loss=0.06431, over 4814.00 frames. ], tot_loss[loss=0.1936, simple_loss=0.2559, pruned_loss=0.06564, over 607426.81 frames. ], batch size: 54, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:09:58,660 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.086e+02 1.678e+02 2.056e+02 2.461e+02 4.455e+02, threshold=4.113e+02, percent-clipped=4.0 +2023-03-26 10:10:22,913 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=46052.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 10:10:22,974 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5982, 1.4449, 1.2996, 1.6327, 1.6242, 1.6814, 0.8819, 1.3302], + device='cuda:1'), covar=tensor([0.2466, 0.2379, 0.2136, 0.1880, 0.1775, 0.1228, 0.3018, 0.2014], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0209, 0.0206, 0.0188, 0.0242, 0.0179, 0.0214, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:10:26,559 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46058.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:10:36,486 INFO [finetune.py:976] (1/7) Epoch 9, batch 250, loss[loss=0.2173, simple_loss=0.2896, pruned_loss=0.07247, over 4902.00 frames. ], tot_loss[loss=0.1964, simple_loss=0.2591, pruned_loss=0.06688, over 683699.03 frames. ], batch size: 36, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:11:09,157 INFO [finetune.py:976] (1/7) Epoch 9, batch 300, loss[loss=0.2251, simple_loss=0.2834, pruned_loss=0.08345, over 4933.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2629, pruned_loss=0.06787, over 743514.32 frames. ], batch size: 33, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:11:12,323 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0 +2023-03-26 10:11:14,967 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.304e+02 1.737e+02 2.021e+02 2.354e+02 3.684e+02, threshold=4.042e+02, percent-clipped=0.0 +2023-03-26 10:11:15,062 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46130.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:11:22,168 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9674, 1.9377, 1.9691, 1.4806, 2.1631, 2.1849, 2.1424, 1.6977], + device='cuda:1'), covar=tensor([0.0583, 0.0606, 0.0801, 0.0877, 0.0615, 0.0638, 0.0568, 0.1068], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0132, 0.0144, 0.0124, 0.0116, 0.0144, 0.0144, 0.0159], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:11:22,231 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-26 10:11:41,958 INFO [finetune.py:976] (1/7) Epoch 9, batch 350, loss[loss=0.26, simple_loss=0.307, pruned_loss=0.1064, over 4736.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2687, pruned_loss=0.07144, over 788367.82 frames. ], batch size: 54, lr: 3.81e-03, grad_scale: 16.0 +2023-03-26 10:11:52,843 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8902, 3.6646, 3.4712, 1.8261, 3.7941, 2.9265, 0.7637, 2.6343], + device='cuda:1'), covar=tensor([0.2611, 0.1417, 0.1460, 0.3173, 0.0872, 0.0928, 0.4381, 0.1325], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0172, 0.0160, 0.0128, 0.0156, 0.0122, 0.0145, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 10:12:11,367 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46215.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:12:17,564 INFO [finetune.py:976] (1/7) Epoch 9, batch 400, loss[loss=0.2415, simple_loss=0.3072, pruned_loss=0.08794, over 4906.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2707, pruned_loss=0.07154, over 827529.87 frames. ], batch size: 43, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:12:23,442 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.951e+01 1.676e+02 2.053e+02 2.418e+02 4.627e+02, threshold=4.106e+02, percent-clipped=2.0 +2023-03-26 10:13:00,669 INFO [finetune.py:976] (1/7) Epoch 9, batch 450, loss[loss=0.1893, simple_loss=0.2587, pruned_loss=0.06, over 4806.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2697, pruned_loss=0.07094, over 857883.17 frames. ], batch size: 39, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:13:03,230 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46276.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:13:35,970 INFO [finetune.py:976] (1/7) Epoch 9, batch 500, loss[loss=0.224, simple_loss=0.2853, pruned_loss=0.08139, over 4757.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2667, pruned_loss=0.07002, over 881574.11 frames. ], batch size: 28, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:13:45,335 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.146e+02 1.586e+02 1.900e+02 2.408e+02 3.619e+02, threshold=3.799e+02, percent-clipped=0.0 +2023-03-26 10:13:54,714 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46337.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:13:59,927 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-26 10:14:08,386 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46358.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:14:17,264 INFO [finetune.py:976] (1/7) Epoch 9, batch 550, loss[loss=0.2259, simple_loss=0.2675, pruned_loss=0.09216, over 4907.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2634, pruned_loss=0.06895, over 896845.46 frames. ], batch size: 36, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:14:39,980 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=46406.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:14:50,106 INFO [finetune.py:976] (1/7) Epoch 9, batch 600, loss[loss=0.2206, simple_loss=0.2718, pruned_loss=0.08469, over 4902.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2631, pruned_loss=0.0686, over 909752.93 frames. ], batch size: 43, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:14:54,848 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.149e+02 1.702e+02 1.969e+02 2.390e+02 4.680e+02, threshold=3.938e+02, percent-clipped=3.0 +2023-03-26 10:14:54,950 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46430.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:15:36,213 INFO [finetune.py:976] (1/7) Epoch 9, batch 650, loss[loss=0.2595, simple_loss=0.3024, pruned_loss=0.1083, over 4749.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2661, pruned_loss=0.06977, over 921045.57 frames. ], batch size: 59, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:15:40,464 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=46478.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:15:51,889 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 10:16:05,400 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46515.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:16:09,545 INFO [finetune.py:976] (1/7) Epoch 9, batch 700, loss[loss=0.2017, simple_loss=0.2603, pruned_loss=0.07158, over 4882.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2674, pruned_loss=0.0699, over 927770.84 frames. ], batch size: 32, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:16:14,889 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.046e+02 1.666e+02 2.010e+02 2.529e+02 4.289e+02, threshold=4.019e+02, percent-clipped=2.0 +2023-03-26 10:16:26,203 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3554, 1.3489, 1.2756, 1.3856, 1.7102, 1.5875, 1.5466, 1.2000], + device='cuda:1'), covar=tensor([0.0330, 0.0291, 0.0627, 0.0324, 0.0224, 0.0458, 0.0259, 0.0372], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0110, 0.0139, 0.0115, 0.0102, 0.0101, 0.0091, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.9912e-05, 8.6186e-05, 1.1125e-04, 9.0722e-05, 8.0368e-05, 7.5266e-05, + 6.8636e-05, 8.3810e-05], device='cuda:1') +2023-03-26 10:16:37,472 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=46563.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:16:42,850 INFO [finetune.py:976] (1/7) Epoch 9, batch 750, loss[loss=0.2218, simple_loss=0.2961, pruned_loss=0.07378, over 4910.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2694, pruned_loss=0.06997, over 933718.72 frames. ], batch size: 37, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:16:58,345 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46595.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:17:03,699 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46602.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:17:15,990 INFO [finetune.py:976] (1/7) Epoch 9, batch 800, loss[loss=0.2047, simple_loss=0.274, pruned_loss=0.06772, over 4754.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2679, pruned_loss=0.06915, over 938513.99 frames. ], batch size: 28, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:17:20,814 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.216e+01 1.561e+02 1.863e+02 2.153e+02 3.377e+02, threshold=3.726e+02, percent-clipped=0.0 +2023-03-26 10:17:22,501 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46632.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:17:39,093 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46656.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 10:17:40,237 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 10:17:43,773 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46663.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:17:49,097 INFO [finetune.py:976] (1/7) Epoch 9, batch 850, loss[loss=0.2014, simple_loss=0.2559, pruned_loss=0.0734, over 4896.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2652, pruned_loss=0.06771, over 944137.15 frames. ], batch size: 35, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:18:34,822 INFO [finetune.py:976] (1/7) Epoch 9, batch 900, loss[loss=0.1921, simple_loss=0.2521, pruned_loss=0.06602, over 4899.00 frames. ], tot_loss[loss=0.1974, simple_loss=0.2616, pruned_loss=0.06658, over 947900.76 frames. ], batch size: 32, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:18:39,658 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.365e+01 1.501e+02 1.768e+02 2.130e+02 3.855e+02, threshold=3.537e+02, percent-clipped=1.0 +2023-03-26 10:19:05,942 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7302, 1.0899, 0.8817, 1.6570, 2.1393, 1.5616, 1.5037, 1.6441], + device='cuda:1'), covar=tensor([0.2117, 0.3337, 0.2719, 0.1758, 0.2535, 0.2797, 0.2241, 0.2975], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0114, 0.0092, 0.0123, 0.0096, 0.0100, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:19:09,052 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4312, 1.4969, 1.7296, 1.7331, 1.5068, 3.3382, 1.3029, 1.5691], + device='cuda:1'), covar=tensor([0.0993, 0.1837, 0.1360, 0.1059, 0.1667, 0.0239, 0.1554, 0.1828], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0076, 0.0079, 0.0093, 0.0084, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 10:19:10,157 INFO [finetune.py:976] (1/7) Epoch 9, batch 950, loss[loss=0.2222, simple_loss=0.2778, pruned_loss=0.08324, over 4912.00 frames. ], tot_loss[loss=0.1976, simple_loss=0.2612, pruned_loss=0.06699, over 950124.39 frames. ], batch size: 36, lr: 3.80e-03, grad_scale: 32.0 +2023-03-26 10:19:33,606 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1656, 1.9941, 2.1944, 0.9211, 2.4161, 2.6159, 2.1057, 1.8514], + device='cuda:1'), covar=tensor([0.1154, 0.0910, 0.0577, 0.0893, 0.0643, 0.0793, 0.0670, 0.0880], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0156, 0.0121, 0.0135, 0.0131, 0.0125, 0.0145, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.5887e-05, 1.1509e-04, 8.7053e-05, 9.7972e-05, 9.4031e-05, 9.1729e-05, + 1.0676e-04, 1.0816e-04], device='cuda:1') +2023-03-26 10:19:36,021 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46810.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:19:44,264 INFO [finetune.py:976] (1/7) Epoch 9, batch 1000, loss[loss=0.198, simple_loss=0.2712, pruned_loss=0.0624, over 4817.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.2632, pruned_loss=0.06761, over 952625.42 frames. ], batch size: 51, lr: 3.80e-03, grad_scale: 32.0 +2023-03-26 10:19:49,056 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.128e+02 1.690e+02 1.992e+02 2.479e+02 5.334e+02, threshold=3.983e+02, percent-clipped=2.0 +2023-03-26 10:20:22,582 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46871.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:20:23,062 INFO [finetune.py:976] (1/7) Epoch 9, batch 1050, loss[loss=0.2249, simple_loss=0.2931, pruned_loss=0.07838, over 4904.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2651, pruned_loss=0.0679, over 951816.30 frames. ], batch size: 43, lr: 3.80e-03, grad_scale: 32.0 +2023-03-26 10:21:05,008 INFO [finetune.py:976] (1/7) Epoch 9, batch 1100, loss[loss=0.228, simple_loss=0.2913, pruned_loss=0.08235, over 4928.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2682, pruned_loss=0.06981, over 953174.45 frames. ], batch size: 38, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:21:10,488 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.274e+02 1.654e+02 2.044e+02 2.534e+02 3.814e+02, threshold=4.089e+02, percent-clipped=0.0 +2023-03-26 10:21:11,167 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46932.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:21:23,028 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46951.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 10:21:28,257 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46958.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:21:37,702 INFO [finetune.py:976] (1/7) Epoch 9, batch 1150, loss[loss=0.1862, simple_loss=0.2469, pruned_loss=0.06279, over 4291.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2685, pruned_loss=0.06987, over 951968.95 frames. ], batch size: 66, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:21:42,585 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=46980.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:22:08,407 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1961, 3.6059, 3.8266, 3.9628, 3.9681, 3.7714, 4.2628, 1.5028], + device='cuda:1'), covar=tensor([0.0732, 0.0739, 0.0756, 0.1001, 0.1038, 0.1198, 0.0655, 0.4807], + device='cuda:1'), in_proj_covar=tensor([0.0345, 0.0241, 0.0273, 0.0291, 0.0329, 0.0280, 0.0298, 0.0292], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:22:10,618 INFO [finetune.py:976] (1/7) Epoch 9, batch 1200, loss[loss=0.1911, simple_loss=0.2588, pruned_loss=0.06173, over 4907.00 frames. ], tot_loss[loss=0.2024, simple_loss=0.2668, pruned_loss=0.06903, over 952507.78 frames. ], batch size: 37, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:22:16,138 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.814e+01 1.625e+02 2.018e+02 2.406e+02 3.989e+02, threshold=4.036e+02, percent-clipped=0.0 +2023-03-26 10:22:20,587 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1357, 1.9776, 1.6394, 1.8559, 2.0888, 1.7599, 2.4243, 2.0854], + device='cuda:1'), covar=tensor([0.1363, 0.2448, 0.3389, 0.2932, 0.2666, 0.1746, 0.2890, 0.1886], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0188, 0.0232, 0.0253, 0.0236, 0.0193, 0.0211, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:22:27,364 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 10:22:43,585 INFO [finetune.py:976] (1/7) Epoch 9, batch 1250, loss[loss=0.1983, simple_loss=0.2611, pruned_loss=0.06774, over 4890.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.263, pruned_loss=0.06771, over 952467.92 frames. ], batch size: 32, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:22:57,425 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.2864, 4.4880, 4.7549, 5.0637, 4.9961, 4.7509, 5.3630, 1.6654], + device='cuda:1'), covar=tensor([0.0678, 0.0862, 0.0685, 0.0824, 0.1115, 0.1420, 0.0503, 0.5489], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0241, 0.0274, 0.0291, 0.0330, 0.0280, 0.0299, 0.0292], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:23:08,186 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8215, 1.1579, 0.7371, 1.7933, 2.1852, 1.5460, 1.6427, 1.5606], + device='cuda:1'), covar=tensor([0.1565, 0.2230, 0.2297, 0.1174, 0.2061, 0.2113, 0.1463, 0.2076], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0096, 0.0113, 0.0092, 0.0122, 0.0096, 0.0100, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:23:21,438 INFO [finetune.py:976] (1/7) Epoch 9, batch 1300, loss[loss=0.2193, simple_loss=0.2709, pruned_loss=0.0839, over 4815.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2609, pruned_loss=0.06767, over 953289.44 frames. ], batch size: 51, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:23:31,783 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.169e+01 1.658e+02 1.956e+02 2.404e+02 5.414e+02, threshold=3.912e+02, percent-clipped=2.0 +2023-03-26 10:23:57,901 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47166.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:24:03,031 INFO [finetune.py:976] (1/7) Epoch 9, batch 1350, loss[loss=0.2604, simple_loss=0.3302, pruned_loss=0.09528, over 4764.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2625, pruned_loss=0.06868, over 951657.47 frames. ], batch size: 54, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:24:36,401 INFO [finetune.py:976] (1/7) Epoch 9, batch 1400, loss[loss=0.2154, simple_loss=0.3052, pruned_loss=0.06276, over 4803.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.266, pruned_loss=0.06957, over 954020.64 frames. ], batch size: 45, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:24:42,787 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.151e+02 1.739e+02 2.002e+02 2.347e+02 4.228e+02, threshold=4.005e+02, percent-clipped=2.0 +2023-03-26 10:24:43,502 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5211, 1.5075, 1.6325, 1.7613, 1.5971, 3.2178, 1.3724, 1.6764], + device='cuda:1'), covar=tensor([0.0976, 0.1705, 0.1204, 0.0957, 0.1544, 0.0262, 0.1462, 0.1588], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0076, 0.0079, 0.0092, 0.0084, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 10:24:46,764 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-03-26 10:24:49,011 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47241.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:24:55,110 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47251.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:24:59,390 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47258.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:25:09,254 INFO [finetune.py:976] (1/7) Epoch 9, batch 1450, loss[loss=0.2075, simple_loss=0.2798, pruned_loss=0.06761, over 4821.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.268, pruned_loss=0.06983, over 954058.89 frames. ], batch size: 39, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:25:25,250 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 10:25:27,569 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=47299.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:25:29,419 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47302.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:25:36,195 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=47306.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:25:36,840 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9347, 1.5678, 2.3857, 3.8583, 2.5979, 2.6404, 0.9412, 3.1078], + device='cuda:1'), covar=tensor([0.1798, 0.1695, 0.1428, 0.0584, 0.0828, 0.1668, 0.1973, 0.0502], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0118, 0.0135, 0.0167, 0.0103, 0.0139, 0.0127, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:25:45,539 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.84 vs. limit=5.0 +2023-03-26 10:25:53,868 INFO [finetune.py:976] (1/7) Epoch 9, batch 1500, loss[loss=0.2116, simple_loss=0.2806, pruned_loss=0.07132, over 4923.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2679, pruned_loss=0.07009, over 953031.41 frames. ], batch size: 38, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:26:00,802 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.103e+02 1.648e+02 1.983e+02 2.305e+02 4.092e+02, threshold=3.967e+02, percent-clipped=1.0 +2023-03-26 10:26:25,123 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9671, 1.8060, 1.8185, 1.9263, 1.4810, 4.4822, 1.8377, 2.3672], + device='cuda:1'), covar=tensor([0.3176, 0.2362, 0.1902, 0.2215, 0.1643, 0.0085, 0.2325, 0.1168], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0119, 0.0123, 0.0116, 0.0098, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 10:26:26,875 INFO [finetune.py:976] (1/7) Epoch 9, batch 1550, loss[loss=0.1854, simple_loss=0.2477, pruned_loss=0.06161, over 4804.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2673, pruned_loss=0.06906, over 955897.38 frames. ], batch size: 45, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:26:33,201 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47379.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:26:50,608 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 10:26:51,164 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8454, 1.9589, 1.9503, 1.2989, 2.0371, 2.0678, 2.0086, 1.7400], + device='cuda:1'), covar=tensor([0.0734, 0.0643, 0.0771, 0.0931, 0.0627, 0.0718, 0.0672, 0.1065], + device='cuda:1'), in_proj_covar=tensor([0.0137, 0.0132, 0.0145, 0.0125, 0.0117, 0.0145, 0.0145, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:26:51,386 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0 +2023-03-26 10:27:00,839 INFO [finetune.py:976] (1/7) Epoch 9, batch 1600, loss[loss=0.155, simple_loss=0.215, pruned_loss=0.0475, over 4805.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.265, pruned_loss=0.0681, over 956538.91 frames. ], batch size: 25, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:27:06,346 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-03-26 10:27:07,814 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.155e+02 1.632e+02 1.991e+02 2.321e+02 5.028e+02, threshold=3.982e+02, percent-clipped=1.0 +2023-03-26 10:27:14,434 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47440.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:27:15,528 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5491, 2.2829, 1.9126, 2.4517, 2.3560, 2.0970, 2.8876, 2.5022], + device='cuda:1'), covar=tensor([0.1351, 0.2637, 0.3420, 0.3379, 0.2932, 0.1783, 0.4360, 0.1944], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0188, 0.0233, 0.0253, 0.0237, 0.0194, 0.0211, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:27:30,687 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47466.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:27:34,229 INFO [finetune.py:976] (1/7) Epoch 9, batch 1650, loss[loss=0.1465, simple_loss=0.2111, pruned_loss=0.04096, over 4721.00 frames. ], tot_loss[loss=0.1967, simple_loss=0.2609, pruned_loss=0.06624, over 958056.87 frames. ], batch size: 23, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:27:50,628 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-26 10:28:02,971 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=47514.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:28:07,770 INFO [finetune.py:976] (1/7) Epoch 9, batch 1700, loss[loss=0.1386, simple_loss=0.2115, pruned_loss=0.03287, over 4797.00 frames. ], tot_loss[loss=0.1967, simple_loss=0.2604, pruned_loss=0.0665, over 957914.15 frames. ], batch size: 25, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:28:13,224 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.017e+02 1.617e+02 1.864e+02 2.209e+02 5.015e+02, threshold=3.728e+02, percent-clipped=2.0 +2023-03-26 10:28:17,983 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2605, 2.0421, 2.1919, 1.0338, 2.2621, 2.6229, 2.1795, 2.0181], + device='cuda:1'), covar=tensor([0.1081, 0.0747, 0.0484, 0.0684, 0.0648, 0.0594, 0.0458, 0.0609], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0156, 0.0121, 0.0135, 0.0132, 0.0125, 0.0146, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.5666e-05, 1.1515e-04, 8.7450e-05, 9.8036e-05, 9.4347e-05, 9.1648e-05, + 1.0683e-04, 1.0838e-04], device='cuda:1') +2023-03-26 10:28:47,731 INFO [finetune.py:976] (1/7) Epoch 9, batch 1750, loss[loss=0.2168, simple_loss=0.2878, pruned_loss=0.07293, over 4762.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2638, pruned_loss=0.06828, over 955755.53 frames. ], batch size: 54, lr: 3.80e-03, grad_scale: 16.0 +2023-03-26 10:29:13,706 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47597.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:29:29,376 INFO [finetune.py:976] (1/7) Epoch 9, batch 1800, loss[loss=0.1969, simple_loss=0.2638, pruned_loss=0.06506, over 4907.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2677, pruned_loss=0.06949, over 957281.80 frames. ], batch size: 36, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:29:34,861 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.172e+02 1.736e+02 1.994e+02 2.573e+02 6.193e+02, threshold=3.989e+02, percent-clipped=4.0 +2023-03-26 10:30:02,647 INFO [finetune.py:976] (1/7) Epoch 9, batch 1850, loss[loss=0.1826, simple_loss=0.2431, pruned_loss=0.06106, over 4704.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2693, pruned_loss=0.07034, over 955753.43 frames. ], batch size: 23, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:30:16,483 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6562, 1.5527, 1.5490, 1.7215, 1.0730, 3.6660, 1.3777, 1.9181], + device='cuda:1'), covar=tensor([0.3398, 0.2421, 0.2107, 0.2258, 0.2018, 0.0163, 0.2555, 0.1332], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0116, 0.0120, 0.0124, 0.0117, 0.0099, 0.0101, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 10:30:30,647 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 10:30:35,893 INFO [finetune.py:976] (1/7) Epoch 9, batch 1900, loss[loss=0.1826, simple_loss=0.2574, pruned_loss=0.05386, over 4746.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2703, pruned_loss=0.07031, over 954977.59 frames. ], batch size: 27, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:30:46,260 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.162e+02 1.608e+02 1.835e+02 2.236e+02 3.803e+02, threshold=3.670e+02, percent-clipped=0.0 +2023-03-26 10:30:53,006 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47735.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:30:54,870 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47738.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:31:06,464 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2811, 2.1730, 2.2467, 0.8424, 2.5185, 2.7643, 2.1953, 2.0998], + device='cuda:1'), covar=tensor([0.0993, 0.0707, 0.0525, 0.0804, 0.0455, 0.0629, 0.0485, 0.0704], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0158, 0.0123, 0.0136, 0.0133, 0.0127, 0.0147, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.6849e-05, 1.1624e-04, 8.8900e-05, 9.9003e-05, 9.5050e-05, 9.2843e-05, + 1.0797e-04, 1.0963e-04], device='cuda:1') +2023-03-26 10:31:21,754 INFO [finetune.py:976] (1/7) Epoch 9, batch 1950, loss[loss=0.2147, simple_loss=0.2649, pruned_loss=0.0823, over 4855.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2681, pruned_loss=0.06934, over 956833.72 frames. ], batch size: 31, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:31:31,520 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47788.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:31:36,903 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.76 vs. limit=2.0 +2023-03-26 10:31:39,143 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47799.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:31:45,551 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.25 vs. limit=5.0 +2023-03-26 10:31:46,707 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47809.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 10:31:55,008 INFO [finetune.py:976] (1/7) Epoch 9, batch 2000, loss[loss=0.1807, simple_loss=0.2507, pruned_loss=0.05532, over 4828.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2654, pruned_loss=0.06887, over 956973.20 frames. ], batch size: 40, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:32:00,455 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.198e+02 1.527e+02 1.824e+02 2.186e+02 3.277e+02, threshold=3.648e+02, percent-clipped=0.0 +2023-03-26 10:32:11,892 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47849.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:32:35,772 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47870.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 10:32:36,819 INFO [finetune.py:976] (1/7) Epoch 9, batch 2050, loss[loss=0.2282, simple_loss=0.2831, pruned_loss=0.08665, over 4822.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2622, pruned_loss=0.06781, over 956117.82 frames. ], batch size: 30, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:32:57,813 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47897.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:33:02,158 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-26 10:33:15,822 INFO [finetune.py:976] (1/7) Epoch 9, batch 2100, loss[loss=0.1767, simple_loss=0.2555, pruned_loss=0.04897, over 4853.00 frames. ], tot_loss[loss=0.199, simple_loss=0.2623, pruned_loss=0.06786, over 955020.13 frames. ], batch size: 49, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:33:21,286 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.115e+02 1.604e+02 1.917e+02 2.370e+02 5.169e+02, threshold=3.834e+02, percent-clipped=3.0 +2023-03-26 10:33:29,783 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=47945.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:33:34,095 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7923, 1.5081, 1.3996, 1.2619, 1.4883, 1.5221, 1.5274, 2.1091], + device='cuda:1'), covar=tensor([0.5036, 0.5130, 0.3923, 0.4431, 0.4535, 0.2830, 0.4446, 0.2245], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0260, 0.0222, 0.0280, 0.0242, 0.0208, 0.0244, 0.0209], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:33:54,987 INFO [finetune.py:976] (1/7) Epoch 9, batch 2150, loss[loss=0.1902, simple_loss=0.2661, pruned_loss=0.05712, over 4738.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2645, pruned_loss=0.06876, over 955904.32 frames. ], batch size: 54, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:34:04,023 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6422, 1.3994, 1.6810, 1.9288, 1.5347, 3.2714, 1.2764, 1.6318], + device='cuda:1'), covar=tensor([0.0925, 0.1861, 0.1284, 0.0939, 0.1688, 0.0254, 0.1672, 0.1795], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 10:34:37,713 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5811, 1.4987, 1.8776, 1.9707, 1.5472, 3.5938, 1.4134, 1.6974], + device='cuda:1'), covar=tensor([0.1092, 0.1990, 0.1169, 0.1049, 0.1859, 0.0311, 0.1735, 0.1845], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0083, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 10:35:00,856 INFO [finetune.py:976] (1/7) Epoch 9, batch 2200, loss[loss=0.1916, simple_loss=0.2574, pruned_loss=0.06288, over 4871.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2679, pruned_loss=0.06979, over 956178.65 frames. ], batch size: 34, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:35:11,848 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.338e+02 1.754e+02 2.052e+02 2.528e+02 4.321e+02, threshold=4.105e+02, percent-clipped=1.0 +2023-03-26 10:35:18,710 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48035.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:35:27,969 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 10:36:02,997 INFO [finetune.py:976] (1/7) Epoch 9, batch 2250, loss[loss=0.2202, simple_loss=0.2828, pruned_loss=0.07876, over 4817.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2681, pruned_loss=0.06929, over 957500.94 frames. ], batch size: 25, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:36:03,123 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48072.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:36:20,662 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=48083.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:36:32,301 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48094.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:36:32,915 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6435, 1.6879, 2.0955, 2.0500, 1.9094, 4.4038, 1.7415, 2.1611], + device='cuda:1'), covar=tensor([0.0993, 0.1734, 0.1088, 0.0988, 0.1537, 0.0166, 0.1394, 0.1522], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0082, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 10:36:41,881 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2842, 2.2094, 2.1134, 2.4774, 3.0073, 2.4651, 2.2527, 1.8291], + device='cuda:1'), covar=tensor([0.2278, 0.1994, 0.1732, 0.1606, 0.1603, 0.0988, 0.2160, 0.1786], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0209, 0.0207, 0.0188, 0.0242, 0.0180, 0.0216, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:37:05,451 INFO [finetune.py:976] (1/7) Epoch 9, batch 2300, loss[loss=0.2043, simple_loss=0.2696, pruned_loss=0.06947, over 4882.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2679, pruned_loss=0.06891, over 957443.10 frames. ], batch size: 35, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:37:16,652 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.508e+01 1.652e+02 1.874e+02 2.360e+02 5.580e+02, threshold=3.748e+02, percent-clipped=1.0 +2023-03-26 10:37:23,199 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48133.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:37:34,216 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48144.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:37:54,849 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48165.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 10:38:01,013 INFO [finetune.py:976] (1/7) Epoch 9, batch 2350, loss[loss=0.2162, simple_loss=0.2757, pruned_loss=0.0784, over 4738.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2655, pruned_loss=0.06796, over 957435.32 frames. ], batch size: 59, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:38:05,717 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7310, 0.7105, 1.7018, 1.5209, 1.4775, 1.4121, 1.4095, 1.5624], + device='cuda:1'), covar=tensor([0.3956, 0.5038, 0.4503, 0.4456, 0.5774, 0.4271, 0.5583, 0.4082], + device='cuda:1'), in_proj_covar=tensor([0.0232, 0.0240, 0.0253, 0.0256, 0.0249, 0.0224, 0.0273, 0.0229], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:38:11,782 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8593, 1.3113, 1.9000, 1.7244, 1.5585, 1.5506, 1.6739, 1.7067], + device='cuda:1'), covar=tensor([0.3758, 0.4551, 0.3599, 0.4086, 0.5242, 0.4053, 0.4846, 0.3700], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0240, 0.0253, 0.0256, 0.0249, 0.0224, 0.0273, 0.0229], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:38:35,577 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48212.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 10:38:42,167 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 10:38:43,033 INFO [finetune.py:976] (1/7) Epoch 9, batch 2400, loss[loss=0.1929, simple_loss=0.2545, pruned_loss=0.0657, over 4939.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2627, pruned_loss=0.06722, over 957297.88 frames. ], batch size: 33, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:38:49,467 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.159e+02 1.602e+02 2.015e+02 2.421e+02 3.465e+02, threshold=4.031e+02, percent-clipped=0.0 +2023-03-26 10:39:18,902 INFO [finetune.py:976] (1/7) Epoch 9, batch 2450, loss[loss=0.2249, simple_loss=0.2871, pruned_loss=0.08133, over 4918.00 frames. ], tot_loss[loss=0.1959, simple_loss=0.2593, pruned_loss=0.06627, over 956710.61 frames. ], batch size: 37, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:39:19,628 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48273.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 10:39:31,826 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3585, 1.5917, 1.7054, 0.9172, 1.5762, 1.8307, 1.8830, 1.4749], + device='cuda:1'), covar=tensor([0.0871, 0.0590, 0.0417, 0.0505, 0.0412, 0.0538, 0.0345, 0.0652], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0156, 0.0121, 0.0135, 0.0132, 0.0126, 0.0146, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.6064e-05, 1.1486e-04, 8.7502e-05, 9.8045e-05, 9.4115e-05, 9.2386e-05, + 1.0737e-04, 1.0829e-04], device='cuda:1') +2023-03-26 10:39:44,216 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-26 10:39:54,551 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7776, 1.6998, 1.4157, 1.4647, 1.6048, 1.5803, 1.6382, 2.2788], + device='cuda:1'), covar=tensor([0.4773, 0.4616, 0.3788, 0.4678, 0.4370, 0.2706, 0.4073, 0.1881], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0260, 0.0222, 0.0280, 0.0243, 0.0209, 0.0245, 0.0210], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:40:01,691 INFO [finetune.py:976] (1/7) Epoch 9, batch 2500, loss[loss=0.1991, simple_loss=0.2792, pruned_loss=0.05952, over 4815.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2627, pruned_loss=0.0685, over 955281.07 frames. ], batch size: 38, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:40:03,384 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48324.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:40:09,017 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.093e+02 1.681e+02 1.962e+02 2.420e+02 5.026e+02, threshold=3.923e+02, percent-clipped=2.0 +2023-03-26 10:40:35,356 INFO [finetune.py:976] (1/7) Epoch 9, batch 2550, loss[loss=0.2041, simple_loss=0.272, pruned_loss=0.06812, over 4820.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2661, pruned_loss=0.06959, over 954913.93 frames. ], batch size: 38, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:40:45,872 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48385.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:40:51,848 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48394.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:40:52,504 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6399, 1.6577, 1.4309, 1.8166, 2.0467, 1.7402, 1.2556, 1.3416], + device='cuda:1'), covar=tensor([0.2490, 0.2225, 0.2131, 0.1867, 0.1989, 0.1412, 0.3006, 0.2261], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0207, 0.0204, 0.0186, 0.0239, 0.0178, 0.0212, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:40:53,649 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48397.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:41:08,895 INFO [finetune.py:976] (1/7) Epoch 9, batch 2600, loss[loss=0.2036, simple_loss=0.2742, pruned_loss=0.06647, over 4843.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.268, pruned_loss=0.06983, over 955225.06 frames. ], batch size: 44, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:41:09,603 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48423.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:41:12,601 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48428.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:41:15,190 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.166e+02 1.754e+02 2.041e+02 2.553e+02 4.015e+02, threshold=4.083e+02, percent-clipped=1.0 +2023-03-26 10:41:23,748 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=48442.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:41:25,435 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48444.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:41:31,558 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0457, 4.5271, 4.2886, 2.5608, 4.5882, 3.4142, 0.7978, 3.1826], + device='cuda:1'), covar=tensor([0.2116, 0.1249, 0.1143, 0.2441, 0.0597, 0.0878, 0.4310, 0.1077], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0173, 0.0159, 0.0128, 0.0156, 0.0122, 0.0147, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 10:41:34,014 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48458.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:41:35,243 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1067, 1.7918, 1.6006, 1.6691, 1.7209, 1.7525, 1.7476, 2.5223], + device='cuda:1'), covar=tensor([0.4777, 0.5711, 0.4149, 0.5205, 0.5232, 0.3012, 0.5061, 0.2036], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0259, 0.0221, 0.0279, 0.0242, 0.0208, 0.0245, 0.0209], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:41:37,652 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9917, 1.9102, 1.9688, 1.4404, 2.0532, 2.1551, 2.1438, 1.5892], + device='cuda:1'), covar=tensor([0.0546, 0.0544, 0.0697, 0.0875, 0.0609, 0.0645, 0.0551, 0.1103], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0132, 0.0144, 0.0124, 0.0117, 0.0144, 0.0144, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:41:38,226 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48465.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 10:41:42,379 INFO [finetune.py:976] (1/7) Epoch 9, batch 2650, loss[loss=0.2111, simple_loss=0.2658, pruned_loss=0.07822, over 4118.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2692, pruned_loss=0.07038, over 953474.05 frames. ], batch size: 65, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:41:56,000 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48484.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:42:06,173 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=48492.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:42:19,478 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=48513.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 10:42:24,896 INFO [finetune.py:976] (1/7) Epoch 9, batch 2700, loss[loss=0.1978, simple_loss=0.2677, pruned_loss=0.06394, over 4810.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2678, pruned_loss=0.06954, over 954301.17 frames. ], batch size: 45, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:42:27,421 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8650, 1.6420, 1.3967, 1.4049, 1.5812, 1.5468, 1.6204, 2.2783], + device='cuda:1'), covar=tensor([0.4851, 0.5149, 0.4027, 0.4848, 0.4469, 0.3115, 0.4719, 0.2026], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0260, 0.0222, 0.0279, 0.0243, 0.0209, 0.0245, 0.0210], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:42:30,333 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.052e+02 1.590e+02 1.873e+02 2.487e+02 4.580e+02, threshold=3.745e+02, percent-clipped=2.0 +2023-03-26 10:43:07,958 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48568.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 10:43:10,318 INFO [finetune.py:976] (1/7) Epoch 9, batch 2750, loss[loss=0.1792, simple_loss=0.24, pruned_loss=0.0592, over 4936.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2648, pruned_loss=0.06809, over 956091.30 frames. ], batch size: 33, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:43:45,668 INFO [finetune.py:976] (1/7) Epoch 9, batch 2800, loss[loss=0.1667, simple_loss=0.2433, pruned_loss=0.04508, over 4921.00 frames. ], tot_loss[loss=0.1973, simple_loss=0.2608, pruned_loss=0.06687, over 955546.15 frames. ], batch size: 43, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:43:51,108 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.018e+02 1.496e+02 1.797e+02 2.211e+02 4.995e+02, threshold=3.593e+02, percent-clipped=1.0 +2023-03-26 10:44:14,938 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.1443, 4.4276, 4.7009, 4.8990, 4.8189, 4.6156, 5.2516, 1.6002], + device='cuda:1'), covar=tensor([0.0785, 0.0912, 0.0705, 0.0925, 0.1281, 0.1440, 0.0590, 0.5550], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0246, 0.0277, 0.0294, 0.0334, 0.0283, 0.0303, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:44:19,114 INFO [finetune.py:976] (1/7) Epoch 9, batch 2850, loss[loss=0.1385, simple_loss=0.21, pruned_loss=0.03352, over 4832.00 frames. ], tot_loss[loss=0.1952, simple_loss=0.259, pruned_loss=0.06566, over 958672.58 frames. ], batch size: 25, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:44:24,040 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48680.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:44:48,775 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7856, 3.8132, 3.6367, 1.8227, 3.9281, 2.8760, 0.6757, 2.6333], + device='cuda:1'), covar=tensor([0.2853, 0.2237, 0.1633, 0.3726, 0.1015, 0.1128, 0.5164, 0.1633], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0173, 0.0160, 0.0128, 0.0156, 0.0122, 0.0147, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 10:44:49,488 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.48 vs. limit=5.0 +2023-03-26 10:45:04,572 INFO [finetune.py:976] (1/7) Epoch 9, batch 2900, loss[loss=0.2475, simple_loss=0.3053, pruned_loss=0.09484, over 4922.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.2633, pruned_loss=0.06787, over 956869.53 frames. ], batch size: 38, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:45:08,312 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48728.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:45:10,029 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.274e+02 1.660e+02 1.926e+02 2.355e+02 4.281e+02, threshold=3.853e+02, percent-clipped=2.0 +2023-03-26 10:45:19,335 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.28 vs. limit=5.0 +2023-03-26 10:45:25,492 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48753.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:45:35,647 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.56 vs. limit=2.0 +2023-03-26 10:45:38,483 INFO [finetune.py:976] (1/7) Epoch 9, batch 2950, loss[loss=0.1792, simple_loss=0.2379, pruned_loss=0.06023, over 4760.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2668, pruned_loss=0.06869, over 955515.65 frames. ], batch size: 23, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:45:40,981 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=48776.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:45:42,828 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48779.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:45:44,060 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7463, 1.5240, 2.0714, 3.5153, 2.3898, 2.3833, 0.8930, 2.7608], + device='cuda:1'), covar=tensor([0.1804, 0.1674, 0.1441, 0.0602, 0.0795, 0.1450, 0.2034, 0.0586], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0133, 0.0164, 0.0101, 0.0138, 0.0125, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:46:03,978 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 10:46:10,722 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8025, 1.4994, 2.3486, 1.3536, 1.9166, 2.0998, 1.3761, 2.1880], + device='cuda:1'), covar=tensor([0.1779, 0.2405, 0.1197, 0.2343, 0.1142, 0.1508, 0.3135, 0.1409], + device='cuda:1'), in_proj_covar=tensor([0.0199, 0.0205, 0.0194, 0.0192, 0.0180, 0.0218, 0.0217, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:46:11,320 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48821.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:46:11,798 INFO [finetune.py:976] (1/7) Epoch 9, batch 3000, loss[loss=0.2515, simple_loss=0.3085, pruned_loss=0.09728, over 4847.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2689, pruned_loss=0.06984, over 955348.42 frames. ], batch size: 44, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:46:11,798 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 10:46:22,396 INFO [finetune.py:1010] (1/7) Epoch 9, validation: loss=0.159, simple_loss=0.2302, pruned_loss=0.04393, over 2265189.00 frames. +2023-03-26 10:46:22,396 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 10:46:27,908 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.631e+02 1.894e+02 2.277e+02 3.777e+02, threshold=3.789e+02, percent-clipped=0.0 +2023-03-26 10:46:32,261 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48838.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:46:51,717 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48868.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 10:46:54,486 INFO [finetune.py:976] (1/7) Epoch 9, batch 3050, loss[loss=0.18, simple_loss=0.2594, pruned_loss=0.05025, over 4722.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2691, pruned_loss=0.0696, over 955232.14 frames. ], batch size: 59, lr: 3.79e-03, grad_scale: 16.0 +2023-03-26 10:46:54,616 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9551, 1.8045, 1.5554, 1.8318, 1.7319, 1.7449, 1.7599, 2.4561], + device='cuda:1'), covar=tensor([0.4966, 0.5473, 0.3872, 0.4847, 0.4744, 0.2935, 0.5209, 0.1900], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0258, 0.0220, 0.0279, 0.0241, 0.0208, 0.0245, 0.0208], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:47:03,398 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48882.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:47:05,848 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8141, 2.5706, 2.0711, 2.4288, 2.4876, 2.3173, 3.0024, 2.7806], + device='cuda:1'), covar=tensor([0.1190, 0.2334, 0.3242, 0.3279, 0.2830, 0.1614, 0.3928, 0.1797], + device='cuda:1'), in_proj_covar=tensor([0.0174, 0.0189, 0.0235, 0.0255, 0.0239, 0.0195, 0.0213, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:47:13,620 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48899.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:47:24,762 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=48916.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 10:47:29,281 INFO [finetune.py:976] (1/7) Epoch 9, batch 3100, loss[loss=0.1696, simple_loss=0.2437, pruned_loss=0.04777, over 4726.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2665, pruned_loss=0.06853, over 955641.31 frames. ], batch size: 23, lr: 3.79e-03, grad_scale: 32.0 +2023-03-26 10:47:36,135 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.216e+02 1.624e+02 1.916e+02 2.206e+02 4.881e+02, threshold=3.833e+02, percent-clipped=2.0 +2023-03-26 10:47:59,943 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-03-26 10:48:07,637 INFO [finetune.py:976] (1/7) Epoch 9, batch 3150, loss[loss=0.1524, simple_loss=0.2202, pruned_loss=0.0423, over 4710.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2633, pruned_loss=0.06749, over 955400.74 frames. ], batch size: 59, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:48:16,688 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48980.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:48:33,966 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6520, 3.3823, 3.2164, 1.5177, 3.5178, 2.6025, 0.8425, 2.2511], + device='cuda:1'), covar=tensor([0.2377, 0.2096, 0.1865, 0.3874, 0.1346, 0.1153, 0.4727, 0.1858], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0173, 0.0159, 0.0127, 0.0156, 0.0122, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 10:48:50,518 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3542, 1.5030, 1.2793, 1.4685, 1.7395, 1.6118, 1.4977, 1.3085], + device='cuda:1'), covar=tensor([0.0391, 0.0297, 0.0514, 0.0310, 0.0206, 0.0498, 0.0349, 0.0374], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0110, 0.0140, 0.0114, 0.0102, 0.0101, 0.0091, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.0280e-05, 8.5761e-05, 1.1131e-04, 8.9855e-05, 8.0041e-05, 7.5145e-05, + 6.8828e-05, 8.3356e-05], device='cuda:1') +2023-03-26 10:48:51,002 INFO [finetune.py:976] (1/7) Epoch 9, batch 3200, loss[loss=0.2202, simple_loss=0.2711, pruned_loss=0.08465, over 4805.00 frames. ], tot_loss[loss=0.1963, simple_loss=0.2599, pruned_loss=0.06631, over 957123.82 frames. ], batch size: 51, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:48:55,662 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=49028.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:48:57,863 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.147e+02 1.665e+02 1.973e+02 2.326e+02 6.022e+02, threshold=3.945e+02, percent-clipped=4.0 +2023-03-26 10:49:03,719 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5015, 1.4970, 1.4114, 1.5506, 1.8067, 1.6814, 1.5889, 1.3104], + device='cuda:1'), covar=tensor([0.0357, 0.0306, 0.0595, 0.0322, 0.0225, 0.0488, 0.0354, 0.0433], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0110, 0.0140, 0.0115, 0.0102, 0.0101, 0.0091, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.0461e-05, 8.5959e-05, 1.1154e-04, 9.0044e-05, 8.0253e-05, 7.5378e-05, + 6.8989e-05, 8.3551e-05], device='cuda:1') +2023-03-26 10:49:08,505 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7253, 1.1599, 0.9912, 1.6626, 2.0577, 1.2921, 1.5233, 1.7028], + device='cuda:1'), covar=tensor([0.1412, 0.2134, 0.1984, 0.1111, 0.1941, 0.2009, 0.1458, 0.1816], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0114, 0.0092, 0.0122, 0.0096, 0.0099, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:49:12,699 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49053.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:49:30,122 INFO [finetune.py:976] (1/7) Epoch 9, batch 3250, loss[loss=0.2144, simple_loss=0.2901, pruned_loss=0.06934, over 4799.00 frames. ], tot_loss[loss=0.1959, simple_loss=0.2596, pruned_loss=0.06605, over 955401.72 frames. ], batch size: 51, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:49:40,157 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49079.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:50:05,715 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=49101.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:50:13,760 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-03-26 10:50:21,990 INFO [finetune.py:976] (1/7) Epoch 9, batch 3300, loss[loss=0.1792, simple_loss=0.2486, pruned_loss=0.05489, over 4754.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.265, pruned_loss=0.0684, over 954777.02 frames. ], batch size: 27, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:50:26,596 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=49127.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:50:28,940 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.155e+02 1.707e+02 1.914e+02 2.346e+02 3.542e+02, threshold=3.827e+02, percent-clipped=0.0 +2023-03-26 10:50:29,718 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9190, 2.6518, 2.4727, 3.0091, 2.6667, 2.6408, 2.6138, 3.6247], + device='cuda:1'), covar=tensor([0.3856, 0.5137, 0.3470, 0.3960, 0.4072, 0.2775, 0.4654, 0.1501], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0260, 0.0223, 0.0281, 0.0243, 0.0210, 0.0247, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 10:50:56,004 INFO [finetune.py:976] (1/7) Epoch 9, batch 3350, loss[loss=0.1963, simple_loss=0.2617, pruned_loss=0.0655, over 4884.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2671, pruned_loss=0.06901, over 954331.60 frames. ], batch size: 32, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:50:59,109 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49177.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:51:17,572 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49194.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:51:29,084 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6771, 1.5385, 1.5456, 1.5835, 1.0175, 3.0139, 1.1685, 1.7217], + device='cuda:1'), covar=tensor([0.3218, 0.2498, 0.2034, 0.2396, 0.1937, 0.0237, 0.2581, 0.1265], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0119, 0.0123, 0.0116, 0.0098, 0.0100, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 10:51:49,753 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-03-26 10:51:50,171 INFO [finetune.py:976] (1/7) Epoch 9, batch 3400, loss[loss=0.216, simple_loss=0.29, pruned_loss=0.07097, over 4915.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.269, pruned_loss=0.06962, over 956083.06 frames. ], batch size: 36, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:51:59,826 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.353e+01 1.606e+02 1.878e+02 2.295e+02 4.525e+02, threshold=3.756e+02, percent-clipped=2.0 +2023-03-26 10:52:10,014 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49236.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:52:45,317 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-03-26 10:52:55,267 INFO [finetune.py:976] (1/7) Epoch 9, batch 3450, loss[loss=0.1967, simple_loss=0.276, pruned_loss=0.05873, over 4815.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2688, pruned_loss=0.06879, over 956640.53 frames. ], batch size: 33, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:53:28,007 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49297.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:53:36,863 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-03-26 10:53:45,123 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-26 10:53:45,602 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 10:53:53,350 INFO [finetune.py:976] (1/7) Epoch 9, batch 3500, loss[loss=0.2034, simple_loss=0.2615, pruned_loss=0.07272, over 4873.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2662, pruned_loss=0.06842, over 954094.56 frames. ], batch size: 34, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:53:58,769 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.331e+01 1.641e+02 1.916e+02 2.289e+02 6.335e+02, threshold=3.833e+02, percent-clipped=2.0 +2023-03-26 10:54:02,504 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-03-26 10:54:34,217 INFO [finetune.py:976] (1/7) Epoch 9, batch 3550, loss[loss=0.2184, simple_loss=0.2825, pruned_loss=0.07717, over 4873.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2641, pruned_loss=0.06856, over 956194.29 frames. ], batch size: 34, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:55:09,813 INFO [finetune.py:976] (1/7) Epoch 9, batch 3600, loss[loss=0.2056, simple_loss=0.2665, pruned_loss=0.07239, over 4824.00 frames. ], tot_loss[loss=0.1983, simple_loss=0.2613, pruned_loss=0.06766, over 954520.67 frames. ], batch size: 40, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:55:09,921 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49422.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:55:15,219 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.012e+02 1.662e+02 2.002e+02 2.382e+02 4.044e+02, threshold=4.004e+02, percent-clipped=1.0 +2023-03-26 10:55:15,940 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49432.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 10:55:43,178 INFO [finetune.py:976] (1/7) Epoch 9, batch 3650, loss[loss=0.1952, simple_loss=0.2634, pruned_loss=0.06353, over 4908.00 frames. ], tot_loss[loss=0.2, simple_loss=0.2633, pruned_loss=0.06839, over 954346.14 frames. ], batch size: 36, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:55:46,362 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49477.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:55:50,094 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49483.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:55:56,198 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49493.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 10:55:56,751 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49494.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:56:00,240 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.23 vs. limit=5.0 +2023-03-26 10:56:17,055 INFO [finetune.py:976] (1/7) Epoch 9, batch 3700, loss[loss=0.2054, simple_loss=0.2801, pruned_loss=0.06535, over 4826.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2661, pruned_loss=0.06871, over 955514.90 frames. ], batch size: 47, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:56:18,949 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=49525.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:56:22,513 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.088e+02 1.782e+02 2.076e+02 2.384e+02 4.659e+02, threshold=4.152e+02, percent-clipped=5.0 +2023-03-26 10:56:29,206 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=49542.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:56:50,562 INFO [finetune.py:976] (1/7) Epoch 9, batch 3750, loss[loss=0.1806, simple_loss=0.2528, pruned_loss=0.05427, over 4897.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2675, pruned_loss=0.0694, over 953882.64 frames. ], batch size: 43, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:56:59,702 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6657, 1.2146, 0.9955, 1.5613, 2.0412, 1.1753, 1.5029, 1.6588], + device='cuda:1'), covar=tensor([0.1510, 0.2145, 0.2037, 0.1215, 0.2024, 0.2070, 0.1488, 0.1942], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0114, 0.0093, 0.0123, 0.0096, 0.0100, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:57:02,675 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49592.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:57:10,807 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 10:57:28,146 INFO [finetune.py:976] (1/7) Epoch 9, batch 3800, loss[loss=0.1767, simple_loss=0.2461, pruned_loss=0.05365, over 4922.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2698, pruned_loss=0.07024, over 953962.91 frames. ], batch size: 41, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:57:39,045 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.087e+02 1.635e+02 1.863e+02 2.259e+02 4.048e+02, threshold=3.725e+02, percent-clipped=0.0 +2023-03-26 10:58:12,405 INFO [finetune.py:976] (1/7) Epoch 9, batch 3850, loss[loss=0.2098, simple_loss=0.2767, pruned_loss=0.07149, over 4794.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2685, pruned_loss=0.06932, over 954937.69 frames. ], batch size: 29, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:58:16,656 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0552, 1.5178, 1.0853, 1.9768, 2.3611, 1.9164, 1.7234, 1.9380], + device='cuda:1'), covar=tensor([0.1107, 0.1680, 0.1887, 0.0929, 0.1704, 0.1867, 0.1180, 0.1634], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0114, 0.0093, 0.0123, 0.0096, 0.0100, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 10:58:17,918 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3524, 1.5224, 1.6632, 0.8680, 1.5708, 1.8546, 1.8263, 1.3960], + device='cuda:1'), covar=tensor([0.0944, 0.0605, 0.0437, 0.0610, 0.0456, 0.0567, 0.0336, 0.0730], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0156, 0.0122, 0.0135, 0.0132, 0.0126, 0.0147, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.6406e-05, 1.1494e-04, 8.7750e-05, 9.8346e-05, 9.4590e-05, 9.2331e-05, + 1.0763e-04, 1.0888e-04], device='cuda:1') +2023-03-26 10:58:30,260 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-26 10:58:31,320 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7668, 1.7681, 1.4729, 1.6405, 2.0088, 2.1073, 1.7192, 1.4278], + device='cuda:1'), covar=tensor([0.0259, 0.0249, 0.0533, 0.0298, 0.0187, 0.0276, 0.0311, 0.0362], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0108, 0.0139, 0.0114, 0.0102, 0.0101, 0.0091, 0.0108], + device='cuda:1'), out_proj_covar=tensor([6.9979e-05, 8.4854e-05, 1.1062e-04, 8.9706e-05, 7.9914e-05, 7.4997e-05, + 6.8586e-05, 8.3039e-05], device='cuda:1') +2023-03-26 10:58:48,064 INFO [finetune.py:976] (1/7) Epoch 9, batch 3900, loss[loss=0.2755, simple_loss=0.3079, pruned_loss=0.1216, over 4325.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2655, pruned_loss=0.06887, over 954902.94 frames. ], batch size: 65, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:58:58,258 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.078e+02 1.638e+02 1.913e+02 2.415e+02 4.821e+02, threshold=3.825e+02, percent-clipped=2.0 +2023-03-26 10:59:31,481 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49766.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:59:36,463 INFO [finetune.py:976] (1/7) Epoch 9, batch 3950, loss[loss=0.2413, simple_loss=0.2861, pruned_loss=0.09828, over 4893.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2618, pruned_loss=0.06764, over 954852.08 frames. ], batch size: 35, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 10:59:40,667 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49778.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 10:59:47,182 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49788.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 11:00:09,535 INFO [finetune.py:976] (1/7) Epoch 9, batch 4000, loss[loss=0.2231, simple_loss=0.2837, pruned_loss=0.08128, over 4935.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2616, pruned_loss=0.06833, over 953787.55 frames. ], batch size: 38, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 11:00:13,717 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49827.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:00:16,513 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.178e+02 1.645e+02 2.017e+02 2.376e+02 6.319e+02, threshold=4.034e+02, percent-clipped=3.0 +2023-03-26 11:00:32,826 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6825, 1.4553, 0.9823, 0.2406, 1.2621, 1.4685, 1.3442, 1.3488], + device='cuda:1'), covar=tensor([0.0822, 0.0763, 0.1303, 0.1964, 0.1339, 0.2274, 0.2282, 0.0873], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0198, 0.0200, 0.0186, 0.0215, 0.0206, 0.0221, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:00:42,824 INFO [finetune.py:976] (1/7) Epoch 9, batch 4050, loss[loss=0.1727, simple_loss=0.2624, pruned_loss=0.04149, over 4847.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.266, pruned_loss=0.07023, over 953918.75 frames. ], batch size: 49, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 11:00:56,949 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7263, 1.4948, 2.0082, 3.2604, 2.1963, 2.4302, 0.8527, 2.5264], + device='cuda:1'), covar=tensor([0.1747, 0.1598, 0.1469, 0.0716, 0.0869, 0.1771, 0.2050, 0.0657], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0133, 0.0165, 0.0102, 0.0138, 0.0126, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:00:56,964 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49892.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:01:01,855 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6872, 1.2114, 0.9514, 1.6036, 1.9699, 1.3384, 1.3971, 1.6170], + device='cuda:1'), covar=tensor([0.1355, 0.2007, 0.1940, 0.1161, 0.1982, 0.2022, 0.1491, 0.1859], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0113, 0.0092, 0.0122, 0.0095, 0.0100, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:01:15,999 INFO [finetune.py:976] (1/7) Epoch 9, batch 4100, loss[loss=0.1665, simple_loss=0.2417, pruned_loss=0.04567, over 4854.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.267, pruned_loss=0.06908, over 956408.56 frames. ], batch size: 31, lr: 3.78e-03, grad_scale: 32.0 +2023-03-26 11:01:22,949 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.254e+02 1.718e+02 2.083e+02 2.512e+02 3.689e+02, threshold=4.166e+02, percent-clipped=0.0 +2023-03-26 11:01:28,936 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=49940.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:01:48,767 INFO [finetune.py:976] (1/7) Epoch 9, batch 4150, loss[loss=0.2258, simple_loss=0.2857, pruned_loss=0.08291, over 4832.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2677, pruned_loss=0.06925, over 954453.18 frames. ], batch size: 47, lr: 3.78e-03, grad_scale: 16.0 +2023-03-26 11:01:53,394 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-03-26 11:02:23,458 INFO [finetune.py:976] (1/7) Epoch 9, batch 4200, loss[loss=0.2396, simple_loss=0.2951, pruned_loss=0.09207, over 4815.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2666, pruned_loss=0.06845, over 952804.02 frames. ], batch size: 38, lr: 3.78e-03, grad_scale: 16.0 +2023-03-26 11:02:31,376 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.147e+02 1.706e+02 2.002e+02 2.506e+02 6.230e+02, threshold=4.003e+02, percent-clipped=2.0 +2023-03-26 11:03:15,797 INFO [finetune.py:976] (1/7) Epoch 9, batch 4250, loss[loss=0.2102, simple_loss=0.2557, pruned_loss=0.08238, over 4176.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2653, pruned_loss=0.06818, over 953975.41 frames. ], batch size: 18, lr: 3.78e-03, grad_scale: 16.0 +2023-03-26 11:03:24,159 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50078.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:03:32,344 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50088.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 11:03:36,482 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7358, 1.6099, 1.5946, 1.7171, 1.3112, 3.5705, 1.4630, 2.1181], + device='cuda:1'), covar=tensor([0.3350, 0.2404, 0.2019, 0.2249, 0.1721, 0.0199, 0.2607, 0.1184], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0114, 0.0118, 0.0122, 0.0115, 0.0097, 0.0099, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0003, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 11:03:53,929 INFO [finetune.py:976] (1/7) Epoch 9, batch 4300, loss[loss=0.2012, simple_loss=0.2691, pruned_loss=0.06663, over 4828.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2627, pruned_loss=0.0673, over 955267.25 frames. ], batch size: 30, lr: 3.78e-03, grad_scale: 16.0 +2023-03-26 11:03:54,007 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50122.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:03:56,364 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=50126.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:03:57,602 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50128.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:04:00,430 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.003e+02 1.608e+02 1.917e+02 2.228e+02 4.011e+02, threshold=3.835e+02, percent-clipped=1.0 +2023-03-26 11:04:03,879 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=50136.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 11:04:49,371 INFO [finetune.py:976] (1/7) Epoch 9, batch 4350, loss[loss=0.1924, simple_loss=0.2509, pruned_loss=0.06696, over 4755.00 frames. ], tot_loss[loss=0.1954, simple_loss=0.2591, pruned_loss=0.06583, over 957213.35 frames. ], batch size: 27, lr: 3.78e-03, grad_scale: 16.0 +2023-03-26 11:05:17,862 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50189.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:05:30,437 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50197.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 11:05:32,599 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.57 vs. limit=5.0 +2023-03-26 11:06:02,455 INFO [finetune.py:976] (1/7) Epoch 9, batch 4400, loss[loss=0.2238, simple_loss=0.2971, pruned_loss=0.07521, over 4749.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2607, pruned_loss=0.06711, over 954687.34 frames. ], batch size: 54, lr: 3.78e-03, grad_scale: 16.0 +2023-03-26 11:06:14,078 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.827e+01 1.714e+02 1.989e+02 2.480e+02 5.028e+02, threshold=3.977e+02, percent-clipped=2.0 +2023-03-26 11:06:24,374 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0374, 1.9000, 1.6273, 1.8074, 1.8816, 1.8106, 1.8498, 2.5447], + device='cuda:1'), covar=tensor([0.4894, 0.5460, 0.3918, 0.4960, 0.4405, 0.2838, 0.5024, 0.1930], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0258, 0.0220, 0.0278, 0.0242, 0.0207, 0.0244, 0.0210], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:06:48,105 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50258.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 11:06:58,869 INFO [finetune.py:976] (1/7) Epoch 9, batch 4450, loss[loss=0.2077, simple_loss=0.2799, pruned_loss=0.06771, over 4810.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2653, pruned_loss=0.06819, over 954479.23 frames. ], batch size: 38, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:07:15,445 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1353, 1.8022, 2.0617, 2.0028, 1.7730, 1.7707, 1.9829, 1.8874], + device='cuda:1'), covar=tensor([0.5070, 0.5746, 0.4613, 0.5511, 0.6822, 0.4887, 0.6688, 0.4524], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0240, 0.0253, 0.0256, 0.0250, 0.0224, 0.0273, 0.0230], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:07:18,305 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7452, 4.0234, 3.7412, 2.0636, 4.0234, 2.9309, 0.6676, 2.8735], + device='cuda:1'), covar=tensor([0.2322, 0.1703, 0.1500, 0.2992, 0.0987, 0.0989, 0.4735, 0.1227], + device='cuda:1'), in_proj_covar=tensor([0.0155, 0.0177, 0.0162, 0.0130, 0.0159, 0.0124, 0.0149, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 11:07:23,648 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9223, 1.8936, 2.0156, 1.3022, 1.9802, 1.9864, 1.8721, 1.6445], + device='cuda:1'), covar=tensor([0.0542, 0.0613, 0.0581, 0.0866, 0.0633, 0.0653, 0.0622, 0.1039], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0132, 0.0143, 0.0123, 0.0117, 0.0143, 0.0142, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:07:32,437 INFO [finetune.py:976] (1/7) Epoch 9, batch 4500, loss[loss=0.1547, simple_loss=0.2241, pruned_loss=0.04261, over 4756.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2656, pruned_loss=0.06816, over 951908.23 frames. ], batch size: 27, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:07:33,711 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4787, 4.0217, 4.2623, 4.2116, 3.9917, 3.9753, 4.6260, 1.7727], + device='cuda:1'), covar=tensor([0.0952, 0.1538, 0.1222, 0.1240, 0.2017, 0.1950, 0.0859, 0.6362], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0244, 0.0275, 0.0291, 0.0330, 0.0281, 0.0301, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:07:33,880 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-26 11:07:38,449 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.049e+02 1.523e+02 1.896e+02 2.480e+02 4.445e+02, threshold=3.793e+02, percent-clipped=2.0 +2023-03-26 11:08:05,991 INFO [finetune.py:976] (1/7) Epoch 9, batch 4550, loss[loss=0.23, simple_loss=0.2863, pruned_loss=0.08689, over 4838.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.268, pruned_loss=0.06928, over 952435.37 frames. ], batch size: 47, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:08:09,789 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8672, 1.7114, 1.6735, 1.7629, 1.5962, 4.5026, 1.8550, 2.5028], + device='cuda:1'), covar=tensor([0.3459, 0.2539, 0.2164, 0.2377, 0.1626, 0.0114, 0.2334, 0.1157], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0119, 0.0122, 0.0115, 0.0097, 0.0099, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 11:08:27,378 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2092, 2.2484, 2.1791, 1.5971, 2.3301, 2.3740, 2.1978, 1.9510], + device='cuda:1'), covar=tensor([0.0582, 0.0537, 0.0692, 0.0830, 0.0531, 0.0633, 0.0591, 0.0983], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0133, 0.0144, 0.0123, 0.0117, 0.0143, 0.0143, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:08:58,491 INFO [finetune.py:976] (1/7) Epoch 9, batch 4600, loss[loss=0.1486, simple_loss=0.2224, pruned_loss=0.03741, over 4871.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2673, pruned_loss=0.06908, over 955149.25 frames. ], batch size: 31, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:08:58,604 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50422.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:09:05,137 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9071, 1.2500, 0.9138, 1.8715, 2.2081, 1.5216, 1.5924, 1.7954], + device='cuda:1'), covar=tensor([0.1456, 0.2209, 0.2152, 0.1132, 0.1846, 0.2067, 0.1571, 0.1964], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0114, 0.0092, 0.0122, 0.0096, 0.0100, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:09:06,260 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.176e+02 1.671e+02 1.983e+02 2.416e+02 3.848e+02, threshold=3.965e+02, percent-clipped=1.0 +2023-03-26 11:09:39,359 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=50470.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:09:40,499 INFO [finetune.py:976] (1/7) Epoch 9, batch 4650, loss[loss=0.2024, simple_loss=0.2608, pruned_loss=0.07199, over 4820.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.2636, pruned_loss=0.06774, over 955220.71 frames. ], batch size: 40, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:09:50,493 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50484.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:10:22,783 INFO [finetune.py:976] (1/7) Epoch 9, batch 4700, loss[loss=0.1664, simple_loss=0.2378, pruned_loss=0.04756, over 4754.00 frames. ], tot_loss[loss=0.1954, simple_loss=0.2595, pruned_loss=0.06563, over 956526.10 frames. ], batch size: 28, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:10:29,351 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.001e+02 1.527e+02 1.850e+02 2.224e+02 3.838e+02, threshold=3.699e+02, percent-clipped=0.0 +2023-03-26 11:10:38,041 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50546.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:10:42,727 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50553.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 11:10:56,119 INFO [finetune.py:976] (1/7) Epoch 9, batch 4750, loss[loss=0.1971, simple_loss=0.2602, pruned_loss=0.06704, over 4800.00 frames. ], tot_loss[loss=0.1945, simple_loss=0.2583, pruned_loss=0.06539, over 954412.48 frames. ], batch size: 29, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:11:14,484 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6008, 1.1625, 0.7209, 1.6161, 2.0324, 1.3213, 1.4758, 1.5902], + device='cuda:1'), covar=tensor([0.1939, 0.2860, 0.2657, 0.1582, 0.2418, 0.2815, 0.2085, 0.2634], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0115, 0.0093, 0.0123, 0.0096, 0.0100, 0.0093], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:11:17,965 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0215, 2.1243, 1.7929, 1.8010, 2.4231, 2.6264, 2.2916, 2.0117], + device='cuda:1'), covar=tensor([0.0305, 0.0340, 0.0525, 0.0363, 0.0274, 0.0445, 0.0242, 0.0387], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0109, 0.0139, 0.0114, 0.0102, 0.0102, 0.0091, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.0129e-05, 8.5551e-05, 1.1073e-04, 8.9883e-05, 8.0043e-05, 7.5339e-05, + 6.9084e-05, 8.3589e-05], device='cuda:1') +2023-03-26 11:11:18,589 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50607.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:11:29,494 INFO [finetune.py:976] (1/7) Epoch 9, batch 4800, loss[loss=0.1674, simple_loss=0.2413, pruned_loss=0.04673, over 4888.00 frames. ], tot_loss[loss=0.1957, simple_loss=0.2599, pruned_loss=0.06578, over 955085.22 frames. ], batch size: 35, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:11:33,377 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.18 vs. limit=5.0 +2023-03-26 11:11:36,103 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.695e+01 1.587e+02 1.907e+02 2.189e+02 3.978e+02, threshold=3.813e+02, percent-clipped=2.0 +2023-03-26 11:11:42,866 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1368, 1.9119, 2.5014, 3.7023, 2.5769, 2.7274, 1.4818, 2.8600], + device='cuda:1'), covar=tensor([0.1809, 0.1494, 0.1293, 0.0512, 0.0805, 0.1346, 0.1783, 0.0706], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0118, 0.0134, 0.0165, 0.0102, 0.0140, 0.0127, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:12:03,073 INFO [finetune.py:976] (1/7) Epoch 9, batch 4850, loss[loss=0.221, simple_loss=0.2941, pruned_loss=0.07391, over 4798.00 frames. ], tot_loss[loss=0.1978, simple_loss=0.2629, pruned_loss=0.06639, over 954594.17 frames. ], batch size: 45, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:12:36,220 INFO [finetune.py:976] (1/7) Epoch 9, batch 4900, loss[loss=0.2723, simple_loss=0.3152, pruned_loss=0.1147, over 4891.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.2645, pruned_loss=0.06731, over 953575.83 frames. ], batch size: 35, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:12:41,848 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1332, 2.0248, 1.5703, 2.0152, 2.0705, 1.7257, 2.3589, 2.0330], + device='cuda:1'), covar=tensor([0.1318, 0.2297, 0.3401, 0.3018, 0.2725, 0.1831, 0.3434, 0.1964], + device='cuda:1'), in_proj_covar=tensor([0.0173, 0.0187, 0.0233, 0.0253, 0.0237, 0.0195, 0.0211, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:12:42,295 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.610e+02 1.915e+02 2.289e+02 4.400e+02, threshold=3.830e+02, percent-clipped=2.0 +2023-03-26 11:13:08,690 INFO [finetune.py:976] (1/7) Epoch 9, batch 4950, loss[loss=0.2661, simple_loss=0.3201, pruned_loss=0.1061, over 4816.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2681, pruned_loss=0.06931, over 955945.51 frames. ], batch size: 38, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:13:09,030 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 11:13:10,728 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-03-26 11:13:15,372 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50782.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:13:16,531 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50784.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:13:28,604 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6125, 1.5089, 1.5551, 1.6148, 1.0884, 3.3103, 1.2646, 1.8022], + device='cuda:1'), covar=tensor([0.3447, 0.2452, 0.2148, 0.2261, 0.1853, 0.0192, 0.2683, 0.1277], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0119, 0.0122, 0.0115, 0.0098, 0.0099, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 11:13:33,475 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50808.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 11:13:51,978 INFO [finetune.py:976] (1/7) Epoch 9, batch 5000, loss[loss=0.2227, simple_loss=0.2674, pruned_loss=0.08904, over 4738.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2665, pruned_loss=0.06891, over 955289.92 frames. ], batch size: 23, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:14:03,064 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.212e+02 1.752e+02 2.044e+02 2.444e+02 6.074e+02, threshold=4.089e+02, percent-clipped=4.0 +2023-03-26 11:14:03,135 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=50832.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:14:05,040 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3037, 2.2239, 2.5687, 1.1592, 2.7137, 2.6985, 2.3721, 2.1161], + device='cuda:1'), covar=tensor([0.1020, 0.0809, 0.0356, 0.0795, 0.0407, 0.0804, 0.0466, 0.0812], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0155, 0.0121, 0.0134, 0.0131, 0.0126, 0.0145, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.5953e-05, 1.1387e-04, 8.7118e-05, 9.7009e-05, 9.3508e-05, 9.1796e-05, + 1.0632e-04, 1.0796e-04], device='cuda:1') +2023-03-26 11:14:13,393 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50843.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:14:14,459 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8133, 1.7724, 1.7601, 1.8090, 1.4631, 3.2996, 1.7271, 2.1400], + device='cuda:1'), covar=tensor([0.2965, 0.2033, 0.1732, 0.1926, 0.1455, 0.0253, 0.2611, 0.1138], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0119, 0.0123, 0.0116, 0.0098, 0.0099, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 11:14:24,224 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50853.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 11:14:30,929 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.06 vs. limit=5.0 +2023-03-26 11:14:37,391 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50869.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 11:14:40,057 INFO [finetune.py:976] (1/7) Epoch 9, batch 5050, loss[loss=0.174, simple_loss=0.2405, pruned_loss=0.05371, over 4739.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2629, pruned_loss=0.06744, over 954978.81 frames. ], batch size: 54, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:14:42,457 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6383, 1.7158, 2.0992, 1.8848, 1.8690, 4.1335, 1.5764, 1.9909], + device='cuda:1'), covar=tensor([0.0918, 0.1678, 0.1110, 0.0931, 0.1427, 0.0148, 0.1372, 0.1514], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0091, 0.0082, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 11:15:00,240 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=50901.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 11:15:00,852 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50902.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:15:21,255 INFO [finetune.py:976] (1/7) Epoch 9, batch 5100, loss[loss=0.1768, simple_loss=0.2273, pruned_loss=0.06314, over 4712.00 frames. ], tot_loss[loss=0.1964, simple_loss=0.2597, pruned_loss=0.06653, over 956748.73 frames. ], batch size: 23, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:15:29,782 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.163e+02 1.571e+02 1.989e+02 2.366e+02 5.072e+02, threshold=3.977e+02, percent-clipped=2.0 +2023-03-26 11:15:38,857 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50946.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:15:55,122 INFO [finetune.py:976] (1/7) Epoch 9, batch 5150, loss[loss=0.1526, simple_loss=0.2225, pruned_loss=0.04132, over 4797.00 frames. ], tot_loss[loss=0.197, simple_loss=0.26, pruned_loss=0.06693, over 953994.69 frames. ], batch size: 29, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:16:18,434 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7827, 1.3614, 1.7775, 1.6758, 1.4959, 1.4766, 1.7154, 1.6483], + device='cuda:1'), covar=tensor([0.4772, 0.4866, 0.4049, 0.4909, 0.5547, 0.4188, 0.5424, 0.3922], + device='cuda:1'), in_proj_covar=tensor([0.0234, 0.0241, 0.0254, 0.0256, 0.0251, 0.0226, 0.0274, 0.0230], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:16:19,643 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51007.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:16:20,326 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8906, 1.3244, 1.7976, 1.7727, 1.5949, 1.5676, 1.7694, 1.6680], + device='cuda:1'), covar=tensor([0.4159, 0.4847, 0.3783, 0.4215, 0.5243, 0.4073, 0.5244, 0.3743], + device='cuda:1'), in_proj_covar=tensor([0.0234, 0.0241, 0.0254, 0.0256, 0.0251, 0.0226, 0.0274, 0.0230], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:16:29,120 INFO [finetune.py:976] (1/7) Epoch 9, batch 5200, loss[loss=0.2439, simple_loss=0.3045, pruned_loss=0.09171, over 4757.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2635, pruned_loss=0.06758, over 955840.32 frames. ], batch size: 59, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:16:37,437 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.247e+02 1.691e+02 2.095e+02 2.506e+02 4.401e+02, threshold=4.191e+02, percent-clipped=2.0 +2023-03-26 11:16:57,046 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 11:17:12,599 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51062.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:17:18,574 INFO [finetune.py:976] (1/7) Epoch 9, batch 5250, loss[loss=0.1796, simple_loss=0.2566, pruned_loss=0.05128, over 4808.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2644, pruned_loss=0.06797, over 951679.03 frames. ], batch size: 51, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:17:49,027 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 11:17:51,198 INFO [finetune.py:976] (1/7) Epoch 9, batch 5300, loss[loss=0.1925, simple_loss=0.2655, pruned_loss=0.0598, over 4799.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2674, pruned_loss=0.06946, over 954416.68 frames. ], batch size: 45, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:17:51,947 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51123.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:17:57,266 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.167e+02 1.689e+02 2.023e+02 2.414e+02 5.734e+02, threshold=4.045e+02, percent-clipped=1.0 +2023-03-26 11:18:01,395 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51138.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:18:19,581 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51164.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 11:18:23,869 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6457, 1.4899, 1.4607, 1.5420, 1.8804, 1.8196, 1.6454, 1.4882], + device='cuda:1'), covar=tensor([0.0309, 0.0282, 0.0553, 0.0259, 0.0195, 0.0337, 0.0285, 0.0317], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0109, 0.0140, 0.0115, 0.0102, 0.0102, 0.0091, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.0650e-05, 8.5377e-05, 1.1116e-04, 9.0303e-05, 8.0044e-05, 7.5559e-05, + 6.9186e-05, 8.3163e-05], device='cuda:1') +2023-03-26 11:18:24,357 INFO [finetune.py:976] (1/7) Epoch 9, batch 5350, loss[loss=0.1961, simple_loss=0.2497, pruned_loss=0.07126, over 4697.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2681, pruned_loss=0.06947, over 953090.69 frames. ], batch size: 23, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:18:24,453 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51172.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 11:18:37,389 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.4591, 2.8163, 2.5983, 1.1682, 2.8383, 2.3662, 2.0126, 2.4492], + device='cuda:1'), covar=tensor([0.0961, 0.1078, 0.1982, 0.2685, 0.1957, 0.2434, 0.2481, 0.1366], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0199, 0.0201, 0.0187, 0.0214, 0.0207, 0.0223, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:18:56,212 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51202.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:19:18,523 INFO [finetune.py:976] (1/7) Epoch 9, batch 5400, loss[loss=0.2279, simple_loss=0.2913, pruned_loss=0.0823, over 4755.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2654, pruned_loss=0.06892, over 952711.76 frames. ], batch size: 54, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:19:26,734 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.913e+01 1.608e+02 1.826e+02 2.251e+02 3.272e+02, threshold=3.651e+02, percent-clipped=0.0 +2023-03-26 11:19:32,695 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51233.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 11:19:48,894 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=51250.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:20:03,230 INFO [finetune.py:976] (1/7) Epoch 9, batch 5450, loss[loss=0.1829, simple_loss=0.2405, pruned_loss=0.06264, over 4750.00 frames. ], tot_loss[loss=0.1979, simple_loss=0.2617, pruned_loss=0.06703, over 952242.65 frames. ], batch size: 28, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:20:31,733 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51302.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:20:38,861 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2896, 2.9220, 3.0421, 3.1901, 3.0244, 2.9395, 3.3452, 0.9757], + device='cuda:1'), covar=tensor([0.1189, 0.1073, 0.1061, 0.1330, 0.1870, 0.1653, 0.1119, 0.5507], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0246, 0.0279, 0.0295, 0.0333, 0.0283, 0.0304, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:20:50,272 INFO [finetune.py:976] (1/7) Epoch 9, batch 5500, loss[loss=0.1645, simple_loss=0.2133, pruned_loss=0.05787, over 4214.00 frames. ], tot_loss[loss=0.1961, simple_loss=0.2594, pruned_loss=0.06638, over 952315.92 frames. ], batch size: 17, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:20:56,825 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.040e+02 1.581e+02 1.869e+02 2.249e+02 3.902e+02, threshold=3.738e+02, percent-clipped=2.0 +2023-03-26 11:21:24,461 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4707, 1.8214, 1.4641, 1.4254, 1.9398, 1.7131, 1.6992, 1.6591], + device='cuda:1'), covar=tensor([0.0471, 0.0333, 0.0554, 0.0388, 0.0447, 0.0720, 0.0326, 0.0405], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0110, 0.0141, 0.0117, 0.0103, 0.0103, 0.0092, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.1207e-05, 8.6271e-05, 1.1212e-04, 9.1581e-05, 8.1033e-05, 7.6470e-05, + 6.9897e-05, 8.4078e-05], device='cuda:1') +2023-03-26 11:21:45,444 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0633, 2.1079, 2.1927, 1.4467, 2.2164, 2.2032, 2.1740, 1.8008], + device='cuda:1'), covar=tensor([0.0592, 0.0570, 0.0618, 0.0837, 0.0547, 0.0641, 0.0585, 0.1006], + device='cuda:1'), in_proj_covar=tensor([0.0137, 0.0133, 0.0145, 0.0124, 0.0118, 0.0144, 0.0144, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:21:48,954 INFO [finetune.py:976] (1/7) Epoch 9, batch 5550, loss[loss=0.1888, simple_loss=0.257, pruned_loss=0.06034, over 4790.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.2622, pruned_loss=0.06804, over 950235.26 frames. ], batch size: 29, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:21:49,843 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-26 11:21:59,537 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51382.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:22:07,975 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51392.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:22:24,404 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51418.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:22:27,580 INFO [finetune.py:976] (1/7) Epoch 9, batch 5600, loss[loss=0.1659, simple_loss=0.2141, pruned_loss=0.05881, over 4033.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2649, pruned_loss=0.06878, over 948249.79 frames. ], batch size: 17, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:22:33,285 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.626e+02 2.005e+02 2.362e+02 4.096e+02, threshold=4.011e+02, percent-clipped=2.0 +2023-03-26 11:22:36,850 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51438.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:22:39,791 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51443.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:22:46,029 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51453.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:22:52,433 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51464.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 11:22:57,075 INFO [finetune.py:976] (1/7) Epoch 9, batch 5650, loss[loss=0.272, simple_loss=0.32, pruned_loss=0.112, over 4888.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.268, pruned_loss=0.0697, over 950262.60 frames. ], batch size: 32, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:23:05,304 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=51486.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:23:10,709 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51495.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:23:16,031 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0123, 1.7665, 2.3997, 3.5763, 2.5787, 2.8134, 1.2589, 2.7593], + device='cuda:1'), covar=tensor([0.1714, 0.1496, 0.1330, 0.0564, 0.0765, 0.1138, 0.1905, 0.0624], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0134, 0.0164, 0.0102, 0.0138, 0.0126, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:23:19,637 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9856, 1.8774, 2.1292, 2.1542, 1.9605, 3.5101, 1.7072, 2.0388], + device='cuda:1'), covar=tensor([0.0763, 0.1378, 0.0839, 0.0716, 0.1229, 0.0321, 0.1134, 0.1253], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0075, 0.0078, 0.0092, 0.0082, 0.0084, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 11:23:20,457 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-26 11:23:20,776 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=51512.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 11:23:26,711 INFO [finetune.py:976] (1/7) Epoch 9, batch 5700, loss[loss=0.1613, simple_loss=0.2127, pruned_loss=0.05497, over 4639.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2639, pruned_loss=0.06954, over 929377.12 frames. ], batch size: 20, lr: 3.77e-03, grad_scale: 16.0 +2023-03-26 11:23:30,370 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51528.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 11:23:32,890 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.987e+01 1.624e+02 1.963e+02 2.341e+02 6.572e+02, threshold=3.927e+02, percent-clipped=1.0 +2023-03-26 11:23:57,292 INFO [finetune.py:976] (1/7) Epoch 10, batch 0, loss[loss=0.1776, simple_loss=0.2418, pruned_loss=0.05668, over 4904.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2418, pruned_loss=0.05668, over 4904.00 frames. ], batch size: 33, lr: 3.76e-03, grad_scale: 16.0 +2023-03-26 11:23:57,292 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 11:24:16,170 INFO [finetune.py:1010] (1/7) Epoch 10, validation: loss=0.1604, simple_loss=0.2317, pruned_loss=0.04451, over 2265189.00 frames. +2023-03-26 11:24:16,171 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 11:24:22,494 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51556.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:24:30,965 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-03-26 11:24:52,581 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1320, 1.4067, 0.7734, 2.0990, 2.3638, 1.8323, 1.7871, 2.0077], + device='cuda:1'), covar=tensor([0.1232, 0.1850, 0.2227, 0.1026, 0.1791, 0.2063, 0.1256, 0.1811], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0113, 0.0092, 0.0121, 0.0095, 0.0100, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:24:58,290 INFO [finetune.py:976] (1/7) Epoch 10, batch 50, loss[loss=0.1793, simple_loss=0.236, pruned_loss=0.06124, over 4745.00 frames. ], tot_loss[loss=0.206, simple_loss=0.27, pruned_loss=0.07097, over 215743.58 frames. ], batch size: 23, lr: 3.76e-03, grad_scale: 16.0 +2023-03-26 11:25:01,140 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51602.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:25:20,208 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.916e+01 1.735e+02 2.131e+02 2.642e+02 7.480e+02, threshold=4.262e+02, percent-clipped=4.0 +2023-03-26 11:25:31,992 INFO [finetune.py:976] (1/7) Epoch 10, batch 100, loss[loss=0.2218, simple_loss=0.2836, pruned_loss=0.07995, over 4902.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2639, pruned_loss=0.06963, over 378554.81 frames. ], batch size: 43, lr: 3.76e-03, grad_scale: 16.0 +2023-03-26 11:25:32,634 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=51650.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:26:04,771 INFO [finetune.py:976] (1/7) Epoch 10, batch 150, loss[loss=0.181, simple_loss=0.2427, pruned_loss=0.05961, over 4844.00 frames. ], tot_loss[loss=0.196, simple_loss=0.2581, pruned_loss=0.06695, over 507434.43 frames. ], batch size: 47, lr: 3.76e-03, grad_scale: 16.0 +2023-03-26 11:26:05,358 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6595, 1.5610, 2.0648, 1.4232, 1.8676, 1.8753, 1.5611, 2.1567], + device='cuda:1'), covar=tensor([0.1563, 0.2294, 0.1578, 0.2240, 0.1150, 0.1617, 0.3275, 0.0928], + device='cuda:1'), in_proj_covar=tensor([0.0200, 0.0207, 0.0197, 0.0194, 0.0181, 0.0218, 0.0219, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:26:12,265 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2998, 2.1122, 1.7300, 2.1834, 2.2472, 1.8819, 2.5546, 2.1832], + device='cuda:1'), covar=tensor([0.1316, 0.2408, 0.3313, 0.2958, 0.2651, 0.1636, 0.3041, 0.1884], + device='cuda:1'), in_proj_covar=tensor([0.0174, 0.0188, 0.0234, 0.0254, 0.0239, 0.0196, 0.0212, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:26:18,708 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51718.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:26:33,393 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.046e+02 1.594e+02 1.858e+02 2.240e+02 3.308e+02, threshold=3.716e+02, percent-clipped=0.0 +2023-03-26 11:26:37,114 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51738.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:26:47,584 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51747.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:26:48,167 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51748.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:26:52,005 INFO [finetune.py:976] (1/7) Epoch 10, batch 200, loss[loss=0.1605, simple_loss=0.229, pruned_loss=0.04599, over 4790.00 frames. ], tot_loss[loss=0.1963, simple_loss=0.2588, pruned_loss=0.06689, over 609179.61 frames. ], batch size: 29, lr: 3.76e-03, grad_scale: 16.0 +2023-03-26 11:27:02,613 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-26 11:27:04,853 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=51766.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:27:10,490 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8080, 1.6711, 1.5572, 1.8874, 2.2966, 1.9604, 1.5602, 1.4833], + device='cuda:1'), covar=tensor([0.2108, 0.2008, 0.1833, 0.1674, 0.1840, 0.1155, 0.2520, 0.1919], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0209, 0.0207, 0.0189, 0.0241, 0.0181, 0.0214, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:27:25,424 INFO [finetune.py:976] (1/7) Epoch 10, batch 250, loss[loss=0.2236, simple_loss=0.3111, pruned_loss=0.06807, over 4845.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2634, pruned_loss=0.06856, over 687194.16 frames. ], batch size: 49, lr: 3.76e-03, grad_scale: 16.0 +2023-03-26 11:27:32,990 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51808.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:27:45,691 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51828.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 11:27:48,003 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.029e+02 1.605e+02 1.930e+02 2.331e+02 5.576e+02, threshold=3.861e+02, percent-clipped=5.0 +2023-03-26 11:27:58,881 INFO [finetune.py:976] (1/7) Epoch 10, batch 300, loss[loss=0.1397, simple_loss=0.2183, pruned_loss=0.03056, over 4806.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.266, pruned_loss=0.0685, over 747072.09 frames. ], batch size: 25, lr: 3.76e-03, grad_scale: 16.0 +2023-03-26 11:28:00,155 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51851.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:28:04,961 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6898, 0.6498, 1.6850, 1.5517, 1.4663, 1.3977, 1.4555, 1.6002], + device='cuda:1'), covar=tensor([0.3646, 0.4661, 0.3783, 0.3976, 0.5081, 0.3867, 0.4748, 0.3572], + device='cuda:1'), in_proj_covar=tensor([0.0235, 0.0241, 0.0255, 0.0257, 0.0252, 0.0227, 0.0276, 0.0230], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:28:17,688 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=51876.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 11:28:31,952 INFO [finetune.py:976] (1/7) Epoch 10, batch 350, loss[loss=0.2426, simple_loss=0.2917, pruned_loss=0.09677, over 4899.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2664, pruned_loss=0.06825, over 791307.65 frames. ], batch size: 32, lr: 3.76e-03, grad_scale: 16.0 +2023-03-26 11:28:46,758 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.54 vs. limit=5.0 +2023-03-26 11:28:54,278 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.052e+02 1.678e+02 2.044e+02 2.443e+02 3.814e+02, threshold=4.089e+02, percent-clipped=0.0 +2023-03-26 11:29:04,642 INFO [finetune.py:976] (1/7) Epoch 10, batch 400, loss[loss=0.2103, simple_loss=0.2695, pruned_loss=0.0756, over 4847.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2672, pruned_loss=0.06798, over 827656.67 frames. ], batch size: 49, lr: 3.76e-03, grad_scale: 16.0 +2023-03-26 11:29:12,579 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0 +2023-03-26 11:29:17,475 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6982, 1.4169, 2.2551, 3.3324, 2.3090, 2.4596, 0.9709, 2.5872], + device='cuda:1'), covar=tensor([0.1683, 0.1623, 0.1264, 0.0646, 0.0783, 0.1514, 0.1895, 0.0579], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0134, 0.0165, 0.0102, 0.0139, 0.0126, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:29:47,994 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.50 vs. limit=5.0 +2023-03-26 11:29:56,981 INFO [finetune.py:976] (1/7) Epoch 10, batch 450, loss[loss=0.2336, simple_loss=0.2791, pruned_loss=0.09403, over 4384.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.265, pruned_loss=0.06714, over 855851.09 frames. ], batch size: 66, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:30:13,862 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.74 vs. limit=5.0 +2023-03-26 11:30:21,154 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.724e+02 2.025e+02 2.574e+02 4.346e+02, threshold=4.050e+02, percent-clipped=1.0 +2023-03-26 11:30:24,958 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52038.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:30:26,199 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52040.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:30:30,961 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52048.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:30:31,462 INFO [finetune.py:976] (1/7) Epoch 10, batch 500, loss[loss=0.2081, simple_loss=0.2737, pruned_loss=0.07125, over 4859.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2626, pruned_loss=0.06671, over 879147.89 frames. ], batch size: 47, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:30:56,807 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=52086.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:31:02,785 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=52096.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:31:04,604 INFO [finetune.py:976] (1/7) Epoch 10, batch 550, loss[loss=0.1672, simple_loss=0.2272, pruned_loss=0.05357, over 4774.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2596, pruned_loss=0.06548, over 897274.18 frames. ], batch size: 26, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:31:05,925 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52101.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:31:07,053 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52103.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:31:10,803 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7756, 1.5758, 1.4218, 1.1656, 1.5863, 1.5890, 1.5496, 2.1583], + device='cuda:1'), covar=tensor([0.4649, 0.4215, 0.3747, 0.4043, 0.3889, 0.2653, 0.3820, 0.1995], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0259, 0.0222, 0.0279, 0.0243, 0.0209, 0.0244, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:31:27,059 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.020e+02 1.591e+02 1.822e+02 2.163e+02 6.487e+02, threshold=3.643e+02, percent-clipped=1.0 +2023-03-26 11:31:37,961 INFO [finetune.py:976] (1/7) Epoch 10, batch 600, loss[loss=0.227, simple_loss=0.2921, pruned_loss=0.0809, over 4817.00 frames. ], tot_loss[loss=0.1976, simple_loss=0.2612, pruned_loss=0.06698, over 909267.31 frames. ], batch size: 40, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:31:39,253 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52151.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:32:10,105 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2270, 1.9845, 1.6767, 2.0092, 2.2560, 1.9074, 2.5621, 2.2380], + device='cuda:1'), covar=tensor([0.1472, 0.2944, 0.3776, 0.2967, 0.2801, 0.1871, 0.3206, 0.2060], + device='cuda:1'), in_proj_covar=tensor([0.0175, 0.0189, 0.0234, 0.0255, 0.0240, 0.0196, 0.0213, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:32:11,831 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7059, 1.5492, 1.5831, 1.6210, 1.2817, 4.1904, 1.6672, 2.0585], + device='cuda:1'), covar=tensor([0.3449, 0.2488, 0.2155, 0.2397, 0.1755, 0.0110, 0.2674, 0.1327], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0123, 0.0116, 0.0099, 0.0099, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 11:32:15,959 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52193.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:32:19,559 INFO [finetune.py:976] (1/7) Epoch 10, batch 650, loss[loss=0.1854, simple_loss=0.2525, pruned_loss=0.05917, over 4915.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2638, pruned_loss=0.0674, over 920689.08 frames. ], batch size: 32, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:32:19,617 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=52199.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:32:33,037 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6116, 1.4503, 1.5444, 1.5621, 0.9272, 3.3545, 1.2982, 1.6906], + device='cuda:1'), covar=tensor([0.3264, 0.2429, 0.2052, 0.2175, 0.1944, 0.0192, 0.2519, 0.1282], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0119, 0.0123, 0.0116, 0.0098, 0.0099, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 11:32:42,612 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.128e+02 1.681e+02 1.969e+02 2.336e+02 3.855e+02, threshold=3.938e+02, percent-clipped=2.0 +2023-03-26 11:32:53,490 INFO [finetune.py:976] (1/7) Epoch 10, batch 700, loss[loss=0.2418, simple_loss=0.2804, pruned_loss=0.1016, over 4748.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2645, pruned_loss=0.06788, over 927828.90 frames. ], batch size: 27, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:32:56,638 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52254.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:33:26,715 INFO [finetune.py:976] (1/7) Epoch 10, batch 750, loss[loss=0.1687, simple_loss=0.2369, pruned_loss=0.05023, over 4766.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2654, pruned_loss=0.0682, over 933003.52 frames. ], batch size: 26, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:33:45,058 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52312.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:34:02,791 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.081e+02 1.612e+02 1.864e+02 2.364e+02 4.342e+02, threshold=3.728e+02, percent-clipped=1.0 +2023-03-26 11:34:06,430 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0757, 1.9788, 1.7258, 2.0742, 1.9732, 1.9067, 1.9161, 2.8384], + device='cuda:1'), covar=tensor([0.4967, 0.5868, 0.3967, 0.5523, 0.5379, 0.2962, 0.5411, 0.1964], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0259, 0.0222, 0.0279, 0.0244, 0.0209, 0.0245, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:34:15,213 INFO [finetune.py:976] (1/7) Epoch 10, batch 800, loss[loss=0.1817, simple_loss=0.2382, pruned_loss=0.06258, over 4823.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2671, pruned_loss=0.06916, over 937864.12 frames. ], batch size: 30, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:34:20,443 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-26 11:34:30,551 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52373.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:34:49,107 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52396.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:34:50,910 INFO [finetune.py:976] (1/7) Epoch 10, batch 850, loss[loss=0.1914, simple_loss=0.2532, pruned_loss=0.06481, over 4804.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2646, pruned_loss=0.06805, over 941804.38 frames. ], batch size: 51, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:34:54,116 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52403.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:35:14,904 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.098e+02 1.556e+02 1.848e+02 2.239e+02 3.627e+02, threshold=3.695e+02, percent-clipped=0.0 +2023-03-26 11:35:17,326 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.0080, 4.3531, 4.5650, 4.8338, 4.7574, 4.4280, 5.0930, 1.4087], + device='cuda:1'), covar=tensor([0.0636, 0.0789, 0.0724, 0.0791, 0.0989, 0.1324, 0.0499, 0.5476], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0245, 0.0274, 0.0289, 0.0326, 0.0279, 0.0299, 0.0292], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:35:25,524 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-26 11:35:25,573 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 11:35:27,911 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 11:35:36,905 INFO [finetune.py:976] (1/7) Epoch 10, batch 900, loss[loss=0.1802, simple_loss=0.2439, pruned_loss=0.05824, over 4756.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.2605, pruned_loss=0.06634, over 945482.64 frames. ], batch size: 54, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:35:38,210 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=52451.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:36:19,948 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.31 vs. limit=5.0 +2023-03-26 11:36:25,193 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2386, 1.2216, 1.6232, 1.1003, 1.2626, 1.3776, 1.2522, 1.5384], + device='cuda:1'), covar=tensor([0.1253, 0.2214, 0.1329, 0.1565, 0.0986, 0.1336, 0.2888, 0.0934], + device='cuda:1'), in_proj_covar=tensor([0.0200, 0.0206, 0.0195, 0.0193, 0.0180, 0.0217, 0.0218, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:36:25,700 INFO [finetune.py:976] (1/7) Epoch 10, batch 950, loss[loss=0.2078, simple_loss=0.2792, pruned_loss=0.06816, over 4804.00 frames. ], tot_loss[loss=0.1961, simple_loss=0.2595, pruned_loss=0.0663, over 949349.03 frames. ], batch size: 45, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:36:31,236 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-26 11:36:43,923 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 11:36:46,733 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.124e+02 1.551e+02 1.918e+02 2.238e+02 5.409e+02, threshold=3.837e+02, percent-clipped=4.0 +2023-03-26 11:37:01,194 INFO [finetune.py:976] (1/7) Epoch 10, batch 1000, loss[loss=0.1976, simple_loss=0.2685, pruned_loss=0.06336, over 4802.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.264, pruned_loss=0.06848, over 952742.40 frames. ], batch size: 51, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:37:01,274 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52549.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:37:04,402 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 11:37:04,954 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52555.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 11:37:26,844 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-03-26 11:38:00,467 INFO [finetune.py:976] (1/7) Epoch 10, batch 1050, loss[loss=0.2098, simple_loss=0.2777, pruned_loss=0.07089, over 4898.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2657, pruned_loss=0.06813, over 951471.89 frames. ], batch size: 32, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:38:21,473 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52616.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 11:38:29,998 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-26 11:38:31,486 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.240e+02 1.591e+02 1.928e+02 2.293e+02 3.930e+02, threshold=3.855e+02, percent-clipped=1.0 +2023-03-26 11:38:43,397 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3037, 2.3968, 2.2744, 1.6827, 2.3038, 2.5771, 2.5077, 1.9521], + device='cuda:1'), covar=tensor([0.0579, 0.0537, 0.0741, 0.0900, 0.0599, 0.0698, 0.0572, 0.1017], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0132, 0.0142, 0.0123, 0.0117, 0.0141, 0.0141, 0.0159], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:38:44,946 INFO [finetune.py:976] (1/7) Epoch 10, batch 1100, loss[loss=0.1555, simple_loss=0.2176, pruned_loss=0.0467, over 4748.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2661, pruned_loss=0.06819, over 951602.91 frames. ], batch size: 23, lr: 3.76e-03, grad_scale: 32.0 +2023-03-26 11:38:59,648 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52668.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:39:10,602 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1737, 2.7650, 2.5168, 1.3813, 2.6312, 2.1257, 2.1059, 2.4190], + device='cuda:1'), covar=tensor([0.0874, 0.0967, 0.1899, 0.2360, 0.2069, 0.2761, 0.2389, 0.1404], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0203, 0.0204, 0.0190, 0.0218, 0.0210, 0.0226, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:39:14,446 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 11:39:16,690 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8652, 1.9543, 1.9589, 2.3665, 2.3775, 2.2573, 1.9337, 1.4811], + device='cuda:1'), covar=tensor([0.2131, 0.1898, 0.1594, 0.1388, 0.2139, 0.1051, 0.2209, 0.1846], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0207, 0.0206, 0.0187, 0.0240, 0.0180, 0.0212, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:39:17,750 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52696.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:39:19,452 INFO [finetune.py:976] (1/7) Epoch 10, batch 1150, loss[loss=0.1585, simple_loss=0.2366, pruned_loss=0.04021, over 4805.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2675, pruned_loss=0.06849, over 954902.18 frames. ], batch size: 25, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:39:19,708 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 11:39:40,535 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 11:39:40,840 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.071e+02 1.654e+02 1.930e+02 2.314e+02 4.484e+02, threshold=3.861e+02, percent-clipped=2.0 +2023-03-26 11:39:48,713 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=52744.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:39:52,581 INFO [finetune.py:976] (1/7) Epoch 10, batch 1200, loss[loss=0.2411, simple_loss=0.3029, pruned_loss=0.08964, over 4741.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2659, pruned_loss=0.06785, over 955527.84 frames. ], batch size: 23, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:39:56,846 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52752.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:40:00,365 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0766, 1.8933, 1.4856, 1.7226, 1.9226, 1.7867, 1.8347, 2.5562], + device='cuda:1'), covar=tensor([0.4514, 0.4930, 0.4038, 0.4911, 0.4359, 0.2907, 0.4530, 0.2048], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0260, 0.0222, 0.0279, 0.0243, 0.0209, 0.0246, 0.0213], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:40:23,856 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9582, 1.7604, 1.4837, 1.5163, 1.7116, 1.6769, 1.6987, 2.4258], + device='cuda:1'), covar=tensor([0.4429, 0.4940, 0.3876, 0.4459, 0.4321, 0.2774, 0.4507, 0.1925], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0260, 0.0222, 0.0279, 0.0243, 0.0209, 0.0246, 0.0213], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:40:35,761 INFO [finetune.py:976] (1/7) Epoch 10, batch 1250, loss[loss=0.2333, simple_loss=0.2911, pruned_loss=0.08772, over 4871.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.263, pruned_loss=0.06661, over 957020.88 frames. ], batch size: 31, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:40:47,077 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52813.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:40:49,488 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1129, 1.6345, 2.3272, 3.7775, 2.7231, 2.6519, 0.7267, 3.0234], + device='cuda:1'), covar=tensor([0.1598, 0.1541, 0.1422, 0.0550, 0.0741, 0.1725, 0.2119, 0.0547], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0135, 0.0165, 0.0102, 0.0139, 0.0126, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:41:05,425 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.045e+02 1.514e+02 1.794e+02 2.223e+02 4.744e+02, threshold=3.588e+02, percent-clipped=2.0 +2023-03-26 11:41:19,445 INFO [finetune.py:976] (1/7) Epoch 10, batch 1300, loss[loss=0.2128, simple_loss=0.256, pruned_loss=0.08478, over 4313.00 frames. ], tot_loss[loss=0.194, simple_loss=0.2588, pruned_loss=0.06463, over 957261.55 frames. ], batch size: 65, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:41:19,552 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52849.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:41:20,260 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-03-26 11:41:51,924 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=52897.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:41:53,102 INFO [finetune.py:976] (1/7) Epoch 10, batch 1350, loss[loss=0.1676, simple_loss=0.2385, pruned_loss=0.04829, over 4727.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.2582, pruned_loss=0.06465, over 956344.99 frames. ], batch size: 23, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:42:01,951 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52911.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 11:42:15,925 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.950e+01 1.660e+02 2.003e+02 2.564e+02 3.985e+02, threshold=4.006e+02, percent-clipped=2.0 +2023-03-26 11:42:16,056 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6745, 1.4454, 1.0445, 0.2564, 1.3534, 1.4274, 1.2957, 1.4075], + device='cuda:1'), covar=tensor([0.0795, 0.0807, 0.1150, 0.1864, 0.1244, 0.2176, 0.2343, 0.0811], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0202, 0.0203, 0.0189, 0.0216, 0.0208, 0.0224, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:42:30,878 INFO [finetune.py:976] (1/7) Epoch 10, batch 1400, loss[loss=0.2271, simple_loss=0.2927, pruned_loss=0.08068, over 4757.00 frames. ], tot_loss[loss=0.1983, simple_loss=0.2629, pruned_loss=0.06692, over 957804.41 frames. ], batch size: 27, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:42:37,905 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-03-26 11:42:48,550 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52968.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:42:49,304 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 11:43:11,423 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5889, 3.2673, 3.1913, 1.7874, 3.4426, 2.6696, 1.4121, 2.4207], + device='cuda:1'), covar=tensor([0.3175, 0.1925, 0.1574, 0.3014, 0.1112, 0.0908, 0.3497, 0.1369], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0175, 0.0161, 0.0129, 0.0157, 0.0123, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 11:43:14,418 INFO [finetune.py:976] (1/7) Epoch 10, batch 1450, loss[loss=0.2767, simple_loss=0.3187, pruned_loss=0.1173, over 4086.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2645, pruned_loss=0.06745, over 954170.06 frames. ], batch size: 65, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:43:22,107 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4507, 2.2641, 1.8542, 2.3936, 2.3423, 1.9905, 2.7683, 2.3216], + device='cuda:1'), covar=tensor([0.1380, 0.2754, 0.3573, 0.3323, 0.2815, 0.1837, 0.3602, 0.2117], + device='cuda:1'), in_proj_covar=tensor([0.0173, 0.0188, 0.0232, 0.0252, 0.0238, 0.0195, 0.0211, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:43:35,023 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=53016.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:43:45,115 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.022e+02 1.605e+02 1.913e+02 2.318e+02 4.347e+02, threshold=3.826e+02, percent-clipped=3.0 +2023-03-26 11:43:55,916 INFO [finetune.py:976] (1/7) Epoch 10, batch 1500, loss[loss=0.2048, simple_loss=0.2637, pruned_loss=0.07294, over 4804.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2651, pruned_loss=0.06766, over 955009.57 frames. ], batch size: 41, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:44:00,677 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53056.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:44:29,472 INFO [finetune.py:976] (1/7) Epoch 10, batch 1550, loss[loss=0.2429, simple_loss=0.2897, pruned_loss=0.09811, over 4804.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2645, pruned_loss=0.06718, over 955215.77 frames. ], batch size: 41, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:44:35,497 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53108.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:44:41,517 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53117.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:44:52,482 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.115e+02 1.652e+02 2.005e+02 2.543e+02 4.651e+02, threshold=4.009e+02, percent-clipped=4.0 +2023-03-26 11:45:03,284 INFO [finetune.py:976] (1/7) Epoch 10, batch 1600, loss[loss=0.1925, simple_loss=0.2682, pruned_loss=0.05837, over 4902.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2622, pruned_loss=0.06638, over 955596.20 frames. ], batch size: 36, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:45:48,099 INFO [finetune.py:976] (1/7) Epoch 10, batch 1650, loss[loss=0.1891, simple_loss=0.2579, pruned_loss=0.06019, over 4882.00 frames. ], tot_loss[loss=0.1954, simple_loss=0.2601, pruned_loss=0.06539, over 956290.40 frames. ], batch size: 35, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:45:50,825 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1392, 1.8805, 1.6875, 1.9399, 1.9089, 1.8344, 1.8690, 2.6344], + device='cuda:1'), covar=tensor([0.4550, 0.5416, 0.3971, 0.4663, 0.4482, 0.2880, 0.4698, 0.1791], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0259, 0.0222, 0.0279, 0.0243, 0.0208, 0.0244, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:45:52,559 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.1592, 4.4868, 4.7546, 4.9752, 4.8690, 4.6139, 5.2471, 1.5307], + device='cuda:1'), covar=tensor([0.0702, 0.0744, 0.0659, 0.0876, 0.1170, 0.1310, 0.0455, 0.5783], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0242, 0.0274, 0.0289, 0.0326, 0.0279, 0.0298, 0.0292], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:45:56,036 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53211.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 11:46:10,716 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.364e+01 1.592e+02 1.774e+02 2.189e+02 3.836e+02, threshold=3.549e+02, percent-clipped=0.0 +2023-03-26 11:46:16,360 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53241.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:46:23,558 INFO [finetune.py:976] (1/7) Epoch 10, batch 1700, loss[loss=0.2143, simple_loss=0.2801, pruned_loss=0.07428, over 4862.00 frames. ], tot_loss[loss=0.1932, simple_loss=0.2576, pruned_loss=0.06437, over 957026.72 frames. ], batch size: 34, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:46:29,716 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=53259.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 11:46:42,690 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7976, 1.5089, 2.1907, 3.5946, 2.4706, 2.4143, 1.2524, 2.7902], + device='cuda:1'), covar=tensor([0.1739, 0.1538, 0.1373, 0.0613, 0.0772, 0.1498, 0.1867, 0.0548], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0118, 0.0134, 0.0165, 0.0102, 0.0139, 0.0127, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:46:56,425 INFO [finetune.py:976] (1/7) Epoch 10, batch 1750, loss[loss=0.3, simple_loss=0.3314, pruned_loss=0.1343, over 4087.00 frames. ], tot_loss[loss=0.1949, simple_loss=0.2594, pruned_loss=0.06521, over 955703.94 frames. ], batch size: 65, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:46:58,860 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53302.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:47:00,081 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53304.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:47:07,251 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53315.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:47:09,893 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 11:47:18,948 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.172e+02 1.605e+02 1.832e+02 2.176e+02 4.638e+02, threshold=3.664e+02, percent-clipped=2.0 +2023-03-26 11:47:29,900 INFO [finetune.py:976] (1/7) Epoch 10, batch 1800, loss[loss=0.2244, simple_loss=0.2807, pruned_loss=0.08403, over 4184.00 frames. ], tot_loss[loss=0.1977, simple_loss=0.263, pruned_loss=0.06623, over 955660.09 frames. ], batch size: 65, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:47:45,397 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53365.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:47:46,639 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53367.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:47:56,028 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53376.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:47:58,562 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2824, 2.2972, 2.0464, 2.3233, 2.8441, 2.3066, 1.9821, 1.7728], + device='cuda:1'), covar=tensor([0.2116, 0.1986, 0.1808, 0.1591, 0.1751, 0.1091, 0.2351, 0.1906], + device='cuda:1'), in_proj_covar=tensor([0.0234, 0.0206, 0.0205, 0.0187, 0.0238, 0.0178, 0.0211, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:48:26,357 INFO [finetune.py:976] (1/7) Epoch 10, batch 1850, loss[loss=0.2175, simple_loss=0.2735, pruned_loss=0.08072, over 4823.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2651, pruned_loss=0.0671, over 956121.55 frames. ], batch size: 33, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:48:32,689 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53408.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:48:35,082 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53412.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:48:51,320 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53428.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 11:48:58,642 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.045e+02 1.752e+02 2.111e+02 2.637e+02 7.323e+02, threshold=4.222e+02, percent-clipped=6.0 +2023-03-26 11:48:59,410 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9189, 1.8027, 1.5363, 1.7142, 1.9294, 1.5907, 2.0927, 1.9000], + device='cuda:1'), covar=tensor([0.1513, 0.2396, 0.3307, 0.2916, 0.2815, 0.1807, 0.3528, 0.1919], + device='cuda:1'), in_proj_covar=tensor([0.0174, 0.0189, 0.0233, 0.0254, 0.0239, 0.0196, 0.0212, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:49:10,459 INFO [finetune.py:976] (1/7) Epoch 10, batch 1900, loss[loss=0.2009, simple_loss=0.2727, pruned_loss=0.06455, over 4923.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2671, pruned_loss=0.06813, over 957052.35 frames. ], batch size: 42, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:49:10,548 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2083, 3.6377, 3.8685, 4.0676, 4.0001, 3.7387, 4.2840, 1.3740], + device='cuda:1'), covar=tensor([0.0820, 0.0795, 0.0892, 0.0975, 0.1200, 0.1474, 0.0756, 0.5330], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0246, 0.0277, 0.0294, 0.0330, 0.0283, 0.0303, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:49:14,804 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=53456.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:49:26,202 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8462, 1.7012, 1.6214, 1.7925, 1.5337, 4.3899, 1.7378, 2.2630], + device='cuda:1'), covar=tensor([0.3271, 0.2412, 0.2116, 0.2223, 0.1609, 0.0136, 0.2477, 0.1231], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0123, 0.0116, 0.0099, 0.0100, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 11:49:43,866 INFO [finetune.py:976] (1/7) Epoch 10, batch 1950, loss[loss=0.1783, simple_loss=0.2447, pruned_loss=0.05599, over 4857.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2647, pruned_loss=0.06671, over 957237.62 frames. ], batch size: 44, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:50:09,883 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.065e+02 1.535e+02 1.778e+02 2.101e+02 3.650e+02, threshold=3.555e+02, percent-clipped=0.0 +2023-03-26 11:50:29,329 INFO [finetune.py:976] (1/7) Epoch 10, batch 2000, loss[loss=0.1757, simple_loss=0.2442, pruned_loss=0.05362, over 4765.00 frames. ], tot_loss[loss=0.1969, simple_loss=0.2618, pruned_loss=0.06601, over 956387.83 frames. ], batch size: 27, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:50:30,081 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8602, 1.0688, 1.7612, 1.6759, 1.5410, 1.4748, 1.5924, 1.6508], + device='cuda:1'), covar=tensor([0.3890, 0.4366, 0.3740, 0.4034, 0.5012, 0.3916, 0.4881, 0.3622], + device='cuda:1'), in_proj_covar=tensor([0.0235, 0.0240, 0.0253, 0.0256, 0.0251, 0.0227, 0.0274, 0.0229], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:50:31,845 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8459, 1.7453, 1.5683, 1.9063, 2.3746, 1.9262, 1.5568, 1.4546], + device='cuda:1'), covar=tensor([0.2143, 0.2028, 0.1876, 0.1615, 0.1774, 0.1234, 0.2491, 0.1922], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0205, 0.0204, 0.0186, 0.0238, 0.0178, 0.0210, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:50:42,549 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-26 11:51:21,830 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53597.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:51:23,501 INFO [finetune.py:976] (1/7) Epoch 10, batch 2050, loss[loss=0.1827, simple_loss=0.2438, pruned_loss=0.06085, over 4745.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2591, pruned_loss=0.06542, over 957667.18 frames. ], batch size: 54, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:51:34,196 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3391, 2.0633, 1.7311, 0.8269, 1.8341, 1.8216, 1.6683, 1.9779], + device='cuda:1'), covar=tensor([0.0745, 0.0752, 0.1323, 0.1880, 0.1382, 0.2033, 0.2105, 0.0826], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0202, 0.0202, 0.0188, 0.0217, 0.0208, 0.0224, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:51:44,830 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.079e+02 1.628e+02 1.908e+02 2.249e+02 5.707e+02, threshold=3.816e+02, percent-clipped=1.0 +2023-03-26 11:51:56,177 INFO [finetune.py:976] (1/7) Epoch 10, batch 2100, loss[loss=0.2601, simple_loss=0.3108, pruned_loss=0.1047, over 4925.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2583, pruned_loss=0.06522, over 957438.57 frames. ], batch size: 38, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:52:03,970 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53660.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:52:10,623 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53671.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:52:37,544 INFO [finetune.py:976] (1/7) Epoch 10, batch 2150, loss[loss=0.2088, simple_loss=0.2939, pruned_loss=0.0618, over 4932.00 frames. ], tot_loss[loss=0.1974, simple_loss=0.2619, pruned_loss=0.06646, over 957841.96 frames. ], batch size: 38, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:52:49,658 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2121, 1.2165, 1.6033, 1.0969, 1.2939, 1.4188, 1.2446, 1.5852], + device='cuda:1'), covar=tensor([0.1325, 0.2238, 0.1254, 0.1488, 0.0936, 0.1239, 0.2801, 0.0942], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0203, 0.0193, 0.0191, 0.0177, 0.0214, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:52:50,926 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7264, 0.7120, 1.6704, 1.5164, 1.4886, 1.4199, 1.4205, 1.5612], + device='cuda:1'), covar=tensor([0.3099, 0.4025, 0.3472, 0.3701, 0.4589, 0.3431, 0.4285, 0.3274], + device='cuda:1'), in_proj_covar=tensor([0.0233, 0.0238, 0.0252, 0.0255, 0.0249, 0.0226, 0.0272, 0.0228], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:52:52,702 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53712.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:52:52,740 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4563, 2.3819, 1.9861, 1.0014, 2.1274, 1.8300, 1.6716, 2.1349], + device='cuda:1'), covar=tensor([0.0957, 0.0803, 0.1686, 0.2214, 0.1781, 0.2560, 0.2487, 0.1166], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0203, 0.0203, 0.0189, 0.0218, 0.0209, 0.0225, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:53:03,977 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53723.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 11:53:19,770 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.135e+02 1.764e+02 2.057e+02 2.459e+02 5.535e+02, threshold=4.114e+02, percent-clipped=2.0 +2023-03-26 11:53:19,871 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.9779, 3.4114, 3.6073, 3.8446, 3.7230, 3.4648, 4.0121, 1.4269], + device='cuda:1'), covar=tensor([0.0684, 0.0833, 0.0747, 0.0780, 0.1001, 0.1394, 0.0697, 0.5049], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0242, 0.0274, 0.0288, 0.0325, 0.0279, 0.0299, 0.0292], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:53:34,458 INFO [finetune.py:976] (1/7) Epoch 10, batch 2200, loss[loss=0.1874, simple_loss=0.2572, pruned_loss=0.05875, over 4855.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2651, pruned_loss=0.06789, over 956956.00 frames. ], batch size: 31, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:53:43,155 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=53760.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:54:07,986 INFO [finetune.py:976] (1/7) Epoch 10, batch 2250, loss[loss=0.1638, simple_loss=0.2282, pruned_loss=0.04974, over 4750.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2655, pruned_loss=0.06749, over 956959.55 frames. ], batch size: 23, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:54:30,209 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.107e+02 1.658e+02 1.958e+02 2.430e+02 3.560e+02, threshold=3.915e+02, percent-clipped=0.0 +2023-03-26 11:54:41,559 INFO [finetune.py:976] (1/7) Epoch 10, batch 2300, loss[loss=0.1903, simple_loss=0.2661, pruned_loss=0.05719, over 4766.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2652, pruned_loss=0.06679, over 957328.80 frames. ], batch size: 28, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:54:50,378 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4376, 1.4004, 1.6616, 2.4950, 1.7339, 2.1961, 0.9077, 2.0546], + device='cuda:1'), covar=tensor([0.1669, 0.1349, 0.1098, 0.0750, 0.0861, 0.1094, 0.1578, 0.0673], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0134, 0.0164, 0.0101, 0.0138, 0.0127, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:55:15,990 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53897.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:55:17,109 INFO [finetune.py:976] (1/7) Epoch 10, batch 2350, loss[loss=0.1913, simple_loss=0.2609, pruned_loss=0.06081, over 4781.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2634, pruned_loss=0.06584, over 958062.11 frames. ], batch size: 29, lr: 3.75e-03, grad_scale: 32.0 +2023-03-26 11:55:47,268 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.820e+01 1.627e+02 1.969e+02 2.442e+02 4.599e+02, threshold=3.938e+02, percent-clipped=2.0 +2023-03-26 11:55:58,276 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=53945.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:56:05,529 INFO [finetune.py:976] (1/7) Epoch 10, batch 2400, loss[loss=0.2099, simple_loss=0.2602, pruned_loss=0.07978, over 4818.00 frames. ], tot_loss[loss=0.1958, simple_loss=0.2605, pruned_loss=0.06551, over 955625.69 frames. ], batch size: 41, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 11:56:15,984 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53960.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:56:24,610 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53971.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:56:36,686 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 11:56:38,459 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.45 vs. limit=2.0 +2023-03-26 11:56:41,931 INFO [finetune.py:976] (1/7) Epoch 10, batch 2450, loss[loss=0.2159, simple_loss=0.2676, pruned_loss=0.08209, over 4094.00 frames. ], tot_loss[loss=0.1942, simple_loss=0.2582, pruned_loss=0.06516, over 954527.21 frames. ], batch size: 65, lr: 3.74e-03, grad_scale: 64.0 +2023-03-26 11:56:49,144 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=54008.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:56:56,921 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=54019.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:56:59,868 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=54023.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 11:57:05,184 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.081e+02 1.615e+02 1.942e+02 2.268e+02 4.833e+02, threshold=3.884e+02, percent-clipped=2.0 +2023-03-26 11:57:16,019 INFO [finetune.py:976] (1/7) Epoch 10, batch 2500, loss[loss=0.2074, simple_loss=0.2709, pruned_loss=0.07198, over 4127.00 frames. ], tot_loss[loss=0.1939, simple_loss=0.2581, pruned_loss=0.06486, over 954837.22 frames. ], batch size: 65, lr: 3.74e-03, grad_scale: 64.0 +2023-03-26 11:57:25,850 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-03-26 11:57:42,139 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=54071.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 11:58:00,118 INFO [finetune.py:976] (1/7) Epoch 10, batch 2550, loss[loss=0.218, simple_loss=0.2694, pruned_loss=0.08325, over 4811.00 frames. ], tot_loss[loss=0.1972, simple_loss=0.2622, pruned_loss=0.06612, over 953328.40 frames. ], batch size: 40, lr: 3.74e-03, grad_scale: 64.0 +2023-03-26 11:58:00,243 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5181, 1.5596, 1.9859, 1.2043, 1.5659, 1.8690, 1.5173, 2.0978], + device='cuda:1'), covar=tensor([0.1252, 0.2108, 0.1236, 0.1724, 0.0935, 0.1244, 0.2653, 0.0868], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0203, 0.0192, 0.0191, 0.0177, 0.0215, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:58:01,487 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7138, 2.4619, 2.3186, 1.3444, 2.4501, 1.9995, 1.8294, 2.3341], + device='cuda:1'), covar=tensor([0.1108, 0.0868, 0.1779, 0.2041, 0.1617, 0.2167, 0.2235, 0.1039], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0200, 0.0201, 0.0187, 0.0215, 0.0207, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:58:02,106 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7433, 1.6858, 1.5441, 1.8746, 2.2982, 1.8430, 1.4800, 1.4594], + device='cuda:1'), covar=tensor([0.2181, 0.2012, 0.1860, 0.1588, 0.1754, 0.1217, 0.2490, 0.1906], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0208, 0.0208, 0.0189, 0.0241, 0.0181, 0.0213, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 11:58:31,770 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.51 vs. limit=5.0 +2023-03-26 11:58:35,815 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.936e+01 1.671e+02 2.051e+02 2.356e+02 3.900e+02, threshold=4.103e+02, percent-clipped=1.0 +2023-03-26 11:58:46,742 INFO [finetune.py:976] (1/7) Epoch 10, batch 2600, loss[loss=0.2078, simple_loss=0.2785, pruned_loss=0.0685, over 4892.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2651, pruned_loss=0.06708, over 954741.94 frames. ], batch size: 43, lr: 3.74e-03, grad_scale: 64.0 +2023-03-26 11:59:19,472 INFO [finetune.py:976] (1/7) Epoch 10, batch 2650, loss[loss=0.2205, simple_loss=0.2846, pruned_loss=0.07821, over 4795.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.266, pruned_loss=0.06714, over 954566.18 frames. ], batch size: 51, lr: 3.74e-03, grad_scale: 64.0 +2023-03-26 11:59:43,742 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.585e+01 1.566e+02 1.779e+02 2.159e+02 3.883e+02, threshold=3.557e+02, percent-clipped=0.0 +2023-03-26 11:59:49,930 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2968, 1.7610, 2.4923, 4.0516, 2.9410, 2.7079, 0.6481, 3.1640], + device='cuda:1'), covar=tensor([0.1558, 0.1516, 0.1455, 0.0543, 0.0637, 0.1466, 0.2209, 0.0483], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0118, 0.0135, 0.0166, 0.0102, 0.0138, 0.0128, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 11:59:53,473 INFO [finetune.py:976] (1/7) Epoch 10, batch 2700, loss[loss=0.1889, simple_loss=0.2536, pruned_loss=0.06213, over 4759.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.265, pruned_loss=0.0663, over 954553.37 frames. ], batch size: 26, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:00:20,006 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.2325, 4.5265, 4.8166, 5.0670, 4.9427, 4.7534, 5.3772, 1.6813], + device='cuda:1'), covar=tensor([0.0781, 0.0842, 0.0713, 0.0832, 0.1270, 0.1503, 0.0539, 0.5303], + device='cuda:1'), in_proj_covar=tensor([0.0345, 0.0243, 0.0274, 0.0287, 0.0325, 0.0279, 0.0299, 0.0291], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:00:26,576 INFO [finetune.py:976] (1/7) Epoch 10, batch 2750, loss[loss=0.1941, simple_loss=0.259, pruned_loss=0.06457, over 4921.00 frames. ], tot_loss[loss=0.1957, simple_loss=0.2614, pruned_loss=0.06501, over 956547.98 frames. ], batch size: 33, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:00:50,902 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.123e+02 1.545e+02 1.929e+02 2.415e+02 3.548e+02, threshold=3.859e+02, percent-clipped=0.0 +2023-03-26 12:00:56,306 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1869, 2.1776, 2.4081, 1.4418, 2.2990, 2.4335, 2.3391, 1.8994], + device='cuda:1'), covar=tensor([0.0599, 0.0554, 0.0555, 0.0910, 0.0547, 0.0570, 0.0534, 0.0969], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0132, 0.0142, 0.0123, 0.0118, 0.0141, 0.0141, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:00:56,335 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9621, 1.6796, 1.5722, 1.9755, 2.4002, 2.1094, 1.5406, 1.5900], + device='cuda:1'), covar=tensor([0.2182, 0.2282, 0.2088, 0.1840, 0.1815, 0.1144, 0.2612, 0.2035], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0209, 0.0209, 0.0190, 0.0242, 0.0181, 0.0214, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:01:01,566 INFO [finetune.py:976] (1/7) Epoch 10, batch 2800, loss[loss=0.1745, simple_loss=0.2305, pruned_loss=0.05927, over 4815.00 frames. ], tot_loss[loss=0.1929, simple_loss=0.2577, pruned_loss=0.06399, over 955605.78 frames. ], batch size: 25, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:01:48,148 INFO [finetune.py:976] (1/7) Epoch 10, batch 2850, loss[loss=0.1774, simple_loss=0.2439, pruned_loss=0.05541, over 4818.00 frames. ], tot_loss[loss=0.1908, simple_loss=0.2557, pruned_loss=0.06302, over 956194.57 frames. ], batch size: 38, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:02:10,454 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.958e+01 1.628e+02 1.894e+02 2.190e+02 3.699e+02, threshold=3.787e+02, percent-clipped=0.0 +2023-03-26 12:02:22,196 INFO [finetune.py:976] (1/7) Epoch 10, batch 2900, loss[loss=0.1972, simple_loss=0.2709, pruned_loss=0.06176, over 4832.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2583, pruned_loss=0.06412, over 955556.86 frames. ], batch size: 30, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:02:57,317 INFO [finetune.py:976] (1/7) Epoch 10, batch 2950, loss[loss=0.174, simple_loss=0.248, pruned_loss=0.05005, over 4760.00 frames. ], tot_loss[loss=0.1965, simple_loss=0.2619, pruned_loss=0.06551, over 954675.58 frames. ], batch size: 28, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:03:18,744 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.146e+02 1.694e+02 2.010e+02 2.318e+02 4.609e+02, threshold=4.019e+02, percent-clipped=2.0 +2023-03-26 12:03:30,484 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2204, 2.0344, 1.8845, 2.3048, 2.0233, 2.0324, 1.9194, 2.6207], + device='cuda:1'), covar=tensor([0.4198, 0.5816, 0.3463, 0.4951, 0.5198, 0.2612, 0.5487, 0.1754], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0259, 0.0222, 0.0279, 0.0243, 0.0208, 0.0246, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:03:40,002 INFO [finetune.py:976] (1/7) Epoch 10, batch 3000, loss[loss=0.1759, simple_loss=0.2503, pruned_loss=0.05071, over 4885.00 frames. ], tot_loss[loss=0.1977, simple_loss=0.2634, pruned_loss=0.066, over 953523.70 frames. ], batch size: 43, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:03:40,002 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 12:03:56,631 INFO [finetune.py:1010] (1/7) Epoch 10, validation: loss=0.1584, simple_loss=0.2295, pruned_loss=0.04366, over 2265189.00 frames. +2023-03-26 12:03:56,632 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6339MB +2023-03-26 12:04:11,721 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1287, 3.5561, 3.7311, 3.9627, 3.8603, 3.6688, 4.2025, 1.2726], + device='cuda:1'), covar=tensor([0.0864, 0.0813, 0.0861, 0.1062, 0.1429, 0.1513, 0.0703, 0.5754], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0246, 0.0278, 0.0292, 0.0331, 0.0283, 0.0303, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:04:29,069 INFO [finetune.py:976] (1/7) Epoch 10, batch 3050, loss[loss=0.1953, simple_loss=0.242, pruned_loss=0.07425, over 4089.00 frames. ], tot_loss[loss=0.1978, simple_loss=0.2636, pruned_loss=0.06604, over 953760.78 frames. ], batch size: 17, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:04:52,092 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.154e+02 1.606e+02 1.839e+02 2.259e+02 4.011e+02, threshold=3.679e+02, percent-clipped=0.0 +2023-03-26 12:04:55,875 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-26 12:05:02,811 INFO [finetune.py:976] (1/7) Epoch 10, batch 3100, loss[loss=0.1862, simple_loss=0.2483, pruned_loss=0.062, over 4818.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.2618, pruned_loss=0.06569, over 954590.77 frames. ], batch size: 39, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:05:14,411 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=54664.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 12:05:36,458 INFO [finetune.py:976] (1/7) Epoch 10, batch 3150, loss[loss=0.168, simple_loss=0.2224, pruned_loss=0.05677, over 4244.00 frames. ], tot_loss[loss=0.1939, simple_loss=0.2586, pruned_loss=0.06461, over 954415.69 frames. ], batch size: 18, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:05:40,007 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6274, 1.6373, 1.9503, 1.8988, 1.6983, 3.7384, 1.4234, 1.7399], + device='cuda:1'), covar=tensor([0.1015, 0.1835, 0.1026, 0.0983, 0.1577, 0.0220, 0.1558, 0.1752], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0091, 0.0082, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:05:49,801 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0787, 1.0242, 1.0013, 0.5498, 0.8248, 1.1702, 1.2421, 1.0051], + device='cuda:1'), covar=tensor([0.0828, 0.0490, 0.0450, 0.0466, 0.0490, 0.0550, 0.0320, 0.0638], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0153, 0.0120, 0.0132, 0.0130, 0.0123, 0.0143, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.4124e-05, 1.1239e-04, 8.6576e-05, 9.5312e-05, 9.3182e-05, 9.0071e-05, + 1.0507e-04, 1.0720e-04], device='cuda:1') +2023-03-26 12:05:54,671 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=54725.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 12:05:59,383 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.009e+02 1.653e+02 1.993e+02 2.298e+02 5.311e+02, threshold=3.986e+02, percent-clipped=2.0 +2023-03-26 12:06:10,110 INFO [finetune.py:976] (1/7) Epoch 10, batch 3200, loss[loss=0.1432, simple_loss=0.2146, pruned_loss=0.03588, over 4819.00 frames. ], tot_loss[loss=0.1901, simple_loss=0.2547, pruned_loss=0.06278, over 955592.70 frames. ], batch size: 38, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:06:53,307 INFO [finetune.py:976] (1/7) Epoch 10, batch 3250, loss[loss=0.2216, simple_loss=0.2765, pruned_loss=0.0834, over 4933.00 frames. ], tot_loss[loss=0.192, simple_loss=0.256, pruned_loss=0.06401, over 953428.87 frames. ], batch size: 33, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:07:26,174 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.100e+02 1.711e+02 2.094e+02 2.546e+02 5.601e+02, threshold=4.189e+02, percent-clipped=2.0 +2023-03-26 12:07:46,362 INFO [finetune.py:976] (1/7) Epoch 10, batch 3300, loss[loss=0.216, simple_loss=0.2744, pruned_loss=0.0788, over 4928.00 frames. ], tot_loss[loss=0.1959, simple_loss=0.2603, pruned_loss=0.06578, over 952307.92 frames. ], batch size: 33, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:07:50,459 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5456, 1.3169, 1.9307, 3.1940, 2.1960, 2.4228, 0.8733, 2.5527], + device='cuda:1'), covar=tensor([0.2107, 0.2153, 0.1731, 0.0941, 0.0928, 0.1704, 0.2256, 0.0762], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0135, 0.0165, 0.0101, 0.0138, 0.0127, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:07:55,829 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=54862.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:08:20,116 INFO [finetune.py:976] (1/7) Epoch 10, batch 3350, loss[loss=0.2189, simple_loss=0.2835, pruned_loss=0.07713, over 4218.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2621, pruned_loss=0.06709, over 949525.00 frames. ], batch size: 66, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:08:25,085 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.74 vs. limit=2.0 +2023-03-26 12:08:47,742 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=54923.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 12:08:57,774 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.164e+02 1.693e+02 1.965e+02 2.442e+02 4.084e+02, threshold=3.930e+02, percent-clipped=0.0 +2023-03-26 12:09:07,543 INFO [finetune.py:976] (1/7) Epoch 10, batch 3400, loss[loss=0.2081, simple_loss=0.2797, pruned_loss=0.0683, over 4763.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.2639, pruned_loss=0.06719, over 951429.71 frames. ], batch size: 26, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:09:10,672 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6002, 3.0210, 2.5150, 1.9379, 2.8051, 2.8718, 2.9218, 2.5606], + device='cuda:1'), covar=tensor([0.0722, 0.0592, 0.0805, 0.0928, 0.0575, 0.0790, 0.0618, 0.0923], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0133, 0.0143, 0.0123, 0.0119, 0.0142, 0.0142, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:09:38,059 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8881, 1.7615, 1.5087, 1.8897, 2.2002, 1.9251, 1.4262, 1.4943], + device='cuda:1'), covar=tensor([0.2015, 0.1930, 0.1894, 0.1540, 0.1715, 0.1136, 0.2552, 0.1903], + device='cuda:1'), in_proj_covar=tensor([0.0235, 0.0207, 0.0206, 0.0188, 0.0239, 0.0180, 0.0213, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:09:56,947 INFO [finetune.py:976] (1/7) Epoch 10, batch 3450, loss[loss=0.1633, simple_loss=0.226, pruned_loss=0.05028, over 4773.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2651, pruned_loss=0.06764, over 952706.04 frames. ], batch size: 26, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:10:16,050 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55020.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 12:10:18,006 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-03-26 12:10:30,764 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.349e+01 1.534e+02 1.957e+02 2.350e+02 5.428e+02, threshold=3.914e+02, percent-clipped=3.0 +2023-03-26 12:10:51,432 INFO [finetune.py:976] (1/7) Epoch 10, batch 3500, loss[loss=0.1921, simple_loss=0.2535, pruned_loss=0.06536, over 4768.00 frames. ], tot_loss[loss=0.1969, simple_loss=0.2618, pruned_loss=0.066, over 954171.90 frames. ], batch size: 27, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:11:03,487 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-26 12:11:36,218 INFO [finetune.py:976] (1/7) Epoch 10, batch 3550, loss[loss=0.2075, simple_loss=0.2605, pruned_loss=0.07728, over 4750.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2581, pruned_loss=0.06432, over 953948.72 frames. ], batch size: 23, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:11:37,523 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1977, 1.2231, 1.1262, 1.2404, 1.5038, 1.4046, 1.2517, 1.1291], + device='cuda:1'), covar=tensor([0.0385, 0.0296, 0.0648, 0.0315, 0.0228, 0.0494, 0.0380, 0.0445], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0109, 0.0139, 0.0114, 0.0101, 0.0103, 0.0092, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.0927e-05, 8.5410e-05, 1.1063e-04, 8.9626e-05, 7.9314e-05, 7.6471e-05, + 6.9713e-05, 8.2996e-05], device='cuda:1') +2023-03-26 12:11:58,141 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2489, 2.0900, 1.6474, 2.0613, 2.0526, 1.9291, 1.9586, 2.7854], + device='cuda:1'), covar=tensor([0.4464, 0.5206, 0.3945, 0.4819, 0.4562, 0.2872, 0.4612, 0.1923], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0259, 0.0222, 0.0278, 0.0242, 0.0209, 0.0244, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:11:58,577 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.369e+01 1.530e+02 1.896e+02 2.280e+02 4.793e+02, threshold=3.791e+02, percent-clipped=5.0 +2023-03-26 12:12:09,210 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55148.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:12:09,742 INFO [finetune.py:976] (1/7) Epoch 10, batch 3600, loss[loss=0.149, simple_loss=0.2251, pruned_loss=0.03642, over 4750.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.256, pruned_loss=0.06386, over 955531.95 frames. ], batch size: 27, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:12:43,419 INFO [finetune.py:976] (1/7) Epoch 10, batch 3650, loss[loss=0.3009, simple_loss=0.3425, pruned_loss=0.1296, over 4814.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2586, pruned_loss=0.06497, over 954625.59 frames. ], batch size: 38, lr: 3.74e-03, grad_scale: 32.0 +2023-03-26 12:12:49,689 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55209.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:12:50,912 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55211.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:12:56,767 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55218.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 12:13:05,140 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7569, 1.6359, 1.5766, 1.5601, 1.9118, 1.8946, 1.7296, 1.5300], + device='cuda:1'), covar=tensor([0.0303, 0.0312, 0.0449, 0.0326, 0.0256, 0.0471, 0.0357, 0.0392], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0109, 0.0138, 0.0114, 0.0101, 0.0102, 0.0091, 0.0107], + device='cuda:1'), out_proj_covar=tensor([7.0504e-05, 8.4845e-05, 1.0998e-04, 8.9048e-05, 7.8876e-05, 7.5876e-05, + 6.9269e-05, 8.2553e-05], device='cuda:1') +2023-03-26 12:13:14,906 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.143e+02 1.616e+02 1.938e+02 2.270e+02 4.700e+02, threshold=3.875e+02, percent-clipped=1.0 +2023-03-26 12:13:26,509 INFO [finetune.py:976] (1/7) Epoch 10, batch 3700, loss[loss=0.1593, simple_loss=0.2171, pruned_loss=0.0508, over 4746.00 frames. ], tot_loss[loss=0.1972, simple_loss=0.2625, pruned_loss=0.06592, over 954176.07 frames. ], batch size: 23, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:13:40,332 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55272.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:14:00,006 INFO [finetune.py:976] (1/7) Epoch 10, batch 3750, loss[loss=0.2393, simple_loss=0.2902, pruned_loss=0.09425, over 4888.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2646, pruned_loss=0.06659, over 955386.74 frames. ], batch size: 43, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:14:16,773 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55320.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 12:14:33,807 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.113e+02 1.583e+02 1.835e+02 2.150e+02 3.880e+02, threshold=3.669e+02, percent-clipped=1.0 +2023-03-26 12:14:45,561 INFO [finetune.py:976] (1/7) Epoch 10, batch 3800, loss[loss=0.2193, simple_loss=0.2787, pruned_loss=0.07995, over 4910.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2656, pruned_loss=0.0669, over 956002.37 frames. ], batch size: 43, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:14:51,160 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1474, 1.3666, 0.9398, 2.1032, 2.4722, 1.8690, 1.6480, 1.8031], + device='cuda:1'), covar=tensor([0.1494, 0.2152, 0.2131, 0.1208, 0.1805, 0.1916, 0.1499, 0.2152], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0114, 0.0093, 0.0123, 0.0096, 0.0100, 0.0092], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:14:56,063 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1654, 1.8914, 1.9673, 0.8925, 2.2106, 2.4644, 2.0303, 1.8553], + device='cuda:1'), covar=tensor([0.1080, 0.0800, 0.0570, 0.0752, 0.0610, 0.0676, 0.0472, 0.0746], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0156, 0.0123, 0.0134, 0.0133, 0.0125, 0.0146, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.5699e-05, 1.1444e-04, 8.8409e-05, 9.7406e-05, 9.4845e-05, 9.1032e-05, + 1.0700e-04, 1.0807e-04], device='cuda:1') +2023-03-26 12:14:57,816 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=55368.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 12:15:27,044 INFO [finetune.py:976] (1/7) Epoch 10, batch 3850, loss[loss=0.1591, simple_loss=0.2442, pruned_loss=0.03703, over 4808.00 frames. ], tot_loss[loss=0.1971, simple_loss=0.2632, pruned_loss=0.06554, over 955483.90 frames. ], batch size: 45, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:15:49,874 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.001e+02 1.578e+02 1.920e+02 2.344e+02 4.809e+02, threshold=3.839e+02, percent-clipped=3.0 +2023-03-26 12:16:01,518 INFO [finetune.py:976] (1/7) Epoch 10, batch 3900, loss[loss=0.1848, simple_loss=0.251, pruned_loss=0.05934, over 4872.00 frames. ], tot_loss[loss=0.1946, simple_loss=0.2603, pruned_loss=0.06445, over 957481.26 frames. ], batch size: 34, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:16:07,869 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3749, 1.2895, 1.5884, 2.4330, 1.6533, 2.1467, 0.9535, 2.0654], + device='cuda:1'), covar=tensor([0.1706, 0.1501, 0.1215, 0.0782, 0.0903, 0.1140, 0.1559, 0.0624], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0136, 0.0166, 0.0102, 0.0139, 0.0127, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:16:23,423 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2664, 1.9766, 2.4808, 4.1544, 2.9678, 2.8599, 0.7807, 3.4404], + device='cuda:1'), covar=tensor([0.1608, 0.1346, 0.1479, 0.0497, 0.0654, 0.1396, 0.2107, 0.0485], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0136, 0.0166, 0.0102, 0.0139, 0.0127, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:16:44,728 INFO [finetune.py:976] (1/7) Epoch 10, batch 3950, loss[loss=0.1569, simple_loss=0.2122, pruned_loss=0.05078, over 4260.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.257, pruned_loss=0.06299, over 957670.03 frames. ], batch size: 18, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:16:48,768 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55504.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:16:58,347 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55518.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 12:16:58,375 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9827, 1.9287, 1.9200, 1.2993, 2.0941, 2.0333, 1.9946, 1.7131], + device='cuda:1'), covar=tensor([0.0662, 0.0725, 0.0848, 0.1033, 0.0654, 0.0839, 0.0755, 0.1127], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0135, 0.0145, 0.0125, 0.0121, 0.0144, 0.0144, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:17:13,743 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.106e+02 1.580e+02 1.857e+02 2.217e+02 3.906e+02, threshold=3.714e+02, percent-clipped=1.0 +2023-03-26 12:17:35,771 INFO [finetune.py:976] (1/7) Epoch 10, batch 4000, loss[loss=0.1835, simple_loss=0.272, pruned_loss=0.04754, over 4815.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.2572, pruned_loss=0.06363, over 958090.95 frames. ], batch size: 41, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:17:35,930 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9463, 1.3822, 1.8588, 1.9113, 1.6538, 1.5997, 1.7960, 1.7338], + device='cuda:1'), covar=tensor([0.4215, 0.4615, 0.3922, 0.4088, 0.5539, 0.4204, 0.5413, 0.3766], + device='cuda:1'), in_proj_covar=tensor([0.0235, 0.0239, 0.0253, 0.0257, 0.0251, 0.0229, 0.0273, 0.0230], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:17:48,291 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=55566.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:17:48,908 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55567.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:17:58,110 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 12:18:09,105 INFO [finetune.py:976] (1/7) Epoch 10, batch 4050, loss[loss=0.1454, simple_loss=0.2117, pruned_loss=0.03958, over 4712.00 frames. ], tot_loss[loss=0.1962, simple_loss=0.2613, pruned_loss=0.06561, over 955441.74 frames. ], batch size: 23, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:18:34,628 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.120e+02 1.736e+02 2.138e+02 2.510e+02 4.140e+02, threshold=4.276e+02, percent-clipped=4.0 +2023-03-26 12:18:44,820 INFO [finetune.py:976] (1/7) Epoch 10, batch 4100, loss[loss=0.1988, simple_loss=0.2758, pruned_loss=0.06091, over 4901.00 frames. ], tot_loss[loss=0.1973, simple_loss=0.2634, pruned_loss=0.06559, over 956452.16 frames. ], batch size: 35, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:19:12,854 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9384, 1.8494, 1.5224, 1.8344, 1.6977, 1.6510, 1.7361, 2.4687], + device='cuda:1'), covar=tensor([0.4440, 0.4379, 0.3733, 0.4313, 0.4650, 0.2880, 0.4373, 0.1852], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0259, 0.0222, 0.0278, 0.0242, 0.0209, 0.0245, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:19:17,503 INFO [finetune.py:976] (1/7) Epoch 10, batch 4150, loss[loss=0.2029, simple_loss=0.2728, pruned_loss=0.06652, over 4921.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2654, pruned_loss=0.06669, over 957074.45 frames. ], batch size: 33, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:19:49,995 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.152e+02 1.688e+02 2.035e+02 2.462e+02 3.895e+02, threshold=4.069e+02, percent-clipped=0.0 +2023-03-26 12:19:59,095 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55748.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:19:59,614 INFO [finetune.py:976] (1/7) Epoch 10, batch 4200, loss[loss=0.2691, simple_loss=0.3116, pruned_loss=0.1133, over 4223.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2662, pruned_loss=0.06714, over 953428.23 frames. ], batch size: 65, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:20:53,374 INFO [finetune.py:976] (1/7) Epoch 10, batch 4250, loss[loss=0.1624, simple_loss=0.2372, pruned_loss=0.04377, over 4786.00 frames. ], tot_loss[loss=0.1982, simple_loss=0.2638, pruned_loss=0.06627, over 954080.26 frames. ], batch size: 51, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:20:57,631 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55804.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:21:06,133 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55809.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:21:38,837 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.929e+01 1.631e+02 1.908e+02 2.201e+02 4.056e+02, threshold=3.816e+02, percent-clipped=0.0 +2023-03-26 12:21:48,082 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55848.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:21:48,566 INFO [finetune.py:976] (1/7) Epoch 10, batch 4300, loss[loss=0.1741, simple_loss=0.2363, pruned_loss=0.05593, over 4724.00 frames. ], tot_loss[loss=0.1962, simple_loss=0.2607, pruned_loss=0.0658, over 954307.55 frames. ], batch size: 23, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:21:49,298 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1860, 2.1092, 1.6843, 1.9844, 2.1503, 1.8434, 2.4234, 2.1296], + device='cuda:1'), covar=tensor([0.1362, 0.2375, 0.3291, 0.2982, 0.2599, 0.1649, 0.3431, 0.1944], + device='cuda:1'), in_proj_covar=tensor([0.0175, 0.0188, 0.0234, 0.0254, 0.0240, 0.0197, 0.0213, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:21:50,369 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=55852.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:22:10,869 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55867.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:22:36,505 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0366, 2.7976, 2.4152, 3.0510, 2.9261, 2.6455, 3.3640, 2.8996], + device='cuda:1'), covar=tensor([0.1197, 0.2334, 0.3134, 0.2933, 0.2537, 0.1571, 0.2674, 0.1954], + device='cuda:1'), in_proj_covar=tensor([0.0175, 0.0188, 0.0233, 0.0253, 0.0239, 0.0196, 0.0213, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:22:36,977 INFO [finetune.py:976] (1/7) Epoch 10, batch 4350, loss[loss=0.1774, simple_loss=0.2332, pruned_loss=0.06078, over 4092.00 frames. ], tot_loss[loss=0.193, simple_loss=0.2575, pruned_loss=0.06424, over 952989.77 frames. ], batch size: 17, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:22:48,854 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55909.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:22:58,948 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=55915.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:23:23,254 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.076e+02 1.622e+02 1.928e+02 2.410e+02 3.855e+02, threshold=3.856e+02, percent-clipped=1.0 +2023-03-26 12:23:37,719 INFO [finetune.py:976] (1/7) Epoch 10, batch 4400, loss[loss=0.1754, simple_loss=0.2525, pruned_loss=0.04917, over 4862.00 frames. ], tot_loss[loss=0.1942, simple_loss=0.2591, pruned_loss=0.06461, over 953741.35 frames. ], batch size: 44, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:24:00,275 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5684, 1.5248, 1.5472, 0.9197, 1.5904, 1.7614, 1.8058, 1.3880], + device='cuda:1'), covar=tensor([0.0731, 0.0508, 0.0506, 0.0469, 0.0401, 0.0465, 0.0279, 0.0549], + device='cuda:1'), in_proj_covar=tensor([0.0128, 0.0154, 0.0122, 0.0133, 0.0131, 0.0125, 0.0144, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.5097e-05, 1.1329e-04, 8.7762e-05, 9.6306e-05, 9.3792e-05, 9.1044e-05, + 1.0582e-04, 1.0722e-04], device='cuda:1') +2023-03-26 12:24:05,734 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5115, 1.4328, 1.3677, 1.5358, 1.1909, 3.1338, 1.2605, 1.8246], + device='cuda:1'), covar=tensor([0.3261, 0.2391, 0.2073, 0.2302, 0.1740, 0.0227, 0.2826, 0.1240], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0120, 0.0123, 0.0116, 0.0098, 0.0099, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:24:11,803 INFO [finetune.py:976] (1/7) Epoch 10, batch 4450, loss[loss=0.1986, simple_loss=0.2649, pruned_loss=0.06611, over 4792.00 frames. ], tot_loss[loss=0.1962, simple_loss=0.262, pruned_loss=0.06516, over 955549.95 frames. ], batch size: 45, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:24:22,917 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5047, 3.1598, 3.0986, 1.3380, 3.3320, 2.4873, 0.7515, 2.2123], + device='cuda:1'), covar=tensor([0.2737, 0.1964, 0.1672, 0.3304, 0.1274, 0.1065, 0.4002, 0.1463], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0158, 0.0127, 0.0154, 0.0121, 0.0144, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 12:24:25,778 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8615, 2.0482, 1.5551, 1.4257, 2.2179, 2.2135, 2.0353, 1.8426], + device='cuda:1'), covar=tensor([0.0415, 0.0329, 0.0537, 0.0403, 0.0305, 0.0574, 0.0329, 0.0401], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0110, 0.0138, 0.0114, 0.0100, 0.0102, 0.0092, 0.0106], + device='cuda:1'), out_proj_covar=tensor([7.0465e-05, 8.5514e-05, 1.1005e-04, 8.9036e-05, 7.8511e-05, 7.5756e-05, + 6.9311e-05, 8.1889e-05], device='cuda:1') +2023-03-26 12:24:36,683 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.054e+02 1.562e+02 1.840e+02 2.258e+02 4.729e+02, threshold=3.681e+02, percent-clipped=2.0 +2023-03-26 12:24:43,415 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9292, 1.8112, 1.5072, 1.5262, 1.7128, 1.7523, 1.7333, 2.3570], + device='cuda:1'), covar=tensor([0.4793, 0.4492, 0.3844, 0.4540, 0.4501, 0.2664, 0.4125, 0.2283], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0257, 0.0221, 0.0277, 0.0241, 0.0208, 0.0244, 0.0212], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:24:46,912 INFO [finetune.py:976] (1/7) Epoch 10, batch 4500, loss[loss=0.2217, simple_loss=0.2858, pruned_loss=0.07884, over 4719.00 frames. ], tot_loss[loss=0.1982, simple_loss=0.2643, pruned_loss=0.06609, over 956853.35 frames. ], batch size: 59, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:25:31,191 INFO [finetune.py:976] (1/7) Epoch 10, batch 4550, loss[loss=0.1995, simple_loss=0.2659, pruned_loss=0.06658, over 4255.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2665, pruned_loss=0.0676, over 956813.59 frames. ], batch size: 65, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:25:34,313 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56104.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:25:53,264 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.089e+02 1.686e+02 1.941e+02 2.447e+02 3.858e+02, threshold=3.882e+02, percent-clipped=3.0 +2023-03-26 12:26:02,541 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56145.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:26:04,873 INFO [finetune.py:976] (1/7) Epoch 10, batch 4600, loss[loss=0.162, simple_loss=0.2202, pruned_loss=0.05187, over 4914.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2651, pruned_loss=0.06716, over 956498.97 frames. ], batch size: 38, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:26:04,994 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9209, 1.7673, 2.0028, 1.2566, 1.9740, 2.0194, 1.9168, 1.6509], + device='cuda:1'), covar=tensor([0.0598, 0.0774, 0.0710, 0.0914, 0.0679, 0.0744, 0.0681, 0.1171], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0134, 0.0144, 0.0125, 0.0120, 0.0144, 0.0144, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:26:40,465 INFO [finetune.py:976] (1/7) Epoch 10, batch 4650, loss[loss=0.1742, simple_loss=0.2341, pruned_loss=0.05712, over 4754.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2619, pruned_loss=0.06652, over 953590.56 frames. ], batch size: 27, lr: 3.73e-03, grad_scale: 32.0 +2023-03-26 12:26:43,715 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56204.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:26:44,965 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56206.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:27:11,819 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.027e+02 1.561e+02 1.851e+02 2.355e+02 3.865e+02, threshold=3.702e+02, percent-clipped=0.0 +2023-03-26 12:27:23,135 INFO [finetune.py:976] (1/7) Epoch 10, batch 4700, loss[loss=0.1754, simple_loss=0.2388, pruned_loss=0.05601, over 4901.00 frames. ], tot_loss[loss=0.1948, simple_loss=0.2589, pruned_loss=0.06537, over 954495.63 frames. ], batch size: 32, lr: 3.73e-03, grad_scale: 64.0 +2023-03-26 12:28:02,724 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2459, 1.3882, 0.8873, 2.1235, 2.4855, 1.7748, 1.7138, 1.9224], + device='cuda:1'), covar=tensor([0.1306, 0.2048, 0.2119, 0.1063, 0.1744, 0.2054, 0.1354, 0.1893], + device='cuda:1'), in_proj_covar=tensor([0.0088, 0.0096, 0.0113, 0.0092, 0.0121, 0.0094, 0.0099, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:28:09,015 INFO [finetune.py:976] (1/7) Epoch 10, batch 4750, loss[loss=0.1792, simple_loss=0.2383, pruned_loss=0.06007, over 4823.00 frames. ], tot_loss[loss=0.1928, simple_loss=0.2564, pruned_loss=0.06456, over 955205.27 frames. ], batch size: 30, lr: 3.73e-03, grad_scale: 64.0 +2023-03-26 12:28:30,218 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.456e+02 1.800e+02 2.273e+02 6.888e+02, threshold=3.601e+02, percent-clipped=2.0 +2023-03-26 12:28:42,335 INFO [finetune.py:976] (1/7) Epoch 10, batch 4800, loss[loss=0.2389, simple_loss=0.3012, pruned_loss=0.08832, over 4907.00 frames. ], tot_loss[loss=0.1952, simple_loss=0.2595, pruned_loss=0.06542, over 955812.74 frames. ], batch size: 36, lr: 3.73e-03, grad_scale: 64.0 +2023-03-26 12:29:06,980 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0369, 1.9635, 1.5598, 1.8594, 1.9694, 1.6711, 2.2684, 2.0389], + device='cuda:1'), covar=tensor([0.1410, 0.2314, 0.3242, 0.2813, 0.2848, 0.1741, 0.3531, 0.1942], + device='cuda:1'), in_proj_covar=tensor([0.0174, 0.0187, 0.0232, 0.0252, 0.0238, 0.0196, 0.0211, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:29:14,946 INFO [finetune.py:976] (1/7) Epoch 10, batch 4850, loss[loss=0.2017, simple_loss=0.2755, pruned_loss=0.06392, over 4887.00 frames. ], tot_loss[loss=0.1978, simple_loss=0.263, pruned_loss=0.06632, over 956008.20 frames. ], batch size: 32, lr: 3.73e-03, grad_scale: 64.0 +2023-03-26 12:29:19,683 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=56404.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:29:22,132 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3933, 2.6553, 2.4701, 1.9244, 2.7211, 2.8761, 2.7831, 2.3413], + device='cuda:1'), covar=tensor([0.0764, 0.0676, 0.0772, 0.0938, 0.0541, 0.0760, 0.0676, 0.1121], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0134, 0.0145, 0.0125, 0.0121, 0.0144, 0.0144, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:29:37,026 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.167e+02 1.738e+02 2.004e+02 2.451e+02 5.164e+02, threshold=4.009e+02, percent-clipped=2.0 +2023-03-26 12:29:48,224 INFO [finetune.py:976] (1/7) Epoch 10, batch 4900, loss[loss=0.205, simple_loss=0.2659, pruned_loss=0.072, over 4828.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2655, pruned_loss=0.0676, over 953790.73 frames. ], batch size: 49, lr: 3.73e-03, grad_scale: 64.0 +2023-03-26 12:29:50,493 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=56452.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:30:09,598 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1125, 1.8899, 1.3625, 0.5293, 1.5886, 1.7311, 1.5502, 1.7289], + device='cuda:1'), covar=tensor([0.0834, 0.0829, 0.1538, 0.2105, 0.1384, 0.2645, 0.2523, 0.0901], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0201, 0.0202, 0.0187, 0.0217, 0.0209, 0.0224, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:30:12,076 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2975, 2.2401, 2.4516, 1.6824, 2.3702, 2.5431, 2.4597, 1.9413], + device='cuda:1'), covar=tensor([0.0587, 0.0625, 0.0635, 0.0877, 0.0554, 0.0708, 0.0616, 0.1057], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0134, 0.0144, 0.0126, 0.0121, 0.0144, 0.0144, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:30:26,420 INFO [finetune.py:976] (1/7) Epoch 10, batch 4950, loss[loss=0.187, simple_loss=0.2415, pruned_loss=0.06627, over 4391.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2663, pruned_loss=0.06754, over 954821.19 frames. ], batch size: 19, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:30:26,690 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-03-26 12:30:32,445 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56501.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:30:34,373 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=56504.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:30:49,789 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56523.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 12:30:55,722 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.081e+02 1.612e+02 1.965e+02 2.275e+02 4.231e+02, threshold=3.931e+02, percent-clipped=1.0 +2023-03-26 12:31:06,812 INFO [finetune.py:976] (1/7) Epoch 10, batch 5000, loss[loss=0.1907, simple_loss=0.2488, pruned_loss=0.06627, over 4090.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.264, pruned_loss=0.06604, over 954754.54 frames. ], batch size: 17, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:31:08,677 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=56552.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:31:29,646 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56584.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 12:31:39,220 INFO [finetune.py:976] (1/7) Epoch 10, batch 5050, loss[loss=0.1707, simple_loss=0.2491, pruned_loss=0.04609, over 4917.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2605, pruned_loss=0.06509, over 953866.94 frames. ], batch size: 37, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:32:04,809 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.243e+02 1.580e+02 1.790e+02 2.049e+02 5.062e+02, threshold=3.579e+02, percent-clipped=1.0 +2023-03-26 12:32:08,161 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.29 vs. limit=5.0 +2023-03-26 12:32:14,689 INFO [finetune.py:976] (1/7) Epoch 10, batch 5100, loss[loss=0.1612, simple_loss=0.2171, pruned_loss=0.05264, over 4174.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.2569, pruned_loss=0.06349, over 954334.72 frames. ], batch size: 65, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:32:34,162 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4885, 1.0901, 0.7497, 1.4066, 1.9491, 0.6926, 1.2953, 1.3763], + device='cuda:1'), covar=tensor([0.1598, 0.2235, 0.1913, 0.1254, 0.2039, 0.2108, 0.1560, 0.2103], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0113, 0.0092, 0.0121, 0.0095, 0.0100, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:32:54,594 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2989, 1.3932, 1.5963, 1.6689, 1.6014, 3.2787, 1.3911, 1.4870], + device='cuda:1'), covar=tensor([0.1080, 0.1790, 0.1182, 0.0954, 0.1467, 0.0219, 0.1439, 0.1688], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0091, 0.0082, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:32:55,109 INFO [finetune.py:976] (1/7) Epoch 10, batch 5150, loss[loss=0.2191, simple_loss=0.2909, pruned_loss=0.07367, over 4898.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2573, pruned_loss=0.06368, over 955501.87 frames. ], batch size: 35, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:33:27,185 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.919e+01 1.632e+02 1.974e+02 2.331e+02 5.610e+02, threshold=3.948e+02, percent-clipped=3.0 +2023-03-26 12:33:36,882 INFO [finetune.py:976] (1/7) Epoch 10, batch 5200, loss[loss=0.1474, simple_loss=0.2375, pruned_loss=0.02866, over 4937.00 frames. ], tot_loss[loss=0.196, simple_loss=0.2615, pruned_loss=0.06527, over 953711.87 frames. ], batch size: 33, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:34:10,223 INFO [finetune.py:976] (1/7) Epoch 10, batch 5250, loss[loss=0.2498, simple_loss=0.3112, pruned_loss=0.09414, over 4905.00 frames. ], tot_loss[loss=0.1982, simple_loss=0.2644, pruned_loss=0.06599, over 953429.88 frames. ], batch size: 36, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:34:11,645 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=56801.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:34:12,865 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56803.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:34:27,056 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3767, 1.2683, 1.7796, 1.8041, 1.4066, 3.3543, 1.1397, 1.3817], + device='cuda:1'), covar=tensor([0.1257, 0.2413, 0.1530, 0.1145, 0.2011, 0.0297, 0.1995, 0.2343], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0091, 0.0082, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:34:32,517 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3884, 1.2770, 1.6694, 2.4140, 1.6768, 2.1764, 0.8918, 2.0180], + device='cuda:1'), covar=tensor([0.1733, 0.1367, 0.1095, 0.0672, 0.0863, 0.1177, 0.1498, 0.0626], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0136, 0.0165, 0.0101, 0.0138, 0.0126, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:34:34,253 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.164e+02 1.706e+02 2.047e+02 2.503e+02 5.084e+02, threshold=4.093e+02, percent-clipped=2.0 +2023-03-26 12:34:43,965 INFO [finetune.py:976] (1/7) Epoch 10, batch 5300, loss[loss=0.2025, simple_loss=0.2813, pruned_loss=0.06185, over 4832.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2668, pruned_loss=0.06677, over 953113.96 frames. ], batch size: 47, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:34:44,026 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=56849.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:34:50,132 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4698, 2.6263, 2.4712, 1.7933, 2.4064, 2.8209, 2.7089, 2.3180], + device='cuda:1'), covar=tensor([0.0584, 0.0582, 0.0746, 0.0895, 0.0849, 0.0647, 0.0599, 0.0941], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0135, 0.0144, 0.0126, 0.0121, 0.0144, 0.0144, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:34:53,679 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56864.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:35:04,614 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56879.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 12:35:17,627 INFO [finetune.py:976] (1/7) Epoch 10, batch 5350, loss[loss=0.1739, simple_loss=0.2461, pruned_loss=0.0508, over 4806.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2655, pruned_loss=0.06589, over 952053.31 frames. ], batch size: 39, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:35:23,289 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56908.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:35:25,159 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5602, 1.4652, 1.5570, 0.9123, 1.6124, 1.5719, 1.5414, 1.3616], + device='cuda:1'), covar=tensor([0.0614, 0.0784, 0.0755, 0.0937, 0.0817, 0.0758, 0.0629, 0.1243], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0135, 0.0144, 0.0126, 0.0121, 0.0144, 0.0144, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:35:49,117 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.924e+01 1.584e+02 1.842e+02 2.192e+02 3.665e+02, threshold=3.684e+02, percent-clipped=0.0 +2023-03-26 12:36:02,275 INFO [finetune.py:976] (1/7) Epoch 10, batch 5400, loss[loss=0.1894, simple_loss=0.2517, pruned_loss=0.06352, over 4843.00 frames. ], tot_loss[loss=0.1956, simple_loss=0.2617, pruned_loss=0.06476, over 951551.46 frames. ], batch size: 49, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:36:15,031 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56969.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:36:32,491 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56993.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:36:33,659 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56995.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:36:35,990 INFO [finetune.py:976] (1/7) Epoch 10, batch 5450, loss[loss=0.2116, simple_loss=0.2619, pruned_loss=0.08061, over 4934.00 frames. ], tot_loss[loss=0.1929, simple_loss=0.2581, pruned_loss=0.06388, over 952710.46 frames. ], batch size: 38, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:36:57,715 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.491e+02 1.807e+02 2.344e+02 4.842e+02, threshold=3.613e+02, percent-clipped=5.0 +2023-03-26 12:37:07,823 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9480, 1.2633, 1.9264, 1.8617, 1.6744, 1.6699, 1.7761, 1.7455], + device='cuda:1'), covar=tensor([0.4117, 0.4600, 0.3871, 0.4295, 0.5330, 0.3880, 0.5142, 0.3805], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0240, 0.0254, 0.0258, 0.0253, 0.0230, 0.0275, 0.0231], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:37:09,490 INFO [finetune.py:976] (1/7) Epoch 10, batch 5500, loss[loss=0.2031, simple_loss=0.2538, pruned_loss=0.07619, over 4773.00 frames. ], tot_loss[loss=0.1912, simple_loss=0.2558, pruned_loss=0.06329, over 952478.39 frames. ], batch size: 28, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:37:12,660 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57054.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:37:13,834 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57056.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:37:19,902 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1585, 1.9520, 1.4898, 0.5762, 1.6924, 1.7654, 1.6118, 1.8097], + device='cuda:1'), covar=tensor([0.0871, 0.0776, 0.1527, 0.2001, 0.1395, 0.2528, 0.2365, 0.0796], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0199, 0.0201, 0.0186, 0.0214, 0.0208, 0.0222, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:37:22,487 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-26 12:37:26,608 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7160, 1.4206, 0.9101, 1.7029, 2.0710, 1.4385, 1.5175, 1.6917], + device='cuda:1'), covar=tensor([0.1380, 0.1860, 0.1872, 0.1099, 0.1920, 0.1861, 0.1410, 0.1775], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0112, 0.0092, 0.0120, 0.0095, 0.0099, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:37:43,350 INFO [finetune.py:976] (1/7) Epoch 10, batch 5550, loss[loss=0.1602, simple_loss=0.2297, pruned_loss=0.04539, over 4885.00 frames. ], tot_loss[loss=0.1928, simple_loss=0.2573, pruned_loss=0.06418, over 951429.96 frames. ], batch size: 32, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:38:06,740 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.869e+01 1.587e+02 1.788e+02 2.090e+02 3.209e+02, threshold=3.576e+02, percent-clipped=0.0 +2023-03-26 12:38:25,603 INFO [finetune.py:976] (1/7) Epoch 10, batch 5600, loss[loss=0.189, simple_loss=0.2696, pruned_loss=0.05413, over 4904.00 frames. ], tot_loss[loss=0.1984, simple_loss=0.2637, pruned_loss=0.06658, over 953598.06 frames. ], batch size: 35, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:38:33,389 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-26 12:38:35,039 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57159.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:38:45,544 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6501, 1.7304, 1.8407, 1.0938, 1.7322, 2.0208, 1.9996, 1.5628], + device='cuda:1'), covar=tensor([0.0975, 0.0606, 0.0472, 0.0604, 0.0532, 0.0633, 0.0367, 0.0660], + device='cuda:1'), in_proj_covar=tensor([0.0128, 0.0154, 0.0121, 0.0133, 0.0131, 0.0124, 0.0143, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.4973e-05, 1.1266e-04, 8.7304e-05, 9.6104e-05, 9.3572e-05, 9.0322e-05, + 1.0497e-04, 1.0664e-04], device='cuda:1') +2023-03-26 12:38:46,663 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57179.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 12:38:50,743 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57186.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:38:54,767 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 12:38:57,554 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57197.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:38:58,645 INFO [finetune.py:976] (1/7) Epoch 10, batch 5650, loss[loss=0.1226, simple_loss=0.1985, pruned_loss=0.02336, over 3964.00 frames. ], tot_loss[loss=0.2, simple_loss=0.2657, pruned_loss=0.06713, over 951785.61 frames. ], batch size: 17, lr: 3.72e-03, grad_scale: 64.0 +2023-03-26 12:39:15,283 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=57227.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 12:39:19,289 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.057e+02 1.551e+02 1.804e+02 2.162e+02 3.713e+02, threshold=3.608e+02, percent-clipped=1.0 +2023-03-26 12:39:21,157 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9339, 1.8648, 2.3029, 2.2004, 2.0208, 3.7955, 1.7859, 2.0752], + device='cuda:1'), covar=tensor([0.0822, 0.1476, 0.0939, 0.0819, 0.1326, 0.0273, 0.1233, 0.1333], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0075, 0.0077, 0.0091, 0.0082, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:39:25,348 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57244.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:39:27,123 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57247.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:39:28,224 INFO [finetune.py:976] (1/7) Epoch 10, batch 5700, loss[loss=0.1886, simple_loss=0.2366, pruned_loss=0.07035, over 3888.00 frames. ], tot_loss[loss=0.1961, simple_loss=0.2603, pruned_loss=0.06598, over 930063.13 frames. ], batch size: 17, lr: 3.72e-03, grad_scale: 32.0 +2023-03-26 12:39:34,005 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57258.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 12:39:36,087 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 12:39:37,767 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57264.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:39:38,594 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-03-26 12:40:00,695 INFO [finetune.py:976] (1/7) Epoch 11, batch 0, loss[loss=0.1997, simple_loss=0.2653, pruned_loss=0.06703, over 4727.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2653, pruned_loss=0.06703, over 4727.00 frames. ], batch size: 59, lr: 3.72e-03, grad_scale: 16.0 +2023-03-26 12:40:00,695 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 12:40:16,055 INFO [finetune.py:1010] (1/7) Epoch 11, validation: loss=0.1597, simple_loss=0.2306, pruned_loss=0.04438, over 2265189.00 frames. +2023-03-26 12:40:16,056 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 12:40:37,196 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57305.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:40:44,112 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0991, 1.0179, 0.9613, 0.4461, 0.8245, 1.1660, 1.1927, 0.9736], + device='cuda:1'), covar=tensor([0.0780, 0.0445, 0.0440, 0.0504, 0.0516, 0.0469, 0.0317, 0.0562], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0152, 0.0120, 0.0132, 0.0130, 0.0123, 0.0142, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.4325e-05, 1.1183e-04, 8.6372e-05, 9.5568e-05, 9.2786e-05, 8.9576e-05, + 1.0424e-04, 1.0590e-04], device='cuda:1') +2023-03-26 12:40:59,547 INFO [finetune.py:976] (1/7) Epoch 11, batch 50, loss[loss=0.2082, simple_loss=0.2615, pruned_loss=0.07748, over 4814.00 frames. ], tot_loss[loss=0.207, simple_loss=0.2713, pruned_loss=0.07133, over 217997.60 frames. ], batch size: 33, lr: 3.72e-03, grad_scale: 16.0 +2023-03-26 12:41:10,025 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.036e+02 1.580e+02 1.868e+02 2.535e+02 4.204e+02, threshold=3.735e+02, percent-clipped=3.0 +2023-03-26 12:41:18,572 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57349.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:41:19,793 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57351.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:41:38,106 INFO [finetune.py:976] (1/7) Epoch 11, batch 100, loss[loss=0.1846, simple_loss=0.2528, pruned_loss=0.05816, over 4890.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.263, pruned_loss=0.06793, over 380946.11 frames. ], batch size: 32, lr: 3.72e-03, grad_scale: 16.0 +2023-03-26 12:41:46,601 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5868, 2.7926, 2.4941, 1.9096, 2.7761, 2.9581, 2.8029, 2.4285], + device='cuda:1'), covar=tensor([0.0627, 0.0540, 0.0799, 0.0860, 0.0510, 0.0687, 0.0673, 0.0943], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0134, 0.0144, 0.0125, 0.0120, 0.0144, 0.0145, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:41:51,430 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57398.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:42:11,567 INFO [finetune.py:976] (1/7) Epoch 11, batch 150, loss[loss=0.1807, simple_loss=0.2363, pruned_loss=0.06251, over 4895.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2565, pruned_loss=0.06511, over 507989.65 frames. ], batch size: 32, lr: 3.72e-03, grad_scale: 16.0 +2023-03-26 12:42:16,966 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.728e+02 2.070e+02 2.489e+02 4.280e+02, threshold=4.140e+02, percent-clipped=3.0 +2023-03-26 12:42:20,077 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8916, 4.3767, 4.2229, 2.2511, 4.4831, 3.3052, 0.7768, 2.9626], + device='cuda:1'), covar=tensor([0.2495, 0.1701, 0.1359, 0.3181, 0.0825, 0.0953, 0.4689, 0.1566], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0158, 0.0127, 0.0155, 0.0121, 0.0144, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 12:42:23,315 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 12:42:31,671 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57459.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:42:31,695 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57459.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:42:33,497 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57462.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:42:35,989 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5614, 3.9482, 4.1521, 4.3468, 4.3330, 3.9875, 4.6253, 1.4465], + device='cuda:1'), covar=tensor([0.0724, 0.0800, 0.0811, 0.0961, 0.1113, 0.1612, 0.0651, 0.5522], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0245, 0.0276, 0.0290, 0.0331, 0.0285, 0.0301, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:42:36,020 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8356, 1.7512, 1.7185, 1.8321, 1.2338, 3.7670, 1.5853, 2.2781], + device='cuda:1'), covar=tensor([0.3223, 0.2459, 0.2050, 0.2225, 0.1834, 0.0172, 0.2497, 0.1172], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0119, 0.0122, 0.0115, 0.0098, 0.0099, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:42:44,007 INFO [finetune.py:976] (1/7) Epoch 11, batch 200, loss[loss=0.2063, simple_loss=0.2726, pruned_loss=0.07007, over 4914.00 frames. ], tot_loss[loss=0.1917, simple_loss=0.2549, pruned_loss=0.06427, over 608765.24 frames. ], batch size: 46, lr: 3.72e-03, grad_scale: 16.0 +2023-03-26 12:42:52,805 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7340, 1.5830, 1.4114, 1.2534, 1.5424, 1.5081, 1.5119, 2.0872], + device='cuda:1'), covar=tensor([0.4463, 0.4219, 0.3503, 0.3973, 0.4027, 0.2555, 0.4104, 0.1895], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0261, 0.0225, 0.0281, 0.0245, 0.0211, 0.0249, 0.0216], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:42:54,062 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.46 vs. limit=5.0 +2023-03-26 12:42:56,409 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57495.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:43:03,705 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=57507.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:43:14,469 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57523.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:43:17,309 INFO [finetune.py:976] (1/7) Epoch 11, batch 250, loss[loss=0.1603, simple_loss=0.2262, pruned_loss=0.04721, over 4773.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.2576, pruned_loss=0.06493, over 684827.68 frames. ], batch size: 28, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:43:22,638 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.287e+02 1.584e+02 1.966e+02 2.356e+02 4.681e+02, threshold=3.932e+02, percent-clipped=1.0 +2023-03-26 12:43:26,174 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9782, 4.3463, 4.1316, 2.4635, 4.4559, 3.3050, 0.9006, 2.8965], + device='cuda:1'), covar=tensor([0.2362, 0.1684, 0.1337, 0.2929, 0.0967, 0.0895, 0.4391, 0.1525], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0173, 0.0158, 0.0128, 0.0155, 0.0121, 0.0144, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 12:43:27,973 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57542.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:43:43,759 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57553.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 12:43:45,623 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57556.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:43:46,877 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0878, 1.4217, 2.0009, 2.0302, 1.8049, 1.6921, 1.9308, 1.8386], + device='cuda:1'), covar=tensor([0.3779, 0.4469, 0.3908, 0.3900, 0.5492, 0.4254, 0.4755, 0.3638], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0239, 0.0254, 0.0257, 0.0252, 0.0229, 0.0274, 0.0230], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:43:53,972 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.7321, 4.0531, 4.3262, 4.5318, 4.4693, 4.1474, 4.8261, 1.4993], + device='cuda:1'), covar=tensor([0.0753, 0.0879, 0.0691, 0.0828, 0.1192, 0.1514, 0.0567, 0.5782], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0247, 0.0277, 0.0292, 0.0333, 0.0287, 0.0302, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:43:55,220 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57564.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:44:08,373 INFO [finetune.py:976] (1/7) Epoch 11, batch 300, loss[loss=0.2269, simple_loss=0.2952, pruned_loss=0.07931, over 4745.00 frames. ], tot_loss[loss=0.1978, simple_loss=0.2627, pruned_loss=0.06646, over 745960.21 frames. ], batch size: 59, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:44:09,233 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.07 vs. limit=5.0 +2023-03-26 12:44:18,956 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8118, 3.2703, 3.2393, 1.9080, 3.4722, 2.6310, 1.2372, 2.3587], + device='cuda:1'), covar=tensor([0.2792, 0.2005, 0.1482, 0.2686, 0.1126, 0.0939, 0.3486, 0.1393], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0159, 0.0128, 0.0156, 0.0121, 0.0144, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 12:44:24,433 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57600.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:44:28,746 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.5469, 3.0703, 2.8126, 1.4330, 2.9440, 2.5524, 2.5315, 2.6389], + device='cuda:1'), covar=tensor([0.0732, 0.0949, 0.1568, 0.2191, 0.1509, 0.2037, 0.1721, 0.1141], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0200, 0.0202, 0.0188, 0.0215, 0.0209, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:44:28,785 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2302, 1.7271, 2.2378, 2.1403, 1.8885, 1.8862, 2.0526, 2.0103], + device='cuda:1'), covar=tensor([0.4474, 0.5294, 0.4243, 0.4685, 0.6060, 0.4602, 0.6016, 0.3896], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0238, 0.0254, 0.0257, 0.0252, 0.0229, 0.0274, 0.0229], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:44:31,733 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=57612.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:44:33,049 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-26 12:44:40,866 INFO [finetune.py:976] (1/7) Epoch 11, batch 350, loss[loss=0.1809, simple_loss=0.2623, pruned_loss=0.04972, over 4895.00 frames. ], tot_loss[loss=0.1967, simple_loss=0.2625, pruned_loss=0.0655, over 792199.64 frames. ], batch size: 37, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:44:40,934 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9150, 3.9607, 3.6775, 1.7462, 4.0507, 2.8880, 0.8617, 2.6364], + device='cuda:1'), covar=tensor([0.2077, 0.1909, 0.1476, 0.3442, 0.0916, 0.1092, 0.4365, 0.1420], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0174, 0.0159, 0.0128, 0.0156, 0.0122, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 12:44:43,933 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.86 vs. limit=5.0 +2023-03-26 12:44:46,738 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.100e+02 1.573e+02 1.819e+02 2.403e+02 4.156e+02, threshold=3.639e+02, percent-clipped=1.0 +2023-03-26 12:44:56,784 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57649.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:44:58,462 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57651.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:45:14,035 INFO [finetune.py:976] (1/7) Epoch 11, batch 400, loss[loss=0.1894, simple_loss=0.2601, pruned_loss=0.0594, over 4827.00 frames. ], tot_loss[loss=0.1967, simple_loss=0.2631, pruned_loss=0.06516, over 828585.35 frames. ], batch size: 47, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:45:26,776 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1728, 2.0222, 2.6865, 1.5854, 2.4735, 2.4454, 1.8457, 2.5527], + device='cuda:1'), covar=tensor([0.1412, 0.1953, 0.1575, 0.2312, 0.0742, 0.1519, 0.2648, 0.0802], + device='cuda:1'), in_proj_covar=tensor([0.0199, 0.0206, 0.0194, 0.0192, 0.0178, 0.0216, 0.0219, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:45:30,903 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=57697.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:45:32,134 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=57699.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:45:46,095 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1743, 1.9998, 1.6875, 2.0303, 2.1595, 1.8673, 2.4545, 2.1882], + device='cuda:1'), covar=tensor([0.1385, 0.2390, 0.3269, 0.2748, 0.2435, 0.1659, 0.2927, 0.1810], + device='cuda:1'), in_proj_covar=tensor([0.0175, 0.0188, 0.0234, 0.0255, 0.0240, 0.0197, 0.0213, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:45:49,608 INFO [finetune.py:976] (1/7) Epoch 11, batch 450, loss[loss=0.1558, simple_loss=0.2325, pruned_loss=0.03956, over 4870.00 frames. ], tot_loss[loss=0.1965, simple_loss=0.2626, pruned_loss=0.06524, over 857785.34 frames. ], batch size: 34, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:45:51,646 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 12:45:53,858 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57733.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:45:55,473 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.158e+02 1.602e+02 1.902e+02 2.220e+02 3.989e+02, threshold=3.804e+02, percent-clipped=2.0 +2023-03-26 12:46:15,573 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57754.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:46:32,832 INFO [finetune.py:976] (1/7) Epoch 11, batch 500, loss[loss=0.1917, simple_loss=0.2474, pruned_loss=0.068, over 4771.00 frames. ], tot_loss[loss=0.1948, simple_loss=0.2603, pruned_loss=0.06465, over 880602.59 frames. ], batch size: 28, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:46:36,676 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4639, 2.3130, 1.7465, 2.3646, 2.3417, 1.9399, 2.8426, 2.4246], + device='cuda:1'), covar=tensor([0.1441, 0.2684, 0.3662, 0.3304, 0.2927, 0.1838, 0.3508, 0.2092], + device='cuda:1'), in_proj_covar=tensor([0.0175, 0.0188, 0.0234, 0.0255, 0.0240, 0.0196, 0.0213, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:46:44,575 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8168, 1.7444, 1.9099, 1.1802, 1.9676, 2.0101, 1.8938, 1.5681], + device='cuda:1'), covar=tensor([0.0597, 0.0639, 0.0648, 0.0895, 0.0556, 0.0659, 0.0605, 0.1081], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0133, 0.0143, 0.0125, 0.0120, 0.0143, 0.0144, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:46:45,705 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57794.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:46:50,001 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6520, 1.6711, 1.4419, 1.6238, 2.0259, 1.8340, 1.7222, 1.4759], + device='cuda:1'), covar=tensor([0.0330, 0.0277, 0.0564, 0.0281, 0.0183, 0.0431, 0.0271, 0.0384], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0108, 0.0138, 0.0113, 0.0100, 0.0102, 0.0092, 0.0107], + device='cuda:1'), out_proj_covar=tensor([7.0581e-05, 8.4258e-05, 1.0973e-04, 8.8673e-05, 7.8363e-05, 7.5950e-05, + 6.9677e-05, 8.2137e-05], device='cuda:1') +2023-03-26 12:47:01,346 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57818.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:47:06,699 INFO [finetune.py:976] (1/7) Epoch 11, batch 550, loss[loss=0.1892, simple_loss=0.2494, pruned_loss=0.06448, over 4705.00 frames. ], tot_loss[loss=0.1926, simple_loss=0.2573, pruned_loss=0.06391, over 895754.97 frames. ], batch size: 23, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:47:11,534 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.185e+02 1.635e+02 1.936e+02 2.160e+02 3.511e+02, threshold=3.871e+02, percent-clipped=0.0 +2023-03-26 12:47:16,783 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57842.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:47:23,737 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57851.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:47:24,991 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57853.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 12:47:28,012 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2804, 2.8741, 2.7147, 1.2977, 2.9237, 2.0416, 0.6851, 1.7867], + device='cuda:1'), covar=tensor([0.2903, 0.2689, 0.1972, 0.3770, 0.1596, 0.1316, 0.4551, 0.1865], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0159, 0.0127, 0.0156, 0.0121, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 12:47:40,101 INFO [finetune.py:976] (1/7) Epoch 11, batch 600, loss[loss=0.2408, simple_loss=0.2926, pruned_loss=0.09449, over 4910.00 frames. ], tot_loss[loss=0.1931, simple_loss=0.2576, pruned_loss=0.06431, over 911826.54 frames. ], batch size: 43, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:47:48,476 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=57890.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:47:56,176 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57900.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:47:56,759 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=57901.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:48:13,627 INFO [finetune.py:976] (1/7) Epoch 11, batch 650, loss[loss=0.2275, simple_loss=0.304, pruned_loss=0.07552, over 4821.00 frames. ], tot_loss[loss=0.1956, simple_loss=0.2608, pruned_loss=0.06518, over 923110.49 frames. ], batch size: 51, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:48:18,498 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.178e+02 1.568e+02 1.897e+02 2.360e+02 4.682e+02, threshold=3.793e+02, percent-clipped=3.0 +2023-03-26 12:48:27,448 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=57948.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:48:30,339 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7282, 1.5892, 1.6047, 1.8034, 2.0547, 1.9150, 1.4409, 1.5386], + device='cuda:1'), covar=tensor([0.1406, 0.1398, 0.1197, 0.1109, 0.1152, 0.0720, 0.1671, 0.1207], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0208, 0.0208, 0.0189, 0.0242, 0.0181, 0.0213, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:48:48,745 INFO [finetune.py:976] (1/7) Epoch 11, batch 700, loss[loss=0.1821, simple_loss=0.2564, pruned_loss=0.05391, over 4764.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.2627, pruned_loss=0.06525, over 929071.79 frames. ], batch size: 54, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:49:01,315 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1862, 2.1074, 1.6372, 2.2386, 2.1430, 1.7947, 2.4934, 2.1848], + device='cuda:1'), covar=tensor([0.1299, 0.2413, 0.3122, 0.2579, 0.2543, 0.1674, 0.3302, 0.1881], + device='cuda:1'), in_proj_covar=tensor([0.0175, 0.0188, 0.0233, 0.0254, 0.0240, 0.0196, 0.0212, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:49:44,561 INFO [finetune.py:976] (1/7) Epoch 11, batch 750, loss[loss=0.2031, simple_loss=0.281, pruned_loss=0.06257, over 4828.00 frames. ], tot_loss[loss=0.199, simple_loss=0.265, pruned_loss=0.0665, over 936406.79 frames. ], batch size: 47, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:49:49,409 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.116e+01 1.579e+02 1.894e+02 2.321e+02 4.436e+02, threshold=3.789e+02, percent-clipped=3.0 +2023-03-26 12:50:02,172 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58054.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:50:18,109 INFO [finetune.py:976] (1/7) Epoch 11, batch 800, loss[loss=0.1608, simple_loss=0.2205, pruned_loss=0.05051, over 4068.00 frames. ], tot_loss[loss=0.1985, simple_loss=0.2648, pruned_loss=0.06609, over 940663.82 frames. ], batch size: 17, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:50:25,463 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58089.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:50:33,868 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=58102.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:50:45,544 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58118.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:50:51,418 INFO [finetune.py:976] (1/7) Epoch 11, batch 850, loss[loss=0.2027, simple_loss=0.2599, pruned_loss=0.07277, over 4758.00 frames. ], tot_loss[loss=0.1971, simple_loss=0.2628, pruned_loss=0.06572, over 942248.10 frames. ], batch size: 54, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:50:56,225 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.394e+01 1.505e+02 1.749e+02 2.082e+02 4.545e+02, threshold=3.498e+02, percent-clipped=2.0 +2023-03-26 12:50:56,922 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3716, 2.0423, 1.6328, 0.7566, 1.8024, 1.8574, 1.7681, 1.8872], + device='cuda:1'), covar=tensor([0.0711, 0.1009, 0.1625, 0.2101, 0.1425, 0.2132, 0.2031, 0.0983], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0201, 0.0202, 0.0188, 0.0216, 0.0209, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:50:59,937 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58141.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:51:01,770 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58144.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:51:05,920 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58151.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:51:08,872 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8491, 3.1674, 3.0568, 1.7499, 3.1998, 2.8009, 2.8463, 2.8203], + device='cuda:1'), covar=tensor([0.0507, 0.0974, 0.1428, 0.2027, 0.1321, 0.1624, 0.1570, 0.1027], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0201, 0.0202, 0.0188, 0.0216, 0.0208, 0.0223, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:51:22,652 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=58166.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:51:33,563 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0718, 1.7347, 2.3609, 3.4331, 2.4118, 2.6651, 1.2363, 2.6920], + device='cuda:1'), covar=tensor([0.1368, 0.1267, 0.1085, 0.0595, 0.0684, 0.2326, 0.1582, 0.0531], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0163, 0.0101, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:51:35,954 INFO [finetune.py:976] (1/7) Epoch 11, batch 900, loss[loss=0.1585, simple_loss=0.2302, pruned_loss=0.04338, over 4749.00 frames. ], tot_loss[loss=0.1939, simple_loss=0.2594, pruned_loss=0.06416, over 944520.88 frames. ], batch size: 27, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:51:57,408 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=58199.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:51:59,461 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58202.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:52:01,286 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58205.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:52:11,680 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7802, 1.7710, 1.6933, 1.6958, 1.4315, 3.2413, 1.7719, 2.1175], + device='cuda:1'), covar=tensor([0.2961, 0.2198, 0.1853, 0.2035, 0.1489, 0.0233, 0.2639, 0.1124], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0121, 0.0123, 0.0115, 0.0098, 0.0099, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:52:17,035 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0436, 1.9021, 1.5302, 1.7437, 1.7971, 1.7384, 1.7549, 2.4877], + device='cuda:1'), covar=tensor([0.4431, 0.4666, 0.3753, 0.4637, 0.4334, 0.2743, 0.4804, 0.1866], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0259, 0.0222, 0.0277, 0.0242, 0.0209, 0.0246, 0.0214], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:52:17,499 INFO [finetune.py:976] (1/7) Epoch 11, batch 950, loss[loss=0.2095, simple_loss=0.2736, pruned_loss=0.0727, over 4916.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.257, pruned_loss=0.06354, over 947334.90 frames. ], batch size: 37, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:52:22,879 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.011e+02 1.516e+02 1.975e+02 2.310e+02 4.008e+02, threshold=3.950e+02, percent-clipped=1.0 +2023-03-26 12:52:51,452 INFO [finetune.py:976] (1/7) Epoch 11, batch 1000, loss[loss=0.2224, simple_loss=0.2803, pruned_loss=0.08232, over 4888.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2595, pruned_loss=0.06526, over 947123.53 frames. ], batch size: 32, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:53:46,405 INFO [finetune.py:976] (1/7) Epoch 11, batch 1050, loss[loss=0.1375, simple_loss=0.2051, pruned_loss=0.03495, over 4757.00 frames. ], tot_loss[loss=0.1967, simple_loss=0.2621, pruned_loss=0.06569, over 947950.16 frames. ], batch size: 26, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:53:51,318 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.197e+02 1.617e+02 2.003e+02 2.375e+02 3.670e+02, threshold=4.006e+02, percent-clipped=0.0 +2023-03-26 12:54:02,544 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.44 vs. limit=5.0 +2023-03-26 12:54:11,828 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1193, 1.3901, 1.4572, 0.7643, 1.2465, 1.5625, 1.6109, 1.2732], + device='cuda:1'), covar=tensor([0.0933, 0.0478, 0.0473, 0.0491, 0.0430, 0.0597, 0.0326, 0.0654], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0152, 0.0119, 0.0131, 0.0129, 0.0122, 0.0141, 0.0144], + device='cuda:1'), out_proj_covar=tensor([9.3441e-05, 1.1144e-04, 8.5891e-05, 9.4605e-05, 9.1906e-05, 8.9067e-05, + 1.0343e-04, 1.0546e-04], device='cuda:1') +2023-03-26 12:54:42,548 INFO [finetune.py:976] (1/7) Epoch 11, batch 1100, loss[loss=0.2344, simple_loss=0.3026, pruned_loss=0.08312, over 4817.00 frames. ], tot_loss[loss=0.1976, simple_loss=0.2631, pruned_loss=0.06605, over 950206.54 frames. ], batch size: 38, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:54:45,071 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.19 vs. limit=5.0 +2023-03-26 12:54:55,686 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58389.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:55:35,370 INFO [finetune.py:976] (1/7) Epoch 11, batch 1150, loss[loss=0.2132, simple_loss=0.2735, pruned_loss=0.07651, over 4879.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2653, pruned_loss=0.06696, over 953406.48 frames. ], batch size: 32, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:55:40,653 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.092e+02 1.672e+02 1.870e+02 2.321e+02 4.403e+02, threshold=3.740e+02, percent-clipped=1.0 +2023-03-26 12:55:41,939 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=58437.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:55:50,942 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7035, 1.6223, 1.4714, 1.4189, 1.8442, 1.5082, 1.8255, 1.6933], + device='cuda:1'), covar=tensor([0.1380, 0.2144, 0.3072, 0.2418, 0.2462, 0.1601, 0.2842, 0.1816], + device='cuda:1'), in_proj_covar=tensor([0.0174, 0.0186, 0.0231, 0.0252, 0.0238, 0.0195, 0.0210, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:55:55,782 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0735, 0.9025, 0.8943, 1.1870, 1.2766, 1.1937, 1.0460, 0.9889], + device='cuda:1'), covar=tensor([0.0306, 0.0308, 0.0630, 0.0276, 0.0259, 0.0399, 0.0318, 0.0394], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0110, 0.0141, 0.0115, 0.0102, 0.0105, 0.0094, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.1784e-05, 8.5371e-05, 1.1186e-04, 8.9972e-05, 7.9767e-05, 7.7718e-05, + 7.0875e-05, 8.4080e-05], device='cuda:1') +2023-03-26 12:55:59,974 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8104, 1.8053, 1.9459, 1.2728, 1.9817, 1.9956, 1.8925, 1.5880], + device='cuda:1'), covar=tensor([0.0575, 0.0729, 0.0630, 0.0876, 0.0670, 0.0681, 0.0636, 0.1202], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0133, 0.0142, 0.0124, 0.0120, 0.0143, 0.0143, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:56:08,441 INFO [finetune.py:976] (1/7) Epoch 11, batch 1200, loss[loss=0.1896, simple_loss=0.2663, pruned_loss=0.05642, over 4814.00 frames. ], tot_loss[loss=0.1987, simple_loss=0.2642, pruned_loss=0.06663, over 955052.72 frames. ], batch size: 40, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:56:15,615 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2527, 1.1718, 1.1776, 1.1801, 1.5842, 1.3667, 1.3440, 1.1282], + device='cuda:1'), covar=tensor([0.0328, 0.0290, 0.0586, 0.0306, 0.0199, 0.0494, 0.0286, 0.0385], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0109, 0.0140, 0.0115, 0.0102, 0.0104, 0.0093, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.1592e-05, 8.5106e-05, 1.1160e-04, 8.9874e-05, 7.9524e-05, 7.7518e-05, + 7.0631e-05, 8.3835e-05], device='cuda:1') +2023-03-26 12:56:21,456 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58497.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:56:23,265 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58500.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:56:40,496 INFO [finetune.py:976] (1/7) Epoch 11, batch 1250, loss[loss=0.1646, simple_loss=0.2276, pruned_loss=0.05082, over 4696.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.2618, pruned_loss=0.06573, over 954932.09 frames. ], batch size: 23, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:56:44,104 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-26 12:56:46,790 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.332e+01 1.581e+02 1.822e+02 2.261e+02 4.369e+02, threshold=3.644e+02, percent-clipped=3.0 +2023-03-26 12:57:15,450 INFO [finetune.py:976] (1/7) Epoch 11, batch 1300, loss[loss=0.1457, simple_loss=0.2114, pruned_loss=0.04001, over 4786.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2581, pruned_loss=0.06418, over 953571.40 frames. ], batch size: 29, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:57:37,603 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4172, 1.2595, 1.8462, 2.7277, 1.7657, 2.0807, 0.8332, 2.1801], + device='cuda:1'), covar=tensor([0.1941, 0.2037, 0.1404, 0.0999, 0.1064, 0.1638, 0.2114, 0.0849], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0162, 0.0100, 0.0137, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:57:48,892 INFO [finetune.py:976] (1/7) Epoch 11, batch 1350, loss[loss=0.1988, simple_loss=0.2548, pruned_loss=0.07144, over 4158.00 frames. ], tot_loss[loss=0.193, simple_loss=0.2579, pruned_loss=0.06407, over 953855.60 frames. ], batch size: 65, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:57:54,728 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.484e+01 1.581e+02 1.914e+02 2.266e+02 4.857e+02, threshold=3.829e+02, percent-clipped=2.0 +2023-03-26 12:58:03,137 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4220, 1.5546, 1.5925, 1.7678, 1.5993, 3.2223, 1.3116, 1.5725], + device='cuda:1'), covar=tensor([0.1015, 0.1648, 0.1153, 0.0929, 0.1496, 0.0279, 0.1517, 0.1681], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0080, 0.0074, 0.0077, 0.0091, 0.0081, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:58:17,007 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1009, 2.0535, 1.5227, 2.1546, 1.9614, 1.7099, 2.4435, 2.1246], + device='cuda:1'), covar=tensor([0.1418, 0.2446, 0.3449, 0.2828, 0.2974, 0.1802, 0.3108, 0.1917], + device='cuda:1'), in_proj_covar=tensor([0.0175, 0.0188, 0.0233, 0.0255, 0.0240, 0.0196, 0.0212, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 12:58:23,957 INFO [finetune.py:976] (1/7) Epoch 11, batch 1400, loss[loss=0.1804, simple_loss=0.2658, pruned_loss=0.04753, over 4806.00 frames. ], tot_loss[loss=0.1961, simple_loss=0.2617, pruned_loss=0.06522, over 954972.04 frames. ], batch size: 41, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:58:26,980 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58681.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:58:31,763 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 12:58:36,380 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4886, 1.6196, 1.7704, 1.8668, 1.5827, 3.4862, 1.3564, 1.6671], + device='cuda:1'), covar=tensor([0.0995, 0.1675, 0.1182, 0.1006, 0.1611, 0.0257, 0.1530, 0.1717], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0077, 0.0091, 0.0081, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:58:50,739 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6555, 1.6047, 1.9738, 1.9104, 1.7855, 4.1966, 1.5097, 1.8907], + device='cuda:1'), covar=tensor([0.0929, 0.1696, 0.1211, 0.1004, 0.1606, 0.0184, 0.1504, 0.1703], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0080, 0.0074, 0.0077, 0.0091, 0.0081, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 12:58:56,042 INFO [finetune.py:976] (1/7) Epoch 11, batch 1450, loss[loss=0.1638, simple_loss=0.2236, pruned_loss=0.05201, over 4327.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2655, pruned_loss=0.06744, over 952980.36 frames. ], batch size: 19, lr: 3.71e-03, grad_scale: 16.0 +2023-03-26 12:59:01,959 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.107e+02 1.669e+02 2.009e+02 2.318e+02 4.324e+02, threshold=4.017e+02, percent-clipped=1.0 +2023-03-26 12:59:07,264 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58742.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:59:35,090 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58775.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 12:59:36,227 INFO [finetune.py:976] (1/7) Epoch 11, batch 1500, loss[loss=0.1673, simple_loss=0.2438, pruned_loss=0.04539, over 4831.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2666, pruned_loss=0.06734, over 954117.68 frames. ], batch size: 30, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 12:59:48,352 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6744, 1.2013, 0.8466, 1.5851, 2.2048, 1.1157, 1.4298, 1.5069], + device='cuda:1'), covar=tensor([0.1697, 0.2214, 0.2130, 0.1335, 0.1945, 0.2067, 0.1566, 0.2274], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0097, 0.0113, 0.0093, 0.0121, 0.0095, 0.0099, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 12:59:57,883 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58797.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:00:04,307 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4390, 1.3290, 1.5673, 2.4621, 1.6493, 2.1138, 0.8449, 2.0322], + device='cuda:1'), covar=tensor([0.1607, 0.1412, 0.1125, 0.0734, 0.0898, 0.1225, 0.1591, 0.0635], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0162, 0.0100, 0.0136, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 13:00:04,322 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58800.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:00:06,117 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8639, 1.4378, 1.0031, 1.7379, 2.1855, 1.4921, 1.6251, 1.7021], + device='cuda:1'), covar=tensor([0.1598, 0.2176, 0.2077, 0.1287, 0.2067, 0.2103, 0.1560, 0.2043], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0113, 0.0093, 0.0121, 0.0095, 0.0099, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 13:00:27,137 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5740, 1.4887, 1.4526, 1.6259, 1.2436, 3.4271, 1.3116, 1.6944], + device='cuda:1'), covar=tensor([0.3404, 0.2378, 0.2115, 0.2258, 0.1653, 0.0181, 0.2736, 0.1384], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0123, 0.0115, 0.0098, 0.0099, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 13:00:34,186 INFO [finetune.py:976] (1/7) Epoch 11, batch 1550, loss[loss=0.1471, simple_loss=0.2308, pruned_loss=0.03172, over 4922.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2655, pruned_loss=0.06635, over 953474.41 frames. ], batch size: 38, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 13:00:39,949 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.185e+02 1.569e+02 1.959e+02 2.197e+02 4.059e+02, threshold=3.918e+02, percent-clipped=1.0 +2023-03-26 13:00:41,217 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58836.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:00:42,438 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1973, 3.6881, 3.8806, 4.0398, 3.9351, 3.6520, 4.2607, 1.3348], + device='cuda:1'), covar=tensor([0.0808, 0.0733, 0.0781, 0.0984, 0.1288, 0.1609, 0.0678, 0.5340], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0244, 0.0275, 0.0292, 0.0332, 0.0285, 0.0301, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:00:47,624 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=58845.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:00:49,459 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=58848.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:01:07,920 INFO [finetune.py:976] (1/7) Epoch 11, batch 1600, loss[loss=0.19, simple_loss=0.2575, pruned_loss=0.06128, over 4916.00 frames. ], tot_loss[loss=0.1967, simple_loss=0.2626, pruned_loss=0.06534, over 954550.26 frames. ], batch size: 37, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 13:01:28,377 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58906.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:01:50,383 INFO [finetune.py:976] (1/7) Epoch 11, batch 1650, loss[loss=0.2016, simple_loss=0.2522, pruned_loss=0.07548, over 4938.00 frames. ], tot_loss[loss=0.1954, simple_loss=0.2607, pruned_loss=0.06502, over 955327.76 frames. ], batch size: 33, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 13:01:51,196 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.76 vs. limit=2.0 +2023-03-26 13:01:55,255 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.149e+02 1.664e+02 1.923e+02 2.390e+02 4.121e+02, threshold=3.846e+02, percent-clipped=1.0 +2023-03-26 13:02:06,920 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0947, 1.7243, 2.4298, 1.6251, 2.2791, 2.3012, 1.6646, 2.3495], + device='cuda:1'), covar=tensor([0.1230, 0.2066, 0.1221, 0.2132, 0.0806, 0.1598, 0.2556, 0.0863], + device='cuda:1'), in_proj_covar=tensor([0.0198, 0.0205, 0.0193, 0.0191, 0.0179, 0.0215, 0.0218, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:02:16,520 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8715, 1.2092, 1.9390, 1.7671, 1.5624, 1.5370, 1.6103, 1.6777], + device='cuda:1'), covar=tensor([0.3512, 0.3961, 0.3074, 0.3775, 0.4672, 0.3660, 0.4464, 0.3307], + device='cuda:1'), in_proj_covar=tensor([0.0235, 0.0236, 0.0251, 0.0254, 0.0252, 0.0227, 0.0271, 0.0228], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:02:18,255 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58967.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:02:24,167 INFO [finetune.py:976] (1/7) Epoch 11, batch 1700, loss[loss=0.167, simple_loss=0.2379, pruned_loss=0.04805, over 4866.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2577, pruned_loss=0.06384, over 956538.96 frames. ], batch size: 31, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 13:02:51,449 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0833, 3.5452, 3.7154, 3.9495, 3.8124, 3.6001, 4.1803, 1.2349], + device='cuda:1'), covar=tensor([0.0929, 0.0927, 0.0817, 0.1152, 0.1454, 0.1746, 0.0764, 0.5776], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0276, 0.0293, 0.0334, 0.0285, 0.0302, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:02:51,460 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5211, 1.4504, 1.9347, 3.1332, 2.0425, 2.1949, 0.9027, 2.4800], + device='cuda:1'), covar=tensor([0.1864, 0.1525, 0.1390, 0.0601, 0.0901, 0.1474, 0.1971, 0.0582], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0163, 0.0100, 0.0137, 0.0125, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 13:02:57,898 INFO [finetune.py:976] (1/7) Epoch 11, batch 1750, loss[loss=0.2289, simple_loss=0.276, pruned_loss=0.09087, over 4833.00 frames. ], tot_loss[loss=0.1949, simple_loss=0.2599, pruned_loss=0.06494, over 957834.24 frames. ], batch size: 30, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 13:03:02,755 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.140e+02 1.620e+02 1.895e+02 2.249e+02 5.052e+02, threshold=3.790e+02, percent-clipped=2.0 +2023-03-26 13:03:04,074 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59037.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:03:19,361 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1629, 2.2873, 2.1817, 1.7908, 1.9655, 2.4976, 2.4780, 2.1046], + device='cuda:1'), covar=tensor([0.0561, 0.0563, 0.0727, 0.0860, 0.1303, 0.0613, 0.0494, 0.0889], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0135, 0.0144, 0.0126, 0.0122, 0.0145, 0.0146, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:03:33,672 INFO [finetune.py:976] (1/7) Epoch 11, batch 1800, loss[loss=0.2759, simple_loss=0.3328, pruned_loss=0.1095, over 4768.00 frames. ], tot_loss[loss=0.1971, simple_loss=0.2627, pruned_loss=0.06571, over 955437.11 frames. ], batch size: 54, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 13:04:19,595 INFO [finetune.py:976] (1/7) Epoch 11, batch 1850, loss[loss=0.2139, simple_loss=0.2788, pruned_loss=0.07455, over 4846.00 frames. ], tot_loss[loss=0.1979, simple_loss=0.2639, pruned_loss=0.06593, over 955286.25 frames. ], batch size: 49, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 13:04:21,192 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 13:04:22,097 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59131.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:04:24,430 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.037e+02 1.668e+02 2.065e+02 2.636e+02 4.490e+02, threshold=4.130e+02, percent-clipped=5.0 +2023-03-26 13:04:43,990 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59164.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:04:57,368 INFO [finetune.py:976] (1/7) Epoch 11, batch 1900, loss[loss=0.1703, simple_loss=0.2388, pruned_loss=0.0509, over 4771.00 frames. ], tot_loss[loss=0.1983, simple_loss=0.2644, pruned_loss=0.0661, over 955119.78 frames. ], batch size: 26, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 13:04:57,728 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-26 13:05:46,805 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59225.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:05:47,935 INFO [finetune.py:976] (1/7) Epoch 11, batch 1950, loss[loss=0.2166, simple_loss=0.2736, pruned_loss=0.07976, over 4760.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.2628, pruned_loss=0.06523, over 955811.93 frames. ], batch size: 27, lr: 3.70e-03, grad_scale: 16.0 +2023-03-26 13:05:59,308 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.090e+02 1.570e+02 1.817e+02 2.294e+02 4.310e+02, threshold=3.633e+02, percent-clipped=1.0 +2023-03-26 13:06:29,859 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59262.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:06:51,936 INFO [finetune.py:976] (1/7) Epoch 11, batch 2000, loss[loss=0.1497, simple_loss=0.2202, pruned_loss=0.03965, over 4877.00 frames. ], tot_loss[loss=0.1944, simple_loss=0.2601, pruned_loss=0.06436, over 953602.65 frames. ], batch size: 31, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:07:03,826 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3762, 2.2474, 1.8776, 2.2044, 2.2973, 1.9898, 2.5654, 2.3227], + device='cuda:1'), covar=tensor([0.1400, 0.2236, 0.3290, 0.2874, 0.2723, 0.1778, 0.3361, 0.1782], + device='cuda:1'), in_proj_covar=tensor([0.0174, 0.0186, 0.0230, 0.0252, 0.0237, 0.0195, 0.0210, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:07:15,788 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0955, 2.1432, 2.1860, 1.6268, 2.1828, 2.3278, 2.2478, 1.8186], + device='cuda:1'), covar=tensor([0.0482, 0.0527, 0.0570, 0.0796, 0.0640, 0.0531, 0.0495, 0.0981], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0133, 0.0142, 0.0125, 0.0121, 0.0143, 0.0144, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:07:37,497 INFO [finetune.py:976] (1/7) Epoch 11, batch 2050, loss[loss=0.1431, simple_loss=0.2102, pruned_loss=0.03798, over 4765.00 frames. ], tot_loss[loss=0.1907, simple_loss=0.2556, pruned_loss=0.06287, over 951426.67 frames. ], batch size: 27, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:07:42,276 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.122e+01 1.513e+02 1.843e+02 2.174e+02 3.611e+02, threshold=3.686e+02, percent-clipped=0.0 +2023-03-26 13:07:44,090 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59337.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:08:00,347 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.84 vs. limit=5.0 +2023-03-26 13:08:17,287 INFO [finetune.py:976] (1/7) Epoch 11, batch 2100, loss[loss=0.2063, simple_loss=0.2706, pruned_loss=0.07099, over 4814.00 frames. ], tot_loss[loss=0.1895, simple_loss=0.2541, pruned_loss=0.06242, over 951258.51 frames. ], batch size: 51, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:08:22,237 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=59385.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:09:08,919 INFO [finetune.py:976] (1/7) Epoch 11, batch 2150, loss[loss=0.2428, simple_loss=0.3025, pruned_loss=0.09149, over 4038.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2569, pruned_loss=0.06314, over 952442.25 frames. ], batch size: 65, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:09:13,326 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59431.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:09:15,687 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.234e+02 1.596e+02 1.893e+02 2.254e+02 5.168e+02, threshold=3.786e+02, percent-clipped=3.0 +2023-03-26 13:09:18,967 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.61 vs. limit=2.0 +2023-03-26 13:09:51,471 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3032, 2.1999, 1.7131, 2.1353, 2.1846, 1.9722, 2.4590, 2.2824], + device='cuda:1'), covar=tensor([0.1355, 0.2315, 0.3302, 0.2998, 0.2690, 0.1688, 0.3067, 0.1950], + device='cuda:1'), in_proj_covar=tensor([0.0175, 0.0186, 0.0231, 0.0253, 0.0238, 0.0195, 0.0211, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:09:54,867 INFO [finetune.py:976] (1/7) Epoch 11, batch 2200, loss[loss=0.1663, simple_loss=0.2338, pruned_loss=0.04939, over 4788.00 frames. ], tot_loss[loss=0.1941, simple_loss=0.26, pruned_loss=0.06411, over 953250.83 frames. ], batch size: 25, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:09:56,162 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=59479.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:10:13,062 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5219, 1.5116, 1.6363, 1.7254, 1.5702, 3.2561, 1.4350, 1.5638], + device='cuda:1'), covar=tensor([0.0952, 0.1834, 0.1157, 0.0944, 0.1620, 0.0324, 0.1508, 0.1729], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0074, 0.0077, 0.0091, 0.0081, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 13:10:13,087 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59502.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:10:15,550 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1613, 2.0781, 2.2641, 1.5215, 2.1352, 2.2904, 2.2981, 1.8419], + device='cuda:1'), covar=tensor([0.0444, 0.0553, 0.0520, 0.0892, 0.0660, 0.0568, 0.0466, 0.0892], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0133, 0.0142, 0.0126, 0.0120, 0.0143, 0.0144, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:10:25,076 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59520.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:10:30,585 INFO [finetune.py:976] (1/7) Epoch 11, batch 2250, loss[loss=0.1876, simple_loss=0.26, pruned_loss=0.05754, over 4756.00 frames. ], tot_loss[loss=0.1955, simple_loss=0.2615, pruned_loss=0.06475, over 952409.92 frames. ], batch size: 28, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:10:31,268 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6519, 3.6508, 3.4811, 1.8111, 3.7638, 2.7458, 0.7980, 2.4117], + device='cuda:1'), covar=tensor([0.2804, 0.1915, 0.1695, 0.3302, 0.1018, 0.1016, 0.4447, 0.1570], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0173, 0.0160, 0.0129, 0.0156, 0.0121, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 13:10:37,559 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.128e+02 1.729e+02 2.023e+02 2.518e+02 3.990e+02, threshold=4.047e+02, percent-clipped=2.0 +2023-03-26 13:11:02,692 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59562.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:11:02,739 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1393, 2.0702, 2.1662, 0.9966, 2.5580, 2.7114, 2.2442, 1.9498], + device='cuda:1'), covar=tensor([0.0940, 0.0681, 0.0459, 0.0694, 0.0380, 0.0610, 0.0413, 0.0678], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0153, 0.0121, 0.0132, 0.0130, 0.0124, 0.0142, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.4078e-05, 1.1222e-04, 8.7132e-05, 9.5434e-05, 9.2410e-05, 9.0035e-05, + 1.0425e-04, 1.0649e-04], device='cuda:1') +2023-03-26 13:11:03,354 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59563.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:11:13,470 INFO [finetune.py:976] (1/7) Epoch 11, batch 2300, loss[loss=0.1974, simple_loss=0.273, pruned_loss=0.0609, over 4820.00 frames. ], tot_loss[loss=0.1962, simple_loss=0.2628, pruned_loss=0.06482, over 954651.35 frames. ], batch size: 38, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:11:20,699 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4652, 1.2526, 1.1282, 1.2763, 1.7579, 1.6399, 1.5373, 1.2429], + device='cuda:1'), covar=tensor([0.0320, 0.0342, 0.0802, 0.0345, 0.0216, 0.0397, 0.0252, 0.0421], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0109, 0.0140, 0.0114, 0.0102, 0.0103, 0.0093, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.0718e-05, 8.5111e-05, 1.1125e-04, 8.9205e-05, 7.9544e-05, 7.6647e-05, + 7.0369e-05, 8.3044e-05], device='cuda:1') +2023-03-26 13:11:22,578 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 13:11:35,295 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=59610.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:11:47,097 INFO [finetune.py:976] (1/7) Epoch 11, batch 2350, loss[loss=0.2396, simple_loss=0.2796, pruned_loss=0.0998, over 4043.00 frames. ], tot_loss[loss=0.196, simple_loss=0.2619, pruned_loss=0.065, over 955741.15 frames. ], batch size: 65, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:11:52,461 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.309e+01 1.451e+02 1.728e+02 2.097e+02 4.600e+02, threshold=3.455e+02, percent-clipped=1.0 +2023-03-26 13:12:19,961 INFO [finetune.py:976] (1/7) Epoch 11, batch 2400, loss[loss=0.1875, simple_loss=0.254, pruned_loss=0.06045, over 4750.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2584, pruned_loss=0.0635, over 955784.94 frames. ], batch size: 27, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:12:53,270 INFO [finetune.py:976] (1/7) Epoch 11, batch 2450, loss[loss=0.1865, simple_loss=0.2506, pruned_loss=0.06123, over 4103.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2543, pruned_loss=0.06194, over 956554.14 frames. ], batch size: 65, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:13:01,211 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.102e+02 1.641e+02 1.877e+02 2.149e+02 5.374e+02, threshold=3.753e+02, percent-clipped=2.0 +2023-03-26 13:13:37,051 INFO [finetune.py:976] (1/7) Epoch 11, batch 2500, loss[loss=0.1749, simple_loss=0.239, pruned_loss=0.05545, over 4740.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2564, pruned_loss=0.06332, over 953558.01 frames. ], batch size: 23, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:14:29,714 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59820.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:14:33,891 INFO [finetune.py:976] (1/7) Epoch 11, batch 2550, loss[loss=0.2016, simple_loss=0.2754, pruned_loss=0.06393, over 4851.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2603, pruned_loss=0.0642, over 953983.18 frames. ], batch size: 44, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:14:34,652 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9828, 1.8667, 1.5442, 1.8398, 1.7792, 1.7400, 1.7649, 2.5503], + device='cuda:1'), covar=tensor([0.4659, 0.5002, 0.3849, 0.4508, 0.4636, 0.2797, 0.4423, 0.1830], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0262, 0.0226, 0.0282, 0.0245, 0.0212, 0.0248, 0.0218], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:14:40,182 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.135e+02 1.636e+02 1.885e+02 2.323e+02 4.849e+02, threshold=3.771e+02, percent-clipped=2.0 +2023-03-26 13:14:41,574 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6335, 1.6857, 1.6677, 1.1136, 1.8496, 2.0343, 1.8727, 1.5145], + device='cuda:1'), covar=tensor([0.0971, 0.0619, 0.0456, 0.0497, 0.0378, 0.0429, 0.0359, 0.0563], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0154, 0.0122, 0.0133, 0.0131, 0.0125, 0.0144, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.5222e-05, 1.1299e-04, 8.8131e-05, 9.6201e-05, 9.3173e-05, 9.1173e-05, + 1.0559e-04, 1.0762e-04], device='cuda:1') +2023-03-26 13:14:49,224 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59848.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:14:57,184 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59858.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:15:03,803 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=59868.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:15:09,218 INFO [finetune.py:976] (1/7) Epoch 11, batch 2600, loss[loss=0.1404, simple_loss=0.205, pruned_loss=0.03794, over 3927.00 frames. ], tot_loss[loss=0.1961, simple_loss=0.2625, pruned_loss=0.06491, over 954795.21 frames. ], batch size: 17, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:15:18,114 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59889.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:15:31,248 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59909.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:15:42,445 INFO [finetune.py:976] (1/7) Epoch 11, batch 2650, loss[loss=0.2265, simple_loss=0.2927, pruned_loss=0.08011, over 4744.00 frames. ], tot_loss[loss=0.1979, simple_loss=0.2641, pruned_loss=0.0658, over 955455.65 frames. ], batch size: 54, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:15:47,333 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.166e+02 1.549e+02 1.976e+02 2.444e+02 3.877e+02, threshold=3.952e+02, percent-clipped=1.0 +2023-03-26 13:15:48,020 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3468, 2.9785, 2.7030, 1.1866, 2.9862, 2.2546, 0.7104, 1.8329], + device='cuda:1'), covar=tensor([0.2473, 0.2155, 0.2096, 0.3791, 0.1504, 0.1144, 0.4305, 0.1809], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0176, 0.0161, 0.0129, 0.0157, 0.0122, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 13:16:03,053 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59950.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:16:29,561 INFO [finetune.py:976] (1/7) Epoch 11, batch 2700, loss[loss=0.1861, simple_loss=0.2405, pruned_loss=0.06579, over 3914.00 frames. ], tot_loss[loss=0.1963, simple_loss=0.2629, pruned_loss=0.06487, over 954595.28 frames. ], batch size: 17, lr: 3.70e-03, grad_scale: 32.0 +2023-03-26 13:16:30,378 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 13:16:41,438 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.1061, 4.4579, 4.6779, 4.9556, 4.8263, 4.5973, 5.2401, 1.6083], + device='cuda:1'), covar=tensor([0.0639, 0.0800, 0.0701, 0.0817, 0.1001, 0.1436, 0.0477, 0.5544], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0242, 0.0274, 0.0292, 0.0329, 0.0284, 0.0299, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:17:04,322 INFO [finetune.py:976] (1/7) Epoch 11, batch 2750, loss[loss=0.2109, simple_loss=0.2778, pruned_loss=0.07202, over 4893.00 frames. ], tot_loss[loss=0.1938, simple_loss=0.2599, pruned_loss=0.06386, over 954756.89 frames. ], batch size: 32, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:17:09,208 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.070e+02 1.603e+02 1.823e+02 2.284e+02 4.397e+02, threshold=3.646e+02, percent-clipped=1.0 +2023-03-26 13:17:12,383 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60040.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:17:21,582 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60052.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:17:30,377 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60065.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:17:37,360 INFO [finetune.py:976] (1/7) Epoch 11, batch 2800, loss[loss=0.2021, simple_loss=0.2614, pruned_loss=0.07138, over 4916.00 frames. ], tot_loss[loss=0.1909, simple_loss=0.2561, pruned_loss=0.06283, over 954887.97 frames. ], batch size: 36, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:17:38,069 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60078.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:17:54,530 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60101.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 13:18:02,754 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60113.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 13:18:10,610 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60126.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:18:11,110 INFO [finetune.py:976] (1/7) Epoch 11, batch 2850, loss[loss=0.2123, simple_loss=0.2767, pruned_loss=0.07397, over 4812.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.2567, pruned_loss=0.06371, over 953709.99 frames. ], batch size: 45, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:18:17,974 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.190e+02 1.579e+02 1.818e+02 2.348e+02 4.165e+02, threshold=3.636e+02, percent-clipped=3.0 +2023-03-26 13:18:21,043 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60139.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:18:39,221 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60158.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:18:51,666 INFO [finetune.py:976] (1/7) Epoch 11, batch 2900, loss[loss=0.3318, simple_loss=0.3707, pruned_loss=0.1464, over 4775.00 frames. ], tot_loss[loss=0.1955, simple_loss=0.2603, pruned_loss=0.06534, over 954173.05 frames. ], batch size: 59, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:19:12,027 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60198.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:19:21,315 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60204.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:19:22,530 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=60206.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:19:51,300 INFO [finetune.py:976] (1/7) Epoch 11, batch 2950, loss[loss=0.215, simple_loss=0.2872, pruned_loss=0.0714, over 4812.00 frames. ], tot_loss[loss=0.1964, simple_loss=0.2624, pruned_loss=0.06523, over 954074.14 frames. ], batch size: 45, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:20:00,140 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.723e+02 2.035e+02 2.444e+02 4.360e+02, threshold=4.070e+02, percent-clipped=6.0 +2023-03-26 13:20:06,678 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60245.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:20:16,624 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60259.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:20:28,427 INFO [finetune.py:976] (1/7) Epoch 11, batch 3000, loss[loss=0.2098, simple_loss=0.2768, pruned_loss=0.0714, over 4835.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2645, pruned_loss=0.06653, over 953758.38 frames. ], batch size: 49, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:20:28,428 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 13:20:34,714 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0065, 1.9164, 1.7670, 1.9574, 2.0230, 1.8122, 2.2680, 1.9783], + device='cuda:1'), covar=tensor([0.1395, 0.2512, 0.2972, 0.2320, 0.2322, 0.1507, 0.3062, 0.1934], + device='cuda:1'), in_proj_covar=tensor([0.0173, 0.0184, 0.0229, 0.0250, 0.0235, 0.0193, 0.0208, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:20:38,898 INFO [finetune.py:1010] (1/7) Epoch 11, validation: loss=0.1572, simple_loss=0.2284, pruned_loss=0.04301, over 2265189.00 frames. +2023-03-26 13:20:38,898 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 13:20:53,545 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9742, 1.7881, 1.5335, 1.6638, 1.6553, 1.6704, 1.6954, 2.4195], + device='cuda:1'), covar=tensor([0.4069, 0.4514, 0.3389, 0.4409, 0.4489, 0.2581, 0.4382, 0.1844], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0260, 0.0223, 0.0279, 0.0243, 0.0210, 0.0247, 0.0216], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:20:56,012 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-03-26 13:21:13,696 INFO [finetune.py:976] (1/7) Epoch 11, batch 3050, loss[loss=0.1488, simple_loss=0.2271, pruned_loss=0.03529, over 4788.00 frames. ], tot_loss[loss=0.1987, simple_loss=0.2648, pruned_loss=0.06628, over 953605.04 frames. ], batch size: 29, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:21:19,478 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.076e+02 1.587e+02 1.939e+02 2.482e+02 4.597e+02, threshold=3.877e+02, percent-clipped=2.0 +2023-03-26 13:21:56,081 INFO [finetune.py:976] (1/7) Epoch 11, batch 3100, loss[loss=0.2037, simple_loss=0.2685, pruned_loss=0.06944, over 4923.00 frames. ], tot_loss[loss=0.1964, simple_loss=0.2621, pruned_loss=0.0653, over 954370.14 frames. ], batch size: 38, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:22:02,402 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-26 13:22:08,704 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60396.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 13:22:16,695 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60408.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 13:22:25,065 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60421.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:22:29,559 INFO [finetune.py:976] (1/7) Epoch 11, batch 3150, loss[loss=0.1988, simple_loss=0.2635, pruned_loss=0.0671, over 4765.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2602, pruned_loss=0.06519, over 955463.04 frames. ], batch size: 26, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:22:34,363 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60434.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:22:34,872 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.045e+02 1.624e+02 1.838e+02 2.200e+02 4.980e+02, threshold=3.676e+02, percent-clipped=1.0 +2023-03-26 13:22:52,327 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0505, 1.9302, 1.6422, 1.8874, 1.8095, 1.7851, 1.8334, 2.5720], + device='cuda:1'), covar=tensor([0.4352, 0.4762, 0.3766, 0.4387, 0.4745, 0.2569, 0.4223, 0.1919], + device='cuda:1'), in_proj_covar=tensor([0.0284, 0.0258, 0.0222, 0.0277, 0.0242, 0.0208, 0.0245, 0.0214], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:22:52,948 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9592, 1.3766, 1.9608, 1.8685, 1.7004, 1.6465, 1.8173, 1.7415], + device='cuda:1'), covar=tensor([0.4066, 0.4596, 0.3695, 0.4357, 0.5357, 0.3902, 0.4989, 0.3773], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0240, 0.0255, 0.0260, 0.0256, 0.0231, 0.0275, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:23:01,693 INFO [finetune.py:976] (1/7) Epoch 11, batch 3200, loss[loss=0.1939, simple_loss=0.2608, pruned_loss=0.06351, over 4709.00 frames. ], tot_loss[loss=0.1926, simple_loss=0.257, pruned_loss=0.06411, over 956596.29 frames. ], batch size: 23, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:23:12,889 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9953, 1.2333, 1.9166, 1.9070, 1.6904, 1.6549, 1.7847, 1.7528], + device='cuda:1'), covar=tensor([0.3808, 0.4458, 0.3564, 0.3983, 0.5067, 0.3768, 0.4719, 0.3622], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0240, 0.0255, 0.0259, 0.0256, 0.0231, 0.0275, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:23:20,602 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60504.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:23:34,301 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3035, 2.2111, 1.6734, 2.4941, 2.2698, 1.8341, 2.8298, 2.2243], + device='cuda:1'), covar=tensor([0.1486, 0.2704, 0.3457, 0.2829, 0.2745, 0.1821, 0.3294, 0.2234], + device='cuda:1'), in_proj_covar=tensor([0.0176, 0.0187, 0.0232, 0.0254, 0.0239, 0.0196, 0.0212, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:23:37,313 INFO [finetune.py:976] (1/7) Epoch 11, batch 3250, loss[loss=0.1754, simple_loss=0.2527, pruned_loss=0.04906, over 4807.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.2581, pruned_loss=0.06463, over 954876.46 frames. ], batch size: 45, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:23:48,950 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.174e+02 1.626e+02 1.982e+02 2.397e+02 3.737e+02, threshold=3.964e+02, percent-clipped=1.0 +2023-03-26 13:23:58,194 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.66 vs. limit=2.0 +2023-03-26 13:23:59,837 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60545.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:24:04,028 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=60552.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:24:05,270 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60554.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:24:27,366 INFO [finetune.py:976] (1/7) Epoch 11, batch 3300, loss[loss=0.215, simple_loss=0.2844, pruned_loss=0.07276, over 4906.00 frames. ], tot_loss[loss=0.197, simple_loss=0.2618, pruned_loss=0.06609, over 953989.99 frames. ], batch size: 37, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:24:45,611 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=60593.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:25:28,937 INFO [finetune.py:976] (1/7) Epoch 11, batch 3350, loss[loss=0.2261, simple_loss=0.3006, pruned_loss=0.07575, over 4801.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2651, pruned_loss=0.06712, over 954317.70 frames. ], batch size: 51, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:25:34,911 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.073e+02 1.701e+02 2.036e+02 2.469e+02 3.577e+02, threshold=4.071e+02, percent-clipped=0.0 +2023-03-26 13:25:46,759 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0615, 2.0392, 2.0662, 1.2677, 2.0693, 2.1438, 2.0019, 1.7481], + device='cuda:1'), covar=tensor([0.0601, 0.0591, 0.0681, 0.1031, 0.0628, 0.0726, 0.0677, 0.1101], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0133, 0.0141, 0.0124, 0.0120, 0.0143, 0.0144, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:26:02,926 INFO [finetune.py:976] (1/7) Epoch 11, batch 3400, loss[loss=0.1884, simple_loss=0.2578, pruned_loss=0.05952, over 4930.00 frames. ], tot_loss[loss=0.1982, simple_loss=0.2636, pruned_loss=0.06633, over 951145.25 frames. ], batch size: 42, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:26:16,541 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60696.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:26:24,750 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5773, 1.6733, 2.0692, 1.9035, 1.7836, 4.3719, 1.4458, 1.8619], + device='cuda:1'), covar=tensor([0.1226, 0.2246, 0.1409, 0.1182, 0.1888, 0.0279, 0.2085, 0.2208], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0081, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 13:26:24,756 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60708.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:26:30,786 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60718.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:26:32,553 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60721.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:26:36,599 INFO [finetune.py:976] (1/7) Epoch 11, batch 3450, loss[loss=0.1859, simple_loss=0.2428, pruned_loss=0.06448, over 4864.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2634, pruned_loss=0.0658, over 952623.40 frames. ], batch size: 31, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:26:41,011 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60734.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:26:41,507 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.902e+01 1.594e+02 1.892e+02 2.253e+02 3.493e+02, threshold=3.783e+02, percent-clipped=0.0 +2023-03-26 13:26:42,944 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-03-26 13:26:52,719 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=60744.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:27:06,422 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=60756.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:27:25,456 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=60769.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:27:36,076 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.76 vs. limit=5.0 +2023-03-26 13:27:36,416 INFO [finetune.py:976] (1/7) Epoch 11, batch 3500, loss[loss=0.1536, simple_loss=0.2257, pruned_loss=0.04075, over 4824.00 frames. ], tot_loss[loss=0.1949, simple_loss=0.2605, pruned_loss=0.06469, over 954964.08 frames. ], batch size: 30, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:27:37,734 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60779.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:27:39,440 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=60782.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:28:00,917 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4534, 1.4809, 1.8241, 1.8446, 1.5675, 3.2829, 1.4026, 1.5905], + device='cuda:1'), covar=tensor([0.0951, 0.1597, 0.1074, 0.0870, 0.1489, 0.0274, 0.1356, 0.1517], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0081, 0.0085, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 13:28:15,226 INFO [finetune.py:976] (1/7) Epoch 11, batch 3550, loss[loss=0.1662, simple_loss=0.233, pruned_loss=0.04968, over 4850.00 frames. ], tot_loss[loss=0.192, simple_loss=0.2571, pruned_loss=0.06343, over 954976.51 frames. ], batch size: 47, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:28:20,663 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.181e+02 1.566e+02 1.863e+02 2.348e+02 4.575e+02, threshold=3.726e+02, percent-clipped=4.0 +2023-03-26 13:28:34,183 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60854.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:28:49,082 INFO [finetune.py:976] (1/7) Epoch 11, batch 3600, loss[loss=0.2019, simple_loss=0.259, pruned_loss=0.07245, over 4793.00 frames. ], tot_loss[loss=0.1902, simple_loss=0.2548, pruned_loss=0.06275, over 955669.87 frames. ], batch size: 51, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:29:17,736 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=60902.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:29:39,497 INFO [finetune.py:976] (1/7) Epoch 11, batch 3650, loss[loss=0.2363, simple_loss=0.3072, pruned_loss=0.0827, over 4815.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.2574, pruned_loss=0.06356, over 952994.68 frames. ], batch size: 40, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:29:44,366 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.012e+02 1.638e+02 1.962e+02 2.312e+02 3.604e+02, threshold=3.924e+02, percent-clipped=0.0 +2023-03-26 13:29:53,020 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-26 13:29:54,162 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0001, 1.8716, 1.6530, 1.9867, 2.5189, 2.0264, 2.0531, 1.4880], + device='cuda:1'), covar=tensor([0.2189, 0.2064, 0.1945, 0.1624, 0.1943, 0.1142, 0.2017, 0.1984], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0208, 0.0209, 0.0190, 0.0243, 0.0183, 0.0213, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:30:29,992 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1634, 1.9607, 2.4481, 1.5850, 2.2864, 2.3952, 1.7399, 2.4617], + device='cuda:1'), covar=tensor([0.1398, 0.1984, 0.1430, 0.2240, 0.0883, 0.1643, 0.2855, 0.0908], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0204, 0.0193, 0.0190, 0.0177, 0.0215, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:30:33,792 INFO [finetune.py:976] (1/7) Epoch 11, batch 3700, loss[loss=0.2157, simple_loss=0.2953, pruned_loss=0.06805, over 4806.00 frames. ], tot_loss[loss=0.1939, simple_loss=0.2602, pruned_loss=0.06374, over 953754.11 frames. ], batch size: 51, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:31:15,829 INFO [finetune.py:976] (1/7) Epoch 11, batch 3750, loss[loss=0.1767, simple_loss=0.235, pruned_loss=0.05916, over 4290.00 frames. ], tot_loss[loss=0.1946, simple_loss=0.2607, pruned_loss=0.06425, over 952181.48 frames. ], batch size: 18, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:31:20,651 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.055e+02 1.587e+02 1.819e+02 2.276e+02 4.586e+02, threshold=3.638e+02, percent-clipped=1.0 +2023-03-26 13:31:38,945 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61061.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:31:47,186 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=61074.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:31:49,400 INFO [finetune.py:976] (1/7) Epoch 11, batch 3800, loss[loss=0.237, simple_loss=0.3023, pruned_loss=0.08588, over 4891.00 frames. ], tot_loss[loss=0.196, simple_loss=0.2624, pruned_loss=0.06477, over 953475.93 frames. ], batch size: 43, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:32:29,708 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61122.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:32:32,630 INFO [finetune.py:976] (1/7) Epoch 11, batch 3850, loss[loss=0.1843, simple_loss=0.2481, pruned_loss=0.06025, over 4905.00 frames. ], tot_loss[loss=0.1946, simple_loss=0.2611, pruned_loss=0.06404, over 956838.91 frames. ], batch size: 36, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:32:37,922 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.075e+02 1.518e+02 1.864e+02 2.279e+02 4.215e+02, threshold=3.727e+02, percent-clipped=1.0 +2023-03-26 13:33:05,949 INFO [finetune.py:976] (1/7) Epoch 11, batch 3900, loss[loss=0.252, simple_loss=0.3035, pruned_loss=0.1003, over 4844.00 frames. ], tot_loss[loss=0.1931, simple_loss=0.2589, pruned_loss=0.06361, over 954900.36 frames. ], batch size: 49, lr: 3.69e-03, grad_scale: 32.0 +2023-03-26 13:33:32,889 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 13:33:39,745 INFO [finetune.py:976] (1/7) Epoch 11, batch 3950, loss[loss=0.1758, simple_loss=0.2387, pruned_loss=0.05646, over 4860.00 frames. ], tot_loss[loss=0.19, simple_loss=0.2554, pruned_loss=0.06234, over 956027.24 frames. ], batch size: 44, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:33:45,061 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.129e+02 1.570e+02 1.907e+02 2.309e+02 4.377e+02, threshold=3.813e+02, percent-clipped=3.0 +2023-03-26 13:34:12,380 INFO [finetune.py:976] (1/7) Epoch 11, batch 4000, loss[loss=0.2079, simple_loss=0.273, pruned_loss=0.07139, over 4803.00 frames. ], tot_loss[loss=0.1896, simple_loss=0.2548, pruned_loss=0.06216, over 956812.38 frames. ], batch size: 51, lr: 3.68e-03, grad_scale: 64.0 +2023-03-26 13:34:30,542 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 13:34:55,751 INFO [finetune.py:976] (1/7) Epoch 11, batch 4050, loss[loss=0.1907, simple_loss=0.2844, pruned_loss=0.04853, over 4814.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2586, pruned_loss=0.06342, over 957044.28 frames. ], batch size: 45, lr: 3.68e-03, grad_scale: 64.0 +2023-03-26 13:35:04,890 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.154e+02 1.652e+02 2.086e+02 2.571e+02 4.987e+02, threshold=4.171e+02, percent-clipped=6.0 +2023-03-26 13:35:41,834 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=61374.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:35:43,531 INFO [finetune.py:976] (1/7) Epoch 11, batch 4100, loss[loss=0.1687, simple_loss=0.242, pruned_loss=0.0477, over 4761.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2614, pruned_loss=0.06435, over 955041.97 frames. ], batch size: 28, lr: 3.68e-03, grad_scale: 64.0 +2023-03-26 13:36:00,261 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6113, 1.2026, 1.0652, 1.5661, 1.9258, 1.3421, 1.4631, 1.6132], + device='cuda:1'), covar=tensor([0.1472, 0.2041, 0.1771, 0.1143, 0.2136, 0.1982, 0.1351, 0.1816], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0114, 0.0094, 0.0121, 0.0095, 0.0100, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 13:36:11,229 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0384, 1.3123, 0.8386, 1.9211, 2.3353, 1.7736, 1.6773, 1.8095], + device='cuda:1'), covar=tensor([0.1383, 0.2007, 0.2107, 0.1121, 0.1864, 0.2001, 0.1326, 0.1859], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0097, 0.0114, 0.0094, 0.0120, 0.0095, 0.0099, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 13:36:11,797 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=61417.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:36:20,055 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=61422.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:36:26,636 INFO [finetune.py:976] (1/7) Epoch 11, batch 4150, loss[loss=0.2835, simple_loss=0.3242, pruned_loss=0.1214, over 4169.00 frames. ], tot_loss[loss=0.1971, simple_loss=0.2635, pruned_loss=0.06532, over 954507.70 frames. ], batch size: 65, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:36:32,502 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.161e+02 1.629e+02 1.982e+02 2.519e+02 5.426e+02, threshold=3.964e+02, percent-clipped=4.0 +2023-03-26 13:36:59,826 INFO [finetune.py:976] (1/7) Epoch 11, batch 4200, loss[loss=0.2208, simple_loss=0.2865, pruned_loss=0.07759, over 4810.00 frames. ], tot_loss[loss=0.197, simple_loss=0.2636, pruned_loss=0.06523, over 954958.96 frames. ], batch size: 41, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:37:25,035 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 13:37:35,243 INFO [finetune.py:976] (1/7) Epoch 11, batch 4250, loss[loss=0.1516, simple_loss=0.2174, pruned_loss=0.04292, over 4789.00 frames. ], tot_loss[loss=0.1952, simple_loss=0.2616, pruned_loss=0.06445, over 956331.91 frames. ], batch size: 29, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:37:45,939 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.771e+01 1.547e+02 1.858e+02 2.245e+02 5.805e+02, threshold=3.715e+02, percent-clipped=2.0 +2023-03-26 13:38:08,343 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8271, 0.9481, 1.7864, 1.6876, 1.5530, 1.4875, 1.5851, 1.6132], + device='cuda:1'), covar=tensor([0.3805, 0.4304, 0.3462, 0.3760, 0.4746, 0.3686, 0.4733, 0.3459], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0238, 0.0253, 0.0258, 0.0255, 0.0230, 0.0274, 0.0231], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:38:15,492 INFO [finetune.py:976] (1/7) Epoch 11, batch 4300, loss[loss=0.1909, simple_loss=0.2498, pruned_loss=0.06602, over 4932.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.2579, pruned_loss=0.06316, over 957103.12 frames. ], batch size: 38, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:38:27,331 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9441, 1.6833, 2.2835, 3.9446, 2.6816, 2.6180, 0.7573, 3.1658], + device='cuda:1'), covar=tensor([0.1739, 0.1421, 0.1390, 0.0513, 0.0730, 0.1724, 0.2021, 0.0474], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0163, 0.0100, 0.0137, 0.0125, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 13:38:32,825 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9246, 2.0920, 1.8736, 1.6337, 2.5235, 2.5857, 2.1410, 2.0391], + device='cuda:1'), covar=tensor([0.0394, 0.0377, 0.0538, 0.0388, 0.0267, 0.0341, 0.0286, 0.0391], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0108, 0.0140, 0.0114, 0.0101, 0.0103, 0.0092, 0.0107], + device='cuda:1'), out_proj_covar=tensor([7.0592e-05, 8.3789e-05, 1.1126e-04, 8.8973e-05, 7.9320e-05, 7.5934e-05, + 6.9595e-05, 8.2030e-05], device='cuda:1') +2023-03-26 13:38:48,430 INFO [finetune.py:976] (1/7) Epoch 11, batch 4350, loss[loss=0.1619, simple_loss=0.2305, pruned_loss=0.04668, over 4767.00 frames. ], tot_loss[loss=0.1892, simple_loss=0.2543, pruned_loss=0.06204, over 956162.03 frames. ], batch size: 26, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:38:54,823 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.000e+02 1.580e+02 1.801e+02 2.212e+02 3.446e+02, threshold=3.603e+02, percent-clipped=0.0 +2023-03-26 13:39:21,855 INFO [finetune.py:976] (1/7) Epoch 11, batch 4400, loss[loss=0.213, simple_loss=0.2723, pruned_loss=0.07685, over 4801.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2561, pruned_loss=0.0633, over 954984.43 frames. ], batch size: 29, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:39:53,805 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=61717.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:40:04,331 INFO [finetune.py:976] (1/7) Epoch 11, batch 4450, loss[loss=0.1734, simple_loss=0.2437, pruned_loss=0.05157, over 4736.00 frames. ], tot_loss[loss=0.1944, simple_loss=0.2601, pruned_loss=0.06438, over 954673.98 frames. ], batch size: 27, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:40:07,495 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61732.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:40:14,298 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.223e+02 1.628e+02 1.977e+02 2.534e+02 3.640e+02, threshold=3.954e+02, percent-clipped=2.0 +2023-03-26 13:40:23,006 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61743.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:40:24,889 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-26 13:40:32,451 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2140, 3.6694, 3.8967, 4.0823, 3.9467, 3.7131, 4.2814, 1.2961], + device='cuda:1'), covar=tensor([0.0928, 0.0958, 0.0823, 0.0984, 0.1400, 0.1716, 0.0649, 0.5837], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0242, 0.0274, 0.0288, 0.0330, 0.0282, 0.0299, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:40:45,900 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=61765.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:40:57,021 INFO [finetune.py:976] (1/7) Epoch 11, batch 4500, loss[loss=0.1772, simple_loss=0.2592, pruned_loss=0.04758, over 4786.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.2626, pruned_loss=0.06529, over 956521.00 frames. ], batch size: 51, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:41:07,849 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61793.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:41:16,091 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61804.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:41:32,576 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7628, 1.5023, 2.0591, 1.3551, 1.9159, 2.0090, 1.4584, 2.0629], + device='cuda:1'), covar=tensor([0.1363, 0.2154, 0.1236, 0.1908, 0.0745, 0.1341, 0.2690, 0.0801], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0204, 0.0192, 0.0190, 0.0178, 0.0215, 0.0215, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:41:33,088 INFO [finetune.py:976] (1/7) Epoch 11, batch 4550, loss[loss=0.2263, simple_loss=0.2904, pruned_loss=0.08111, over 4755.00 frames. ], tot_loss[loss=0.1987, simple_loss=0.2644, pruned_loss=0.06652, over 956160.01 frames. ], batch size: 54, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:41:43,495 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.535e+01 1.607e+02 1.951e+02 2.245e+02 3.846e+02, threshold=3.902e+02, percent-clipped=0.0 +2023-03-26 13:41:51,200 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6035, 1.5346, 1.5041, 1.6061, 1.2791, 2.7310, 1.2564, 1.6903], + device='cuda:1'), covar=tensor([0.2898, 0.2088, 0.1878, 0.1963, 0.1519, 0.0323, 0.2826, 0.1135], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0119, 0.0123, 0.0114, 0.0098, 0.0098, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 13:42:15,220 INFO [finetune.py:976] (1/7) Epoch 11, batch 4600, loss[loss=0.1851, simple_loss=0.2412, pruned_loss=0.0645, over 4827.00 frames. ], tot_loss[loss=0.1969, simple_loss=0.2629, pruned_loss=0.06541, over 955664.96 frames. ], batch size: 30, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:42:40,442 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61914.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:42:48,594 INFO [finetune.py:976] (1/7) Epoch 11, batch 4650, loss[loss=0.1626, simple_loss=0.2362, pruned_loss=0.04456, over 4888.00 frames. ], tot_loss[loss=0.1946, simple_loss=0.2598, pruned_loss=0.06469, over 955476.01 frames. ], batch size: 35, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:42:56,047 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.057e+02 1.606e+02 1.934e+02 2.317e+02 5.626e+02, threshold=3.867e+02, percent-clipped=3.0 +2023-03-26 13:43:31,741 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61975.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:43:32,825 INFO [finetune.py:976] (1/7) Epoch 11, batch 4700, loss[loss=0.1703, simple_loss=0.2287, pruned_loss=0.05598, over 4917.00 frames. ], tot_loss[loss=0.1908, simple_loss=0.2556, pruned_loss=0.06299, over 954475.19 frames. ], batch size: 43, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:43:57,639 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6200, 1.2062, 0.8909, 1.5261, 2.0889, 1.0377, 1.4057, 1.5900], + device='cuda:1'), covar=tensor([0.1407, 0.2056, 0.1824, 0.1163, 0.1818, 0.1844, 0.1373, 0.1843], + device='cuda:1'), in_proj_covar=tensor([0.0088, 0.0095, 0.0112, 0.0093, 0.0119, 0.0094, 0.0098, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 13:44:17,472 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8725, 1.8563, 1.6418, 2.0843, 2.4532, 2.1320, 1.6235, 1.5600], + device='cuda:1'), covar=tensor([0.2297, 0.2041, 0.2029, 0.1696, 0.1648, 0.1071, 0.2421, 0.2042], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0207, 0.0208, 0.0189, 0.0242, 0.0181, 0.0212, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:44:19,771 INFO [finetune.py:976] (1/7) Epoch 11, batch 4750, loss[loss=0.2272, simple_loss=0.3009, pruned_loss=0.07681, over 4833.00 frames. ], tot_loss[loss=0.1912, simple_loss=0.2555, pruned_loss=0.06343, over 954382.23 frames. ], batch size: 51, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:44:25,601 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.066e+02 1.474e+02 1.769e+02 2.148e+02 4.944e+02, threshold=3.539e+02, percent-clipped=1.0 +2023-03-26 13:44:50,485 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8890, 1.4055, 1.9064, 1.8547, 1.6325, 1.5923, 1.7743, 1.7165], + device='cuda:1'), covar=tensor([0.4376, 0.4613, 0.3794, 0.4270, 0.5433, 0.4106, 0.5104, 0.3630], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0238, 0.0253, 0.0259, 0.0255, 0.0230, 0.0274, 0.0231], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:44:53,404 INFO [finetune.py:976] (1/7) Epoch 11, batch 4800, loss[loss=0.193, simple_loss=0.254, pruned_loss=0.066, over 4889.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2588, pruned_loss=0.06391, over 955367.10 frames. ], batch size: 32, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:45:06,469 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62088.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:45:13,197 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62099.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:45:39,061 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 13:45:41,900 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62122.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:45:50,588 INFO [finetune.py:976] (1/7) Epoch 11, batch 4850, loss[loss=0.2067, simple_loss=0.2745, pruned_loss=0.06944, over 4823.00 frames. ], tot_loss[loss=0.1956, simple_loss=0.2619, pruned_loss=0.06463, over 956066.68 frames. ], batch size: 39, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:46:01,539 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.225e+02 1.730e+02 2.037e+02 2.587e+02 8.043e+02, threshold=4.075e+02, percent-clipped=4.0 +2023-03-26 13:46:45,228 INFO [finetune.py:976] (1/7) Epoch 11, batch 4900, loss[loss=0.2373, simple_loss=0.285, pruned_loss=0.09476, over 4762.00 frames. ], tot_loss[loss=0.1969, simple_loss=0.2629, pruned_loss=0.06549, over 954412.02 frames. ], batch size: 26, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:46:48,914 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62183.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:47:49,178 INFO [finetune.py:976] (1/7) Epoch 11, batch 4950, loss[loss=0.2496, simple_loss=0.3048, pruned_loss=0.09722, over 4849.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2633, pruned_loss=0.06579, over 952926.87 frames. ], batch size: 31, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:47:56,653 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.284e+02 1.728e+02 2.029e+02 2.471e+02 5.736e+02, threshold=4.057e+02, percent-clipped=2.0 +2023-03-26 13:48:18,908 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62270.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:48:21,277 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62273.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:48:24,039 INFO [finetune.py:976] (1/7) Epoch 11, batch 5000, loss[loss=0.1841, simple_loss=0.2554, pruned_loss=0.05639, over 4816.00 frames. ], tot_loss[loss=0.1959, simple_loss=0.2618, pruned_loss=0.065, over 952570.23 frames. ], batch size: 39, lr: 3.68e-03, grad_scale: 32.0 +2023-03-26 13:48:47,502 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7666, 1.4476, 0.7668, 1.6218, 2.1903, 1.4122, 1.6254, 1.7926], + device='cuda:1'), covar=tensor([0.1369, 0.1942, 0.2185, 0.1180, 0.1827, 0.1887, 0.1331, 0.1759], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0113, 0.0093, 0.0119, 0.0094, 0.0099, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 13:48:52,579 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0197, 0.8950, 0.9649, 1.1139, 1.2302, 1.1180, 1.0250, 0.9492], + device='cuda:1'), covar=tensor([0.0360, 0.0300, 0.0579, 0.0241, 0.0260, 0.0453, 0.0318, 0.0400], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0109, 0.0141, 0.0114, 0.0102, 0.0104, 0.0092, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.1651e-05, 8.4679e-05, 1.1181e-04, 8.9213e-05, 7.9381e-05, 7.6856e-05, + 6.9525e-05, 8.3006e-05], device='cuda:1') +2023-03-26 13:48:57,126 INFO [finetune.py:976] (1/7) Epoch 11, batch 5050, loss[loss=0.1604, simple_loss=0.2309, pruned_loss=0.04497, over 4841.00 frames. ], tot_loss[loss=0.1942, simple_loss=0.2595, pruned_loss=0.06445, over 952989.08 frames. ], batch size: 25, lr: 3.68e-03, grad_scale: 16.0 +2023-03-26 13:49:02,453 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62334.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:49:04,171 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.067e+02 1.504e+02 1.759e+02 2.068e+02 4.473e+02, threshold=3.518e+02, percent-clipped=1.0 +2023-03-26 13:49:17,256 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3341, 3.7966, 3.9831, 4.1717, 4.1162, 3.8743, 4.3996, 1.4677], + device='cuda:1'), covar=tensor([0.0756, 0.0842, 0.0795, 0.0881, 0.1024, 0.1455, 0.0627, 0.5357], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0244, 0.0277, 0.0291, 0.0332, 0.0284, 0.0303, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:49:17,882 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4751, 2.2367, 1.9164, 0.9580, 2.1279, 1.9411, 1.6541, 2.0686], + device='cuda:1'), covar=tensor([0.0823, 0.0984, 0.1677, 0.2087, 0.1525, 0.2162, 0.2326, 0.1077], + device='cuda:1'), in_proj_covar=tensor([0.0166, 0.0197, 0.0200, 0.0184, 0.0214, 0.0206, 0.0221, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:49:31,104 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4459, 2.2778, 2.0181, 2.4117, 2.2557, 2.2678, 2.1824, 3.3048], + device='cuda:1'), covar=tensor([0.4452, 0.5647, 0.3647, 0.4923, 0.4825, 0.2702, 0.5432, 0.1736], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0223, 0.0277, 0.0243, 0.0210, 0.0247, 0.0216], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:49:32,190 INFO [finetune.py:976] (1/7) Epoch 11, batch 5100, loss[loss=0.1877, simple_loss=0.2488, pruned_loss=0.06329, over 4814.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.2571, pruned_loss=0.06355, over 954962.22 frames. ], batch size: 41, lr: 3.68e-03, grad_scale: 16.0 +2023-03-26 13:49:40,035 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62388.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:49:42,316 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62391.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:49:47,642 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62399.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:50:05,684 INFO [finetune.py:976] (1/7) Epoch 11, batch 5150, loss[loss=0.1946, simple_loss=0.2446, pruned_loss=0.07235, over 4133.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2576, pruned_loss=0.06455, over 953538.78 frames. ], batch size: 18, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:50:12,135 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=62436.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:50:12,674 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.110e+02 1.578e+02 2.001e+02 2.432e+02 3.455e+02, threshold=4.003e+02, percent-clipped=0.0 +2023-03-26 13:50:16,386 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7978, 4.0262, 3.7586, 1.9743, 4.0721, 2.9734, 0.8148, 2.8169], + device='cuda:1'), covar=tensor([0.2778, 0.2015, 0.1812, 0.3436, 0.0986, 0.1094, 0.4866, 0.1488], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0158, 0.0128, 0.0155, 0.0121, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 13:50:26,762 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=62447.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:50:30,449 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62452.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:50:55,269 INFO [finetune.py:976] (1/7) Epoch 11, batch 5200, loss[loss=0.2217, simple_loss=0.2928, pruned_loss=0.0753, over 4095.00 frames. ], tot_loss[loss=0.1967, simple_loss=0.2618, pruned_loss=0.06574, over 952984.72 frames. ], batch size: 65, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:50:56,961 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62478.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:51:13,606 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1285, 3.6041, 3.7699, 3.9540, 3.8804, 3.6444, 4.2046, 1.2499], + device='cuda:1'), covar=tensor([0.0831, 0.0823, 0.0930, 0.1004, 0.1193, 0.1583, 0.0740, 0.5264], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0279, 0.0292, 0.0333, 0.0287, 0.0304, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:51:22,138 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-03-26 13:51:32,734 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8830, 1.7356, 1.6844, 1.7612, 1.6062, 4.3708, 1.7288, 2.2115], + device='cuda:1'), covar=tensor([0.3237, 0.2407, 0.2080, 0.2345, 0.1579, 0.0114, 0.2611, 0.1247], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0119, 0.0122, 0.0114, 0.0098, 0.0098, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 13:51:36,854 INFO [finetune.py:976] (1/7) Epoch 11, batch 5250, loss[loss=0.1828, simple_loss=0.2638, pruned_loss=0.05088, over 4308.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2644, pruned_loss=0.06669, over 953267.61 frames. ], batch size: 65, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:51:54,392 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.247e+02 1.618e+02 1.949e+02 2.406e+02 7.235e+02, threshold=3.897e+02, percent-clipped=3.0 +2023-03-26 13:52:03,522 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62545.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:52:19,534 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62570.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:52:23,690 INFO [finetune.py:976] (1/7) Epoch 11, batch 5300, loss[loss=0.2199, simple_loss=0.2805, pruned_loss=0.07967, over 4803.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2646, pruned_loss=0.06679, over 950765.11 frames. ], batch size: 51, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:52:29,600 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62585.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:52:43,020 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2417, 2.8779, 2.8138, 1.1661, 3.0046, 2.2002, 0.7401, 1.7579], + device='cuda:1'), covar=tensor([0.2482, 0.2105, 0.1816, 0.3399, 0.1418, 0.1098, 0.3937, 0.1608], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0175, 0.0160, 0.0129, 0.0157, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 13:52:44,283 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62606.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:52:52,195 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=62618.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:52:57,602 INFO [finetune.py:976] (1/7) Epoch 11, batch 5350, loss[loss=0.1966, simple_loss=0.2612, pruned_loss=0.06597, over 4723.00 frames. ], tot_loss[loss=0.1969, simple_loss=0.2629, pruned_loss=0.06546, over 951565.27 frames. ], batch size: 54, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:52:58,902 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62629.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:53:04,200 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.067e+02 1.504e+02 1.845e+02 2.238e+02 3.589e+02, threshold=3.690e+02, percent-clipped=0.0 +2023-03-26 13:53:10,791 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62646.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:53:24,286 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62666.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:53:30,763 INFO [finetune.py:976] (1/7) Epoch 11, batch 5400, loss[loss=0.1904, simple_loss=0.2516, pruned_loss=0.06458, over 4934.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2589, pruned_loss=0.06318, over 954573.28 frames. ], batch size: 33, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:53:30,834 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1185, 4.8773, 4.6344, 2.8881, 5.0058, 3.8565, 1.1522, 3.6468], + device='cuda:1'), covar=tensor([0.2264, 0.1589, 0.1319, 0.2744, 0.0762, 0.0752, 0.4330, 0.1182], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0174, 0.0159, 0.0128, 0.0156, 0.0122, 0.0146, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 13:53:50,848 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5108, 2.2818, 2.0076, 1.0400, 2.2320, 1.9799, 1.7700, 2.1220], + device='cuda:1'), covar=tensor([0.0833, 0.0770, 0.1568, 0.1912, 0.1187, 0.2013, 0.1936, 0.0865], + device='cuda:1'), in_proj_covar=tensor([0.0166, 0.0198, 0.0201, 0.0185, 0.0215, 0.0207, 0.0221, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:54:04,661 INFO [finetune.py:976] (1/7) Epoch 11, batch 5450, loss[loss=0.2086, simple_loss=0.2612, pruned_loss=0.07796, over 4929.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2563, pruned_loss=0.06247, over 954363.87 frames. ], batch size: 46, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:54:04,775 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62727.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:54:07,834 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1536, 1.7999, 2.4324, 1.5759, 2.1375, 2.3898, 1.6884, 2.5322], + device='cuda:1'), covar=tensor([0.1532, 0.2248, 0.1566, 0.2200, 0.1013, 0.1503, 0.2946, 0.1008], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0206, 0.0194, 0.0191, 0.0179, 0.0215, 0.0217, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:54:10,763 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.513e+01 1.463e+02 1.876e+02 2.335e+02 4.427e+02, threshold=3.751e+02, percent-clipped=2.0 +2023-03-26 13:54:17,807 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62747.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:54:36,188 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62773.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:54:38,553 INFO [finetune.py:976] (1/7) Epoch 11, batch 5500, loss[loss=0.2192, simple_loss=0.2788, pruned_loss=0.07978, over 4804.00 frames. ], tot_loss[loss=0.189, simple_loss=0.254, pruned_loss=0.06201, over 953917.45 frames. ], batch size: 51, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:54:39,228 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62778.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:55:12,334 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=62826.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:55:12,913 INFO [finetune.py:976] (1/7) Epoch 11, batch 5550, loss[loss=0.1818, simple_loss=0.2292, pruned_loss=0.06721, over 4407.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2558, pruned_loss=0.06272, over 953986.47 frames. ], batch size: 19, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:55:17,890 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62834.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:55:18,637 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7842, 3.7958, 3.5933, 1.8790, 3.8907, 2.8552, 0.7328, 2.5657], + device='cuda:1'), covar=tensor([0.2283, 0.1906, 0.1596, 0.3148, 0.0963, 0.0968, 0.4599, 0.1506], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0158, 0.0128, 0.0155, 0.0121, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 13:55:19,878 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.162e+02 1.580e+02 1.841e+02 2.336e+02 5.980e+02, threshold=3.683e+02, percent-clipped=6.0 +2023-03-26 13:55:52,817 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8505, 3.3307, 3.5159, 3.7480, 3.6141, 3.3429, 3.8792, 1.2956], + device='cuda:1'), covar=tensor([0.0822, 0.0832, 0.0859, 0.0946, 0.1220, 0.1677, 0.0880, 0.5076], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0243, 0.0275, 0.0289, 0.0329, 0.0283, 0.0302, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:56:07,693 INFO [finetune.py:976] (1/7) Epoch 11, batch 5600, loss[loss=0.1653, simple_loss=0.2387, pruned_loss=0.04598, over 4791.00 frames. ], tot_loss[loss=0.1945, simple_loss=0.2605, pruned_loss=0.0642, over 952098.12 frames. ], batch size: 26, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:56:22,195 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62901.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:56:22,822 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8775, 1.9422, 1.7938, 2.2154, 2.4002, 2.2154, 1.7877, 1.5754], + device='cuda:1'), covar=tensor([0.2215, 0.1874, 0.1839, 0.1495, 0.1812, 0.1114, 0.2195, 0.1965], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0209, 0.0209, 0.0190, 0.0244, 0.0183, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:56:37,252 INFO [finetune.py:976] (1/7) Epoch 11, batch 5650, loss[loss=0.1492, simple_loss=0.2146, pruned_loss=0.04196, over 4700.00 frames. ], tot_loss[loss=0.1959, simple_loss=0.2629, pruned_loss=0.06448, over 952010.92 frames. ], batch size: 23, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:56:38,503 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62929.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:56:48,736 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.326e+01 1.606e+02 1.910e+02 2.279e+02 4.497e+02, threshold=3.820e+02, percent-clipped=2.0 +2023-03-26 13:56:49,468 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.65 vs. limit=2.0 +2023-03-26 13:56:50,560 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2680, 3.7025, 3.9117, 4.0717, 4.0170, 3.7342, 4.3774, 1.4252], + device='cuda:1'), covar=tensor([0.0826, 0.0845, 0.0804, 0.1038, 0.1299, 0.1552, 0.0694, 0.5383], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0242, 0.0274, 0.0288, 0.0328, 0.0282, 0.0300, 0.0292], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:56:51,135 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62941.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:57:18,740 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9858, 3.3492, 2.8886, 2.2309, 3.0738, 3.3786, 3.2459, 3.0186], + device='cuda:1'), covar=tensor([0.0623, 0.0500, 0.0778, 0.0864, 0.0719, 0.0650, 0.0601, 0.0861], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0134, 0.0141, 0.0125, 0.0121, 0.0144, 0.0143, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:57:23,398 INFO [finetune.py:976] (1/7) Epoch 11, batch 5700, loss[loss=0.1292, simple_loss=0.1897, pruned_loss=0.03436, over 4307.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.2593, pruned_loss=0.0641, over 935235.68 frames. ], batch size: 19, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:57:23,434 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=62977.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:57:54,981 INFO [finetune.py:976] (1/7) Epoch 12, batch 0, loss[loss=0.1683, simple_loss=0.2392, pruned_loss=0.04872, over 4819.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2392, pruned_loss=0.04872, over 4819.00 frames. ], batch size: 25, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:57:54,981 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 13:58:03,883 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1073, 1.7997, 1.6758, 1.6808, 1.7798, 1.7654, 1.7301, 2.4659], + device='cuda:1'), covar=tensor([0.4371, 0.5228, 0.3945, 0.4577, 0.4769, 0.2701, 0.4841, 0.2020], + device='cuda:1'), in_proj_covar=tensor([0.0284, 0.0260, 0.0222, 0.0275, 0.0242, 0.0209, 0.0245, 0.0216], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:58:11,586 INFO [finetune.py:1010] (1/7) Epoch 12, validation: loss=0.16, simple_loss=0.2305, pruned_loss=0.04472, over 2265189.00 frames. +2023-03-26 13:58:11,586 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 13:58:19,009 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0257, 1.9556, 2.0805, 1.5492, 1.9151, 2.1580, 2.1467, 1.7139], + device='cuda:1'), covar=tensor([0.0488, 0.0538, 0.0563, 0.0780, 0.1025, 0.0492, 0.0485, 0.0950], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0134, 0.0141, 0.0125, 0.0121, 0.0143, 0.0143, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:58:22,060 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63022.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:58:37,034 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.014e+02 1.590e+02 1.966e+02 2.351e+02 4.424e+02, threshold=3.931e+02, percent-clipped=2.0 +2023-03-26 13:58:37,203 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3997, 2.1987, 1.9105, 2.4241, 2.1866, 2.1640, 2.1407, 3.2186], + device='cuda:1'), covar=tensor([0.4577, 0.6096, 0.3803, 0.4952, 0.4945, 0.2662, 0.4944, 0.1735], + device='cuda:1'), in_proj_covar=tensor([0.0284, 0.0259, 0.0222, 0.0275, 0.0242, 0.0208, 0.0245, 0.0215], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:58:48,195 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 13:58:49,959 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63047.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:59:00,529 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 13:59:00,856 INFO [finetune.py:976] (1/7) Epoch 12, batch 50, loss[loss=0.1752, simple_loss=0.2482, pruned_loss=0.05114, over 4907.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2659, pruned_loss=0.068, over 212941.62 frames. ], batch size: 36, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 13:59:10,740 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0741, 1.8150, 2.1143, 1.3664, 2.1442, 2.1387, 2.1583, 1.4071], + device='cuda:1'), covar=tensor([0.0770, 0.0864, 0.0770, 0.1125, 0.0647, 0.0782, 0.0764, 0.1893], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0134, 0.0141, 0.0125, 0.0121, 0.0144, 0.0143, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 13:59:42,643 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=63095.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 13:59:51,323 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.63 vs. limit=5.0 +2023-03-26 13:59:54,686 INFO [finetune.py:976] (1/7) Epoch 12, batch 100, loss[loss=0.2048, simple_loss=0.2611, pruned_loss=0.07427, over 4799.00 frames. ], tot_loss[loss=0.1936, simple_loss=0.2591, pruned_loss=0.06406, over 378688.93 frames. ], batch size: 51, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 14:00:15,391 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63129.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:00:21,134 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.120e+02 1.723e+02 1.978e+02 2.544e+02 5.107e+02, threshold=3.957e+02, percent-clipped=1.0 +2023-03-26 14:00:50,143 INFO [finetune.py:976] (1/7) Epoch 12, batch 150, loss[loss=0.1947, simple_loss=0.2546, pruned_loss=0.06744, over 4939.00 frames. ], tot_loss[loss=0.1907, simple_loss=0.2549, pruned_loss=0.0632, over 507807.85 frames. ], batch size: 38, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 14:01:47,573 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63201.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:01:56,083 INFO [finetune.py:976] (1/7) Epoch 12, batch 200, loss[loss=0.1748, simple_loss=0.2484, pruned_loss=0.05062, over 4938.00 frames. ], tot_loss[loss=0.1902, simple_loss=0.2544, pruned_loss=0.063, over 606596.59 frames. ], batch size: 42, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 14:02:17,458 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63221.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:02:32,829 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.005e+02 1.551e+02 1.870e+02 2.223e+02 3.918e+02, threshold=3.740e+02, percent-clipped=0.0 +2023-03-26 14:02:41,477 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63241.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:02:46,789 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=63249.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:02:51,341 INFO [finetune.py:976] (1/7) Epoch 12, batch 250, loss[loss=0.2116, simple_loss=0.2787, pruned_loss=0.07232, over 4906.00 frames. ], tot_loss[loss=0.1938, simple_loss=0.2582, pruned_loss=0.06467, over 683406.26 frames. ], batch size: 36, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 14:03:08,673 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63282.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:03:13,376 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=63289.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:03:23,972 INFO [finetune.py:976] (1/7) Epoch 12, batch 300, loss[loss=0.123, simple_loss=0.1958, pruned_loss=0.02512, over 4686.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2608, pruned_loss=0.06487, over 744851.25 frames. ], batch size: 23, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 14:03:40,227 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63322.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:03:51,184 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.087e+02 1.663e+02 2.076e+02 2.406e+02 5.777e+02, threshold=4.151e+02, percent-clipped=4.0 +2023-03-26 14:03:55,998 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4912, 2.4236, 1.8947, 2.4935, 2.3993, 2.1148, 2.8830, 2.5240], + device='cuda:1'), covar=tensor([0.1276, 0.2297, 0.3035, 0.2829, 0.2584, 0.1612, 0.3303, 0.1772], + device='cuda:1'), in_proj_covar=tensor([0.0176, 0.0188, 0.0234, 0.0255, 0.0242, 0.0198, 0.0213, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:04:08,573 INFO [finetune.py:976] (1/7) Epoch 12, batch 350, loss[loss=0.2943, simple_loss=0.353, pruned_loss=0.1178, over 4924.00 frames. ], tot_loss[loss=0.196, simple_loss=0.2622, pruned_loss=0.06489, over 792781.87 frames. ], batch size: 38, lr: 3.67e-03, grad_scale: 16.0 +2023-03-26 14:04:27,600 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=63370.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:04:59,619 INFO [finetune.py:976] (1/7) Epoch 12, batch 400, loss[loss=0.197, simple_loss=0.2744, pruned_loss=0.05985, over 4892.00 frames. ], tot_loss[loss=0.197, simple_loss=0.2634, pruned_loss=0.06528, over 829917.80 frames. ], batch size: 32, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:05:02,626 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3393, 1.4277, 1.3762, 0.7621, 1.3636, 1.5868, 1.6151, 1.2665], + device='cuda:1'), covar=tensor([0.0934, 0.0605, 0.0522, 0.0578, 0.0444, 0.0617, 0.0296, 0.0641], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0152, 0.0121, 0.0131, 0.0130, 0.0125, 0.0142, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.3084e-05, 1.1128e-04, 8.7420e-05, 9.4828e-05, 9.2063e-05, 9.0664e-05, + 1.0373e-04, 1.0627e-04], device='cuda:1') +2023-03-26 14:05:10,704 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63420.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:05:11,948 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63422.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:05:16,641 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63429.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:05:21,322 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.052e+02 1.591e+02 1.854e+02 2.332e+02 4.296e+02, threshold=3.709e+02, percent-clipped=1.0 +2023-03-26 14:05:26,457 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 14:05:38,140 INFO [finetune.py:976] (1/7) Epoch 12, batch 450, loss[loss=0.1787, simple_loss=0.2426, pruned_loss=0.05734, over 4815.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2612, pruned_loss=0.06436, over 858067.38 frames. ], batch size: 41, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:05:57,264 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=63477.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:05:59,774 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63481.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:06:00,977 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63483.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:06:15,172 INFO [finetune.py:976] (1/7) Epoch 12, batch 500, loss[loss=0.2225, simple_loss=0.2707, pruned_loss=0.08717, over 4715.00 frames. ], tot_loss[loss=0.1947, simple_loss=0.2599, pruned_loss=0.0648, over 879612.35 frames. ], batch size: 59, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:06:37,050 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.336e+01 1.553e+02 1.855e+02 2.331e+02 4.193e+02, threshold=3.711e+02, percent-clipped=1.0 +2023-03-26 14:06:48,875 INFO [finetune.py:976] (1/7) Epoch 12, batch 550, loss[loss=0.2201, simple_loss=0.2741, pruned_loss=0.08306, over 4936.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2577, pruned_loss=0.0644, over 897009.54 frames. ], batch size: 38, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:06:58,444 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63569.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:07:03,796 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63577.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:07:10,272 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63586.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:07:22,327 INFO [finetune.py:976] (1/7) Epoch 12, batch 600, loss[loss=0.2313, simple_loss=0.2944, pruned_loss=0.0841, over 4827.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2577, pruned_loss=0.06445, over 909849.32 frames. ], batch size: 33, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:07:28,305 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7604, 3.3828, 3.3822, 1.7794, 3.5405, 2.8448, 1.0926, 2.5068], + device='cuda:1'), covar=tensor([0.2753, 0.2270, 0.1616, 0.3207, 0.1057, 0.0932, 0.3899, 0.1496], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0159, 0.0128, 0.0156, 0.0122, 0.0146, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 14:07:33,068 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8348, 1.6486, 2.1156, 1.3421, 1.8054, 1.9926, 1.5827, 2.1852], + device='cuda:1'), covar=tensor([0.1384, 0.2173, 0.1271, 0.1969, 0.0916, 0.1562, 0.2645, 0.0960], + device='cuda:1'), in_proj_covar=tensor([0.0199, 0.0207, 0.0195, 0.0192, 0.0180, 0.0217, 0.0220, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:07:34,380 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 14:07:40,195 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63630.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:07:44,849 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.064e+02 1.685e+02 2.017e+02 2.531e+02 3.696e+02, threshold=4.034e+02, percent-clipped=0.0 +2023-03-26 14:07:51,112 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63647.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:07:55,868 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2191, 2.8192, 2.7212, 1.2325, 2.9886, 2.1696, 0.8257, 1.7774], + device='cuda:1'), covar=tensor([0.2795, 0.2128, 0.1745, 0.3376, 0.1306, 0.1082, 0.3712, 0.1620], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0172, 0.0158, 0.0127, 0.0155, 0.0121, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 14:07:56,391 INFO [finetune.py:976] (1/7) Epoch 12, batch 650, loss[loss=0.2264, simple_loss=0.3007, pruned_loss=0.0761, over 4737.00 frames. ], tot_loss[loss=0.1944, simple_loss=0.2594, pruned_loss=0.0647, over 919889.56 frames. ], batch size: 59, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:08:29,866 INFO [finetune.py:976] (1/7) Epoch 12, batch 700, loss[loss=0.2253, simple_loss=0.2987, pruned_loss=0.07594, over 4919.00 frames. ], tot_loss[loss=0.1968, simple_loss=0.2625, pruned_loss=0.06556, over 929041.25 frames. ], batch size: 33, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:08:56,516 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0085, 0.9367, 0.9195, 1.1003, 1.2250, 1.1292, 1.0325, 0.9495], + device='cuda:1'), covar=tensor([0.0371, 0.0289, 0.0627, 0.0321, 0.0292, 0.0413, 0.0311, 0.0388], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0108, 0.0139, 0.0113, 0.0101, 0.0104, 0.0093, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.1737e-05, 8.4206e-05, 1.1054e-04, 8.8651e-05, 7.9175e-05, 7.7043e-05, + 7.0120e-05, 8.3489e-05], device='cuda:1') +2023-03-26 14:08:59,836 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.754e+02 2.049e+02 2.499e+02 4.974e+02, threshold=4.098e+02, percent-clipped=3.0 +2023-03-26 14:09:11,205 INFO [finetune.py:976] (1/7) Epoch 12, batch 750, loss[loss=0.1948, simple_loss=0.2752, pruned_loss=0.0572, over 4750.00 frames. ], tot_loss[loss=0.1972, simple_loss=0.2632, pruned_loss=0.06561, over 934086.20 frames. ], batch size: 54, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:09:25,556 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63776.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:09:26,770 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63778.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:09:56,448 INFO [finetune.py:976] (1/7) Epoch 12, batch 800, loss[loss=0.2108, simple_loss=0.2752, pruned_loss=0.07315, over 4831.00 frames. ], tot_loss[loss=0.1963, simple_loss=0.2624, pruned_loss=0.06505, over 936941.45 frames. ], batch size: 39, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:10:02,695 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 14:10:04,892 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63810.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:10:12,963 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3041, 1.3236, 1.6158, 1.0835, 1.2259, 1.4866, 1.3108, 1.6046], + device='cuda:1'), covar=tensor([0.1325, 0.2273, 0.1136, 0.1499, 0.0932, 0.1338, 0.2755, 0.0805], + device='cuda:1'), in_proj_covar=tensor([0.0198, 0.0206, 0.0194, 0.0191, 0.0180, 0.0215, 0.0218, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:10:25,050 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.63 vs. limit=5.0 +2023-03-26 14:10:25,500 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8695, 3.3629, 3.5509, 3.7427, 3.6388, 3.4332, 3.9232, 1.1394], + device='cuda:1'), covar=tensor([0.0925, 0.0898, 0.1030, 0.1098, 0.1326, 0.1559, 0.0839, 0.5244], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0243, 0.0275, 0.0288, 0.0329, 0.0281, 0.0299, 0.0292], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:10:26,005 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.160e+02 1.587e+02 1.868e+02 2.134e+02 3.136e+02, threshold=3.736e+02, percent-clipped=1.0 +2023-03-26 14:10:32,460 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63845.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:10:38,485 INFO [finetune.py:976] (1/7) Epoch 12, batch 850, loss[loss=0.1725, simple_loss=0.2397, pruned_loss=0.05269, over 4820.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.2595, pruned_loss=0.06396, over 939008.54 frames. ], batch size: 39, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:10:51,304 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63871.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:10:54,969 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63877.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:11:22,731 INFO [finetune.py:976] (1/7) Epoch 12, batch 900, loss[loss=0.1438, simple_loss=0.2188, pruned_loss=0.03444, over 4869.00 frames. ], tot_loss[loss=0.1901, simple_loss=0.256, pruned_loss=0.06213, over 944312.93 frames. ], batch size: 31, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:11:23,442 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63906.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:11:29,443 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63916.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:11:34,077 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1732, 2.1528, 2.3595, 0.9562, 2.5741, 2.7988, 2.4114, 2.0488], + device='cuda:1'), covar=tensor([0.0869, 0.0716, 0.0446, 0.0716, 0.0419, 0.0660, 0.0358, 0.0755], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0151, 0.0120, 0.0131, 0.0129, 0.0124, 0.0142, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.2876e-05, 1.1070e-04, 8.6682e-05, 9.4505e-05, 9.1498e-05, 9.0379e-05, + 1.0370e-04, 1.0551e-04], device='cuda:1') +2023-03-26 14:11:35,894 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=63925.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:11:35,904 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63925.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:11:44,046 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.021e+02 1.611e+02 1.873e+02 2.372e+02 4.297e+02, threshold=3.747e+02, percent-clipped=2.0 +2023-03-26 14:11:47,183 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63942.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:11:56,453 INFO [finetune.py:976] (1/7) Epoch 12, batch 950, loss[loss=0.1568, simple_loss=0.2287, pruned_loss=0.04248, over 4791.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2541, pruned_loss=0.06145, over 946871.66 frames. ], batch size: 25, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:12:10,869 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63977.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:12:20,433 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0917, 2.0210, 1.5892, 1.8815, 2.0007, 1.7241, 2.2635, 2.0700], + device='cuda:1'), covar=tensor([0.1290, 0.2293, 0.3263, 0.2853, 0.2893, 0.1732, 0.3468, 0.1911], + device='cuda:1'), in_proj_covar=tensor([0.0177, 0.0188, 0.0233, 0.0255, 0.0241, 0.0198, 0.0212, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:12:31,104 INFO [finetune.py:976] (1/7) Epoch 12, batch 1000, loss[loss=0.2188, simple_loss=0.2939, pruned_loss=0.07188, over 4935.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2576, pruned_loss=0.06277, over 947457.79 frames. ], batch size: 38, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:12:43,086 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64023.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:12:51,952 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.200e+02 1.649e+02 1.875e+02 2.259e+02 3.443e+02, threshold=3.751e+02, percent-clipped=0.0 +2023-03-26 14:13:04,254 INFO [finetune.py:976] (1/7) Epoch 12, batch 1050, loss[loss=0.1856, simple_loss=0.2586, pruned_loss=0.05632, over 4895.00 frames. ], tot_loss[loss=0.195, simple_loss=0.262, pruned_loss=0.06398, over 949992.45 frames. ], batch size: 43, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:13:17,522 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64076.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:13:19,205 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64078.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:13:22,964 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64084.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:13:37,901 INFO [finetune.py:976] (1/7) Epoch 12, batch 1100, loss[loss=0.2022, simple_loss=0.2832, pruned_loss=0.06055, over 4813.00 frames. ], tot_loss[loss=0.1965, simple_loss=0.2637, pruned_loss=0.0646, over 952805.65 frames. ], batch size: 45, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:13:53,894 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=64124.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:13:55,103 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=64126.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:14:05,897 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.091e+02 1.584e+02 1.925e+02 2.329e+02 4.054e+02, threshold=3.850e+02, percent-clipped=2.0 +2023-03-26 14:14:07,277 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9188, 2.1728, 1.8330, 1.8701, 2.4602, 2.4671, 2.1384, 2.0433], + device='cuda:1'), covar=tensor([0.0441, 0.0321, 0.0475, 0.0335, 0.0266, 0.0500, 0.0324, 0.0322], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0109, 0.0140, 0.0114, 0.0102, 0.0104, 0.0093, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.1942e-05, 8.4593e-05, 1.1111e-04, 8.8651e-05, 7.9508e-05, 7.6980e-05, + 7.0662e-05, 8.3594e-05], device='cuda:1') +2023-03-26 14:14:09,231 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.93 vs. limit=5.0 +2023-03-26 14:14:17,902 INFO [finetune.py:976] (1/7) Epoch 12, batch 1150, loss[loss=0.1535, simple_loss=0.2268, pruned_loss=0.04013, over 4778.00 frames. ], tot_loss[loss=0.1957, simple_loss=0.2639, pruned_loss=0.06378, over 954587.49 frames. ], batch size: 25, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:14:25,687 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64166.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:14:31,030 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5598, 1.5241, 1.4400, 1.5011, 1.0974, 3.1001, 1.1947, 1.5681], + device='cuda:1'), covar=tensor([0.3166, 0.2220, 0.2023, 0.2163, 0.1730, 0.0245, 0.2778, 0.1263], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0120, 0.0123, 0.0115, 0.0098, 0.0098, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 14:14:48,868 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64201.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:14:56,319 INFO [finetune.py:976] (1/7) Epoch 12, batch 1200, loss[loss=0.1383, simple_loss=0.2055, pruned_loss=0.03554, over 4705.00 frames. ], tot_loss[loss=0.1939, simple_loss=0.2613, pruned_loss=0.06328, over 954364.08 frames. ], batch size: 23, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:15:14,939 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64225.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:15:24,860 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64232.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:15:31,342 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.004e+02 1.575e+02 1.833e+02 2.193e+02 5.344e+02, threshold=3.667e+02, percent-clipped=2.0 +2023-03-26 14:15:34,436 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64242.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:15:43,192 INFO [finetune.py:976] (1/7) Epoch 12, batch 1250, loss[loss=0.1931, simple_loss=0.2468, pruned_loss=0.0697, over 4144.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.2588, pruned_loss=0.06282, over 952777.19 frames. ], batch size: 65, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:15:55,100 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64272.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:15:55,714 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=64273.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:16:09,179 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=64290.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:16:11,100 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64293.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:16:27,223 INFO [finetune.py:976] (1/7) Epoch 12, batch 1300, loss[loss=0.2172, simple_loss=0.2848, pruned_loss=0.07476, over 4912.00 frames. ], tot_loss[loss=0.1898, simple_loss=0.2561, pruned_loss=0.06177, over 954567.68 frames. ], batch size: 36, lr: 3.66e-03, grad_scale: 16.0 +2023-03-26 14:16:42,135 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-03-26 14:16:48,476 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.098e+02 1.610e+02 1.842e+02 2.244e+02 4.381e+02, threshold=3.684e+02, percent-clipped=1.0 +2023-03-26 14:16:53,418 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64345.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:16:55,222 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6618, 1.5336, 2.1016, 3.4813, 2.4348, 2.4857, 1.0628, 2.7288], + device='cuda:1'), covar=tensor([0.1928, 0.1546, 0.1438, 0.0608, 0.0834, 0.1536, 0.1935, 0.0594], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0116, 0.0135, 0.0165, 0.0101, 0.0138, 0.0127, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 14:16:59,917 INFO [finetune.py:976] (1/7) Epoch 12, batch 1350, loss[loss=0.2649, simple_loss=0.3227, pruned_loss=0.1036, over 4828.00 frames. ], tot_loss[loss=0.1909, simple_loss=0.2566, pruned_loss=0.06261, over 955732.20 frames. ], batch size: 47, lr: 3.66e-03, grad_scale: 32.0 +2023-03-26 14:17:16,046 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64379.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:17:33,430 INFO [finetune.py:976] (1/7) Epoch 12, batch 1400, loss[loss=0.189, simple_loss=0.2695, pruned_loss=0.0542, over 4808.00 frames. ], tot_loss[loss=0.1945, simple_loss=0.2603, pruned_loss=0.06431, over 954279.25 frames. ], batch size: 45, lr: 3.66e-03, grad_scale: 32.0 +2023-03-26 14:17:34,186 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64406.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:17:46,545 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9530, 1.7539, 2.3589, 1.6389, 2.2073, 2.3506, 1.6467, 2.4872], + device='cuda:1'), covar=tensor([0.1480, 0.2079, 0.1563, 0.2076, 0.0901, 0.1419, 0.2646, 0.0793], + device='cuda:1'), in_proj_covar=tensor([0.0196, 0.0205, 0.0195, 0.0191, 0.0179, 0.0214, 0.0218, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:17:54,256 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.198e+02 1.617e+02 1.936e+02 2.295e+02 3.610e+02, threshold=3.872e+02, percent-clipped=0.0 +2023-03-26 14:18:06,657 INFO [finetune.py:976] (1/7) Epoch 12, batch 1450, loss[loss=0.1813, simple_loss=0.2481, pruned_loss=0.05725, over 4815.00 frames. ], tot_loss[loss=0.196, simple_loss=0.2623, pruned_loss=0.06483, over 953796.17 frames. ], batch size: 25, lr: 3.66e-03, grad_scale: 32.0 +2023-03-26 14:18:13,333 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64465.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:18:13,904 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64466.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:18:37,444 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64501.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:18:39,730 INFO [finetune.py:976] (1/7) Epoch 12, batch 1500, loss[loss=0.1844, simple_loss=0.2555, pruned_loss=0.05664, over 4914.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2646, pruned_loss=0.06567, over 954675.70 frames. ], batch size: 33, lr: 3.66e-03, grad_scale: 32.0 +2023-03-26 14:18:46,092 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=64514.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:18:54,916 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64526.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:19:01,494 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.120e+02 1.738e+02 2.083e+02 2.672e+02 4.064e+02, threshold=4.165e+02, percent-clipped=1.0 +2023-03-26 14:19:01,633 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0898, 1.9359, 1.4663, 0.6572, 1.7031, 1.7146, 1.5477, 1.7785], + device='cuda:1'), covar=tensor([0.0882, 0.0608, 0.1276, 0.1760, 0.1027, 0.2118, 0.1894, 0.0751], + device='cuda:1'), in_proj_covar=tensor([0.0166, 0.0199, 0.0201, 0.0186, 0.0214, 0.0208, 0.0222, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:19:15,859 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=64549.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:19:19,446 INFO [finetune.py:976] (1/7) Epoch 12, batch 1550, loss[loss=0.2026, simple_loss=0.2724, pruned_loss=0.0664, over 4838.00 frames. ], tot_loss[loss=0.1976, simple_loss=0.2645, pruned_loss=0.0654, over 954029.47 frames. ], batch size: 44, lr: 3.66e-03, grad_scale: 32.0 +2023-03-26 14:19:30,793 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.0009, 4.3076, 4.5876, 4.7941, 4.7178, 4.4211, 5.1121, 1.6206], + device='cuda:1'), covar=tensor([0.0666, 0.0776, 0.0758, 0.0817, 0.1086, 0.1377, 0.0523, 0.5474], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0243, 0.0276, 0.0291, 0.0331, 0.0283, 0.0301, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:19:33,886 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64572.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:19:45,159 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64588.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:19:45,836 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64589.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:19:56,453 INFO [finetune.py:976] (1/7) Epoch 12, batch 1600, loss[loss=0.22, simple_loss=0.2713, pruned_loss=0.08438, over 4778.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2617, pruned_loss=0.06419, over 955070.03 frames. ], batch size: 28, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:20:08,114 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=64620.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:20:30,241 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.026e+02 1.633e+02 1.922e+02 2.431e+02 4.177e+02, threshold=3.845e+02, percent-clipped=1.0 +2023-03-26 14:20:41,557 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64648.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:20:43,267 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64650.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:20:49,612 INFO [finetune.py:976] (1/7) Epoch 12, batch 1650, loss[loss=0.1792, simple_loss=0.2472, pruned_loss=0.05563, over 4824.00 frames. ], tot_loss[loss=0.192, simple_loss=0.258, pruned_loss=0.063, over 954200.29 frames. ], batch size: 40, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:20:51,520 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8034, 4.0516, 3.7396, 2.1032, 4.1068, 3.1660, 0.9237, 2.8886], + device='cuda:1'), covar=tensor([0.2350, 0.1714, 0.1769, 0.3380, 0.1102, 0.0941, 0.4572, 0.1614], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0173, 0.0158, 0.0128, 0.0155, 0.0121, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 14:21:05,196 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64679.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:21:19,528 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64701.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:21:21,791 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64703.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:21:22,888 INFO [finetune.py:976] (1/7) Epoch 12, batch 1700, loss[loss=0.2197, simple_loss=0.2787, pruned_loss=0.08031, over 4936.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2563, pruned_loss=0.06251, over 955020.90 frames. ], batch size: 38, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:21:27,402 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64709.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:21:36,623 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.40 vs. limit=5.0 +2023-03-26 14:21:46,814 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=64727.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:21:53,446 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.135e+02 1.591e+02 1.930e+02 2.225e+02 5.420e+02, threshold=3.861e+02, percent-clipped=2.0 +2023-03-26 14:21:58,922 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64745.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:22:03,092 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64752.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:22:05,308 INFO [finetune.py:976] (1/7) Epoch 12, batch 1750, loss[loss=0.239, simple_loss=0.3029, pruned_loss=0.08752, over 4874.00 frames. ], tot_loss[loss=0.1938, simple_loss=0.2597, pruned_loss=0.06391, over 954862.08 frames. ], batch size: 44, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:22:10,821 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64764.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:22:33,847 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6006, 0.7171, 1.6004, 1.5420, 1.4207, 1.3591, 1.4205, 1.4861], + device='cuda:1'), covar=tensor([0.3848, 0.4205, 0.3508, 0.3722, 0.4707, 0.3666, 0.4611, 0.3463], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0238, 0.0255, 0.0259, 0.0256, 0.0231, 0.0274, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:22:36,992 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8051, 2.4638, 1.9642, 0.9803, 2.2044, 2.1784, 2.0087, 2.2118], + device='cuda:1'), covar=tensor([0.0599, 0.0729, 0.1612, 0.2002, 0.1527, 0.1924, 0.1855, 0.0860], + device='cuda:1'), in_proj_covar=tensor([0.0165, 0.0198, 0.0200, 0.0184, 0.0214, 0.0206, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:22:38,079 INFO [finetune.py:976] (1/7) Epoch 12, batch 1800, loss[loss=0.1889, simple_loss=0.2542, pruned_loss=0.06178, over 4931.00 frames. ], tot_loss[loss=0.1941, simple_loss=0.2602, pruned_loss=0.06398, over 951008.01 frames. ], batch size: 33, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:22:38,793 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64806.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:22:43,537 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64813.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:22:48,350 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64821.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:22:58,985 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.145e+02 1.622e+02 1.968e+02 2.271e+02 4.247e+02, threshold=3.936e+02, percent-clipped=1.0 +2023-03-26 14:23:11,389 INFO [finetune.py:976] (1/7) Epoch 12, batch 1850, loss[loss=0.2209, simple_loss=0.2818, pruned_loss=0.07995, over 4920.00 frames. ], tot_loss[loss=0.1936, simple_loss=0.2606, pruned_loss=0.0633, over 952419.16 frames. ], batch size: 33, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:23:22,194 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8996, 1.0157, 1.8347, 1.7860, 1.6323, 1.5763, 1.6547, 1.6709], + device='cuda:1'), covar=tensor([0.3572, 0.4405, 0.3769, 0.3731, 0.4958, 0.3594, 0.4743, 0.3670], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0237, 0.0255, 0.0259, 0.0256, 0.0231, 0.0274, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:23:33,507 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64888.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:23:45,142 INFO [finetune.py:976] (1/7) Epoch 12, batch 1900, loss[loss=0.2309, simple_loss=0.2883, pruned_loss=0.08681, over 4849.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2625, pruned_loss=0.06378, over 953299.91 frames. ], batch size: 31, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:23:45,878 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8979, 1.6346, 1.4298, 1.1955, 1.6251, 1.6386, 1.5762, 2.2065], + device='cuda:1'), covar=tensor([0.4390, 0.4081, 0.3795, 0.4176, 0.3979, 0.2631, 0.4022, 0.2019], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0258, 0.0223, 0.0275, 0.0242, 0.0209, 0.0246, 0.0216], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:23:54,108 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1443, 1.9723, 1.6896, 1.8356, 1.8143, 1.8546, 1.8153, 2.6968], + device='cuda:1'), covar=tensor([0.4118, 0.4637, 0.3737, 0.4674, 0.4196, 0.2537, 0.4755, 0.1766], + device='cuda:1'), in_proj_covar=tensor([0.0284, 0.0258, 0.0223, 0.0275, 0.0242, 0.0210, 0.0246, 0.0216], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:23:56,717 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 14:23:59,623 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1961, 2.1505, 2.1726, 1.4910, 2.1448, 2.3588, 2.2574, 1.8176], + device='cuda:1'), covar=tensor([0.0611, 0.0644, 0.0744, 0.1012, 0.0632, 0.0688, 0.0629, 0.1056], + device='cuda:1'), in_proj_covar=tensor([0.0137, 0.0135, 0.0143, 0.0126, 0.0123, 0.0144, 0.0145, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:24:05,579 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=64936.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:24:06,597 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.024e+02 1.653e+02 1.911e+02 2.364e+02 4.358e+02, threshold=3.822e+02, percent-clipped=3.0 +2023-03-26 14:24:12,012 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64945.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:24:18,921 INFO [finetune.py:976] (1/7) Epoch 12, batch 1950, loss[loss=0.193, simple_loss=0.2606, pruned_loss=0.06267, over 4760.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.261, pruned_loss=0.06324, over 952648.86 frames. ], batch size: 28, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:24:40,796 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64976.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:24:58,090 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65001.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:24:59,916 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65004.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:25:00,451 INFO [finetune.py:976] (1/7) Epoch 12, batch 2000, loss[loss=0.1737, simple_loss=0.2388, pruned_loss=0.05424, over 4762.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2583, pruned_loss=0.06242, over 953137.45 frames. ], batch size: 28, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:25:05,488 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 14:25:21,641 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.067e+02 1.609e+02 1.880e+02 2.233e+02 7.388e+02, threshold=3.760e+02, percent-clipped=1.0 +2023-03-26 14:25:21,788 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65037.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:25:34,387 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65049.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:25:42,375 INFO [finetune.py:976] (1/7) Epoch 12, batch 2050, loss[loss=0.2389, simple_loss=0.294, pruned_loss=0.0919, over 4712.00 frames. ], tot_loss[loss=0.189, simple_loss=0.255, pruned_loss=0.06153, over 954089.56 frames. ], batch size: 59, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:25:42,500 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65055.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:25:45,412 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65059.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:26:04,613 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.93 vs. limit=5.0 +2023-03-26 14:26:12,789 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4695, 1.3975, 1.9375, 2.8844, 1.9499, 2.1337, 1.0580, 2.3360], + device='cuda:1'), covar=tensor([0.1693, 0.1459, 0.1204, 0.0581, 0.0807, 0.1447, 0.1663, 0.0544], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0164, 0.0100, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 14:26:21,859 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65101.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:26:22,504 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2643, 2.1083, 1.8438, 1.1941, 2.0721, 1.9220, 1.7552, 2.0415], + device='cuda:1'), covar=tensor([0.0822, 0.0688, 0.1363, 0.1600, 0.1022, 0.1505, 0.1635, 0.0816], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0200, 0.0201, 0.0186, 0.0215, 0.0208, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:26:24,690 INFO [finetune.py:976] (1/7) Epoch 12, batch 2100, loss[loss=0.2359, simple_loss=0.3008, pruned_loss=0.08549, over 4865.00 frames. ], tot_loss[loss=0.1896, simple_loss=0.2551, pruned_loss=0.06199, over 954538.37 frames. ], batch size: 44, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:26:27,107 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65108.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:26:32,509 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65116.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:26:35,491 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65121.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:26:52,593 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.054e+02 1.674e+02 1.985e+02 2.398e+02 5.597e+02, threshold=3.971e+02, percent-clipped=1.0 +2023-03-26 14:27:08,169 INFO [finetune.py:976] (1/7) Epoch 12, batch 2150, loss[loss=0.1757, simple_loss=0.2478, pruned_loss=0.05176, over 4828.00 frames. ], tot_loss[loss=0.1907, simple_loss=0.257, pruned_loss=0.06221, over 955223.56 frames. ], batch size: 33, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:27:16,037 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.66 vs. limit=5.0 +2023-03-26 14:27:17,728 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65169.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:27:34,348 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.52 vs. limit=5.0 +2023-03-26 14:27:41,483 INFO [finetune.py:976] (1/7) Epoch 12, batch 2200, loss[loss=0.2268, simple_loss=0.2818, pruned_loss=0.0859, over 4814.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2599, pruned_loss=0.06331, over 954977.48 frames. ], batch size: 38, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:28:03,263 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.110e+02 1.670e+02 2.055e+02 2.491e+02 4.530e+02, threshold=4.111e+02, percent-clipped=2.0 +2023-03-26 14:28:08,783 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65245.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:28:15,258 INFO [finetune.py:976] (1/7) Epoch 12, batch 2250, loss[loss=0.2386, simple_loss=0.2992, pruned_loss=0.08902, over 4745.00 frames. ], tot_loss[loss=0.1956, simple_loss=0.2624, pruned_loss=0.06435, over 956548.68 frames. ], batch size: 59, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:28:27,725 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5192, 1.4709, 1.3135, 1.4116, 1.7674, 1.6999, 1.4555, 1.3243], + device='cuda:1'), covar=tensor([0.0288, 0.0283, 0.0598, 0.0307, 0.0240, 0.0394, 0.0341, 0.0381], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0108, 0.0139, 0.0114, 0.0102, 0.0104, 0.0093, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.1310e-05, 8.4070e-05, 1.1085e-04, 8.8622e-05, 7.9826e-05, 7.6731e-05, + 7.0512e-05, 8.3052e-05], device='cuda:1') +2023-03-26 14:28:36,008 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65285.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:28:36,182 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.94 vs. limit=5.0 +2023-03-26 14:28:39,801 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-26 14:28:41,278 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65293.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:28:48,506 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65304.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:28:49,046 INFO [finetune.py:976] (1/7) Epoch 12, batch 2300, loss[loss=0.2077, simple_loss=0.2645, pruned_loss=0.07542, over 4886.00 frames. ], tot_loss[loss=0.1955, simple_loss=0.2629, pruned_loss=0.06404, over 957329.48 frames. ], batch size: 35, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:28:50,944 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0687, 3.5175, 3.7302, 3.9059, 3.8105, 3.5911, 4.1590, 1.2646], + device='cuda:1'), covar=tensor([0.0807, 0.0859, 0.0844, 0.1066, 0.1211, 0.1573, 0.0767, 0.5690], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0243, 0.0275, 0.0291, 0.0330, 0.0282, 0.0301, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:29:07,484 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65332.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:29:10,444 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.717e+01 1.673e+02 1.949e+02 2.267e+02 6.743e+02, threshold=3.897e+02, percent-clipped=1.0 +2023-03-26 14:29:16,556 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65346.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:29:20,077 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65352.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:29:22,340 INFO [finetune.py:976] (1/7) Epoch 12, batch 2350, loss[loss=0.2045, simple_loss=0.2619, pruned_loss=0.07358, over 4823.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2611, pruned_loss=0.0638, over 956856.52 frames. ], batch size: 30, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:29:24,815 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65359.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:29:38,187 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2232, 1.9525, 2.6725, 1.7741, 2.4194, 2.3449, 1.9446, 2.5204], + device='cuda:1'), covar=tensor([0.1188, 0.1872, 0.1406, 0.1812, 0.0619, 0.1328, 0.2211, 0.0819], + device='cuda:1'), in_proj_covar=tensor([0.0198, 0.0208, 0.0197, 0.0194, 0.0179, 0.0217, 0.0219, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:30:02,647 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65401.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:30:04,996 INFO [finetune.py:976] (1/7) Epoch 12, batch 2400, loss[loss=0.1953, simple_loss=0.2621, pruned_loss=0.06421, over 4918.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2574, pruned_loss=0.0627, over 954112.71 frames. ], batch size: 36, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:30:06,251 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65407.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:30:06,316 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6448, 1.6539, 1.7825, 0.8965, 1.7599, 2.0575, 1.9876, 1.5563], + device='cuda:1'), covar=tensor([0.0957, 0.0685, 0.0527, 0.0635, 0.0471, 0.0560, 0.0382, 0.0733], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0152, 0.0121, 0.0131, 0.0130, 0.0126, 0.0142, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.2970e-05, 1.1143e-04, 8.6971e-05, 9.4606e-05, 9.2613e-05, 9.1432e-05, + 1.0393e-04, 1.0595e-04], device='cuda:1') +2023-03-26 14:30:07,395 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65408.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:30:09,181 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65411.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:30:26,324 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.174e+02 1.534e+02 1.885e+02 2.327e+02 5.518e+02, threshold=3.771e+02, percent-clipped=1.0 +2023-03-26 14:30:34,672 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65449.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:30:38,806 INFO [finetune.py:976] (1/7) Epoch 12, batch 2450, loss[loss=0.191, simple_loss=0.2622, pruned_loss=0.05992, over 4756.00 frames. ], tot_loss[loss=0.1895, simple_loss=0.2549, pruned_loss=0.06201, over 955951.68 frames. ], batch size: 54, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:30:39,468 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65456.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:31:31,329 INFO [finetune.py:976] (1/7) Epoch 12, batch 2500, loss[loss=0.1645, simple_loss=0.2283, pruned_loss=0.05037, over 4754.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.2577, pruned_loss=0.06308, over 956294.56 frames. ], batch size: 26, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:31:49,235 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65532.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:31:52,647 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.473e+01 1.650e+02 2.020e+02 2.341e+02 4.049e+02, threshold=4.040e+02, percent-clipped=1.0 +2023-03-26 14:31:55,254 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 14:32:06,903 INFO [finetune.py:976] (1/7) Epoch 12, batch 2550, loss[loss=0.1782, simple_loss=0.2459, pruned_loss=0.05527, over 4819.00 frames. ], tot_loss[loss=0.1952, simple_loss=0.2614, pruned_loss=0.06455, over 953959.97 frames. ], batch size: 30, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:32:27,324 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3274, 2.2642, 2.1893, 2.4615, 2.9364, 2.4110, 2.2177, 1.8731], + device='cuda:1'), covar=tensor([0.2055, 0.1838, 0.1643, 0.1479, 0.1518, 0.0972, 0.1940, 0.1735], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0208, 0.0210, 0.0189, 0.0242, 0.0183, 0.0213, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:32:40,909 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65593.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 14:32:43,297 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3281, 1.8815, 2.7413, 1.8165, 2.4410, 2.4395, 1.8840, 2.5853], + device='cuda:1'), covar=tensor([0.1006, 0.2042, 0.1181, 0.1760, 0.0572, 0.1185, 0.2485, 0.0655], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0206, 0.0194, 0.0191, 0.0179, 0.0216, 0.0217, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:32:48,713 INFO [finetune.py:976] (1/7) Epoch 12, batch 2600, loss[loss=0.2375, simple_loss=0.297, pruned_loss=0.08895, over 4919.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2616, pruned_loss=0.0645, over 951861.97 frames. ], batch size: 38, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:32:56,498 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4327, 3.8496, 4.0345, 4.2336, 4.1719, 3.9223, 4.5018, 1.3949], + device='cuda:1'), covar=tensor([0.0663, 0.0822, 0.0709, 0.0803, 0.1075, 0.1432, 0.0665, 0.5378], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0242, 0.0276, 0.0291, 0.0330, 0.0282, 0.0300, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:33:06,626 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65632.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:33:08,912 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3457, 2.5429, 2.2511, 1.7668, 2.4197, 2.7077, 2.4301, 2.1917], + device='cuda:1'), covar=tensor([0.0603, 0.0621, 0.0800, 0.0956, 0.0858, 0.0692, 0.0687, 0.1010], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0135, 0.0142, 0.0125, 0.0123, 0.0143, 0.0145, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:33:10,010 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.524e+01 1.645e+02 2.049e+02 2.494e+02 4.393e+02, threshold=4.097e+02, percent-clipped=1.0 +2023-03-26 14:33:10,741 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1759, 1.3151, 1.4316, 0.7211, 1.2424, 1.5892, 1.5911, 1.3093], + device='cuda:1'), covar=tensor([0.0854, 0.0585, 0.0415, 0.0464, 0.0462, 0.0500, 0.0361, 0.0595], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0153, 0.0122, 0.0132, 0.0130, 0.0126, 0.0143, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.3271e-05, 1.1166e-04, 8.7761e-05, 9.4853e-05, 9.2776e-05, 9.1318e-05, + 1.0414e-04, 1.0645e-04], device='cuda:1') +2023-03-26 14:33:12,535 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65641.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:33:22,368 INFO [finetune.py:976] (1/7) Epoch 12, batch 2650, loss[loss=0.1896, simple_loss=0.2688, pruned_loss=0.05522, over 4852.00 frames. ], tot_loss[loss=0.1972, simple_loss=0.2638, pruned_loss=0.06529, over 952340.51 frames. ], batch size: 44, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:33:38,618 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65680.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:33:43,846 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6159, 1.4893, 2.0407, 3.1454, 2.1716, 2.2619, 0.9311, 2.5370], + device='cuda:1'), covar=tensor([0.1732, 0.1469, 0.1226, 0.0673, 0.0780, 0.1454, 0.1849, 0.0611], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0116, 0.0135, 0.0165, 0.0101, 0.0138, 0.0126, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 14:33:47,582 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 14:33:55,611 INFO [finetune.py:976] (1/7) Epoch 12, batch 2700, loss[loss=0.1963, simple_loss=0.2621, pruned_loss=0.06526, over 4862.00 frames. ], tot_loss[loss=0.1949, simple_loss=0.2617, pruned_loss=0.06408, over 950247.31 frames. ], batch size: 34, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:33:59,758 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65711.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:34:17,009 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.115e+02 1.547e+02 1.884e+02 2.200e+02 3.210e+02, threshold=3.769e+02, percent-clipped=0.0 +2023-03-26 14:34:30,260 INFO [finetune.py:976] (1/7) Epoch 12, batch 2750, loss[loss=0.2141, simple_loss=0.2718, pruned_loss=0.07819, over 4839.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2589, pruned_loss=0.0633, over 949338.06 frames. ], batch size: 47, lr: 3.65e-03, grad_scale: 32.0 +2023-03-26 14:34:38,001 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65759.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:35:20,815 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4230, 3.8461, 4.0655, 4.2670, 4.2144, 3.8920, 4.4802, 1.4475], + device='cuda:1'), covar=tensor([0.0719, 0.0778, 0.0793, 0.0808, 0.1080, 0.1352, 0.0616, 0.5497], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0243, 0.0277, 0.0292, 0.0332, 0.0283, 0.0302, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:35:30,777 INFO [finetune.py:976] (1/7) Epoch 12, batch 2800, loss[loss=0.1691, simple_loss=0.2354, pruned_loss=0.05137, over 4848.00 frames. ], tot_loss[loss=0.1905, simple_loss=0.2561, pruned_loss=0.06247, over 952322.29 frames. ], batch size: 44, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:35:52,227 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.914e+01 1.578e+02 1.887e+02 2.176e+02 5.167e+02, threshold=3.774e+02, percent-clipped=1.0 +2023-03-26 14:35:52,958 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65838.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:36:04,208 INFO [finetune.py:976] (1/7) Epoch 12, batch 2850, loss[loss=0.1475, simple_loss=0.2321, pruned_loss=0.03146, over 4781.00 frames. ], tot_loss[loss=0.1875, simple_loss=0.2533, pruned_loss=0.0609, over 950858.72 frames. ], batch size: 29, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:36:36,962 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65888.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:36:45,225 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65899.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:36:48,785 INFO [finetune.py:976] (1/7) Epoch 12, batch 2900, loss[loss=0.2087, simple_loss=0.2791, pruned_loss=0.06912, over 4818.00 frames. ], tot_loss[loss=0.1909, simple_loss=0.2571, pruned_loss=0.06237, over 952518.18 frames. ], batch size: 39, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:37:10,262 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.038e+02 1.527e+02 1.842e+02 2.377e+02 4.547e+02, threshold=3.684e+02, percent-clipped=3.0 +2023-03-26 14:37:10,411 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8113, 2.5595, 1.9505, 1.1743, 2.1705, 2.2934, 2.1045, 2.2374], + device='cuda:1'), covar=tensor([0.0837, 0.0771, 0.1592, 0.1937, 0.1623, 0.1762, 0.1976, 0.0948], + device='cuda:1'), in_proj_covar=tensor([0.0166, 0.0198, 0.0200, 0.0185, 0.0215, 0.0207, 0.0223, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:37:12,784 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65941.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:37:27,693 INFO [finetune.py:976] (1/7) Epoch 12, batch 2950, loss[loss=0.2241, simple_loss=0.2821, pruned_loss=0.08305, over 4774.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.2587, pruned_loss=0.06281, over 949885.35 frames. ], batch size: 27, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:37:42,551 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.53 vs. limit=5.0 +2023-03-26 14:38:02,422 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=65989.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:38:26,400 INFO [finetune.py:976] (1/7) Epoch 12, batch 3000, loss[loss=0.2024, simple_loss=0.2737, pruned_loss=0.06559, over 4822.00 frames. ], tot_loss[loss=0.1938, simple_loss=0.2601, pruned_loss=0.06375, over 950875.56 frames. ], batch size: 39, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:38:26,400 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 14:38:31,366 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1588, 2.0490, 1.5911, 0.7619, 1.8137, 1.8283, 1.7081, 1.8934], + device='cuda:1'), covar=tensor([0.0840, 0.0637, 0.1543, 0.1868, 0.1343, 0.2092, 0.2265, 0.0884], + device='cuda:1'), in_proj_covar=tensor([0.0166, 0.0198, 0.0201, 0.0185, 0.0214, 0.0207, 0.0223, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:38:37,072 INFO [finetune.py:1010] (1/7) Epoch 12, validation: loss=0.1571, simple_loss=0.2281, pruned_loss=0.04309, over 2265189.00 frames. +2023-03-26 14:38:37,073 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 14:38:53,469 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 14:38:58,511 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.011e+02 1.619e+02 1.943e+02 2.343e+02 4.325e+02, threshold=3.886e+02, percent-clipped=3.0 +2023-03-26 14:39:21,446 INFO [finetune.py:976] (1/7) Epoch 12, batch 3050, loss[loss=0.1527, simple_loss=0.2287, pruned_loss=0.03839, over 4793.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2606, pruned_loss=0.06398, over 950440.84 frames. ], batch size: 45, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:39:32,829 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.90 vs. limit=5.0 +2023-03-26 14:39:53,019 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 14:40:36,353 INFO [finetune.py:976] (1/7) Epoch 12, batch 3100, loss[loss=0.172, simple_loss=0.2459, pruned_loss=0.0491, over 4824.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.2585, pruned_loss=0.06282, over 952313.44 frames. ], batch size: 38, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:41:19,789 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.081e+02 1.628e+02 1.972e+02 2.398e+02 4.316e+02, threshold=3.945e+02, percent-clipped=3.0 +2023-03-26 14:41:27,578 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66144.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:41:40,144 INFO [finetune.py:976] (1/7) Epoch 12, batch 3150, loss[loss=0.15, simple_loss=0.209, pruned_loss=0.04553, over 4716.00 frames. ], tot_loss[loss=0.19, simple_loss=0.2556, pruned_loss=0.06223, over 950786.71 frames. ], batch size: 23, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:41:43,352 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66160.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:42:24,556 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66188.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 14:42:32,946 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66194.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 14:42:35,815 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2119, 2.9124, 2.7603, 1.2126, 2.9694, 2.2255, 0.6444, 1.8313], + device='cuda:1'), covar=tensor([0.2389, 0.2147, 0.1781, 0.3303, 0.1383, 0.1073, 0.3964, 0.1646], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0175, 0.0161, 0.0129, 0.0156, 0.0122, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 14:42:40,455 INFO [finetune.py:976] (1/7) Epoch 12, batch 3200, loss[loss=0.1814, simple_loss=0.2346, pruned_loss=0.06414, over 4913.00 frames. ], tot_loss[loss=0.1877, simple_loss=0.2525, pruned_loss=0.06144, over 949365.98 frames. ], batch size: 36, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:42:40,564 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66205.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:42:51,181 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66221.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:43:01,311 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=66236.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:43:01,827 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.059e+02 1.602e+02 1.878e+02 2.419e+02 4.134e+02, threshold=3.755e+02, percent-clipped=1.0 +2023-03-26 14:43:03,245 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.71 vs. limit=2.0 +2023-03-26 14:43:11,565 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-03-26 14:43:14,218 INFO [finetune.py:976] (1/7) Epoch 12, batch 3250, loss[loss=0.1984, simple_loss=0.2571, pruned_loss=0.06984, over 4875.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2529, pruned_loss=0.06198, over 949448.83 frames. ], batch size: 31, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:43:14,949 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4145, 3.7773, 4.0352, 4.2481, 4.1805, 3.8646, 4.4643, 1.3557], + device='cuda:1'), covar=tensor([0.0645, 0.0797, 0.0761, 0.0895, 0.0991, 0.1478, 0.0677, 0.5336], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0242, 0.0274, 0.0289, 0.0327, 0.0282, 0.0299, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:43:16,203 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66258.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:43:28,793 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8075, 1.2175, 0.8788, 1.7636, 2.1460, 1.4828, 1.4878, 1.6275], + device='cuda:1'), covar=tensor([0.1539, 0.2311, 0.2158, 0.1271, 0.1966, 0.2024, 0.1683, 0.2069], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0112, 0.0092, 0.0120, 0.0094, 0.0100, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 14:43:42,484 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8056, 3.9963, 3.7543, 1.8152, 4.0412, 3.1516, 0.8849, 2.7624], + device='cuda:1'), covar=tensor([0.2143, 0.1477, 0.1257, 0.3053, 0.0780, 0.0789, 0.3919, 0.1336], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0175, 0.0160, 0.0129, 0.0156, 0.0122, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 14:43:48,271 INFO [finetune.py:976] (1/7) Epoch 12, batch 3300, loss[loss=0.217, simple_loss=0.2885, pruned_loss=0.07271, over 4818.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2572, pruned_loss=0.06289, over 952031.46 frames. ], batch size: 41, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:43:56,980 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66319.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:44:14,652 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.247e+02 1.642e+02 1.904e+02 2.370e+02 4.024e+02, threshold=3.808e+02, percent-clipped=2.0 +2023-03-26 14:44:29,722 INFO [finetune.py:976] (1/7) Epoch 12, batch 3350, loss[loss=0.1859, simple_loss=0.2584, pruned_loss=0.05668, over 4849.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.2585, pruned_loss=0.06261, over 954574.86 frames. ], batch size: 44, lr: 3.64e-03, grad_scale: 64.0 +2023-03-26 14:44:41,256 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7593, 1.2254, 0.9935, 1.6766, 2.0358, 1.2989, 1.4816, 1.7000], + device='cuda:1'), covar=tensor([0.1351, 0.2047, 0.1841, 0.1096, 0.1944, 0.1935, 0.1390, 0.1779], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0113, 0.0092, 0.0120, 0.0094, 0.0100, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 14:45:02,613 INFO [finetune.py:976] (1/7) Epoch 12, batch 3400, loss[loss=0.1869, simple_loss=0.2637, pruned_loss=0.05503, over 4818.00 frames. ], tot_loss[loss=0.1947, simple_loss=0.2613, pruned_loss=0.06409, over 954540.42 frames. ], batch size: 38, lr: 3.64e-03, grad_scale: 64.0 +2023-03-26 14:45:10,999 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5776, 1.4498, 1.4638, 1.4608, 0.8550, 2.8979, 1.0104, 1.5489], + device='cuda:1'), covar=tensor([0.3279, 0.2467, 0.2030, 0.2331, 0.2028, 0.0251, 0.2730, 0.1309], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0124, 0.0115, 0.0098, 0.0098, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 14:45:12,838 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4003, 1.2794, 1.3438, 1.2853, 0.8145, 2.2700, 0.7388, 1.2592], + device='cuda:1'), covar=tensor([0.3416, 0.2495, 0.2108, 0.2486, 0.2024, 0.0384, 0.2790, 0.1311], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0124, 0.0115, 0.0098, 0.0098, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 14:45:14,664 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0663, 1.8624, 1.6246, 1.8615, 1.6908, 1.6992, 1.7453, 2.5485], + device='cuda:1'), covar=tensor([0.3832, 0.4341, 0.3374, 0.4081, 0.4464, 0.2525, 0.4089, 0.1619], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0224, 0.0277, 0.0244, 0.0211, 0.0247, 0.0218], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:45:24,436 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.132e+02 1.693e+02 1.994e+02 2.432e+02 3.824e+02, threshold=3.988e+02, percent-clipped=2.0 +2023-03-26 14:45:28,899 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2399, 1.2915, 1.4489, 1.0545, 1.2426, 1.3754, 1.2291, 1.5152], + device='cuda:1'), covar=tensor([0.0944, 0.1492, 0.1076, 0.1288, 0.0690, 0.1054, 0.2227, 0.0634], + device='cuda:1'), in_proj_covar=tensor([0.0195, 0.0204, 0.0192, 0.0189, 0.0177, 0.0213, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:45:36,095 INFO [finetune.py:976] (1/7) Epoch 12, batch 3450, loss[loss=0.189, simple_loss=0.2569, pruned_loss=0.06059, over 4786.00 frames. ], tot_loss[loss=0.1946, simple_loss=0.2617, pruned_loss=0.06379, over 955212.39 frames. ], batch size: 51, lr: 3.64e-03, grad_scale: 64.0 +2023-03-26 14:46:02,769 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66494.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 14:46:06,887 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66500.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 14:46:09,847 INFO [finetune.py:976] (1/7) Epoch 12, batch 3500, loss[loss=0.179, simple_loss=0.2396, pruned_loss=0.05914, over 4914.00 frames. ], tot_loss[loss=0.1935, simple_loss=0.26, pruned_loss=0.06353, over 956148.07 frames. ], batch size: 46, lr: 3.64e-03, grad_scale: 64.0 +2023-03-26 14:46:20,797 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66516.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:46:23,834 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66521.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:46:36,430 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.070e+02 1.655e+02 1.937e+02 2.486e+02 6.010e+02, threshold=3.875e+02, percent-clipped=2.0 +2023-03-26 14:46:40,002 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=66542.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:46:57,616 INFO [finetune.py:976] (1/7) Epoch 12, batch 3550, loss[loss=0.2366, simple_loss=0.2882, pruned_loss=0.09247, over 4852.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2579, pruned_loss=0.06375, over 956050.09 frames. ], batch size: 49, lr: 3.64e-03, grad_scale: 64.0 +2023-03-26 14:47:23,063 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66582.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 14:47:43,651 INFO [finetune.py:976] (1/7) Epoch 12, batch 3600, loss[loss=0.1708, simple_loss=0.24, pruned_loss=0.05077, over 4747.00 frames. ], tot_loss[loss=0.1913, simple_loss=0.2558, pruned_loss=0.06335, over 954839.52 frames. ], batch size: 54, lr: 3.64e-03, grad_scale: 64.0 +2023-03-26 14:47:53,904 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66614.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:47:56,180 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2753, 3.7049, 3.8900, 4.0891, 4.0321, 3.7660, 4.3576, 1.3847], + device='cuda:1'), covar=tensor([0.0825, 0.0821, 0.0932, 0.1119, 0.1228, 0.1622, 0.0695, 0.5607], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0244, 0.0277, 0.0292, 0.0330, 0.0283, 0.0301, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:48:02,929 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 14:48:08,736 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.078e+02 1.581e+02 1.999e+02 2.430e+02 3.919e+02, threshold=3.999e+02, percent-clipped=1.0 +2023-03-26 14:48:21,114 INFO [finetune.py:976] (1/7) Epoch 12, batch 3650, loss[loss=0.2378, simple_loss=0.3077, pruned_loss=0.08391, over 4874.00 frames. ], tot_loss[loss=0.1924, simple_loss=0.257, pruned_loss=0.06387, over 953902.34 frames. ], batch size: 34, lr: 3.64e-03, grad_scale: 64.0 +2023-03-26 14:48:34,264 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.37 vs. limit=5.0 +2023-03-26 14:48:36,558 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2395, 2.4015, 2.3129, 1.5740, 2.2476, 2.6526, 2.4847, 2.1110], + device='cuda:1'), covar=tensor([0.0695, 0.0577, 0.0763, 0.0975, 0.1024, 0.0627, 0.0626, 0.0980], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0133, 0.0141, 0.0123, 0.0122, 0.0142, 0.0142, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:48:51,366 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1731, 1.8051, 2.6318, 1.5933, 2.1998, 2.2969, 1.7798, 2.4469], + device='cuda:1'), covar=tensor([0.1219, 0.1933, 0.1588, 0.2215, 0.0814, 0.1545, 0.2704, 0.0867], + device='cuda:1'), in_proj_covar=tensor([0.0196, 0.0206, 0.0195, 0.0192, 0.0179, 0.0215, 0.0219, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:48:54,888 INFO [finetune.py:976] (1/7) Epoch 12, batch 3700, loss[loss=0.2173, simple_loss=0.2862, pruned_loss=0.07416, over 4911.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2601, pruned_loss=0.0643, over 955392.02 frames. ], batch size: 37, lr: 3.64e-03, grad_scale: 64.0 +2023-03-26 14:49:14,937 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66731.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:49:18,458 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.046e+02 1.675e+02 1.999e+02 2.466e+02 3.717e+02, threshold=3.997e+02, percent-clipped=0.0 +2023-03-26 14:49:32,890 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8660, 1.6601, 1.6828, 1.7499, 1.6188, 4.4790, 1.7381, 2.3288], + device='cuda:1'), covar=tensor([0.3201, 0.2478, 0.2045, 0.2263, 0.1518, 0.0114, 0.2414, 0.1191], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0124, 0.0116, 0.0098, 0.0098, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 14:49:38,724 INFO [finetune.py:976] (1/7) Epoch 12, batch 3750, loss[loss=0.2421, simple_loss=0.2978, pruned_loss=0.09323, over 4717.00 frames. ], tot_loss[loss=0.1966, simple_loss=0.2624, pruned_loss=0.06544, over 956586.02 frames. ], batch size: 59, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:49:51,424 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-26 14:50:01,267 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2660, 1.6093, 1.5682, 0.8537, 1.8435, 1.8055, 1.9103, 1.6257], + device='cuda:1'), covar=tensor([0.0812, 0.0758, 0.0601, 0.0690, 0.0514, 0.0773, 0.0470, 0.0847], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0154, 0.0124, 0.0133, 0.0131, 0.0127, 0.0144, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.4138e-05, 1.1268e-04, 8.9223e-05, 9.5632e-05, 9.2745e-05, 9.2492e-05, + 1.0535e-04, 1.0724e-04], device='cuda:1') +2023-03-26 14:50:03,079 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66792.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:50:08,991 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66800.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:50:12,436 INFO [finetune.py:976] (1/7) Epoch 12, batch 3800, loss[loss=0.1888, simple_loss=0.2705, pruned_loss=0.05354, over 4855.00 frames. ], tot_loss[loss=0.1973, simple_loss=0.2635, pruned_loss=0.06558, over 955331.69 frames. ], batch size: 44, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:50:19,278 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66816.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:50:34,226 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.175e+02 1.668e+02 2.101e+02 2.666e+02 4.038e+02, threshold=4.202e+02, percent-clipped=1.0 +2023-03-26 14:50:40,358 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=66848.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:50:45,464 INFO [finetune.py:976] (1/7) Epoch 12, batch 3850, loss[loss=0.2057, simple_loss=0.2701, pruned_loss=0.07071, over 4915.00 frames. ], tot_loss[loss=0.1958, simple_loss=0.2619, pruned_loss=0.06491, over 954000.20 frames. ], batch size: 37, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:50:51,372 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=66864.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:50:53,290 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-03-26 14:51:00,116 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66877.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 14:51:18,885 INFO [finetune.py:976] (1/7) Epoch 12, batch 3900, loss[loss=0.1887, simple_loss=0.2569, pruned_loss=0.06023, over 4800.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2586, pruned_loss=0.064, over 952207.45 frames. ], batch size: 45, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:51:24,971 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66914.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:51:40,886 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.325e+01 1.578e+02 1.785e+02 2.294e+02 5.103e+02, threshold=3.570e+02, percent-clipped=1.0 +2023-03-26 14:51:51,222 INFO [finetune.py:976] (1/7) Epoch 12, batch 3950, loss[loss=0.1853, simple_loss=0.236, pruned_loss=0.06728, over 4780.00 frames. ], tot_loss[loss=0.1901, simple_loss=0.2551, pruned_loss=0.06261, over 953778.48 frames. ], batch size: 26, lr: 3.64e-03, grad_scale: 32.0 +2023-03-26 14:51:53,020 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66957.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:51:58,481 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=66962.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:52:44,781 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7901, 1.6120, 1.4894, 1.8579, 2.2237, 1.8782, 1.4433, 1.4485], + device='cuda:1'), covar=tensor([0.2097, 0.2071, 0.1890, 0.1531, 0.1746, 0.1158, 0.2500, 0.1960], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0208, 0.0211, 0.0191, 0.0242, 0.0183, 0.0214, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:52:47,107 INFO [finetune.py:976] (1/7) Epoch 12, batch 4000, loss[loss=0.1462, simple_loss=0.2144, pruned_loss=0.03897, over 4766.00 frames. ], tot_loss[loss=0.1898, simple_loss=0.2546, pruned_loss=0.06253, over 953109.18 frames. ], batch size: 26, lr: 3.63e-03, grad_scale: 32.0 +2023-03-26 14:52:55,764 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 14:53:04,484 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67018.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 14:53:08,755 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6526, 1.5601, 1.4924, 1.5693, 1.4865, 4.2126, 1.5154, 1.8653], + device='cuda:1'), covar=tensor([0.3321, 0.2510, 0.2220, 0.2394, 0.1606, 0.0115, 0.2850, 0.1458], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0124, 0.0115, 0.0098, 0.0098, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 14:53:18,612 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.442e+01 1.595e+02 2.014e+02 2.521e+02 4.335e+02, threshold=4.027e+02, percent-clipped=3.0 +2023-03-26 14:53:28,873 INFO [finetune.py:976] (1/7) Epoch 12, batch 4050, loss[loss=0.2238, simple_loss=0.2892, pruned_loss=0.07914, over 4808.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2587, pruned_loss=0.06395, over 952486.90 frames. ], batch size: 38, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:53:35,421 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7458, 1.6635, 1.4449, 1.0885, 2.0311, 2.1217, 1.9574, 1.7153], + device='cuda:1'), covar=tensor([0.0406, 0.0474, 0.0779, 0.0538, 0.0338, 0.0660, 0.0387, 0.0566], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0110, 0.0140, 0.0114, 0.0103, 0.0105, 0.0095, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.2504e-05, 8.5134e-05, 1.1139e-04, 8.8885e-05, 8.0178e-05, 7.8044e-05, + 7.1610e-05, 8.4141e-05], device='cuda:1') +2023-03-26 14:53:47,878 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67083.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:53:50,803 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67087.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 14:53:51,582 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.96 vs. limit=5.0 +2023-03-26 14:54:02,028 INFO [finetune.py:976] (1/7) Epoch 12, batch 4100, loss[loss=0.1643, simple_loss=0.2414, pruned_loss=0.04363, over 4792.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2617, pruned_loss=0.06417, over 953930.92 frames. ], batch size: 29, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:54:17,491 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2351, 2.2101, 2.3790, 1.6267, 2.3005, 2.5587, 2.3970, 2.0176], + device='cuda:1'), covar=tensor([0.0634, 0.0608, 0.0663, 0.0917, 0.0660, 0.0641, 0.0615, 0.1034], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0133, 0.0141, 0.0124, 0.0122, 0.0141, 0.0141, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:54:29,988 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.152e+02 1.725e+02 1.998e+02 2.409e+02 3.172e+02, threshold=3.997e+02, percent-clipped=0.0 +2023-03-26 14:54:34,072 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67144.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:54:44,209 INFO [finetune.py:976] (1/7) Epoch 12, batch 4150, loss[loss=0.176, simple_loss=0.2297, pruned_loss=0.06119, over 3943.00 frames. ], tot_loss[loss=0.1962, simple_loss=0.2627, pruned_loss=0.06489, over 951235.59 frames. ], batch size: 17, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:54:48,609 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5282, 1.9323, 1.4519, 1.5328, 2.0826, 1.9591, 1.8215, 1.7589], + device='cuda:1'), covar=tensor([0.0616, 0.0312, 0.0643, 0.0346, 0.0294, 0.0695, 0.0378, 0.0438], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0109, 0.0139, 0.0113, 0.0102, 0.0104, 0.0094, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2156e-05, 8.4520e-05, 1.1046e-04, 8.8186e-05, 7.9655e-05, 7.7236e-05, + 7.1229e-05, 8.3393e-05], device='cuda:1') +2023-03-26 14:54:59,053 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67177.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:54:59,101 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7437, 1.6528, 1.4806, 1.8288, 2.3008, 1.9124, 1.4816, 1.4144], + device='cuda:1'), covar=tensor([0.2415, 0.2107, 0.2158, 0.1827, 0.1823, 0.1262, 0.2619, 0.2275], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0206, 0.0210, 0.0190, 0.0240, 0.0182, 0.0213, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:55:01,460 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0687, 3.4920, 3.7642, 3.8462, 3.8795, 3.6434, 4.1026, 1.8563], + device='cuda:1'), covar=tensor([0.0717, 0.0775, 0.0687, 0.0870, 0.0977, 0.1229, 0.0660, 0.4077], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0241, 0.0274, 0.0289, 0.0326, 0.0280, 0.0298, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:55:17,545 INFO [finetune.py:976] (1/7) Epoch 12, batch 4200, loss[loss=0.2214, simple_loss=0.2795, pruned_loss=0.08163, over 4787.00 frames. ], tot_loss[loss=0.1955, simple_loss=0.2619, pruned_loss=0.0645, over 951239.93 frames. ], batch size: 29, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:55:30,583 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=67225.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:55:39,456 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.066e+02 1.549e+02 1.852e+02 2.427e+02 4.145e+02, threshold=3.704e+02, percent-clipped=1.0 +2023-03-26 14:55:40,670 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5869, 1.6908, 1.2937, 1.5700, 1.9400, 1.8318, 1.5912, 1.4279], + device='cuda:1'), covar=tensor([0.0382, 0.0275, 0.0580, 0.0325, 0.0208, 0.0499, 0.0385, 0.0430], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0109, 0.0139, 0.0113, 0.0102, 0.0104, 0.0095, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2388e-05, 8.4727e-05, 1.1069e-04, 8.8460e-05, 7.9847e-05, 7.7403e-05, + 7.1499e-05, 8.3761e-05], device='cuda:1') +2023-03-26 14:55:50,256 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 14:55:50,530 INFO [finetune.py:976] (1/7) Epoch 12, batch 4250, loss[loss=0.1719, simple_loss=0.2206, pruned_loss=0.06163, over 4270.00 frames. ], tot_loss[loss=0.1934, simple_loss=0.2591, pruned_loss=0.06387, over 951476.46 frames. ], batch size: 18, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:56:09,122 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1016, 1.8096, 2.1541, 2.1004, 1.8055, 1.8231, 2.0021, 1.9388], + device='cuda:1'), covar=tensor([0.4404, 0.4294, 0.3547, 0.4364, 0.5151, 0.4235, 0.5373, 0.3719], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0237, 0.0253, 0.0259, 0.0256, 0.0231, 0.0273, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 14:56:32,184 INFO [finetune.py:976] (1/7) Epoch 12, batch 4300, loss[loss=0.1678, simple_loss=0.2276, pruned_loss=0.05397, over 4702.00 frames. ], tot_loss[loss=0.1901, simple_loss=0.2556, pruned_loss=0.06232, over 950008.00 frames. ], batch size: 23, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:56:37,167 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67313.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 14:56:54,399 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.954e+01 1.550e+02 1.913e+02 2.348e+02 5.397e+02, threshold=3.825e+02, percent-clipped=3.0 +2023-03-26 14:57:05,081 INFO [finetune.py:976] (1/7) Epoch 12, batch 4350, loss[loss=0.1914, simple_loss=0.2536, pruned_loss=0.06462, over 4760.00 frames. ], tot_loss[loss=0.1887, simple_loss=0.2534, pruned_loss=0.06203, over 950781.43 frames. ], batch size: 28, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:57:23,351 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1643, 1.7471, 2.5873, 4.1783, 2.8964, 2.9433, 0.9444, 3.4951], + device='cuda:1'), covar=tensor([0.1609, 0.1517, 0.1327, 0.0492, 0.0723, 0.1383, 0.1989, 0.0372], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0116, 0.0134, 0.0165, 0.0101, 0.0137, 0.0126, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 14:57:28,520 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67387.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 14:57:39,911 INFO [finetune.py:976] (1/7) Epoch 12, batch 4400, loss[loss=0.1434, simple_loss=0.209, pruned_loss=0.03894, over 4197.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.255, pruned_loss=0.06312, over 950842.58 frames. ], batch size: 18, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:57:58,957 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-03-26 14:58:14,140 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=67435.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 14:58:16,974 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.120e+02 1.656e+02 1.972e+02 2.339e+02 4.406e+02, threshold=3.944e+02, percent-clipped=2.0 +2023-03-26 14:58:17,064 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67439.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:58:24,284 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1882, 1.3326, 1.3947, 0.7012, 1.3204, 1.5604, 1.6031, 1.2893], + device='cuda:1'), covar=tensor([0.0902, 0.0515, 0.0467, 0.0518, 0.0439, 0.0537, 0.0306, 0.0616], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0155, 0.0123, 0.0133, 0.0132, 0.0127, 0.0145, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.3892e-05, 1.1321e-04, 8.8936e-05, 9.5962e-05, 9.3784e-05, 9.2232e-05, + 1.0589e-04, 1.0709e-04], device='cuda:1') +2023-03-26 14:58:30,809 INFO [finetune.py:976] (1/7) Epoch 12, batch 4450, loss[loss=0.1741, simple_loss=0.2469, pruned_loss=0.0506, over 4885.00 frames. ], tot_loss[loss=0.1936, simple_loss=0.259, pruned_loss=0.06409, over 949539.17 frames. ], batch size: 32, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:59:03,971 INFO [finetune.py:976] (1/7) Epoch 12, batch 4500, loss[loss=0.2126, simple_loss=0.2789, pruned_loss=0.07312, over 4197.00 frames. ], tot_loss[loss=0.1963, simple_loss=0.2624, pruned_loss=0.06514, over 951429.50 frames. ], batch size: 65, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:59:04,075 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1731, 1.6589, 0.7838, 2.1032, 2.4775, 1.9039, 1.9339, 2.0362], + device='cuda:1'), covar=tensor([0.1384, 0.1876, 0.2198, 0.1067, 0.1856, 0.1951, 0.1352, 0.1988], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0113, 0.0092, 0.0121, 0.0094, 0.0100, 0.0091], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 14:59:09,850 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67513.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 14:59:20,666 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4923, 1.3805, 1.3452, 1.4685, 0.8683, 3.0626, 1.0555, 1.4546], + device='cuda:1'), covar=tensor([0.3617, 0.2638, 0.2391, 0.2562, 0.2199, 0.0245, 0.2832, 0.1481], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0121, 0.0124, 0.0116, 0.0098, 0.0098, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 14:59:24,343 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7418, 1.6143, 1.5916, 1.6759, 1.2505, 3.4565, 1.4132, 1.8444], + device='cuda:1'), covar=tensor([0.3543, 0.2479, 0.2178, 0.2389, 0.1886, 0.0211, 0.2519, 0.1272], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0121, 0.0124, 0.0116, 0.0098, 0.0098, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 14:59:26,032 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.163e+02 1.686e+02 1.980e+02 2.352e+02 4.001e+02, threshold=3.961e+02, percent-clipped=1.0 +2023-03-26 14:59:37,241 INFO [finetune.py:976] (1/7) Epoch 12, batch 4550, loss[loss=0.2364, simple_loss=0.2931, pruned_loss=0.08984, over 4879.00 frames. ], tot_loss[loss=0.1973, simple_loss=0.2633, pruned_loss=0.06562, over 951176.58 frames. ], batch size: 43, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 14:59:56,256 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67574.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 15:00:19,986 INFO [finetune.py:976] (1/7) Epoch 12, batch 4600, loss[loss=0.1988, simple_loss=0.2626, pruned_loss=0.06753, over 4836.00 frames. ], tot_loss[loss=0.1959, simple_loss=0.2622, pruned_loss=0.06481, over 952425.51 frames. ], batch size: 49, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:00:24,951 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67613.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:00:24,998 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4919, 2.1951, 1.9283, 2.5329, 2.2110, 2.1552, 2.1978, 3.1056], + device='cuda:1'), covar=tensor([0.4402, 0.6063, 0.3806, 0.4867, 0.4646, 0.2624, 0.5044, 0.1750], + device='cuda:1'), in_proj_covar=tensor([0.0284, 0.0259, 0.0223, 0.0276, 0.0243, 0.0210, 0.0246, 0.0218], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:00:26,033 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5648, 3.9519, 4.1604, 4.3723, 4.2977, 4.0082, 4.6464, 1.5766], + device='cuda:1'), covar=tensor([0.0732, 0.0832, 0.0791, 0.1010, 0.1276, 0.1606, 0.0648, 0.5380], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0242, 0.0276, 0.0289, 0.0329, 0.0280, 0.0299, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:00:42,114 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.172e+02 1.477e+02 1.878e+02 2.272e+02 4.960e+02, threshold=3.756e+02, percent-clipped=1.0 +2023-03-26 15:00:53,234 INFO [finetune.py:976] (1/7) Epoch 12, batch 4650, loss[loss=0.1576, simple_loss=0.2251, pruned_loss=0.04507, over 4934.00 frames. ], tot_loss[loss=0.1932, simple_loss=0.2593, pruned_loss=0.06358, over 954512.97 frames. ], batch size: 33, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:00:54,116 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-26 15:00:56,979 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=67661.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:01:10,464 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67680.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:01:17,715 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67692.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:01:31,301 INFO [finetune.py:976] (1/7) Epoch 12, batch 4700, loss[loss=0.1411, simple_loss=0.2078, pruned_loss=0.03719, over 4902.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2565, pruned_loss=0.06275, over 953096.40 frames. ], batch size: 32, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:01:38,779 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.40 vs. limit=5.0 +2023-03-26 15:01:56,970 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.008e+02 1.553e+02 1.823e+02 2.116e+02 3.808e+02, threshold=3.646e+02, percent-clipped=1.0 +2023-03-26 15:01:57,076 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67739.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:01:58,325 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67741.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:02:05,966 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67753.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 15:02:07,553 INFO [finetune.py:976] (1/7) Epoch 12, batch 4750, loss[loss=0.1948, simple_loss=0.2612, pruned_loss=0.0642, over 4833.00 frames. ], tot_loss[loss=0.1895, simple_loss=0.2543, pruned_loss=0.06237, over 952615.62 frames. ], batch size: 39, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:02:28,907 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=67787.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:02:30,168 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9340, 1.9384, 2.0148, 1.4175, 2.0001, 2.1822, 2.1016, 1.6801], + device='cuda:1'), covar=tensor([0.0682, 0.0604, 0.0717, 0.0872, 0.0680, 0.0636, 0.0594, 0.1184], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0133, 0.0141, 0.0124, 0.0122, 0.0141, 0.0142, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:02:33,864 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6878, 1.5003, 1.3726, 1.5982, 2.2759, 1.7670, 1.5743, 1.3843], + device='cuda:1'), covar=tensor([0.2445, 0.2339, 0.2307, 0.2041, 0.1842, 0.1410, 0.2568, 0.2247], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0206, 0.0210, 0.0190, 0.0240, 0.0182, 0.0213, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:02:40,328 INFO [finetune.py:976] (1/7) Epoch 12, batch 4800, loss[loss=0.1902, simple_loss=0.2546, pruned_loss=0.06289, over 4863.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.257, pruned_loss=0.06373, over 954588.58 frames. ], batch size: 31, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:02:51,809 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5047, 1.6874, 1.6668, 1.0003, 1.7208, 1.9551, 1.9951, 1.4885], + device='cuda:1'), covar=tensor([0.0980, 0.0600, 0.0580, 0.0534, 0.0555, 0.0550, 0.0325, 0.0639], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0155, 0.0124, 0.0133, 0.0133, 0.0128, 0.0146, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.4294e-05, 1.1356e-04, 8.9099e-05, 9.5651e-05, 9.4133e-05, 9.2839e-05, + 1.0625e-04, 1.0724e-04], device='cuda:1') +2023-03-26 15:02:53,511 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1316, 2.0731, 1.6642, 2.2606, 2.0333, 1.7138, 2.5397, 2.0945], + device='cuda:1'), covar=tensor([0.1490, 0.2272, 0.3418, 0.2704, 0.2860, 0.1871, 0.3433, 0.1983], + device='cuda:1'), in_proj_covar=tensor([0.0179, 0.0188, 0.0234, 0.0257, 0.0243, 0.0200, 0.0214, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:03:07,513 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.086e+02 1.756e+02 1.975e+02 2.556e+02 4.813e+02, threshold=3.950e+02, percent-clipped=3.0 +2023-03-26 15:03:10,216 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-26 15:03:16,792 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8909, 1.8445, 1.9551, 1.2505, 1.9494, 2.1185, 1.9749, 1.6334], + device='cuda:1'), covar=tensor([0.0660, 0.0659, 0.0683, 0.0956, 0.0585, 0.0662, 0.0662, 0.1055], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0134, 0.0143, 0.0125, 0.0123, 0.0142, 0.0143, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:03:25,929 INFO [finetune.py:976] (1/7) Epoch 12, batch 4850, loss[loss=0.2138, simple_loss=0.2877, pruned_loss=0.06995, over 4830.00 frames. ], tot_loss[loss=0.1932, simple_loss=0.2595, pruned_loss=0.06342, over 953373.27 frames. ], batch size: 33, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:03:39,908 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67869.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 15:04:03,159 INFO [finetune.py:976] (1/7) Epoch 12, batch 4900, loss[loss=0.2097, simple_loss=0.2643, pruned_loss=0.07757, over 4759.00 frames. ], tot_loss[loss=0.1945, simple_loss=0.261, pruned_loss=0.06397, over 952245.04 frames. ], batch size: 28, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:04:26,938 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.121e+02 1.717e+02 1.971e+02 2.418e+02 4.222e+02, threshold=3.942e+02, percent-clipped=1.0 +2023-03-26 15:04:36,657 INFO [finetune.py:976] (1/7) Epoch 12, batch 4950, loss[loss=0.1939, simple_loss=0.2652, pruned_loss=0.06128, over 4901.00 frames. ], tot_loss[loss=0.1958, simple_loss=0.2629, pruned_loss=0.06433, over 953817.07 frames. ], batch size: 36, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:04:53,988 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67981.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:05:12,322 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.58 vs. limit=5.0 +2023-03-26 15:05:17,958 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6827, 1.5178, 0.9843, 0.2548, 1.3224, 1.5135, 1.3973, 1.3941], + device='cuda:1'), covar=tensor([0.0852, 0.0753, 0.1308, 0.1847, 0.1371, 0.2332, 0.2159, 0.0853], + device='cuda:1'), in_proj_covar=tensor([0.0166, 0.0198, 0.0200, 0.0186, 0.0214, 0.0207, 0.0222, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:05:20,898 INFO [finetune.py:976] (1/7) Epoch 12, batch 5000, loss[loss=0.1657, simple_loss=0.237, pruned_loss=0.0472, over 4900.00 frames. ], tot_loss[loss=0.1939, simple_loss=0.2608, pruned_loss=0.06353, over 953610.47 frames. ], batch size: 35, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:05:21,634 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5130, 1.5895, 1.3513, 1.5350, 1.8639, 1.7744, 1.5934, 1.3694], + device='cuda:1'), covar=tensor([0.0348, 0.0307, 0.0601, 0.0340, 0.0232, 0.0585, 0.0323, 0.0423], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0109, 0.0141, 0.0114, 0.0103, 0.0105, 0.0095, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.2583e-05, 8.4741e-05, 1.1154e-04, 8.8723e-05, 8.0193e-05, 7.7910e-05, + 7.1981e-05, 8.4292e-05], device='cuda:1') +2023-03-26 15:05:41,203 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68036.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:05:43,413 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.964e+01 1.543e+02 1.867e+02 2.301e+02 3.447e+02, threshold=3.734e+02, percent-clipped=0.0 +2023-03-26 15:05:46,316 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68042.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:05:50,383 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68048.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 15:05:54,513 INFO [finetune.py:976] (1/7) Epoch 12, batch 5050, loss[loss=0.1717, simple_loss=0.2339, pruned_loss=0.0548, over 4751.00 frames. ], tot_loss[loss=0.1925, simple_loss=0.2583, pruned_loss=0.06333, over 954007.09 frames. ], batch size: 27, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:06:15,507 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68087.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:06:27,690 INFO [finetune.py:976] (1/7) Epoch 12, batch 5100, loss[loss=0.1796, simple_loss=0.239, pruned_loss=0.06008, over 4074.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2544, pruned_loss=0.0616, over 954854.83 frames. ], batch size: 65, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:06:59,402 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.056e+02 1.565e+02 1.837e+02 2.198e+02 4.078e+02, threshold=3.675e+02, percent-clipped=2.0 +2023-03-26 15:07:05,468 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68148.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 15:07:10,948 INFO [finetune.py:976] (1/7) Epoch 12, batch 5150, loss[loss=0.1782, simple_loss=0.2441, pruned_loss=0.05614, over 4782.00 frames. ], tot_loss[loss=0.188, simple_loss=0.2535, pruned_loss=0.06119, over 952529.66 frames. ], batch size: 26, lr: 3.63e-03, grad_scale: 16.0 +2023-03-26 15:07:19,524 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68169.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:07:20,151 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.3245, 2.9186, 2.5408, 1.3851, 2.7701, 2.4235, 2.1565, 2.3886], + device='cuda:1'), covar=tensor([0.0999, 0.0938, 0.1988, 0.2274, 0.1715, 0.2167, 0.2203, 0.1370], + device='cuda:1'), in_proj_covar=tensor([0.0165, 0.0198, 0.0199, 0.0185, 0.0213, 0.0207, 0.0222, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:07:23,372 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-03-26 15:07:38,463 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 15:07:43,719 INFO [finetune.py:976] (1/7) Epoch 12, batch 5200, loss[loss=0.3138, simple_loss=0.3542, pruned_loss=0.1367, over 4783.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2569, pruned_loss=0.06256, over 952378.54 frames. ], batch size: 59, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:07:51,085 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=68217.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:08:05,785 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.176e+02 1.664e+02 1.889e+02 2.252e+02 3.665e+02, threshold=3.778e+02, percent-clipped=0.0 +2023-03-26 15:08:12,437 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0944, 3.5695, 3.7704, 3.8334, 3.8627, 3.6627, 4.1615, 1.7410], + device='cuda:1'), covar=tensor([0.0832, 0.0806, 0.0857, 0.1035, 0.1310, 0.1399, 0.0772, 0.4772], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0242, 0.0276, 0.0291, 0.0329, 0.0282, 0.0300, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:08:16,523 INFO [finetune.py:976] (1/7) Epoch 12, batch 5250, loss[loss=0.1757, simple_loss=0.2535, pruned_loss=0.04894, over 4749.00 frames. ], tot_loss[loss=0.1933, simple_loss=0.2595, pruned_loss=0.06358, over 949372.28 frames. ], batch size: 27, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:08:24,221 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6355, 1.2769, 1.7888, 1.9003, 1.5183, 3.3719, 1.1253, 1.3969], + device='cuda:1'), covar=tensor([0.1080, 0.2444, 0.1648, 0.1179, 0.1973, 0.0313, 0.2132, 0.2502], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0075, 0.0078, 0.0092, 0.0081, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 15:09:03,120 INFO [finetune.py:976] (1/7) Epoch 12, batch 5300, loss[loss=0.1135, simple_loss=0.1835, pruned_loss=0.02175, over 4455.00 frames. ], tot_loss[loss=0.1931, simple_loss=0.2597, pruned_loss=0.06328, over 949144.71 frames. ], batch size: 19, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:09:24,994 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68336.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:09:25,579 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68337.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:09:26,706 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.294e+02 1.838e+02 2.123e+02 2.651e+02 4.524e+02, threshold=4.245e+02, percent-clipped=5.0 +2023-03-26 15:09:32,247 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68348.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:09:36,484 INFO [finetune.py:976] (1/7) Epoch 12, batch 5350, loss[loss=0.1763, simple_loss=0.2485, pruned_loss=0.05204, over 4917.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.2592, pruned_loss=0.06254, over 951308.66 frames. ], batch size: 37, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:09:50,619 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3014, 1.6383, 1.6854, 0.8553, 1.6462, 1.8642, 1.9145, 1.5065], + device='cuda:1'), covar=tensor([0.0982, 0.0575, 0.0532, 0.0576, 0.0488, 0.0631, 0.0355, 0.0756], + device='cuda:1'), in_proj_covar=tensor([0.0128, 0.0157, 0.0125, 0.0134, 0.0134, 0.0129, 0.0147, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.4814e-05, 1.1453e-04, 9.0404e-05, 9.6472e-05, 9.4720e-05, 9.3655e-05, + 1.0719e-04, 1.0831e-04], device='cuda:1') +2023-03-26 15:09:55,984 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=68384.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:09:56,636 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.3327, 2.9752, 2.6788, 1.4966, 2.8371, 2.4306, 2.3063, 2.5116], + device='cuda:1'), covar=tensor([0.0748, 0.0848, 0.1619, 0.2181, 0.1517, 0.1849, 0.1863, 0.1177], + device='cuda:1'), in_proj_covar=tensor([0.0165, 0.0197, 0.0198, 0.0184, 0.0212, 0.0206, 0.0221, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:10:04,679 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=68396.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:10:06,566 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68399.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:10:10,285 INFO [finetune.py:976] (1/7) Epoch 12, batch 5400, loss[loss=0.2053, simple_loss=0.265, pruned_loss=0.07275, over 4824.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.258, pruned_loss=0.06252, over 952329.92 frames. ], batch size: 38, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:10:40,850 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.097e+02 1.541e+02 1.801e+02 2.082e+02 4.267e+02, threshold=3.602e+02, percent-clipped=1.0 +2023-03-26 15:10:44,356 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68443.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 15:10:51,598 INFO [finetune.py:976] (1/7) Epoch 12, batch 5450, loss[loss=0.1801, simple_loss=0.2531, pruned_loss=0.05352, over 4824.00 frames. ], tot_loss[loss=0.1895, simple_loss=0.2556, pruned_loss=0.06163, over 953532.40 frames. ], batch size: 41, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:10:54,764 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68460.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:11:24,504 INFO [finetune.py:976] (1/7) Epoch 12, batch 5500, loss[loss=0.1847, simple_loss=0.2587, pruned_loss=0.05536, over 4809.00 frames. ], tot_loss[loss=0.187, simple_loss=0.2527, pruned_loss=0.06061, over 956229.71 frames. ], batch size: 33, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:11:29,583 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-03-26 15:11:47,045 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.070e+02 1.509e+02 1.942e+02 2.407e+02 6.603e+02, threshold=3.884e+02, percent-clipped=3.0 +2023-03-26 15:11:59,915 INFO [finetune.py:976] (1/7) Epoch 12, batch 5550, loss[loss=0.2016, simple_loss=0.2603, pruned_loss=0.07151, over 4874.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2542, pruned_loss=0.06153, over 956215.49 frames. ], batch size: 34, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:12:23,727 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68578.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:12:39,652 INFO [finetune.py:976] (1/7) Epoch 12, batch 5600, loss[loss=0.1671, simple_loss=0.2495, pruned_loss=0.04234, over 4736.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2575, pruned_loss=0.0626, over 954426.01 frames. ], batch size: 59, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:12:58,320 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68637.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:12:59,422 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.238e+02 1.664e+02 1.965e+02 2.319e+02 3.885e+02, threshold=3.931e+02, percent-clipped=1.0 +2023-03-26 15:12:59,531 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68639.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:13:09,174 INFO [finetune.py:976] (1/7) Epoch 12, batch 5650, loss[loss=0.2115, simple_loss=0.2786, pruned_loss=0.07222, over 4896.00 frames. ], tot_loss[loss=0.1941, simple_loss=0.261, pruned_loss=0.06353, over 954661.45 frames. ], batch size: 35, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:13:21,417 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8763, 1.7072, 1.5925, 1.9563, 2.1342, 1.9876, 1.3042, 1.6037], + device='cuda:1'), covar=tensor([0.2110, 0.2023, 0.1977, 0.1668, 0.1576, 0.1099, 0.2575, 0.1873], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0207, 0.0210, 0.0190, 0.0240, 0.0182, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:13:22,612 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4176, 1.6653, 1.7848, 0.9976, 1.6748, 1.8539, 1.9373, 1.5441], + device='cuda:1'), covar=tensor([0.0954, 0.0620, 0.0443, 0.0600, 0.0466, 0.0721, 0.0351, 0.0618], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0156, 0.0125, 0.0133, 0.0133, 0.0129, 0.0147, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.4255e-05, 1.1419e-04, 9.0117e-05, 9.6202e-05, 9.4549e-05, 9.3702e-05, + 1.0676e-04, 1.0790e-04], device='cuda:1') +2023-03-26 15:13:27,857 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=68685.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:13:27,925 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68685.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:13:41,829 INFO [finetune.py:976] (1/7) Epoch 12, batch 5700, loss[loss=0.17, simple_loss=0.2314, pruned_loss=0.05429, over 4386.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2585, pruned_loss=0.06301, over 940267.18 frames. ], batch size: 19, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:13:44,352 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2573, 1.7114, 2.1696, 2.1141, 1.9189, 1.8960, 2.0396, 1.9961], + device='cuda:1'), covar=tensor([0.3605, 0.4125, 0.3659, 0.3617, 0.4979, 0.3800, 0.4742, 0.3342], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0238, 0.0255, 0.0261, 0.0257, 0.0233, 0.0275, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:14:27,874 INFO [finetune.py:976] (1/7) Epoch 13, batch 0, loss[loss=0.2134, simple_loss=0.2829, pruned_loss=0.07196, over 4884.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2829, pruned_loss=0.07196, over 4884.00 frames. ], batch size: 43, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:14:27,875 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 15:14:42,135 INFO [finetune.py:1010] (1/7) Epoch 13, validation: loss=0.1598, simple_loss=0.23, pruned_loss=0.04482, over 2265189.00 frames. +2023-03-26 15:14:42,135 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 15:14:47,267 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.040e+02 1.546e+02 1.915e+02 2.253e+02 4.332e+02, threshold=3.830e+02, percent-clipped=1.0 +2023-03-26 15:14:49,809 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68743.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:14:52,139 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68746.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:14:56,350 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0873, 1.0105, 1.0040, 0.4659, 0.9085, 1.1494, 1.1872, 0.9954], + device='cuda:1'), covar=tensor([0.0973, 0.0538, 0.0497, 0.0565, 0.0489, 0.0597, 0.0358, 0.0632], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0155, 0.0123, 0.0132, 0.0132, 0.0127, 0.0145, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.3373e-05, 1.1311e-04, 8.8848e-05, 9.5292e-05, 9.3380e-05, 9.2446e-05, + 1.0547e-04, 1.0665e-04], device='cuda:1') +2023-03-26 15:14:58,510 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68755.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:15:13,117 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0975, 1.8819, 2.1637, 1.6744, 2.1417, 2.4471, 2.1837, 1.4598], + device='cuda:1'), covar=tensor([0.0654, 0.0897, 0.0778, 0.1056, 0.0709, 0.0597, 0.0720, 0.1671], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0133, 0.0142, 0.0125, 0.0122, 0.0141, 0.0142, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:15:15,978 INFO [finetune.py:976] (1/7) Epoch 13, batch 50, loss[loss=0.1719, simple_loss=0.2407, pruned_loss=0.05154, over 4818.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2643, pruned_loss=0.06538, over 215700.84 frames. ], batch size: 39, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:15:21,852 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=68791.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 15:15:22,505 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5999, 2.3951, 1.8244, 0.9889, 2.0356, 1.9901, 1.9085, 2.1397], + device='cuda:1'), covar=tensor([0.0793, 0.0803, 0.1564, 0.2042, 0.1406, 0.2134, 0.2010, 0.0943], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0195, 0.0196, 0.0184, 0.0211, 0.0205, 0.0220, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:15:27,194 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2153, 1.6472, 2.2080, 2.0853, 1.8978, 1.8752, 2.0579, 2.0044], + device='cuda:1'), covar=tensor([0.3793, 0.4456, 0.3733, 0.3977, 0.5638, 0.4235, 0.5174, 0.3641], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0237, 0.0254, 0.0260, 0.0256, 0.0232, 0.0273, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:15:56,576 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1021, 2.0320, 1.6512, 2.0362, 2.0709, 1.7521, 2.3378, 2.0913], + device='cuda:1'), covar=tensor([0.1250, 0.1983, 0.3109, 0.2520, 0.2391, 0.1727, 0.3193, 0.1856], + device='cuda:1'), in_proj_covar=tensor([0.0177, 0.0186, 0.0231, 0.0253, 0.0240, 0.0198, 0.0211, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:15:57,664 INFO [finetune.py:976] (1/7) Epoch 13, batch 100, loss[loss=0.1779, simple_loss=0.25, pruned_loss=0.05286, over 4908.00 frames. ], tot_loss[loss=0.1911, simple_loss=0.2569, pruned_loss=0.06264, over 379659.00 frames. ], batch size: 37, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:16:02,754 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.134e+02 1.681e+02 1.901e+02 2.429e+02 4.753e+02, threshold=3.802e+02, percent-clipped=2.0 +2023-03-26 15:16:07,215 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 15:16:31,426 INFO [finetune.py:976] (1/7) Epoch 13, batch 150, loss[loss=0.1426, simple_loss=0.2065, pruned_loss=0.03935, over 4773.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2514, pruned_loss=0.0609, over 509613.47 frames. ], batch size: 26, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:16:59,138 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5675, 1.4496, 1.4216, 1.4966, 0.8766, 2.9927, 0.9499, 1.3853], + device='cuda:1'), covar=tensor([0.3458, 0.2490, 0.2187, 0.2354, 0.2128, 0.0258, 0.2752, 0.1443], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0120, 0.0123, 0.0115, 0.0098, 0.0098, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 15:17:05,107 INFO [finetune.py:976] (1/7) Epoch 13, batch 200, loss[loss=0.1784, simple_loss=0.2455, pruned_loss=0.05566, over 4903.00 frames. ], tot_loss[loss=0.187, simple_loss=0.2512, pruned_loss=0.06144, over 609171.26 frames. ], batch size: 37, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:17:05,764 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68934.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:17:09,209 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.222e+02 1.603e+02 1.930e+02 2.189e+02 8.191e+02, threshold=3.861e+02, percent-clipped=2.0 +2023-03-26 15:17:46,315 INFO [finetune.py:976] (1/7) Epoch 13, batch 250, loss[loss=0.1831, simple_loss=0.2608, pruned_loss=0.05272, over 4799.00 frames. ], tot_loss[loss=0.1903, simple_loss=0.2549, pruned_loss=0.0628, over 687248.66 frames. ], batch size: 41, lr: 3.62e-03, grad_scale: 16.0 +2023-03-26 15:18:05,425 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6353, 1.1434, 0.8143, 1.5451, 1.9677, 1.3653, 1.3093, 1.5139], + device='cuda:1'), covar=tensor([0.2003, 0.2974, 0.2517, 0.1632, 0.2479, 0.2567, 0.2148, 0.2731], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0111, 0.0091, 0.0119, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 15:18:19,711 INFO [finetune.py:976] (1/7) Epoch 13, batch 300, loss[loss=0.2013, simple_loss=0.2666, pruned_loss=0.06798, over 4698.00 frames. ], tot_loss[loss=0.1928, simple_loss=0.2587, pruned_loss=0.06345, over 747212.43 frames. ], batch size: 59, lr: 3.62e-03, grad_scale: 32.0 +2023-03-26 15:18:23,316 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.215e+02 1.585e+02 1.877e+02 2.328e+02 4.201e+02, threshold=3.755e+02, percent-clipped=2.0 +2023-03-26 15:18:24,576 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69041.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:18:25,855 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69043.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:18:30,413 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8531, 2.0118, 1.6472, 1.6388, 2.2984, 2.2396, 1.9912, 1.8075], + device='cuda:1'), covar=tensor([0.0375, 0.0345, 0.0534, 0.0350, 0.0350, 0.0684, 0.0340, 0.0445], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0108, 0.0139, 0.0113, 0.0101, 0.0104, 0.0094, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.2202e-05, 8.3653e-05, 1.1029e-04, 8.7830e-05, 7.8874e-05, 7.7218e-05, + 7.1315e-05, 8.3284e-05], device='cuda:1') +2023-03-26 15:18:34,496 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69055.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:18:55,352 INFO [finetune.py:976] (1/7) Epoch 13, batch 350, loss[loss=0.189, simple_loss=0.2646, pruned_loss=0.05666, over 4902.00 frames. ], tot_loss[loss=0.1931, simple_loss=0.2596, pruned_loss=0.06329, over 792282.11 frames. ], batch size: 36, lr: 3.62e-03, grad_scale: 32.0 +2023-03-26 15:19:18,508 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=69103.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:19:19,633 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69104.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:19:41,449 INFO [finetune.py:976] (1/7) Epoch 13, batch 400, loss[loss=0.1872, simple_loss=0.2672, pruned_loss=0.05365, over 4903.00 frames. ], tot_loss[loss=0.193, simple_loss=0.2601, pruned_loss=0.06292, over 829345.97 frames. ], batch size: 37, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:19:50,065 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.154e+02 1.689e+02 1.999e+02 2.345e+02 4.076e+02, threshold=3.998e+02, percent-clipped=3.0 +2023-03-26 15:20:09,243 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-26 15:20:09,757 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69162.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:20:13,410 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69168.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:20:23,370 INFO [finetune.py:976] (1/7) Epoch 13, batch 450, loss[loss=0.1965, simple_loss=0.2568, pruned_loss=0.06809, over 4818.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2587, pruned_loss=0.06202, over 857484.53 frames. ], batch size: 40, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:20:29,474 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8111, 1.7410, 1.5720, 1.8513, 2.2143, 1.9318, 1.4296, 1.4995], + device='cuda:1'), covar=tensor([0.2332, 0.2086, 0.2066, 0.1717, 0.1678, 0.1141, 0.2503, 0.1975], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0206, 0.0209, 0.0190, 0.0239, 0.0181, 0.0212, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:20:30,139 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-26 15:21:04,193 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69223.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:21:05,419 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69225.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:21:07,850 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69229.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:21:10,194 INFO [finetune.py:976] (1/7) Epoch 13, batch 500, loss[loss=0.1958, simple_loss=0.2487, pruned_loss=0.07141, over 4832.00 frames. ], tot_loss[loss=0.1889, simple_loss=0.2559, pruned_loss=0.06095, over 881585.10 frames. ], batch size: 30, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:21:10,905 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69234.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:21:14,297 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.151e+02 1.659e+02 1.928e+02 2.205e+02 4.798e+02, threshold=3.855e+02, percent-clipped=1.0 +2023-03-26 15:21:18,077 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69245.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:21:37,026 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69273.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:21:43,321 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=69282.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:21:43,877 INFO [finetune.py:976] (1/7) Epoch 13, batch 550, loss[loss=0.1621, simple_loss=0.2315, pruned_loss=0.0463, over 4812.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2523, pruned_loss=0.05962, over 899135.58 frames. ], batch size: 25, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:21:45,829 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69286.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:21:48,832 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-26 15:21:56,595 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0428, 1.9861, 1.8734, 2.0343, 1.7364, 3.9413, 1.9003, 2.4289], + device='cuda:1'), covar=tensor([0.3307, 0.2423, 0.2012, 0.2326, 0.1575, 0.0202, 0.2187, 0.1080], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0119, 0.0123, 0.0114, 0.0098, 0.0097, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 15:21:59,421 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69306.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 15:22:08,848 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69320.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:22:17,310 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.01 vs. limit=2.0 +2023-03-26 15:22:17,555 INFO [finetune.py:976] (1/7) Epoch 13, batch 600, loss[loss=0.227, simple_loss=0.2865, pruned_loss=0.08379, over 4823.00 frames. ], tot_loss[loss=0.1869, simple_loss=0.2535, pruned_loss=0.06013, over 909153.82 frames. ], batch size: 33, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:22:18,296 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69334.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 15:22:21,204 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.059e+02 1.536e+02 1.861e+02 2.296e+02 3.946e+02, threshold=3.721e+02, percent-clipped=1.0 +2023-03-26 15:22:22,512 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69341.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:22:50,007 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69368.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:22:58,838 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69381.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:22:59,983 INFO [finetune.py:976] (1/7) Epoch 13, batch 650, loss[loss=0.1891, simple_loss=0.2684, pruned_loss=0.0549, over 4808.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2577, pruned_loss=0.06252, over 917481.98 frames. ], batch size: 39, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:23:03,692 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=69389.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:23:06,745 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69394.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:23:10,353 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69399.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:23:18,900 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8994, 1.9003, 1.7721, 2.1119, 2.2370, 2.1438, 1.5118, 1.6460], + device='cuda:1'), covar=tensor([0.2550, 0.1997, 0.1988, 0.1668, 0.1949, 0.1178, 0.2787, 0.2070], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0207, 0.0210, 0.0190, 0.0241, 0.0182, 0.0213, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:23:30,650 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69429.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:23:33,419 INFO [finetune.py:976] (1/7) Epoch 13, batch 700, loss[loss=0.2267, simple_loss=0.2888, pruned_loss=0.08227, over 4811.00 frames. ], tot_loss[loss=0.1924, simple_loss=0.2592, pruned_loss=0.06284, over 928337.58 frames. ], batch size: 39, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:23:37,540 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.135e+02 1.702e+02 1.957e+02 2.425e+02 4.096e+02, threshold=3.913e+02, percent-clipped=2.0 +2023-03-26 15:23:47,859 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69455.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:23:54,772 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8896, 1.8184, 1.7451, 1.7840, 1.4583, 3.7297, 1.6622, 2.1043], + device='cuda:1'), covar=tensor([0.2775, 0.2075, 0.1745, 0.1983, 0.1443, 0.0212, 0.2534, 0.1180], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0120, 0.0123, 0.0115, 0.0098, 0.0098, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 15:23:58,421 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1193, 3.5431, 3.7320, 3.9485, 3.8705, 3.6035, 4.2116, 1.3850], + device='cuda:1'), covar=tensor([0.0820, 0.0878, 0.0870, 0.1060, 0.1297, 0.1713, 0.0765, 0.5189], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0243, 0.0277, 0.0292, 0.0329, 0.0283, 0.0302, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:24:06,514 INFO [finetune.py:976] (1/7) Epoch 13, batch 750, loss[loss=0.2445, simple_loss=0.312, pruned_loss=0.08854, over 4890.00 frames. ], tot_loss[loss=0.1941, simple_loss=0.2612, pruned_loss=0.06348, over 933979.41 frames. ], batch size: 43, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:24:40,542 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69518.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:24:43,530 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6536, 1.1965, 0.8146, 1.5371, 2.0193, 1.3296, 1.3786, 1.5857], + device='cuda:1'), covar=tensor([0.1702, 0.2576, 0.2681, 0.1450, 0.2228, 0.2843, 0.1936, 0.2297], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0112, 0.0091, 0.0120, 0.0094, 0.0099, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 15:24:44,560 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69524.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:24:44,819 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0 +2023-03-26 15:24:50,486 INFO [finetune.py:976] (1/7) Epoch 13, batch 800, loss[loss=0.1705, simple_loss=0.2353, pruned_loss=0.05288, over 4866.00 frames. ], tot_loss[loss=0.1936, simple_loss=0.2609, pruned_loss=0.06318, over 937548.59 frames. ], batch size: 34, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:24:57,839 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.091e+02 1.694e+02 1.982e+02 2.355e+02 4.334e+02, threshold=3.964e+02, percent-clipped=1.0 +2023-03-26 15:25:08,474 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7876, 1.8008, 1.6409, 2.0584, 2.1124, 1.9842, 1.6425, 1.5085], + device='cuda:1'), covar=tensor([0.2168, 0.1860, 0.1742, 0.1436, 0.1967, 0.1163, 0.2404, 0.1827], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0207, 0.0209, 0.0190, 0.0240, 0.0182, 0.0212, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:25:47,577 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69581.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:25:48,750 INFO [finetune.py:976] (1/7) Epoch 13, batch 850, loss[loss=0.2004, simple_loss=0.2662, pruned_loss=0.06727, over 4863.00 frames. ], tot_loss[loss=0.1904, simple_loss=0.2575, pruned_loss=0.06163, over 940664.57 frames. ], batch size: 34, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:25:52,441 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4319, 1.7033, 1.7865, 0.9435, 1.7483, 1.9904, 2.0006, 1.5267], + device='cuda:1'), covar=tensor([0.0868, 0.0550, 0.0470, 0.0516, 0.0473, 0.0516, 0.0283, 0.0644], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0153, 0.0123, 0.0130, 0.0130, 0.0126, 0.0143, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.2788e-05, 1.1173e-04, 8.8324e-05, 9.3617e-05, 9.2401e-05, 9.1099e-05, + 1.0450e-04, 1.0596e-04], device='cuda:1') +2023-03-26 15:26:03,127 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69601.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:26:06,192 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5960, 2.5082, 2.0218, 2.7938, 2.4431, 2.1266, 3.0697, 2.5758], + device='cuda:1'), covar=tensor([0.1341, 0.2461, 0.3352, 0.2739, 0.2906, 0.1747, 0.3470, 0.1785], + device='cuda:1'), in_proj_covar=tensor([0.0180, 0.0188, 0.0234, 0.0256, 0.0244, 0.0200, 0.0214, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:26:19,784 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6755, 0.7124, 1.7045, 1.5954, 1.5127, 1.4376, 1.5069, 1.6083], + device='cuda:1'), covar=tensor([0.3893, 0.4172, 0.3622, 0.3781, 0.4953, 0.3532, 0.4644, 0.3351], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0238, 0.0255, 0.0261, 0.0257, 0.0233, 0.0275, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:26:21,352 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69629.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:26:24,227 INFO [finetune.py:976] (1/7) Epoch 13, batch 900, loss[loss=0.2133, simple_loss=0.2627, pruned_loss=0.08195, over 4234.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2544, pruned_loss=0.06064, over 945432.67 frames. ], batch size: 65, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:26:27,889 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.145e+02 1.604e+02 1.856e+02 2.224e+02 3.601e+02, threshold=3.711e+02, percent-clipped=0.0 +2023-03-26 15:26:55,509 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69674.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:26:56,706 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69676.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:27:06,462 INFO [finetune.py:976] (1/7) Epoch 13, batch 950, loss[loss=0.1896, simple_loss=0.2605, pruned_loss=0.05932, over 4822.00 frames. ], tot_loss[loss=0.188, simple_loss=0.2542, pruned_loss=0.06092, over 947228.43 frames. ], batch size: 39, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:27:29,366 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69699.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:28:02,967 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69724.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:28:08,358 INFO [finetune.py:976] (1/7) Epoch 13, batch 1000, loss[loss=0.1591, simple_loss=0.2211, pruned_loss=0.04853, over 4718.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2569, pruned_loss=0.06259, over 948244.95 frames. ], batch size: 23, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:28:10,721 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69735.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:28:13,002 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.598e+02 1.856e+02 2.406e+02 4.029e+02, threshold=3.712e+02, percent-clipped=2.0 +2023-03-26 15:28:17,951 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=69747.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:28:20,309 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69750.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:28:21,482 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3906, 1.2568, 1.2744, 0.8487, 1.1956, 1.4230, 1.4597, 1.1707], + device='cuda:1'), covar=tensor([0.0751, 0.0486, 0.0450, 0.0434, 0.0442, 0.0525, 0.0292, 0.0523], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0153, 0.0123, 0.0130, 0.0130, 0.0126, 0.0144, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.3083e-05, 1.1158e-04, 8.8272e-05, 9.3607e-05, 9.2106e-05, 9.1281e-05, + 1.0479e-04, 1.0585e-04], device='cuda:1') +2023-03-26 15:28:26,978 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0034, 1.3288, 0.8181, 1.9167, 2.2430, 1.6136, 1.7620, 1.5769], + device='cuda:1'), covar=tensor([0.1420, 0.2076, 0.2229, 0.1163, 0.1942, 0.2030, 0.1374, 0.2098], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0096, 0.0113, 0.0092, 0.0121, 0.0094, 0.0099, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 15:28:52,947 INFO [finetune.py:976] (1/7) Epoch 13, batch 1050, loss[loss=0.196, simple_loss=0.2588, pruned_loss=0.0666, over 4775.00 frames. ], tot_loss[loss=0.1935, simple_loss=0.2606, pruned_loss=0.06324, over 950324.36 frames. ], batch size: 26, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:29:03,453 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8060, 1.4251, 1.9146, 1.7559, 1.5735, 1.5101, 1.7160, 1.6790], + device='cuda:1'), covar=tensor([0.3871, 0.4013, 0.3394, 0.3706, 0.4870, 0.3856, 0.4603, 0.3170], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0237, 0.0254, 0.0261, 0.0257, 0.0232, 0.0274, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:29:05,248 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0151, 2.0100, 1.7454, 2.1153, 2.5823, 2.1366, 1.7514, 1.6138], + device='cuda:1'), covar=tensor([0.2094, 0.1811, 0.1819, 0.1603, 0.1527, 0.1068, 0.2295, 0.1991], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0206, 0.0210, 0.0189, 0.0239, 0.0182, 0.0212, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:29:38,099 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69818.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:29:48,096 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69824.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:29:59,171 INFO [finetune.py:976] (1/7) Epoch 13, batch 1100, loss[loss=0.1623, simple_loss=0.2286, pruned_loss=0.04799, over 4721.00 frames. ], tot_loss[loss=0.1948, simple_loss=0.2617, pruned_loss=0.06394, over 951586.01 frames. ], batch size: 23, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:30:02,885 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.063e+02 1.609e+02 1.898e+02 2.282e+02 6.010e+02, threshold=3.795e+02, percent-clipped=2.0 +2023-03-26 15:30:35,884 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=69866.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:30:43,312 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=69872.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:30:51,921 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69881.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:30:53,489 INFO [finetune.py:976] (1/7) Epoch 13, batch 1150, loss[loss=0.2258, simple_loss=0.283, pruned_loss=0.08437, over 4890.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2621, pruned_loss=0.0642, over 951530.06 frames. ], batch size: 32, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:31:12,298 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69901.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:31:42,210 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=69929.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:31:42,238 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69929.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 15:31:44,609 INFO [finetune.py:976] (1/7) Epoch 13, batch 1200, loss[loss=0.1419, simple_loss=0.2171, pruned_loss=0.03335, over 4839.00 frames. ], tot_loss[loss=0.1944, simple_loss=0.2614, pruned_loss=0.06366, over 949855.84 frames. ], batch size: 49, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:31:48,756 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.031e+02 1.603e+02 1.893e+02 2.321e+02 3.158e+02, threshold=3.786e+02, percent-clipped=0.0 +2023-03-26 15:31:55,835 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=69949.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 15:32:13,204 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69976.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:32:13,759 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=69977.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:32:17,837 INFO [finetune.py:976] (1/7) Epoch 13, batch 1250, loss[loss=0.1832, simple_loss=0.245, pruned_loss=0.06073, over 4893.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2579, pruned_loss=0.06242, over 951317.38 frames. ], batch size: 35, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:32:46,509 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=70024.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:32:46,536 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70024.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:32:50,169 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=70030.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:32:52,397 INFO [finetune.py:976] (1/7) Epoch 13, batch 1300, loss[loss=0.1511, simple_loss=0.2169, pruned_loss=0.04269, over 4820.00 frames. ], tot_loss[loss=0.1882, simple_loss=0.2545, pruned_loss=0.06097, over 951307.23 frames. ], batch size: 25, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:32:56,053 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.095e+02 1.649e+02 1.897e+02 2.309e+02 4.234e+02, threshold=3.795e+02, percent-clipped=2.0 +2023-03-26 15:33:03,807 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70050.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:33:19,134 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=70072.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:33:24,096 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9424, 1.7323, 2.2764, 1.6326, 2.0996, 2.3739, 1.7578, 2.3706], + device='cuda:1'), covar=tensor([0.1536, 0.2129, 0.1720, 0.2154, 0.0915, 0.1497, 0.2696, 0.0998], + device='cuda:1'), in_proj_covar=tensor([0.0196, 0.0205, 0.0195, 0.0193, 0.0178, 0.0214, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:33:25,839 INFO [finetune.py:976] (1/7) Epoch 13, batch 1350, loss[loss=0.2119, simple_loss=0.2779, pruned_loss=0.07298, over 4899.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2545, pruned_loss=0.0612, over 954507.42 frames. ], batch size: 35, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:33:36,430 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=70098.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:33:53,176 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70110.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:34:08,077 INFO [finetune.py:976] (1/7) Epoch 13, batch 1400, loss[loss=0.2064, simple_loss=0.2702, pruned_loss=0.07128, over 4764.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2575, pruned_loss=0.06224, over 955224.66 frames. ], batch size: 27, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:34:12,153 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.202e+02 1.588e+02 1.939e+02 2.393e+02 8.943e+02, threshold=3.877e+02, percent-clipped=1.0 +2023-03-26 15:34:18,797 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2504, 1.2516, 1.5548, 1.1298, 1.2335, 1.4966, 1.2732, 1.6634], + device='cuda:1'), covar=tensor([0.1330, 0.2021, 0.1228, 0.1484, 0.0888, 0.1335, 0.2721, 0.0761], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0206, 0.0196, 0.0194, 0.0179, 0.0216, 0.0218, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:34:34,228 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70171.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:34:34,395 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-26 15:34:41,771 INFO [finetune.py:976] (1/7) Epoch 13, batch 1450, loss[loss=0.1791, simple_loss=0.2488, pruned_loss=0.05472, over 4741.00 frames. ], tot_loss[loss=0.1931, simple_loss=0.2607, pruned_loss=0.06281, over 955990.32 frames. ], batch size: 27, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:35:02,484 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-26 15:35:05,315 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4825, 3.8631, 4.1049, 4.3377, 4.2086, 3.9201, 4.5630, 1.3625], + device='cuda:1'), covar=tensor([0.0787, 0.0824, 0.0835, 0.0933, 0.1334, 0.1802, 0.0736, 0.5899], + device='cuda:1'), in_proj_covar=tensor([0.0354, 0.0245, 0.0278, 0.0293, 0.0332, 0.0284, 0.0304, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:35:13,599 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-26 15:35:26,431 INFO [finetune.py:976] (1/7) Epoch 13, batch 1500, loss[loss=0.2214, simple_loss=0.2826, pruned_loss=0.08015, over 4890.00 frames. ], tot_loss[loss=0.1953, simple_loss=0.2626, pruned_loss=0.06401, over 955035.78 frames. ], batch size: 35, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:35:29,033 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2435, 1.2062, 1.6258, 1.1537, 1.2817, 1.5084, 1.2311, 1.6386], + device='cuda:1'), covar=tensor([0.1235, 0.2056, 0.1069, 0.1252, 0.0916, 0.0974, 0.2741, 0.0739], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0205, 0.0195, 0.0193, 0.0178, 0.0214, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:35:30,134 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.195e+02 1.613e+02 1.899e+02 2.364e+02 4.350e+02, threshold=3.798e+02, percent-clipped=1.0 +2023-03-26 15:35:43,991 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-03-26 15:35:46,877 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70260.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:36:10,535 INFO [finetune.py:976] (1/7) Epoch 13, batch 1550, loss[loss=0.1896, simple_loss=0.2551, pruned_loss=0.06201, over 4905.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2624, pruned_loss=0.06375, over 956023.43 frames. ], batch size: 36, lr: 3.61e-03, grad_scale: 32.0 +2023-03-26 15:36:20,268 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8205, 1.8627, 1.6583, 1.9126, 2.2799, 1.9918, 1.6328, 1.4895], + device='cuda:1'), covar=tensor([0.2031, 0.1781, 0.1764, 0.1507, 0.1654, 0.1094, 0.2247, 0.1764], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0206, 0.0209, 0.0189, 0.0239, 0.0182, 0.0212, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:36:31,366 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 15:36:48,580 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.50 vs. limit=2.0 +2023-03-26 15:36:49,628 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70321.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:36:58,935 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70330.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 15:37:00,629 INFO [finetune.py:976] (1/7) Epoch 13, batch 1600, loss[loss=0.2079, simple_loss=0.2645, pruned_loss=0.07571, over 4874.00 frames. ], tot_loss[loss=0.1937, simple_loss=0.2608, pruned_loss=0.06337, over 956709.82 frames. ], batch size: 34, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:37:04,737 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.057e+02 1.529e+02 1.873e+02 2.318e+02 5.550e+02, threshold=3.745e+02, percent-clipped=4.0 +2023-03-26 15:37:29,068 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1934, 2.3205, 2.0743, 2.3269, 2.6994, 2.3087, 2.2162, 1.7043], + device='cuda:1'), covar=tensor([0.2191, 0.1841, 0.1793, 0.1600, 0.1828, 0.1071, 0.2048, 0.1927], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0206, 0.0208, 0.0189, 0.0238, 0.0182, 0.0211, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:37:30,810 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=70378.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 15:37:34,182 INFO [finetune.py:976] (1/7) Epoch 13, batch 1650, loss[loss=0.1813, simple_loss=0.254, pruned_loss=0.0543, over 4758.00 frames. ], tot_loss[loss=0.1925, simple_loss=0.2587, pruned_loss=0.06313, over 956009.32 frames. ], batch size: 27, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:37:37,312 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9359, 1.4791, 1.9839, 1.8891, 1.6961, 1.6717, 1.8389, 1.8105], + device='cuda:1'), covar=tensor([0.4031, 0.4242, 0.3405, 0.3735, 0.4854, 0.3729, 0.4641, 0.3345], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0237, 0.0255, 0.0261, 0.0258, 0.0233, 0.0275, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:37:43,255 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8122, 1.5571, 1.9648, 1.3229, 1.6938, 1.9445, 1.5011, 2.0977], + device='cuda:1'), covar=tensor([0.1259, 0.2079, 0.1256, 0.1770, 0.0953, 0.1196, 0.2986, 0.0854], + device='cuda:1'), in_proj_covar=tensor([0.0196, 0.0205, 0.0195, 0.0193, 0.0179, 0.0214, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:38:08,087 INFO [finetune.py:976] (1/7) Epoch 13, batch 1700, loss[loss=0.1859, simple_loss=0.2429, pruned_loss=0.06438, over 4701.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2549, pruned_loss=0.06161, over 956282.92 frames. ], batch size: 59, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:38:11,732 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.265e+02 1.610e+02 1.926e+02 2.276e+02 4.227e+02, threshold=3.852e+02, percent-clipped=1.0 +2023-03-26 15:38:30,201 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=70466.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:38:41,456 INFO [finetune.py:976] (1/7) Epoch 13, batch 1750, loss[loss=0.2094, simple_loss=0.2715, pruned_loss=0.07368, over 4912.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2549, pruned_loss=0.06101, over 953718.47 frames. ], batch size: 36, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:39:24,239 INFO [finetune.py:976] (1/7) Epoch 13, batch 1800, loss[loss=0.1944, simple_loss=0.2615, pruned_loss=0.06362, over 4748.00 frames. ], tot_loss[loss=0.191, simple_loss=0.258, pruned_loss=0.06199, over 953038.09 frames. ], batch size: 54, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:39:28,345 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.316e+01 1.597e+02 2.051e+02 2.548e+02 3.844e+02, threshold=4.101e+02, percent-clipped=0.0 +2023-03-26 15:39:52,779 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3128, 1.2912, 1.7260, 2.4205, 1.6843, 2.0829, 1.0215, 2.0622], + device='cuda:1'), covar=tensor([0.1705, 0.1454, 0.1076, 0.0642, 0.0912, 0.1321, 0.1521, 0.0619], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0118, 0.0136, 0.0167, 0.0102, 0.0140, 0.0128, 0.0103], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 15:39:54,652 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5866, 1.4486, 1.3301, 1.5787, 1.5993, 1.5804, 0.9993, 1.3564], + device='cuda:1'), covar=tensor([0.2260, 0.2205, 0.2062, 0.1739, 0.1695, 0.1279, 0.2676, 0.2046], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0205, 0.0207, 0.0188, 0.0237, 0.0181, 0.0210, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:39:58,058 INFO [finetune.py:976] (1/7) Epoch 13, batch 1850, loss[loss=0.1786, simple_loss=0.246, pruned_loss=0.05555, over 4818.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2595, pruned_loss=0.06256, over 953185.54 frames. ], batch size: 40, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:40:26,914 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=70616.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:40:42,677 INFO [finetune.py:976] (1/7) Epoch 13, batch 1900, loss[loss=0.1375, simple_loss=0.2056, pruned_loss=0.03476, over 4749.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2594, pruned_loss=0.06169, over 955343.90 frames. ], batch size: 23, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:40:46,778 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.033e+02 1.570e+02 1.884e+02 2.217e+02 6.026e+02, threshold=3.769e+02, percent-clipped=2.0 +2023-03-26 15:40:54,169 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1462, 1.2206, 1.5847, 1.0321, 1.2311, 1.4361, 1.2377, 1.6182], + device='cuda:1'), covar=tensor([0.1203, 0.1988, 0.1109, 0.1429, 0.0751, 0.1071, 0.2485, 0.0693], + device='cuda:1'), in_proj_covar=tensor([0.0195, 0.0204, 0.0193, 0.0191, 0.0177, 0.0213, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:41:27,334 INFO [finetune.py:976] (1/7) Epoch 13, batch 1950, loss[loss=0.1488, simple_loss=0.2236, pruned_loss=0.03697, over 4800.00 frames. ], tot_loss[loss=0.19, simple_loss=0.2583, pruned_loss=0.06089, over 956555.47 frames. ], batch size: 25, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:41:47,164 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9653, 4.2641, 4.4828, 4.7087, 4.6780, 4.4184, 5.0383, 1.5396], + device='cuda:1'), covar=tensor([0.0738, 0.0861, 0.0791, 0.0956, 0.1150, 0.1513, 0.0586, 0.5751], + device='cuda:1'), in_proj_covar=tensor([0.0354, 0.0245, 0.0278, 0.0293, 0.0332, 0.0284, 0.0304, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:42:06,899 INFO [finetune.py:976] (1/7) Epoch 13, batch 2000, loss[loss=0.1884, simple_loss=0.2552, pruned_loss=0.06078, over 4909.00 frames. ], tot_loss[loss=0.1889, simple_loss=0.2563, pruned_loss=0.06079, over 957196.40 frames. ], batch size: 46, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:42:15,812 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.177e+02 1.535e+02 1.807e+02 2.194e+02 3.140e+02, threshold=3.615e+02, percent-clipped=0.0 +2023-03-26 15:42:36,877 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70766.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:42:48,489 INFO [finetune.py:976] (1/7) Epoch 13, batch 2050, loss[loss=0.1524, simple_loss=0.2185, pruned_loss=0.0431, over 4868.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2535, pruned_loss=0.06001, over 958089.81 frames. ], batch size: 31, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:42:53,941 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-26 15:43:07,660 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3312, 1.9255, 2.3518, 2.1879, 1.9409, 1.9527, 2.1889, 2.0621], + device='cuda:1'), covar=tensor([0.4248, 0.4685, 0.3572, 0.4625, 0.5612, 0.4200, 0.5285, 0.3610], + device='cuda:1'), in_proj_covar=tensor([0.0241, 0.0238, 0.0256, 0.0263, 0.0260, 0.0234, 0.0276, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:43:09,365 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=70814.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:43:22,315 INFO [finetune.py:976] (1/7) Epoch 13, batch 2100, loss[loss=0.1865, simple_loss=0.2524, pruned_loss=0.0603, over 4246.00 frames. ], tot_loss[loss=0.1875, simple_loss=0.2538, pruned_loss=0.0606, over 956402.78 frames. ], batch size: 65, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:43:26,462 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.827e+01 1.609e+02 1.892e+02 2.240e+02 3.187e+02, threshold=3.783e+02, percent-clipped=0.0 +2023-03-26 15:43:28,933 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9791, 1.8995, 1.6847, 2.1169, 2.4149, 2.0914, 1.5726, 1.6201], + device='cuda:1'), covar=tensor([0.2116, 0.1985, 0.1909, 0.1536, 0.1619, 0.1150, 0.2448, 0.1859], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0206, 0.0208, 0.0189, 0.0238, 0.0181, 0.0212, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:43:35,028 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6426, 1.4632, 1.1241, 0.3645, 1.4069, 1.4707, 1.3580, 1.4142], + device='cuda:1'), covar=tensor([0.0872, 0.0722, 0.1076, 0.1652, 0.1192, 0.2046, 0.1976, 0.0726], + device='cuda:1'), in_proj_covar=tensor([0.0165, 0.0195, 0.0199, 0.0185, 0.0212, 0.0205, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:43:56,100 INFO [finetune.py:976] (1/7) Epoch 13, batch 2150, loss[loss=0.2033, simple_loss=0.2755, pruned_loss=0.06554, over 4869.00 frames. ], tot_loss[loss=0.1896, simple_loss=0.2567, pruned_loss=0.06127, over 956020.30 frames. ], batch size: 34, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:44:13,699 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-03-26 15:44:27,518 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.62 vs. limit=5.0 +2023-03-26 15:44:35,060 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70916.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:44:38,206 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0 +2023-03-26 15:44:46,790 INFO [finetune.py:976] (1/7) Epoch 13, batch 2200, loss[loss=0.176, simple_loss=0.2448, pruned_loss=0.05359, over 4744.00 frames. ], tot_loss[loss=0.1906, simple_loss=0.2582, pruned_loss=0.06152, over 955305.89 frames. ], batch size: 27, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:44:50,484 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.006e+02 1.701e+02 1.958e+02 2.316e+02 4.574e+02, threshold=3.916e+02, percent-clipped=1.0 +2023-03-26 15:45:07,656 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=70964.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:45:19,185 INFO [finetune.py:976] (1/7) Epoch 13, batch 2250, loss[loss=0.2361, simple_loss=0.3014, pruned_loss=0.08537, over 4762.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2592, pruned_loss=0.06185, over 953523.81 frames. ], batch size: 54, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:45:26,620 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70992.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:46:02,059 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=71030.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:46:03,709 INFO [finetune.py:976] (1/7) Epoch 13, batch 2300, loss[loss=0.2306, simple_loss=0.2788, pruned_loss=0.09124, over 4033.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.26, pruned_loss=0.06218, over 955009.35 frames. ], batch size: 17, lr: 3.60e-03, grad_scale: 64.0 +2023-03-26 15:46:08,249 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.237e+01 1.685e+02 2.000e+02 2.324e+02 3.629e+02, threshold=3.999e+02, percent-clipped=0.0 +2023-03-26 15:46:23,745 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71053.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:46:49,201 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.9877, 3.4272, 3.5967, 3.8436, 3.7243, 3.5216, 4.0466, 1.3659], + device='cuda:1'), covar=tensor([0.0806, 0.0871, 0.0828, 0.0986, 0.1293, 0.1536, 0.0777, 0.5170], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0277, 0.0292, 0.0331, 0.0283, 0.0303, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:46:59,589 INFO [finetune.py:976] (1/7) Epoch 13, batch 2350, loss[loss=0.1696, simple_loss=0.2393, pruned_loss=0.04995, over 4828.00 frames. ], tot_loss[loss=0.1908, simple_loss=0.2582, pruned_loss=0.06173, over 956427.02 frames. ], batch size: 49, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:47:10,245 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71091.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:47:37,772 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 15:47:42,748 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 15:47:48,028 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8620, 1.6911, 2.3703, 1.6317, 2.0682, 2.2213, 1.6801, 2.3772], + device='cuda:1'), covar=tensor([0.1292, 0.1821, 0.1172, 0.1762, 0.0754, 0.1152, 0.2419, 0.0697], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0206, 0.0195, 0.0192, 0.0179, 0.0215, 0.0218, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:48:00,798 INFO [finetune.py:976] (1/7) Epoch 13, batch 2400, loss[loss=0.1626, simple_loss=0.2171, pruned_loss=0.05406, over 4872.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2553, pruned_loss=0.06095, over 954844.78 frames. ], batch size: 34, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:48:09,284 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.076e+01 1.502e+02 1.791e+02 2.104e+02 3.987e+02, threshold=3.583e+02, percent-clipped=0.0 +2023-03-26 15:48:52,070 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1950, 1.2831, 1.3729, 0.6534, 1.2191, 1.5692, 1.5233, 1.3117], + device='cuda:1'), covar=tensor([0.0978, 0.0620, 0.0487, 0.0504, 0.0472, 0.0545, 0.0325, 0.0669], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0151, 0.0122, 0.0129, 0.0130, 0.0125, 0.0142, 0.0144], + device='cuda:1'), out_proj_covar=tensor([9.2336e-05, 1.1032e-04, 8.7560e-05, 9.2980e-05, 9.2196e-05, 9.0872e-05, + 1.0334e-04, 1.0480e-04], device='cuda:1') +2023-03-26 15:49:05,622 INFO [finetune.py:976] (1/7) Epoch 13, batch 2450, loss[loss=0.2061, simple_loss=0.2658, pruned_loss=0.07322, over 4831.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2518, pruned_loss=0.05918, over 956764.46 frames. ], batch size: 39, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:50:04,538 INFO [finetune.py:976] (1/7) Epoch 13, batch 2500, loss[loss=0.1981, simple_loss=0.2687, pruned_loss=0.06376, over 4819.00 frames. ], tot_loss[loss=0.1862, simple_loss=0.2529, pruned_loss=0.0598, over 956846.59 frames. ], batch size: 40, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:50:08,818 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.134e+02 1.629e+02 1.890e+02 2.415e+02 4.682e+02, threshold=3.780e+02, percent-clipped=4.0 +2023-03-26 15:50:41,417 INFO [finetune.py:976] (1/7) Epoch 13, batch 2550, loss[loss=0.1593, simple_loss=0.2407, pruned_loss=0.03895, over 4779.00 frames. ], tot_loss[loss=0.1897, simple_loss=0.2568, pruned_loss=0.06129, over 956280.67 frames. ], batch size: 29, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:51:02,608 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 15:51:22,574 INFO [finetune.py:976] (1/7) Epoch 13, batch 2600, loss[loss=0.2204, simple_loss=0.2827, pruned_loss=0.07907, over 4802.00 frames. ], tot_loss[loss=0.1921, simple_loss=0.2589, pruned_loss=0.06267, over 956134.07 frames. ], batch size: 40, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:51:24,596 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9135, 1.8219, 1.7366, 2.1070, 2.5191, 2.1278, 1.6895, 1.5607], + device='cuda:1'), covar=tensor([0.2353, 0.2100, 0.1995, 0.1755, 0.1673, 0.1122, 0.2352, 0.2020], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0208, 0.0210, 0.0190, 0.0240, 0.0183, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:51:26,381 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5185, 1.3653, 1.3769, 1.4371, 0.9510, 3.1773, 1.2060, 1.5644], + device='cuda:1'), covar=tensor([0.3455, 0.2468, 0.2255, 0.2442, 0.2125, 0.0208, 0.2905, 0.1426], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0121, 0.0124, 0.0115, 0.0098, 0.0098, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 15:51:26,870 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.157e+02 1.678e+02 1.922e+02 2.428e+02 5.321e+02, threshold=3.843e+02, percent-clipped=3.0 +2023-03-26 15:51:26,988 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2639, 1.3579, 1.4991, 1.5049, 1.4497, 2.6296, 1.1966, 1.4538], + device='cuda:1'), covar=tensor([0.0802, 0.1300, 0.0844, 0.0710, 0.1211, 0.0387, 0.1157, 0.1293], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0075, 0.0078, 0.0093, 0.0082, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 15:51:31,778 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71348.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:51:42,225 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5326, 1.4015, 1.3061, 1.4409, 1.6806, 1.6235, 1.4883, 1.2636], + device='cuda:1'), covar=tensor([0.0327, 0.0263, 0.0422, 0.0243, 0.0226, 0.0325, 0.0253, 0.0309], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0110, 0.0141, 0.0113, 0.0102, 0.0106, 0.0096, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.3445e-05, 8.5145e-05, 1.1165e-04, 8.8115e-05, 7.9788e-05, 7.8314e-05, + 7.2229e-05, 8.3939e-05], device='cuda:1') +2023-03-26 15:51:43,483 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.64 vs. limit=2.0 +2023-03-26 15:51:55,370 INFO [finetune.py:976] (1/7) Epoch 13, batch 2650, loss[loss=0.2235, simple_loss=0.29, pruned_loss=0.07853, over 4922.00 frames. ], tot_loss[loss=0.1926, simple_loss=0.2599, pruned_loss=0.0626, over 954692.09 frames. ], batch size: 42, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:51:58,310 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71386.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:52:24,639 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-26 15:52:29,325 INFO [finetune.py:976] (1/7) Epoch 13, batch 2700, loss[loss=0.1777, simple_loss=0.2472, pruned_loss=0.05406, over 4899.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2593, pruned_loss=0.06196, over 956314.97 frames. ], batch size: 37, lr: 3.60e-03, grad_scale: 32.0 +2023-03-26 15:52:34,539 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.057e+02 1.578e+02 1.884e+02 2.307e+02 4.300e+02, threshold=3.769e+02, percent-clipped=2.0 +2023-03-26 15:52:35,316 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1635, 1.9302, 1.7369, 1.9293, 1.8765, 1.8511, 1.9235, 2.6685], + device='cuda:1'), covar=tensor([0.4183, 0.4674, 0.3570, 0.4443, 0.4312, 0.2535, 0.4190, 0.1781], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0260, 0.0223, 0.0278, 0.0246, 0.0212, 0.0247, 0.0221], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:53:02,922 INFO [finetune.py:976] (1/7) Epoch 13, batch 2750, loss[loss=0.1929, simple_loss=0.2545, pruned_loss=0.0656, over 4771.00 frames. ], tot_loss[loss=0.1896, simple_loss=0.2567, pruned_loss=0.06128, over 957065.74 frames. ], batch size: 28, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:53:34,354 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.77 vs. limit=2.0 +2023-03-26 15:53:36,642 INFO [finetune.py:976] (1/7) Epoch 13, batch 2800, loss[loss=0.1837, simple_loss=0.2509, pruned_loss=0.05822, over 4873.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2528, pruned_loss=0.06004, over 956170.67 frames. ], batch size: 34, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:53:40,881 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.138e+02 1.564e+02 1.863e+02 2.304e+02 3.302e+02, threshold=3.726e+02, percent-clipped=0.0 +2023-03-26 15:53:45,022 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2584, 2.0655, 1.9166, 2.4319, 2.5425, 2.2396, 2.1220, 1.6599], + device='cuda:1'), covar=tensor([0.2191, 0.2046, 0.1894, 0.1571, 0.1921, 0.1082, 0.2022, 0.1919], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0207, 0.0210, 0.0190, 0.0240, 0.0183, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:54:23,056 INFO [finetune.py:976] (1/7) Epoch 13, batch 2850, loss[loss=0.2505, simple_loss=0.3134, pruned_loss=0.09379, over 4896.00 frames. ], tot_loss[loss=0.1861, simple_loss=0.2521, pruned_loss=0.0601, over 956544.06 frames. ], batch size: 37, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:54:52,360 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=71616.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:55:03,555 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 15:55:06,323 INFO [finetune.py:976] (1/7) Epoch 13, batch 2900, loss[loss=0.1259, simple_loss=0.1921, pruned_loss=0.02979, over 3999.00 frames. ], tot_loss[loss=0.187, simple_loss=0.2534, pruned_loss=0.06032, over 953427.23 frames. ], batch size: 17, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:55:15,495 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.635e+01 1.661e+02 1.944e+02 2.530e+02 6.475e+02, threshold=3.888e+02, percent-clipped=5.0 +2023-03-26 15:55:24,622 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=71648.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:55:37,978 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7046, 1.6619, 1.4636, 1.7947, 2.1729, 1.8198, 1.5305, 1.3705], + device='cuda:1'), covar=tensor([0.2186, 0.2081, 0.1952, 0.1676, 0.1746, 0.1250, 0.2375, 0.1995], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0208, 0.0210, 0.0190, 0.0241, 0.0183, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:55:49,609 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71677.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:55:58,582 INFO [finetune.py:976] (1/7) Epoch 13, batch 2950, loss[loss=0.1967, simple_loss=0.265, pruned_loss=0.06423, over 4795.00 frames. ], tot_loss[loss=0.1887, simple_loss=0.2565, pruned_loss=0.06045, over 953736.76 frames. ], batch size: 54, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:56:00,489 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=71686.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:56:09,944 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4227, 1.9310, 2.7870, 1.5747, 2.5405, 2.6727, 1.9632, 2.8304], + device='cuda:1'), covar=tensor([0.1324, 0.2215, 0.1590, 0.2429, 0.0860, 0.1640, 0.2684, 0.0826], + device='cuda:1'), in_proj_covar=tensor([0.0197, 0.0207, 0.0196, 0.0193, 0.0180, 0.0217, 0.0219, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:56:11,115 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=71696.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:56:39,793 INFO [finetune.py:976] (1/7) Epoch 13, batch 3000, loss[loss=0.1716, simple_loss=0.2431, pruned_loss=0.05003, over 4762.00 frames. ], tot_loss[loss=0.1903, simple_loss=0.2578, pruned_loss=0.06142, over 953839.00 frames. ], batch size: 28, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:56:39,793 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 15:56:46,169 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6287, 1.5582, 1.5676, 1.5858, 0.9980, 2.9710, 1.1365, 1.6825], + device='cuda:1'), covar=tensor([0.3372, 0.2389, 0.2015, 0.2419, 0.1929, 0.0261, 0.2669, 0.1212], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0121, 0.0124, 0.0116, 0.0099, 0.0098, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 15:56:48,194 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9748, 1.1526, 2.0227, 1.8482, 1.7690, 1.6454, 1.7675, 1.8118], + device='cuda:1'), covar=tensor([0.3854, 0.4355, 0.3915, 0.3906, 0.5485, 0.3993, 0.4866, 0.3657], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0240, 0.0256, 0.0265, 0.0262, 0.0236, 0.0277, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:56:50,410 INFO [finetune.py:1010] (1/7) Epoch 13, validation: loss=0.1572, simple_loss=0.2278, pruned_loss=0.04333, over 2265189.00 frames. +2023-03-26 15:56:50,411 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 15:56:51,093 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=71734.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:56:55,666 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.204e+02 1.624e+02 1.953e+02 2.376e+02 4.887e+02, threshold=3.907e+02, percent-clipped=1.0 +2023-03-26 15:57:00,774 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 15:57:15,026 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1908, 1.3652, 1.4232, 0.8048, 1.3295, 1.5698, 1.6215, 1.3142], + device='cuda:1'), covar=tensor([0.0914, 0.0521, 0.0447, 0.0475, 0.0451, 0.0587, 0.0315, 0.0640], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0152, 0.0121, 0.0129, 0.0130, 0.0126, 0.0141, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.2790e-05, 1.1073e-04, 8.7231e-05, 9.2568e-05, 9.1921e-05, 9.1667e-05, + 1.0295e-04, 1.0495e-04], device='cuda:1') +2023-03-26 15:57:22,729 INFO [finetune.py:976] (1/7) Epoch 13, batch 3050, loss[loss=0.1971, simple_loss=0.2685, pruned_loss=0.06279, over 4859.00 frames. ], tot_loss[loss=0.1918, simple_loss=0.2593, pruned_loss=0.06213, over 954480.31 frames. ], batch size: 31, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:57:55,478 INFO [finetune.py:976] (1/7) Epoch 13, batch 3100, loss[loss=0.2154, simple_loss=0.2704, pruned_loss=0.0802, over 4740.00 frames. ], tot_loss[loss=0.1898, simple_loss=0.257, pruned_loss=0.06126, over 955301.89 frames. ], batch size: 59, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:58:01,083 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.937e+01 1.560e+02 1.843e+02 2.215e+02 5.565e+02, threshold=3.687e+02, percent-clipped=1.0 +2023-03-26 15:58:11,157 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.6747, 4.0216, 4.2486, 4.4652, 4.4205, 4.1265, 4.7435, 1.4712], + device='cuda:1'), covar=tensor([0.0703, 0.0751, 0.0829, 0.0838, 0.1142, 0.1607, 0.0603, 0.5533], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0244, 0.0276, 0.0291, 0.0331, 0.0282, 0.0303, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:58:29,153 INFO [finetune.py:976] (1/7) Epoch 13, batch 3150, loss[loss=0.1475, simple_loss=0.211, pruned_loss=0.04202, over 4752.00 frames. ], tot_loss[loss=0.1878, simple_loss=0.2546, pruned_loss=0.06052, over 955817.70 frames. ], batch size: 23, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:58:43,977 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1792, 2.1333, 1.8897, 2.3240, 2.0987, 2.0567, 2.0480, 2.9091], + device='cuda:1'), covar=tensor([0.3889, 0.4930, 0.3401, 0.4484, 0.4666, 0.2401, 0.4794, 0.1592], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0260, 0.0224, 0.0278, 0.0246, 0.0212, 0.0247, 0.0221], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 15:58:53,741 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=71919.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:59:03,051 INFO [finetune.py:976] (1/7) Epoch 13, batch 3200, loss[loss=0.1302, simple_loss=0.2116, pruned_loss=0.02443, over 4907.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2516, pruned_loss=0.05948, over 955916.51 frames. ], batch size: 35, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 15:59:07,310 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.138e+02 1.561e+02 1.912e+02 2.265e+02 3.518e+02, threshold=3.824e+02, percent-clipped=0.0 +2023-03-26 15:59:40,761 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71972.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:59:54,381 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71980.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 15:59:56,073 INFO [finetune.py:976] (1/7) Epoch 13, batch 3250, loss[loss=0.1683, simple_loss=0.24, pruned_loss=0.0483, over 4786.00 frames. ], tot_loss[loss=0.1867, simple_loss=0.2533, pruned_loss=0.06007, over 954733.61 frames. ], batch size: 26, lr: 3.59e-03, grad_scale: 32.0 +2023-03-26 16:00:39,651 INFO [finetune.py:976] (1/7) Epoch 13, batch 3300, loss[loss=0.1755, simple_loss=0.2539, pruned_loss=0.0485, over 4879.00 frames. ], tot_loss[loss=0.1875, simple_loss=0.2548, pruned_loss=0.06007, over 953770.87 frames. ], batch size: 31, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:00:44,477 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.095e+02 1.593e+02 1.995e+02 2.341e+02 5.205e+02, threshold=3.991e+02, percent-clipped=4.0 +2023-03-26 16:01:29,182 INFO [finetune.py:976] (1/7) Epoch 13, batch 3350, loss[loss=0.2217, simple_loss=0.2794, pruned_loss=0.08205, over 4861.00 frames. ], tot_loss[loss=0.1905, simple_loss=0.2584, pruned_loss=0.06127, over 953582.35 frames. ], batch size: 44, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:02:07,913 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72129.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:02:11,039 INFO [finetune.py:976] (1/7) Epoch 13, batch 3400, loss[loss=0.1977, simple_loss=0.271, pruned_loss=0.06222, over 4918.00 frames. ], tot_loss[loss=0.1917, simple_loss=0.2599, pruned_loss=0.06179, over 955598.93 frames. ], batch size: 33, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:02:16,763 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.109e+02 1.708e+02 2.008e+02 2.371e+02 4.954e+02, threshold=4.015e+02, percent-clipped=4.0 +2023-03-26 16:02:26,716 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5573, 1.6786, 1.7763, 1.0182, 1.7950, 2.0500, 1.9555, 1.5399], + device='cuda:1'), covar=tensor([0.0993, 0.0670, 0.0448, 0.0521, 0.0400, 0.0493, 0.0356, 0.0661], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0153, 0.0123, 0.0129, 0.0131, 0.0127, 0.0143, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.3820e-05, 1.1148e-04, 8.8218e-05, 9.3196e-05, 9.2677e-05, 9.2107e-05, + 1.0403e-04, 1.0600e-04], device='cuda:1') +2023-03-26 16:02:44,554 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4318, 1.5078, 1.7222, 1.7908, 1.5697, 3.3016, 1.2730, 1.5129], + device='cuda:1'), covar=tensor([0.0977, 0.1773, 0.1256, 0.0957, 0.1564, 0.0234, 0.1585, 0.1759], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0074, 0.0078, 0.0092, 0.0082, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:02:49,868 INFO [finetune.py:976] (1/7) Epoch 13, batch 3450, loss[loss=0.2317, simple_loss=0.2972, pruned_loss=0.08304, over 4776.00 frames. ], tot_loss[loss=0.1907, simple_loss=0.259, pruned_loss=0.06125, over 957126.38 frames. ], batch size: 29, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:02:55,191 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=72190.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:03:12,696 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2889, 1.9606, 1.4777, 0.5934, 1.7586, 1.9178, 1.7526, 1.8049], + device='cuda:1'), covar=tensor([0.0762, 0.0816, 0.1462, 0.2080, 0.1237, 0.2115, 0.2162, 0.0883], + device='cuda:1'), in_proj_covar=tensor([0.0164, 0.0194, 0.0198, 0.0183, 0.0210, 0.0205, 0.0220, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:03:23,347 INFO [finetune.py:976] (1/7) Epoch 13, batch 3500, loss[loss=0.2079, simple_loss=0.2696, pruned_loss=0.07309, over 4911.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2565, pruned_loss=0.06083, over 955847.60 frames. ], batch size: 36, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:03:29,064 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.069e+02 1.641e+02 1.993e+02 2.438e+02 4.377e+02, threshold=3.986e+02, percent-clipped=2.0 +2023-03-26 16:03:46,437 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0673, 1.4851, 2.0204, 1.8947, 1.7739, 1.7119, 1.8677, 1.8969], + device='cuda:1'), covar=tensor([0.3430, 0.4081, 0.3409, 0.3665, 0.4970, 0.3657, 0.4400, 0.3306], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0237, 0.0254, 0.0263, 0.0259, 0.0234, 0.0274, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:03:49,880 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72272.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:03:51,664 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8410, 3.3839, 3.5243, 3.6954, 3.6226, 3.4050, 3.9060, 1.3155], + device='cuda:1'), covar=tensor([0.0848, 0.0810, 0.0937, 0.0998, 0.1265, 0.1503, 0.0856, 0.4761], + device='cuda:1'), in_proj_covar=tensor([0.0345, 0.0241, 0.0274, 0.0288, 0.0328, 0.0278, 0.0300, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:03:51,667 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72275.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:03:56,421 INFO [finetune.py:976] (1/7) Epoch 13, batch 3550, loss[loss=0.1963, simple_loss=0.2481, pruned_loss=0.07227, over 4834.00 frames. ], tot_loss[loss=0.1876, simple_loss=0.2537, pruned_loss=0.06076, over 956941.55 frames. ], batch size: 38, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:03:57,345 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-03-26 16:04:37,446 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=72320.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:04:47,547 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72326.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:04:51,694 INFO [finetune.py:976] (1/7) Epoch 13, batch 3600, loss[loss=0.2021, simple_loss=0.266, pruned_loss=0.06907, over 4806.00 frames. ], tot_loss[loss=0.1865, simple_loss=0.2519, pruned_loss=0.06061, over 956200.76 frames. ], batch size: 41, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:04:58,281 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.845e+01 1.525e+02 1.754e+02 2.048e+02 3.586e+02, threshold=3.507e+02, percent-clipped=0.0 +2023-03-26 16:05:42,453 INFO [finetune.py:976] (1/7) Epoch 13, batch 3650, loss[loss=0.1386, simple_loss=0.2099, pruned_loss=0.03359, over 4782.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2539, pruned_loss=0.06094, over 952990.83 frames. ], batch size: 26, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:05:50,458 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=72387.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:06:11,592 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2081, 1.9275, 1.3466, 0.6263, 1.6625, 1.7983, 1.6560, 1.7945], + device='cuda:1'), covar=tensor([0.0844, 0.0772, 0.1726, 0.2016, 0.1431, 0.2304, 0.2346, 0.0952], + device='cuda:1'), in_proj_covar=tensor([0.0165, 0.0195, 0.0199, 0.0184, 0.0212, 0.0207, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:06:54,368 INFO [finetune.py:976] (1/7) Epoch 13, batch 3700, loss[loss=0.1699, simple_loss=0.2429, pruned_loss=0.04844, over 4817.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.259, pruned_loss=0.06238, over 954488.05 frames. ], batch size: 51, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:06:57,035 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.71 vs. limit=5.0 +2023-03-26 16:07:04,387 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.108e+02 1.616e+02 1.915e+02 2.308e+02 4.437e+02, threshold=3.829e+02, percent-clipped=1.0 +2023-03-26 16:07:08,806 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9835, 1.2022, 1.9317, 1.8245, 1.7344, 1.6668, 1.7001, 1.8130], + device='cuda:1'), covar=tensor([0.3390, 0.3937, 0.3585, 0.3407, 0.4868, 0.3540, 0.4548, 0.3457], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0238, 0.0255, 0.0264, 0.0260, 0.0235, 0.0276, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:07:27,867 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2460, 2.1580, 2.3621, 1.5040, 2.2153, 2.4327, 2.1938, 1.8644], + device='cuda:1'), covar=tensor([0.0611, 0.0607, 0.0622, 0.0868, 0.0653, 0.0650, 0.0631, 0.1016], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0133, 0.0142, 0.0124, 0.0124, 0.0142, 0.0142, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:07:30,285 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4648, 1.5512, 2.0078, 1.7879, 1.6436, 3.9312, 1.3896, 1.6144], + device='cuda:1'), covar=tensor([0.0980, 0.1741, 0.1306, 0.0987, 0.1585, 0.0240, 0.1538, 0.1823], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0077, 0.0091, 0.0081, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:07:37,532 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2545, 2.0966, 1.7016, 0.8567, 1.9333, 1.8061, 1.6135, 1.8902], + device='cuda:1'), covar=tensor([0.0822, 0.0668, 0.1522, 0.1885, 0.1243, 0.2323, 0.2271, 0.0957], + device='cuda:1'), in_proj_covar=tensor([0.0166, 0.0196, 0.0200, 0.0185, 0.0212, 0.0207, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:07:52,818 INFO [finetune.py:976] (1/7) Epoch 13, batch 3750, loss[loss=0.1611, simple_loss=0.2402, pruned_loss=0.04104, over 4894.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2598, pruned_loss=0.06239, over 952446.87 frames. ], batch size: 32, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:07:54,136 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72485.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:08:19,400 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7137, 1.5007, 2.1146, 3.3587, 2.3634, 2.4131, 1.0122, 2.7675], + device='cuda:1'), covar=tensor([0.1696, 0.1446, 0.1256, 0.0593, 0.0727, 0.1609, 0.1821, 0.0520], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0115, 0.0132, 0.0163, 0.0100, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 16:08:29,282 INFO [finetune.py:976] (1/7) Epoch 13, batch 3800, loss[loss=0.1873, simple_loss=0.2531, pruned_loss=0.06077, over 4758.00 frames. ], tot_loss[loss=0.1925, simple_loss=0.2607, pruned_loss=0.06219, over 954129.03 frames. ], batch size: 27, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:08:33,554 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2175, 1.8711, 2.5953, 1.6171, 2.2722, 2.3771, 1.7838, 2.6616], + device='cuda:1'), covar=tensor([0.1367, 0.1986, 0.1843, 0.2407, 0.0962, 0.1679, 0.2770, 0.0847], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0204, 0.0193, 0.0190, 0.0177, 0.0213, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:08:34,665 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.445e+01 1.578e+02 1.803e+02 2.155e+02 3.901e+02, threshold=3.607e+02, percent-clipped=1.0 +2023-03-26 16:08:56,819 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72575.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:09:02,584 INFO [finetune.py:976] (1/7) Epoch 13, batch 3850, loss[loss=0.1895, simple_loss=0.2449, pruned_loss=0.06703, over 4829.00 frames. ], tot_loss[loss=0.1897, simple_loss=0.2577, pruned_loss=0.06083, over 953548.77 frames. ], batch size: 25, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:09:30,590 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=72623.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:09:45,334 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2119, 2.1973, 1.9822, 2.2620, 2.8619, 2.2022, 2.2039, 1.6599], + device='cuda:1'), covar=tensor([0.2232, 0.1927, 0.1894, 0.1672, 0.1697, 0.1115, 0.2047, 0.1987], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0205, 0.0208, 0.0189, 0.0238, 0.0182, 0.0212, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:09:46,306 INFO [finetune.py:976] (1/7) Epoch 13, batch 3900, loss[loss=0.1839, simple_loss=0.2413, pruned_loss=0.06327, over 4825.00 frames. ], tot_loss[loss=0.1878, simple_loss=0.2548, pruned_loss=0.06037, over 954661.93 frames. ], batch size: 30, lr: 3.59e-03, grad_scale: 16.0 +2023-03-26 16:09:47,003 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3276, 2.9039, 2.8209, 1.2094, 3.0406, 2.1386, 0.6587, 1.9341], + device='cuda:1'), covar=tensor([0.2245, 0.2191, 0.1778, 0.3907, 0.1276, 0.1216, 0.4547, 0.1751], + device='cuda:1'), in_proj_covar=tensor([0.0148, 0.0172, 0.0157, 0.0127, 0.0155, 0.0120, 0.0144, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 16:09:51,188 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.402e+01 1.493e+02 1.751e+02 2.217e+02 3.590e+02, threshold=3.501e+02, percent-clipped=0.0 +2023-03-26 16:09:52,438 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1941, 1.7769, 2.2793, 2.0530, 1.8573, 1.8705, 2.0741, 1.9515], + device='cuda:1'), covar=tensor([0.4272, 0.4730, 0.3432, 0.4564, 0.5443, 0.3913, 0.5334, 0.3476], + device='cuda:1'), in_proj_covar=tensor([0.0241, 0.0238, 0.0256, 0.0264, 0.0261, 0.0236, 0.0276, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:09:56,481 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5361, 3.3313, 3.1689, 1.4434, 3.4785, 2.5108, 0.7142, 2.2754], + device='cuda:1'), covar=tensor([0.2367, 0.2155, 0.1732, 0.3679, 0.1110, 0.1094, 0.4572, 0.1785], + device='cuda:1'), in_proj_covar=tensor([0.0149, 0.0172, 0.0158, 0.0127, 0.0156, 0.0120, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 16:09:58,343 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5035, 1.3042, 1.2047, 1.4378, 1.6598, 1.4763, 0.9346, 1.2556], + device='cuda:1'), covar=tensor([0.2238, 0.2231, 0.2252, 0.1820, 0.1645, 0.1397, 0.2670, 0.2033], + device='cuda:1'), in_proj_covar=tensor([0.0236, 0.0205, 0.0208, 0.0189, 0.0238, 0.0182, 0.0211, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:10:18,069 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72682.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:10:18,645 INFO [finetune.py:976] (1/7) Epoch 13, batch 3950, loss[loss=0.1421, simple_loss=0.2123, pruned_loss=0.03601, over 4697.00 frames. ], tot_loss[loss=0.1847, simple_loss=0.2514, pruned_loss=0.05898, over 955811.14 frames. ], batch size: 23, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:10:22,808 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4312, 1.7238, 2.1584, 1.8505, 1.7542, 4.1426, 1.3487, 1.8051], + device='cuda:1'), covar=tensor([0.0928, 0.1639, 0.1107, 0.0973, 0.1513, 0.0183, 0.1519, 0.1643], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0073, 0.0077, 0.0091, 0.0081, 0.0085, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:10:50,320 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9161, 1.4879, 2.0156, 1.9029, 1.7111, 1.6493, 1.8414, 1.7698], + device='cuda:1'), covar=tensor([0.4275, 0.4391, 0.3405, 0.3847, 0.4917, 0.3842, 0.4783, 0.3551], + device='cuda:1'), in_proj_covar=tensor([0.0241, 0.0238, 0.0256, 0.0264, 0.0260, 0.0236, 0.0275, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:11:10,484 INFO [finetune.py:976] (1/7) Epoch 13, batch 4000, loss[loss=0.2148, simple_loss=0.2799, pruned_loss=0.0748, over 4824.00 frames. ], tot_loss[loss=0.1852, simple_loss=0.2516, pruned_loss=0.05944, over 956214.06 frames. ], batch size: 39, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:11:16,799 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.220e+02 1.596e+02 1.921e+02 2.181e+02 4.609e+02, threshold=3.842e+02, percent-clipped=3.0 +2023-03-26 16:11:44,632 INFO [finetune.py:976] (1/7) Epoch 13, batch 4050, loss[loss=0.1545, simple_loss=0.2138, pruned_loss=0.04759, over 4037.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2558, pruned_loss=0.06141, over 953556.32 frames. ], batch size: 17, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:11:46,039 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72785.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:12:15,995 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1453, 1.8019, 1.9108, 0.8010, 2.2532, 2.5123, 1.8839, 1.8268], + device='cuda:1'), covar=tensor([0.1174, 0.1455, 0.0654, 0.0912, 0.0696, 0.0641, 0.0638, 0.0788], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0153, 0.0122, 0.0129, 0.0131, 0.0126, 0.0142, 0.0145], + device='cuda:1'), out_proj_covar=tensor([9.2800e-05, 1.1140e-04, 8.8100e-05, 9.2586e-05, 9.2798e-05, 9.1376e-05, + 1.0365e-04, 1.0513e-04], device='cuda:1') +2023-03-26 16:12:26,290 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-03-26 16:12:28,583 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 16:12:36,322 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2832, 2.6429, 2.2809, 1.8590, 2.5220, 2.7315, 2.6372, 2.2590], + device='cuda:1'), covar=tensor([0.0691, 0.0594, 0.0810, 0.0932, 0.0748, 0.0655, 0.0627, 0.0976], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0132, 0.0140, 0.0122, 0.0123, 0.0140, 0.0140, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:12:39,938 INFO [finetune.py:976] (1/7) Epoch 13, batch 4100, loss[loss=0.1879, simple_loss=0.2554, pruned_loss=0.06022, over 4766.00 frames. ], tot_loss[loss=0.1905, simple_loss=0.258, pruned_loss=0.06153, over 954203.06 frames. ], batch size: 28, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:12:39,998 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=72833.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:12:45,296 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.089e+02 1.592e+02 1.875e+02 2.230e+02 3.624e+02, threshold=3.749e+02, percent-clipped=0.0 +2023-03-26 16:13:02,152 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8562, 1.0241, 1.7995, 1.7116, 1.5982, 1.5109, 1.6482, 1.6192], + device='cuda:1'), covar=tensor([0.3519, 0.4175, 0.3573, 0.3632, 0.4839, 0.3612, 0.4186, 0.3359], + device='cuda:1'), in_proj_covar=tensor([0.0241, 0.0237, 0.0256, 0.0263, 0.0260, 0.0235, 0.0275, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:13:11,582 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9075, 1.5216, 1.9883, 1.4417, 1.8605, 2.0315, 1.4634, 2.2249], + device='cuda:1'), covar=tensor([0.1144, 0.2164, 0.1312, 0.1975, 0.0819, 0.1414, 0.2682, 0.0779], + device='cuda:1'), in_proj_covar=tensor([0.0195, 0.0205, 0.0192, 0.0191, 0.0178, 0.0213, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:13:13,291 INFO [finetune.py:976] (1/7) Epoch 13, batch 4150, loss[loss=0.185, simple_loss=0.2531, pruned_loss=0.05842, over 4864.00 frames. ], tot_loss[loss=0.1908, simple_loss=0.2588, pruned_loss=0.06142, over 952665.14 frames. ], batch size: 31, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:13:26,379 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72894.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:14:01,531 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5002, 1.5022, 1.8719, 1.8607, 1.5941, 3.4937, 1.3179, 1.6294], + device='cuda:1'), covar=tensor([0.0916, 0.1761, 0.1045, 0.0931, 0.1627, 0.0224, 0.1507, 0.1754], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0073, 0.0077, 0.0091, 0.0081, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:14:03,790 INFO [finetune.py:976] (1/7) Epoch 13, batch 4200, loss[loss=0.2034, simple_loss=0.2741, pruned_loss=0.06637, over 4810.00 frames. ], tot_loss[loss=0.1912, simple_loss=0.2597, pruned_loss=0.06134, over 952392.50 frames. ], batch size: 40, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:14:08,711 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.693e+01 1.496e+02 1.812e+02 2.169e+02 4.504e+02, threshold=3.624e+02, percent-clipped=2.0 +2023-03-26 16:14:18,160 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8695, 1.4179, 1.9452, 1.8343, 1.6626, 1.5933, 1.7528, 1.7553], + device='cuda:1'), covar=tensor([0.4140, 0.4193, 0.3474, 0.3895, 0.5064, 0.3831, 0.4762, 0.3406], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0238, 0.0256, 0.0264, 0.0262, 0.0236, 0.0277, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:14:18,693 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=72955.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:14:30,147 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72964.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 16:14:37,442 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7548, 1.3805, 2.0097, 1.3834, 1.7123, 1.9295, 1.3894, 2.0356], + device='cuda:1'), covar=tensor([0.1049, 0.2202, 0.1002, 0.1479, 0.0780, 0.1165, 0.2747, 0.0746], + device='cuda:1'), in_proj_covar=tensor([0.0195, 0.0206, 0.0193, 0.0191, 0.0179, 0.0214, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:14:52,081 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72982.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:14:53,055 INFO [finetune.py:976] (1/7) Epoch 13, batch 4250, loss[loss=0.2125, simple_loss=0.2517, pruned_loss=0.08666, over 4815.00 frames. ], tot_loss[loss=0.1892, simple_loss=0.2571, pruned_loss=0.06065, over 952202.09 frames. ], batch size: 30, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:15:21,368 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73025.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 16:15:24,375 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=73030.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:15:26,658 INFO [finetune.py:976] (1/7) Epoch 13, batch 4300, loss[loss=0.1369, simple_loss=0.2045, pruned_loss=0.03466, over 4800.00 frames. ], tot_loss[loss=0.1871, simple_loss=0.2543, pruned_loss=0.05988, over 953301.34 frames. ], batch size: 29, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:15:27,968 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8493, 1.7957, 1.9375, 1.1528, 1.9866, 1.9648, 1.9864, 1.6359], + device='cuda:1'), covar=tensor([0.0578, 0.0657, 0.0630, 0.0993, 0.0754, 0.0690, 0.0574, 0.1133], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0132, 0.0141, 0.0123, 0.0123, 0.0141, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:15:31,990 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.052e+02 1.498e+02 1.782e+02 2.254e+02 4.055e+02, threshold=3.563e+02, percent-clipped=2.0 +2023-03-26 16:15:59,435 INFO [finetune.py:976] (1/7) Epoch 13, batch 4350, loss[loss=0.2266, simple_loss=0.2868, pruned_loss=0.08315, over 4753.00 frames. ], tot_loss[loss=0.1837, simple_loss=0.2509, pruned_loss=0.05821, over 954145.66 frames. ], batch size: 59, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:16:06,478 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5868, 1.3825, 2.2079, 3.2806, 2.1543, 2.3417, 0.9870, 2.6021], + device='cuda:1'), covar=tensor([0.1779, 0.1517, 0.1169, 0.0603, 0.0861, 0.1378, 0.1837, 0.0589], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0115, 0.0133, 0.0165, 0.0101, 0.0138, 0.0126, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 16:16:18,127 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7420, 0.6899, 1.6980, 1.6431, 1.5590, 1.4565, 1.5528, 1.6231], + device='cuda:1'), covar=tensor([0.3147, 0.3702, 0.3029, 0.2926, 0.3954, 0.3035, 0.3520, 0.2956], + device='cuda:1'), in_proj_covar=tensor([0.0241, 0.0237, 0.0255, 0.0263, 0.0261, 0.0235, 0.0275, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:16:27,417 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2110, 1.7313, 2.2265, 2.1209, 1.8273, 1.9049, 2.0927, 1.9796], + device='cuda:1'), covar=tensor([0.4184, 0.4491, 0.3310, 0.3921, 0.5330, 0.4065, 0.4682, 0.3329], + device='cuda:1'), in_proj_covar=tensor([0.0241, 0.0237, 0.0255, 0.0263, 0.0261, 0.0235, 0.0275, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:16:34,973 INFO [finetune.py:976] (1/7) Epoch 13, batch 4400, loss[loss=0.1587, simple_loss=0.241, pruned_loss=0.03825, over 4865.00 frames. ], tot_loss[loss=0.1848, simple_loss=0.2522, pruned_loss=0.05873, over 954658.81 frames. ], batch size: 44, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:16:40,310 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.328e+01 1.433e+02 1.829e+02 2.142e+02 3.915e+02, threshold=3.659e+02, percent-clipped=1.0 +2023-03-26 16:16:49,379 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-26 16:17:08,723 INFO [finetune.py:976] (1/7) Epoch 13, batch 4450, loss[loss=0.1902, simple_loss=0.2704, pruned_loss=0.05501, over 4889.00 frames. ], tot_loss[loss=0.1892, simple_loss=0.2572, pruned_loss=0.06065, over 951953.53 frames. ], batch size: 32, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:17:43,370 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.68 vs. limit=5.0 +2023-03-26 16:17:53,147 INFO [finetune.py:976] (1/7) Epoch 13, batch 4500, loss[loss=0.1723, simple_loss=0.2514, pruned_loss=0.04659, over 4894.00 frames. ], tot_loss[loss=0.1908, simple_loss=0.2585, pruned_loss=0.06151, over 952251.26 frames. ], batch size: 35, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:17:55,818 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.14 vs. limit=5.0 +2023-03-26 16:17:58,019 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.220e+02 1.732e+02 2.105e+02 2.505e+02 4.470e+02, threshold=4.210e+02, percent-clipped=3.0 +2023-03-26 16:18:04,471 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73250.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:18:10,477 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6987, 1.6570, 1.5919, 1.6768, 1.2954, 3.0703, 1.4878, 1.8175], + device='cuda:1'), covar=tensor([0.3013, 0.2047, 0.1835, 0.2027, 0.1623, 0.0277, 0.2709, 0.1201], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0115, 0.0121, 0.0124, 0.0115, 0.0098, 0.0098, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:18:26,891 INFO [finetune.py:976] (1/7) Epoch 13, batch 4550, loss[loss=0.2173, simple_loss=0.2952, pruned_loss=0.06968, over 4741.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2598, pruned_loss=0.06163, over 953252.12 frames. ], batch size: 54, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:18:34,954 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=73293.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:19:07,530 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73320.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 16:19:19,977 INFO [finetune.py:976] (1/7) Epoch 13, batch 4600, loss[loss=0.1857, simple_loss=0.249, pruned_loss=0.06123, over 4830.00 frames. ], tot_loss[loss=0.1903, simple_loss=0.2584, pruned_loss=0.06107, over 953778.46 frames. ], batch size: 38, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:19:24,888 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.809e+01 1.600e+02 1.901e+02 2.194e+02 3.702e+02, threshold=3.803e+02, percent-clipped=0.0 +2023-03-26 16:19:42,014 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73354.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:20:11,744 INFO [finetune.py:976] (1/7) Epoch 13, batch 4650, loss[loss=0.1605, simple_loss=0.2282, pruned_loss=0.04643, over 4914.00 frames. ], tot_loss[loss=0.189, simple_loss=0.2563, pruned_loss=0.06086, over 953611.25 frames. ], batch size: 36, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:20:45,672 INFO [finetune.py:976] (1/7) Epoch 13, batch 4700, loss[loss=0.1639, simple_loss=0.2297, pruned_loss=0.04903, over 4811.00 frames. ], tot_loss[loss=0.1862, simple_loss=0.2532, pruned_loss=0.05958, over 955767.99 frames. ], batch size: 51, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:20:48,327 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-26 16:20:50,433 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.031e+02 1.614e+02 1.909e+02 2.257e+02 3.771e+02, threshold=3.817e+02, percent-clipped=0.0 +2023-03-26 16:21:18,783 INFO [finetune.py:976] (1/7) Epoch 13, batch 4750, loss[loss=0.1971, simple_loss=0.2608, pruned_loss=0.06675, over 4903.00 frames. ], tot_loss[loss=0.185, simple_loss=0.2517, pruned_loss=0.05917, over 952826.37 frames. ], batch size: 35, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:21:34,975 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5256, 1.3618, 1.3286, 1.4429, 1.7122, 1.6256, 1.4354, 1.3244], + device='cuda:1'), covar=tensor([0.0308, 0.0285, 0.0552, 0.0260, 0.0251, 0.0417, 0.0304, 0.0393], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0108, 0.0138, 0.0112, 0.0100, 0.0104, 0.0095, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.2436e-05, 8.3543e-05, 1.0937e-04, 8.6628e-05, 7.8073e-05, 7.6998e-05, + 7.1394e-05, 8.2834e-05], device='cuda:1') +2023-03-26 16:21:51,902 INFO [finetune.py:976] (1/7) Epoch 13, batch 4800, loss[loss=0.2266, simple_loss=0.2955, pruned_loss=0.07883, over 4818.00 frames. ], tot_loss[loss=0.1874, simple_loss=0.2539, pruned_loss=0.0604, over 953098.33 frames. ], batch size: 38, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:21:56,135 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0454, 2.0170, 1.5606, 1.9108, 2.0073, 1.7240, 2.3177, 2.0499], + device='cuda:1'), covar=tensor([0.1418, 0.2113, 0.3253, 0.2829, 0.2720, 0.1686, 0.3419, 0.1962], + device='cuda:1'), in_proj_covar=tensor([0.0179, 0.0187, 0.0234, 0.0254, 0.0243, 0.0199, 0.0212, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:21:57,193 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.195e+02 1.637e+02 2.007e+02 2.318e+02 3.852e+02, threshold=4.014e+02, percent-clipped=1.0 +2023-03-26 16:22:03,308 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=73550.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:22:24,805 INFO [finetune.py:976] (1/7) Epoch 13, batch 4850, loss[loss=0.2007, simple_loss=0.2705, pruned_loss=0.06548, over 4749.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2582, pruned_loss=0.06194, over 954718.97 frames. ], batch size: 59, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:22:30,077 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=73590.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:22:37,205 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=73598.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:22:39,695 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2740, 2.9050, 2.7735, 1.1300, 3.0264, 2.2985, 0.7947, 1.7970], + device='cuda:1'), covar=tensor([0.2388, 0.2054, 0.1693, 0.3292, 0.1321, 0.1067, 0.3754, 0.1580], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0174, 0.0160, 0.0128, 0.0157, 0.0122, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 16:23:00,184 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=73620.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 16:23:08,517 INFO [finetune.py:976] (1/7) Epoch 13, batch 4900, loss[loss=0.1833, simple_loss=0.2609, pruned_loss=0.05283, over 4892.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2598, pruned_loss=0.06242, over 955965.39 frames. ], batch size: 36, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:23:14,277 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.113e+02 1.751e+02 2.108e+02 2.596e+02 5.059e+02, threshold=4.217e+02, percent-clipped=3.0 +2023-03-26 16:23:19,728 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73649.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:23:21,031 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73651.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:23:31,876 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=73668.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 16:23:36,191 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 16:23:41,382 INFO [finetune.py:976] (1/7) Epoch 13, batch 4950, loss[loss=0.1685, simple_loss=0.2344, pruned_loss=0.05131, over 4743.00 frames. ], tot_loss[loss=0.1926, simple_loss=0.2605, pruned_loss=0.0623, over 955916.64 frames. ], batch size: 23, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:24:24,430 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=73732.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:24:24,559 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 16:24:24,937 INFO [finetune.py:976] (1/7) Epoch 13, batch 5000, loss[loss=0.1578, simple_loss=0.2218, pruned_loss=0.04685, over 4848.00 frames. ], tot_loss[loss=0.1918, simple_loss=0.2594, pruned_loss=0.06214, over 955887.68 frames. ], batch size: 47, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:24:33,743 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.586e+02 1.888e+02 2.371e+02 3.310e+02, threshold=3.776e+02, percent-clipped=1.0 +2023-03-26 16:24:41,323 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5072, 2.8117, 2.6475, 2.0263, 2.8473, 2.9082, 2.9035, 2.4624], + device='cuda:1'), covar=tensor([0.0594, 0.0544, 0.0676, 0.0855, 0.0522, 0.0719, 0.0578, 0.0899], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0132, 0.0140, 0.0122, 0.0122, 0.0140, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:24:47,752 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0696, 1.3796, 1.9293, 1.9744, 1.7128, 1.6620, 1.8498, 1.8356], + device='cuda:1'), covar=tensor([0.3540, 0.3827, 0.3286, 0.3406, 0.4821, 0.3704, 0.4317, 0.3215], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0236, 0.0254, 0.0263, 0.0261, 0.0235, 0.0275, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:25:17,330 INFO [finetune.py:976] (1/7) Epoch 13, batch 5050, loss[loss=0.169, simple_loss=0.2201, pruned_loss=0.05894, over 3891.00 frames. ], tot_loss[loss=0.189, simple_loss=0.2561, pruned_loss=0.061, over 954146.01 frames. ], batch size: 16, lr: 3.58e-03, grad_scale: 16.0 +2023-03-26 16:25:26,630 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73793.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:25:53,899 INFO [finetune.py:976] (1/7) Epoch 13, batch 5100, loss[loss=0.159, simple_loss=0.2299, pruned_loss=0.04402, over 4802.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2531, pruned_loss=0.06005, over 955554.80 frames. ], batch size: 25, lr: 3.57e-03, grad_scale: 16.0 +2023-03-26 16:25:59,154 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.804e+01 1.435e+02 1.752e+02 2.086e+02 3.868e+02, threshold=3.504e+02, percent-clipped=1.0 +2023-03-26 16:26:27,642 INFO [finetune.py:976] (1/7) Epoch 13, batch 5150, loss[loss=0.2138, simple_loss=0.2844, pruned_loss=0.07161, over 4821.00 frames. ], tot_loss[loss=0.1871, simple_loss=0.2534, pruned_loss=0.06042, over 956101.91 frames. ], batch size: 39, lr: 3.57e-03, grad_scale: 16.0 +2023-03-26 16:26:37,937 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-03-26 16:26:47,156 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9962, 1.9294, 2.0741, 1.5947, 2.0229, 2.1530, 2.1026, 1.6243], + device='cuda:1'), covar=tensor([0.0524, 0.0622, 0.0599, 0.0784, 0.0686, 0.0650, 0.0568, 0.1115], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0134, 0.0142, 0.0124, 0.0123, 0.0142, 0.0142, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:26:50,206 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5889, 1.5237, 1.5403, 1.5678, 1.0215, 3.0147, 1.1185, 1.6665], + device='cuda:1'), covar=tensor([0.3536, 0.2613, 0.2110, 0.2507, 0.1986, 0.0260, 0.2779, 0.1304], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0123, 0.0115, 0.0098, 0.0098, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:27:01,331 INFO [finetune.py:976] (1/7) Epoch 13, batch 5200, loss[loss=0.2149, simple_loss=0.285, pruned_loss=0.07234, over 4830.00 frames. ], tot_loss[loss=0.1918, simple_loss=0.2588, pruned_loss=0.06239, over 957143.48 frames. ], batch size: 33, lr: 3.57e-03, grad_scale: 16.0 +2023-03-26 16:27:06,218 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.103e+02 1.675e+02 1.952e+02 2.217e+02 3.649e+02, threshold=3.904e+02, percent-clipped=2.0 +2023-03-26 16:27:09,798 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73946.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:27:11,693 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=73949.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:27:30,172 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 16:27:34,703 INFO [finetune.py:976] (1/7) Epoch 13, batch 5250, loss[loss=0.1935, simple_loss=0.2586, pruned_loss=0.06422, over 4828.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2595, pruned_loss=0.06184, over 958540.09 frames. ], batch size: 39, lr: 3.57e-03, grad_scale: 16.0 +2023-03-26 16:27:43,267 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.55 vs. limit=5.0 +2023-03-26 16:27:44,360 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=73997.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:28:11,317 INFO [finetune.py:976] (1/7) Epoch 13, batch 5300, loss[loss=0.1984, simple_loss=0.2651, pruned_loss=0.06587, over 4809.00 frames. ], tot_loss[loss=0.1924, simple_loss=0.26, pruned_loss=0.06236, over 958498.24 frames. ], batch size: 40, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:28:17,127 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.042e+02 1.638e+02 2.059e+02 2.515e+02 4.122e+02, threshold=4.117e+02, percent-clipped=3.0 +2023-03-26 16:28:33,137 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74064.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:28:45,018 INFO [finetune.py:976] (1/7) Epoch 13, batch 5350, loss[loss=0.2387, simple_loss=0.2912, pruned_loss=0.09308, over 4153.00 frames. ], tot_loss[loss=0.1929, simple_loss=0.2602, pruned_loss=0.06286, over 956312.73 frames. ], batch size: 18, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:28:48,556 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74088.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:28:53,872 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0181, 1.8226, 1.6507, 1.5907, 1.7527, 1.7667, 1.7705, 2.4937], + device='cuda:1'), covar=tensor([0.3953, 0.4309, 0.3293, 0.4080, 0.4226, 0.2400, 0.4051, 0.1640], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0259, 0.0223, 0.0275, 0.0245, 0.0212, 0.0247, 0.0222], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:29:12,979 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74125.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:29:18,157 INFO [finetune.py:976] (1/7) Epoch 13, batch 5400, loss[loss=0.1724, simple_loss=0.2459, pruned_loss=0.04941, over 4923.00 frames. ], tot_loss[loss=0.1913, simple_loss=0.2574, pruned_loss=0.06258, over 954655.10 frames. ], batch size: 38, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:29:27,916 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.066e+02 1.534e+02 1.853e+02 2.261e+02 4.254e+02, threshold=3.706e+02, percent-clipped=1.0 +2023-03-26 16:29:54,617 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4384, 1.5735, 1.2842, 1.5052, 1.8524, 1.7279, 1.6063, 1.3864], + device='cuda:1'), covar=tensor([0.0372, 0.0274, 0.0521, 0.0279, 0.0199, 0.0462, 0.0261, 0.0368], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0108, 0.0139, 0.0113, 0.0101, 0.0105, 0.0095, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2691e-05, 8.4038e-05, 1.1032e-04, 8.7711e-05, 7.8829e-05, 7.7503e-05, + 7.1767e-05, 8.3444e-05], device='cuda:1') +2023-03-26 16:30:02,346 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4005, 1.3329, 1.2608, 1.3397, 1.6580, 1.4846, 1.4234, 1.2406], + device='cuda:1'), covar=tensor([0.0326, 0.0250, 0.0533, 0.0274, 0.0237, 0.0444, 0.0282, 0.0374], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0139, 0.0113, 0.0101, 0.0105, 0.0095, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2787e-05, 8.4178e-05, 1.1047e-04, 8.7819e-05, 7.8915e-05, 7.7659e-05, + 7.1894e-05, 8.3549e-05], device='cuda:1') +2023-03-26 16:30:11,887 INFO [finetune.py:976] (1/7) Epoch 13, batch 5450, loss[loss=0.1632, simple_loss=0.2254, pruned_loss=0.05051, over 4717.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2547, pruned_loss=0.0612, over 955642.36 frames. ], batch size: 23, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:30:56,457 INFO [finetune.py:976] (1/7) Epoch 13, batch 5500, loss[loss=0.1873, simple_loss=0.2582, pruned_loss=0.0582, over 4904.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.253, pruned_loss=0.06073, over 956274.22 frames. ], batch size: 43, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:31:01,343 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74240.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:31:01,847 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.610e+01 1.534e+02 1.911e+02 2.187e+02 5.924e+02, threshold=3.822e+02, percent-clipped=2.0 +2023-03-26 16:31:04,993 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74246.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:31:30,481 INFO [finetune.py:976] (1/7) Epoch 13, batch 5550, loss[loss=0.1836, simple_loss=0.2556, pruned_loss=0.0558, over 4791.00 frames. ], tot_loss[loss=0.1881, simple_loss=0.2539, pruned_loss=0.06115, over 956638.24 frames. ], batch size: 29, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:31:37,727 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=74294.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:31:42,489 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74301.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:32:02,276 INFO [finetune.py:976] (1/7) Epoch 13, batch 5600, loss[loss=0.1878, simple_loss=0.2621, pruned_loss=0.05675, over 4898.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2581, pruned_loss=0.062, over 957451.59 frames. ], batch size: 35, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:32:06,847 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.439e+01 1.528e+02 1.833e+02 2.251e+02 4.644e+02, threshold=3.666e+02, percent-clipped=1.0 +2023-03-26 16:32:31,595 INFO [finetune.py:976] (1/7) Epoch 13, batch 5650, loss[loss=0.2229, simple_loss=0.3034, pruned_loss=0.07117, over 4819.00 frames. ], tot_loss[loss=0.1945, simple_loss=0.262, pruned_loss=0.06343, over 957744.83 frames. ], batch size: 45, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:32:35,019 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74388.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:32:54,213 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74420.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:32:57,287 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 16:33:00,740 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0769, 4.2132, 4.0345, 2.5702, 4.2422, 3.4175, 1.3340, 3.1342], + device='cuda:1'), covar=tensor([0.2115, 0.2082, 0.1675, 0.3035, 0.1113, 0.0938, 0.4279, 0.1538], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0173, 0.0161, 0.0128, 0.0156, 0.0121, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 16:33:01,853 INFO [finetune.py:976] (1/7) Epoch 13, batch 5700, loss[loss=0.1758, simple_loss=0.2321, pruned_loss=0.0597, over 4319.00 frames. ], tot_loss[loss=0.1912, simple_loss=0.2574, pruned_loss=0.06246, over 939069.07 frames. ], batch size: 19, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:33:03,647 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=74436.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:33:06,487 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.627e+01 1.487e+02 1.916e+02 2.529e+02 4.839e+02, threshold=3.833e+02, percent-clipped=5.0 +2023-03-26 16:33:31,118 INFO [finetune.py:976] (1/7) Epoch 14, batch 0, loss[loss=0.1768, simple_loss=0.2554, pruned_loss=0.0491, over 4924.00 frames. ], tot_loss[loss=0.1768, simple_loss=0.2554, pruned_loss=0.0491, over 4924.00 frames. ], batch size: 42, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:33:31,118 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 16:33:41,691 INFO [finetune.py:1010] (1/7) Epoch 14, validation: loss=0.1582, simple_loss=0.2295, pruned_loss=0.04344, over 2265189.00 frames. +2023-03-26 16:33:41,691 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 16:33:53,659 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1737, 1.2873, 1.1767, 1.3012, 1.4630, 2.4686, 1.2468, 1.4509], + device='cuda:1'), covar=tensor([0.1061, 0.2021, 0.1162, 0.1015, 0.1780, 0.0430, 0.1692, 0.1860], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0077, 0.0091, 0.0080, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:34:14,919 INFO [finetune.py:976] (1/7) Epoch 14, batch 50, loss[loss=0.2068, simple_loss=0.2701, pruned_loss=0.07175, over 4864.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2648, pruned_loss=0.06615, over 217671.42 frames. ], batch size: 34, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:34:42,636 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.553e+01 1.578e+02 1.920e+02 2.248e+02 3.729e+02, threshold=3.841e+02, percent-clipped=1.0 +2023-03-26 16:35:04,284 INFO [finetune.py:976] (1/7) Epoch 14, batch 100, loss[loss=0.2282, simple_loss=0.2819, pruned_loss=0.08721, over 4899.00 frames. ], tot_loss[loss=0.1908, simple_loss=0.2555, pruned_loss=0.06306, over 380257.12 frames. ], batch size: 37, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:35:04,425 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2134, 2.0820, 1.8283, 2.0646, 1.9746, 1.9730, 1.9862, 2.6955], + device='cuda:1'), covar=tensor([0.3913, 0.4769, 0.3678, 0.3921, 0.4125, 0.2504, 0.3820, 0.1806], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0261, 0.0225, 0.0277, 0.0247, 0.0213, 0.0248, 0.0223], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:35:31,042 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74595.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:35:31,620 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74596.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:35:49,899 INFO [finetune.py:976] (1/7) Epoch 14, batch 150, loss[loss=0.1435, simple_loss=0.2064, pruned_loss=0.04032, over 4816.00 frames. ], tot_loss[loss=0.1849, simple_loss=0.2491, pruned_loss=0.06034, over 506925.92 frames. ], batch size: 41, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:36:22,702 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.045e+02 1.552e+02 1.795e+02 2.139e+02 3.747e+02, threshold=3.589e+02, percent-clipped=0.0 +2023-03-26 16:36:32,536 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74656.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:36:35,947 INFO [finetune.py:976] (1/7) Epoch 14, batch 200, loss[loss=0.2044, simple_loss=0.2704, pruned_loss=0.06925, over 3992.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2482, pruned_loss=0.05943, over 606636.98 frames. ], batch size: 65, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:37:09,842 INFO [finetune.py:976] (1/7) Epoch 14, batch 250, loss[loss=0.1714, simple_loss=0.243, pruned_loss=0.04991, over 4813.00 frames. ], tot_loss[loss=0.1896, simple_loss=0.2557, pruned_loss=0.06175, over 685506.73 frames. ], batch size: 51, lr: 3.57e-03, grad_scale: 32.0 +2023-03-26 16:37:15,966 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74720.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:37:30,097 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.719e+02 2.081e+02 2.468e+02 4.342e+02, threshold=4.162e+02, percent-clipped=2.0 +2023-03-26 16:37:41,614 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0100, 1.8898, 1.6208, 1.7578, 1.7958, 1.7842, 1.8060, 2.5086], + device='cuda:1'), covar=tensor([0.3946, 0.4497, 0.3467, 0.4221, 0.4322, 0.2519, 0.4112, 0.1698], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0259, 0.0224, 0.0275, 0.0246, 0.0213, 0.0247, 0.0222], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:37:42,695 INFO [finetune.py:976] (1/7) Epoch 14, batch 300, loss[loss=0.2175, simple_loss=0.2795, pruned_loss=0.0777, over 4758.00 frames. ], tot_loss[loss=0.189, simple_loss=0.2562, pruned_loss=0.06088, over 745652.48 frames. ], batch size: 28, lr: 3.56e-03, grad_scale: 32.0 +2023-03-26 16:37:48,010 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=74768.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:38:16,372 INFO [finetune.py:976] (1/7) Epoch 14, batch 350, loss[loss=0.2312, simple_loss=0.2883, pruned_loss=0.08707, over 4781.00 frames. ], tot_loss[loss=0.1909, simple_loss=0.2583, pruned_loss=0.06174, over 792699.16 frames. ], batch size: 51, lr: 3.56e-03, grad_scale: 32.0 +2023-03-26 16:38:36,813 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.618e+01 1.648e+02 1.967e+02 2.475e+02 5.107e+02, threshold=3.933e+02, percent-clipped=3.0 +2023-03-26 16:38:39,428 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-26 16:38:49,809 INFO [finetune.py:976] (1/7) Epoch 14, batch 400, loss[loss=0.169, simple_loss=0.2474, pruned_loss=0.0453, over 4834.00 frames. ], tot_loss[loss=0.1915, simple_loss=0.2598, pruned_loss=0.06166, over 830462.96 frames. ], batch size: 49, lr: 3.56e-03, grad_scale: 32.0 +2023-03-26 16:39:13,569 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74896.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:39:22,425 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74909.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:39:23,515 INFO [finetune.py:976] (1/7) Epoch 14, batch 450, loss[loss=0.2, simple_loss=0.2708, pruned_loss=0.06456, over 4901.00 frames. ], tot_loss[loss=0.1901, simple_loss=0.2582, pruned_loss=0.06097, over 857056.17 frames. ], batch size: 36, lr: 3.56e-03, grad_scale: 32.0 +2023-03-26 16:39:35,429 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9831, 1.8730, 1.7288, 1.9013, 1.4601, 4.5625, 1.6686, 2.3001], + device='cuda:1'), covar=tensor([0.3290, 0.2476, 0.2216, 0.2354, 0.1679, 0.0131, 0.2575, 0.1248], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0121, 0.0124, 0.0116, 0.0098, 0.0098, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:39:43,596 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.160e+02 1.587e+02 1.868e+02 2.260e+02 4.285e+02, threshold=3.737e+02, percent-clipped=2.0 +2023-03-26 16:39:45,923 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=74944.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:39:50,146 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74951.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:39:58,628 INFO [finetune.py:976] (1/7) Epoch 14, batch 500, loss[loss=0.1688, simple_loss=0.2392, pruned_loss=0.04915, over 4806.00 frames. ], tot_loss[loss=0.1883, simple_loss=0.2556, pruned_loss=0.06051, over 877074.59 frames. ], batch size: 45, lr: 3.56e-03, grad_scale: 32.0 +2023-03-26 16:40:00,595 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74964.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:40:08,981 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74970.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:40:19,257 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3516, 3.7448, 3.9617, 4.1718, 4.1224, 3.9361, 4.4847, 1.3706], + device='cuda:1'), covar=tensor([0.0836, 0.0854, 0.0796, 0.1077, 0.1272, 0.1475, 0.0683, 0.5918], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0243, 0.0276, 0.0289, 0.0332, 0.0280, 0.0300, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:40:45,475 INFO [finetune.py:976] (1/7) Epoch 14, batch 550, loss[loss=0.1685, simple_loss=0.2397, pruned_loss=0.04864, over 4909.00 frames. ], tot_loss[loss=0.1859, simple_loss=0.2525, pruned_loss=0.05965, over 894336.30 frames. ], batch size: 46, lr: 3.56e-03, grad_scale: 32.0 +2023-03-26 16:40:57,879 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75025.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:41:02,048 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7690, 1.6546, 1.5793, 1.7766, 1.3193, 3.5913, 1.3891, 1.8437], + device='cuda:1'), covar=tensor([0.3297, 0.2570, 0.2190, 0.2294, 0.1704, 0.0201, 0.2573, 0.1354], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0124, 0.0115, 0.0097, 0.0097, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:41:03,960 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.25 vs. limit=5.0 +2023-03-26 16:41:09,612 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.017e+02 1.564e+02 1.846e+02 2.204e+02 7.411e+02, threshold=3.691e+02, percent-clipped=3.0 +2023-03-26 16:41:32,953 INFO [finetune.py:976] (1/7) Epoch 14, batch 600, loss[loss=0.1969, simple_loss=0.2704, pruned_loss=0.06167, over 4751.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.2538, pruned_loss=0.06041, over 906698.75 frames. ], batch size: 54, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:41:59,068 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7383, 2.4223, 2.0528, 0.9761, 2.1759, 2.1358, 1.9137, 2.2328], + device='cuda:1'), covar=tensor([0.0855, 0.0881, 0.1658, 0.2151, 0.1620, 0.2157, 0.2090, 0.1078], + device='cuda:1'), in_proj_covar=tensor([0.0165, 0.0193, 0.0197, 0.0182, 0.0210, 0.0205, 0.0221, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:42:10,221 INFO [finetune.py:976] (1/7) Epoch 14, batch 650, loss[loss=0.172, simple_loss=0.249, pruned_loss=0.04749, over 4744.00 frames. ], tot_loss[loss=0.192, simple_loss=0.259, pruned_loss=0.06253, over 917331.25 frames. ], batch size: 54, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:42:30,911 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.139e+02 1.620e+02 1.922e+02 2.248e+02 3.855e+02, threshold=3.845e+02, percent-clipped=1.0 +2023-03-26 16:42:43,795 INFO [finetune.py:976] (1/7) Epoch 14, batch 700, loss[loss=0.1916, simple_loss=0.2707, pruned_loss=0.05624, over 4895.00 frames. ], tot_loss[loss=0.1949, simple_loss=0.2624, pruned_loss=0.06372, over 926934.58 frames. ], batch size: 35, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:43:16,883 INFO [finetune.py:976] (1/7) Epoch 14, batch 750, loss[loss=0.1816, simple_loss=0.2619, pruned_loss=0.05062, over 4840.00 frames. ], tot_loss[loss=0.1945, simple_loss=0.2625, pruned_loss=0.06319, over 933751.22 frames. ], batch size: 47, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:43:28,283 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=75228.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:43:37,703 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.931e+01 1.583e+02 1.834e+02 2.163e+02 4.783e+02, threshold=3.668e+02, percent-clipped=1.0 +2023-03-26 16:43:44,240 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75251.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:43:50,665 INFO [finetune.py:976] (1/7) Epoch 14, batch 800, loss[loss=0.1438, simple_loss=0.2221, pruned_loss=0.03269, over 4820.00 frames. ], tot_loss[loss=0.1923, simple_loss=0.2606, pruned_loss=0.062, over 938378.49 frames. ], batch size: 33, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:43:52,031 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8103, 1.6754, 1.5548, 1.9174, 2.1636, 1.8479, 1.5926, 1.4648], + device='cuda:1'), covar=tensor([0.2018, 0.2053, 0.1791, 0.1541, 0.1682, 0.1195, 0.2326, 0.1903], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0206, 0.0208, 0.0189, 0.0238, 0.0183, 0.0212, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:43:53,645 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75265.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:43:56,681 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=75270.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:44:09,552 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75289.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:44:16,557 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=75299.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:44:24,289 INFO [finetune.py:976] (1/7) Epoch 14, batch 850, loss[loss=0.1869, simple_loss=0.2488, pruned_loss=0.06252, over 4827.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2589, pruned_loss=0.06214, over 940871.30 frames. ], batch size: 30, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:44:30,266 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75320.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:44:37,473 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75331.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 16:44:44,944 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.704e+01 1.586e+02 1.985e+02 2.275e+02 3.825e+02, threshold=3.970e+02, percent-clipped=2.0 +2023-03-26 16:44:57,417 INFO [finetune.py:976] (1/7) Epoch 14, batch 900, loss[loss=0.1786, simple_loss=0.2483, pruned_loss=0.05447, over 4805.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2554, pruned_loss=0.0609, over 945003.28 frames. ], batch size: 45, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:45:44,925 INFO [finetune.py:976] (1/7) Epoch 14, batch 950, loss[loss=0.2072, simple_loss=0.2755, pruned_loss=0.06942, over 4829.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2546, pruned_loss=0.06111, over 946259.64 frames. ], batch size: 40, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:45:54,651 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0910, 1.8894, 1.4102, 0.5897, 1.6236, 1.7683, 1.5433, 1.7655], + device='cuda:1'), covar=tensor([0.0895, 0.0821, 0.1501, 0.1934, 0.1416, 0.2409, 0.2354, 0.0939], + device='cuda:1'), in_proj_covar=tensor([0.0165, 0.0194, 0.0197, 0.0183, 0.0211, 0.0206, 0.0221, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:46:05,784 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.942e+01 1.566e+02 1.871e+02 2.243e+02 4.539e+02, threshold=3.743e+02, percent-clipped=1.0 +2023-03-26 16:46:18,858 INFO [finetune.py:976] (1/7) Epoch 14, batch 1000, loss[loss=0.2362, simple_loss=0.2952, pruned_loss=0.08858, over 4918.00 frames. ], tot_loss[loss=0.1913, simple_loss=0.2582, pruned_loss=0.06225, over 948960.39 frames. ], batch size: 36, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:46:27,934 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8486, 1.3255, 1.8515, 1.8177, 1.6412, 1.5761, 1.7212, 1.6735], + device='cuda:1'), covar=tensor([0.3988, 0.4125, 0.3365, 0.3625, 0.4927, 0.3769, 0.4318, 0.3208], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0237, 0.0255, 0.0264, 0.0262, 0.0236, 0.0276, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:47:07,211 INFO [finetune.py:976] (1/7) Epoch 14, batch 1050, loss[loss=0.1527, simple_loss=0.2235, pruned_loss=0.04094, over 4881.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.2591, pruned_loss=0.06239, over 950520.76 frames. ], batch size: 32, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:47:31,085 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.109e+01 1.606e+02 2.003e+02 2.356e+02 8.983e+02, threshold=4.007e+02, percent-clipped=2.0 +2023-03-26 16:47:44,012 INFO [finetune.py:976] (1/7) Epoch 14, batch 1100, loss[loss=0.1612, simple_loss=0.2259, pruned_loss=0.0482, over 4736.00 frames. ], tot_loss[loss=0.1927, simple_loss=0.2604, pruned_loss=0.06253, over 952210.68 frames. ], batch size: 23, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:47:45,245 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0564, 1.9034, 1.7400, 1.9184, 1.3814, 4.5128, 1.7472, 2.4261], + device='cuda:1'), covar=tensor([0.3063, 0.2276, 0.2022, 0.2182, 0.1678, 0.0118, 0.2255, 0.1054], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0124, 0.0115, 0.0097, 0.0097, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:47:47,088 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75565.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:47:59,614 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75584.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:48:09,083 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2622, 3.7499, 3.9051, 4.1122, 4.0554, 3.7889, 4.3912, 1.3330], + device='cuda:1'), covar=tensor([0.0761, 0.0756, 0.0789, 0.0916, 0.1181, 0.1555, 0.0669, 0.5463], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0244, 0.0277, 0.0290, 0.0332, 0.0282, 0.0301, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:48:14,157 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 16:48:18,069 INFO [finetune.py:976] (1/7) Epoch 14, batch 1150, loss[loss=0.1486, simple_loss=0.2236, pruned_loss=0.03677, over 4788.00 frames. ], tot_loss[loss=0.1943, simple_loss=0.2623, pruned_loss=0.0632, over 953842.16 frames. ], batch size: 26, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:48:19,335 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=75613.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:48:24,054 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75620.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:48:28,194 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75626.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 16:48:38,760 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.806e+01 1.586e+02 1.986e+02 2.337e+02 5.787e+02, threshold=3.972e+02, percent-clipped=2.0 +2023-03-26 16:48:51,183 INFO [finetune.py:976] (1/7) Epoch 14, batch 1200, loss[loss=0.1405, simple_loss=0.2246, pruned_loss=0.02818, over 4899.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.2596, pruned_loss=0.06212, over 956236.13 frames. ], batch size: 35, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:48:52,369 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6659, 1.4662, 1.9934, 1.3054, 1.6720, 1.8734, 1.4511, 2.0830], + device='cuda:1'), covar=tensor([0.1190, 0.2163, 0.1064, 0.1500, 0.0854, 0.1093, 0.2867, 0.0800], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0204, 0.0191, 0.0189, 0.0175, 0.0213, 0.0215, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:48:56,394 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=75668.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:49:03,630 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4234, 2.2867, 2.7708, 1.6361, 2.4479, 2.5727, 1.9263, 2.8706], + device='cuda:1'), covar=tensor([0.1386, 0.1801, 0.1614, 0.2422, 0.0906, 0.1644, 0.2756, 0.0911], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0204, 0.0191, 0.0189, 0.0176, 0.0214, 0.0215, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:49:15,491 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.64 vs. limit=2.0 +2023-03-26 16:49:24,700 INFO [finetune.py:976] (1/7) Epoch 14, batch 1250, loss[loss=0.1758, simple_loss=0.2509, pruned_loss=0.05038, over 4836.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2558, pruned_loss=0.06067, over 957224.16 frames. ], batch size: 30, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:49:27,177 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.77 vs. limit=2.0 +2023-03-26 16:49:45,240 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.316e+01 1.497e+02 1.829e+02 2.293e+02 4.240e+02, threshold=3.659e+02, percent-clipped=2.0 +2023-03-26 16:49:56,724 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8265, 1.2402, 1.8822, 1.7407, 1.6139, 1.5348, 1.7204, 1.7428], + device='cuda:1'), covar=tensor([0.4063, 0.4073, 0.3339, 0.3677, 0.4676, 0.3534, 0.4822, 0.3186], + device='cuda:1'), in_proj_covar=tensor([0.0243, 0.0238, 0.0256, 0.0266, 0.0263, 0.0237, 0.0277, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:49:57,806 INFO [finetune.py:976] (1/7) Epoch 14, batch 1300, loss[loss=0.1662, simple_loss=0.238, pruned_loss=0.04721, over 4751.00 frames. ], tot_loss[loss=0.1862, simple_loss=0.2529, pruned_loss=0.05974, over 957365.21 frames. ], batch size: 27, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:50:14,942 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7208, 1.6483, 1.5762, 1.5990, 1.0664, 3.5392, 1.3361, 1.8815], + device='cuda:1'), covar=tensor([0.3183, 0.2409, 0.2052, 0.2420, 0.1854, 0.0181, 0.2751, 0.1281], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0120, 0.0124, 0.0115, 0.0097, 0.0097, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:50:31,734 INFO [finetune.py:976] (1/7) Epoch 14, batch 1350, loss[loss=0.1911, simple_loss=0.2426, pruned_loss=0.0698, over 4142.00 frames. ], tot_loss[loss=0.1876, simple_loss=0.2539, pruned_loss=0.0606, over 957246.99 frames. ], batch size: 18, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:50:35,361 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7840, 1.1101, 1.8168, 1.7454, 1.5891, 1.4997, 1.6788, 1.6617], + device='cuda:1'), covar=tensor([0.3773, 0.4121, 0.3356, 0.3726, 0.4609, 0.3480, 0.4302, 0.3111], + device='cuda:1'), in_proj_covar=tensor([0.0243, 0.0238, 0.0256, 0.0266, 0.0263, 0.0236, 0.0276, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:50:52,070 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4366, 1.3081, 1.5165, 0.8733, 1.4470, 1.4914, 1.4895, 1.1915], + device='cuda:1'), covar=tensor([0.0550, 0.0760, 0.0617, 0.0897, 0.0812, 0.0706, 0.0578, 0.1226], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0132, 0.0141, 0.0123, 0.0123, 0.0141, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:51:07,733 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.659e+01 1.636e+02 1.953e+02 2.256e+02 6.748e+02, threshold=3.906e+02, percent-clipped=1.0 +2023-03-26 16:51:19,718 INFO [finetune.py:976] (1/7) Epoch 14, batch 1400, loss[loss=0.1973, simple_loss=0.2761, pruned_loss=0.05922, over 4741.00 frames. ], tot_loss[loss=0.1887, simple_loss=0.2559, pruned_loss=0.0608, over 958696.45 frames. ], batch size: 59, lr: 3.56e-03, grad_scale: 16.0 +2023-03-26 16:51:35,779 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75884.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:51:53,107 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0741, 1.9932, 1.6601, 2.0606, 1.8186, 1.8140, 1.8547, 2.5654], + device='cuda:1'), covar=tensor([0.4021, 0.4993, 0.3699, 0.4227, 0.4526, 0.2684, 0.4663, 0.1796], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0260, 0.0226, 0.0278, 0.0247, 0.0214, 0.0249, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:51:53,578 INFO [finetune.py:976] (1/7) Epoch 14, batch 1450, loss[loss=0.1675, simple_loss=0.2474, pruned_loss=0.0438, over 4804.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2588, pruned_loss=0.06164, over 958124.69 frames. ], batch size: 40, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:52:08,066 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75926.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 16:52:17,408 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=75932.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:52:27,770 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.094e+02 1.664e+02 1.939e+02 2.609e+02 1.085e+03, threshold=3.877e+02, percent-clipped=5.0 +2023-03-26 16:52:44,458 INFO [finetune.py:976] (1/7) Epoch 14, batch 1500, loss[loss=0.1765, simple_loss=0.2595, pruned_loss=0.04671, over 4832.00 frames. ], tot_loss[loss=0.192, simple_loss=0.2603, pruned_loss=0.06187, over 959346.19 frames. ], batch size: 49, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:52:52,970 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=75974.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:53:14,332 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.55 vs. limit=5.0 +2023-03-26 16:53:19,432 INFO [finetune.py:976] (1/7) Epoch 14, batch 1550, loss[loss=0.1676, simple_loss=0.2333, pruned_loss=0.05098, over 4848.00 frames. ], tot_loss[loss=0.1916, simple_loss=0.2598, pruned_loss=0.06173, over 957645.30 frames. ], batch size: 44, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:53:33,199 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6503, 2.3022, 2.9021, 1.7551, 2.7054, 2.8371, 2.1232, 3.1070], + device='cuda:1'), covar=tensor([0.1383, 0.1893, 0.1470, 0.2352, 0.0838, 0.1490, 0.2594, 0.0765], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0204, 0.0192, 0.0189, 0.0176, 0.0213, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:53:40,209 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.569e+01 1.489e+02 1.761e+02 2.263e+02 3.823e+02, threshold=3.522e+02, percent-clipped=0.0 +2023-03-26 16:53:42,218 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 16:53:53,247 INFO [finetune.py:976] (1/7) Epoch 14, batch 1600, loss[loss=0.1394, simple_loss=0.2194, pruned_loss=0.02964, over 4765.00 frames. ], tot_loss[loss=0.1883, simple_loss=0.2563, pruned_loss=0.06012, over 956780.37 frames. ], batch size: 27, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:54:14,228 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.4283, 2.9218, 2.6470, 1.3861, 2.9368, 2.4341, 2.3569, 2.5112], + device='cuda:1'), covar=tensor([0.0913, 0.0954, 0.1780, 0.2309, 0.1630, 0.2084, 0.2109, 0.1241], + device='cuda:1'), in_proj_covar=tensor([0.0166, 0.0196, 0.0198, 0.0184, 0.0212, 0.0206, 0.0220, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:54:19,132 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.53 vs. limit=5.0 +2023-03-26 16:54:26,639 INFO [finetune.py:976] (1/7) Epoch 14, batch 1650, loss[loss=0.1558, simple_loss=0.2291, pruned_loss=0.04127, over 4911.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2542, pruned_loss=0.0597, over 956805.93 frames. ], batch size: 37, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:54:41,410 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.2788, 4.5998, 4.8059, 5.1147, 5.0651, 4.7626, 5.4344, 1.6495], + device='cuda:1'), covar=tensor([0.0624, 0.0761, 0.0789, 0.0713, 0.1014, 0.1469, 0.0476, 0.5487], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0244, 0.0277, 0.0291, 0.0331, 0.0282, 0.0302, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:54:47,811 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.271e+01 1.580e+02 1.872e+02 2.187e+02 4.946e+02, threshold=3.744e+02, percent-clipped=3.0 +2023-03-26 16:54:48,112 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.73 vs. limit=5.0 +2023-03-26 16:55:00,263 INFO [finetune.py:976] (1/7) Epoch 14, batch 1700, loss[loss=0.1686, simple_loss=0.2357, pruned_loss=0.05077, over 4784.00 frames. ], tot_loss[loss=0.185, simple_loss=0.252, pruned_loss=0.05902, over 957962.17 frames. ], batch size: 26, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:55:04,624 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 16:55:34,234 INFO [finetune.py:976] (1/7) Epoch 14, batch 1750, loss[loss=0.2548, simple_loss=0.3037, pruned_loss=0.103, over 4843.00 frames. ], tot_loss[loss=0.1876, simple_loss=0.2545, pruned_loss=0.06033, over 957405.86 frames. ], batch size: 49, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:55:34,374 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0764, 1.8343, 1.6532, 2.0182, 2.7377, 2.0301, 1.9015, 1.5135], + device='cuda:1'), covar=tensor([0.2214, 0.2116, 0.1928, 0.1727, 0.1707, 0.1153, 0.2179, 0.2019], + device='cuda:1'), in_proj_covar=tensor([0.0237, 0.0205, 0.0209, 0.0188, 0.0239, 0.0183, 0.0212, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:55:38,095 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.45 vs. limit=2.0 +2023-03-26 16:55:42,827 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2101, 1.8791, 2.2350, 2.1261, 1.8836, 1.8924, 2.1030, 2.0314], + device='cuda:1'), covar=tensor([0.4434, 0.4681, 0.3613, 0.4198, 0.5710, 0.4277, 0.5357, 0.3690], + device='cuda:1'), in_proj_covar=tensor([0.0243, 0.0239, 0.0257, 0.0266, 0.0264, 0.0238, 0.0278, 0.0236], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 16:55:55,253 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.088e+02 1.620e+02 1.973e+02 2.349e+02 4.562e+02, threshold=3.945e+02, percent-clipped=4.0 +2023-03-26 16:56:17,745 INFO [finetune.py:976] (1/7) Epoch 14, batch 1800, loss[loss=0.2095, simple_loss=0.2789, pruned_loss=0.07006, over 4776.00 frames. ], tot_loss[loss=0.1901, simple_loss=0.258, pruned_loss=0.06115, over 954531.53 frames. ], batch size: 29, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:56:27,648 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=76269.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 16:56:47,390 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=76294.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:56:58,604 INFO [finetune.py:976] (1/7) Epoch 14, batch 1850, loss[loss=0.2194, simple_loss=0.281, pruned_loss=0.0789, over 4892.00 frames. ], tot_loss[loss=0.1917, simple_loss=0.2593, pruned_loss=0.062, over 954845.58 frames. ], batch size: 35, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:57:06,616 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=76323.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 16:57:11,382 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=76330.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 16:57:19,099 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.107e+02 1.514e+02 1.907e+02 2.299e+02 3.483e+02, threshold=3.815e+02, percent-clipped=0.0 +2023-03-26 16:57:35,309 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=76355.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 16:57:38,872 INFO [finetune.py:976] (1/7) Epoch 14, batch 1900, loss[loss=0.1725, simple_loss=0.2441, pruned_loss=0.05043, over 4825.00 frames. ], tot_loss[loss=0.1919, simple_loss=0.2598, pruned_loss=0.06198, over 954103.70 frames. ], batch size: 30, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:57:51,179 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5515, 1.3997, 1.2697, 1.4611, 1.8016, 1.7538, 1.5560, 1.2999], + device='cuda:1'), covar=tensor([0.0319, 0.0337, 0.0587, 0.0359, 0.0204, 0.0424, 0.0337, 0.0455], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0110, 0.0141, 0.0114, 0.0101, 0.0106, 0.0096, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.3246e-05, 8.4799e-05, 1.1190e-04, 8.8473e-05, 7.8749e-05, 7.8704e-05, + 7.2519e-05, 8.3123e-05], device='cuda:1') +2023-03-26 16:57:57,164 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=76384.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 16:58:15,528 INFO [finetune.py:976] (1/7) Epoch 14, batch 1950, loss[loss=0.2042, simple_loss=0.261, pruned_loss=0.07374, over 4826.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2575, pruned_loss=0.06055, over 955531.32 frames. ], batch size: 30, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:58:23,932 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8430, 1.6860, 1.6322, 1.7870, 1.1397, 3.8174, 1.6112, 2.1237], + device='cuda:1'), covar=tensor([0.3320, 0.2448, 0.2093, 0.2407, 0.1912, 0.0174, 0.2461, 0.1223], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0120, 0.0124, 0.0115, 0.0097, 0.0097, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 16:58:30,420 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7364, 1.3662, 2.1523, 3.3717, 2.3241, 2.3974, 1.0456, 2.6859], + device='cuda:1'), covar=tensor([0.1815, 0.1733, 0.1397, 0.0560, 0.0853, 0.1417, 0.2088, 0.0553], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0164, 0.0100, 0.0138, 0.0126, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 16:58:35,771 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.938e+01 1.459e+02 1.786e+02 2.082e+02 3.715e+02, threshold=3.572e+02, percent-clipped=0.0 +2023-03-26 16:58:49,143 INFO [finetune.py:976] (1/7) Epoch 14, batch 2000, loss[loss=0.2072, simple_loss=0.267, pruned_loss=0.07373, over 4771.00 frames. ], tot_loss[loss=0.188, simple_loss=0.2553, pruned_loss=0.06034, over 955795.70 frames. ], batch size: 29, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:59:22,189 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4121, 1.2435, 1.2394, 1.3257, 1.6415, 1.5422, 1.3802, 1.2067], + device='cuda:1'), covar=tensor([0.0334, 0.0359, 0.0580, 0.0320, 0.0238, 0.0476, 0.0366, 0.0497], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0141, 0.0113, 0.0101, 0.0106, 0.0096, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.3122e-05, 8.4542e-05, 1.1162e-04, 8.8179e-05, 7.8642e-05, 7.8399e-05, + 7.2321e-05, 8.2873e-05], device='cuda:1') +2023-03-26 16:59:22,663 INFO [finetune.py:976] (1/7) Epoch 14, batch 2050, loss[loss=0.181, simple_loss=0.2373, pruned_loss=0.06231, over 4807.00 frames. ], tot_loss[loss=0.1852, simple_loss=0.2516, pruned_loss=0.05938, over 955150.11 frames. ], batch size: 25, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 16:59:42,964 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.833e+01 1.460e+02 1.798e+02 2.157e+02 5.136e+02, threshold=3.595e+02, percent-clipped=3.0 +2023-03-26 16:59:56,036 INFO [finetune.py:976] (1/7) Epoch 14, batch 2100, loss[loss=0.161, simple_loss=0.2253, pruned_loss=0.04835, over 4710.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2512, pruned_loss=0.05883, over 955511.49 frames. ], batch size: 23, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:00:29,580 INFO [finetune.py:976] (1/7) Epoch 14, batch 2150, loss[loss=0.1756, simple_loss=0.2444, pruned_loss=0.05344, over 4186.00 frames. ], tot_loss[loss=0.1871, simple_loss=0.2548, pruned_loss=0.05967, over 956025.84 frames. ], batch size: 65, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:00:38,760 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76625.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 17:00:50,408 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.070e+02 1.680e+02 1.855e+02 2.274e+02 3.771e+02, threshold=3.710e+02, percent-clipped=2.0 +2023-03-26 17:00:55,366 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76650.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:01:02,483 INFO [finetune.py:976] (1/7) Epoch 14, batch 2200, loss[loss=0.208, simple_loss=0.264, pruned_loss=0.07601, over 4747.00 frames. ], tot_loss[loss=0.1874, simple_loss=0.2557, pruned_loss=0.05949, over 954810.78 frames. ], batch size: 54, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:01:04,317 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5819, 1.6531, 1.3816, 1.6542, 2.0785, 1.9407, 1.7000, 1.3882], + device='cuda:1'), covar=tensor([0.0350, 0.0300, 0.0627, 0.0313, 0.0188, 0.0410, 0.0302, 0.0436], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0141, 0.0113, 0.0101, 0.0106, 0.0096, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.3127e-05, 8.4621e-05, 1.1185e-04, 8.8257e-05, 7.8480e-05, 7.8400e-05, + 7.2463e-05, 8.2778e-05], device='cuda:1') +2023-03-26 17:01:21,647 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76679.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 17:01:57,530 INFO [finetune.py:976] (1/7) Epoch 14, batch 2250, loss[loss=0.1859, simple_loss=0.2351, pruned_loss=0.06835, over 4041.00 frames. ], tot_loss[loss=0.1886, simple_loss=0.2571, pruned_loss=0.06, over 954880.01 frames. ], batch size: 17, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:02:18,736 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.580e+02 1.848e+02 2.151e+02 3.368e+02, threshold=3.695e+02, percent-clipped=0.0 +2023-03-26 17:02:31,272 INFO [finetune.py:976] (1/7) Epoch 14, batch 2300, loss[loss=0.1847, simple_loss=0.2344, pruned_loss=0.06753, over 4018.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2581, pruned_loss=0.06031, over 955751.60 frames. ], batch size: 17, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:03:06,755 INFO [finetune.py:976] (1/7) Epoch 14, batch 2350, loss[loss=0.1951, simple_loss=0.2628, pruned_loss=0.06372, over 4894.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2546, pruned_loss=0.05916, over 954680.83 frames. ], batch size: 32, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:03:25,947 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=76839.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:03:28,174 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.075e+02 1.531e+02 1.863e+02 2.217e+02 4.521e+02, threshold=3.725e+02, percent-clipped=2.0 +2023-03-26 17:03:40,658 INFO [finetune.py:976] (1/7) Epoch 14, batch 2400, loss[loss=0.176, simple_loss=0.2403, pruned_loss=0.0558, over 4798.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2517, pruned_loss=0.05806, over 957433.29 frames. ], batch size: 29, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:03:55,832 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-03-26 17:04:06,926 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=76900.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:04:14,026 INFO [finetune.py:976] (1/7) Epoch 14, batch 2450, loss[loss=0.1969, simple_loss=0.2635, pruned_loss=0.06512, over 4753.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2483, pruned_loss=0.05675, over 954939.78 frames. ], batch size: 59, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:04:23,096 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=76925.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 17:04:34,661 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.650e+01 1.627e+02 1.914e+02 2.448e+02 4.488e+02, threshold=3.829e+02, percent-clipped=3.0 +2023-03-26 17:04:39,568 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8738, 1.1975, 1.6808, 1.7001, 1.5653, 1.5421, 1.6310, 1.6350], + device='cuda:1'), covar=tensor([0.4836, 0.4872, 0.4256, 0.4496, 0.5682, 0.4392, 0.5524, 0.4148], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0238, 0.0256, 0.0266, 0.0263, 0.0236, 0.0277, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:04:40,118 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=76950.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:04:47,674 INFO [finetune.py:976] (1/7) Epoch 14, batch 2500, loss[loss=0.2018, simple_loss=0.2684, pruned_loss=0.06763, over 4942.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2518, pruned_loss=0.05852, over 954092.26 frames. ], batch size: 33, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:04:55,524 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=76973.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 17:04:59,670 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=76979.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 17:05:12,270 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=76998.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:05:12,303 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4762, 1.2668, 1.8984, 3.1997, 2.0516, 2.1652, 0.8501, 2.5714], + device='cuda:1'), covar=tensor([0.1959, 0.1735, 0.1436, 0.0611, 0.0935, 0.1524, 0.2037, 0.0603], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0163, 0.0100, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 17:05:21,356 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-03-26 17:05:21,742 INFO [finetune.py:976] (1/7) Epoch 14, batch 2550, loss[loss=0.2023, simple_loss=0.2662, pruned_loss=0.06916, over 4813.00 frames. ], tot_loss[loss=0.187, simple_loss=0.2551, pruned_loss=0.05948, over 955008.27 frames. ], batch size: 51, lr: 3.55e-03, grad_scale: 16.0 +2023-03-26 17:05:30,855 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1317, 2.1290, 2.0457, 1.4672, 2.2001, 2.2599, 2.1583, 1.8942], + device='cuda:1'), covar=tensor([0.0656, 0.0689, 0.0763, 0.0888, 0.0599, 0.0769, 0.0706, 0.1079], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0132, 0.0142, 0.0123, 0.0124, 0.0141, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:05:32,003 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=77027.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 17:05:42,424 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.113e+02 1.599e+02 1.863e+02 2.531e+02 5.028e+02, threshold=3.725e+02, percent-clipped=2.0 +2023-03-26 17:05:55,387 INFO [finetune.py:976] (1/7) Epoch 14, batch 2600, loss[loss=0.2197, simple_loss=0.2905, pruned_loss=0.07447, over 4826.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2566, pruned_loss=0.05961, over 957553.78 frames. ], batch size: 39, lr: 3.55e-03, grad_scale: 32.0 +2023-03-26 17:05:59,779 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4616, 1.5509, 1.2681, 1.5243, 1.9337, 1.6807, 1.5034, 1.3275], + device='cuda:1'), covar=tensor([0.0324, 0.0283, 0.0581, 0.0274, 0.0189, 0.0533, 0.0275, 0.0368], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0141, 0.0113, 0.0101, 0.0106, 0.0096, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.3136e-05, 8.4297e-05, 1.1158e-04, 8.7759e-05, 7.8453e-05, 7.8424e-05, + 7.2521e-05, 8.2393e-05], device='cuda:1') +2023-03-26 17:06:12,033 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8049, 1.6959, 1.4440, 1.8721, 2.3290, 1.9215, 1.6077, 1.4096], + device='cuda:1'), covar=tensor([0.2180, 0.2041, 0.2003, 0.1654, 0.1844, 0.1209, 0.2380, 0.1965], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0208, 0.0211, 0.0191, 0.0241, 0.0184, 0.0215, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:06:16,139 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.9891, 3.4593, 3.5771, 3.8450, 3.7439, 3.5338, 4.0954, 1.3071], + device='cuda:1'), covar=tensor([0.0818, 0.0968, 0.0885, 0.1015, 0.1203, 0.1543, 0.0799, 0.5344], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0244, 0.0278, 0.0293, 0.0333, 0.0283, 0.0302, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:06:18,019 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77095.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:06:30,495 INFO [finetune.py:976] (1/7) Epoch 14, batch 2650, loss[loss=0.2101, simple_loss=0.2849, pruned_loss=0.06759, over 4814.00 frames. ], tot_loss[loss=0.1896, simple_loss=0.2584, pruned_loss=0.06042, over 956560.36 frames. ], batch size: 38, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:07:08,841 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.161e+02 1.606e+02 1.948e+02 2.371e+02 3.624e+02, threshold=3.895e+02, percent-clipped=0.0 +2023-03-26 17:07:22,877 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77156.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 17:07:26,327 INFO [finetune.py:976] (1/7) Epoch 14, batch 2700, loss[loss=0.2142, simple_loss=0.2885, pruned_loss=0.06995, over 4765.00 frames. ], tot_loss[loss=0.189, simple_loss=0.2578, pruned_loss=0.06007, over 956070.12 frames. ], batch size: 54, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:07:57,702 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77195.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:08:07,956 INFO [finetune.py:976] (1/7) Epoch 14, batch 2750, loss[loss=0.2165, simple_loss=0.2765, pruned_loss=0.07819, over 4836.00 frames. ], tot_loss[loss=0.1859, simple_loss=0.2543, pruned_loss=0.05877, over 957716.27 frames. ], batch size: 49, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:08:08,061 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77211.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:08:28,394 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.127e+02 1.514e+02 1.871e+02 2.163e+02 3.576e+02, threshold=3.742e+02, percent-clipped=0.0 +2023-03-26 17:08:40,953 INFO [finetune.py:976] (1/7) Epoch 14, batch 2800, loss[loss=0.1426, simple_loss=0.2139, pruned_loss=0.03569, over 4830.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2508, pruned_loss=0.05766, over 956690.21 frames. ], batch size: 49, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:08:48,257 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77272.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:08:48,850 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77273.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:08:52,180 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77277.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:09:14,628 INFO [finetune.py:976] (1/7) Epoch 14, batch 2850, loss[loss=0.2272, simple_loss=0.3043, pruned_loss=0.07505, over 4835.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2502, pruned_loss=0.05816, over 954591.08 frames. ], batch size: 47, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:09:19,614 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4096, 1.4970, 2.0551, 1.8657, 1.7169, 3.7506, 1.3259, 1.6802], + device='cuda:1'), covar=tensor([0.0967, 0.1689, 0.1244, 0.0921, 0.1412, 0.0235, 0.1487, 0.1614], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0080, 0.0073, 0.0077, 0.0091, 0.0080, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 17:09:26,880 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-03-26 17:09:29,753 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77334.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:09:32,166 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77338.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:09:34,910 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.563e+02 1.884e+02 2.320e+02 5.201e+02, threshold=3.768e+02, percent-clipped=2.0 +2023-03-26 17:09:47,962 INFO [finetune.py:976] (1/7) Epoch 14, batch 2900, loss[loss=0.1801, simple_loss=0.2523, pruned_loss=0.05396, over 4778.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2549, pruned_loss=0.06041, over 956285.09 frames. ], batch size: 28, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:10:14,875 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9182, 1.6875, 1.5545, 1.3232, 1.6736, 1.6960, 1.6200, 2.2017], + device='cuda:1'), covar=tensor([0.4071, 0.4368, 0.3407, 0.4054, 0.3720, 0.2441, 0.3810, 0.1901], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0261, 0.0226, 0.0279, 0.0248, 0.0215, 0.0250, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:10:21,778 INFO [finetune.py:976] (1/7) Epoch 14, batch 2950, loss[loss=0.2289, simple_loss=0.294, pruned_loss=0.08191, over 4747.00 frames. ], tot_loss[loss=0.1908, simple_loss=0.2585, pruned_loss=0.06152, over 957876.43 frames. ], batch size: 54, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:10:23,127 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9532, 1.8518, 1.6698, 1.5272, 2.0189, 1.6977, 1.8609, 1.9638], + device='cuda:1'), covar=tensor([0.1445, 0.1968, 0.3168, 0.2401, 0.2477, 0.1675, 0.2813, 0.1826], + device='cuda:1'), in_proj_covar=tensor([0.0179, 0.0187, 0.0232, 0.0252, 0.0243, 0.0198, 0.0212, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:10:41,985 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.681e+01 1.651e+02 1.924e+02 2.203e+02 4.754e+02, threshold=3.848e+02, percent-clipped=2.0 +2023-03-26 17:10:48,018 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77451.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 17:10:54,978 INFO [finetune.py:976] (1/7) Epoch 14, batch 3000, loss[loss=0.1776, simple_loss=0.2587, pruned_loss=0.04821, over 4815.00 frames. ], tot_loss[loss=0.192, simple_loss=0.2598, pruned_loss=0.06208, over 956098.78 frames. ], batch size: 33, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:10:54,978 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 17:11:02,158 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0822, 1.7896, 1.6947, 1.7196, 1.7402, 1.7531, 1.7475, 2.4728], + device='cuda:1'), covar=tensor([0.3937, 0.4724, 0.3354, 0.3953, 0.4397, 0.2425, 0.4228, 0.1815], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0262, 0.0227, 0.0279, 0.0249, 0.0216, 0.0251, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:11:02,992 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8188, 3.3841, 3.4809, 3.7161, 3.5583, 3.4560, 3.8999, 1.3167], + device='cuda:1'), covar=tensor([0.0933, 0.1021, 0.1002, 0.0912, 0.1636, 0.1638, 0.0865, 0.5110], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0243, 0.0276, 0.0292, 0.0332, 0.0283, 0.0299, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:11:09,361 INFO [finetune.py:1010] (1/7) Epoch 14, validation: loss=0.1563, simple_loss=0.2268, pruned_loss=0.04293, over 2265189.00 frames. +2023-03-26 17:11:09,361 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 17:11:15,694 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-03-26 17:11:34,064 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77495.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:11:44,250 INFO [finetune.py:976] (1/7) Epoch 14, batch 3050, loss[loss=0.2078, simple_loss=0.2583, pruned_loss=0.07871, over 4067.00 frames. ], tot_loss[loss=0.191, simple_loss=0.2592, pruned_loss=0.06144, over 955440.90 frames. ], batch size: 65, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:12:13,224 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.117e+02 1.567e+02 1.800e+02 2.244e+02 5.193e+02, threshold=3.600e+02, percent-clipped=2.0 +2023-03-26 17:12:13,930 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=77543.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:12:35,684 INFO [finetune.py:976] (1/7) Epoch 14, batch 3100, loss[loss=0.1514, simple_loss=0.2244, pruned_loss=0.03919, over 4882.00 frames. ], tot_loss[loss=0.188, simple_loss=0.2566, pruned_loss=0.05973, over 955681.12 frames. ], batch size: 43, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:12:39,907 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77567.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:13:22,406 INFO [finetune.py:976] (1/7) Epoch 14, batch 3150, loss[loss=0.1306, simple_loss=0.2016, pruned_loss=0.02985, over 4819.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2553, pruned_loss=0.06029, over 955893.17 frames. ], batch size: 39, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:13:24,479 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 17:13:25,544 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77616.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:13:35,434 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77629.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:13:37,875 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77633.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:13:43,292 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.167e+02 1.643e+02 1.968e+02 2.398e+02 4.679e+02, threshold=3.936e+02, percent-clipped=3.0 +2023-03-26 17:13:52,405 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.98 vs. limit=5.0 +2023-03-26 17:13:56,367 INFO [finetune.py:976] (1/7) Epoch 14, batch 3200, loss[loss=0.2522, simple_loss=0.3141, pruned_loss=0.09511, over 4153.00 frames. ], tot_loss[loss=0.185, simple_loss=0.252, pruned_loss=0.059, over 955634.59 frames. ], batch size: 66, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:14:06,643 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77677.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:14:06,655 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77677.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:14:13,957 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-03-26 17:14:22,279 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.82 vs. limit=5.0 +2023-03-26 17:14:29,511 INFO [finetune.py:976] (1/7) Epoch 14, batch 3250, loss[loss=0.1862, simple_loss=0.2544, pruned_loss=0.05902, over 4699.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2513, pruned_loss=0.05872, over 956206.34 frames. ], batch size: 23, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:14:47,325 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77738.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:14:49,600 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.063e+02 1.623e+02 1.885e+02 2.287e+02 7.301e+02, threshold=3.769e+02, percent-clipped=4.0 +2023-03-26 17:14:55,615 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77751.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 17:14:58,680 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8514, 1.5630, 2.1523, 1.5110, 2.0353, 2.1951, 1.5894, 2.2328], + device='cuda:1'), covar=tensor([0.1434, 0.2335, 0.1504, 0.2000, 0.0878, 0.1382, 0.2721, 0.0884], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0204, 0.0191, 0.0189, 0.0176, 0.0212, 0.0215, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:15:02,067 INFO [finetune.py:976] (1/7) Epoch 14, batch 3300, loss[loss=0.1782, simple_loss=0.2472, pruned_loss=0.05463, over 4757.00 frames. ], tot_loss[loss=0.1872, simple_loss=0.2546, pruned_loss=0.05985, over 956927.42 frames. ], batch size: 27, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:15:15,340 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8629, 3.3638, 3.5310, 3.7288, 3.6145, 3.3940, 3.9443, 1.1808], + device='cuda:1'), covar=tensor([0.0967, 0.1026, 0.0936, 0.1140, 0.1435, 0.1558, 0.0827, 0.5593], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0243, 0.0276, 0.0294, 0.0332, 0.0283, 0.0299, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:15:27,675 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=77799.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:15:35,621 INFO [finetune.py:976] (1/7) Epoch 14, batch 3350, loss[loss=0.1719, simple_loss=0.233, pruned_loss=0.05537, over 4764.00 frames. ], tot_loss[loss=0.1894, simple_loss=0.257, pruned_loss=0.06086, over 956499.36 frames. ], batch size: 23, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:15:57,262 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.144e+02 1.683e+02 1.958e+02 2.277e+02 5.309e+02, threshold=3.915e+02, percent-clipped=1.0 +2023-03-26 17:16:09,330 INFO [finetune.py:976] (1/7) Epoch 14, batch 3400, loss[loss=0.2234, simple_loss=0.2977, pruned_loss=0.07453, over 4795.00 frames. ], tot_loss[loss=0.1922, simple_loss=0.2601, pruned_loss=0.06215, over 956087.98 frames. ], batch size: 25, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:16:18,540 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77867.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:16:34,986 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5373, 1.3191, 1.9682, 3.1459, 2.0819, 2.0495, 0.8859, 2.5023], + device='cuda:1'), covar=tensor([0.1756, 0.1485, 0.1295, 0.0536, 0.0838, 0.1555, 0.1922, 0.0535], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0115, 0.0133, 0.0164, 0.0101, 0.0137, 0.0125, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 17:16:40,646 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9872, 1.2679, 0.8821, 1.8417, 2.2996, 1.7834, 1.4750, 1.8993], + device='cuda:1'), covar=tensor([0.1372, 0.1992, 0.2169, 0.1198, 0.1878, 0.1994, 0.1472, 0.1787], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0111, 0.0092, 0.0119, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 17:16:43,709 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5763, 1.6750, 2.1239, 1.9035, 1.8074, 4.0820, 1.4856, 1.9360], + device='cuda:1'), covar=tensor([0.0939, 0.1612, 0.1241, 0.0932, 0.1453, 0.0173, 0.1376, 0.1571], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0074, 0.0078, 0.0092, 0.0081, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 17:16:51,417 INFO [finetune.py:976] (1/7) Epoch 14, batch 3450, loss[loss=0.1907, simple_loss=0.2628, pruned_loss=0.05932, over 4807.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.2594, pruned_loss=0.06173, over 955004.34 frames. ], batch size: 41, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:16:53,841 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=77915.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:16:59,125 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7835, 1.6856, 1.4861, 1.8566, 2.2315, 1.8256, 1.5194, 1.4509], + device='cuda:1'), covar=tensor([0.1951, 0.1826, 0.1745, 0.1488, 0.1516, 0.1150, 0.2363, 0.1760], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0207, 0.0209, 0.0190, 0.0240, 0.0184, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:17:03,227 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77929.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:17:05,736 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77933.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:17:10,905 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2040, 2.0020, 2.6599, 1.8197, 2.2822, 2.5138, 1.9790, 2.5756], + device='cuda:1'), covar=tensor([0.1115, 0.1518, 0.1266, 0.1627, 0.0690, 0.0955, 0.1990, 0.0629], + device='cuda:1'), in_proj_covar=tensor([0.0196, 0.0204, 0.0192, 0.0190, 0.0176, 0.0214, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:17:11,406 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.160e+01 1.485e+02 1.825e+02 2.093e+02 4.049e+02, threshold=3.650e+02, percent-clipped=1.0 +2023-03-26 17:17:33,222 INFO [finetune.py:976] (1/7) Epoch 14, batch 3500, loss[loss=0.2079, simple_loss=0.2732, pruned_loss=0.07125, over 4811.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2564, pruned_loss=0.06056, over 955601.43 frames. ], batch size: 39, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:17:40,927 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77972.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:17:44,499 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=77977.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:17:46,349 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9434, 4.6236, 4.3761, 2.4189, 4.6552, 3.5806, 0.8967, 3.3389], + device='cuda:1'), covar=tensor([0.2377, 0.1401, 0.1297, 0.2946, 0.0716, 0.0728, 0.4486, 0.1244], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0160, 0.0127, 0.0156, 0.0122, 0.0145, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 17:17:46,952 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=77981.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:18:20,969 INFO [finetune.py:976] (1/7) Epoch 14, batch 3550, loss[loss=0.203, simple_loss=0.2703, pruned_loss=0.06779, over 4834.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2531, pruned_loss=0.05912, over 955252.33 frames. ], batch size: 33, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:18:28,314 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3174, 1.5001, 1.5858, 0.7694, 1.5286, 1.7755, 1.7825, 1.4257], + device='cuda:1'), covar=tensor([0.0874, 0.0658, 0.0483, 0.0523, 0.0439, 0.0521, 0.0309, 0.0672], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0154, 0.0123, 0.0131, 0.0131, 0.0128, 0.0143, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.4085e-05, 1.1219e-04, 8.8343e-05, 9.3823e-05, 9.3014e-05, 9.2521e-05, + 1.0405e-04, 1.0646e-04], device='cuda:1') +2023-03-26 17:18:29,456 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6879, 1.4525, 2.0522, 2.9776, 2.0495, 2.1874, 0.8862, 2.4431], + device='cuda:1'), covar=tensor([0.1565, 0.1352, 0.1149, 0.0579, 0.0848, 0.1221, 0.1879, 0.0570], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0163, 0.0100, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 17:18:31,787 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.10 vs. limit=2.0 +2023-03-26 17:18:34,113 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8759, 1.6508, 1.4729, 1.2779, 1.5904, 1.5835, 1.5870, 2.1801], + device='cuda:1'), covar=tensor([0.3579, 0.3472, 0.3050, 0.3532, 0.3662, 0.2197, 0.3520, 0.1616], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0226, 0.0277, 0.0247, 0.0213, 0.0249, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:18:35,858 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78033.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:18:41,165 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.505e+02 1.960e+02 2.472e+02 4.194e+02, threshold=3.920e+02, percent-clipped=4.0 +2023-03-26 17:18:42,559 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1710, 2.0145, 1.5944, 1.8917, 2.0844, 1.8124, 2.3239, 2.1554], + device='cuda:1'), covar=tensor([0.1301, 0.2260, 0.3235, 0.2753, 0.2682, 0.1646, 0.3806, 0.1699], + device='cuda:1'), in_proj_covar=tensor([0.0179, 0.0188, 0.0234, 0.0253, 0.0244, 0.0199, 0.0212, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:18:54,340 INFO [finetune.py:976] (1/7) Epoch 14, batch 3600, loss[loss=0.1576, simple_loss=0.2252, pruned_loss=0.04498, over 4741.00 frames. ], tot_loss[loss=0.1836, simple_loss=0.2504, pruned_loss=0.0584, over 954372.36 frames. ], batch size: 26, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:19:20,452 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.94 vs. limit=5.0 +2023-03-26 17:19:28,414 INFO [finetune.py:976] (1/7) Epoch 14, batch 3650, loss[loss=0.2052, simple_loss=0.2745, pruned_loss=0.06794, over 4809.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2526, pruned_loss=0.05929, over 954621.41 frames. ], batch size: 41, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:19:48,730 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.050e+02 1.711e+02 2.033e+02 2.406e+02 8.151e+02, threshold=4.067e+02, percent-clipped=4.0 +2023-03-26 17:20:02,233 INFO [finetune.py:976] (1/7) Epoch 14, batch 3700, loss[loss=0.2006, simple_loss=0.2645, pruned_loss=0.0684, over 4891.00 frames. ], tot_loss[loss=0.1869, simple_loss=0.2552, pruned_loss=0.05928, over 954948.48 frames. ], batch size: 35, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:20:35,985 INFO [finetune.py:976] (1/7) Epoch 14, batch 3750, loss[loss=0.2074, simple_loss=0.2663, pruned_loss=0.07428, over 4921.00 frames. ], tot_loss[loss=0.1875, simple_loss=0.2563, pruned_loss=0.05937, over 956364.45 frames. ], batch size: 33, lr: 3.54e-03, grad_scale: 32.0 +2023-03-26 17:20:37,451 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-26 17:20:49,435 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78232.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:20:55,808 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.186e+02 1.631e+02 1.949e+02 2.239e+02 4.423e+02, threshold=3.899e+02, percent-clipped=1.0 +2023-03-26 17:20:57,717 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7416, 4.4069, 4.1132, 2.3674, 4.4532, 3.4774, 0.9625, 2.9985], + device='cuda:1'), covar=tensor([0.2534, 0.1347, 0.1256, 0.2769, 0.0717, 0.0758, 0.4186, 0.1321], + device='cuda:1'), in_proj_covar=tensor([0.0148, 0.0170, 0.0157, 0.0125, 0.0154, 0.0120, 0.0143, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 17:21:01,253 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-03-26 17:21:08,216 INFO [finetune.py:976] (1/7) Epoch 14, batch 3800, loss[loss=0.2123, simple_loss=0.2888, pruned_loss=0.06795, over 4813.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2577, pruned_loss=0.06043, over 954417.13 frames. ], batch size: 33, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:21:15,894 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78272.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:21:28,609 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6723, 1.4394, 0.9131, 0.2299, 1.1727, 1.4644, 1.4207, 1.3054], + device='cuda:1'), covar=tensor([0.0894, 0.0835, 0.1457, 0.2049, 0.1501, 0.2536, 0.2232, 0.0977], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0195, 0.0199, 0.0183, 0.0213, 0.0207, 0.0222, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:21:31,058 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78293.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 17:21:47,135 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 17:21:49,333 INFO [finetune.py:976] (1/7) Epoch 14, batch 3850, loss[loss=0.188, simple_loss=0.2484, pruned_loss=0.06376, over 4921.00 frames. ], tot_loss[loss=0.1872, simple_loss=0.256, pruned_loss=0.05921, over 955280.18 frames. ], batch size: 33, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:21:55,866 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=78320.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:22:03,792 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78333.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:22:10,189 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.042e+02 1.518e+02 1.784e+02 2.158e+02 4.566e+02, threshold=3.568e+02, percent-clipped=2.0 +2023-03-26 17:22:22,149 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8498, 4.7267, 4.4547, 2.4151, 4.7691, 3.8456, 1.0300, 3.3826], + device='cuda:1'), covar=tensor([0.2530, 0.1503, 0.1425, 0.3173, 0.0776, 0.0720, 0.4781, 0.1338], + device='cuda:1'), in_proj_covar=tensor([0.0149, 0.0171, 0.0158, 0.0126, 0.0155, 0.0120, 0.0144, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 17:22:22,702 INFO [finetune.py:976] (1/7) Epoch 14, batch 3900, loss[loss=0.1766, simple_loss=0.2499, pruned_loss=0.05161, over 4910.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.2525, pruned_loss=0.0584, over 955090.68 frames. ], batch size: 36, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:22:35,916 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1651, 2.1365, 2.0626, 1.5947, 2.1828, 2.2400, 2.2870, 1.8620], + device='cuda:1'), covar=tensor([0.0505, 0.0561, 0.0664, 0.0905, 0.0613, 0.0642, 0.0508, 0.1000], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0132, 0.0141, 0.0123, 0.0123, 0.0140, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:22:45,628 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=78381.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:22:48,129 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.0018, 4.3779, 4.5455, 4.8285, 4.7349, 4.4465, 5.1575, 1.6353], + device='cuda:1'), covar=tensor([0.0679, 0.0909, 0.0663, 0.0912, 0.1124, 0.1448, 0.0441, 0.5596], + device='cuda:1'), in_proj_covar=tensor([0.0344, 0.0241, 0.0273, 0.0291, 0.0330, 0.0281, 0.0298, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:23:02,766 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.60 vs. limit=5.0 +2023-03-26 17:23:09,953 INFO [finetune.py:976] (1/7) Epoch 14, batch 3950, loss[loss=0.1871, simple_loss=0.2384, pruned_loss=0.06795, over 4864.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2496, pruned_loss=0.05766, over 955757.58 frames. ], batch size: 44, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:23:37,906 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.052e+02 1.587e+02 1.893e+02 2.259e+02 3.905e+02, threshold=3.786e+02, percent-clipped=1.0 +2023-03-26 17:23:50,392 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78460.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:23:50,893 INFO [finetune.py:976] (1/7) Epoch 14, batch 4000, loss[loss=0.1783, simple_loss=0.2444, pruned_loss=0.0561, over 4815.00 frames. ], tot_loss[loss=0.1828, simple_loss=0.2498, pruned_loss=0.0579, over 954258.71 frames. ], batch size: 25, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:24:24,848 INFO [finetune.py:976] (1/7) Epoch 14, batch 4050, loss[loss=0.1621, simple_loss=0.2187, pruned_loss=0.05269, over 4035.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2538, pruned_loss=0.0597, over 953779.48 frames. ], batch size: 17, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:24:31,537 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78521.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:24:45,460 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.285e+01 1.581e+02 1.919e+02 2.315e+02 3.488e+02, threshold=3.837e+02, percent-clipped=0.0 +2023-03-26 17:24:46,225 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9576, 1.8458, 1.7550, 2.0171, 2.4373, 2.0742, 1.6485, 1.6972], + device='cuda:1'), covar=tensor([0.2092, 0.2027, 0.1899, 0.1600, 0.1457, 0.1062, 0.2413, 0.1945], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0208, 0.0210, 0.0192, 0.0242, 0.0185, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:24:54,867 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78556.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:24:57,772 INFO [finetune.py:976] (1/7) Epoch 14, batch 4100, loss[loss=0.1776, simple_loss=0.2504, pruned_loss=0.0524, over 4926.00 frames. ], tot_loss[loss=0.1881, simple_loss=0.2559, pruned_loss=0.06022, over 953173.83 frames. ], batch size: 33, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:25:02,531 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4853, 1.3784, 1.7519, 1.8204, 1.5709, 3.3038, 1.3786, 1.5046], + device='cuda:1'), covar=tensor([0.1032, 0.1814, 0.1253, 0.0913, 0.1626, 0.0222, 0.1508, 0.1782], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0074, 0.0077, 0.0092, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 17:25:08,930 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9316, 1.8684, 1.4878, 1.7082, 1.8116, 1.7172, 1.7785, 2.4753], + device='cuda:1'), covar=tensor([0.4013, 0.4070, 0.3353, 0.3959, 0.3978, 0.2456, 0.3874, 0.1657], + device='cuda:1'), in_proj_covar=tensor([0.0284, 0.0259, 0.0224, 0.0276, 0.0246, 0.0213, 0.0248, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:25:16,500 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78588.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 17:25:31,551 INFO [finetune.py:976] (1/7) Epoch 14, batch 4150, loss[loss=0.1904, simple_loss=0.2772, pruned_loss=0.05183, over 4900.00 frames. ], tot_loss[loss=0.1894, simple_loss=0.2572, pruned_loss=0.06076, over 952394.93 frames. ], batch size: 37, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:25:35,308 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78617.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:25:36,539 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 17:25:52,377 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.171e+02 1.587e+02 1.869e+02 2.218e+02 3.242e+02, threshold=3.739e+02, percent-clipped=0.0 +2023-03-26 17:26:04,875 INFO [finetune.py:976] (1/7) Epoch 14, batch 4200, loss[loss=0.2284, simple_loss=0.2958, pruned_loss=0.0805, over 4827.00 frames. ], tot_loss[loss=0.1897, simple_loss=0.2577, pruned_loss=0.06082, over 953101.51 frames. ], batch size: 49, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:26:28,494 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9053, 3.4105, 3.4830, 1.8875, 3.7148, 2.8953, 1.1882, 2.5664], + device='cuda:1'), covar=tensor([0.2504, 0.1805, 0.1236, 0.2666, 0.0847, 0.0898, 0.3519, 0.1266], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0160, 0.0127, 0.0157, 0.0122, 0.0146, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 17:26:37,998 INFO [finetune.py:976] (1/7) Epoch 14, batch 4250, loss[loss=0.2063, simple_loss=0.2464, pruned_loss=0.08311, over 2917.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2562, pruned_loss=0.06044, over 951156.94 frames. ], batch size: 12, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:26:41,605 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5707, 1.4241, 2.1112, 3.1637, 2.1876, 2.3159, 0.8848, 2.5105], + device='cuda:1'), covar=tensor([0.1616, 0.1458, 0.1181, 0.0578, 0.0738, 0.1593, 0.1819, 0.0576], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0117, 0.0133, 0.0164, 0.0101, 0.0138, 0.0126, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 17:26:42,212 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9445, 1.6317, 2.3522, 1.5187, 2.0187, 2.2604, 1.6334, 2.3551], + device='cuda:1'), covar=tensor([0.1365, 0.2059, 0.1156, 0.2012, 0.0882, 0.1415, 0.2678, 0.0870], + device='cuda:1'), in_proj_covar=tensor([0.0195, 0.0205, 0.0191, 0.0191, 0.0176, 0.0213, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:27:05,885 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.090e+02 1.501e+02 1.722e+02 2.085e+02 5.543e+02, threshold=3.444e+02, percent-clipped=2.0 +2023-03-26 17:27:21,259 INFO [finetune.py:976] (1/7) Epoch 14, batch 4300, loss[loss=0.1792, simple_loss=0.2427, pruned_loss=0.05781, over 4750.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2533, pruned_loss=0.05976, over 950798.78 frames. ], batch size: 26, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:27:50,498 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 17:27:59,985 INFO [finetune.py:976] (1/7) Epoch 14, batch 4350, loss[loss=0.1586, simple_loss=0.2316, pruned_loss=0.04276, over 4817.00 frames. ], tot_loss[loss=0.1828, simple_loss=0.2494, pruned_loss=0.05808, over 950844.93 frames. ], batch size: 39, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:28:06,790 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78816.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:28:32,853 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2484, 2.7152, 2.5015, 1.3518, 2.7479, 2.2930, 2.0632, 2.5355], + device='cuda:1'), covar=tensor([0.1265, 0.0915, 0.1775, 0.2221, 0.1649, 0.2142, 0.2298, 0.1114], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0194, 0.0199, 0.0183, 0.0213, 0.0206, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:28:34,426 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.179e+02 1.620e+02 1.919e+02 2.276e+02 4.946e+02, threshold=3.838e+02, percent-clipped=3.0 +2023-03-26 17:28:54,136 INFO [finetune.py:976] (1/7) Epoch 14, batch 4400, loss[loss=0.2272, simple_loss=0.3001, pruned_loss=0.0772, over 4840.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2506, pruned_loss=0.0591, over 948907.77 frames. ], batch size: 47, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:29:00,738 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2667, 1.3109, 0.6423, 1.9732, 2.5126, 1.6920, 1.7735, 2.0072], + device='cuda:1'), covar=tensor([0.1376, 0.2169, 0.2225, 0.1137, 0.1758, 0.1958, 0.1421, 0.1873], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0111, 0.0092, 0.0119, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 17:29:12,615 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78888.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:29:27,938 INFO [finetune.py:976] (1/7) Epoch 14, batch 4450, loss[loss=0.2227, simple_loss=0.2816, pruned_loss=0.08195, over 4819.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2557, pruned_loss=0.06062, over 949719.83 frames. ], batch size: 40, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:29:28,595 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78912.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:29:44,121 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=78936.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:29:48,694 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.157e+02 1.623e+02 2.075e+02 2.454e+02 4.700e+02, threshold=4.150e+02, percent-clipped=3.0 +2023-03-26 17:29:51,180 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.53 vs. limit=5.0 +2023-03-26 17:30:01,641 INFO [finetune.py:976] (1/7) Epoch 14, batch 4500, loss[loss=0.1751, simple_loss=0.2405, pruned_loss=0.05484, over 4740.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2569, pruned_loss=0.06089, over 951268.00 frames. ], batch size: 54, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:30:04,138 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78965.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:30:34,870 INFO [finetune.py:976] (1/7) Epoch 14, batch 4550, loss[loss=0.1797, simple_loss=0.2474, pruned_loss=0.05603, over 4742.00 frames. ], tot_loss[loss=0.1912, simple_loss=0.2589, pruned_loss=0.06173, over 952928.85 frames. ], batch size: 59, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:30:44,021 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79026.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:30:54,526 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.032e+02 1.627e+02 1.835e+02 2.182e+02 4.419e+02, threshold=3.671e+02, percent-clipped=1.0 +2023-03-26 17:31:08,637 INFO [finetune.py:976] (1/7) Epoch 14, batch 4600, loss[loss=0.2192, simple_loss=0.2712, pruned_loss=0.08364, over 4863.00 frames. ], tot_loss[loss=0.1905, simple_loss=0.2588, pruned_loss=0.0611, over 953516.20 frames. ], batch size: 34, lr: 3.53e-03, grad_scale: 64.0 +2023-03-26 17:31:42,442 INFO [finetune.py:976] (1/7) Epoch 14, batch 4650, loss[loss=0.1356, simple_loss=0.214, pruned_loss=0.02856, over 4795.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2562, pruned_loss=0.06032, over 954019.70 frames. ], batch size: 26, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:31:45,565 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79116.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:32:02,945 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.138e+02 1.520e+02 1.886e+02 2.415e+02 4.094e+02, threshold=3.771e+02, percent-clipped=1.0 +2023-03-26 17:32:24,493 INFO [finetune.py:976] (1/7) Epoch 14, batch 4700, loss[loss=0.1847, simple_loss=0.2551, pruned_loss=0.05715, over 4755.00 frames. ], tot_loss[loss=0.1855, simple_loss=0.2528, pruned_loss=0.05911, over 954195.13 frames. ], batch size: 26, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:32:26,306 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=79164.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:32:44,433 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0200, 1.9813, 1.6409, 1.8867, 1.8021, 1.7924, 1.9168, 2.5790], + device='cuda:1'), covar=tensor([0.4177, 0.4386, 0.3358, 0.4184, 0.4200, 0.2476, 0.4026, 0.1861], + device='cuda:1'), in_proj_covar=tensor([0.0284, 0.0259, 0.0224, 0.0276, 0.0246, 0.0212, 0.0248, 0.0224], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:32:58,003 INFO [finetune.py:976] (1/7) Epoch 14, batch 4750, loss[loss=0.1665, simple_loss=0.2383, pruned_loss=0.04736, over 4926.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2495, pruned_loss=0.05763, over 953209.15 frames. ], batch size: 33, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:32:59,181 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79212.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:32:59,786 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2728, 2.9175, 3.0529, 3.2181, 3.0410, 2.9049, 3.3280, 1.0446], + device='cuda:1'), covar=tensor([0.1016, 0.1035, 0.0940, 0.1057, 0.1487, 0.1714, 0.1043, 0.5118], + device='cuda:1'), in_proj_covar=tensor([0.0344, 0.0239, 0.0271, 0.0288, 0.0328, 0.0279, 0.0296, 0.0292], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:33:32,020 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.558e+01 1.654e+02 1.996e+02 2.359e+02 6.861e+02, threshold=3.993e+02, percent-clipped=2.0 +2023-03-26 17:33:51,148 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=79260.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:33:51,678 INFO [finetune.py:976] (1/7) Epoch 14, batch 4800, loss[loss=0.1842, simple_loss=0.2586, pruned_loss=0.05489, over 4784.00 frames. ], tot_loss[loss=0.1861, simple_loss=0.2533, pruned_loss=0.0594, over 952365.46 frames. ], batch size: 29, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:34:10,283 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79285.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 17:34:24,522 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0922, 3.5123, 3.7389, 3.9274, 3.8359, 3.6338, 4.1511, 1.4252], + device='cuda:1'), covar=tensor([0.0726, 0.0866, 0.0793, 0.0928, 0.1175, 0.1408, 0.0670, 0.5258], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0241, 0.0272, 0.0290, 0.0330, 0.0280, 0.0297, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:34:27,365 INFO [finetune.py:976] (1/7) Epoch 14, batch 4850, loss[loss=0.2851, simple_loss=0.3334, pruned_loss=0.1184, over 4821.00 frames. ], tot_loss[loss=0.1902, simple_loss=0.2584, pruned_loss=0.06104, over 954221.45 frames. ], batch size: 40, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:34:35,470 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79321.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:34:42,728 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79333.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:34:49,133 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.151e+02 1.639e+02 1.937e+02 2.312e+02 4.640e+02, threshold=3.873e+02, percent-clipped=1.0 +2023-03-26 17:34:51,057 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79346.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 17:34:58,210 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7476, 1.7014, 1.6023, 1.8794, 2.1880, 1.8268, 1.5088, 1.4409], + device='cuda:1'), covar=tensor([0.2055, 0.1920, 0.1728, 0.1513, 0.1688, 0.1153, 0.2386, 0.1778], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0208, 0.0212, 0.0191, 0.0242, 0.0185, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:35:00,513 INFO [finetune.py:976] (1/7) Epoch 14, batch 4900, loss[loss=0.2184, simple_loss=0.2895, pruned_loss=0.07365, over 4802.00 frames. ], tot_loss[loss=0.1914, simple_loss=0.26, pruned_loss=0.06135, over 955459.60 frames. ], batch size: 45, lr: 3.53e-03, grad_scale: 32.0 +2023-03-26 17:35:03,437 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79364.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:35:09,188 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.73 vs. limit=2.0 +2023-03-26 17:35:11,496 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79375.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 17:35:23,209 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79394.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:35:34,335 INFO [finetune.py:976] (1/7) Epoch 14, batch 4950, loss[loss=0.187, simple_loss=0.2622, pruned_loss=0.05594, over 4898.00 frames. ], tot_loss[loss=0.1913, simple_loss=0.2602, pruned_loss=0.06121, over 955180.39 frames. ], batch size: 36, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:35:45,280 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79425.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:35:51,949 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79436.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 17:35:55,999 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.049e+02 1.541e+02 1.877e+02 2.434e+02 3.585e+02, threshold=3.755e+02, percent-clipped=0.0 +2023-03-26 17:36:07,907 INFO [finetune.py:976] (1/7) Epoch 14, batch 5000, loss[loss=0.1926, simple_loss=0.2631, pruned_loss=0.06108, over 4793.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2577, pruned_loss=0.06025, over 953821.22 frames. ], batch size: 51, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:36:28,540 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.98 vs. limit=5.0 +2023-03-26 17:36:35,020 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2134, 1.8154, 2.2528, 2.1370, 1.8949, 1.8742, 2.0859, 1.9777], + device='cuda:1'), covar=tensor([0.4511, 0.4527, 0.3430, 0.4228, 0.5332, 0.4173, 0.5081, 0.3487], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0241, 0.0258, 0.0268, 0.0266, 0.0240, 0.0280, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:36:39,078 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3757, 3.2543, 3.1088, 1.4260, 3.3409, 2.4265, 0.9553, 2.2322], + device='cuda:1'), covar=tensor([0.2692, 0.1910, 0.1736, 0.3565, 0.1238, 0.1142, 0.4145, 0.1637], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0173, 0.0160, 0.0128, 0.0157, 0.0122, 0.0146, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 17:36:41,414 INFO [finetune.py:976] (1/7) Epoch 14, batch 5050, loss[loss=0.206, simple_loss=0.2619, pruned_loss=0.07504, over 4901.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2548, pruned_loss=0.0592, over 954914.28 frames. ], batch size: 32, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:37:02,697 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.152e+02 1.527e+02 1.776e+02 2.127e+02 3.568e+02, threshold=3.553e+02, percent-clipped=0.0 +2023-03-26 17:37:14,552 INFO [finetune.py:976] (1/7) Epoch 14, batch 5100, loss[loss=0.1948, simple_loss=0.2594, pruned_loss=0.06509, over 4904.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2505, pruned_loss=0.05729, over 955535.98 frames. ], batch size: 32, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:37:15,413 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-26 17:37:38,099 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79585.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:37:57,944 INFO [finetune.py:976] (1/7) Epoch 14, batch 5150, loss[loss=0.1618, simple_loss=0.2383, pruned_loss=0.04268, over 4823.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2498, pruned_loss=0.0574, over 955103.24 frames. ], batch size: 25, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:38:04,629 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79621.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:38:07,108 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0927, 1.9550, 1.6256, 1.7706, 2.0179, 1.7480, 2.1757, 2.0353], + device='cuda:1'), covar=tensor([0.1356, 0.2007, 0.3223, 0.2777, 0.2946, 0.1720, 0.3916, 0.1880], + device='cuda:1'), in_proj_covar=tensor([0.0180, 0.0187, 0.0234, 0.0254, 0.0244, 0.0199, 0.0212, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:38:20,366 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79641.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 17:38:21,527 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.081e+02 1.601e+02 1.924e+02 2.369e+02 3.228e+02, threshold=3.849e+02, percent-clipped=0.0 +2023-03-26 17:38:23,474 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79646.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:38:38,715 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79657.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:38:41,523 INFO [finetune.py:976] (1/7) Epoch 14, batch 5200, loss[loss=0.1597, simple_loss=0.2286, pruned_loss=0.04536, over 4770.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.2529, pruned_loss=0.05818, over 956630.18 frames. ], batch size: 28, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:38:50,974 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=79669.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:38:51,647 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6284, 1.7316, 1.3629, 1.6767, 2.0708, 1.9673, 1.6718, 1.4417], + device='cuda:1'), covar=tensor([0.0299, 0.0264, 0.0630, 0.0286, 0.0202, 0.0436, 0.0273, 0.0371], + device='cuda:1'), in_proj_covar=tensor([0.0095, 0.0110, 0.0143, 0.0115, 0.0101, 0.0108, 0.0097, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.3523e-05, 8.4905e-05, 1.1309e-04, 8.9135e-05, 7.9226e-05, 7.9770e-05, + 7.3288e-05, 8.3465e-05], device='cuda:1') +2023-03-26 17:39:10,032 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79689.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:39:27,297 INFO [finetune.py:976] (1/7) Epoch 14, batch 5250, loss[loss=0.174, simple_loss=0.2408, pruned_loss=0.05359, over 4749.00 frames. ], tot_loss[loss=0.1861, simple_loss=0.2547, pruned_loss=0.05878, over 953637.08 frames. ], batch size: 28, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:39:32,154 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79718.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:39:33,310 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79720.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:39:40,419 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79731.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 17:39:41,702 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6502, 1.4746, 1.0142, 0.2889, 1.2282, 1.4245, 1.3533, 1.3424], + device='cuda:1'), covar=tensor([0.1024, 0.0909, 0.1548, 0.2120, 0.1567, 0.2722, 0.2572, 0.1093], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0196, 0.0200, 0.0184, 0.0214, 0.0208, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:39:43,453 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-26 17:39:49,075 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.058e+02 1.683e+02 1.987e+02 2.478e+02 3.642e+02, threshold=3.974e+02, percent-clipped=0.0 +2023-03-26 17:39:57,674 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3807, 2.2163, 2.0699, 1.2881, 2.1948, 1.8973, 1.8258, 2.1895], + device='cuda:1'), covar=tensor([0.0912, 0.0722, 0.1535, 0.1692, 0.1180, 0.1594, 0.1651, 0.0763], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0196, 0.0201, 0.0184, 0.0215, 0.0208, 0.0225, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:39:59,971 INFO [finetune.py:976] (1/7) Epoch 14, batch 5300, loss[loss=0.1545, simple_loss=0.2161, pruned_loss=0.04641, over 4263.00 frames. ], tot_loss[loss=0.1881, simple_loss=0.2565, pruned_loss=0.05983, over 951092.46 frames. ], batch size: 18, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:40:00,705 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79762.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:40:33,375 INFO [finetune.py:976] (1/7) Epoch 14, batch 5350, loss[loss=0.211, simple_loss=0.2739, pruned_loss=0.07409, over 4767.00 frames. ], tot_loss[loss=0.1892, simple_loss=0.2579, pruned_loss=0.06027, over 952110.95 frames. ], batch size: 28, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:40:41,375 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79823.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:40:55,393 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.090e+02 1.503e+02 1.776e+02 2.291e+02 4.117e+02, threshold=3.553e+02, percent-clipped=3.0 +2023-03-26 17:41:06,813 INFO [finetune.py:976] (1/7) Epoch 14, batch 5400, loss[loss=0.1736, simple_loss=0.2399, pruned_loss=0.05365, over 4899.00 frames. ], tot_loss[loss=0.1874, simple_loss=0.2557, pruned_loss=0.0595, over 953353.43 frames. ], batch size: 36, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:41:40,249 INFO [finetune.py:976] (1/7) Epoch 14, batch 5450, loss[loss=0.1546, simple_loss=0.2122, pruned_loss=0.04846, over 4917.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2538, pruned_loss=0.05971, over 953389.44 frames. ], batch size: 36, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:41:43,364 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8595, 4.0067, 3.7603, 1.9226, 4.1001, 3.0318, 0.6854, 2.8369], + device='cuda:1'), covar=tensor([0.2373, 0.1900, 0.1715, 0.3637, 0.1057, 0.1128, 0.5052, 0.1691], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0174, 0.0161, 0.0128, 0.0158, 0.0123, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 17:41:59,674 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79941.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:41:59,692 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79941.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 17:42:00,807 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.485e+02 1.773e+02 2.175e+02 4.141e+02, threshold=3.546e+02, percent-clipped=3.0 +2023-03-26 17:42:10,064 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0564, 1.5602, 2.0341, 1.9830, 1.8135, 1.7843, 1.9651, 1.9391], + device='cuda:1'), covar=tensor([0.3990, 0.4203, 0.3498, 0.4074, 0.5167, 0.3738, 0.4828, 0.3378], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0240, 0.0257, 0.0267, 0.0266, 0.0238, 0.0279, 0.0236], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:42:14,248 INFO [finetune.py:976] (1/7) Epoch 14, batch 5500, loss[loss=0.164, simple_loss=0.2184, pruned_loss=0.05478, over 4202.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2501, pruned_loss=0.05804, over 954179.73 frames. ], batch size: 18, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:42:15,705 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.07 vs. limit=5.0 +2023-03-26 17:42:32,369 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=79989.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 17:42:32,387 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79989.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:42:34,932 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.61 vs. limit=2.0 +2023-03-26 17:42:54,668 INFO [finetune.py:976] (1/7) Epoch 14, batch 5550, loss[loss=0.1772, simple_loss=0.2362, pruned_loss=0.05911, over 4730.00 frames. ], tot_loss[loss=0.1838, simple_loss=0.251, pruned_loss=0.05833, over 956159.24 frames. ], batch size: 23, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:42:55,963 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80013.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:42:56,076 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.21 vs. limit=5.0 +2023-03-26 17:43:00,266 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80020.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:43:07,413 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80031.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 17:43:11,544 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=80037.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:43:15,065 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.061e+02 1.609e+02 1.862e+02 2.441e+02 4.163e+02, threshold=3.724e+02, percent-clipped=2.0 +2023-03-26 17:43:25,569 INFO [finetune.py:976] (1/7) Epoch 14, batch 5600, loss[loss=0.151, simple_loss=0.2303, pruned_loss=0.03585, over 4925.00 frames. ], tot_loss[loss=0.1862, simple_loss=0.2546, pruned_loss=0.05893, over 957529.69 frames. ], batch size: 37, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:43:29,672 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=80068.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:43:35,674 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-26 17:43:36,034 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=80079.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 17:44:12,671 INFO [finetune.py:976] (1/7) Epoch 14, batch 5650, loss[loss=0.2122, simple_loss=0.2733, pruned_loss=0.07554, over 4826.00 frames. ], tot_loss[loss=0.1881, simple_loss=0.257, pruned_loss=0.05963, over 956803.00 frames. ], batch size: 33, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:44:21,798 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80118.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:44:41,677 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.115e+02 1.565e+02 1.838e+02 2.201e+02 4.652e+02, threshold=3.676e+02, percent-clipped=2.0 +2023-03-26 17:44:56,307 INFO [finetune.py:976] (1/7) Epoch 14, batch 5700, loss[loss=0.1477, simple_loss=0.2189, pruned_loss=0.03823, over 4473.00 frames. ], tot_loss[loss=0.1859, simple_loss=0.2537, pruned_loss=0.05908, over 938254.04 frames. ], batch size: 19, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:45:28,231 INFO [finetune.py:976] (1/7) Epoch 15, batch 0, loss[loss=0.1885, simple_loss=0.2565, pruned_loss=0.06023, over 4917.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2565, pruned_loss=0.06023, over 4917.00 frames. ], batch size: 33, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:45:28,231 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 17:45:30,649 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2977, 2.0591, 1.4923, 0.5854, 1.8773, 1.9253, 1.7663, 1.9375], + device='cuda:1'), covar=tensor([0.1020, 0.0841, 0.1602, 0.2203, 0.1337, 0.2487, 0.2470, 0.0876], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0195, 0.0200, 0.0183, 0.0213, 0.0207, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:45:42,542 INFO [finetune.py:1010] (1/7) Epoch 15, validation: loss=0.1586, simple_loss=0.2288, pruned_loss=0.0442, over 2265189.00 frames. +2023-03-26 17:45:42,542 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 17:45:42,684 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8681, 1.7380, 1.5433, 1.4270, 1.8802, 1.5950, 1.8211, 1.8407], + device='cuda:1'), covar=tensor([0.1515, 0.2230, 0.3188, 0.2686, 0.2834, 0.1854, 0.3104, 0.1967], + device='cuda:1'), in_proj_covar=tensor([0.0180, 0.0187, 0.0234, 0.0255, 0.0245, 0.0200, 0.0213, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:46:12,249 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8770, 1.3240, 0.7917, 1.7717, 2.0763, 1.5125, 1.7375, 1.7268], + device='cuda:1'), covar=tensor([0.1501, 0.2157, 0.2136, 0.1255, 0.2082, 0.2149, 0.1409, 0.1970], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0111, 0.0093, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 17:46:15,190 INFO [finetune.py:976] (1/7) Epoch 15, batch 50, loss[loss=0.1662, simple_loss=0.2448, pruned_loss=0.04376, over 4813.00 frames. ], tot_loss[loss=0.1924, simple_loss=0.2599, pruned_loss=0.06245, over 215311.07 frames. ], batch size: 38, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:46:17,640 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80241.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:46:17,692 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0412, 1.9631, 1.3953, 1.9445, 2.0111, 1.6789, 2.6266, 2.0166], + device='cuda:1'), covar=tensor([0.1475, 0.2202, 0.3690, 0.3134, 0.2899, 0.1821, 0.2731, 0.1950], + device='cuda:1'), in_proj_covar=tensor([0.0180, 0.0187, 0.0234, 0.0254, 0.0244, 0.0199, 0.0213, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:46:18,753 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.025e+02 1.470e+02 1.896e+02 2.201e+02 3.299e+02, threshold=3.792e+02, percent-clipped=0.0 +2023-03-26 17:46:21,714 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5655, 1.4965, 1.3815, 1.5515, 1.0059, 3.3861, 1.2562, 1.6026], + device='cuda:1'), covar=tensor([0.3909, 0.2739, 0.2400, 0.2767, 0.2107, 0.0226, 0.2679, 0.1414], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0116, 0.0120, 0.0124, 0.0115, 0.0098, 0.0097, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 17:46:45,357 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80283.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:46:48,316 INFO [finetune.py:976] (1/7) Epoch 15, batch 100, loss[loss=0.1859, simple_loss=0.2646, pruned_loss=0.05353, over 4903.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2553, pruned_loss=0.06165, over 380461.99 frames. ], batch size: 36, lr: 3.52e-03, grad_scale: 32.0 +2023-03-26 17:46:48,982 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=80289.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:47:05,100 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80313.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:47:05,681 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7291, 1.0913, 0.8708, 1.6271, 2.0478, 1.5229, 1.4602, 1.5512], + device='cuda:1'), covar=tensor([0.1466, 0.2366, 0.2067, 0.1329, 0.1926, 0.1991, 0.1647, 0.2066], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0111, 0.0093, 0.0120, 0.0094, 0.0099, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 17:47:21,594 INFO [finetune.py:976] (1/7) Epoch 15, batch 150, loss[loss=0.1663, simple_loss=0.2385, pruned_loss=0.04702, over 4837.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2485, pruned_loss=0.05897, over 508921.69 frames. ], batch size: 33, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:47:25,132 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.875e+01 1.589e+02 1.860e+02 2.189e+02 4.694e+02, threshold=3.721e+02, percent-clipped=1.0 +2023-03-26 17:47:25,879 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80344.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:47:36,663 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=80361.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:47:42,668 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7008, 1.6221, 1.3442, 1.5060, 1.9407, 1.8325, 1.6103, 1.4165], + device='cuda:1'), covar=tensor([0.0280, 0.0286, 0.0603, 0.0317, 0.0200, 0.0409, 0.0357, 0.0376], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0108, 0.0141, 0.0113, 0.0100, 0.0107, 0.0097, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.2362e-05, 8.3734e-05, 1.1152e-04, 8.7637e-05, 7.8303e-05, 7.9081e-05, + 7.2584e-05, 8.2632e-05], device='cuda:1') +2023-03-26 17:47:54,529 INFO [finetune.py:976] (1/7) Epoch 15, batch 200, loss[loss=0.193, simple_loss=0.2625, pruned_loss=0.06178, over 4818.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2495, pruned_loss=0.05967, over 608707.85 frames. ], batch size: 33, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:47:58,243 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80393.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:48:16,693 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80418.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:48:22,178 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6151, 1.4939, 1.3368, 1.6458, 1.9795, 1.7125, 1.2386, 1.3593], + device='cuda:1'), covar=tensor([0.2469, 0.2323, 0.2282, 0.1883, 0.1822, 0.1406, 0.2725, 0.2141], + device='cuda:1'), in_proj_covar=tensor([0.0238, 0.0206, 0.0210, 0.0190, 0.0240, 0.0184, 0.0214, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:48:22,246 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 17:48:34,168 INFO [finetune.py:976] (1/7) Epoch 15, batch 250, loss[loss=0.201, simple_loss=0.2749, pruned_loss=0.06352, over 4106.00 frames. ], tot_loss[loss=0.1875, simple_loss=0.2527, pruned_loss=0.06112, over 683699.03 frames. ], batch size: 65, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:48:37,160 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.156e+02 1.638e+02 2.049e+02 2.410e+02 5.367e+02, threshold=4.098e+02, percent-clipped=2.0 +2023-03-26 17:48:48,463 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80454.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:48:58,605 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80463.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:49:00,356 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=80466.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:49:02,261 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4368, 1.4677, 1.2471, 1.4221, 1.7816, 1.6247, 1.4499, 1.3507], + device='cuda:1'), covar=tensor([0.0366, 0.0300, 0.0584, 0.0303, 0.0199, 0.0474, 0.0313, 0.0398], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0142, 0.0114, 0.0101, 0.0107, 0.0097, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2716e-05, 8.4215e-05, 1.1218e-04, 8.8156e-05, 7.8620e-05, 7.9391e-05, + 7.2822e-05, 8.3131e-05], device='cuda:1') +2023-03-26 17:49:16,720 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.57 vs. limit=5.0 +2023-03-26 17:49:20,044 INFO [finetune.py:976] (1/7) Epoch 15, batch 300, loss[loss=0.1772, simple_loss=0.2617, pruned_loss=0.04636, over 4728.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2553, pruned_loss=0.06117, over 743592.47 frames. ], batch size: 54, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:49:24,814 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80494.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:49:46,726 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1155, 2.7201, 2.4878, 1.2347, 2.6912, 2.2103, 2.1606, 2.4597], + device='cuda:1'), covar=tensor([0.0847, 0.0819, 0.1642, 0.2151, 0.1645, 0.2041, 0.1879, 0.1168], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0194, 0.0200, 0.0182, 0.0213, 0.0206, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:50:03,912 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80524.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:50:04,536 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1665, 1.9056, 2.6380, 1.5916, 2.3267, 2.4538, 1.8544, 2.6094], + device='cuda:1'), covar=tensor([0.1279, 0.1964, 0.1518, 0.2100, 0.0819, 0.1363, 0.2429, 0.0802], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0203, 0.0191, 0.0189, 0.0175, 0.0213, 0.0217, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:50:12,916 INFO [finetune.py:976] (1/7) Epoch 15, batch 350, loss[loss=0.2943, simple_loss=0.3394, pruned_loss=0.1246, over 4730.00 frames. ], tot_loss[loss=0.1905, simple_loss=0.2579, pruned_loss=0.06151, over 791364.59 frames. ], batch size: 59, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:50:14,266 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2839, 2.1601, 2.7167, 1.6362, 2.4832, 2.5726, 1.9180, 2.7497], + device='cuda:1'), covar=tensor([0.1515, 0.1876, 0.1568, 0.2134, 0.0841, 0.1528, 0.2394, 0.0832], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0203, 0.0191, 0.0190, 0.0175, 0.0213, 0.0217, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:50:16,427 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.008e+02 1.511e+02 1.809e+02 2.185e+02 3.892e+02, threshold=3.618e+02, percent-clipped=0.0 +2023-03-26 17:50:24,822 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80555.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:50:30,809 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-26 17:50:47,435 INFO [finetune.py:976] (1/7) Epoch 15, batch 400, loss[loss=0.1821, simple_loss=0.2582, pruned_loss=0.05306, over 4862.00 frames. ], tot_loss[loss=0.1908, simple_loss=0.2585, pruned_loss=0.06153, over 826478.23 frames. ], batch size: 34, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:51:29,121 INFO [finetune.py:976] (1/7) Epoch 15, batch 450, loss[loss=0.1726, simple_loss=0.2424, pruned_loss=0.05139, over 4917.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2546, pruned_loss=0.0595, over 855112.80 frames. ], batch size: 46, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:51:29,776 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80639.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:51:32,679 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.016e+02 1.580e+02 1.854e+02 2.177e+02 4.594e+02, threshold=3.707e+02, percent-clipped=2.0 +2023-03-26 17:52:03,114 INFO [finetune.py:976] (1/7) Epoch 15, batch 500, loss[loss=0.1641, simple_loss=0.23, pruned_loss=0.04903, over 4862.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2529, pruned_loss=0.05884, over 879285.78 frames. ], batch size: 31, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:52:21,496 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4609, 1.4406, 1.1393, 1.2708, 1.7198, 1.6614, 1.5244, 1.3155], + device='cuda:1'), covar=tensor([0.0390, 0.0403, 0.0871, 0.0421, 0.0282, 0.0555, 0.0324, 0.0450], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0142, 0.0114, 0.0101, 0.0107, 0.0097, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.3258e-05, 8.4568e-05, 1.1262e-04, 8.8343e-05, 7.8933e-05, 7.9349e-05, + 7.2951e-05, 8.3589e-05], device='cuda:1') +2023-03-26 17:52:26,221 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1682, 3.6597, 3.8054, 4.0461, 3.9523, 3.7147, 4.2571, 1.2428], + device='cuda:1'), covar=tensor([0.0792, 0.0800, 0.0833, 0.0937, 0.1173, 0.1444, 0.0707, 0.5672], + device='cuda:1'), in_proj_covar=tensor([0.0345, 0.0242, 0.0271, 0.0290, 0.0329, 0.0280, 0.0296, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:52:35,993 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80736.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:52:37,122 INFO [finetune.py:976] (1/7) Epoch 15, batch 550, loss[loss=0.1732, simple_loss=0.2399, pruned_loss=0.05326, over 4888.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2499, pruned_loss=0.05815, over 895083.98 frames. ], batch size: 35, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:52:37,328 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.63 vs. limit=2.0 +2023-03-26 17:52:40,201 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.067e+02 1.496e+02 1.725e+02 2.011e+02 3.976e+02, threshold=3.451e+02, percent-clipped=1.0 +2023-03-26 17:52:44,391 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80749.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:53:01,318 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80773.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:53:10,746 INFO [finetune.py:976] (1/7) Epoch 15, batch 600, loss[loss=0.1639, simple_loss=0.2332, pruned_loss=0.04727, over 4758.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2509, pruned_loss=0.0588, over 909283.52 frames. ], batch size: 28, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:53:16,313 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80797.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:53:32,700 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80819.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:53:44,641 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80834.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:53:47,023 INFO [finetune.py:976] (1/7) Epoch 15, batch 650, loss[loss=0.1298, simple_loss=0.2066, pruned_loss=0.02649, over 4772.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.254, pruned_loss=0.05956, over 919027.12 frames. ], batch size: 28, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:53:50,575 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.196e+02 1.643e+02 1.965e+02 2.358e+02 6.399e+02, threshold=3.929e+02, percent-clipped=5.0 +2023-03-26 17:53:54,879 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80850.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:54:20,405 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5033, 1.4260, 1.2234, 1.3740, 1.7614, 1.6823, 1.4372, 1.2494], + device='cuda:1'), covar=tensor([0.0358, 0.0365, 0.0699, 0.0393, 0.0278, 0.0459, 0.0441, 0.0471], + device='cuda:1'), in_proj_covar=tensor([0.0095, 0.0110, 0.0143, 0.0115, 0.0102, 0.0108, 0.0098, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.3803e-05, 8.5310e-05, 1.1342e-04, 8.8934e-05, 7.9583e-05, 7.9946e-05, + 7.3518e-05, 8.4109e-05], device='cuda:1') +2023-03-26 17:54:27,023 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.16 vs. limit=5.0 +2023-03-26 17:54:28,239 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.55 vs. limit=5.0 +2023-03-26 17:54:29,241 INFO [finetune.py:976] (1/7) Epoch 15, batch 700, loss[loss=0.1863, simple_loss=0.2592, pruned_loss=0.05668, over 4769.00 frames. ], tot_loss[loss=0.187, simple_loss=0.2551, pruned_loss=0.05944, over 927454.42 frames. ], batch size: 28, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:55:23,118 INFO [finetune.py:976] (1/7) Epoch 15, batch 750, loss[loss=0.2123, simple_loss=0.2821, pruned_loss=0.07128, over 4712.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2565, pruned_loss=0.06013, over 933362.18 frames. ], batch size: 54, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:55:23,819 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80939.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:55:26,160 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.122e+02 1.628e+02 1.856e+02 2.303e+02 3.612e+02, threshold=3.712e+02, percent-clipped=0.0 +2023-03-26 17:55:56,364 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=80987.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:55:56,907 INFO [finetune.py:976] (1/7) Epoch 15, batch 800, loss[loss=0.1841, simple_loss=0.2587, pruned_loss=0.05477, over 4834.00 frames. ], tot_loss[loss=0.1883, simple_loss=0.2569, pruned_loss=0.05979, over 937044.07 frames. ], batch size: 47, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:56:02,706 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.74 vs. limit=5.0 +2023-03-26 17:56:38,263 INFO [finetune.py:976] (1/7) Epoch 15, batch 850, loss[loss=0.1669, simple_loss=0.2421, pruned_loss=0.04589, over 4918.00 frames. ], tot_loss[loss=0.1877, simple_loss=0.2561, pruned_loss=0.05967, over 938275.11 frames. ], batch size: 43, lr: 3.51e-03, grad_scale: 32.0 +2023-03-26 17:56:41,289 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.001e+02 1.679e+02 1.976e+02 2.340e+02 3.768e+02, threshold=3.952e+02, percent-clipped=2.0 +2023-03-26 17:56:44,971 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81049.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:57:11,961 INFO [finetune.py:976] (1/7) Epoch 15, batch 900, loss[loss=0.1822, simple_loss=0.2511, pruned_loss=0.05666, over 4859.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2536, pruned_loss=0.05878, over 943262.98 frames. ], batch size: 44, lr: 3.51e-03, grad_scale: 64.0 +2023-03-26 17:57:14,452 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81092.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:57:17,456 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=81097.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:57:25,170 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4852, 1.2571, 1.3150, 1.3941, 1.7121, 1.5715, 1.4141, 1.2880], + device='cuda:1'), covar=tensor([0.0361, 0.0336, 0.0634, 0.0323, 0.0291, 0.0476, 0.0363, 0.0416], + device='cuda:1'), in_proj_covar=tensor([0.0095, 0.0111, 0.0144, 0.0115, 0.0102, 0.0109, 0.0098, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4032e-05, 8.5559e-05, 1.1394e-04, 8.9218e-05, 7.9871e-05, 8.0383e-05, + 7.3880e-05, 8.4238e-05], device='cuda:1') +2023-03-26 17:57:29,365 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81116.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:57:31,678 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81119.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:57:38,666 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81129.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:57:45,614 INFO [finetune.py:976] (1/7) Epoch 15, batch 950, loss[loss=0.1859, simple_loss=0.2606, pruned_loss=0.05564, over 4907.00 frames. ], tot_loss[loss=0.1849, simple_loss=0.2524, pruned_loss=0.05869, over 945650.08 frames. ], batch size: 37, lr: 3.51e-03, grad_scale: 64.0 +2023-03-26 17:57:48,668 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.129e+01 1.456e+02 1.848e+02 2.216e+02 5.430e+02, threshold=3.695e+02, percent-clipped=2.0 +2023-03-26 17:57:53,011 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81150.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:58:03,766 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=81167.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:58:11,301 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81177.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:58:19,387 INFO [finetune.py:976] (1/7) Epoch 15, batch 1000, loss[loss=0.1984, simple_loss=0.2662, pruned_loss=0.06532, over 4855.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2539, pruned_loss=0.0597, over 948499.89 frames. ], batch size: 31, lr: 3.51e-03, grad_scale: 64.0 +2023-03-26 17:58:25,510 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=81198.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:58:37,141 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2509, 1.1867, 1.5502, 1.0069, 1.3261, 1.4448, 1.1838, 1.5801], + device='cuda:1'), covar=tensor([0.1299, 0.2294, 0.1240, 0.1544, 0.0935, 0.1233, 0.3043, 0.0825], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0203, 0.0190, 0.0188, 0.0175, 0.0212, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 17:58:52,895 INFO [finetune.py:976] (1/7) Epoch 15, batch 1050, loss[loss=0.124, simple_loss=0.1849, pruned_loss=0.03152, over 4722.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2568, pruned_loss=0.06036, over 951084.16 frames. ], batch size: 23, lr: 3.51e-03, grad_scale: 64.0 +2023-03-26 17:58:56,385 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.183e+02 1.566e+02 1.800e+02 2.282e+02 3.514e+02, threshold=3.601e+02, percent-clipped=0.0 +2023-03-26 17:58:56,479 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9769, 4.5719, 4.2529, 2.2896, 4.6532, 3.7174, 0.9766, 3.3257], + device='cuda:1'), covar=tensor([0.2288, 0.1722, 0.1362, 0.3098, 0.0826, 0.0783, 0.4494, 0.1199], + device='cuda:1'), in_proj_covar=tensor([0.0149, 0.0172, 0.0158, 0.0127, 0.0155, 0.0121, 0.0144, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 17:59:31,696 INFO [finetune.py:976] (1/7) Epoch 15, batch 1100, loss[loss=0.2223, simple_loss=0.2914, pruned_loss=0.07653, over 4800.00 frames. ], tot_loss[loss=0.1899, simple_loss=0.2584, pruned_loss=0.06063, over 954417.33 frames. ], batch size: 51, lr: 3.51e-03, grad_scale: 64.0 +2023-03-26 17:59:43,362 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81299.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 17:59:49,385 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81301.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:00:16,529 INFO [finetune.py:976] (1/7) Epoch 15, batch 1150, loss[loss=0.173, simple_loss=0.2486, pruned_loss=0.04875, over 4851.00 frames. ], tot_loss[loss=0.1898, simple_loss=0.2586, pruned_loss=0.06052, over 953742.51 frames. ], batch size: 31, lr: 3.51e-03, grad_scale: 64.0 +2023-03-26 18:00:21,042 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.37 vs. limit=5.0 +2023-03-26 18:00:23,240 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.122e+02 1.635e+02 2.084e+02 2.407e+02 3.907e+02, threshold=4.168e+02, percent-clipped=1.0 +2023-03-26 18:00:41,613 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81360.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:00:42,871 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81362.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:00:59,899 INFO [finetune.py:976] (1/7) Epoch 15, batch 1200, loss[loss=0.1879, simple_loss=0.251, pruned_loss=0.06238, over 4745.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2563, pruned_loss=0.05978, over 950294.77 frames. ], batch size: 54, lr: 3.51e-03, grad_scale: 64.0 +2023-03-26 18:01:03,413 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81392.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:01:27,558 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81429.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:01:35,102 INFO [finetune.py:976] (1/7) Epoch 15, batch 1250, loss[loss=0.1806, simple_loss=0.2413, pruned_loss=0.05995, over 4934.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2537, pruned_loss=0.0585, over 953170.02 frames. ], batch size: 33, lr: 3.51e-03, grad_scale: 64.0 +2023-03-26 18:01:40,098 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=81440.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:01:42,327 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.103e+01 1.546e+02 1.830e+02 2.259e+02 3.665e+02, threshold=3.660e+02, percent-clipped=0.0 +2023-03-26 18:01:44,923 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.56 vs. limit=2.0 +2023-03-26 18:02:05,446 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81472.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:02:08,448 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=81477.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:02:09,234 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.59 vs. limit=5.0 +2023-03-26 18:02:15,564 INFO [finetune.py:976] (1/7) Epoch 15, batch 1300, loss[loss=0.1502, simple_loss=0.2185, pruned_loss=0.04096, over 4772.00 frames. ], tot_loss[loss=0.1821, simple_loss=0.2498, pruned_loss=0.05727, over 953325.23 frames. ], batch size: 28, lr: 3.50e-03, grad_scale: 64.0 +2023-03-26 18:02:24,853 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3575, 2.1063, 1.7499, 0.8144, 1.9002, 1.8996, 1.7528, 1.8786], + device='cuda:1'), covar=tensor([0.0875, 0.0788, 0.1490, 0.1955, 0.1364, 0.2026, 0.1958, 0.0982], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0193, 0.0198, 0.0182, 0.0212, 0.0205, 0.0223, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:02:49,395 INFO [finetune.py:976] (1/7) Epoch 15, batch 1350, loss[loss=0.1836, simple_loss=0.2534, pruned_loss=0.05686, over 4833.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2505, pruned_loss=0.05804, over 953473.19 frames. ], batch size: 51, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:02:53,475 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.113e+02 1.608e+02 1.859e+02 2.257e+02 3.880e+02, threshold=3.719e+02, percent-clipped=1.0 +2023-03-26 18:02:53,618 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1411, 0.9608, 1.0167, 0.4808, 0.8575, 1.1239, 1.1496, 0.9940], + device='cuda:1'), covar=tensor([0.0838, 0.0522, 0.0472, 0.0494, 0.0503, 0.0546, 0.0314, 0.0541], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0152, 0.0123, 0.0129, 0.0130, 0.0127, 0.0142, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.3142e-05, 1.1050e-04, 8.8533e-05, 9.2727e-05, 9.2418e-05, 9.1947e-05, + 1.0279e-04, 1.0546e-04], device='cuda:1') +2023-03-26 18:03:08,487 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7761, 2.0181, 1.6587, 1.7568, 2.3288, 2.2728, 2.0085, 1.8580], + device='cuda:1'), covar=tensor([0.0382, 0.0327, 0.0579, 0.0324, 0.0255, 0.0485, 0.0337, 0.0387], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0142, 0.0113, 0.0100, 0.0107, 0.0097, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2706e-05, 8.4356e-05, 1.1256e-04, 8.7489e-05, 7.8312e-05, 7.9289e-05, + 7.2909e-05, 8.3219e-05], device='cuda:1') +2023-03-26 18:03:10,267 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1064, 0.9816, 0.9955, 0.5018, 0.9015, 1.1469, 1.2011, 0.9882], + device='cuda:1'), covar=tensor([0.0923, 0.0561, 0.0507, 0.0527, 0.0530, 0.0579, 0.0374, 0.0686], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0152, 0.0123, 0.0129, 0.0130, 0.0127, 0.0142, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.3033e-05, 1.1043e-04, 8.8504e-05, 9.2639e-05, 9.2251e-05, 9.1907e-05, + 1.0273e-04, 1.0539e-04], device='cuda:1') +2023-03-26 18:03:22,722 INFO [finetune.py:976] (1/7) Epoch 15, batch 1400, loss[loss=0.1866, simple_loss=0.2695, pruned_loss=0.05191, over 4732.00 frames. ], tot_loss[loss=0.1869, simple_loss=0.2547, pruned_loss=0.05957, over 954507.46 frames. ], batch size: 54, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:03:36,113 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0721, 4.7975, 4.6007, 2.5510, 4.8609, 3.8629, 0.8071, 3.3194], + device='cuda:1'), covar=tensor([0.1955, 0.1552, 0.1165, 0.3073, 0.0623, 0.0766, 0.4622, 0.1412], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0174, 0.0158, 0.0128, 0.0156, 0.0121, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 18:03:36,754 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1391, 3.6104, 3.8446, 3.9882, 3.9120, 3.6071, 4.2169, 1.3511], + device='cuda:1'), covar=tensor([0.0828, 0.0931, 0.0812, 0.1012, 0.1241, 0.1561, 0.0689, 0.5463], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0242, 0.0271, 0.0290, 0.0328, 0.0280, 0.0295, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:03:56,014 INFO [finetune.py:976] (1/7) Epoch 15, batch 1450, loss[loss=0.2159, simple_loss=0.2723, pruned_loss=0.07978, over 4887.00 frames. ], tot_loss[loss=0.1872, simple_loss=0.2556, pruned_loss=0.05945, over 955003.14 frames. ], batch size: 32, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:04:00,100 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.106e+02 1.620e+02 1.887e+02 2.237e+02 3.719e+02, threshold=3.774e+02, percent-clipped=1.0 +2023-03-26 18:04:01,466 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0647, 1.9847, 1.7314, 1.9802, 1.7421, 4.6917, 1.8495, 2.3771], + device='cuda:1'), covar=tensor([0.3079, 0.2363, 0.2021, 0.2232, 0.1501, 0.0106, 0.2211, 0.1151], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0120, 0.0124, 0.0115, 0.0097, 0.0097, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 18:04:07,951 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81655.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:04:09,556 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81657.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:04:18,075 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81670.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:04:29,491 INFO [finetune.py:976] (1/7) Epoch 15, batch 1500, loss[loss=0.1888, simple_loss=0.2605, pruned_loss=0.05862, over 4311.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.2571, pruned_loss=0.06019, over 954450.69 frames. ], batch size: 65, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:05:16,702 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81731.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 18:05:20,802 INFO [finetune.py:976] (1/7) Epoch 15, batch 1550, loss[loss=0.259, simple_loss=0.3039, pruned_loss=0.1071, over 4270.00 frames. ], tot_loss[loss=0.1892, simple_loss=0.2579, pruned_loss=0.06028, over 954190.54 frames. ], batch size: 66, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:05:24,955 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.055e+02 1.530e+02 1.898e+02 2.293e+02 4.636e+02, threshold=3.795e+02, percent-clipped=1.0 +2023-03-26 18:05:42,308 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6218, 2.3414, 2.0476, 2.8864, 2.5896, 2.2817, 3.2449, 2.5731], + device='cuda:1'), covar=tensor([0.1538, 0.2619, 0.3351, 0.2818, 0.2712, 0.2034, 0.2844, 0.2062], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0189, 0.0237, 0.0257, 0.0248, 0.0201, 0.0215, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:05:50,090 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81772.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:06:03,756 INFO [finetune.py:976] (1/7) Epoch 15, batch 1600, loss[loss=0.1775, simple_loss=0.2419, pruned_loss=0.05654, over 4759.00 frames. ], tot_loss[loss=0.188, simple_loss=0.2557, pruned_loss=0.06009, over 954286.25 frames. ], batch size: 59, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:06:25,742 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=81820.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:06:37,140 INFO [finetune.py:976] (1/7) Epoch 15, batch 1650, loss[loss=0.1682, simple_loss=0.2322, pruned_loss=0.05207, over 4858.00 frames. ], tot_loss[loss=0.1838, simple_loss=0.2514, pruned_loss=0.05813, over 955784.27 frames. ], batch size: 47, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:06:40,762 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.362e+01 1.564e+02 1.826e+02 2.251e+02 4.924e+02, threshold=3.651e+02, percent-clipped=3.0 +2023-03-26 18:06:54,457 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6077, 1.7884, 2.2048, 2.0336, 1.8850, 4.2328, 1.6482, 1.9150], + device='cuda:1'), covar=tensor([0.1005, 0.1716, 0.1128, 0.0879, 0.1504, 0.0182, 0.1391, 0.1662], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0074, 0.0078, 0.0092, 0.0081, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 18:06:56,933 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1159, 1.3709, 1.2194, 1.3175, 1.4406, 2.4593, 1.2597, 1.5001], + device='cuda:1'), covar=tensor([0.0970, 0.1822, 0.1098, 0.0942, 0.1666, 0.0393, 0.1476, 0.1640], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0081, 0.0074, 0.0078, 0.0092, 0.0081, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 18:07:18,076 INFO [finetune.py:976] (1/7) Epoch 15, batch 1700, loss[loss=0.2006, simple_loss=0.2722, pruned_loss=0.06456, over 4854.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2497, pruned_loss=0.05753, over 955180.19 frames. ], batch size: 49, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:07:51,480 INFO [finetune.py:976] (1/7) Epoch 15, batch 1750, loss[loss=0.182, simple_loss=0.252, pruned_loss=0.05598, over 4893.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2527, pruned_loss=0.05896, over 954831.36 frames. ], batch size: 35, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:07:55,583 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.016e+02 1.510e+02 1.914e+02 2.293e+02 4.004e+02, threshold=3.828e+02, percent-clipped=1.0 +2023-03-26 18:08:02,972 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81955.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:08:04,187 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81957.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:08:24,920 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81987.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:08:25,411 INFO [finetune.py:976] (1/7) Epoch 15, batch 1800, loss[loss=0.1695, simple_loss=0.243, pruned_loss=0.04803, over 4885.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.2557, pruned_loss=0.05949, over 955612.95 frames. ], batch size: 32, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:08:36,207 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=82003.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:08:37,856 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=82005.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:08:43,868 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.66 vs. limit=2.0 +2023-03-26 18:08:52,498 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82026.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 18:09:00,021 INFO [finetune.py:976] (1/7) Epoch 15, batch 1850, loss[loss=0.2157, simple_loss=0.282, pruned_loss=0.07467, over 4157.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2578, pruned_loss=0.06037, over 955258.58 frames. ], batch size: 65, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:09:03,672 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.098e+02 1.664e+02 1.894e+02 2.440e+02 3.763e+02, threshold=3.787e+02, percent-clipped=0.0 +2023-03-26 18:09:06,715 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82048.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:09:11,030 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82055.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:09:13,458 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8188, 1.6499, 2.0833, 1.3800, 1.9188, 2.0634, 1.5940, 2.2144], + device='cuda:1'), covar=tensor([0.1302, 0.1963, 0.1311, 0.1735, 0.0858, 0.1267, 0.2703, 0.0802], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0204, 0.0193, 0.0190, 0.0176, 0.0213, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:09:33,282 INFO [finetune.py:976] (1/7) Epoch 15, batch 1900, loss[loss=0.1567, simple_loss=0.2177, pruned_loss=0.04788, over 4752.00 frames. ], tot_loss[loss=0.1889, simple_loss=0.2576, pruned_loss=0.06011, over 954049.96 frames. ], batch size: 23, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:09:42,402 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 18:09:51,836 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82116.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:09:58,151 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2919, 4.9527, 4.6405, 2.6931, 4.9740, 4.0797, 1.0003, 3.4861], + device='cuda:1'), covar=tensor([0.1967, 0.1384, 0.1300, 0.2653, 0.0660, 0.0661, 0.4282, 0.1189], + device='cuda:1'), in_proj_covar=tensor([0.0149, 0.0172, 0.0157, 0.0127, 0.0156, 0.0121, 0.0144, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 18:10:16,117 INFO [finetune.py:976] (1/7) Epoch 15, batch 1950, loss[loss=0.1583, simple_loss=0.2462, pruned_loss=0.03519, over 4870.00 frames. ], tot_loss[loss=0.1869, simple_loss=0.2557, pruned_loss=0.05906, over 953007.47 frames. ], batch size: 34, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:10:24,222 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.964e+01 1.525e+02 1.906e+02 2.226e+02 4.434e+02, threshold=3.812e+02, percent-clipped=2.0 +2023-03-26 18:10:32,052 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9465, 1.8298, 2.0389, 1.5961, 2.0997, 2.2928, 2.1603, 1.4249], + device='cuda:1'), covar=tensor([0.0697, 0.0857, 0.0790, 0.0940, 0.0669, 0.0658, 0.0678, 0.1782], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0135, 0.0142, 0.0124, 0.0125, 0.0142, 0.0142, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:11:01,528 INFO [finetune.py:976] (1/7) Epoch 15, batch 2000, loss[loss=0.1843, simple_loss=0.2427, pruned_loss=0.06293, over 4840.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2532, pruned_loss=0.05848, over 953632.05 frames. ], batch size: 47, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:11:25,758 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 18:11:38,383 INFO [finetune.py:976] (1/7) Epoch 15, batch 2050, loss[loss=0.1577, simple_loss=0.2273, pruned_loss=0.044, over 4825.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2498, pruned_loss=0.05745, over 954253.09 frames. ], batch size: 39, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:11:42,512 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.084e+02 1.417e+02 1.727e+02 2.274e+02 4.171e+02, threshold=3.454e+02, percent-clipped=1.0 +2023-03-26 18:12:04,058 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2242, 1.3748, 0.8284, 2.1149, 2.4975, 1.8177, 1.7716, 1.9748], + device='cuda:1'), covar=tensor([0.1324, 0.2056, 0.2033, 0.1035, 0.1670, 0.1993, 0.1350, 0.1810], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0096, 0.0112, 0.0094, 0.0121, 0.0095, 0.0100, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 18:12:24,922 INFO [finetune.py:976] (1/7) Epoch 15, batch 2100, loss[loss=0.235, simple_loss=0.3, pruned_loss=0.08502, over 4029.00 frames. ], tot_loss[loss=0.1813, simple_loss=0.2488, pruned_loss=0.05693, over 953343.35 frames. ], batch size: 65, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:12:50,152 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 18:12:54,220 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82326.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:13:02,490 INFO [finetune.py:976] (1/7) Epoch 15, batch 2150, loss[loss=0.198, simple_loss=0.2735, pruned_loss=0.06127, over 4863.00 frames. ], tot_loss[loss=0.184, simple_loss=0.2521, pruned_loss=0.05796, over 953257.38 frames. ], batch size: 31, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:13:06,113 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82343.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:13:06,660 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.137e+02 1.626e+02 1.861e+02 2.291e+02 4.001e+02, threshold=3.721e+02, percent-clipped=2.0 +2023-03-26 18:13:07,995 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82346.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:13:09,400 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 18:13:21,269 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7408, 3.2588, 3.4549, 3.5868, 3.5083, 3.3256, 3.8024, 1.3094], + device='cuda:1'), covar=tensor([0.0793, 0.0909, 0.0883, 0.1087, 0.1274, 0.1462, 0.0850, 0.5054], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0243, 0.0275, 0.0291, 0.0330, 0.0283, 0.0297, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:13:25,989 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=82374.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:13:31,378 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4868, 2.3875, 1.8638, 2.5445, 2.2758, 2.0245, 2.8373, 2.5272], + device='cuda:1'), covar=tensor([0.1329, 0.2326, 0.3125, 0.2735, 0.2812, 0.1705, 0.3900, 0.1717], + device='cuda:1'), in_proj_covar=tensor([0.0181, 0.0187, 0.0235, 0.0254, 0.0247, 0.0200, 0.0213, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:13:35,325 INFO [finetune.py:976] (1/7) Epoch 15, batch 2200, loss[loss=0.174, simple_loss=0.24, pruned_loss=0.05403, over 4816.00 frames. ], tot_loss[loss=0.1869, simple_loss=0.2555, pruned_loss=0.05921, over 954869.63 frames. ], batch size: 25, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:13:48,054 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82407.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:13:50,413 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82411.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:13:53,680 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-26 18:14:03,073 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0 +2023-03-26 18:14:08,110 INFO [finetune.py:976] (1/7) Epoch 15, batch 2250, loss[loss=0.1613, simple_loss=0.2293, pruned_loss=0.04659, over 3992.00 frames. ], tot_loss[loss=0.1867, simple_loss=0.2556, pruned_loss=0.05895, over 952043.32 frames. ], batch size: 17, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:14:08,234 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9079, 1.9255, 2.3879, 1.5853, 2.1087, 2.3363, 1.8426, 2.4684], + device='cuda:1'), covar=tensor([0.1493, 0.1666, 0.1447, 0.1917, 0.0957, 0.1411, 0.2433, 0.0888], + device='cuda:1'), in_proj_covar=tensor([0.0196, 0.0205, 0.0195, 0.0192, 0.0178, 0.0215, 0.0218, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:14:12,178 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.061e+02 1.505e+02 1.711e+02 2.071e+02 3.892e+02, threshold=3.421e+02, percent-clipped=2.0 +2023-03-26 18:14:15,779 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0 +2023-03-26 18:14:31,201 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8864, 1.6939, 2.2467, 1.5007, 1.9253, 2.2424, 1.6505, 2.3683], + device='cuda:1'), covar=tensor([0.1388, 0.1952, 0.1374, 0.1887, 0.0948, 0.1366, 0.2803, 0.0853], + device='cuda:1'), in_proj_covar=tensor([0.0196, 0.0205, 0.0194, 0.0192, 0.0178, 0.0214, 0.0218, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:14:41,717 INFO [finetune.py:976] (1/7) Epoch 15, batch 2300, loss[loss=0.1799, simple_loss=0.247, pruned_loss=0.05638, over 4900.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2562, pruned_loss=0.05864, over 953492.18 frames. ], batch size: 36, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:14:45,294 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82493.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 18:15:17,446 INFO [finetune.py:976] (1/7) Epoch 15, batch 2350, loss[loss=0.2102, simple_loss=0.2493, pruned_loss=0.08557, over 4749.00 frames. ], tot_loss[loss=0.1859, simple_loss=0.2544, pruned_loss=0.05868, over 953650.95 frames. ], batch size: 54, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:15:21,099 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.644e+02 1.984e+02 2.390e+02 4.799e+02, threshold=3.967e+02, percent-clipped=3.0 +2023-03-26 18:15:28,712 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82554.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 18:15:44,888 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82569.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:16:00,853 INFO [finetune.py:976] (1/7) Epoch 15, batch 2400, loss[loss=0.1836, simple_loss=0.2472, pruned_loss=0.05995, over 4811.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2512, pruned_loss=0.05768, over 954624.07 frames. ], batch size: 51, lr: 3.50e-03, grad_scale: 32.0 +2023-03-26 18:16:29,664 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.98 vs. limit=5.0 +2023-03-26 18:16:37,310 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82630.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:16:42,548 INFO [finetune.py:976] (1/7) Epoch 15, batch 2450, loss[loss=0.1754, simple_loss=0.2412, pruned_loss=0.05477, over 4745.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.248, pruned_loss=0.05645, over 955017.28 frames. ], batch size: 27, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:16:45,674 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82643.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:16:46,162 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.277e+01 1.596e+02 1.937e+02 2.264e+02 4.235e+02, threshold=3.875e+02, percent-clipped=1.0 +2023-03-26 18:16:56,845 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4179, 1.4830, 1.1627, 1.3728, 1.7597, 1.6468, 1.4228, 1.2916], + device='cuda:1'), covar=tensor([0.0316, 0.0271, 0.0648, 0.0307, 0.0212, 0.0424, 0.0322, 0.0347], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0142, 0.0113, 0.0100, 0.0107, 0.0097, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2822e-05, 8.4377e-05, 1.1262e-04, 8.7428e-05, 7.7991e-05, 7.9048e-05, + 7.3282e-05, 8.3210e-05], device='cuda:1') +2023-03-26 18:17:18,146 INFO [finetune.py:976] (1/7) Epoch 15, batch 2500, loss[loss=0.1573, simple_loss=0.2436, pruned_loss=0.03548, over 4770.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2505, pruned_loss=0.05795, over 956178.27 frames. ], batch size: 27, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:17:20,058 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=82691.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:17:35,742 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82702.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:17:46,342 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82711.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:18:03,677 INFO [finetune.py:976] (1/7) Epoch 15, batch 2550, loss[loss=0.2704, simple_loss=0.3272, pruned_loss=0.1068, over 4830.00 frames. ], tot_loss[loss=0.1863, simple_loss=0.2545, pruned_loss=0.0591, over 955571.21 frames. ], batch size: 47, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:18:04,371 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7894, 4.2999, 4.1892, 2.1722, 4.3238, 3.1965, 1.0420, 3.0018], + device='cuda:1'), covar=tensor([0.2523, 0.1663, 0.1151, 0.2868, 0.0728, 0.0897, 0.3991, 0.1306], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0174, 0.0159, 0.0128, 0.0157, 0.0122, 0.0145, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 18:18:07,768 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.099e+02 1.574e+02 1.850e+02 2.269e+02 4.152e+02, threshold=3.700e+02, percent-clipped=3.0 +2023-03-26 18:18:17,863 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=82759.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:18:33,525 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1297, 1.7854, 2.0965, 2.1080, 1.8615, 1.8769, 2.0476, 1.9095], + device='cuda:1'), covar=tensor([0.4565, 0.4499, 0.3598, 0.4350, 0.5379, 0.4142, 0.5406, 0.3616], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0239, 0.0257, 0.0268, 0.0267, 0.0240, 0.0280, 0.0236], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:18:36,859 INFO [finetune.py:976] (1/7) Epoch 15, batch 2600, loss[loss=0.1983, simple_loss=0.2684, pruned_loss=0.06407, over 4793.00 frames. ], tot_loss[loss=0.1871, simple_loss=0.2559, pruned_loss=0.05913, over 954767.72 frames. ], batch size: 29, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:18:45,280 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4781, 1.0781, 0.8510, 1.3343, 1.8753, 0.7521, 1.2601, 1.3559], + device='cuda:1'), covar=tensor([0.1547, 0.2222, 0.1817, 0.1319, 0.2005, 0.2119, 0.1521, 0.1994], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0096, 0.0112, 0.0093, 0.0121, 0.0095, 0.0100, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 18:19:10,655 INFO [finetune.py:976] (1/7) Epoch 15, batch 2650, loss[loss=0.1868, simple_loss=0.2562, pruned_loss=0.05872, over 4898.00 frames. ], tot_loss[loss=0.188, simple_loss=0.2567, pruned_loss=0.05964, over 953001.92 frames. ], batch size: 36, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:19:14,268 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.056e+02 1.579e+02 1.879e+02 2.251e+02 6.929e+02, threshold=3.759e+02, percent-clipped=2.0 +2023-03-26 18:19:16,178 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5650, 1.5594, 2.1707, 1.7847, 1.6807, 3.9960, 1.4220, 1.7111], + device='cuda:1'), covar=tensor([0.1030, 0.1840, 0.1279, 0.1051, 0.1662, 0.0213, 0.1528, 0.1773], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0082, 0.0074, 0.0078, 0.0093, 0.0082, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 18:19:17,848 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82849.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 18:19:43,179 INFO [finetune.py:976] (1/7) Epoch 15, batch 2700, loss[loss=0.1505, simple_loss=0.2226, pruned_loss=0.03917, over 4751.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.2561, pruned_loss=0.05921, over 953352.53 frames. ], batch size: 27, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:20:08,048 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82925.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:20:12,220 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5326, 1.4466, 1.9945, 2.9237, 1.9415, 2.2554, 0.9197, 2.4891], + device='cuda:1'), covar=tensor([0.1705, 0.1452, 0.1151, 0.0678, 0.0825, 0.1262, 0.1804, 0.0527], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0132, 0.0162, 0.0099, 0.0137, 0.0123, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 18:20:15,446 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-03-26 18:20:16,387 INFO [finetune.py:976] (1/7) Epoch 15, batch 2750, loss[loss=0.1842, simple_loss=0.2496, pruned_loss=0.05943, over 4834.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2534, pruned_loss=0.05859, over 954618.37 frames. ], batch size: 49, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:20:20,499 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.092e+02 1.574e+02 1.758e+02 2.107e+02 4.076e+02, threshold=3.515e+02, percent-clipped=2.0 +2023-03-26 18:20:33,228 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82963.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:20:49,686 INFO [finetune.py:976] (1/7) Epoch 15, batch 2800, loss[loss=0.1563, simple_loss=0.2155, pruned_loss=0.04853, over 4798.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2496, pruned_loss=0.0572, over 955812.78 frames. ], batch size: 25, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:21:07,072 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83002.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:21:21,921 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83024.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:21:37,765 INFO [finetune.py:976] (1/7) Epoch 15, batch 2850, loss[loss=0.1237, simple_loss=0.2004, pruned_loss=0.02344, over 4779.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2479, pruned_loss=0.05684, over 954583.93 frames. ], batch size: 26, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:21:41,396 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.061e+02 1.604e+02 1.866e+02 2.227e+02 4.125e+02, threshold=3.733e+02, percent-clipped=3.0 +2023-03-26 18:21:48,005 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6394, 1.6009, 1.6067, 0.9012, 1.7392, 1.9182, 1.8969, 1.4885], + device='cuda:1'), covar=tensor([0.1041, 0.0725, 0.0521, 0.0626, 0.0527, 0.0613, 0.0403, 0.0721], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0151, 0.0123, 0.0128, 0.0129, 0.0126, 0.0140, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.1886e-05, 1.0965e-04, 8.8038e-05, 9.1719e-05, 9.1178e-05, 9.0974e-05, + 1.0081e-04, 1.0551e-04], device='cuda:1') +2023-03-26 18:21:48,542 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=83050.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:22:05,115 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1832, 2.0750, 2.7536, 1.5935, 2.2182, 2.5638, 1.9013, 2.7289], + device='cuda:1'), covar=tensor([0.1604, 0.2134, 0.1529, 0.2303, 0.1108, 0.1677, 0.2874, 0.1121], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0203, 0.0192, 0.0189, 0.0174, 0.0212, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:22:15,027 INFO [finetune.py:976] (1/7) Epoch 15, batch 2900, loss[loss=0.2085, simple_loss=0.2733, pruned_loss=0.07189, over 4749.00 frames. ], tot_loss[loss=0.183, simple_loss=0.2504, pruned_loss=0.05776, over 953138.52 frames. ], batch size: 59, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:22:16,420 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.66 vs. limit=5.0 +2023-03-26 18:22:33,003 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1525, 1.2449, 1.2681, 1.2897, 1.4338, 2.4425, 1.1490, 1.4017], + device='cuda:1'), covar=tensor([0.1062, 0.1990, 0.1118, 0.0980, 0.1696, 0.0368, 0.1680, 0.1843], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0074, 0.0077, 0.0092, 0.0081, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 18:22:57,707 INFO [finetune.py:976] (1/7) Epoch 15, batch 2950, loss[loss=0.2221, simple_loss=0.2898, pruned_loss=0.07718, over 4758.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.2531, pruned_loss=0.05802, over 954119.99 frames. ], batch size: 54, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:23:01,331 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.293e+02 1.748e+02 2.030e+02 2.368e+02 3.585e+02, threshold=4.059e+02, percent-clipped=0.0 +2023-03-26 18:23:08,895 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83149.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 18:23:26,496 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-03-26 18:23:33,517 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6728, 1.5565, 2.1763, 3.5314, 2.4042, 2.5167, 1.0037, 2.9587], + device='cuda:1'), covar=tensor([0.1768, 0.1495, 0.1369, 0.0582, 0.0781, 0.1380, 0.1898, 0.0454], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0133, 0.0165, 0.0101, 0.0138, 0.0125, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 18:23:37,596 INFO [finetune.py:976] (1/7) Epoch 15, batch 3000, loss[loss=0.1803, simple_loss=0.2533, pruned_loss=0.05365, over 4884.00 frames. ], tot_loss[loss=0.1876, simple_loss=0.2556, pruned_loss=0.05984, over 952323.16 frames. ], batch size: 43, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:23:37,596 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 18:23:45,392 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6713, 1.6808, 2.0474, 1.2637, 1.7494, 1.9129, 1.6095, 2.1616], + device='cuda:1'), covar=tensor([0.1439, 0.2254, 0.1386, 0.1744, 0.0890, 0.1336, 0.2844, 0.0877], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0203, 0.0192, 0.0190, 0.0175, 0.0212, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:23:48,368 INFO [finetune.py:1010] (1/7) Epoch 15, validation: loss=0.1564, simple_loss=0.2269, pruned_loss=0.04296, over 2265189.00 frames. +2023-03-26 18:23:48,369 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 18:23:59,745 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83196.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:24:00,266 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=83197.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 18:24:00,990 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.61 vs. limit=5.0 +2023-03-26 18:24:21,543 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83225.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:24:22,867 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.63 vs. limit=5.0 +2023-03-26 18:24:30,316 INFO [finetune.py:976] (1/7) Epoch 15, batch 3050, loss[loss=0.1884, simple_loss=0.2665, pruned_loss=0.05513, over 4927.00 frames. ], tot_loss[loss=0.1881, simple_loss=0.257, pruned_loss=0.05961, over 954077.19 frames. ], batch size: 42, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:24:34,914 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.542e+02 1.763e+02 2.135e+02 3.801e+02, threshold=3.526e+02, percent-clipped=0.0 +2023-03-26 18:24:44,043 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83257.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:24:54,192 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=83273.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:24:54,975 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 18:25:01,251 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83284.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:25:04,096 INFO [finetune.py:976] (1/7) Epoch 15, batch 3100, loss[loss=0.1761, simple_loss=0.2609, pruned_loss=0.04566, over 4904.00 frames. ], tot_loss[loss=0.1854, simple_loss=0.2543, pruned_loss=0.05827, over 954416.55 frames. ], batch size: 37, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:25:07,185 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.71 vs. limit=2.0 +2023-03-26 18:25:15,161 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9371, 4.7453, 4.4828, 2.6047, 4.8604, 3.7820, 0.8237, 3.4525], + device='cuda:1'), covar=tensor([0.2361, 0.1608, 0.1311, 0.2979, 0.0746, 0.0788, 0.4962, 0.1323], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0175, 0.0160, 0.0128, 0.0158, 0.0123, 0.0146, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 18:25:15,900 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.57 vs. limit=5.0 +2023-03-26 18:25:24,780 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83319.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:25:37,273 INFO [finetune.py:976] (1/7) Epoch 15, batch 3150, loss[loss=0.1458, simple_loss=0.2206, pruned_loss=0.03557, over 4911.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2522, pruned_loss=0.05813, over 952771.73 frames. ], batch size: 43, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:25:41,380 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.111e+02 1.524e+02 1.821e+02 2.258e+02 3.585e+02, threshold=3.643e+02, percent-clipped=2.0 +2023-03-26 18:25:42,587 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83345.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:26:12,519 INFO [finetune.py:976] (1/7) Epoch 15, batch 3200, loss[loss=0.1613, simple_loss=0.2323, pruned_loss=0.04516, over 4815.00 frames. ], tot_loss[loss=0.1815, simple_loss=0.2491, pruned_loss=0.05694, over 954786.96 frames. ], batch size: 41, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:26:25,189 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.11 vs. limit=5.0 +2023-03-26 18:26:47,736 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.14 vs. limit=5.0 +2023-03-26 18:26:55,876 INFO [finetune.py:976] (1/7) Epoch 15, batch 3250, loss[loss=0.2115, simple_loss=0.269, pruned_loss=0.07697, over 4805.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2495, pruned_loss=0.0572, over 955149.27 frames. ], batch size: 51, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:27:00,083 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.648e+01 1.539e+02 1.854e+02 2.232e+02 3.646e+02, threshold=3.708e+02, percent-clipped=1.0 +2023-03-26 18:27:03,147 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2665, 2.0907, 1.7402, 2.1257, 2.1304, 1.8449, 2.4683, 2.2586], + device='cuda:1'), covar=tensor([0.1299, 0.2269, 0.3215, 0.2731, 0.2557, 0.1749, 0.3278, 0.1759], + device='cuda:1'), in_proj_covar=tensor([0.0180, 0.0186, 0.0234, 0.0252, 0.0243, 0.0200, 0.0211, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:27:29,589 INFO [finetune.py:976] (1/7) Epoch 15, batch 3300, loss[loss=0.1504, simple_loss=0.2353, pruned_loss=0.03275, over 4767.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2533, pruned_loss=0.05846, over 955572.01 frames. ], batch size: 54, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:27:37,999 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 18:27:49,499 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8249, 1.3249, 1.8993, 1.7376, 1.5433, 1.5425, 1.6870, 1.7523], + device='cuda:1'), covar=tensor([0.4130, 0.4156, 0.3419, 0.4257, 0.4979, 0.4019, 0.4836, 0.3240], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0239, 0.0257, 0.0269, 0.0268, 0.0240, 0.0281, 0.0236], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:28:07,466 INFO [finetune.py:976] (1/7) Epoch 15, batch 3350, loss[loss=0.1749, simple_loss=0.2383, pruned_loss=0.05577, over 4829.00 frames. ], tot_loss[loss=0.1878, simple_loss=0.2562, pruned_loss=0.05965, over 954567.43 frames. ], batch size: 30, lr: 3.49e-03, grad_scale: 64.0 +2023-03-26 18:28:14,626 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.071e+02 1.792e+02 2.041e+02 2.510e+02 5.102e+02, threshold=4.082e+02, percent-clipped=3.0 +2023-03-26 18:28:21,872 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83552.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:28:32,664 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5434, 1.4567, 1.4419, 1.6134, 0.9797, 3.4066, 1.2291, 1.6945], + device='cuda:1'), covar=tensor([0.3331, 0.2601, 0.2185, 0.2310, 0.1918, 0.0184, 0.2679, 0.1345], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0116, 0.0120, 0.0124, 0.0115, 0.0098, 0.0097, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 18:28:54,246 INFO [finetune.py:976] (1/7) Epoch 15, batch 3400, loss[loss=0.2221, simple_loss=0.2912, pruned_loss=0.07644, over 4892.00 frames. ], tot_loss[loss=0.1898, simple_loss=0.2582, pruned_loss=0.06077, over 954750.14 frames. ], batch size: 43, lr: 3.49e-03, grad_scale: 64.0 +2023-03-26 18:29:24,881 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83619.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:29:37,314 INFO [finetune.py:976] (1/7) Epoch 15, batch 3450, loss[loss=0.1905, simple_loss=0.2592, pruned_loss=0.06088, over 4812.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2571, pruned_loss=0.05984, over 955709.01 frames. ], batch size: 41, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:29:39,049 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83640.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:29:42,000 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.066e+02 1.651e+02 1.928e+02 2.236e+02 3.717e+02, threshold=3.855e+02, percent-clipped=0.0 +2023-03-26 18:29:52,358 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.11 vs. limit=5.0 +2023-03-26 18:29:57,379 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=83667.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:30:11,001 INFO [finetune.py:976] (1/7) Epoch 15, batch 3500, loss[loss=0.1776, simple_loss=0.2345, pruned_loss=0.06035, over 4323.00 frames. ], tot_loss[loss=0.1867, simple_loss=0.2546, pruned_loss=0.05945, over 953889.83 frames. ], batch size: 65, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:30:18,307 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.40 vs. limit=5.0 +2023-03-26 18:30:38,978 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 18:30:44,676 INFO [finetune.py:976] (1/7) Epoch 15, batch 3550, loss[loss=0.149, simple_loss=0.2107, pruned_loss=0.04367, over 4371.00 frames. ], tot_loss[loss=0.1827, simple_loss=0.2506, pruned_loss=0.05746, over 952920.76 frames. ], batch size: 19, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:30:49,420 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.222e+01 1.573e+02 1.880e+02 2.102e+02 4.250e+02, threshold=3.760e+02, percent-clipped=2.0 +2023-03-26 18:30:51,914 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9683, 1.5590, 2.3591, 4.0919, 2.5652, 2.9216, 0.9881, 3.4529], + device='cuda:1'), covar=tensor([0.2186, 0.2259, 0.1698, 0.0768, 0.0987, 0.1491, 0.2391, 0.0567], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0118, 0.0133, 0.0165, 0.0101, 0.0138, 0.0125, 0.0103], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 18:30:53,740 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.6116, 4.1214, 4.3317, 4.4320, 4.4149, 4.1259, 4.6728, 2.0067], + device='cuda:1'), covar=tensor([0.0572, 0.0829, 0.0656, 0.0755, 0.0872, 0.1348, 0.0621, 0.4263], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0244, 0.0276, 0.0292, 0.0333, 0.0284, 0.0300, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:30:59,115 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83760.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:31:18,472 INFO [finetune.py:976] (1/7) Epoch 15, batch 3600, loss[loss=0.1791, simple_loss=0.237, pruned_loss=0.06062, over 4197.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2474, pruned_loss=0.05645, over 951615.67 frames. ], batch size: 18, lr: 3.49e-03, grad_scale: 32.0 +2023-03-26 18:31:48,191 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83821.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:31:48,833 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9019, 1.5640, 2.0573, 1.8652, 1.7000, 1.6520, 1.8411, 1.8565], + device='cuda:1'), covar=tensor([0.3737, 0.3640, 0.2804, 0.3593, 0.4280, 0.3648, 0.4079, 0.2750], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0239, 0.0258, 0.0269, 0.0268, 0.0240, 0.0281, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:31:51,814 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5926, 0.6910, 1.6127, 1.5281, 1.4039, 1.3299, 1.4786, 1.4825], + device='cuda:1'), covar=tensor([0.3456, 0.3581, 0.3033, 0.3230, 0.4069, 0.3395, 0.3652, 0.2997], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0239, 0.0258, 0.0269, 0.0268, 0.0240, 0.0281, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:31:59,979 INFO [finetune.py:976] (1/7) Epoch 15, batch 3650, loss[loss=0.2044, simple_loss=0.269, pruned_loss=0.0699, over 4934.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.251, pruned_loss=0.05836, over 952819.91 frames. ], batch size: 33, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:32:04,777 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.180e+02 1.603e+02 1.941e+02 2.306e+02 4.863e+02, threshold=3.882e+02, percent-clipped=1.0 +2023-03-26 18:32:09,555 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83852.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:32:10,159 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6223, 3.8927, 3.6654, 1.9285, 4.0310, 3.1062, 0.9741, 2.7171], + device='cuda:1'), covar=tensor([0.2632, 0.1894, 0.1385, 0.3411, 0.0974, 0.0921, 0.4557, 0.1582], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0174, 0.0159, 0.0128, 0.0158, 0.0122, 0.0146, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 18:32:33,847 INFO [finetune.py:976] (1/7) Epoch 15, batch 3700, loss[loss=0.1881, simple_loss=0.2622, pruned_loss=0.05694, over 4800.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2546, pruned_loss=0.05929, over 951452.51 frames. ], batch size: 51, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:32:41,696 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=83900.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:32:55,984 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5972, 1.5258, 1.3354, 1.7289, 1.9306, 1.6438, 1.2434, 1.3333], + device='cuda:1'), covar=tensor([0.2225, 0.2071, 0.1928, 0.1494, 0.1729, 0.1251, 0.2655, 0.1898], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0208, 0.0210, 0.0191, 0.0242, 0.0185, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:32:57,255 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0 +2023-03-26 18:33:07,588 INFO [finetune.py:976] (1/7) Epoch 15, batch 3750, loss[loss=0.1938, simple_loss=0.266, pruned_loss=0.0608, over 4886.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2565, pruned_loss=0.06023, over 949845.53 frames. ], batch size: 35, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:33:08,908 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83940.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:33:11,798 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.005e+02 1.596e+02 1.977e+02 2.275e+02 5.079e+02, threshold=3.955e+02, percent-clipped=1.0 +2023-03-26 18:33:14,836 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83949.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:33:53,426 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83985.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:33:55,619 INFO [finetune.py:976] (1/7) Epoch 15, batch 3800, loss[loss=0.1503, simple_loss=0.2224, pruned_loss=0.03911, over 4758.00 frames. ], tot_loss[loss=0.1895, simple_loss=0.258, pruned_loss=0.0605, over 952092.12 frames. ], batch size: 28, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:33:55,680 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=83988.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:33:58,154 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4642, 1.4992, 1.2147, 1.5293, 1.8860, 1.6807, 1.4747, 1.3614], + device='cuda:1'), covar=tensor([0.0353, 0.0321, 0.0630, 0.0292, 0.0206, 0.0422, 0.0322, 0.0373], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0108, 0.0140, 0.0112, 0.0099, 0.0106, 0.0097, 0.0107], + device='cuda:1'), out_proj_covar=tensor([7.1812e-05, 8.3370e-05, 1.1110e-04, 8.6647e-05, 7.7353e-05, 7.7958e-05, + 7.2769e-05, 8.1821e-05], device='cuda:1') +2023-03-26 18:34:11,203 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84010.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 18:34:22,269 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4916, 1.5016, 1.9976, 2.6853, 1.8094, 2.2672, 1.3009, 2.2632], + device='cuda:1'), covar=tensor([0.1753, 0.1688, 0.1165, 0.0983, 0.0872, 0.1776, 0.1592, 0.0736], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0132, 0.0163, 0.0100, 0.0137, 0.0124, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 18:34:36,933 INFO [finetune.py:976] (1/7) Epoch 15, batch 3850, loss[loss=0.1665, simple_loss=0.2322, pruned_loss=0.05042, over 4896.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2551, pruned_loss=0.0588, over 953565.42 frames. ], batch size: 35, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:34:41,716 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.557e+01 1.496e+02 1.862e+02 2.338e+02 3.560e+02, threshold=3.724e+02, percent-clipped=0.0 +2023-03-26 18:34:42,466 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84046.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:35:10,033 INFO [finetune.py:976] (1/7) Epoch 15, batch 3900, loss[loss=0.1923, simple_loss=0.2596, pruned_loss=0.06246, over 4866.00 frames. ], tot_loss[loss=0.1852, simple_loss=0.2534, pruned_loss=0.05845, over 954089.84 frames. ], batch size: 31, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:35:16,523 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84097.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:35:28,909 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84116.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:35:43,580 INFO [finetune.py:976] (1/7) Epoch 15, batch 3950, loss[loss=0.1863, simple_loss=0.24, pruned_loss=0.06632, over 4918.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2499, pruned_loss=0.05739, over 954556.52 frames. ], batch size: 43, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:35:47,769 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.097e+02 1.469e+02 1.885e+02 2.278e+02 4.120e+02, threshold=3.770e+02, percent-clipped=1.0 +2023-03-26 18:35:51,955 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 18:35:57,230 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84158.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:36:13,533 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 18:36:16,808 INFO [finetune.py:976] (1/7) Epoch 15, batch 4000, loss[loss=0.2144, simple_loss=0.279, pruned_loss=0.07493, over 4870.00 frames. ], tot_loss[loss=0.1829, simple_loss=0.2499, pruned_loss=0.05793, over 951706.79 frames. ], batch size: 34, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:36:24,513 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84199.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:36:26,859 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6172, 1.5813, 1.8034, 0.9249, 1.8351, 1.9657, 1.8871, 1.5316], + device='cuda:1'), covar=tensor([0.0891, 0.0670, 0.0425, 0.0642, 0.0422, 0.0617, 0.0361, 0.0623], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0151, 0.0124, 0.0128, 0.0130, 0.0126, 0.0141, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.1550e-05, 1.0959e-04, 8.8792e-05, 9.1931e-05, 9.2165e-05, 9.0842e-05, + 1.0179e-04, 1.0597e-04], device='cuda:1') +2023-03-26 18:36:57,622 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84236.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:36:59,238 INFO [finetune.py:976] (1/7) Epoch 15, batch 4050, loss[loss=0.1934, simple_loss=0.252, pruned_loss=0.06745, over 4720.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2525, pruned_loss=0.05888, over 950240.18 frames. ], batch size: 23, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:37:07,827 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.087e+02 1.586e+02 1.909e+02 2.268e+02 5.729e+02, threshold=3.818e+02, percent-clipped=2.0 +2023-03-26 18:37:21,559 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84260.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:37:39,952 INFO [finetune.py:976] (1/7) Epoch 15, batch 4100, loss[loss=0.1935, simple_loss=0.2495, pruned_loss=0.06876, over 4892.00 frames. ], tot_loss[loss=0.1877, simple_loss=0.2561, pruned_loss=0.05969, over 950929.18 frames. ], batch size: 35, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:37:46,460 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84297.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:37:52,188 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84305.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 18:38:08,848 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.8996, 4.2873, 4.4765, 4.7390, 4.6530, 4.3431, 5.0115, 1.5292], + device='cuda:1'), covar=tensor([0.0782, 0.0803, 0.0809, 0.0819, 0.1170, 0.1432, 0.0528, 0.5794], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0243, 0.0275, 0.0292, 0.0333, 0.0283, 0.0298, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:38:11,869 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84336.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:38:13,433 INFO [finetune.py:976] (1/7) Epoch 15, batch 4150, loss[loss=0.2058, simple_loss=0.2831, pruned_loss=0.06425, over 4822.00 frames. ], tot_loss[loss=0.1892, simple_loss=0.2578, pruned_loss=0.06029, over 953965.03 frames. ], batch size: 47, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:38:15,327 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84341.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:38:18,112 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.618e+02 1.997e+02 2.307e+02 7.274e+02, threshold=3.993e+02, percent-clipped=3.0 +2023-03-26 18:38:46,761 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1201, 2.0770, 1.6901, 2.1460, 2.0492, 1.8043, 2.4841, 2.1116], + device='cuda:1'), covar=tensor([0.1236, 0.2156, 0.2759, 0.2316, 0.2405, 0.1638, 0.2814, 0.1770], + device='cuda:1'), in_proj_covar=tensor([0.0181, 0.0187, 0.0234, 0.0253, 0.0245, 0.0201, 0.0212, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:38:50,306 INFO [finetune.py:976] (1/7) Epoch 15, batch 4200, loss[loss=0.1592, simple_loss=0.2228, pruned_loss=0.0478, over 4750.00 frames. ], tot_loss[loss=0.1894, simple_loss=0.2582, pruned_loss=0.06025, over 954017.38 frames. ], batch size: 26, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:39:04,821 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84397.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:39:17,864 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84416.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:39:31,951 INFO [finetune.py:976] (1/7) Epoch 15, batch 4250, loss[loss=0.2121, simple_loss=0.2743, pruned_loss=0.07494, over 4816.00 frames. ], tot_loss[loss=0.1883, simple_loss=0.2566, pruned_loss=0.05997, over 955367.30 frames. ], batch size: 41, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:39:36,665 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.324e+01 1.559e+02 1.825e+02 2.300e+02 4.289e+02, threshold=3.650e+02, percent-clipped=1.0 +2023-03-26 18:39:47,481 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84453.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:39:58,790 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=84464.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:39:59,514 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9676, 1.0769, 1.9397, 1.8155, 1.7232, 1.6668, 1.7269, 1.8526], + device='cuda:1'), covar=tensor([0.3480, 0.3913, 0.3394, 0.3391, 0.4481, 0.3332, 0.3970, 0.3091], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0238, 0.0257, 0.0268, 0.0267, 0.0240, 0.0279, 0.0236], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:40:12,587 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6116, 2.3190, 2.9495, 1.7452, 2.6129, 2.8803, 2.2340, 3.1466], + device='cuda:1'), covar=tensor([0.1209, 0.1786, 0.1494, 0.2007, 0.0947, 0.1479, 0.2212, 0.0824], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0203, 0.0193, 0.0189, 0.0176, 0.0212, 0.0216, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:40:14,294 INFO [finetune.py:976] (1/7) Epoch 15, batch 4300, loss[loss=0.1719, simple_loss=0.246, pruned_loss=0.04885, over 4935.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2541, pruned_loss=0.05937, over 957441.28 frames. ], batch size: 33, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:40:39,092 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84525.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:40:42,150 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8855, 1.7097, 1.6259, 1.9692, 2.3684, 1.9365, 1.5381, 1.5831], + device='cuda:1'), covar=tensor([0.1938, 0.2036, 0.1791, 0.1486, 0.1683, 0.1201, 0.2495, 0.1844], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0206, 0.0208, 0.0191, 0.0240, 0.0184, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:40:47,840 INFO [finetune.py:976] (1/7) Epoch 15, batch 4350, loss[loss=0.207, simple_loss=0.2595, pruned_loss=0.07728, over 4343.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2505, pruned_loss=0.05804, over 956726.56 frames. ], batch size: 19, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:40:52,223 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.160e+02 1.502e+02 1.820e+02 2.196e+02 3.984e+02, threshold=3.641e+02, percent-clipped=2.0 +2023-03-26 18:40:58,847 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84555.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:41:14,761 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1666, 1.9440, 2.0833, 0.7973, 2.3289, 2.5016, 2.2603, 1.8917], + device='cuda:1'), covar=tensor([0.0845, 0.0698, 0.0506, 0.0730, 0.0442, 0.0667, 0.0377, 0.0808], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0152, 0.0124, 0.0130, 0.0131, 0.0126, 0.0142, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.2176e-05, 1.1062e-04, 8.9285e-05, 9.2691e-05, 9.2909e-05, 9.1185e-05, + 1.0240e-04, 1.0641e-04], device='cuda:1') +2023-03-26 18:41:19,643 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84586.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:41:21,103 INFO [finetune.py:976] (1/7) Epoch 15, batch 4400, loss[loss=0.2063, simple_loss=0.2822, pruned_loss=0.06523, over 4844.00 frames. ], tot_loss[loss=0.1845, simple_loss=0.2517, pruned_loss=0.05865, over 957124.48 frames. ], batch size: 49, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:41:24,129 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84592.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:41:32,778 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84605.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:41:45,291 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3017, 2.1911, 1.8296, 0.8886, 1.9500, 1.8217, 1.6560, 1.8783], + device='cuda:1'), covar=tensor([0.0961, 0.0830, 0.1606, 0.1918, 0.1525, 0.1799, 0.1936, 0.1059], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0195, 0.0199, 0.0183, 0.0213, 0.0207, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:41:51,831 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84633.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:41:54,825 INFO [finetune.py:976] (1/7) Epoch 15, batch 4450, loss[loss=0.1905, simple_loss=0.267, pruned_loss=0.05701, over 4918.00 frames. ], tot_loss[loss=0.1872, simple_loss=0.255, pruned_loss=0.05966, over 954452.28 frames. ], batch size: 36, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:41:57,719 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3475, 3.7785, 3.9475, 4.2386, 4.0841, 3.8310, 4.4189, 1.4029], + device='cuda:1'), covar=tensor([0.0787, 0.0850, 0.0884, 0.0911, 0.1213, 0.1572, 0.0713, 0.5607], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0244, 0.0276, 0.0292, 0.0334, 0.0284, 0.0299, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:41:57,738 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84641.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:42:01,984 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.076e+02 1.595e+02 1.952e+02 2.292e+02 4.719e+02, threshold=3.904e+02, percent-clipped=1.0 +2023-03-26 18:42:07,100 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=84653.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:42:22,150 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1361, 1.9898, 1.6090, 2.0337, 1.9473, 1.7131, 2.3224, 2.1581], + device='cuda:1'), covar=tensor([0.1371, 0.2091, 0.3239, 0.2842, 0.2763, 0.1839, 0.3881, 0.1665], + device='cuda:1'), in_proj_covar=tensor([0.0181, 0.0187, 0.0233, 0.0253, 0.0244, 0.0201, 0.0211, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:42:46,434 INFO [finetune.py:976] (1/7) Epoch 15, batch 4500, loss[loss=0.1718, simple_loss=0.2512, pruned_loss=0.04618, over 4824.00 frames. ], tot_loss[loss=0.1869, simple_loss=0.2554, pruned_loss=0.05924, over 954201.96 frames. ], batch size: 39, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:42:47,110 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=84689.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:42:49,431 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84692.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:42:50,722 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84694.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:43:09,611 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0764, 1.9890, 1.5955, 1.8743, 1.9235, 1.9028, 1.9440, 2.6270], + device='cuda:1'), covar=tensor([0.4250, 0.4038, 0.3617, 0.4149, 0.4094, 0.2655, 0.3908, 0.1941], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0260, 0.0226, 0.0276, 0.0248, 0.0215, 0.0249, 0.0227], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:43:20,117 INFO [finetune.py:976] (1/7) Epoch 15, batch 4550, loss[loss=0.21, simple_loss=0.28, pruned_loss=0.07002, over 4821.00 frames. ], tot_loss[loss=0.1895, simple_loss=0.2575, pruned_loss=0.06077, over 953166.07 frames. ], batch size: 33, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:43:25,284 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.592e+02 2.005e+02 2.406e+02 4.528e+02, threshold=4.009e+02, percent-clipped=3.0 +2023-03-26 18:43:30,230 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84753.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:43:31,579 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 18:43:53,702 INFO [finetune.py:976] (1/7) Epoch 15, batch 4600, loss[loss=0.1465, simple_loss=0.2302, pruned_loss=0.03134, over 4864.00 frames. ], tot_loss[loss=0.1888, simple_loss=0.257, pruned_loss=0.06028, over 954145.07 frames. ], batch size: 31, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:44:06,847 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84800.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:44:07,394 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=84801.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:44:36,214 INFO [finetune.py:976] (1/7) Epoch 15, batch 4650, loss[loss=0.2307, simple_loss=0.2846, pruned_loss=0.08841, over 4849.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.2549, pruned_loss=0.05983, over 956055.35 frames. ], batch size: 49, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:44:40,389 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.176e+02 1.584e+02 1.933e+02 2.372e+02 3.946e+02, threshold=3.865e+02, percent-clipped=0.0 +2023-03-26 18:44:47,567 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84855.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:44:56,259 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84861.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:45:17,966 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84881.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:45:26,374 INFO [finetune.py:976] (1/7) Epoch 15, batch 4700, loss[loss=0.1309, simple_loss=0.2, pruned_loss=0.03088, over 4821.00 frames. ], tot_loss[loss=0.1845, simple_loss=0.2516, pruned_loss=0.05873, over 956457.84 frames. ], batch size: 41, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:45:29,334 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84892.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:45:36,039 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=84903.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:45:56,884 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.61 vs. limit=5.0 +2023-03-26 18:45:59,760 INFO [finetune.py:976] (1/7) Epoch 15, batch 4750, loss[loss=0.1685, simple_loss=0.2455, pruned_loss=0.04571, over 4916.00 frames. ], tot_loss[loss=0.1814, simple_loss=0.2482, pruned_loss=0.05729, over 955349.98 frames. ], batch size: 36, lr: 3.48e-03, grad_scale: 32.0 +2023-03-26 18:46:01,520 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=84940.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:46:04,957 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.000e+02 1.651e+02 1.892e+02 2.436e+02 4.596e+02, threshold=3.784e+02, percent-clipped=1.0 +2023-03-26 18:46:08,767 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84951.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:46:09,984 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-03-26 18:46:33,699 INFO [finetune.py:976] (1/7) Epoch 15, batch 4800, loss[loss=0.1642, simple_loss=0.2455, pruned_loss=0.04142, over 4832.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2509, pruned_loss=0.05846, over 954729.77 frames. ], batch size: 47, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:46:34,386 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84989.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:46:36,741 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84992.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:46:50,495 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85012.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:47:07,603 INFO [finetune.py:976] (1/7) Epoch 15, batch 4850, loss[loss=0.2024, simple_loss=0.2819, pruned_loss=0.06143, over 4317.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2545, pruned_loss=0.05933, over 955178.71 frames. ], batch size: 65, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:47:08,859 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=85040.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:47:12,313 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.155e+02 1.585e+02 1.858e+02 2.141e+02 6.123e+02, threshold=3.716e+02, percent-clipped=1.0 +2023-03-26 18:47:13,700 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-03-26 18:47:50,160 INFO [finetune.py:976] (1/7) Epoch 15, batch 4900, loss[loss=0.1773, simple_loss=0.2486, pruned_loss=0.05299, over 4694.00 frames. ], tot_loss[loss=0.1878, simple_loss=0.2564, pruned_loss=0.05965, over 956495.20 frames. ], batch size: 59, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:48:06,017 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85107.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:48:11,356 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85115.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:48:26,707 INFO [finetune.py:976] (1/7) Epoch 15, batch 4950, loss[loss=0.2019, simple_loss=0.2734, pruned_loss=0.06518, over 4819.00 frames. ], tot_loss[loss=0.1899, simple_loss=0.2585, pruned_loss=0.06068, over 955268.09 frames. ], batch size: 39, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:48:31,434 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.158e+02 1.624e+02 1.884e+02 2.194e+02 3.725e+02, threshold=3.769e+02, percent-clipped=1.0 +2023-03-26 18:48:39,185 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85156.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:48:41,104 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6683, 1.5371, 1.5234, 1.5507, 1.2269, 3.6981, 1.5532, 1.8953], + device='cuda:1'), covar=tensor([0.3997, 0.3274, 0.2345, 0.2869, 0.1785, 0.0279, 0.2341, 0.1216], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0120, 0.0124, 0.0115, 0.0097, 0.0096, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 18:48:47,021 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3537, 1.1371, 1.1619, 1.2278, 1.5851, 1.4818, 1.2813, 1.1399], + device='cuda:1'), covar=tensor([0.0371, 0.0398, 0.0716, 0.0358, 0.0247, 0.0547, 0.0419, 0.0492], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0143, 0.0113, 0.0100, 0.0108, 0.0099, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2824e-05, 8.4645e-05, 1.1351e-04, 8.7623e-05, 7.8282e-05, 7.9539e-05, + 7.4001e-05, 8.3085e-05], device='cuda:1') +2023-03-26 18:48:47,028 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85168.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 18:48:52,390 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85176.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:48:55,814 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85181.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:49:00,447 INFO [finetune.py:976] (1/7) Epoch 15, batch 5000, loss[loss=0.1877, simple_loss=0.2538, pruned_loss=0.06078, over 4817.00 frames. ], tot_loss[loss=0.1885, simple_loss=0.2567, pruned_loss=0.06012, over 954719.39 frames. ], batch size: 30, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:49:36,184 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=85229.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:49:41,575 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.0340, 4.4160, 4.6053, 4.8238, 4.7146, 4.4759, 5.2004, 1.5730], + device='cuda:1'), covar=tensor([0.0779, 0.0924, 0.0784, 0.0983, 0.1345, 0.1586, 0.0516, 0.6250], + device='cuda:1'), in_proj_covar=tensor([0.0345, 0.0242, 0.0272, 0.0289, 0.0331, 0.0279, 0.0296, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:49:42,106 INFO [finetune.py:976] (1/7) Epoch 15, batch 5050, loss[loss=0.1538, simple_loss=0.2208, pruned_loss=0.04337, over 4916.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2542, pruned_loss=0.05955, over 956516.72 frames. ], batch size: 37, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:49:46,815 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.063e+02 1.593e+02 1.872e+02 2.269e+02 5.264e+02, threshold=3.745e+02, percent-clipped=1.0 +2023-03-26 18:50:22,631 INFO [finetune.py:976] (1/7) Epoch 15, batch 5100, loss[loss=0.1702, simple_loss=0.2353, pruned_loss=0.05253, over 4897.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.2512, pruned_loss=0.0585, over 957265.55 frames. ], batch size: 32, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:50:23,311 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85289.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:50:37,543 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8585, 1.7775, 1.5998, 1.7700, 2.1468, 1.9747, 1.7668, 1.5802], + device='cuda:1'), covar=tensor([0.0343, 0.0280, 0.0515, 0.0284, 0.0173, 0.0433, 0.0323, 0.0378], + device='cuda:1'), in_proj_covar=tensor([0.0093, 0.0108, 0.0142, 0.0112, 0.0099, 0.0106, 0.0097, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.2447e-05, 8.3598e-05, 1.1219e-04, 8.6659e-05, 7.7477e-05, 7.8535e-05, + 7.2975e-05, 8.2294e-05], device='cuda:1') +2023-03-26 18:50:42,799 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85307.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:50:56,453 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8222, 1.6729, 1.4859, 1.8497, 2.2660, 1.8050, 1.6134, 1.4733], + device='cuda:1'), covar=tensor([0.1873, 0.1837, 0.1749, 0.1424, 0.1599, 0.1235, 0.2255, 0.1800], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0206, 0.0209, 0.0191, 0.0240, 0.0184, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:51:00,529 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4833, 3.8861, 4.1127, 4.3529, 4.2577, 3.9746, 4.6136, 1.3392], + device='cuda:1'), covar=tensor([0.0779, 0.0898, 0.0821, 0.0992, 0.1151, 0.1414, 0.0560, 0.5598], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0243, 0.0272, 0.0289, 0.0331, 0.0280, 0.0296, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:51:02,802 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=85337.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:51:03,346 INFO [finetune.py:976] (1/7) Epoch 15, batch 5150, loss[loss=0.1742, simple_loss=0.2414, pruned_loss=0.05345, over 4768.00 frames. ], tot_loss[loss=0.184, simple_loss=0.2508, pruned_loss=0.05864, over 955611.14 frames. ], batch size: 28, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:51:08,102 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.937e+01 1.526e+02 1.888e+02 2.256e+02 3.382e+02, threshold=3.776e+02, percent-clipped=0.0 +2023-03-26 18:51:26,533 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4712, 1.5050, 1.3038, 1.5893, 1.8391, 1.7161, 1.4817, 1.3376], + device='cuda:1'), covar=tensor([0.0320, 0.0292, 0.0587, 0.0276, 0.0185, 0.0527, 0.0348, 0.0391], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0143, 0.0113, 0.0100, 0.0107, 0.0098, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2722e-05, 8.4184e-05, 1.1293e-04, 8.7189e-05, 7.7897e-05, 7.9203e-05, + 7.3262e-05, 8.2853e-05], device='cuda:1') +2023-03-26 18:51:27,813 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.66 vs. limit=2.0 +2023-03-26 18:51:33,067 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0234, 1.6858, 2.3135, 1.4196, 2.0521, 2.1514, 1.6486, 2.3618], + device='cuda:1'), covar=tensor([0.1225, 0.1947, 0.1281, 0.2029, 0.0876, 0.1496, 0.2710, 0.0877], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0201, 0.0191, 0.0189, 0.0175, 0.0212, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:51:37,046 INFO [finetune.py:976] (1/7) Epoch 15, batch 5200, loss[loss=0.2155, simple_loss=0.2914, pruned_loss=0.06978, over 4901.00 frames. ], tot_loss[loss=0.1857, simple_loss=0.2531, pruned_loss=0.05916, over 954083.46 frames. ], batch size: 37, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:52:04,191 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85428.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:52:10,614 INFO [finetune.py:976] (1/7) Epoch 15, batch 5250, loss[loss=0.1781, simple_loss=0.2512, pruned_loss=0.05252, over 4902.00 frames. ], tot_loss[loss=0.1859, simple_loss=0.2538, pruned_loss=0.059, over 953458.21 frames. ], batch size: 37, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:52:15,822 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.083e+02 1.657e+02 1.928e+02 2.523e+02 8.274e+02, threshold=3.856e+02, percent-clipped=2.0 +2023-03-26 18:52:23,094 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85456.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:52:27,681 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85463.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 18:52:33,027 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85471.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:52:40,804 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85483.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:52:44,235 INFO [finetune.py:976] (1/7) Epoch 15, batch 5300, loss[loss=0.255, simple_loss=0.3258, pruned_loss=0.09212, over 4917.00 frames. ], tot_loss[loss=0.1863, simple_loss=0.2549, pruned_loss=0.05885, over 954446.92 frames. ], batch size: 42, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:52:44,965 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85489.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:52:54,980 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=85504.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:52:55,152 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 18:53:14,949 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9025, 1.8386, 1.6614, 1.8539, 1.2302, 4.6727, 1.6197, 1.9305], + device='cuda:1'), covar=tensor([0.3220, 0.2446, 0.2150, 0.2207, 0.1820, 0.0087, 0.2290, 0.1368], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0119, 0.0123, 0.0114, 0.0097, 0.0096, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 18:53:19,692 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85530.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:53:24,908 INFO [finetune.py:976] (1/7) Epoch 15, batch 5350, loss[loss=0.1463, simple_loss=0.2131, pruned_loss=0.03978, over 4428.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2554, pruned_loss=0.05913, over 953452.91 frames. ], batch size: 19, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:53:28,695 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85544.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:53:29,176 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.094e+02 1.542e+02 1.806e+02 2.197e+02 4.190e+02, threshold=3.613e+02, percent-clipped=2.0 +2023-03-26 18:53:58,022 INFO [finetune.py:976] (1/7) Epoch 15, batch 5400, loss[loss=0.1675, simple_loss=0.2344, pruned_loss=0.05032, over 4821.00 frames. ], tot_loss[loss=0.186, simple_loss=0.254, pruned_loss=0.05904, over 955911.09 frames. ], batch size: 33, lr: 3.47e-03, grad_scale: 32.0 +2023-03-26 18:54:00,460 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85591.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:54:11,188 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85607.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:54:23,525 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85625.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:54:31,764 INFO [finetune.py:976] (1/7) Epoch 15, batch 5450, loss[loss=0.1405, simple_loss=0.2066, pruned_loss=0.0372, over 4419.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2516, pruned_loss=0.05805, over 957219.38 frames. ], batch size: 19, lr: 3.47e-03, grad_scale: 64.0 +2023-03-26 18:54:40,638 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0264, 1.9287, 1.6778, 1.8481, 2.0917, 1.7977, 2.2388, 1.9957], + device='cuda:1'), covar=tensor([0.1272, 0.1965, 0.2763, 0.2267, 0.2266, 0.1630, 0.2614, 0.1640], + device='cuda:1'), in_proj_covar=tensor([0.0182, 0.0189, 0.0236, 0.0256, 0.0246, 0.0203, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:54:41,082 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.884e+01 1.516e+02 1.902e+02 2.390e+02 5.288e+02, threshold=3.804e+02, percent-clipped=4.0 +2023-03-26 18:54:52,061 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=85655.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:55:04,738 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-03-26 18:55:16,865 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85686.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 18:55:17,973 INFO [finetune.py:976] (1/7) Epoch 15, batch 5500, loss[loss=0.1684, simple_loss=0.2471, pruned_loss=0.04483, over 4770.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2492, pruned_loss=0.05721, over 955738.80 frames. ], batch size: 27, lr: 3.47e-03, grad_scale: 64.0 +2023-03-26 18:55:39,096 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2746, 3.7050, 3.8626, 4.0502, 4.0208, 3.7901, 4.3340, 1.4656], + device='cuda:1'), covar=tensor([0.0774, 0.0861, 0.0798, 0.0950, 0.1179, 0.1503, 0.0708, 0.5357], + device='cuda:1'), in_proj_covar=tensor([0.0344, 0.0241, 0.0271, 0.0287, 0.0328, 0.0278, 0.0295, 0.0291], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:56:02,558 INFO [finetune.py:976] (1/7) Epoch 15, batch 5550, loss[loss=0.2362, simple_loss=0.2751, pruned_loss=0.09859, over 4691.00 frames. ], tot_loss[loss=0.1829, simple_loss=0.2504, pruned_loss=0.05773, over 951834.05 frames. ], batch size: 23, lr: 3.47e-03, grad_scale: 64.0 +2023-03-26 18:56:06,721 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.623e+01 1.589e+02 1.875e+02 2.150e+02 4.153e+02, threshold=3.750e+02, percent-clipped=1.0 +2023-03-26 18:56:19,300 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85763.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:56:24,653 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85771.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:56:25,811 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4729, 1.4448, 1.2078, 1.3657, 1.8166, 1.7547, 1.5175, 1.3460], + device='cuda:1'), covar=tensor([0.0398, 0.0355, 0.0695, 0.0367, 0.0218, 0.0477, 0.0312, 0.0414], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0110, 0.0144, 0.0114, 0.0101, 0.0108, 0.0098, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.3179e-05, 8.4923e-05, 1.1389e-04, 8.8074e-05, 7.8637e-05, 7.9502e-05, + 7.3766e-05, 8.3262e-05], device='cuda:1') +2023-03-26 18:56:27,462 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1904, 2.0093, 2.4655, 4.0375, 2.9458, 2.6761, 0.9441, 3.3471], + device='cuda:1'), covar=tensor([0.1604, 0.1334, 0.1329, 0.0433, 0.0658, 0.1632, 0.1980, 0.0442], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0134, 0.0165, 0.0101, 0.0140, 0.0126, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 18:56:27,518 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5499, 2.3873, 1.9988, 2.3658, 2.4330, 2.1103, 2.8877, 2.4810], + device='cuda:1'), covar=tensor([0.1284, 0.2326, 0.2880, 0.3075, 0.2603, 0.1619, 0.3307, 0.1764], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0190, 0.0236, 0.0257, 0.0246, 0.0204, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:56:32,618 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85784.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:56:34,954 INFO [finetune.py:976] (1/7) Epoch 15, batch 5600, loss[loss=0.1536, simple_loss=0.2422, pruned_loss=0.03251, over 4801.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.2524, pruned_loss=0.05786, over 952032.35 frames. ], batch size: 51, lr: 3.47e-03, grad_scale: 64.0 +2023-03-26 18:56:48,426 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=85811.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:56:53,122 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=85819.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:57:04,148 INFO [finetune.py:976] (1/7) Epoch 15, batch 5650, loss[loss=0.2557, simple_loss=0.3051, pruned_loss=0.1032, over 4809.00 frames. ], tot_loss[loss=0.1872, simple_loss=0.256, pruned_loss=0.05924, over 952157.49 frames. ], batch size: 38, lr: 3.47e-03, grad_scale: 64.0 +2023-03-26 18:57:04,770 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85839.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:57:08,244 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.126e+02 1.563e+02 1.888e+02 2.328e+02 3.522e+02, threshold=3.776e+02, percent-clipped=0.0 +2023-03-26 18:57:22,658 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-03-26 18:57:26,026 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85875.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:57:32,549 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85886.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:57:33,723 INFO [finetune.py:976] (1/7) Epoch 15, batch 5700, loss[loss=0.1748, simple_loss=0.2268, pruned_loss=0.06138, over 4578.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2537, pruned_loss=0.05898, over 937860.41 frames. ], batch size: 20, lr: 3.47e-03, grad_scale: 64.0 +2023-03-26 18:57:42,824 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.99 vs. limit=2.0 +2023-03-26 18:58:02,781 INFO [finetune.py:976] (1/7) Epoch 16, batch 0, loss[loss=0.2423, simple_loss=0.2891, pruned_loss=0.09773, over 4204.00 frames. ], tot_loss[loss=0.2423, simple_loss=0.2891, pruned_loss=0.09773, over 4204.00 frames. ], batch size: 66, lr: 3.46e-03, grad_scale: 64.0 +2023-03-26 18:58:02,781 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 18:58:10,659 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7874, 1.3548, 0.9477, 1.6021, 2.0678, 1.1502, 1.6078, 1.5978], + device='cuda:1'), covar=tensor([0.1360, 0.1924, 0.1758, 0.1168, 0.1780, 0.1943, 0.1269, 0.1915], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0110, 0.0093, 0.0119, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 18:58:17,920 INFO [finetune.py:1010] (1/7) Epoch 16, validation: loss=0.1572, simple_loss=0.2278, pruned_loss=0.04329, over 2265189.00 frames. +2023-03-26 18:58:17,921 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 18:58:22,810 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7081, 1.9589, 1.6422, 1.6327, 2.2494, 2.2854, 1.9426, 1.8985], + device='cuda:1'), covar=tensor([0.0436, 0.0312, 0.0620, 0.0344, 0.0325, 0.0489, 0.0364, 0.0359], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0109, 0.0143, 0.0113, 0.0100, 0.0107, 0.0097, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2658e-05, 8.4242e-05, 1.1341e-04, 8.7361e-05, 7.8285e-05, 7.8926e-05, + 7.3140e-05, 8.2796e-05], device='cuda:1') +2023-03-26 18:58:25,989 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 18:58:26,968 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85930.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 18:58:29,353 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0077, 1.9444, 1.8419, 1.9499, 1.4572, 3.9199, 1.7145, 2.2168], + device='cuda:1'), covar=tensor([0.2990, 0.2183, 0.1867, 0.2101, 0.1565, 0.0226, 0.2103, 0.1093], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0114, 0.0118, 0.0123, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 18:58:30,589 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85936.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:58:36,423 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.566e+02 1.783e+02 2.274e+02 8.459e+02, threshold=3.567e+02, percent-clipped=4.0 +2023-03-26 18:58:40,698 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4037, 2.3693, 1.8777, 2.6496, 2.5097, 1.9902, 3.0153, 2.4473], + device='cuda:1'), covar=tensor([0.1272, 0.2226, 0.2968, 0.2425, 0.2494, 0.1661, 0.2842, 0.1822], + device='cuda:1'), in_proj_covar=tensor([0.0182, 0.0188, 0.0234, 0.0254, 0.0244, 0.0202, 0.0212, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 18:58:44,314 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85957.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:58:49,654 INFO [finetune.py:976] (1/7) Epoch 16, batch 50, loss[loss=0.1551, simple_loss=0.2299, pruned_loss=0.0401, over 4816.00 frames. ], tot_loss[loss=0.1883, simple_loss=0.2554, pruned_loss=0.06055, over 214284.07 frames. ], batch size: 47, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 18:58:59,307 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85981.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 18:59:05,873 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85991.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 18:59:14,743 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86002.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:59:17,228 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 18:59:23,676 INFO [finetune.py:976] (1/7) Epoch 16, batch 100, loss[loss=0.1968, simple_loss=0.2597, pruned_loss=0.067, over 4855.00 frames. ], tot_loss[loss=0.1864, simple_loss=0.2532, pruned_loss=0.05978, over 381159.68 frames. ], batch size: 31, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 18:59:24,985 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86018.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 18:59:44,128 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.132e+02 1.611e+02 1.877e+02 2.147e+02 3.763e+02, threshold=3.754e+02, percent-clipped=3.0 +2023-03-26 18:59:47,928 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86052.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:00:00,115 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86063.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:00:06,620 INFO [finetune.py:976] (1/7) Epoch 16, batch 150, loss[loss=0.1436, simple_loss=0.2273, pruned_loss=0.02994, over 4766.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2482, pruned_loss=0.05792, over 506398.95 frames. ], batch size: 28, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:00:08,579 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86069.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:00:17,302 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.86 vs. limit=2.0 +2023-03-26 19:00:27,238 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86084.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:00:37,233 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3864, 1.4515, 1.6735, 1.6658, 1.5400, 3.1753, 1.3266, 1.5703], + device='cuda:1'), covar=tensor([0.0961, 0.1793, 0.1113, 0.0951, 0.1541, 0.0242, 0.1450, 0.1626], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0073, 0.0078, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:00:50,270 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86113.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:00:51,952 INFO [finetune.py:976] (1/7) Epoch 16, batch 200, loss[loss=0.2114, simple_loss=0.2728, pruned_loss=0.07505, over 4788.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2481, pruned_loss=0.05846, over 605774.06 frames. ], batch size: 51, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:01:04,016 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86130.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:01:05,146 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86132.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:01:09,827 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86139.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:01:14,012 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.089e+02 1.565e+02 1.801e+02 2.285e+02 3.660e+02, threshold=3.601e+02, percent-clipped=0.0 +2023-03-26 19:01:27,190 INFO [finetune.py:976] (1/7) Epoch 16, batch 250, loss[loss=0.1918, simple_loss=0.2633, pruned_loss=0.06017, over 4808.00 frames. ], tot_loss[loss=0.183, simple_loss=0.2498, pruned_loss=0.05816, over 683280.40 frames. ], batch size: 45, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:01:34,326 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9577, 1.8036, 1.9045, 1.3114, 1.8824, 2.0195, 2.0174, 1.5324], + device='cuda:1'), covar=tensor([0.0603, 0.0738, 0.0749, 0.0954, 0.0747, 0.0748, 0.0597, 0.1237], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0134, 0.0140, 0.0122, 0.0123, 0.0139, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:01:40,818 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86186.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:01:41,403 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86187.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:01:45,501 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3510, 3.8006, 4.0088, 4.2013, 4.1222, 3.8037, 4.4561, 1.4518], + device='cuda:1'), covar=tensor([0.0765, 0.0759, 0.0775, 0.0974, 0.1152, 0.1476, 0.0646, 0.5350], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0243, 0.0274, 0.0291, 0.0332, 0.0281, 0.0298, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:02:00,655 INFO [finetune.py:976] (1/7) Epoch 16, batch 300, loss[loss=0.1361, simple_loss=0.2048, pruned_loss=0.03367, over 4802.00 frames. ], tot_loss[loss=0.1867, simple_loss=0.2543, pruned_loss=0.05957, over 743081.46 frames. ], batch size: 25, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:02:11,158 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86231.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:02:12,977 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86234.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:02:13,732 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.96 vs. limit=5.0 +2023-03-26 19:02:20,698 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.190e+02 1.625e+02 1.967e+02 2.251e+02 5.649e+02, threshold=3.935e+02, percent-clipped=3.0 +2023-03-26 19:02:20,831 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86246.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:02:23,809 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7841, 3.7867, 3.5539, 1.7336, 3.8986, 2.9250, 0.8906, 2.6920], + device='cuda:1'), covar=tensor([0.2333, 0.1767, 0.1433, 0.3364, 0.0918, 0.0995, 0.4205, 0.1421], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0173, 0.0159, 0.0127, 0.0157, 0.0122, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 19:02:34,277 INFO [finetune.py:976] (1/7) Epoch 16, batch 350, loss[loss=0.213, simple_loss=0.2853, pruned_loss=0.07035, over 4910.00 frames. ], tot_loss[loss=0.1893, simple_loss=0.2577, pruned_loss=0.06048, over 791318.89 frames. ], batch size: 36, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:02:39,873 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1924, 1.9736, 2.1074, 0.8598, 2.3318, 2.5817, 2.1431, 1.8733], + device='cuda:1'), covar=tensor([0.0947, 0.0794, 0.0521, 0.0831, 0.0437, 0.0806, 0.0507, 0.0752], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0151, 0.0124, 0.0128, 0.0129, 0.0127, 0.0142, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.1856e-05, 1.0970e-04, 8.8804e-05, 9.1725e-05, 9.1320e-05, 9.1600e-05, + 1.0220e-04, 1.0557e-04], device='cuda:1') +2023-03-26 19:02:44,517 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86281.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:02:47,962 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86286.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 19:03:01,133 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86307.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:03:05,210 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86313.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:03:07,485 INFO [finetune.py:976] (1/7) Epoch 16, batch 400, loss[loss=0.1603, simple_loss=0.2358, pruned_loss=0.04238, over 4843.00 frames. ], tot_loss[loss=0.1877, simple_loss=0.2572, pruned_loss=0.05905, over 828106.73 frames. ], batch size: 44, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:03:08,671 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7611, 1.3168, 0.8594, 1.5366, 2.0579, 1.4648, 1.5534, 1.5411], + device='cuda:1'), covar=tensor([0.1372, 0.1931, 0.1890, 0.1215, 0.1857, 0.2000, 0.1292, 0.1819], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0110, 0.0093, 0.0119, 0.0094, 0.0099, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:03:15,903 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86329.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:03:34,375 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.307e+01 1.538e+02 1.774e+02 2.172e+02 4.200e+02, threshold=3.548e+02, percent-clipped=1.0 +2023-03-26 19:03:35,736 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4929, 1.5299, 1.5396, 0.8512, 1.6408, 1.7767, 1.8439, 1.4101], + device='cuda:1'), covar=tensor([0.0841, 0.0575, 0.0504, 0.0517, 0.0405, 0.0535, 0.0307, 0.0622], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0151, 0.0124, 0.0128, 0.0130, 0.0127, 0.0142, 0.0146], + device='cuda:1'), out_proj_covar=tensor([9.2080e-05, 1.1004e-04, 8.8955e-05, 9.1800e-05, 9.1824e-05, 9.1781e-05, + 1.0246e-04, 1.0591e-04], device='cuda:1') +2023-03-26 19:03:45,094 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86358.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:03:50,331 INFO [finetune.py:976] (1/7) Epoch 16, batch 450, loss[loss=0.1773, simple_loss=0.2338, pruned_loss=0.06042, over 4819.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2542, pruned_loss=0.05799, over 856128.41 frames. ], batch size: 40, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:04:18,721 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86408.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:04:24,032 INFO [finetune.py:976] (1/7) Epoch 16, batch 500, loss[loss=0.1361, simple_loss=0.2051, pruned_loss=0.03362, over 4850.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2517, pruned_loss=0.05755, over 876526.77 frames. ], batch size: 49, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:04:30,006 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86425.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:04:40,627 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86440.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:04:43,601 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86445.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:04:44,070 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.779e+01 1.556e+02 1.875e+02 2.226e+02 4.465e+02, threshold=3.750e+02, percent-clipped=2.0 +2023-03-26 19:04:57,163 INFO [finetune.py:976] (1/7) Epoch 16, batch 550, loss[loss=0.158, simple_loss=0.2329, pruned_loss=0.04159, over 4783.00 frames. ], tot_loss[loss=0.1828, simple_loss=0.2502, pruned_loss=0.0577, over 890075.86 frames. ], batch size: 26, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:05:31,549 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86501.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:05:39,630 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86506.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 19:05:50,204 INFO [finetune.py:976] (1/7) Epoch 16, batch 600, loss[loss=0.2315, simple_loss=0.2935, pruned_loss=0.08475, over 4846.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2498, pruned_loss=0.05762, over 903608.96 frames. ], batch size: 47, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:06:04,168 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86531.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:06:14,603 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.101e+02 1.575e+02 1.922e+02 2.222e+02 3.111e+02, threshold=3.844e+02, percent-clipped=0.0 +2023-03-26 19:06:15,338 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6491, 1.2728, 0.8769, 1.4596, 2.0115, 1.2810, 1.4800, 1.4174], + device='cuda:1'), covar=tensor([0.1706, 0.2320, 0.2182, 0.1381, 0.2228, 0.2152, 0.1695, 0.2197], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0111, 0.0093, 0.0119, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:06:26,818 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4349, 1.3530, 1.2964, 1.4335, 1.0302, 2.8830, 1.0715, 1.5183], + device='cuda:1'), covar=tensor([0.3586, 0.2690, 0.2354, 0.2556, 0.1911, 0.0287, 0.2912, 0.1383], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0120, 0.0124, 0.0115, 0.0097, 0.0096, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:06:27,336 INFO [finetune.py:976] (1/7) Epoch 16, batch 650, loss[loss=0.2248, simple_loss=0.2951, pruned_loss=0.07721, over 4806.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2522, pruned_loss=0.05823, over 914278.97 frames. ], batch size: 45, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:06:36,233 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86579.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:06:36,304 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5907, 1.5449, 1.2842, 1.5684, 1.8896, 1.6891, 1.5484, 1.3244], + device='cuda:1'), covar=tensor([0.0313, 0.0289, 0.0575, 0.0281, 0.0184, 0.0511, 0.0344, 0.0397], + device='cuda:1'), in_proj_covar=tensor([0.0092, 0.0106, 0.0141, 0.0111, 0.0098, 0.0105, 0.0096, 0.0107], + device='cuda:1'), out_proj_covar=tensor([7.1539e-05, 8.2477e-05, 1.1165e-04, 8.5674e-05, 7.6561e-05, 7.7482e-05, + 7.2488e-05, 8.1761e-05], device='cuda:1') +2023-03-26 19:06:37,380 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86580.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:06:39,246 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7124, 1.5676, 1.4404, 1.7917, 1.9540, 1.7588, 1.2426, 1.4453], + device='cuda:1'), covar=tensor([0.2014, 0.1900, 0.1816, 0.1463, 0.1551, 0.1125, 0.2368, 0.1717], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0208, 0.0211, 0.0191, 0.0242, 0.0184, 0.0214, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:06:41,053 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86586.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:06:52,264 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86602.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:06:59,424 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86613.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:07:01,141 INFO [finetune.py:976] (1/7) Epoch 16, batch 700, loss[loss=0.1998, simple_loss=0.2777, pruned_loss=0.06098, over 4914.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.254, pruned_loss=0.05878, over 922837.43 frames. ], batch size: 42, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:07:13,496 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86634.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:07:17,794 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86641.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:07:21,644 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.125e+02 1.513e+02 1.866e+02 2.326e+02 3.823e+02, threshold=3.732e+02, percent-clipped=0.0 +2023-03-26 19:07:29,575 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86658.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:07:31,327 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86661.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:07:34,796 INFO [finetune.py:976] (1/7) Epoch 16, batch 750, loss[loss=0.1701, simple_loss=0.2456, pruned_loss=0.04731, over 4796.00 frames. ], tot_loss[loss=0.1877, simple_loss=0.2561, pruned_loss=0.05961, over 929518.36 frames. ], batch size: 45, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:08:02,109 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86706.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:08:03,383 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86708.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:08:08,617 INFO [finetune.py:976] (1/7) Epoch 16, batch 800, loss[loss=0.1567, simple_loss=0.2324, pruned_loss=0.04054, over 4896.00 frames. ], tot_loss[loss=0.1862, simple_loss=0.2554, pruned_loss=0.05852, over 937181.91 frames. ], batch size: 43, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:08:14,179 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86725.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:08:28,782 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.094e+02 1.501e+02 1.840e+02 2.208e+02 4.378e+02, threshold=3.681e+02, percent-clipped=4.0 +2023-03-26 19:08:40,243 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86756.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:08:47,823 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-03-26 19:08:49,459 INFO [finetune.py:976] (1/7) Epoch 16, batch 850, loss[loss=0.1773, simple_loss=0.2485, pruned_loss=0.05303, over 4768.00 frames. ], tot_loss[loss=0.1861, simple_loss=0.2543, pruned_loss=0.05897, over 942094.13 frames. ], batch size: 28, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:08:54,211 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86773.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:09:09,592 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86796.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:09:13,594 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86801.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:09:21,921 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86814.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:09:23,014 INFO [finetune.py:976] (1/7) Epoch 16, batch 900, loss[loss=0.1695, simple_loss=0.2358, pruned_loss=0.05161, over 4907.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2505, pruned_loss=0.05698, over 946130.87 frames. ], batch size: 43, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:09:43,094 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.048e+02 1.502e+02 1.904e+02 2.205e+02 3.944e+02, threshold=3.808e+02, percent-clipped=2.0 +2023-03-26 19:09:56,619 INFO [finetune.py:976] (1/7) Epoch 16, batch 950, loss[loss=0.2025, simple_loss=0.2678, pruned_loss=0.06864, over 4910.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2501, pruned_loss=0.05754, over 948577.65 frames. ], batch size: 37, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:10:02,691 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86875.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:10:04,452 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86878.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:10:20,382 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86902.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:10:40,037 INFO [finetune.py:976] (1/7) Epoch 16, batch 1000, loss[loss=0.2369, simple_loss=0.3042, pruned_loss=0.08479, over 4751.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2529, pruned_loss=0.05866, over 950639.13 frames. ], batch size: 54, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:11:01,989 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86936.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:11:08,286 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86939.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:11:12,979 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.150e+02 1.633e+02 1.920e+02 2.320e+02 4.350e+02, threshold=3.840e+02, percent-clipped=2.0 +2023-03-26 19:11:20,312 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=86950.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:11:34,824 INFO [finetune.py:976] (1/7) Epoch 16, batch 1050, loss[loss=0.1902, simple_loss=0.2609, pruned_loss=0.05977, over 4924.00 frames. ], tot_loss[loss=0.1863, simple_loss=0.255, pruned_loss=0.05886, over 952076.54 frames. ], batch size: 42, lr: 3.46e-03, grad_scale: 32.0 +2023-03-26 19:11:36,214 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8477, 1.2299, 1.8811, 1.8270, 1.6372, 1.5750, 1.7800, 1.7021], + device='cuda:1'), covar=tensor([0.3680, 0.3951, 0.3456, 0.3754, 0.4832, 0.3628, 0.4368, 0.3202], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0238, 0.0257, 0.0269, 0.0267, 0.0240, 0.0280, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:11:58,504 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87002.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 19:12:08,331 INFO [finetune.py:976] (1/7) Epoch 16, batch 1100, loss[loss=0.1825, simple_loss=0.2586, pruned_loss=0.05322, over 4715.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2563, pruned_loss=0.05864, over 952551.08 frames. ], batch size: 59, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:12:27,409 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.131e+02 1.613e+02 1.835e+02 2.273e+02 4.124e+02, threshold=3.670e+02, percent-clipped=1.0 +2023-03-26 19:12:34,662 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-26 19:12:39,657 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87063.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:12:41,762 INFO [finetune.py:976] (1/7) Epoch 16, batch 1150, loss[loss=0.1919, simple_loss=0.2698, pruned_loss=0.05707, over 4798.00 frames. ], tot_loss[loss=0.1874, simple_loss=0.2567, pruned_loss=0.05903, over 953662.80 frames. ], batch size: 40, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:13:00,933 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87096.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:13:04,445 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87101.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:13:10,407 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 19:13:15,217 INFO [finetune.py:976] (1/7) Epoch 16, batch 1200, loss[loss=0.1543, simple_loss=0.229, pruned_loss=0.03978, over 4929.00 frames. ], tot_loss[loss=0.1865, simple_loss=0.2552, pruned_loss=0.0589, over 953761.19 frames. ], batch size: 38, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:13:20,744 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1574, 2.0916, 1.7467, 2.1333, 1.9149, 1.9480, 1.9780, 2.6883], + device='cuda:1'), covar=tensor([0.3882, 0.4557, 0.3478, 0.4143, 0.4381, 0.2521, 0.4307, 0.1826], + device='cuda:1'), in_proj_covar=tensor([0.0283, 0.0260, 0.0224, 0.0273, 0.0247, 0.0214, 0.0248, 0.0226], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:13:33,241 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=87144.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:13:33,360 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-03-26 19:13:34,371 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.093e+01 1.544e+02 1.890e+02 2.251e+02 4.242e+02, threshold=3.781e+02, percent-clipped=1.0 +2023-03-26 19:13:36,661 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=87149.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:13:50,131 INFO [finetune.py:976] (1/7) Epoch 16, batch 1250, loss[loss=0.1507, simple_loss=0.2219, pruned_loss=0.03979, over 4838.00 frames. ], tot_loss[loss=0.184, simple_loss=0.2525, pruned_loss=0.05778, over 953055.87 frames. ], batch size: 47, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:13:53,128 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87170.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:13:54,372 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5793, 1.5278, 1.4235, 1.5875, 0.9502, 2.9043, 1.0479, 1.5415], + device='cuda:1'), covar=tensor([0.3243, 0.2429, 0.2077, 0.2219, 0.1855, 0.0239, 0.2651, 0.1241], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0121, 0.0124, 0.0115, 0.0097, 0.0097, 0.0097], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:14:17,428 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5244, 2.3963, 1.9599, 2.7519, 2.5338, 2.1239, 3.0111, 2.5007], + device='cuda:1'), covar=tensor([0.1386, 0.2413, 0.3119, 0.2435, 0.2538, 0.1737, 0.3088, 0.1878], + device='cuda:1'), in_proj_covar=tensor([0.0182, 0.0188, 0.0233, 0.0253, 0.0244, 0.0202, 0.0211, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:14:31,062 INFO [finetune.py:976] (1/7) Epoch 16, batch 1300, loss[loss=0.1852, simple_loss=0.2532, pruned_loss=0.0586, over 4909.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2489, pruned_loss=0.05644, over 955106.77 frames. ], batch size: 35, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:14:32,531 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 19:14:38,939 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 19:14:39,364 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4823, 1.3431, 1.9587, 2.8711, 1.8567, 2.2032, 1.0101, 2.3750], + device='cuda:1'), covar=tensor([0.1693, 0.1532, 0.1145, 0.0691, 0.0895, 0.1354, 0.1710, 0.0533], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0118, 0.0135, 0.0166, 0.0102, 0.0140, 0.0126, 0.0103], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:14:41,197 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87230.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:14:43,594 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87234.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:14:44,804 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87236.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:14:51,269 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.006e+02 1.578e+02 1.944e+02 2.250e+02 4.130e+02, threshold=3.887e+02, percent-clipped=1.0 +2023-03-26 19:14:57,987 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3565, 1.3462, 1.4866, 1.5976, 1.4307, 2.8583, 1.3447, 1.4359], + device='cuda:1'), covar=tensor([0.1030, 0.1860, 0.1222, 0.0995, 0.1696, 0.0311, 0.1520, 0.1812], + device='cuda:1'), in_proj_covar=tensor([0.0077, 0.0082, 0.0075, 0.0078, 0.0093, 0.0081, 0.0086, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:15:04,397 INFO [finetune.py:976] (1/7) Epoch 16, batch 1350, loss[loss=0.1363, simple_loss=0.2077, pruned_loss=0.03245, over 4824.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2498, pruned_loss=0.05705, over 953332.24 frames. ], batch size: 30, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:15:14,035 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 19:15:16,950 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=87284.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:15:21,323 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87291.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:15:44,163 INFO [finetune.py:976] (1/7) Epoch 16, batch 1400, loss[loss=0.1938, simple_loss=0.2714, pruned_loss=0.05814, over 4907.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.2526, pruned_loss=0.05778, over 952206.24 frames. ], batch size: 36, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:15:50,981 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7810, 1.2716, 0.8275, 1.5676, 2.0022, 1.4362, 1.5087, 1.5566], + device='cuda:1'), covar=tensor([0.1443, 0.2056, 0.2052, 0.1242, 0.2004, 0.1982, 0.1463, 0.1881], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0119, 0.0093, 0.0099, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:16:19,253 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.049e+02 1.619e+02 1.873e+02 2.376e+02 3.982e+02, threshold=3.745e+02, percent-clipped=1.0 +2023-03-26 19:16:31,528 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87358.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:16:32,169 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5834, 1.4827, 1.9049, 1.9648, 1.6834, 3.4452, 1.5427, 1.5604], + device='cuda:1'), covar=tensor([0.0984, 0.1874, 0.1196, 0.0915, 0.1578, 0.0269, 0.1427, 0.1767], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0075, 0.0078, 0.0093, 0.0081, 0.0086, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:16:38,511 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87362.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:16:41,376 INFO [finetune.py:976] (1/7) Epoch 16, batch 1450, loss[loss=0.1898, simple_loss=0.2617, pruned_loss=0.05891, over 4792.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.255, pruned_loss=0.05831, over 953762.81 frames. ], batch size: 51, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:17:18,468 INFO [finetune.py:976] (1/7) Epoch 16, batch 1500, loss[loss=0.1714, simple_loss=0.2465, pruned_loss=0.04812, over 4800.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2573, pruned_loss=0.05922, over 954664.09 frames. ], batch size: 25, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:17:23,389 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87423.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:17:39,129 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.097e+02 1.635e+02 2.033e+02 2.421e+02 4.092e+02, threshold=4.066e+02, percent-clipped=1.0 +2023-03-26 19:17:48,802 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87461.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:17:52,196 INFO [finetune.py:976] (1/7) Epoch 16, batch 1550, loss[loss=0.1599, simple_loss=0.2361, pruned_loss=0.04187, over 4867.00 frames. ], tot_loss[loss=0.188, simple_loss=0.257, pruned_loss=0.05948, over 952391.44 frames. ], batch size: 34, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:17:55,185 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87470.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:18:18,819 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87505.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:18:25,477 INFO [finetune.py:976] (1/7) Epoch 16, batch 1600, loss[loss=0.1877, simple_loss=0.2538, pruned_loss=0.06079, over 4816.00 frames. ], tot_loss[loss=0.1857, simple_loss=0.2542, pruned_loss=0.05862, over 952247.07 frames. ], batch size: 41, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:18:27,231 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=87518.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:18:29,742 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87522.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:18:38,561 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87534.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:18:46,617 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.707e+01 1.405e+02 1.628e+02 1.991e+02 3.372e+02, threshold=3.256e+02, percent-clipped=0.0 +2023-03-26 19:18:53,352 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7533, 1.7701, 1.7160, 2.0324, 2.2979, 2.0981, 1.7512, 1.5093], + device='cuda:1'), covar=tensor([0.2364, 0.1988, 0.1751, 0.1577, 0.1905, 0.1168, 0.2356, 0.2013], + device='cuda:1'), in_proj_covar=tensor([0.0239, 0.0208, 0.0210, 0.0191, 0.0242, 0.0184, 0.0214, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:18:59,343 INFO [finetune.py:976] (1/7) Epoch 16, batch 1650, loss[loss=0.1915, simple_loss=0.253, pruned_loss=0.06504, over 4869.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2517, pruned_loss=0.05804, over 953859.22 frames. ], batch size: 49, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:18:59,462 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87566.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:19:00,021 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6053, 3.5353, 3.4423, 1.6406, 3.7099, 2.7355, 1.0685, 2.6178], + device='cuda:1'), covar=tensor([0.2369, 0.2069, 0.1580, 0.3384, 0.1128, 0.1114, 0.3885, 0.1524], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0174, 0.0159, 0.0128, 0.0156, 0.0122, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 19:19:10,144 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=87582.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:19:11,438 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87584.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:19:13,585 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87586.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:19:17,432 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87589.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:19:36,464 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87605.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:19:43,482 INFO [finetune.py:976] (1/7) Epoch 16, batch 1700, loss[loss=0.1389, simple_loss=0.2066, pruned_loss=0.03559, over 4770.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2498, pruned_loss=0.0571, over 953779.26 frames. ], batch size: 26, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:19:46,220 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-03-26 19:20:03,778 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87645.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:20:04,232 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.164e+02 1.606e+02 1.879e+02 2.372e+02 9.403e+02, threshold=3.758e+02, percent-clipped=6.0 +2023-03-26 19:20:06,857 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87650.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:20:11,625 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87658.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:20:16,841 INFO [finetune.py:976] (1/7) Epoch 16, batch 1750, loss[loss=0.1808, simple_loss=0.2532, pruned_loss=0.05416, over 4892.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2517, pruned_loss=0.05808, over 955112.56 frames. ], batch size: 35, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:20:16,974 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87666.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:20:39,440 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.66 vs. limit=2.0 +2023-03-26 19:20:44,145 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=87706.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 19:20:50,640 INFO [finetune.py:976] (1/7) Epoch 16, batch 1800, loss[loss=0.1724, simple_loss=0.2381, pruned_loss=0.05337, over 4762.00 frames. ], tot_loss[loss=0.186, simple_loss=0.255, pruned_loss=0.05843, over 954969.67 frames. ], batch size: 26, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:20:51,918 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87718.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:21:13,298 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.163e+02 1.635e+02 1.907e+02 2.399e+02 5.758e+02, threshold=3.813e+02, percent-clipped=1.0 +2023-03-26 19:21:36,198 INFO [finetune.py:976] (1/7) Epoch 16, batch 1850, loss[loss=0.1998, simple_loss=0.2651, pruned_loss=0.06726, over 4890.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.255, pruned_loss=0.05834, over 953615.44 frames. ], batch size: 35, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:22:21,976 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0 +2023-03-26 19:22:22,284 INFO [finetune.py:976] (1/7) Epoch 16, batch 1900, loss[loss=0.1578, simple_loss=0.2388, pruned_loss=0.03841, over 4822.00 frames. ], tot_loss[loss=0.1874, simple_loss=0.2563, pruned_loss=0.0592, over 952193.77 frames. ], batch size: 39, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:22:23,001 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87817.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:22:25,473 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7468, 1.2910, 0.8703, 1.6334, 2.0860, 1.2728, 1.5705, 1.7124], + device='cuda:1'), covar=tensor([0.1392, 0.1963, 0.1859, 0.1095, 0.1794, 0.2039, 0.1302, 0.1783], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0119, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:22:31,504 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.10 vs. limit=5.0 +2023-03-26 19:22:41,851 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.250e+02 1.530e+02 1.792e+02 2.215e+02 4.706e+02, threshold=3.584e+02, percent-clipped=3.0 +2023-03-26 19:22:45,942 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1173, 5.2746, 5.0314, 3.7917, 5.3352, 4.2676, 2.4348, 4.5703], + device='cuda:1'), covar=tensor([0.1467, 0.1418, 0.1423, 0.1940, 0.0742, 0.0643, 0.3081, 0.0845], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0174, 0.0158, 0.0128, 0.0156, 0.0122, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 19:22:53,011 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87861.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:22:55,973 INFO [finetune.py:976] (1/7) Epoch 16, batch 1950, loss[loss=0.2127, simple_loss=0.2759, pruned_loss=0.07474, over 4818.00 frames. ], tot_loss[loss=0.1852, simple_loss=0.2542, pruned_loss=0.05805, over 951648.26 frames. ], batch size: 38, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:23:09,168 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87886.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:23:24,843 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1280, 1.3079, 1.2904, 1.3482, 1.4822, 2.4307, 1.2433, 1.4463], + device='cuda:1'), covar=tensor([0.1032, 0.1868, 0.1085, 0.0942, 0.1575, 0.0368, 0.1503, 0.1682], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0074, 0.0078, 0.0092, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:23:27,316 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9138, 1.7574, 1.5424, 1.4283, 1.9439, 1.6739, 1.8618, 1.8578], + device='cuda:1'), covar=tensor([0.1480, 0.2163, 0.3133, 0.2638, 0.2556, 0.1792, 0.2829, 0.1820], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0189, 0.0235, 0.0254, 0.0246, 0.0203, 0.0212, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:23:29,598 INFO [finetune.py:976] (1/7) Epoch 16, batch 2000, loss[loss=0.1961, simple_loss=0.2544, pruned_loss=0.06895, over 4915.00 frames. ], tot_loss[loss=0.1852, simple_loss=0.2533, pruned_loss=0.05853, over 953638.38 frames. ], batch size: 37, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:23:41,098 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=87934.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:23:45,280 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87940.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:23:48,827 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87945.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:23:49,306 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.076e+02 1.562e+02 1.812e+02 2.199e+02 5.123e+02, threshold=3.624e+02, percent-clipped=1.0 +2023-03-26 19:23:59,908 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87961.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:24:02,864 INFO [finetune.py:976] (1/7) Epoch 16, batch 2050, loss[loss=0.1646, simple_loss=0.2323, pruned_loss=0.04845, over 4894.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2499, pruned_loss=0.05744, over 955693.15 frames. ], batch size: 32, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:24:37,647 INFO [finetune.py:976] (1/7) Epoch 16, batch 2100, loss[loss=0.1835, simple_loss=0.2454, pruned_loss=0.06079, over 4878.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2498, pruned_loss=0.05773, over 955567.48 frames. ], batch size: 31, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:24:41,375 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88018.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:24:44,989 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4378, 1.3897, 1.9820, 2.8484, 1.8893, 2.1716, 1.0377, 2.3157], + device='cuda:1'), covar=tensor([0.1811, 0.1421, 0.1091, 0.0508, 0.0860, 0.1493, 0.1676, 0.0583], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0134, 0.0164, 0.0101, 0.0139, 0.0125, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:24:56,042 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0 +2023-03-26 19:24:59,877 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.339e+01 1.671e+02 1.963e+02 2.403e+02 4.597e+02, threshold=3.926e+02, percent-clipped=2.0 +2023-03-26 19:25:13,434 INFO [finetune.py:976] (1/7) Epoch 16, batch 2150, loss[loss=0.2022, simple_loss=0.2738, pruned_loss=0.06534, over 4901.00 frames. ], tot_loss[loss=0.1857, simple_loss=0.2532, pruned_loss=0.0591, over 954960.22 frames. ], batch size: 43, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:25:13,499 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=88066.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:25:30,388 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-03-26 19:25:35,013 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88100.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:25:44,869 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7172, 3.8238, 3.5947, 2.0042, 3.9983, 2.8978, 0.9267, 2.6664], + device='cuda:1'), covar=tensor([0.2584, 0.1973, 0.1553, 0.3050, 0.0902, 0.0949, 0.4366, 0.1454], + device='cuda:1'), in_proj_covar=tensor([0.0149, 0.0173, 0.0158, 0.0127, 0.0155, 0.0121, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 19:25:46,508 INFO [finetune.py:976] (1/7) Epoch 16, batch 2200, loss[loss=0.1669, simple_loss=0.2389, pruned_loss=0.04742, over 4814.00 frames. ], tot_loss[loss=0.1873, simple_loss=0.2555, pruned_loss=0.05952, over 955531.24 frames. ], batch size: 38, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:25:47,207 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88117.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:26:05,785 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.079e+02 1.658e+02 1.940e+02 2.471e+02 6.986e+02, threshold=3.880e+02, percent-clipped=3.0 +2023-03-26 19:26:15,392 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88161.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:26:15,439 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88161.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:26:18,734 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=88165.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:26:19,283 INFO [finetune.py:976] (1/7) Epoch 16, batch 2250, loss[loss=0.219, simple_loss=0.2896, pruned_loss=0.07422, over 4865.00 frames. ], tot_loss[loss=0.1879, simple_loss=0.2565, pruned_loss=0.05961, over 955911.85 frames. ], batch size: 34, lr: 3.45e-03, grad_scale: 32.0 +2023-03-26 19:26:56,473 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=88209.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:27:05,825 INFO [finetune.py:976] (1/7) Epoch 16, batch 2300, loss[loss=0.1803, simple_loss=0.2379, pruned_loss=0.06138, over 4711.00 frames. ], tot_loss[loss=0.1884, simple_loss=0.2574, pruned_loss=0.05967, over 956873.72 frames. ], batch size: 23, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:27:29,754 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88240.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:27:33,304 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88245.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:27:34,392 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.097e+02 1.462e+02 1.840e+02 2.103e+02 3.666e+02, threshold=3.679e+02, percent-clipped=0.0 +2023-03-26 19:27:43,885 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88261.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:27:47,285 INFO [finetune.py:976] (1/7) Epoch 16, batch 2350, loss[loss=0.127, simple_loss=0.2005, pruned_loss=0.0268, over 4703.00 frames. ], tot_loss[loss=0.1848, simple_loss=0.2539, pruned_loss=0.05787, over 956439.59 frames. ], batch size: 23, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:27:50,968 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0688, 1.9046, 1.6655, 1.9955, 1.8390, 1.8517, 1.8492, 2.6437], + device='cuda:1'), covar=tensor([0.3965, 0.4713, 0.3480, 0.3942, 0.4092, 0.2565, 0.4257, 0.1621], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0225, 0.0274, 0.0247, 0.0214, 0.0249, 0.0226], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:28:01,661 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=88288.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:28:03,565 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88291.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 19:28:04,696 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=88293.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:28:07,105 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1205, 1.9218, 1.6061, 1.8015, 1.9036, 1.8798, 1.9493, 2.6529], + device='cuda:1'), covar=tensor([0.3946, 0.3989, 0.3290, 0.3776, 0.3876, 0.2492, 0.3471, 0.1730], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0225, 0.0274, 0.0247, 0.0214, 0.0249, 0.0226], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:28:15,423 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=88309.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:28:20,094 INFO [finetune.py:976] (1/7) Epoch 16, batch 2400, loss[loss=0.1854, simple_loss=0.2483, pruned_loss=0.06125, over 4816.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2505, pruned_loss=0.05664, over 955816.93 frames. ], batch size: 51, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:28:40,399 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.006e+02 1.525e+02 1.766e+02 2.106e+02 3.774e+02, threshold=3.532e+02, percent-clipped=1.0 +2023-03-26 19:28:44,079 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88352.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 19:28:52,861 INFO [finetune.py:976] (1/7) Epoch 16, batch 2450, loss[loss=0.1931, simple_loss=0.2562, pruned_loss=0.06495, over 4823.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.248, pruned_loss=0.05607, over 955829.43 frames. ], batch size: 38, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:29:26,833 INFO [finetune.py:976] (1/7) Epoch 16, batch 2500, loss[loss=0.1857, simple_loss=0.2647, pruned_loss=0.05337, over 4827.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2497, pruned_loss=0.05713, over 953980.94 frames. ], batch size: 40, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:29:48,234 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.125e+02 1.743e+02 2.000e+02 2.601e+02 5.270e+02, threshold=4.000e+02, percent-clipped=5.0 +2023-03-26 19:29:54,227 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=88456.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:30:00,708 INFO [finetune.py:976] (1/7) Epoch 16, batch 2550, loss[loss=0.1525, simple_loss=0.2256, pruned_loss=0.03976, over 4879.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.2532, pruned_loss=0.05804, over 954118.81 frames. ], batch size: 32, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:30:33,907 INFO [finetune.py:976] (1/7) Epoch 16, batch 2600, loss[loss=0.2103, simple_loss=0.2927, pruned_loss=0.06392, over 4811.00 frames. ], tot_loss[loss=0.1861, simple_loss=0.255, pruned_loss=0.05863, over 950776.94 frames. ], batch size: 38, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:30:34,172 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-26 19:30:55,463 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.436e+01 1.666e+02 1.940e+02 2.279e+02 3.712e+02, threshold=3.880e+02, percent-clipped=0.0 +2023-03-26 19:30:59,282 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0823, 2.0146, 1.6255, 2.0324, 2.0129, 1.7179, 2.3881, 2.1139], + device='cuda:1'), covar=tensor([0.1364, 0.2194, 0.3138, 0.2893, 0.2640, 0.1707, 0.3088, 0.1813], + device='cuda:1'), in_proj_covar=tensor([0.0182, 0.0187, 0.0234, 0.0252, 0.0244, 0.0201, 0.0212, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:31:07,484 INFO [finetune.py:976] (1/7) Epoch 16, batch 2650, loss[loss=0.1775, simple_loss=0.2509, pruned_loss=0.0521, over 4891.00 frames. ], tot_loss[loss=0.187, simple_loss=0.256, pruned_loss=0.05897, over 951706.84 frames. ], batch size: 35, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:31:12,342 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7918, 3.3664, 3.4866, 3.6875, 3.5574, 3.3097, 3.8624, 1.2435], + device='cuda:1'), covar=tensor([0.0830, 0.0866, 0.0938, 0.0957, 0.1296, 0.1696, 0.0864, 0.5280], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0245, 0.0277, 0.0292, 0.0335, 0.0283, 0.0298, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:31:41,338 INFO [finetune.py:976] (1/7) Epoch 16, batch 2700, loss[loss=0.1793, simple_loss=0.2448, pruned_loss=0.05688, over 4734.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2547, pruned_loss=0.05797, over 953949.58 frames. ], batch size: 54, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:31:50,280 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7941, 1.6313, 2.1209, 1.3878, 1.8516, 2.0117, 1.5876, 2.2603], + device='cuda:1'), covar=tensor([0.1286, 0.1795, 0.1181, 0.1738, 0.0884, 0.1291, 0.2526, 0.0800], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0204, 0.0192, 0.0189, 0.0177, 0.0213, 0.0217, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:32:05,538 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.033e+02 1.504e+02 1.815e+02 2.296e+02 4.078e+02, threshold=3.631e+02, percent-clipped=1.0 +2023-03-26 19:32:05,618 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=88647.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:32:26,939 INFO [finetune.py:976] (1/7) Epoch 16, batch 2750, loss[loss=0.1681, simple_loss=0.2366, pruned_loss=0.04978, over 4887.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2509, pruned_loss=0.05679, over 954044.60 frames. ], batch size: 32, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:32:42,717 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.65 vs. limit=2.0 +2023-03-26 19:32:46,901 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 19:33:17,052 INFO [finetune.py:976] (1/7) Epoch 16, batch 2800, loss[loss=0.1861, simple_loss=0.2569, pruned_loss=0.05762, over 4737.00 frames. ], tot_loss[loss=0.181, simple_loss=0.2489, pruned_loss=0.05654, over 954654.00 frames. ], batch size: 27, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:33:37,857 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.207e+02 1.516e+02 1.815e+02 2.286e+02 3.246e+02, threshold=3.631e+02, percent-clipped=0.0 +2023-03-26 19:33:43,830 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 19:33:44,962 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88756.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:33:50,910 INFO [finetune.py:976] (1/7) Epoch 16, batch 2850, loss[loss=0.1821, simple_loss=0.2552, pruned_loss=0.05454, over 4781.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2464, pruned_loss=0.05516, over 955658.56 frames. ], batch size: 59, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:33:58,347 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.72 vs. limit=5.0 +2023-03-26 19:34:17,621 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=88804.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:34:24,821 INFO [finetune.py:976] (1/7) Epoch 16, batch 2900, loss[loss=0.1825, simple_loss=0.2647, pruned_loss=0.05022, over 4852.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2492, pruned_loss=0.0562, over 954187.31 frames. ], batch size: 47, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:34:39,340 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4094, 1.4809, 1.5994, 1.5285, 1.6059, 3.0489, 1.3587, 1.5957], + device='cuda:1'), covar=tensor([0.1027, 0.1787, 0.1079, 0.1031, 0.1638, 0.0301, 0.1549, 0.1775], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0073, 0.0077, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:34:44,084 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8021, 1.3203, 0.9675, 1.7666, 2.2124, 1.4099, 1.6190, 1.5980], + device='cuda:1'), covar=tensor([0.1439, 0.2099, 0.1887, 0.1171, 0.1786, 0.1823, 0.1444, 0.1965], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0096, 0.0111, 0.0093, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:34:45,211 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.025e+02 1.632e+02 1.968e+02 2.500e+02 4.348e+02, threshold=3.936e+02, percent-clipped=6.0 +2023-03-26 19:34:58,812 INFO [finetune.py:976] (1/7) Epoch 16, batch 2950, loss[loss=0.1886, simple_loss=0.2642, pruned_loss=0.05646, over 4904.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2507, pruned_loss=0.05643, over 953886.44 frames. ], batch size: 36, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:35:02,006 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6815, 1.6493, 1.5101, 1.6451, 1.2309, 4.0382, 1.5549, 1.9089], + device='cuda:1'), covar=tensor([0.3255, 0.2402, 0.2135, 0.2300, 0.1841, 0.0152, 0.2582, 0.1264], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0120, 0.0124, 0.0114, 0.0097, 0.0096, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:35:32,643 INFO [finetune.py:976] (1/7) Epoch 16, batch 3000, loss[loss=0.2484, simple_loss=0.3171, pruned_loss=0.08989, over 4841.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.253, pruned_loss=0.05771, over 955077.94 frames. ], batch size: 44, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:35:32,643 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 19:35:49,227 INFO [finetune.py:1010] (1/7) Epoch 16, validation: loss=0.1563, simple_loss=0.2263, pruned_loss=0.04316, over 2265189.00 frames. +2023-03-26 19:35:49,227 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 19:35:54,021 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1886, 1.5723, 0.9779, 2.1077, 2.4295, 1.8772, 1.8991, 1.9228], + device='cuda:1'), covar=tensor([0.1385, 0.1996, 0.1887, 0.1077, 0.1775, 0.1783, 0.1335, 0.1889], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0095, 0.0111, 0.0093, 0.0119, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:36:10,682 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.136e+02 1.661e+02 1.990e+02 2.439e+02 3.546e+02, threshold=3.980e+02, percent-clipped=0.0 +2023-03-26 19:36:10,785 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88947.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:36:23,201 INFO [finetune.py:976] (1/7) Epoch 16, batch 3050, loss[loss=0.1688, simple_loss=0.2518, pruned_loss=0.04286, over 4822.00 frames. ], tot_loss[loss=0.185, simple_loss=0.2542, pruned_loss=0.05789, over 954831.66 frames. ], batch size: 38, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:36:43,519 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=88995.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:36:57,478 INFO [finetune.py:976] (1/7) Epoch 16, batch 3100, loss[loss=0.157, simple_loss=0.2292, pruned_loss=0.04239, over 4791.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2532, pruned_loss=0.0577, over 955490.13 frames. ], batch size: 29, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:37:06,377 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89027.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:37:10,258 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-26 19:37:14,839 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8577, 4.0656, 3.9600, 1.9643, 4.1345, 3.0720, 1.0580, 2.9980], + device='cuda:1'), covar=tensor([0.2256, 0.1541, 0.1252, 0.3068, 0.0824, 0.0946, 0.4052, 0.1283], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0176, 0.0160, 0.0129, 0.0158, 0.0124, 0.0148, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 19:37:20,959 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.966e+01 1.506e+02 1.838e+02 2.198e+02 3.411e+02, threshold=3.676e+02, percent-clipped=0.0 +2023-03-26 19:37:33,687 INFO [finetune.py:976] (1/7) Epoch 16, batch 3150, loss[loss=0.1772, simple_loss=0.2542, pruned_loss=0.05012, over 4855.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2509, pruned_loss=0.057, over 955851.10 frames. ], batch size: 44, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:37:33,768 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1524, 3.6294, 3.7922, 4.0362, 3.9354, 3.6390, 4.2172, 1.3151], + device='cuda:1'), covar=tensor([0.0808, 0.0928, 0.0846, 0.0882, 0.1245, 0.1558, 0.0692, 0.5668], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0246, 0.0277, 0.0294, 0.0335, 0.0283, 0.0298, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:37:56,585 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89088.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:38:22,856 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.52 vs. limit=5.0 +2023-03-26 19:38:25,684 INFO [finetune.py:976] (1/7) Epoch 16, batch 3200, loss[loss=0.114, simple_loss=0.1842, pruned_loss=0.02194, over 4840.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2475, pruned_loss=0.05535, over 958380.38 frames. ], batch size: 25, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:38:50,100 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.129e+02 1.609e+02 1.908e+02 2.339e+02 4.086e+02, threshold=3.816e+02, percent-clipped=1.0 +2023-03-26 19:39:02,086 INFO [finetune.py:976] (1/7) Epoch 16, batch 3250, loss[loss=0.2114, simple_loss=0.2765, pruned_loss=0.07318, over 4830.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2483, pruned_loss=0.0558, over 958087.69 frames. ], batch size: 39, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:39:04,012 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89169.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:39:35,940 INFO [finetune.py:976] (1/7) Epoch 16, batch 3300, loss[loss=0.1543, simple_loss=0.2253, pruned_loss=0.04164, over 4755.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.2535, pruned_loss=0.05791, over 959048.59 frames. ], batch size: 28, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:39:40,985 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 19:39:44,489 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7016, 1.2016, 0.7986, 1.5160, 2.0701, 1.1147, 1.4175, 1.5729], + device='cuda:1'), covar=tensor([0.1665, 0.2328, 0.2141, 0.1401, 0.2069, 0.2182, 0.1588, 0.2024], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0096, 0.0111, 0.0093, 0.0119, 0.0095, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:39:45,113 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89230.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:39:47,277 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4316, 3.8878, 4.0529, 4.2761, 4.1631, 3.9695, 4.4992, 1.4659], + device='cuda:1'), covar=tensor([0.0761, 0.0880, 0.0843, 0.1022, 0.1340, 0.1459, 0.0682, 0.5344], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0247, 0.0278, 0.0295, 0.0336, 0.0284, 0.0299, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:39:56,763 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.235e+02 1.765e+02 2.004e+02 2.308e+02 3.942e+02, threshold=4.007e+02, percent-clipped=1.0 +2023-03-26 19:40:04,629 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1458, 1.7822, 2.1265, 2.0855, 1.8042, 1.8170, 2.0799, 1.9479], + device='cuda:1'), covar=tensor([0.4379, 0.4328, 0.3374, 0.4333, 0.5422, 0.4080, 0.4873, 0.3467], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0238, 0.0258, 0.0271, 0.0269, 0.0243, 0.0279, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:40:09,186 INFO [finetune.py:976] (1/7) Epoch 16, batch 3350, loss[loss=0.1693, simple_loss=0.242, pruned_loss=0.04829, over 4736.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2557, pruned_loss=0.05893, over 958806.35 frames. ], batch size: 54, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:40:33,043 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 19:40:42,689 INFO [finetune.py:976] (1/7) Epoch 16, batch 3400, loss[loss=0.1637, simple_loss=0.239, pruned_loss=0.04416, over 4792.00 frames. ], tot_loss[loss=0.1891, simple_loss=0.2585, pruned_loss=0.05985, over 958684.08 frames. ], batch size: 51, lr: 3.44e-03, grad_scale: 32.0 +2023-03-26 19:41:12,550 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.032e+02 1.581e+02 1.832e+02 2.219e+02 5.301e+02, threshold=3.664e+02, percent-clipped=1.0 +2023-03-26 19:41:24,424 INFO [finetune.py:976] (1/7) Epoch 16, batch 3450, loss[loss=0.1624, simple_loss=0.2296, pruned_loss=0.04767, over 4901.00 frames. ], tot_loss[loss=0.1889, simple_loss=0.2583, pruned_loss=0.05973, over 957143.20 frames. ], batch size: 35, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:41:28,725 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-26 19:41:29,811 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89374.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:41:35,737 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89383.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:41:35,760 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6935, 1.5811, 2.1549, 3.4812, 2.4529, 2.4797, 1.0044, 2.7978], + device='cuda:1'), covar=tensor([0.1655, 0.1362, 0.1269, 0.0534, 0.0735, 0.1309, 0.1806, 0.0477], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0134, 0.0166, 0.0101, 0.0139, 0.0125, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:41:55,395 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1151, 3.6324, 3.7570, 3.9495, 3.8651, 3.6408, 4.2064, 1.4097], + device='cuda:1'), covar=tensor([0.0700, 0.0881, 0.0803, 0.0920, 0.1183, 0.1488, 0.0688, 0.4932], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0246, 0.0277, 0.0295, 0.0335, 0.0283, 0.0299, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:41:58,330 INFO [finetune.py:976] (1/7) Epoch 16, batch 3500, loss[loss=0.1787, simple_loss=0.2352, pruned_loss=0.06116, over 4825.00 frames. ], tot_loss[loss=0.1878, simple_loss=0.2562, pruned_loss=0.05975, over 955282.17 frames. ], batch size: 33, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:42:04,421 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89425.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:42:10,964 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89435.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:42:13,399 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89439.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:42:18,651 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.936e+01 1.517e+02 1.946e+02 2.225e+02 4.216e+02, threshold=3.891e+02, percent-clipped=3.0 +2023-03-26 19:42:29,435 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8985, 1.8106, 1.9817, 1.2240, 2.0244, 1.9868, 1.9196, 1.6257], + device='cuda:1'), covar=tensor([0.0634, 0.0723, 0.0639, 0.0966, 0.0748, 0.0685, 0.0602, 0.1232], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0136, 0.0144, 0.0125, 0.0125, 0.0143, 0.0144, 0.0167], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:42:31,135 INFO [finetune.py:976] (1/7) Epoch 16, batch 3550, loss[loss=0.2021, simple_loss=0.2717, pruned_loss=0.06626, over 4928.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2515, pruned_loss=0.05777, over 956079.76 frames. ], batch size: 43, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:42:43,947 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89486.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:42:53,369 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89500.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:42:59,273 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89509.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:43:06,067 INFO [finetune.py:976] (1/7) Epoch 16, batch 3600, loss[loss=0.1488, simple_loss=0.2181, pruned_loss=0.0397, over 4870.00 frames. ], tot_loss[loss=0.1806, simple_loss=0.2486, pruned_loss=0.05637, over 958266.22 frames. ], batch size: 34, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:43:14,211 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89525.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:43:36,202 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.57 vs. limit=2.0 +2023-03-26 19:43:37,733 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.924e+01 1.526e+02 1.890e+02 2.215e+02 3.895e+02, threshold=3.780e+02, percent-clipped=1.0 +2023-03-26 19:43:51,622 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 19:44:03,076 INFO [finetune.py:976] (1/7) Epoch 16, batch 3650, loss[loss=0.1754, simple_loss=0.2385, pruned_loss=0.0562, over 4698.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2497, pruned_loss=0.05676, over 955329.28 frames. ], batch size: 23, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:44:06,112 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89570.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:44:09,334 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-26 19:44:27,365 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8915, 1.7568, 1.5320, 1.4135, 1.9276, 1.6472, 1.8463, 1.8528], + device='cuda:1'), covar=tensor([0.1441, 0.2011, 0.3269, 0.2424, 0.2627, 0.1721, 0.2724, 0.1805], + device='cuda:1'), in_proj_covar=tensor([0.0182, 0.0188, 0.0234, 0.0253, 0.0244, 0.0202, 0.0211, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:44:36,728 INFO [finetune.py:976] (1/7) Epoch 16, batch 3700, loss[loss=0.1947, simple_loss=0.2665, pruned_loss=0.06143, over 4756.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2522, pruned_loss=0.05728, over 956577.01 frames. ], batch size: 27, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:44:47,062 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3936, 2.0842, 2.7922, 1.5942, 2.3995, 2.7277, 1.8957, 2.9014], + device='cuda:1'), covar=tensor([0.1380, 0.2143, 0.1614, 0.2425, 0.1013, 0.1477, 0.2793, 0.0873], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0203, 0.0190, 0.0189, 0.0176, 0.0211, 0.0216, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:44:50,580 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5089, 1.5523, 2.0311, 1.8421, 1.8079, 3.9332, 1.4522, 1.6926], + device='cuda:1'), covar=tensor([0.1011, 0.1824, 0.1308, 0.0991, 0.1676, 0.0199, 0.1656, 0.1900], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0073, 0.0077, 0.0092, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:44:56,689 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-03-26 19:44:57,073 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.548e+01 1.593e+02 1.994e+02 2.376e+02 3.738e+02, threshold=3.989e+02, percent-clipped=0.0 +2023-03-26 19:45:03,064 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6358, 1.4854, 1.9140, 1.2934, 1.6382, 1.8533, 1.4600, 2.0622], + device='cuda:1'), covar=tensor([0.1128, 0.1948, 0.1176, 0.1632, 0.0866, 0.1098, 0.2526, 0.0702], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0203, 0.0190, 0.0189, 0.0175, 0.0211, 0.0216, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:45:04,293 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0786, 2.0636, 2.1378, 1.4225, 2.0912, 2.2654, 2.2202, 1.7956], + device='cuda:1'), covar=tensor([0.0662, 0.0649, 0.0706, 0.0967, 0.0703, 0.0705, 0.0639, 0.1153], + device='cuda:1'), in_proj_covar=tensor([0.0137, 0.0137, 0.0145, 0.0126, 0.0126, 0.0143, 0.0144, 0.0168], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:45:10,208 INFO [finetune.py:976] (1/7) Epoch 16, batch 3750, loss[loss=0.22, simple_loss=0.2917, pruned_loss=0.07413, over 4904.00 frames. ], tot_loss[loss=0.1849, simple_loss=0.2542, pruned_loss=0.05783, over 955748.31 frames. ], batch size: 37, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:45:19,312 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89680.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:45:21,100 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89683.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:45:43,312 INFO [finetune.py:976] (1/7) Epoch 16, batch 3800, loss[loss=0.1501, simple_loss=0.2299, pruned_loss=0.03517, over 4757.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2552, pruned_loss=0.05801, over 955067.86 frames. ], batch size: 27, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:45:52,762 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89730.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:45:53,366 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=89731.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:46:00,037 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89741.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:46:03,522 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.091e+02 1.549e+02 1.882e+02 2.353e+02 4.344e+02, threshold=3.764e+02, percent-clipped=2.0 +2023-03-26 19:46:09,911 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 19:46:19,023 INFO [finetune.py:976] (1/7) Epoch 16, batch 3850, loss[loss=0.1777, simple_loss=0.2453, pruned_loss=0.05504, over 4886.00 frames. ], tot_loss[loss=0.184, simple_loss=0.2535, pruned_loss=0.05718, over 956169.68 frames. ], batch size: 32, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:46:29,108 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89781.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:46:38,078 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89795.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 19:46:38,101 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7474, 1.3033, 0.8644, 1.6523, 2.0661, 1.5500, 1.4662, 1.6309], + device='cuda:1'), covar=tensor([0.1734, 0.2629, 0.2411, 0.1425, 0.2215, 0.2514, 0.1877, 0.2333], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0095, 0.0111, 0.0093, 0.0119, 0.0095, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 19:46:38,743 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7464, 1.5792, 1.4301, 1.8166, 2.3416, 1.8548, 1.6899, 1.4248], + device='cuda:1'), covar=tensor([0.2312, 0.2339, 0.2201, 0.1786, 0.1670, 0.1358, 0.2463, 0.2241], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0208, 0.0210, 0.0191, 0.0242, 0.0184, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:46:39,353 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6348, 2.4618, 1.9171, 1.0062, 2.1488, 2.0080, 1.8227, 2.2077], + device='cuda:1'), covar=tensor([0.1029, 0.0753, 0.1761, 0.2116, 0.1562, 0.2360, 0.2300, 0.0994], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0195, 0.0200, 0.0182, 0.0213, 0.0207, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:46:44,627 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2781, 2.8800, 2.9952, 3.1962, 3.0519, 2.8815, 3.3070, 0.9960], + device='cuda:1'), covar=tensor([0.1009, 0.0979, 0.1049, 0.1086, 0.1478, 0.1583, 0.1063, 0.5015], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0245, 0.0276, 0.0293, 0.0332, 0.0281, 0.0298, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:46:52,158 INFO [finetune.py:976] (1/7) Epoch 16, batch 3900, loss[loss=0.2219, simple_loss=0.288, pruned_loss=0.07785, over 4823.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2515, pruned_loss=0.05685, over 958718.14 frames. ], batch size: 30, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:46:58,247 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89825.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:47:12,472 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.084e+02 1.500e+02 1.857e+02 2.274e+02 5.172e+02, threshold=3.715e+02, percent-clipped=1.0 +2023-03-26 19:47:15,644 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7352, 1.6149, 1.5532, 1.5827, 2.0628, 1.9346, 1.6678, 1.5316], + device='cuda:1'), covar=tensor([0.0330, 0.0323, 0.0516, 0.0309, 0.0195, 0.0406, 0.0350, 0.0384], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0109, 0.0145, 0.0114, 0.0101, 0.0109, 0.0100, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.4242e-05, 8.4452e-05, 1.1452e-04, 8.7791e-05, 7.8347e-05, 8.0138e-05, + 7.4841e-05, 8.3367e-05], device='cuda:1') +2023-03-26 19:47:23,888 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89865.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:47:23,923 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89865.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:47:24,440 INFO [finetune.py:976] (1/7) Epoch 16, batch 3950, loss[loss=0.1438, simple_loss=0.2157, pruned_loss=0.03595, over 4896.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2484, pruned_loss=0.05566, over 959511.96 frames. ], batch size: 32, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:47:29,792 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=89873.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:47:30,483 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1893, 2.1594, 1.7819, 2.0373, 2.2186, 1.9310, 2.5422, 2.2114], + device='cuda:1'), covar=tensor([0.1435, 0.2171, 0.2990, 0.2777, 0.2548, 0.1727, 0.2724, 0.1891], + device='cuda:1'), in_proj_covar=tensor([0.0181, 0.0187, 0.0233, 0.0252, 0.0243, 0.0201, 0.0211, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:47:47,668 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5551, 2.7384, 2.5241, 1.9109, 2.7646, 2.9036, 2.8545, 2.4362], + device='cuda:1'), covar=tensor([0.0638, 0.0528, 0.0729, 0.0866, 0.0545, 0.0652, 0.0551, 0.0929], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0137, 0.0144, 0.0126, 0.0126, 0.0143, 0.0144, 0.0167], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:47:57,663 INFO [finetune.py:976] (1/7) Epoch 16, batch 4000, loss[loss=0.2116, simple_loss=0.2766, pruned_loss=0.07333, over 4929.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2479, pruned_loss=0.05548, over 958761.01 frames. ], batch size: 38, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:48:04,736 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89926.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 19:48:08,779 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0612, 1.9020, 2.0049, 1.4648, 2.0886, 2.1181, 2.1467, 1.6248], + device='cuda:1'), covar=tensor([0.0544, 0.0607, 0.0669, 0.0881, 0.0588, 0.0607, 0.0517, 0.1102], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0136, 0.0143, 0.0125, 0.0125, 0.0143, 0.0143, 0.0167], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:48:18,302 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.466e+01 1.676e+02 1.974e+02 2.469e+02 4.779e+02, threshold=3.947e+02, percent-clipped=6.0 +2023-03-26 19:48:32,943 INFO [finetune.py:976] (1/7) Epoch 16, batch 4050, loss[loss=0.2279, simple_loss=0.2958, pruned_loss=0.08, over 4813.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2519, pruned_loss=0.05738, over 958694.43 frames. ], batch size: 45, lr: 3.43e-03, grad_scale: 64.0 +2023-03-26 19:49:22,545 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8661, 1.1171, 1.8039, 1.8202, 1.6139, 1.5589, 1.7213, 1.7022], + device='cuda:1'), covar=tensor([0.3240, 0.3580, 0.3401, 0.3375, 0.4530, 0.3545, 0.3904, 0.3055], + device='cuda:1'), in_proj_covar=tensor([0.0249, 0.0240, 0.0260, 0.0272, 0.0271, 0.0245, 0.0281, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:49:29,800 INFO [finetune.py:976] (1/7) Epoch 16, batch 4100, loss[loss=0.1506, simple_loss=0.2276, pruned_loss=0.03682, over 4759.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2537, pruned_loss=0.05755, over 957869.95 frames. ], batch size: 27, lr: 3.43e-03, grad_scale: 64.0 +2023-03-26 19:49:30,002 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.81 vs. limit=5.0 +2023-03-26 19:49:43,248 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90030.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:49:46,954 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90036.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:49:54,440 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.029e+02 1.559e+02 1.839e+02 2.160e+02 6.359e+02, threshold=3.678e+02, percent-clipped=1.0 +2023-03-26 19:50:06,382 INFO [finetune.py:976] (1/7) Epoch 16, batch 4150, loss[loss=0.1659, simple_loss=0.2182, pruned_loss=0.05686, over 4127.00 frames. ], tot_loss[loss=0.1854, simple_loss=0.2544, pruned_loss=0.05818, over 956671.61 frames. ], batch size: 17, lr: 3.43e-03, grad_scale: 64.0 +2023-03-26 19:50:14,190 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=90078.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:50:16,529 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90081.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:50:26,430 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90095.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:50:39,512 INFO [finetune.py:976] (1/7) Epoch 16, batch 4200, loss[loss=0.1828, simple_loss=0.2486, pruned_loss=0.05852, over 4881.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2551, pruned_loss=0.05807, over 956046.81 frames. ], batch size: 32, lr: 3.43e-03, grad_scale: 64.0 +2023-03-26 19:50:44,325 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4046, 1.4198, 1.7743, 1.7578, 1.6124, 3.2248, 1.2485, 1.5136], + device='cuda:1'), covar=tensor([0.1008, 0.1829, 0.1115, 0.0938, 0.1598, 0.0274, 0.1597, 0.1804], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0074, 0.0078, 0.0092, 0.0081, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:50:47,969 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=90129.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:50:48,624 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90130.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:50:57,830 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=90143.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 19:51:00,648 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.087e+02 1.547e+02 1.785e+02 2.134e+02 3.751e+02, threshold=3.570e+02, percent-clipped=1.0 +2023-03-26 19:51:12,326 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90165.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:51:12,839 INFO [finetune.py:976] (1/7) Epoch 16, batch 4250, loss[loss=0.2173, simple_loss=0.268, pruned_loss=0.08334, over 4933.00 frames. ], tot_loss[loss=0.183, simple_loss=0.2521, pruned_loss=0.05698, over 955552.27 frames. ], batch size: 33, lr: 3.43e-03, grad_scale: 64.0 +2023-03-26 19:51:29,547 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90191.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:51:34,308 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0712, 1.9674, 1.6486, 1.8265, 2.0665, 1.7265, 2.2101, 2.0466], + device='cuda:1'), covar=tensor([0.1451, 0.2083, 0.3210, 0.2583, 0.2649, 0.1840, 0.2984, 0.1871], + device='cuda:1'), in_proj_covar=tensor([0.0182, 0.0187, 0.0234, 0.0252, 0.0244, 0.0202, 0.0212, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:51:43,669 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=90213.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:51:45,896 INFO [finetune.py:976] (1/7) Epoch 16, batch 4300, loss[loss=0.2022, simple_loss=0.2574, pruned_loss=0.07354, over 4137.00 frames. ], tot_loss[loss=0.1821, simple_loss=0.2506, pruned_loss=0.0568, over 953650.86 frames. ], batch size: 65, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:51:48,954 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90221.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 19:51:56,175 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90232.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:52:07,166 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.180e+02 1.551e+02 1.797e+02 2.190e+02 3.764e+02, threshold=3.594e+02, percent-clipped=1.0 +2023-03-26 19:52:18,553 INFO [finetune.py:976] (1/7) Epoch 16, batch 4350, loss[loss=0.2027, simple_loss=0.2565, pruned_loss=0.07446, over 4821.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2475, pruned_loss=0.05565, over 955779.61 frames. ], batch size: 30, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:52:34,164 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7746, 1.7219, 1.6189, 1.8006, 1.3350, 3.8173, 1.5395, 1.9366], + device='cuda:1'), covar=tensor([0.3243, 0.2553, 0.2134, 0.2353, 0.1794, 0.0206, 0.2481, 0.1332], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0115, 0.0120, 0.0123, 0.0114, 0.0097, 0.0096, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 19:52:36,980 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90293.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:52:51,897 INFO [finetune.py:976] (1/7) Epoch 16, batch 4400, loss[loss=0.2179, simple_loss=0.2959, pruned_loss=0.06999, over 4811.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2471, pruned_loss=0.05531, over 954547.84 frames. ], batch size: 51, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:53:04,943 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90336.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:53:13,597 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.069e+02 1.586e+02 1.845e+02 2.241e+02 4.760e+02, threshold=3.689e+02, percent-clipped=4.0 +2023-03-26 19:53:14,914 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9986, 1.2770, 1.8901, 1.8695, 1.6553, 1.6504, 1.8181, 1.8096], + device='cuda:1'), covar=tensor([0.3909, 0.4017, 0.3434, 0.3851, 0.5022, 0.4002, 0.4391, 0.3307], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0238, 0.0258, 0.0270, 0.0270, 0.0243, 0.0280, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:53:15,500 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90351.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:53:25,405 INFO [finetune.py:976] (1/7) Epoch 16, batch 4450, loss[loss=0.2041, simple_loss=0.2751, pruned_loss=0.06659, over 4912.00 frames. ], tot_loss[loss=0.1814, simple_loss=0.2502, pruned_loss=0.05629, over 953898.00 frames. ], batch size: 36, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:53:37,365 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=90384.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:54:05,567 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90412.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:54:07,882 INFO [finetune.py:976] (1/7) Epoch 16, batch 4500, loss[loss=0.1674, simple_loss=0.2449, pruned_loss=0.04497, over 4759.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2505, pruned_loss=0.05658, over 949710.60 frames. ], batch size: 26, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:54:23,611 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-26 19:54:40,947 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.141e+02 1.659e+02 2.056e+02 2.631e+02 3.688e+02, threshold=4.111e+02, percent-clipped=0.0 +2023-03-26 19:54:53,410 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90459.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:55:01,923 INFO [finetune.py:976] (1/7) Epoch 16, batch 4550, loss[loss=0.1792, simple_loss=0.2747, pruned_loss=0.04187, over 4894.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2515, pruned_loss=0.05652, over 948998.31 frames. ], batch size: 36, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:55:18,093 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90486.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:55:24,689 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90496.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:55:38,625 INFO [finetune.py:976] (1/7) Epoch 16, batch 4600, loss[loss=0.1849, simple_loss=0.2502, pruned_loss=0.05981, over 4404.00 frames. ], tot_loss[loss=0.1817, simple_loss=0.2512, pruned_loss=0.0561, over 950645.12 frames. ], batch size: 19, lr: 3.43e-03, grad_scale: 32.0 +2023-03-26 19:55:41,654 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90520.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:55:42,192 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90521.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 19:55:59,292 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.556e+01 1.469e+02 1.715e+02 2.012e+02 4.010e+02, threshold=3.429e+02, percent-clipped=0.0 +2023-03-26 19:56:01,264 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0472, 1.5329, 2.0331, 2.0230, 1.8017, 1.7244, 2.0149, 1.8496], + device='cuda:1'), covar=tensor([0.3516, 0.3799, 0.3320, 0.3488, 0.4941, 0.3889, 0.4424, 0.3164], + device='cuda:1'), in_proj_covar=tensor([0.0248, 0.0240, 0.0259, 0.0272, 0.0271, 0.0245, 0.0282, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:56:06,288 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90557.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:56:11,543 INFO [finetune.py:976] (1/7) Epoch 16, batch 4650, loss[loss=0.1715, simple_loss=0.2322, pruned_loss=0.05543, over 4766.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2506, pruned_loss=0.05666, over 951197.33 frames. ], batch size: 28, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 19:56:13,931 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=90569.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:56:26,169 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90588.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:56:45,054 INFO [finetune.py:976] (1/7) Epoch 16, batch 4700, loss[loss=0.1702, simple_loss=0.2379, pruned_loss=0.05131, over 4874.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2476, pruned_loss=0.05575, over 953833.39 frames. ], batch size: 31, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 19:57:01,518 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9971, 4.8872, 4.6141, 2.9811, 5.0336, 3.9107, 1.1830, 3.6335], + device='cuda:1'), covar=tensor([0.2129, 0.1919, 0.1290, 0.2646, 0.0674, 0.0703, 0.4079, 0.1230], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0175, 0.0159, 0.0129, 0.0157, 0.0123, 0.0146, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 19:57:05,662 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.103e+02 1.593e+02 1.858e+02 2.163e+02 3.767e+02, threshold=3.717e+02, percent-clipped=1.0 +2023-03-26 19:57:18,492 INFO [finetune.py:976] (1/7) Epoch 16, batch 4750, loss[loss=0.1552, simple_loss=0.2268, pruned_loss=0.04184, over 4354.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2459, pruned_loss=0.05533, over 952046.08 frames. ], batch size: 19, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 19:57:45,612 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90707.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:57:52,456 INFO [finetune.py:976] (1/7) Epoch 16, batch 4800, loss[loss=0.2771, simple_loss=0.3257, pruned_loss=0.1143, over 4805.00 frames. ], tot_loss[loss=0.181, simple_loss=0.2489, pruned_loss=0.05657, over 951045.27 frames. ], batch size: 41, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 19:58:00,877 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 19:58:13,286 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.231e+02 1.629e+02 1.979e+02 2.321e+02 4.531e+02, threshold=3.957e+02, percent-clipped=1.0 +2023-03-26 19:58:18,761 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3797, 2.3249, 2.1310, 2.5495, 2.9360, 2.5807, 2.4457, 1.8984], + device='cuda:1'), covar=tensor([0.2123, 0.1886, 0.1816, 0.1483, 0.1682, 0.1000, 0.1949, 0.1812], + device='cuda:1'), in_proj_covar=tensor([0.0241, 0.0209, 0.0211, 0.0191, 0.0243, 0.0186, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:58:25,071 INFO [finetune.py:976] (1/7) Epoch 16, batch 4850, loss[loss=0.1928, simple_loss=0.2629, pruned_loss=0.06134, over 4779.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2516, pruned_loss=0.0568, over 952139.08 frames. ], batch size: 28, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 19:58:39,070 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90786.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:58:57,852 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90815.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:58:58,402 INFO [finetune.py:976] (1/7) Epoch 16, batch 4900, loss[loss=0.1946, simple_loss=0.262, pruned_loss=0.06364, over 4860.00 frames. ], tot_loss[loss=0.1854, simple_loss=0.2543, pruned_loss=0.05822, over 954019.57 frames. ], batch size: 34, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 19:59:04,115 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8806, 1.7620, 1.5244, 1.5902, 1.9404, 1.6296, 1.8653, 1.9012], + device='cuda:1'), covar=tensor([0.1431, 0.1964, 0.2975, 0.2323, 0.2461, 0.1717, 0.2918, 0.1673], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0187, 0.0234, 0.0253, 0.0245, 0.0202, 0.0212, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:59:11,163 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=90834.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:59:11,785 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1981, 3.6494, 3.8049, 4.0832, 3.9633, 3.6628, 4.2518, 1.3183], + device='cuda:1'), covar=tensor([0.0758, 0.0913, 0.0843, 0.0916, 0.1216, 0.1659, 0.0741, 0.5537], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0242, 0.0274, 0.0291, 0.0331, 0.0278, 0.0296, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:59:24,120 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.061e+02 1.543e+02 1.926e+02 2.205e+02 3.945e+02, threshold=3.852e+02, percent-clipped=0.0 +2023-03-26 19:59:25,324 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90849.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:59:26,597 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2499, 2.2454, 1.8548, 2.2735, 2.1334, 2.1036, 2.1225, 3.1132], + device='cuda:1'), covar=tensor([0.3880, 0.4507, 0.3494, 0.4443, 0.4545, 0.2466, 0.4516, 0.1551], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0225, 0.0275, 0.0248, 0.0216, 0.0249, 0.0228], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 19:59:27,116 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90852.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 19:59:42,911 INFO [finetune.py:976] (1/7) Epoch 16, batch 4950, loss[loss=0.2005, simple_loss=0.2688, pruned_loss=0.06609, over 4734.00 frames. ], tot_loss[loss=0.1863, simple_loss=0.2556, pruned_loss=0.05848, over 952491.00 frames. ], batch size: 59, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:00:12,900 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90888.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:00:34,905 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90910.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:00:38,891 INFO [finetune.py:976] (1/7) Epoch 16, batch 5000, loss[loss=0.151, simple_loss=0.2099, pruned_loss=0.04611, over 4186.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2527, pruned_loss=0.05712, over 952441.32 frames. ], batch size: 17, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:00:53,036 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=90936.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:01:00,204 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.882e+01 1.531e+02 1.843e+02 2.134e+02 5.620e+02, threshold=3.687e+02, percent-clipped=2.0 +2023-03-26 20:01:11,975 INFO [finetune.py:976] (1/7) Epoch 16, batch 5050, loss[loss=0.1694, simple_loss=0.2326, pruned_loss=0.05304, over 4758.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2511, pruned_loss=0.05697, over 953889.31 frames. ], batch size: 26, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:01:40,196 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91007.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:01:45,573 INFO [finetune.py:976] (1/7) Epoch 16, batch 5100, loss[loss=0.1803, simple_loss=0.2329, pruned_loss=0.06389, over 4838.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2477, pruned_loss=0.05561, over 954764.94 frames. ], batch size: 30, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:02:07,770 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.054e+02 1.471e+02 1.778e+02 2.096e+02 3.940e+02, threshold=3.556e+02, percent-clipped=2.0 +2023-03-26 20:02:12,113 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=91055.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:02:19,222 INFO [finetune.py:976] (1/7) Epoch 16, batch 5150, loss[loss=0.2324, simple_loss=0.2935, pruned_loss=0.08571, over 4831.00 frames. ], tot_loss[loss=0.18, simple_loss=0.2482, pruned_loss=0.05589, over 953257.31 frames. ], batch size: 40, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:02:52,455 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91115.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:02:52,957 INFO [finetune.py:976] (1/7) Epoch 16, batch 5200, loss[loss=0.1944, simple_loss=0.2668, pruned_loss=0.06097, over 4797.00 frames. ], tot_loss[loss=0.1836, simple_loss=0.252, pruned_loss=0.05764, over 949928.84 frames. ], batch size: 29, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:03:14,342 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.009e+02 1.594e+02 1.966e+02 2.465e+02 4.658e+02, threshold=3.932e+02, percent-clipped=3.0 +2023-03-26 20:03:17,838 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91152.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:03:24,389 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=91163.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:03:26,645 INFO [finetune.py:976] (1/7) Epoch 16, batch 5250, loss[loss=0.1513, simple_loss=0.205, pruned_loss=0.04879, over 4057.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2531, pruned_loss=0.05779, over 948856.62 frames. ], batch size: 17, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:03:49,259 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=91200.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:03:53,266 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91205.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:03:59,840 INFO [finetune.py:976] (1/7) Epoch 16, batch 5300, loss[loss=0.1947, simple_loss=0.2761, pruned_loss=0.05663, over 4846.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2531, pruned_loss=0.05741, over 948262.12 frames. ], batch size: 44, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:04:00,037 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.38 vs. limit=5.0 +2023-03-26 20:04:02,265 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0994, 1.8305, 2.4897, 3.9838, 2.7285, 2.7247, 0.7186, 3.1866], + device='cuda:1'), covar=tensor([0.1676, 0.1374, 0.1342, 0.0486, 0.0796, 0.1361, 0.2133, 0.0494], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0134, 0.0166, 0.0102, 0.0138, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:04:15,514 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-26 20:04:21,113 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.091e+02 1.597e+02 1.843e+02 2.222e+02 3.769e+02, threshold=3.686e+02, percent-clipped=0.0 +2023-03-26 20:04:33,500 INFO [finetune.py:976] (1/7) Epoch 16, batch 5350, loss[loss=0.1841, simple_loss=0.2506, pruned_loss=0.05879, over 4919.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2538, pruned_loss=0.05725, over 946826.36 frames. ], batch size: 38, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:04:50,623 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91285.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:05:29,812 INFO [finetune.py:976] (1/7) Epoch 16, batch 5400, loss[loss=0.1686, simple_loss=0.2358, pruned_loss=0.05074, over 4793.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.251, pruned_loss=0.0564, over 947789.08 frames. ], batch size: 29, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:05:58,068 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2781, 1.9587, 2.5315, 4.2686, 3.0324, 2.8753, 1.0791, 3.4811], + device='cuda:1'), covar=tensor([0.1476, 0.1397, 0.1313, 0.0435, 0.0670, 0.1371, 0.1825, 0.0366], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0165, 0.0101, 0.0138, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:06:01,741 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91346.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:06:03,346 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.781e+01 1.583e+02 1.812e+02 2.291e+02 3.767e+02, threshold=3.624e+02, percent-clipped=1.0 +2023-03-26 20:06:15,750 INFO [finetune.py:976] (1/7) Epoch 16, batch 5450, loss[loss=0.1609, simple_loss=0.2304, pruned_loss=0.0457, over 4937.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2507, pruned_loss=0.05707, over 950879.52 frames. ], batch size: 38, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:06:49,418 INFO [finetune.py:976] (1/7) Epoch 16, batch 5500, loss[loss=0.1613, simple_loss=0.2251, pruned_loss=0.04881, over 4814.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.2458, pruned_loss=0.05492, over 953724.87 frames. ], batch size: 25, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:07:10,222 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.789e+01 1.460e+02 1.744e+02 2.187e+02 6.443e+02, threshold=3.488e+02, percent-clipped=1.0 +2023-03-26 20:07:17,451 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3322, 2.1892, 1.7721, 0.9025, 1.9359, 1.8136, 1.6853, 2.0234], + device='cuda:1'), covar=tensor([0.0891, 0.0719, 0.1466, 0.1956, 0.1374, 0.1854, 0.1932, 0.0899], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0194, 0.0198, 0.0182, 0.0211, 0.0205, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:07:22,092 INFO [finetune.py:976] (1/7) Epoch 16, batch 5550, loss[loss=0.16, simple_loss=0.2304, pruned_loss=0.04478, over 4779.00 frames. ], tot_loss[loss=0.1806, simple_loss=0.2483, pruned_loss=0.0565, over 953969.67 frames. ], batch size: 28, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:07:46,396 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6555, 3.8921, 3.6341, 1.8108, 3.9080, 2.9196, 0.9426, 2.6703], + device='cuda:1'), covar=tensor([0.2616, 0.2044, 0.1334, 0.3551, 0.1002, 0.1101, 0.4422, 0.1574], + device='cuda:1'), in_proj_covar=tensor([0.0149, 0.0173, 0.0157, 0.0127, 0.0156, 0.0121, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 20:07:47,603 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91505.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:07:53,906 INFO [finetune.py:976] (1/7) Epoch 16, batch 5600, loss[loss=0.2217, simple_loss=0.2988, pruned_loss=0.07229, over 4910.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2514, pruned_loss=0.05673, over 956490.07 frames. ], batch size: 36, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:08:08,981 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91542.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 20:08:13,217 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.075e+02 1.603e+02 1.969e+02 2.458e+02 5.397e+02, threshold=3.938e+02, percent-clipped=5.0 +2023-03-26 20:08:16,171 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=91553.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:08:23,625 INFO [finetune.py:976] (1/7) Epoch 16, batch 5650, loss[loss=0.2555, simple_loss=0.3221, pruned_loss=0.09441, over 4850.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.254, pruned_loss=0.05717, over 955790.72 frames. ], batch size: 47, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:08:45,660 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91603.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 20:08:53,175 INFO [finetune.py:976] (1/7) Epoch 16, batch 5700, loss[loss=0.1456, simple_loss=0.2095, pruned_loss=0.04084, over 4409.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2506, pruned_loss=0.05661, over 937435.17 frames. ], batch size: 19, lr: 3.42e-03, grad_scale: 32.0 +2023-03-26 20:09:04,626 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-26 20:09:07,888 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91641.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:09:21,367 INFO [finetune.py:976] (1/7) Epoch 17, batch 0, loss[loss=0.2171, simple_loss=0.2841, pruned_loss=0.07509, over 4892.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2841, pruned_loss=0.07509, over 4892.00 frames. ], batch size: 36, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:09:21,367 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 20:09:23,584 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8449, 1.0960, 1.9082, 1.8246, 1.6822, 1.6312, 1.7081, 1.8096], + device='cuda:1'), covar=tensor([0.4228, 0.4304, 0.3827, 0.3970, 0.5424, 0.3799, 0.5089, 0.3204], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0240, 0.0259, 0.0271, 0.0270, 0.0245, 0.0282, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:09:23,647 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8583, 3.4173, 3.5405, 3.7089, 3.6081, 3.3619, 3.9429, 1.3190], + device='cuda:1'), covar=tensor([0.0939, 0.0895, 0.0952, 0.1157, 0.1531, 0.1762, 0.0773, 0.5319], + device='cuda:1'), in_proj_covar=tensor([0.0345, 0.0241, 0.0272, 0.0290, 0.0330, 0.0278, 0.0296, 0.0291], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:09:24,169 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5117, 1.2787, 1.3447, 1.4745, 1.7781, 1.6154, 1.3645, 1.2155], + device='cuda:1'), covar=tensor([0.0401, 0.0366, 0.0624, 0.0327, 0.0223, 0.0480, 0.0385, 0.0491], + device='cuda:1'), in_proj_covar=tensor([0.0095, 0.0108, 0.0144, 0.0113, 0.0100, 0.0108, 0.0098, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.3808e-05, 8.3862e-05, 1.1375e-04, 8.6997e-05, 7.7713e-05, 7.9723e-05, + 7.3573e-05, 8.3274e-05], device='cuda:1') +2023-03-26 20:09:32,014 INFO [finetune.py:1010] (1/7) Epoch 17, validation: loss=0.1591, simple_loss=0.2283, pruned_loss=0.04492, over 2265189.00 frames. +2023-03-26 20:09:32,014 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 20:09:35,491 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.774e+01 1.479e+02 1.757e+02 2.057e+02 5.096e+02, threshold=3.514e+02, percent-clipped=1.0 +2023-03-26 20:10:00,082 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.64 vs. limit=2.0 +2023-03-26 20:10:07,336 INFO [finetune.py:976] (1/7) Epoch 17, batch 50, loss[loss=0.1357, simple_loss=0.211, pruned_loss=0.03023, over 4762.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2553, pruned_loss=0.05914, over 215313.74 frames. ], batch size: 27, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:10:16,233 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3794, 2.2718, 1.6966, 0.7828, 1.9284, 1.8599, 1.6071, 1.9800], + device='cuda:1'), covar=tensor([0.0774, 0.0696, 0.1420, 0.1834, 0.1285, 0.2095, 0.2320, 0.0893], + device='cuda:1'), in_proj_covar=tensor([0.0167, 0.0194, 0.0198, 0.0181, 0.0211, 0.0205, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:10:23,534 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-03-26 20:10:32,214 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6207, 1.2978, 1.9987, 3.3724, 2.2725, 2.4608, 1.0158, 2.8405], + device='cuda:1'), covar=tensor([0.2077, 0.2157, 0.1684, 0.1026, 0.0965, 0.1547, 0.2214, 0.0635], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0132, 0.0164, 0.0100, 0.0137, 0.0123, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:10:52,787 INFO [finetune.py:976] (1/7) Epoch 17, batch 100, loss[loss=0.1605, simple_loss=0.2306, pruned_loss=0.04521, over 4820.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2517, pruned_loss=0.05847, over 378556.60 frames. ], batch size: 40, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:11:01,250 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.123e+02 1.600e+02 1.810e+02 2.096e+02 3.529e+02, threshold=3.620e+02, percent-clipped=1.0 +2023-03-26 20:11:09,176 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-03-26 20:11:37,625 INFO [finetune.py:976] (1/7) Epoch 17, batch 150, loss[loss=0.1733, simple_loss=0.2575, pruned_loss=0.04452, over 4868.00 frames. ], tot_loss[loss=0.181, simple_loss=0.2477, pruned_loss=0.05713, over 508867.29 frames. ], batch size: 34, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:12:11,016 INFO [finetune.py:976] (1/7) Epoch 17, batch 200, loss[loss=0.157, simple_loss=0.2274, pruned_loss=0.04329, over 4871.00 frames. ], tot_loss[loss=0.1806, simple_loss=0.2469, pruned_loss=0.05712, over 609879.18 frames. ], batch size: 31, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:12:14,524 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.078e+02 1.595e+02 1.938e+02 2.273e+02 4.627e+02, threshold=3.876e+02, percent-clipped=4.0 +2023-03-26 20:12:43,715 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91891.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:12:44,795 INFO [finetune.py:976] (1/7) Epoch 17, batch 250, loss[loss=0.1955, simple_loss=0.2526, pruned_loss=0.06923, over 4773.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.252, pruned_loss=0.05909, over 685568.59 frames. ], batch size: 26, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:12:47,892 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91898.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 20:13:14,609 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6369, 2.6342, 2.5717, 1.9498, 2.8567, 2.9713, 2.9165, 2.0538], + device='cuda:1'), covar=tensor([0.0800, 0.0752, 0.0876, 0.1072, 0.0623, 0.0839, 0.0804, 0.1763], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0134, 0.0141, 0.0124, 0.0124, 0.0140, 0.0141, 0.0165], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:13:17,048 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91941.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:13:18,231 INFO [finetune.py:976] (1/7) Epoch 17, batch 300, loss[loss=0.2072, simple_loss=0.2793, pruned_loss=0.0676, over 4821.00 frames. ], tot_loss[loss=0.1869, simple_loss=0.2549, pruned_loss=0.05942, over 746381.44 frames. ], batch size: 33, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:13:21,759 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.873e+01 1.605e+02 2.003e+02 2.239e+02 3.510e+02, threshold=4.006e+02, percent-clipped=0.0 +2023-03-26 20:13:21,989 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-26 20:13:24,424 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91952.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:13:26,036 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6669, 3.7241, 3.5668, 1.9154, 3.8180, 2.9455, 1.1384, 2.6364], + device='cuda:1'), covar=tensor([0.2347, 0.2778, 0.1582, 0.3649, 0.1169, 0.1060, 0.4542, 0.1774], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0174, 0.0159, 0.0128, 0.0158, 0.0123, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 20:13:27,176 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91955.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:13:33,609 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5555, 1.5076, 1.4269, 1.4703, 0.9978, 3.2509, 1.3034, 1.7090], + device='cuda:1'), covar=tensor([0.3189, 0.2402, 0.2133, 0.2342, 0.1845, 0.0223, 0.2529, 0.1186], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0123, 0.0113, 0.0096, 0.0096, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 20:13:40,209 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5267, 1.4340, 1.2390, 1.5411, 1.8430, 1.6696, 1.4215, 1.3237], + device='cuda:1'), covar=tensor([0.0285, 0.0294, 0.0624, 0.0260, 0.0194, 0.0426, 0.0351, 0.0383], + device='cuda:1'), in_proj_covar=tensor([0.0095, 0.0108, 0.0144, 0.0113, 0.0100, 0.0108, 0.0098, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.4055e-05, 8.3752e-05, 1.1391e-04, 8.7247e-05, 7.7784e-05, 7.9726e-05, + 7.3843e-05, 8.2929e-05], device='cuda:1') +2023-03-26 20:13:42,102 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.83 vs. limit=5.0 +2023-03-26 20:13:44,461 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91981.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 20:13:49,290 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=91989.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:13:51,731 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-26 20:13:52,146 INFO [finetune.py:976] (1/7) Epoch 17, batch 350, loss[loss=0.2031, simple_loss=0.2766, pruned_loss=0.06487, over 4828.00 frames. ], tot_loss[loss=0.1868, simple_loss=0.2556, pruned_loss=0.05893, over 793025.58 frames. ], batch size: 30, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:14:09,830 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92016.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:14:19,916 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9663, 4.6748, 4.3576, 2.7289, 4.6805, 3.7092, 1.1946, 3.3908], + device='cuda:1'), covar=tensor([0.2222, 0.1869, 0.1314, 0.2616, 0.0787, 0.0755, 0.4301, 0.1237], + device='cuda:1'), in_proj_covar=tensor([0.0148, 0.0172, 0.0157, 0.0127, 0.0156, 0.0121, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 20:14:26,168 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92042.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 20:14:26,650 INFO [finetune.py:976] (1/7) Epoch 17, batch 400, loss[loss=0.1775, simple_loss=0.264, pruned_loss=0.04546, over 4824.00 frames. ], tot_loss[loss=0.187, simple_loss=0.2564, pruned_loss=0.0588, over 829104.38 frames. ], batch size: 39, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:14:30,187 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.886e+01 1.544e+02 1.847e+02 2.163e+02 3.487e+02, threshold=3.695e+02, percent-clipped=0.0 +2023-03-26 20:14:37,414 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 20:15:00,227 INFO [finetune.py:976] (1/7) Epoch 17, batch 450, loss[loss=0.1992, simple_loss=0.2603, pruned_loss=0.06912, over 4864.00 frames. ], tot_loss[loss=0.1842, simple_loss=0.2537, pruned_loss=0.05739, over 856920.13 frames. ], batch size: 31, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:15:33,735 INFO [finetune.py:976] (1/7) Epoch 17, batch 500, loss[loss=0.1578, simple_loss=0.2326, pruned_loss=0.04155, over 4874.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2514, pruned_loss=0.0569, over 880651.59 frames. ], batch size: 34, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:15:37,217 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.042e+02 1.578e+02 1.863e+02 2.269e+02 4.074e+02, threshold=3.727e+02, percent-clipped=2.0 +2023-03-26 20:16:01,200 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8448, 1.7904, 1.5721, 1.4260, 1.8755, 1.5892, 1.8591, 1.8036], + device='cuda:1'), covar=tensor([0.1365, 0.1969, 0.2944, 0.2234, 0.2453, 0.1647, 0.3060, 0.1721], + device='cuda:1'), in_proj_covar=tensor([0.0181, 0.0186, 0.0233, 0.0250, 0.0242, 0.0200, 0.0211, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:16:05,164 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5995, 1.4746, 2.0570, 3.2824, 2.1085, 2.3913, 1.1171, 2.6785], + device='cuda:1'), covar=tensor([0.1750, 0.1425, 0.1313, 0.0559, 0.0892, 0.1318, 0.1651, 0.0543], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0165, 0.0102, 0.0138, 0.0124, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:16:30,567 INFO [finetune.py:976] (1/7) Epoch 17, batch 550, loss[loss=0.1891, simple_loss=0.2613, pruned_loss=0.05847, over 4756.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2489, pruned_loss=0.05637, over 898256.99 frames. ], batch size: 54, lr: 3.41e-03, grad_scale: 32.0 +2023-03-26 20:16:33,716 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92198.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 20:17:13,333 INFO [finetune.py:976] (1/7) Epoch 17, batch 600, loss[loss=0.1502, simple_loss=0.2288, pruned_loss=0.03582, over 4817.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2488, pruned_loss=0.05641, over 909288.98 frames. ], batch size: 30, lr: 3.41e-03, grad_scale: 64.0 +2023-03-26 20:17:15,214 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=92246.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 20:17:15,824 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92247.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:17:16,357 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.142e+02 1.709e+02 1.980e+02 2.349e+02 5.069e+02, threshold=3.960e+02, percent-clipped=5.0 +2023-03-26 20:17:47,090 INFO [finetune.py:976] (1/7) Epoch 17, batch 650, loss[loss=0.1271, simple_loss=0.1951, pruned_loss=0.02949, over 4757.00 frames. ], tot_loss[loss=0.1837, simple_loss=0.252, pruned_loss=0.05769, over 920321.43 frames. ], batch size: 28, lr: 3.41e-03, grad_scale: 64.0 +2023-03-26 20:17:58,667 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92311.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:18:16,765 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92337.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 20:18:20,698 INFO [finetune.py:976] (1/7) Epoch 17, batch 700, loss[loss=0.1556, simple_loss=0.2233, pruned_loss=0.04396, over 4793.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.2531, pruned_loss=0.0576, over 927811.64 frames. ], batch size: 25, lr: 3.41e-03, grad_scale: 64.0 +2023-03-26 20:18:23,723 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.116e+02 1.553e+02 1.844e+02 2.135e+02 3.970e+02, threshold=3.688e+02, percent-clipped=1.0 +2023-03-26 20:18:54,353 INFO [finetune.py:976] (1/7) Epoch 17, batch 750, loss[loss=0.2041, simple_loss=0.2633, pruned_loss=0.07246, over 4740.00 frames. ], tot_loss[loss=0.1845, simple_loss=0.254, pruned_loss=0.05753, over 936542.14 frames. ], batch size: 54, lr: 3.41e-03, grad_scale: 64.0 +2023-03-26 20:19:04,326 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9071, 1.7702, 2.2566, 3.7961, 2.5825, 2.5801, 1.0621, 3.0371], + device='cuda:1'), covar=tensor([0.1720, 0.1452, 0.1331, 0.0462, 0.0721, 0.1266, 0.1877, 0.0517], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0164, 0.0101, 0.0136, 0.0124, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:19:24,926 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 20:19:28,149 INFO [finetune.py:976] (1/7) Epoch 17, batch 800, loss[loss=0.1719, simple_loss=0.2515, pruned_loss=0.04614, over 4769.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.253, pruned_loss=0.0568, over 940663.64 frames. ], batch size: 26, lr: 3.41e-03, grad_scale: 64.0 +2023-03-26 20:19:31,195 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.199e+02 1.753e+02 1.963e+02 2.342e+02 4.288e+02, threshold=3.926e+02, percent-clipped=2.0 +2023-03-26 20:19:51,697 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.57 vs. limit=5.0 +2023-03-26 20:20:01,478 INFO [finetune.py:976] (1/7) Epoch 17, batch 850, loss[loss=0.1168, simple_loss=0.1814, pruned_loss=0.02611, over 4229.00 frames. ], tot_loss[loss=0.1815, simple_loss=0.2508, pruned_loss=0.05609, over 945602.31 frames. ], batch size: 18, lr: 3.41e-03, grad_scale: 64.0 +2023-03-26 20:20:03,320 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6086, 1.4885, 1.4887, 1.4821, 0.9462, 2.9235, 1.0765, 1.3858], + device='cuda:1'), covar=tensor([0.3388, 0.2498, 0.2160, 0.2477, 0.1973, 0.0247, 0.2613, 0.1435], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0123, 0.0113, 0.0096, 0.0096, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 20:20:09,331 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6752, 3.7356, 3.5293, 1.7811, 3.8109, 2.9393, 0.7938, 2.7091], + device='cuda:1'), covar=tensor([0.2441, 0.1854, 0.1541, 0.3166, 0.0963, 0.1019, 0.4544, 0.1454], + device='cuda:1'), in_proj_covar=tensor([0.0149, 0.0173, 0.0158, 0.0128, 0.0156, 0.0122, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 20:20:18,441 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5107, 1.4772, 1.7856, 1.7435, 1.6234, 3.4012, 1.4370, 1.6324], + device='cuda:1'), covar=tensor([0.0884, 0.1684, 0.1020, 0.0925, 0.1477, 0.0223, 0.1357, 0.1601], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0078, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 20:20:35,310 INFO [finetune.py:976] (1/7) Epoch 17, batch 900, loss[loss=0.1522, simple_loss=0.2117, pruned_loss=0.04632, over 4940.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2472, pruned_loss=0.05476, over 947773.84 frames. ], batch size: 33, lr: 3.41e-03, grad_scale: 64.0 +2023-03-26 20:20:38,329 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92547.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:20:38,821 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.037e+02 1.480e+02 1.791e+02 2.296e+02 4.324e+02, threshold=3.582e+02, percent-clipped=2.0 +2023-03-26 20:20:46,212 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8013, 1.2182, 1.8056, 1.7918, 1.5913, 1.5848, 1.6609, 1.7057], + device='cuda:1'), covar=tensor([0.3879, 0.4101, 0.3447, 0.3786, 0.5071, 0.3707, 0.4553, 0.3218], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0239, 0.0257, 0.0271, 0.0269, 0.0244, 0.0281, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:21:06,941 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0060, 1.9710, 1.9848, 1.3954, 2.0990, 2.0794, 2.0873, 1.7086], + device='cuda:1'), covar=tensor([0.0578, 0.0619, 0.0688, 0.0897, 0.0578, 0.0632, 0.0598, 0.1074], + device='cuda:1'), in_proj_covar=tensor([0.0135, 0.0136, 0.0143, 0.0125, 0.0125, 0.0142, 0.0143, 0.0167], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:21:15,094 INFO [finetune.py:976] (1/7) Epoch 17, batch 950, loss[loss=0.1814, simple_loss=0.252, pruned_loss=0.05541, over 4755.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.2465, pruned_loss=0.05468, over 948019.48 frames. ], batch size: 27, lr: 3.40e-03, grad_scale: 64.0 +2023-03-26 20:21:16,913 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=92595.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:21:37,230 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92611.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:21:54,785 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92625.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:22:05,399 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92637.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 20:22:13,800 INFO [finetune.py:976] (1/7) Epoch 17, batch 1000, loss[loss=0.2011, simple_loss=0.2723, pruned_loss=0.06492, over 4743.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2503, pruned_loss=0.05663, over 948885.28 frames. ], batch size: 54, lr: 3.40e-03, grad_scale: 64.0 +2023-03-26 20:22:20,426 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.492e+01 1.711e+02 2.074e+02 2.603e+02 6.251e+02, threshold=4.148e+02, percent-clipped=4.0 +2023-03-26 20:22:27,818 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=92659.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:22:27,830 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8316, 4.3085, 4.1096, 2.2102, 4.4297, 3.5722, 0.6180, 2.9339], + device='cuda:1'), covar=tensor([0.2579, 0.1565, 0.1261, 0.3040, 0.0732, 0.0698, 0.4539, 0.1400], + device='cuda:1'), in_proj_covar=tensor([0.0149, 0.0174, 0.0158, 0.0128, 0.0156, 0.0122, 0.0145, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 20:22:45,203 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=92685.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 20:22:45,246 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3827, 2.5997, 2.3062, 1.9444, 2.4937, 2.6593, 2.5637, 2.2655], + device='cuda:1'), covar=tensor([0.0651, 0.0587, 0.0779, 0.0876, 0.0719, 0.0779, 0.0678, 0.0980], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0135, 0.0142, 0.0124, 0.0124, 0.0141, 0.0143, 0.0166], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:22:45,858 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92686.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:22:50,928 INFO [finetune.py:976] (1/7) Epoch 17, batch 1050, loss[loss=0.1658, simple_loss=0.2267, pruned_loss=0.05248, over 4931.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2526, pruned_loss=0.05676, over 949889.30 frames. ], batch size: 33, lr: 3.40e-03, grad_scale: 64.0 +2023-03-26 20:23:08,354 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92720.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:23:09,630 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.33 vs. limit=5.0 +2023-03-26 20:23:23,722 INFO [finetune.py:976] (1/7) Epoch 17, batch 1100, loss[loss=0.2081, simple_loss=0.2746, pruned_loss=0.07083, over 4819.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2548, pruned_loss=0.05836, over 950018.66 frames. ], batch size: 33, lr: 3.40e-03, grad_scale: 64.0 +2023-03-26 20:23:27,193 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.157e+02 1.694e+02 2.013e+02 2.338e+02 4.806e+02, threshold=4.026e+02, percent-clipped=2.0 +2023-03-26 20:23:37,599 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-26 20:23:48,913 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92781.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:23:57,181 INFO [finetune.py:976] (1/7) Epoch 17, batch 1150, loss[loss=0.2034, simple_loss=0.2678, pruned_loss=0.06948, over 4903.00 frames. ], tot_loss[loss=0.1865, simple_loss=0.2558, pruned_loss=0.05862, over 951870.37 frames. ], batch size: 36, lr: 3.40e-03, grad_scale: 64.0 +2023-03-26 20:24:14,642 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 20:24:22,927 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92831.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:24:31,079 INFO [finetune.py:976] (1/7) Epoch 17, batch 1200, loss[loss=0.1797, simple_loss=0.2392, pruned_loss=0.06007, over 4929.00 frames. ], tot_loss[loss=0.1854, simple_loss=0.2545, pruned_loss=0.05819, over 949995.46 frames. ], batch size: 33, lr: 3.40e-03, grad_scale: 64.0 +2023-03-26 20:24:34,573 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.909e+01 1.547e+02 1.742e+02 2.125e+02 5.044e+02, threshold=3.483e+02, percent-clipped=2.0 +2023-03-26 20:24:56,040 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.6841, 4.0275, 4.2199, 4.4829, 4.4506, 4.1970, 4.7341, 1.6461], + device='cuda:1'), covar=tensor([0.0642, 0.0904, 0.0887, 0.0946, 0.1087, 0.1543, 0.0620, 0.5314], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0242, 0.0273, 0.0290, 0.0329, 0.0278, 0.0298, 0.0293], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:24:57,195 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0013, 1.9310, 1.7782, 2.1852, 2.6705, 2.1960, 1.8342, 1.6649], + device='cuda:1'), covar=tensor([0.2162, 0.1882, 0.1807, 0.1690, 0.1605, 0.1076, 0.2273, 0.2045], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0208, 0.0211, 0.0190, 0.0241, 0.0185, 0.0215, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:25:03,716 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92892.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:25:04,693 INFO [finetune.py:976] (1/7) Epoch 17, batch 1250, loss[loss=0.2026, simple_loss=0.2511, pruned_loss=0.07706, over 4728.00 frames. ], tot_loss[loss=0.183, simple_loss=0.2511, pruned_loss=0.05746, over 950654.33 frames. ], batch size: 23, lr: 3.40e-03, grad_scale: 64.0 +2023-03-26 20:25:08,426 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92899.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:25:29,746 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92931.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:25:37,457 INFO [finetune.py:976] (1/7) Epoch 17, batch 1300, loss[loss=0.1565, simple_loss=0.2264, pruned_loss=0.04335, over 4823.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2484, pruned_loss=0.05629, over 951792.34 frames. ], batch size: 25, lr: 3.40e-03, grad_scale: 64.0 +2023-03-26 20:25:41,344 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.030e+02 1.503e+02 1.790e+02 2.154e+02 4.064e+02, threshold=3.581e+02, percent-clipped=1.0 +2023-03-26 20:25:46,123 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3725, 1.3614, 1.7038, 2.4599, 1.6979, 2.1601, 0.9530, 2.1288], + device='cuda:1'), covar=tensor([0.1773, 0.1426, 0.1095, 0.0789, 0.0920, 0.1292, 0.1537, 0.0611], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0164, 0.0101, 0.0137, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:25:49,677 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5326, 1.4433, 1.4245, 1.4133, 0.8502, 2.3036, 0.7462, 1.2119], + device='cuda:1'), covar=tensor([0.3274, 0.2468, 0.2237, 0.2501, 0.1890, 0.0361, 0.2634, 0.1360], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0123, 0.0113, 0.0096, 0.0096, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 20:25:49,703 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92960.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:25:59,955 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92975.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:26:00,566 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7090, 1.5627, 2.0129, 1.9139, 1.7961, 3.5919, 1.4918, 1.7093], + device='cuda:1'), covar=tensor([0.0894, 0.1857, 0.1048, 0.0970, 0.1524, 0.0206, 0.1534, 0.1738], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0078, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 20:26:03,544 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92981.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:26:10,744 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92992.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:26:11,263 INFO [finetune.py:976] (1/7) Epoch 17, batch 1350, loss[loss=0.1536, simple_loss=0.2313, pruned_loss=0.03798, over 4909.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.2484, pruned_loss=0.05598, over 953238.38 frames. ], batch size: 36, lr: 3.40e-03, grad_scale: 64.0 +2023-03-26 20:26:23,862 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93010.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:26:51,524 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93036.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:26:51,538 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93036.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:26:52,262 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.20 vs. limit=5.0 +2023-03-26 20:26:59,858 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4867, 2.3155, 2.0281, 2.2463, 2.1939, 2.2346, 2.1992, 2.9076], + device='cuda:1'), covar=tensor([0.3816, 0.4701, 0.3294, 0.4159, 0.4180, 0.2506, 0.4264, 0.1683], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0225, 0.0276, 0.0249, 0.0217, 0.0249, 0.0229], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:27:00,929 INFO [finetune.py:976] (1/7) Epoch 17, batch 1400, loss[loss=0.1646, simple_loss=0.2426, pruned_loss=0.04324, over 4912.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2518, pruned_loss=0.05669, over 953411.70 frames. ], batch size: 37, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:27:08,967 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.549e+02 1.883e+02 2.310e+02 4.523e+02, threshold=3.767e+02, percent-clipped=3.0 +2023-03-26 20:27:34,715 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93071.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:27:39,783 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93076.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:27:50,103 INFO [finetune.py:976] (1/7) Epoch 17, batch 1450, loss[loss=0.1527, simple_loss=0.2178, pruned_loss=0.04383, over 4735.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2522, pruned_loss=0.05631, over 953489.75 frames. ], batch size: 23, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:27:53,126 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93097.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 20:28:05,999 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93115.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:28:17,833 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1343, 1.3009, 1.3863, 0.7419, 1.2796, 1.5767, 1.5998, 1.2974], + device='cuda:1'), covar=tensor([0.0875, 0.0587, 0.0532, 0.0500, 0.0539, 0.0541, 0.0347, 0.0625], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0151, 0.0123, 0.0127, 0.0130, 0.0128, 0.0142, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.1545e-05, 1.0941e-04, 8.8441e-05, 9.0325e-05, 9.1730e-05, 9.2104e-05, + 1.0263e-04, 1.0613e-04], device='cuda:1') +2023-03-26 20:28:23,787 INFO [finetune.py:976] (1/7) Epoch 17, batch 1500, loss[loss=0.1727, simple_loss=0.251, pruned_loss=0.04723, over 4749.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2544, pruned_loss=0.05715, over 953658.28 frames. ], batch size: 27, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:28:27,863 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.132e+02 1.651e+02 1.993e+02 2.270e+02 5.642e+02, threshold=3.987e+02, percent-clipped=1.0 +2023-03-26 20:28:40,283 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3998, 2.2328, 1.7479, 2.1407, 2.1315, 1.9002, 2.4778, 2.2868], + device='cuda:1'), covar=tensor([0.1137, 0.2128, 0.2774, 0.2763, 0.2426, 0.1565, 0.3931, 0.1527], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0189, 0.0234, 0.0253, 0.0245, 0.0202, 0.0213, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:28:47,139 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93176.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:28:53,710 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93187.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:28:57,278 INFO [finetune.py:976] (1/7) Epoch 17, batch 1550, loss[loss=0.1964, simple_loss=0.2696, pruned_loss=0.0616, over 4792.00 frames. ], tot_loss[loss=0.185, simple_loss=0.2551, pruned_loss=0.05751, over 954178.48 frames. ], batch size: 45, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:29:13,179 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0567, 1.9307, 1.9939, 1.4130, 1.9990, 2.1033, 2.1227, 1.6833], + device='cuda:1'), covar=tensor([0.0506, 0.0581, 0.0646, 0.0837, 0.0677, 0.0593, 0.0487, 0.1016], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0135, 0.0143, 0.0124, 0.0125, 0.0142, 0.0143, 0.0166], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:29:30,932 INFO [finetune.py:976] (1/7) Epoch 17, batch 1600, loss[loss=0.2411, simple_loss=0.2971, pruned_loss=0.0926, over 4297.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.253, pruned_loss=0.05705, over 953430.91 frames. ], batch size: 65, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:29:34,594 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.059e+02 1.546e+02 1.807e+02 2.216e+02 3.989e+02, threshold=3.613e+02, percent-clipped=1.0 +2023-03-26 20:29:38,722 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93255.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:29:47,628 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 20:29:57,408 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93281.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:30:01,029 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93287.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:30:04,615 INFO [finetune.py:976] (1/7) Epoch 17, batch 1650, loss[loss=0.1864, simple_loss=0.2483, pruned_loss=0.06225, over 4818.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2509, pruned_loss=0.0565, over 955064.48 frames. ], batch size: 45, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:30:10,844 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0561, 0.9300, 0.9557, 0.4497, 0.8636, 1.1615, 1.1638, 0.9964], + device='cuda:1'), covar=tensor([0.1064, 0.0709, 0.0611, 0.0565, 0.0656, 0.0687, 0.0490, 0.0762], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0151, 0.0124, 0.0127, 0.0131, 0.0128, 0.0143, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.2006e-05, 1.0978e-04, 8.8585e-05, 9.0733e-05, 9.2222e-05, 9.2385e-05, + 1.0295e-04, 1.0677e-04], device='cuda:1') +2023-03-26 20:30:13,743 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9527, 1.2675, 2.0241, 1.9182, 1.7734, 1.7071, 1.8065, 1.8636], + device='cuda:1'), covar=tensor([0.3821, 0.4380, 0.3294, 0.3610, 0.4485, 0.3458, 0.4518, 0.3082], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0239, 0.0257, 0.0272, 0.0270, 0.0245, 0.0282, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:30:28,975 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=93329.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:30:30,200 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93331.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:30:38,308 INFO [finetune.py:976] (1/7) Epoch 17, batch 1700, loss[loss=0.1746, simple_loss=0.2498, pruned_loss=0.04964, over 4886.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2491, pruned_loss=0.05596, over 955247.75 frames. ], batch size: 32, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:30:41,941 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.003e+02 1.487e+02 1.694e+02 2.142e+02 3.933e+02, threshold=3.388e+02, percent-clipped=2.0 +2023-03-26 20:30:43,408 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.07 vs. limit=5.0 +2023-03-26 20:30:53,342 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93366.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:31:00,349 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93376.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:31:11,619 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93392.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 20:31:12,182 INFO [finetune.py:976] (1/7) Epoch 17, batch 1750, loss[loss=0.1972, simple_loss=0.2704, pruned_loss=0.06203, over 4889.00 frames. ], tot_loss[loss=0.1821, simple_loss=0.251, pruned_loss=0.0566, over 953760.45 frames. ], batch size: 32, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:31:13,564 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9086, 1.6418, 2.1583, 1.5245, 1.9526, 2.0771, 1.5608, 2.2915], + device='cuda:1'), covar=tensor([0.1255, 0.2055, 0.1274, 0.1725, 0.0905, 0.1564, 0.2768, 0.0743], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0203, 0.0190, 0.0189, 0.0176, 0.0213, 0.0217, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:31:33,283 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=93424.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:31:48,780 INFO [finetune.py:976] (1/7) Epoch 17, batch 1800, loss[loss=0.2042, simple_loss=0.2651, pruned_loss=0.0716, over 4901.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2526, pruned_loss=0.05698, over 951968.39 frames. ], batch size: 35, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:31:56,895 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.045e+02 1.552e+02 1.846e+02 2.179e+02 3.576e+02, threshold=3.692e+02, percent-clipped=3.0 +2023-03-26 20:32:20,903 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93471.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:32:40,551 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93487.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:32:49,593 INFO [finetune.py:976] (1/7) Epoch 17, batch 1850, loss[loss=0.2274, simple_loss=0.2963, pruned_loss=0.07932, over 4864.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.254, pruned_loss=0.05743, over 953297.36 frames. ], batch size: 44, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:33:02,763 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8173, 1.2387, 0.8679, 1.5952, 2.1189, 1.5159, 1.4379, 1.6088], + device='cuda:1'), covar=tensor([0.1313, 0.2040, 0.1914, 0.1143, 0.1859, 0.1969, 0.1389, 0.1812], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0092, 0.0118, 0.0094, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:33:20,424 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=93535.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:33:26,141 INFO [finetune.py:976] (1/7) Epoch 17, batch 1900, loss[loss=0.1725, simple_loss=0.255, pruned_loss=0.045, over 4829.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2558, pruned_loss=0.05792, over 954602.99 frames. ], batch size: 47, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:33:30,350 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.241e+02 1.618e+02 1.925e+02 2.327e+02 3.543e+02, threshold=3.851e+02, percent-clipped=0.0 +2023-03-26 20:33:34,113 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93555.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:33:55,469 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93587.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:33:59,455 INFO [finetune.py:976] (1/7) Epoch 17, batch 1950, loss[loss=0.1731, simple_loss=0.2462, pruned_loss=0.04998, over 4828.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2538, pruned_loss=0.05645, over 956394.49 frames. ], batch size: 33, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:34:06,614 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=93603.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:34:24,529 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93631.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:34:27,389 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=93635.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:34:32,610 INFO [finetune.py:976] (1/7) Epoch 17, batch 2000, loss[loss=0.176, simple_loss=0.2482, pruned_loss=0.05194, over 4901.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2519, pruned_loss=0.0564, over 955290.65 frames. ], batch size: 35, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:34:36,884 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 20:34:37,203 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.489e+01 1.528e+02 1.753e+02 2.103e+02 5.258e+02, threshold=3.506e+02, percent-clipped=1.0 +2023-03-26 20:34:48,138 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93666.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:34:56,475 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=93679.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:35:04,204 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.72 vs. limit=2.0 +2023-03-26 20:35:05,823 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93692.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:35:06,327 INFO [finetune.py:976] (1/7) Epoch 17, batch 2050, loss[loss=0.1573, simple_loss=0.2276, pruned_loss=0.04354, over 4913.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2475, pruned_loss=0.05466, over 956172.93 frames. ], batch size: 37, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:35:07,614 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4652, 1.3341, 1.8942, 2.8310, 1.9958, 2.2856, 0.9303, 2.3599], + device='cuda:1'), covar=tensor([0.2033, 0.1957, 0.1490, 0.0931, 0.0970, 0.1380, 0.2046, 0.0791], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0165, 0.0101, 0.0136, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:35:20,566 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=93714.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:35:37,806 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=93740.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:35:39,567 INFO [finetune.py:976] (1/7) Epoch 17, batch 2100, loss[loss=0.1921, simple_loss=0.2585, pruned_loss=0.06285, over 4911.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2462, pruned_loss=0.05404, over 957792.72 frames. ], batch size: 37, lr: 3.40e-03, grad_scale: 32.0 +2023-03-26 20:35:43,619 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.402e+01 1.568e+02 1.860e+02 2.232e+02 5.340e+02, threshold=3.720e+02, percent-clipped=4.0 +2023-03-26 20:35:57,994 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93770.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:35:58,584 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93771.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:36:13,275 INFO [finetune.py:976] (1/7) Epoch 17, batch 2150, loss[loss=0.1974, simple_loss=0.2466, pruned_loss=0.07412, over 3992.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2507, pruned_loss=0.05621, over 954996.23 frames. ], batch size: 17, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:36:31,169 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=93819.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:36:38,614 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93831.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:36:47,327 INFO [finetune.py:976] (1/7) Epoch 17, batch 2200, loss[loss=0.2309, simple_loss=0.2901, pruned_loss=0.08581, over 4895.00 frames. ], tot_loss[loss=0.1838, simple_loss=0.2528, pruned_loss=0.05738, over 953160.07 frames. ], batch size: 35, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:36:51,479 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.198e+02 1.569e+02 1.869e+02 2.308e+02 4.137e+02, threshold=3.738e+02, percent-clipped=1.0 +2023-03-26 20:36:51,687 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 20:37:36,156 INFO [finetune.py:976] (1/7) Epoch 17, batch 2250, loss[loss=0.1701, simple_loss=0.2458, pruned_loss=0.04717, over 4863.00 frames. ], tot_loss[loss=0.185, simple_loss=0.2546, pruned_loss=0.05774, over 952540.09 frames. ], batch size: 31, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:38:30,053 INFO [finetune.py:976] (1/7) Epoch 17, batch 2300, loss[loss=0.1399, simple_loss=0.2125, pruned_loss=0.0337, over 4771.00 frames. ], tot_loss[loss=0.1837, simple_loss=0.2535, pruned_loss=0.05694, over 953090.82 frames. ], batch size: 25, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:38:34,185 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.189e+02 1.601e+02 1.890e+02 2.328e+02 3.292e+02, threshold=3.781e+02, percent-clipped=0.0 +2023-03-26 20:38:52,002 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 20:38:53,833 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-03-26 20:38:56,095 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93981.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:39:02,936 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-03-26 20:39:03,825 INFO [finetune.py:976] (1/7) Epoch 17, batch 2350, loss[loss=0.1435, simple_loss=0.2215, pruned_loss=0.03274, over 4888.00 frames. ], tot_loss[loss=0.1806, simple_loss=0.2507, pruned_loss=0.05524, over 953987.19 frames. ], batch size: 35, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:39:22,361 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-26 20:39:37,884 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=94042.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:39:38,383 INFO [finetune.py:976] (1/7) Epoch 17, batch 2400, loss[loss=0.1915, simple_loss=0.2546, pruned_loss=0.06414, over 4856.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2488, pruned_loss=0.05498, over 954563.01 frames. ], batch size: 49, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:39:42,505 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.024e+02 1.475e+02 1.781e+02 2.110e+02 4.538e+02, threshold=3.563e+02, percent-clipped=2.0 +2023-03-26 20:40:11,636 INFO [finetune.py:976] (1/7) Epoch 17, batch 2450, loss[loss=0.1953, simple_loss=0.2561, pruned_loss=0.06731, over 4764.00 frames. ], tot_loss[loss=0.1787, simple_loss=0.2473, pruned_loss=0.0551, over 951414.03 frames. ], batch size: 59, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:40:18,300 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4623, 1.3760, 1.9495, 3.1465, 2.1088, 2.3146, 0.9890, 2.6653], + device='cuda:1'), covar=tensor([0.1801, 0.1509, 0.1303, 0.0609, 0.0845, 0.1334, 0.1807, 0.0530], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0165, 0.0101, 0.0136, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:40:34,650 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94126.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:40:42,586 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 20:40:45,406 INFO [finetune.py:976] (1/7) Epoch 17, batch 2500, loss[loss=0.2079, simple_loss=0.2726, pruned_loss=0.07159, over 4810.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2487, pruned_loss=0.05533, over 950252.20 frames. ], batch size: 38, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:40:49,550 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.530e+01 1.683e+02 1.909e+02 2.220e+02 4.342e+02, threshold=3.819e+02, percent-clipped=2.0 +2023-03-26 20:40:54,057 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.83 vs. limit=5.0 +2023-03-26 20:41:18,602 INFO [finetune.py:976] (1/7) Epoch 17, batch 2550, loss[loss=0.2286, simple_loss=0.2789, pruned_loss=0.08911, over 4830.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2512, pruned_loss=0.05618, over 950163.40 frames. ], batch size: 30, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:41:28,912 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4782, 1.1506, 0.7414, 1.3396, 1.8707, 0.7223, 1.2575, 1.3748], + device='cuda:1'), covar=tensor([0.1455, 0.2050, 0.1745, 0.1169, 0.1975, 0.2032, 0.1414, 0.1865], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0119, 0.0095, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:41:38,788 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=94223.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 20:41:51,285 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2016, 2.1254, 1.7216, 1.9628, 2.1400, 1.8209, 2.3488, 2.1698], + device='cuda:1'), covar=tensor([0.1261, 0.1923, 0.2880, 0.2553, 0.2431, 0.1605, 0.2996, 0.1679], + device='cuda:1'), in_proj_covar=tensor([0.0182, 0.0187, 0.0232, 0.0252, 0.0243, 0.0201, 0.0212, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:41:52,380 INFO [finetune.py:976] (1/7) Epoch 17, batch 2600, loss[loss=0.1563, simple_loss=0.226, pruned_loss=0.04327, over 4894.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2521, pruned_loss=0.05643, over 949993.98 frames. ], batch size: 35, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:41:56,016 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.038e+02 1.647e+02 1.955e+02 2.265e+02 3.573e+02, threshold=3.911e+02, percent-clipped=0.0 +2023-03-26 20:41:56,111 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6702, 1.2293, 0.8748, 1.5317, 2.2045, 1.3728, 1.3224, 1.5706], + device='cuda:1'), covar=tensor([0.1536, 0.2207, 0.2010, 0.1360, 0.1731, 0.1970, 0.1578, 0.1954], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0119, 0.0095, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:42:01,547 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-03-26 20:42:13,960 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5742, 2.3349, 3.0195, 1.9475, 2.7110, 2.9179, 2.1845, 2.9754], + device='cuda:1'), covar=tensor([0.1336, 0.1878, 0.1349, 0.2052, 0.0930, 0.1433, 0.2667, 0.0877], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0203, 0.0189, 0.0189, 0.0176, 0.0212, 0.0216, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:42:19,682 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3597, 1.4539, 1.4648, 0.9192, 1.4498, 1.7024, 1.7055, 1.3480], + device='cuda:1'), covar=tensor([0.1035, 0.0640, 0.0543, 0.0539, 0.0455, 0.0610, 0.0312, 0.0728], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0153, 0.0125, 0.0129, 0.0132, 0.0130, 0.0145, 0.0150], + device='cuda:1'), out_proj_covar=tensor([9.3128e-05, 1.1091e-04, 8.9296e-05, 9.1824e-05, 9.3226e-05, 9.3535e-05, + 1.0426e-04, 1.0797e-04], device='cuda:1') +2023-03-26 20:42:19,685 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=94284.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 20:42:24,401 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 20:42:25,432 INFO [finetune.py:976] (1/7) Epoch 17, batch 2650, loss[loss=0.1929, simple_loss=0.2612, pruned_loss=0.06235, over 4902.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2541, pruned_loss=0.05727, over 949718.22 frames. ], batch size: 32, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:43:12,615 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94337.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:43:20,898 INFO [finetune.py:976] (1/7) Epoch 17, batch 2700, loss[loss=0.1825, simple_loss=0.246, pruned_loss=0.05945, over 4685.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2522, pruned_loss=0.0564, over 950144.76 frames. ], batch size: 23, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:43:24,087 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=94348.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:43:28,204 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.025e+02 1.514e+02 1.766e+02 2.145e+02 4.618e+02, threshold=3.532e+02, percent-clipped=2.0 +2023-03-26 20:43:29,551 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5207, 1.3307, 1.5112, 1.0063, 1.5595, 1.5217, 1.4843, 1.1801], + device='cuda:1'), covar=tensor([0.0748, 0.1016, 0.0744, 0.1032, 0.1009, 0.0801, 0.0761, 0.1924], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0136, 0.0142, 0.0124, 0.0125, 0.0141, 0.0142, 0.0165], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:44:10,178 INFO [finetune.py:976] (1/7) Epoch 17, batch 2750, loss[loss=0.1782, simple_loss=0.2466, pruned_loss=0.05492, over 4926.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2503, pruned_loss=0.05573, over 951701.08 frames. ], batch size: 33, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:44:11,593 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-26 20:44:15,311 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2559, 2.1769, 2.1360, 2.4538, 2.8622, 2.4560, 2.3182, 2.0345], + device='cuda:1'), covar=tensor([0.2140, 0.1976, 0.1800, 0.1523, 0.1436, 0.1112, 0.2025, 0.1932], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0207, 0.0211, 0.0191, 0.0240, 0.0184, 0.0214, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:44:20,188 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=94409.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 20:44:31,938 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94426.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:44:43,081 INFO [finetune.py:976] (1/7) Epoch 17, batch 2800, loss[loss=0.207, simple_loss=0.2608, pruned_loss=0.07662, over 4828.00 frames. ], tot_loss[loss=0.1781, simple_loss=0.2469, pruned_loss=0.05468, over 952900.42 frames. ], batch size: 33, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:44:47,182 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.577e+01 1.587e+02 1.887e+02 2.313e+02 4.372e+02, threshold=3.775e+02, percent-clipped=5.0 +2023-03-26 20:44:55,813 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.7002, 1.8041, 1.7901, 1.3106, 1.8759, 2.1494, 2.0002, 1.5950], + device='cuda:1'), covar=tensor([0.0978, 0.0588, 0.0568, 0.0526, 0.0476, 0.0509, 0.0370, 0.0735], + device='cuda:1'), in_proj_covar=tensor([0.0127, 0.0154, 0.0125, 0.0129, 0.0133, 0.0131, 0.0145, 0.0150], + device='cuda:1'), out_proj_covar=tensor([9.3347e-05, 1.1134e-04, 8.9849e-05, 9.2042e-05, 9.3824e-05, 9.4270e-05, + 1.0477e-04, 1.0837e-04], device='cuda:1') +2023-03-26 20:45:02,861 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=94474.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:45:09,939 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0655, 2.1818, 1.8149, 1.9240, 2.6166, 2.7483, 2.0671, 2.1191], + device='cuda:1'), covar=tensor([0.0310, 0.0355, 0.0537, 0.0374, 0.0256, 0.0379, 0.0484, 0.0383], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0106, 0.0141, 0.0111, 0.0098, 0.0107, 0.0097, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.2620e-05, 8.2064e-05, 1.1144e-04, 8.5567e-05, 7.6462e-05, 7.8809e-05, + 7.2894e-05, 8.2584e-05], device='cuda:1') +2023-03-26 20:45:16,199 INFO [finetune.py:976] (1/7) Epoch 17, batch 2850, loss[loss=0.1977, simple_loss=0.2799, pruned_loss=0.05776, over 4849.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2469, pruned_loss=0.05501, over 954677.65 frames. ], batch size: 47, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:45:49,613 INFO [finetune.py:976] (1/7) Epoch 17, batch 2900, loss[loss=0.1866, simple_loss=0.2751, pruned_loss=0.04902, over 4792.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2503, pruned_loss=0.056, over 954705.88 frames. ], batch size: 51, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:45:53,201 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.550e+02 1.801e+02 2.117e+02 3.911e+02, threshold=3.601e+02, percent-clipped=1.0 +2023-03-26 20:46:12,362 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94579.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 20:46:22,372 INFO [finetune.py:976] (1/7) Epoch 17, batch 2950, loss[loss=0.2071, simple_loss=0.2801, pruned_loss=0.06706, over 4815.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2529, pruned_loss=0.05663, over 954738.66 frames. ], batch size: 38, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:46:36,954 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-03-26 20:46:46,973 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5685, 1.5054, 1.4599, 1.5282, 1.0943, 3.2497, 1.4263, 1.7331], + device='cuda:1'), covar=tensor([0.3196, 0.2425, 0.2070, 0.2289, 0.1808, 0.0241, 0.2646, 0.1294], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0114, 0.0120, 0.0122, 0.0113, 0.0096, 0.0096, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 20:46:48,231 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0504, 2.0318, 2.0154, 1.5186, 2.0059, 2.0936, 2.1023, 1.7216], + device='cuda:1'), covar=tensor([0.0607, 0.0581, 0.0674, 0.0885, 0.0678, 0.0706, 0.0593, 0.1015], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0134, 0.0140, 0.0123, 0.0124, 0.0140, 0.0141, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:46:52,125 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94637.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:46:56,158 INFO [finetune.py:976] (1/7) Epoch 17, batch 3000, loss[loss=0.1591, simple_loss=0.228, pruned_loss=0.04508, over 4831.00 frames. ], tot_loss[loss=0.1854, simple_loss=0.2554, pruned_loss=0.05767, over 956415.66 frames. ], batch size: 30, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:46:56,158 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 20:47:06,773 INFO [finetune.py:1010] (1/7) Epoch 17, validation: loss=0.1562, simple_loss=0.2257, pruned_loss=0.04335, over 2265189.00 frames. +2023-03-26 20:47:06,773 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 20:47:09,446 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-26 20:47:10,420 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.006e+02 1.605e+02 1.916e+02 2.337e+02 3.800e+02, threshold=3.832e+02, percent-clipped=2.0 +2023-03-26 20:47:33,733 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=94685.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:47:34,413 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6223, 1.5048, 1.3752, 1.7071, 1.9400, 1.6575, 1.2629, 1.3989], + device='cuda:1'), covar=tensor([0.2255, 0.2074, 0.1963, 0.1607, 0.1598, 0.1288, 0.2512, 0.1901], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0207, 0.0211, 0.0190, 0.0239, 0.0184, 0.0214, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:47:36,719 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-03-26 20:47:38,495 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9785, 1.9946, 1.9791, 1.4550, 1.9957, 2.0315, 2.0490, 1.6649], + device='cuda:1'), covar=tensor([0.0589, 0.0602, 0.0699, 0.0870, 0.0710, 0.0780, 0.0637, 0.1147], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0134, 0.0140, 0.0123, 0.0123, 0.0140, 0.0141, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:47:38,997 INFO [finetune.py:976] (1/7) Epoch 17, batch 3050, loss[loss=0.1718, simple_loss=0.2562, pruned_loss=0.04367, over 4811.00 frames. ], tot_loss[loss=0.1848, simple_loss=0.2553, pruned_loss=0.05719, over 956974.56 frames. ], batch size: 40, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:47:46,847 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94704.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 20:48:19,141 INFO [finetune.py:976] (1/7) Epoch 17, batch 3100, loss[loss=0.1508, simple_loss=0.2123, pruned_loss=0.04467, over 4915.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2521, pruned_loss=0.05595, over 956795.60 frames. ], batch size: 46, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:48:27,685 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.157e+02 1.533e+02 1.793e+02 2.269e+02 8.706e+02, threshold=3.585e+02, percent-clipped=3.0 +2023-03-26 20:48:38,014 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0311, 0.9696, 0.9408, 1.1772, 1.2363, 1.1746, 1.0183, 0.9361], + device='cuda:1'), covar=tensor([0.0344, 0.0307, 0.0640, 0.0287, 0.0249, 0.0398, 0.0315, 0.0414], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0106, 0.0141, 0.0111, 0.0098, 0.0107, 0.0097, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.2801e-05, 8.1927e-05, 1.1155e-04, 8.5702e-05, 7.6516e-05, 7.8845e-05, + 7.2744e-05, 8.2848e-05], device='cuda:1') +2023-03-26 20:48:38,691 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0 +2023-03-26 20:49:07,467 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4701, 1.4736, 2.1008, 3.1534, 2.1758, 2.4106, 1.2569, 2.5531], + device='cuda:1'), covar=tensor([0.1833, 0.1589, 0.1286, 0.0615, 0.0789, 0.1207, 0.1644, 0.0567], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0115, 0.0133, 0.0164, 0.0100, 0.0135, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:49:17,518 INFO [finetune.py:976] (1/7) Epoch 17, batch 3150, loss[loss=0.1833, simple_loss=0.2492, pruned_loss=0.05869, over 4799.00 frames. ], tot_loss[loss=0.18, simple_loss=0.2493, pruned_loss=0.05534, over 954839.91 frames. ], batch size: 29, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:49:51,401 INFO [finetune.py:976] (1/7) Epoch 17, batch 3200, loss[loss=0.1413, simple_loss=0.2136, pruned_loss=0.03456, over 4937.00 frames. ], tot_loss[loss=0.1773, simple_loss=0.2461, pruned_loss=0.05428, over 954796.22 frames. ], batch size: 38, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:49:55,534 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.020e+02 1.463e+02 1.766e+02 2.027e+02 4.168e+02, threshold=3.532e+02, percent-clipped=1.0 +2023-03-26 20:50:16,321 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94879.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 20:50:25,247 INFO [finetune.py:976] (1/7) Epoch 17, batch 3250, loss[loss=0.1925, simple_loss=0.2694, pruned_loss=0.05781, over 4913.00 frames. ], tot_loss[loss=0.1793, simple_loss=0.2483, pruned_loss=0.05516, over 956776.09 frames. ], batch size: 43, lr: 3.39e-03, grad_scale: 32.0 +2023-03-26 20:50:48,425 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=94927.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 20:50:58,674 INFO [finetune.py:976] (1/7) Epoch 17, batch 3300, loss[loss=0.2019, simple_loss=0.2715, pruned_loss=0.0661, over 4811.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.2497, pruned_loss=0.05536, over 954859.66 frames. ], batch size: 45, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 20:51:02,380 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.215e+02 1.785e+02 2.188e+02 2.532e+02 5.228e+02, threshold=4.375e+02, percent-clipped=4.0 +2023-03-26 20:51:16,200 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-26 20:51:16,625 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7999, 1.6052, 1.4119, 1.2738, 1.5659, 1.5717, 1.5864, 2.1725], + device='cuda:1'), covar=tensor([0.3619, 0.3500, 0.3008, 0.3525, 0.3700, 0.2216, 0.3332, 0.1748], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0260, 0.0227, 0.0275, 0.0250, 0.0219, 0.0250, 0.0230], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:51:23,779 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8517, 1.7151, 1.7126, 1.8267, 1.4005, 4.4201, 1.6882, 2.0372], + device='cuda:1'), covar=tensor([0.3163, 0.2477, 0.2059, 0.2306, 0.1697, 0.0152, 0.2377, 0.1255], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0122, 0.0113, 0.0096, 0.0096, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 20:51:32,707 INFO [finetune.py:976] (1/7) Epoch 17, batch 3350, loss[loss=0.1935, simple_loss=0.2655, pruned_loss=0.06071, over 4899.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.252, pruned_loss=0.05637, over 954751.43 frames. ], batch size: 35, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 20:51:36,490 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7104, 3.6085, 3.4371, 1.7219, 3.7423, 2.8552, 0.8487, 2.6434], + device='cuda:1'), covar=tensor([0.2603, 0.1918, 0.1585, 0.3194, 0.0958, 0.0954, 0.4126, 0.1363], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0176, 0.0159, 0.0129, 0.0159, 0.0124, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 20:51:40,201 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95004.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:51:43,875 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-26 20:52:06,010 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3065, 2.2242, 1.8908, 2.3184, 2.1010, 2.1154, 2.1789, 3.0985], + device='cuda:1'), covar=tensor([0.4131, 0.5047, 0.3743, 0.4615, 0.4803, 0.2724, 0.4511, 0.1749], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0261, 0.0228, 0.0277, 0.0251, 0.0219, 0.0251, 0.0231], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:52:06,489 INFO [finetune.py:976] (1/7) Epoch 17, batch 3400, loss[loss=0.15, simple_loss=0.2358, pruned_loss=0.03207, over 4781.00 frames. ], tot_loss[loss=0.183, simple_loss=0.2527, pruned_loss=0.05669, over 953431.39 frames. ], batch size: 29, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:52:10,131 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.060e+02 1.553e+02 1.863e+02 2.086e+02 3.757e+02, threshold=3.727e+02, percent-clipped=0.0 +2023-03-26 20:52:12,047 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=95052.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:52:16,226 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2466, 2.0449, 1.9295, 2.2856, 2.6932, 2.2986, 2.1030, 1.8532], + device='cuda:1'), covar=tensor([0.1948, 0.1862, 0.1774, 0.1460, 0.1428, 0.1074, 0.1973, 0.1835], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0210, 0.0213, 0.0192, 0.0242, 0.0187, 0.0216, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:52:19,900 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8628, 1.6140, 2.0136, 1.3158, 1.9661, 1.9785, 1.5334, 2.1182], + device='cuda:1'), covar=tensor([0.1219, 0.2003, 0.1374, 0.2031, 0.0735, 0.1391, 0.2725, 0.0797], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0204, 0.0190, 0.0190, 0.0177, 0.0212, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:52:28,680 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.54 vs. limit=5.0 +2023-03-26 20:52:40,293 INFO [finetune.py:976] (1/7) Epoch 17, batch 3450, loss[loss=0.185, simple_loss=0.2449, pruned_loss=0.06258, over 4736.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2534, pruned_loss=0.05678, over 953412.54 frames. ], batch size: 54, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:52:45,928 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5396, 2.4461, 1.9038, 2.6030, 2.4503, 2.0200, 2.9216, 2.5125], + device='cuda:1'), covar=tensor([0.1288, 0.2105, 0.3055, 0.2615, 0.2624, 0.1671, 0.3069, 0.1824], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0188, 0.0234, 0.0252, 0.0244, 0.0202, 0.0212, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:52:47,677 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95105.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 20:53:10,939 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 20:53:13,090 INFO [finetune.py:976] (1/7) Epoch 17, batch 3500, loss[loss=0.1921, simple_loss=0.2548, pruned_loss=0.06471, over 4897.00 frames. ], tot_loss[loss=0.1814, simple_loss=0.2507, pruned_loss=0.05605, over 954709.17 frames. ], batch size: 32, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:53:17,182 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.619e+02 1.940e+02 2.279e+02 3.817e+02, threshold=3.880e+02, percent-clipped=1.0 +2023-03-26 20:53:35,211 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95166.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 20:53:47,158 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9106, 3.7536, 3.5868, 1.8063, 3.8503, 2.8869, 0.9892, 2.6641], + device='cuda:1'), covar=tensor([0.2065, 0.2203, 0.1719, 0.3525, 0.1182, 0.1034, 0.4667, 0.1621], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0176, 0.0160, 0.0129, 0.0159, 0.0124, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 20:54:05,084 INFO [finetune.py:976] (1/7) Epoch 17, batch 3550, loss[loss=0.1787, simple_loss=0.2481, pruned_loss=0.05467, over 4754.00 frames. ], tot_loss[loss=0.1793, simple_loss=0.2481, pruned_loss=0.05527, over 955732.56 frames. ], batch size: 27, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:54:51,064 INFO [finetune.py:976] (1/7) Epoch 17, batch 3600, loss[loss=0.2173, simple_loss=0.2695, pruned_loss=0.08259, over 4846.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2459, pruned_loss=0.05443, over 956952.36 frames. ], batch size: 49, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:54:54,645 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.130e+02 1.580e+02 1.871e+02 2.182e+02 4.206e+02, threshold=3.742e+02, percent-clipped=1.0 +2023-03-26 20:55:01,982 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95260.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:55:06,900 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95268.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:55:24,756 INFO [finetune.py:976] (1/7) Epoch 17, batch 3650, loss[loss=0.1391, simple_loss=0.2124, pruned_loss=0.03294, over 4889.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2474, pruned_loss=0.0547, over 956433.28 frames. ], batch size: 32, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:55:42,980 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95321.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:55:46,738 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.67 vs. limit=5.0 +2023-03-26 20:55:47,159 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6208, 1.3255, 0.7909, 1.6287, 2.2168, 1.3594, 1.4788, 1.6509], + device='cuda:1'), covar=tensor([0.1890, 0.2740, 0.2442, 0.1559, 0.2037, 0.2430, 0.1959, 0.2606], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0095, 0.0110, 0.0092, 0.0119, 0.0095, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 20:55:48,238 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95329.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:55:55,797 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95339.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:55:58,601 INFO [finetune.py:976] (1/7) Epoch 17, batch 3700, loss[loss=0.1797, simple_loss=0.2535, pruned_loss=0.05299, over 4744.00 frames. ], tot_loss[loss=0.18, simple_loss=0.2502, pruned_loss=0.05493, over 955623.70 frames. ], batch size: 59, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:56:02,229 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.130e+02 1.605e+02 1.907e+02 2.409e+02 3.957e+02, threshold=3.813e+02, percent-clipped=4.0 +2023-03-26 20:56:31,738 INFO [finetune.py:976] (1/7) Epoch 17, batch 3750, loss[loss=0.1812, simple_loss=0.2651, pruned_loss=0.04861, over 4887.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2522, pruned_loss=0.05569, over 955146.38 frames. ], batch size: 43, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:56:36,733 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95400.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:56:51,864 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8716, 1.4365, 1.9476, 1.8806, 1.6318, 1.6137, 1.8107, 1.7751], + device='cuda:1'), covar=tensor([0.4091, 0.4312, 0.3339, 0.3732, 0.5125, 0.3803, 0.4611, 0.3282], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0238, 0.0257, 0.0272, 0.0269, 0.0245, 0.0280, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:57:04,528 INFO [finetune.py:976] (1/7) Epoch 17, batch 3800, loss[loss=0.1978, simple_loss=0.2683, pruned_loss=0.06372, over 4868.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2539, pruned_loss=0.05656, over 954931.19 frames. ], batch size: 34, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:57:09,550 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.108e+02 1.488e+02 1.737e+02 2.235e+02 4.648e+02, threshold=3.475e+02, percent-clipped=3.0 +2023-03-26 20:57:16,359 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0560, 0.9645, 1.0031, 0.4828, 0.9699, 1.1707, 1.1958, 0.9966], + device='cuda:1'), covar=tensor([0.1034, 0.0718, 0.0685, 0.0601, 0.0699, 0.0853, 0.0596, 0.0840], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0152, 0.0124, 0.0128, 0.0132, 0.0130, 0.0144, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.2530e-05, 1.1043e-04, 8.8856e-05, 9.0985e-05, 9.2837e-05, 9.3164e-05, + 1.0394e-04, 1.0727e-04], device='cuda:1') +2023-03-26 20:57:16,895 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95461.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 20:57:37,562 INFO [finetune.py:976] (1/7) Epoch 17, batch 3850, loss[loss=0.1669, simple_loss=0.2405, pruned_loss=0.04661, over 4839.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2533, pruned_loss=0.05651, over 955796.81 frames. ], batch size: 49, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:57:42,253 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7443, 1.7118, 1.4519, 1.7182, 2.1546, 1.9889, 1.6448, 1.5562], + device='cuda:1'), covar=tensor([0.0262, 0.0324, 0.0576, 0.0300, 0.0177, 0.0527, 0.0367, 0.0402], + device='cuda:1'), in_proj_covar=tensor([0.0095, 0.0108, 0.0143, 0.0113, 0.0099, 0.0109, 0.0099, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4014e-05, 8.3327e-05, 1.1320e-04, 8.7090e-05, 7.7144e-05, 8.0629e-05, + 7.3834e-05, 8.4177e-05], device='cuda:1') +2023-03-26 20:58:10,766 INFO [finetune.py:976] (1/7) Epoch 17, batch 3900, loss[loss=0.1443, simple_loss=0.2249, pruned_loss=0.03183, over 4748.00 frames. ], tot_loss[loss=0.1815, simple_loss=0.2511, pruned_loss=0.05595, over 956866.30 frames. ], batch size: 27, lr: 3.38e-03, grad_scale: 64.0 +2023-03-26 20:58:15,379 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.225e+02 1.550e+02 1.834e+02 2.229e+02 4.290e+02, threshold=3.669e+02, percent-clipped=3.0 +2023-03-26 20:58:46,333 INFO [finetune.py:976] (1/7) Epoch 17, batch 3950, loss[loss=0.1584, simple_loss=0.2346, pruned_loss=0.04112, over 4830.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2489, pruned_loss=0.05498, over 956562.21 frames. ], batch size: 39, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 20:58:51,438 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1613, 2.0336, 1.7134, 1.9874, 1.8939, 1.8872, 1.9074, 2.7228], + device='cuda:1'), covar=tensor([0.3888, 0.4539, 0.3311, 0.3985, 0.4249, 0.2492, 0.4040, 0.1698], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0262, 0.0227, 0.0277, 0.0251, 0.0219, 0.0251, 0.0231], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:59:04,354 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95616.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:59:05,025 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95617.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:59:07,528 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8778, 1.8015, 1.6220, 1.9696, 2.3840, 1.9412, 1.7086, 1.6094], + device='cuda:1'), covar=tensor([0.1623, 0.1576, 0.1531, 0.1281, 0.1340, 0.1047, 0.2173, 0.1510], + device='cuda:1'), in_proj_covar=tensor([0.0240, 0.0208, 0.0211, 0.0190, 0.0240, 0.0186, 0.0214, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 20:59:13,740 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95624.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 20:59:34,007 INFO [finetune.py:976] (1/7) Epoch 17, batch 4000, loss[loss=0.1922, simple_loss=0.2548, pruned_loss=0.06481, over 4920.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2469, pruned_loss=0.05495, over 951430.08 frames. ], batch size: 38, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 20:59:42,361 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.075e+02 1.540e+02 1.979e+02 2.285e+02 3.877e+02, threshold=3.958e+02, percent-clipped=2.0 +2023-03-26 21:00:12,621 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95678.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:00:22,563 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5853, 1.4398, 1.5993, 0.7714, 1.5624, 1.5095, 1.5077, 1.4008], + device='cuda:1'), covar=tensor([0.0595, 0.0856, 0.0694, 0.1048, 0.0854, 0.0811, 0.0683, 0.1280], + device='cuda:1'), in_proj_covar=tensor([0.0136, 0.0137, 0.0143, 0.0125, 0.0126, 0.0143, 0.0144, 0.0166], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:00:26,082 INFO [finetune.py:976] (1/7) Epoch 17, batch 4050, loss[loss=0.2279, simple_loss=0.3004, pruned_loss=0.0777, over 4861.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2504, pruned_loss=0.05598, over 952327.14 frames. ], batch size: 44, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 21:00:27,355 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95695.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:00:36,727 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6286, 1.6340, 2.1284, 3.2302, 2.1823, 2.2938, 0.9258, 2.6337], + device='cuda:1'), covar=tensor([0.1741, 0.1382, 0.1245, 0.0549, 0.0807, 0.1383, 0.1859, 0.0512], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0165, 0.0101, 0.0135, 0.0123, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:00:37,325 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6241, 1.5180, 2.1374, 3.2667, 2.1651, 2.2843, 1.3227, 2.7074], + device='cuda:1'), covar=tensor([0.1875, 0.1664, 0.1400, 0.0769, 0.0946, 0.1702, 0.1768, 0.0569], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0165, 0.0101, 0.0135, 0.0123, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:00:59,870 INFO [finetune.py:976] (1/7) Epoch 17, batch 4100, loss[loss=0.1544, simple_loss=0.2207, pruned_loss=0.0441, over 3997.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2531, pruned_loss=0.05661, over 951752.72 frames. ], batch size: 17, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 21:01:04,067 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.117e+02 1.600e+02 1.864e+02 2.304e+02 4.240e+02, threshold=3.729e+02, percent-clipped=2.0 +2023-03-26 21:01:12,319 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95761.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 21:01:33,030 INFO [finetune.py:976] (1/7) Epoch 17, batch 4150, loss[loss=0.1395, simple_loss=0.2099, pruned_loss=0.03457, over 4723.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2552, pruned_loss=0.05806, over 952697.39 frames. ], batch size: 23, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 21:01:44,396 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=95809.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:01:44,409 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0451, 3.6467, 3.7872, 3.7111, 3.5868, 3.5327, 4.1811, 1.3050], + device='cuda:1'), covar=tensor([0.1003, 0.1367, 0.1189, 0.1511, 0.1813, 0.2156, 0.1168, 0.6680], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0246, 0.0277, 0.0293, 0.0336, 0.0282, 0.0302, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:01:46,220 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95812.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 21:02:06,757 INFO [finetune.py:976] (1/7) Epoch 17, batch 4200, loss[loss=0.1897, simple_loss=0.2646, pruned_loss=0.05736, over 4816.00 frames. ], tot_loss[loss=0.1862, simple_loss=0.256, pruned_loss=0.05815, over 954111.19 frames. ], batch size: 39, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 21:02:09,292 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95847.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:02:09,897 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5922, 1.1486, 0.7992, 1.4600, 1.9589, 1.0688, 1.3138, 1.4932], + device='cuda:1'), covar=tensor([0.1460, 0.2089, 0.1866, 0.1239, 0.1878, 0.2008, 0.1477, 0.1897], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0118, 0.0094, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:02:11,507 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.143e+02 1.538e+02 1.932e+02 2.354e+02 8.206e+02, threshold=3.863e+02, percent-clipped=2.0 +2023-03-26 21:02:27,890 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95873.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:02:29,078 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7592, 2.3874, 2.9652, 1.7189, 2.6777, 2.8951, 2.1559, 3.0365], + device='cuda:1'), covar=tensor([0.1212, 0.1796, 0.1370, 0.2344, 0.0913, 0.1460, 0.2671, 0.0744], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0203, 0.0189, 0.0189, 0.0176, 0.0211, 0.0215, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:02:39,921 INFO [finetune.py:976] (1/7) Epoch 17, batch 4250, loss[loss=0.168, simple_loss=0.2355, pruned_loss=0.05031, over 4728.00 frames. ], tot_loss[loss=0.1837, simple_loss=0.2533, pruned_loss=0.05708, over 953505.54 frames. ], batch size: 54, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 21:02:43,527 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7945, 3.3275, 3.4807, 3.6760, 3.5736, 3.3615, 3.8695, 1.2604], + device='cuda:1'), covar=tensor([0.0863, 0.0905, 0.0949, 0.1028, 0.1304, 0.1642, 0.0908, 0.5356], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0245, 0.0276, 0.0292, 0.0336, 0.0282, 0.0301, 0.0295], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:02:50,094 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95908.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:02:55,856 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95916.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:03:02,179 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95924.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:03:12,436 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0626, 1.7666, 2.3908, 1.4581, 2.1410, 2.3036, 1.6699, 2.5605], + device='cuda:1'), covar=tensor([0.1216, 0.2045, 0.1320, 0.2233, 0.0887, 0.1504, 0.2637, 0.0683], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0203, 0.0189, 0.0189, 0.0176, 0.0211, 0.0215, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:03:13,534 INFO [finetune.py:976] (1/7) Epoch 17, batch 4300, loss[loss=0.1438, simple_loss=0.2115, pruned_loss=0.03804, over 4722.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2504, pruned_loss=0.05595, over 953320.40 frames. ], batch size: 54, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 21:03:18,258 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.024e+02 1.465e+02 1.654e+02 2.123e+02 3.225e+02, threshold=3.308e+02, percent-clipped=0.0 +2023-03-26 21:03:27,285 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=95964.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:03:32,111 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 21:03:33,601 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=95972.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:03:34,210 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95973.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:03:39,469 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95980.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:03:47,254 INFO [finetune.py:976] (1/7) Epoch 17, batch 4350, loss[loss=0.1505, simple_loss=0.2174, pruned_loss=0.04178, over 4825.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.2467, pruned_loss=0.05459, over 954766.64 frames. ], batch size: 30, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 21:03:48,532 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95995.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:03:54,756 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.05 vs. limit=5.0 +2023-03-26 21:04:19,590 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4216, 2.3654, 1.7920, 0.9182, 2.0313, 1.8214, 1.7668, 2.1419], + device='cuda:1'), covar=tensor([0.0869, 0.0797, 0.1712, 0.2020, 0.1477, 0.2384, 0.2153, 0.0915], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0196, 0.0199, 0.0183, 0.0212, 0.0207, 0.0223, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:04:20,836 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96041.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:04:21,941 INFO [finetune.py:976] (1/7) Epoch 17, batch 4400, loss[loss=0.1743, simple_loss=0.2478, pruned_loss=0.05033, over 4901.00 frames. ], tot_loss[loss=0.1787, simple_loss=0.2478, pruned_loss=0.05477, over 956068.84 frames. ], batch size: 32, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 21:04:22,001 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=96043.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:04:28,706 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.398e+01 1.490e+02 1.749e+02 2.200e+02 3.209e+02, threshold=3.497e+02, percent-clipped=0.0 +2023-03-26 21:04:59,177 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96080.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:05:00,458 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-26 21:05:13,436 INFO [finetune.py:976] (1/7) Epoch 17, batch 4450, loss[loss=0.1494, simple_loss=0.2247, pruned_loss=0.03703, over 4752.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2515, pruned_loss=0.05612, over 954566.20 frames. ], batch size: 28, lr: 3.38e-03, grad_scale: 32.0 +2023-03-26 21:05:41,020 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-26 21:05:58,466 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96141.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:05:59,552 INFO [finetune.py:976] (1/7) Epoch 17, batch 4500, loss[loss=0.1748, simple_loss=0.2582, pruned_loss=0.04569, over 4817.00 frames. ], tot_loss[loss=0.1839, simple_loss=0.2537, pruned_loss=0.05701, over 952631.51 frames. ], batch size: 38, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:06:03,840 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.121e+02 1.724e+02 1.946e+02 2.358e+02 4.504e+02, threshold=3.891e+02, percent-clipped=3.0 +2023-03-26 21:06:13,987 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96165.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:06:15,770 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96168.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 21:06:17,801 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 21:06:33,251 INFO [finetune.py:976] (1/7) Epoch 17, batch 4550, loss[loss=0.1973, simple_loss=0.2631, pruned_loss=0.06576, over 4853.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2533, pruned_loss=0.05674, over 954292.72 frames. ], batch size: 31, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:06:39,497 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96203.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 21:06:54,570 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96226.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:07:07,173 INFO [finetune.py:976] (1/7) Epoch 17, batch 4600, loss[loss=0.1569, simple_loss=0.2339, pruned_loss=0.03991, over 4747.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2525, pruned_loss=0.05622, over 955269.64 frames. ], batch size: 54, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:07:07,999 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 21:07:11,422 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.099e+02 1.530e+02 1.886e+02 2.340e+02 4.335e+02, threshold=3.772e+02, percent-clipped=2.0 +2023-03-26 21:07:26,486 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96273.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:07:40,044 INFO [finetune.py:976] (1/7) Epoch 17, batch 4650, loss[loss=0.1643, simple_loss=0.2377, pruned_loss=0.04546, over 4816.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2512, pruned_loss=0.05627, over 956081.50 frames. ], batch size: 38, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:07:57,957 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 21:07:58,449 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=96321.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:08:00,272 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96324.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:08:08,017 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96336.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:08:13,204 INFO [finetune.py:976] (1/7) Epoch 17, batch 4700, loss[loss=0.1837, simple_loss=0.2399, pruned_loss=0.0637, over 4256.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2484, pruned_loss=0.0555, over 955352.78 frames. ], batch size: 66, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:08:18,315 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.127e+02 1.531e+02 1.882e+02 2.216e+02 4.319e+02, threshold=3.764e+02, percent-clipped=2.0 +2023-03-26 21:08:19,715 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.14 vs. limit=5.0 +2023-03-26 21:08:40,441 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96385.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:08:43,985 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96390.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:08:45,650 INFO [finetune.py:976] (1/7) Epoch 17, batch 4750, loss[loss=0.1812, simple_loss=0.2546, pruned_loss=0.05387, over 4823.00 frames. ], tot_loss[loss=0.1787, simple_loss=0.247, pruned_loss=0.05522, over 957209.63 frames. ], batch size: 30, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:08:54,530 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-03-26 21:09:09,986 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5344, 1.6412, 1.8898, 1.7486, 1.6506, 3.3875, 1.4010, 1.5806], + device='cuda:1'), covar=tensor([0.0910, 0.1630, 0.1137, 0.0930, 0.1495, 0.0234, 0.1446, 0.1712], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0078, 0.0091, 0.0081, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:09:13,606 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9953, 1.9648, 1.5825, 1.8920, 1.7951, 1.7862, 1.8692, 2.5506], + device='cuda:1'), covar=tensor([0.4271, 0.4503, 0.3623, 0.4072, 0.4544, 0.2708, 0.4337, 0.1865], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0228, 0.0278, 0.0253, 0.0220, 0.0252, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:09:14,756 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96436.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:09:19,420 INFO [finetune.py:976] (1/7) Epoch 17, batch 4800, loss[loss=0.2092, simple_loss=0.2766, pruned_loss=0.07089, over 4901.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2504, pruned_loss=0.05641, over 958351.73 frames. ], batch size: 32, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:09:25,022 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.112e+02 1.609e+02 1.875e+02 2.422e+02 6.864e+02, threshold=3.750e+02, percent-clipped=2.0 +2023-03-26 21:09:25,773 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96451.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:09:36,645 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96468.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 21:09:54,757 INFO [finetune.py:976] (1/7) Epoch 17, batch 4850, loss[loss=0.1834, simple_loss=0.2476, pruned_loss=0.05958, over 4836.00 frames. ], tot_loss[loss=0.1845, simple_loss=0.2538, pruned_loss=0.05763, over 957805.51 frames. ], batch size: 30, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:10:02,903 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96503.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:10:15,683 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=96516.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:10:18,746 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96521.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:10:45,278 INFO [finetune.py:976] (1/7) Epoch 17, batch 4900, loss[loss=0.2221, simple_loss=0.2797, pruned_loss=0.0822, over 4885.00 frames. ], tot_loss[loss=0.1851, simple_loss=0.2543, pruned_loss=0.05794, over 956629.23 frames. ], batch size: 35, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:10:54,622 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.014e+02 1.592e+02 1.896e+02 2.164e+02 3.347e+02, threshold=3.792e+02, percent-clipped=0.0 +2023-03-26 21:10:55,802 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=96551.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:11:26,123 INFO [finetune.py:976] (1/7) Epoch 17, batch 4950, loss[loss=0.1741, simple_loss=0.2458, pruned_loss=0.05113, over 4758.00 frames. ], tot_loss[loss=0.1859, simple_loss=0.2551, pruned_loss=0.0583, over 954077.92 frames. ], batch size: 26, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:11:55,611 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96636.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:11:59,761 INFO [finetune.py:976] (1/7) Epoch 17, batch 5000, loss[loss=0.156, simple_loss=0.2236, pruned_loss=0.04424, over 4866.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.254, pruned_loss=0.05765, over 952390.84 frames. ], batch size: 34, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:12:04,410 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.088e+02 1.531e+02 1.819e+02 2.156e+02 3.437e+02, threshold=3.638e+02, percent-clipped=0.0 +2023-03-26 21:12:24,596 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96680.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:12:26,964 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=96684.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:12:33,329 INFO [finetune.py:976] (1/7) Epoch 17, batch 5050, loss[loss=0.1875, simple_loss=0.2532, pruned_loss=0.06085, over 4916.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2517, pruned_loss=0.05743, over 955050.98 frames. ], batch size: 36, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:12:51,177 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-26 21:13:00,181 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0 +2023-03-26 21:13:01,806 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96736.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:13:06,444 INFO [finetune.py:976] (1/7) Epoch 17, batch 5100, loss[loss=0.1291, simple_loss=0.2096, pruned_loss=0.02434, over 4755.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2488, pruned_loss=0.05615, over 955566.79 frames. ], batch size: 27, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:13:08,316 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96746.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:13:10,589 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.091e+02 1.513e+02 1.848e+02 2.246e+02 3.685e+02, threshold=3.695e+02, percent-clipped=1.0 +2023-03-26 21:13:28,946 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96776.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:13:32,465 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4152, 1.3237, 1.6664, 2.4366, 1.5784, 2.1872, 0.9461, 2.0691], + device='cuda:1'), covar=tensor([0.1753, 0.1442, 0.1172, 0.0718, 0.0947, 0.1068, 0.1527, 0.0676], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0135, 0.0166, 0.0102, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:13:33,646 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=96784.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:13:39,090 INFO [finetune.py:976] (1/7) Epoch 17, batch 5150, loss[loss=0.1911, simple_loss=0.2662, pruned_loss=0.05802, over 4820.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.25, pruned_loss=0.05721, over 956145.36 frames. ], batch size: 30, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:13:57,076 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9729, 1.7943, 1.5624, 1.6770, 1.7338, 1.7047, 1.7801, 2.4908], + device='cuda:1'), covar=tensor([0.3937, 0.4365, 0.3362, 0.3714, 0.3906, 0.2617, 0.3846, 0.1665], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0225, 0.0274, 0.0250, 0.0218, 0.0250, 0.0229], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:13:58,209 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96821.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:14:08,378 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96837.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:14:11,884 INFO [finetune.py:976] (1/7) Epoch 17, batch 5200, loss[loss=0.1904, simple_loss=0.2607, pruned_loss=0.06002, over 4750.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2534, pruned_loss=0.0586, over 955011.38 frames. ], batch size: 59, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:14:16,599 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.064e+02 1.635e+02 1.977e+02 2.300e+02 5.939e+02, threshold=3.955e+02, percent-clipped=5.0 +2023-03-26 21:14:29,994 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=96869.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:14:30,734 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-26 21:14:44,941 INFO [finetune.py:976] (1/7) Epoch 17, batch 5250, loss[loss=0.185, simple_loss=0.2555, pruned_loss=0.05722, over 4910.00 frames. ], tot_loss[loss=0.1874, simple_loss=0.2561, pruned_loss=0.05941, over 955118.97 frames. ], batch size: 37, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:15:21,072 INFO [finetune.py:976] (1/7) Epoch 17, batch 5300, loss[loss=0.1674, simple_loss=0.2458, pruned_loss=0.04452, over 4904.00 frames. ], tot_loss[loss=0.1878, simple_loss=0.2572, pruned_loss=0.0592, over 955787.83 frames. ], batch size: 36, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:15:23,074 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.63 vs. limit=2.0 +2023-03-26 21:15:30,008 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.125e+02 1.644e+02 1.959e+02 2.379e+02 3.599e+02, threshold=3.918e+02, percent-clipped=0.0 +2023-03-26 21:16:00,367 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96980.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:16:17,642 INFO [finetune.py:976] (1/7) Epoch 17, batch 5350, loss[loss=0.1378, simple_loss=0.2148, pruned_loss=0.03042, over 4796.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2561, pruned_loss=0.0586, over 954366.59 frames. ], batch size: 25, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:16:25,312 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97000.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:16:35,475 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-26 21:16:44,613 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=97028.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:16:54,582 INFO [finetune.py:976] (1/7) Epoch 17, batch 5400, loss[loss=0.1324, simple_loss=0.2103, pruned_loss=0.02724, over 4744.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.252, pruned_loss=0.05732, over 953214.08 frames. ], batch size: 23, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:16:56,523 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97046.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:16:58,799 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.012e+02 1.591e+02 1.860e+02 2.348e+02 4.043e+02, threshold=3.721e+02, percent-clipped=1.0 +2023-03-26 21:17:06,279 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97061.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:17:27,398 INFO [finetune.py:976] (1/7) Epoch 17, batch 5450, loss[loss=0.1444, simple_loss=0.206, pruned_loss=0.0414, over 4773.00 frames. ], tot_loss[loss=0.181, simple_loss=0.2492, pruned_loss=0.05639, over 956257.94 frames. ], batch size: 27, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:17:28,070 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=97094.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:17:52,423 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97132.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:17:56,001 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7938, 1.4014, 0.6075, 1.6074, 2.2642, 1.4263, 1.5838, 1.6924], + device='cuda:1'), covar=tensor([0.1518, 0.2100, 0.2164, 0.1288, 0.1874, 0.2125, 0.1578, 0.1965], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0095, 0.0110, 0.0092, 0.0119, 0.0094, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:18:00,547 INFO [finetune.py:976] (1/7) Epoch 17, batch 5500, loss[loss=0.1605, simple_loss=0.2409, pruned_loss=0.04007, over 4888.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2465, pruned_loss=0.055, over 957309.00 frames. ], batch size: 43, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:18:04,745 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.221e+01 1.533e+02 1.769e+02 2.042e+02 3.202e+02, threshold=3.539e+02, percent-clipped=0.0 +2023-03-26 21:18:20,549 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-26 21:18:33,719 INFO [finetune.py:976] (1/7) Epoch 17, batch 5550, loss[loss=0.2802, simple_loss=0.3218, pruned_loss=0.1193, over 4054.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2495, pruned_loss=0.05648, over 954817.67 frames. ], batch size: 65, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:18:34,908 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8038, 1.6749, 1.7463, 1.2708, 1.7999, 1.9183, 1.8758, 1.4547], + device='cuda:1'), covar=tensor([0.0559, 0.0630, 0.0609, 0.0818, 0.0757, 0.0536, 0.0508, 0.1112], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0134, 0.0140, 0.0122, 0.0124, 0.0139, 0.0141, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:19:05,110 INFO [finetune.py:976] (1/7) Epoch 17, batch 5600, loss[loss=0.2191, simple_loss=0.2927, pruned_loss=0.07274, over 4910.00 frames. ], tot_loss[loss=0.1837, simple_loss=0.2528, pruned_loss=0.05734, over 955669.99 frames. ], batch size: 36, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:19:09,079 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.154e+02 1.641e+02 1.914e+02 2.406e+02 4.422e+02, threshold=3.827e+02, percent-clipped=1.0 +2023-03-26 21:19:14,873 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97260.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:19:29,798 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3838, 1.2897, 1.3164, 1.2785, 0.8344, 2.1828, 0.8202, 1.1951], + device='cuda:1'), covar=tensor([0.2954, 0.2262, 0.1945, 0.2284, 0.1846, 0.0407, 0.2951, 0.1276], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0121, 0.0123, 0.0113, 0.0096, 0.0096, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:19:34,291 INFO [finetune.py:976] (1/7) Epoch 17, batch 5650, loss[loss=0.2125, simple_loss=0.2899, pruned_loss=0.06754, over 4900.00 frames. ], tot_loss[loss=0.1866, simple_loss=0.2563, pruned_loss=0.05847, over 956380.17 frames. ], batch size: 37, lr: 3.37e-03, grad_scale: 32.0 +2023-03-26 21:19:51,463 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97321.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:20:04,110 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97342.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:20:04,605 INFO [finetune.py:976] (1/7) Epoch 17, batch 5700, loss[loss=0.1504, simple_loss=0.2022, pruned_loss=0.04932, over 4371.00 frames. ], tot_loss[loss=0.1832, simple_loss=0.2516, pruned_loss=0.05739, over 937787.10 frames. ], batch size: 19, lr: 3.36e-03, grad_scale: 32.0 +2023-03-26 21:20:05,878 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7547, 1.6371, 1.5920, 1.6854, 1.2137, 2.9907, 1.2691, 1.6947], + device='cuda:1'), covar=tensor([0.3072, 0.2360, 0.2054, 0.2211, 0.1706, 0.0294, 0.2468, 0.1255], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0121, 0.0123, 0.0113, 0.0096, 0.0096, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:20:07,045 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97347.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:20:08,735 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.528e+01 1.533e+02 1.739e+02 2.212e+02 3.283e+02, threshold=3.478e+02, percent-clipped=0.0 +2023-03-26 21:20:12,314 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97356.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:20:35,933 INFO [finetune.py:976] (1/7) Epoch 18, batch 0, loss[loss=0.1739, simple_loss=0.2497, pruned_loss=0.04904, over 4919.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2497, pruned_loss=0.04904, over 4919.00 frames. ], batch size: 38, lr: 3.36e-03, grad_scale: 32.0 +2023-03-26 21:20:35,933 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 21:20:46,783 INFO [finetune.py:1010] (1/7) Epoch 18, validation: loss=0.1584, simple_loss=0.2281, pruned_loss=0.0444, over 2265189.00 frames. +2023-03-26 21:20:46,783 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 21:20:49,164 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97374.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:21:26,105 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97403.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 21:21:30,202 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97408.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:21:47,682 INFO [finetune.py:976] (1/7) Epoch 18, batch 50, loss[loss=0.2035, simple_loss=0.2768, pruned_loss=0.06509, over 4783.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2472, pruned_loss=0.05209, over 213603.15 frames. ], batch size: 51, lr: 3.36e-03, grad_scale: 32.0 +2023-03-26 21:21:58,962 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97432.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:22:00,780 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97435.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:22:09,806 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.163e+01 1.563e+02 1.902e+02 2.308e+02 3.615e+02, threshold=3.804e+02, percent-clipped=1.0 +2023-03-26 21:22:25,130 INFO [finetune.py:976] (1/7) Epoch 18, batch 100, loss[loss=0.1744, simple_loss=0.2468, pruned_loss=0.05102, over 4867.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2452, pruned_loss=0.054, over 378905.33 frames. ], batch size: 34, lr: 3.36e-03, grad_scale: 32.0 +2023-03-26 21:22:31,665 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=97480.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:22:52,386 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8054, 1.6664, 1.6556, 1.7181, 1.2205, 3.6164, 1.4802, 2.0142], + device='cuda:1'), covar=tensor([0.3227, 0.2372, 0.2022, 0.2244, 0.1736, 0.0186, 0.2484, 0.1183], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0114, 0.0119, 0.0122, 0.0112, 0.0095, 0.0095, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:22:58,721 INFO [finetune.py:976] (1/7) Epoch 18, batch 150, loss[loss=0.1578, simple_loss=0.2331, pruned_loss=0.04129, over 4830.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2412, pruned_loss=0.0531, over 507490.53 frames. ], batch size: 30, lr: 3.36e-03, grad_scale: 32.0 +2023-03-26 21:23:13,595 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0206, 4.9753, 4.7175, 2.5684, 4.9817, 3.8502, 0.9179, 3.6435], + device='cuda:1'), covar=tensor([0.2261, 0.1831, 0.1446, 0.3327, 0.0757, 0.0886, 0.4698, 0.1307], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0175, 0.0158, 0.0128, 0.0158, 0.0122, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 21:23:17,208 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.032e+02 1.559e+02 1.792e+02 2.227e+02 6.409e+02, threshold=3.584e+02, percent-clipped=2.0 +2023-03-26 21:23:20,339 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97555.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:23:32,346 INFO [finetune.py:976] (1/7) Epoch 18, batch 200, loss[loss=0.1973, simple_loss=0.2602, pruned_loss=0.06718, over 4774.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2428, pruned_loss=0.05434, over 607292.48 frames. ], batch size: 54, lr: 3.36e-03, grad_scale: 32.0 +2023-03-26 21:23:43,739 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0096, 1.3910, 0.8067, 1.9678, 2.3197, 1.6664, 1.5386, 1.9284], + device='cuda:1'), covar=tensor([0.1465, 0.2088, 0.2034, 0.1152, 0.1861, 0.2055, 0.1504, 0.1860], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0095, 0.0110, 0.0092, 0.0119, 0.0094, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:24:01,851 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97616.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:24:01,907 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97616.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:24:05,307 INFO [finetune.py:976] (1/7) Epoch 18, batch 250, loss[loss=0.182, simple_loss=0.2618, pruned_loss=0.05104, over 4920.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2469, pruned_loss=0.05515, over 685734.32 frames. ], batch size: 42, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:24:16,419 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3664, 1.2861, 1.6855, 2.4574, 1.6133, 2.1210, 0.9285, 2.0688], + device='cuda:1'), covar=tensor([0.1789, 0.1529, 0.1180, 0.0755, 0.0991, 0.1257, 0.1578, 0.0636], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0134, 0.0164, 0.0100, 0.0136, 0.0123, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:24:18,252 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5815, 1.1391, 0.8149, 1.5160, 2.0120, 1.3825, 1.2965, 1.5555], + device='cuda:1'), covar=tensor([0.1585, 0.2129, 0.1986, 0.1227, 0.2010, 0.2186, 0.1541, 0.1944], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0119, 0.0094, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:24:24,121 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.114e+02 1.615e+02 1.960e+02 2.417e+02 4.168e+02, threshold=3.921e+02, percent-clipped=3.0 +2023-03-26 21:24:27,908 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97656.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:24:37,889 INFO [finetune.py:976] (1/7) Epoch 18, batch 300, loss[loss=0.1652, simple_loss=0.2404, pruned_loss=0.04504, over 4748.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2509, pruned_loss=0.0557, over 746714.07 frames. ], batch size: 27, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:24:40,275 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97674.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:24:56,232 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97698.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 21:24:59,261 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97703.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:24:59,859 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=97704.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:25:10,597 INFO [finetune.py:976] (1/7) Epoch 18, batch 350, loss[loss=0.17, simple_loss=0.2577, pruned_loss=0.04117, over 4860.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2527, pruned_loss=0.05673, over 793995.17 frames. ], batch size: 49, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:25:17,545 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97730.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:25:20,594 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97735.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:25:28,928 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0094, 2.7307, 2.3909, 3.0897, 2.8364, 2.6800, 3.3432, 2.9127], + device='cuda:1'), covar=tensor([0.1286, 0.2426, 0.3022, 0.2634, 0.2566, 0.1667, 0.2583, 0.1753], + device='cuda:1'), in_proj_covar=tensor([0.0184, 0.0188, 0.0235, 0.0254, 0.0246, 0.0203, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:25:30,594 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.123e+02 1.611e+02 1.882e+02 2.387e+02 3.928e+02, threshold=3.763e+02, percent-clipped=1.0 +2023-03-26 21:25:43,299 INFO [finetune.py:976] (1/7) Epoch 18, batch 400, loss[loss=0.1651, simple_loss=0.2346, pruned_loss=0.0478, over 4877.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2532, pruned_loss=0.05683, over 829477.39 frames. ], batch size: 43, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:25:46,124 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7719, 1.2548, 0.8661, 1.7270, 2.1160, 1.5500, 1.5186, 1.7272], + device='cuda:1'), covar=tensor([0.1389, 0.1966, 0.1906, 0.1067, 0.1732, 0.1947, 0.1364, 0.1757], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0095, 0.0110, 0.0092, 0.0119, 0.0094, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:25:52,684 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.67 vs. limit=5.0 +2023-03-26 21:26:07,810 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-26 21:26:22,831 INFO [finetune.py:976] (1/7) Epoch 18, batch 450, loss[loss=0.1403, simple_loss=0.2029, pruned_loss=0.03889, over 4259.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2519, pruned_loss=0.05639, over 857660.57 frames. ], batch size: 18, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:26:52,266 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3212, 2.1623, 1.7819, 2.2256, 2.2100, 2.0016, 2.6325, 2.3217], + device='cuda:1'), covar=tensor([0.1257, 0.2107, 0.3160, 0.2721, 0.2530, 0.1622, 0.2758, 0.1638], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0187, 0.0234, 0.0253, 0.0244, 0.0201, 0.0213, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:27:01,042 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.011e+02 1.519e+02 1.772e+02 2.128e+02 3.513e+02, threshold=3.544e+02, percent-clipped=0.0 +2023-03-26 21:27:16,933 INFO [finetune.py:976] (1/7) Epoch 18, batch 500, loss[loss=0.1748, simple_loss=0.2449, pruned_loss=0.05233, over 4905.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2498, pruned_loss=0.05619, over 879788.36 frames. ], batch size: 37, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:27:29,760 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97888.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:27:30,610 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.99 vs. limit=2.0 +2023-03-26 21:27:44,610 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97911.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:27:47,670 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97916.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:27:50,632 INFO [finetune.py:976] (1/7) Epoch 18, batch 550, loss[loss=0.1416, simple_loss=0.2163, pruned_loss=0.0334, over 4765.00 frames. ], tot_loss[loss=0.179, simple_loss=0.247, pruned_loss=0.05546, over 897530.52 frames. ], batch size: 26, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:28:19,273 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97949.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:28:19,731 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.294e+01 1.530e+02 1.816e+02 2.060e+02 3.951e+02, threshold=3.633e+02, percent-clipped=3.0 +2023-03-26 21:28:28,260 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=97964.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:28:32,532 INFO [finetune.py:976] (1/7) Epoch 18, batch 600, loss[loss=0.1743, simple_loss=0.2426, pruned_loss=0.05297, over 4811.00 frames. ], tot_loss[loss=0.1788, simple_loss=0.2469, pruned_loss=0.05532, over 908202.71 frames. ], batch size: 39, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:28:51,970 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97998.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:28:56,757 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98003.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:29:07,660 INFO [finetune.py:976] (1/7) Epoch 18, batch 650, loss[loss=0.2339, simple_loss=0.3087, pruned_loss=0.07951, over 4908.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2496, pruned_loss=0.05566, over 919729.04 frames. ], batch size: 43, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:29:13,278 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98030.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 21:29:13,291 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98030.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:29:13,334 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5239, 1.6011, 1.3235, 1.5323, 1.9263, 1.7536, 1.5056, 1.3975], + device='cuda:1'), covar=tensor([0.0336, 0.0299, 0.0582, 0.0329, 0.0195, 0.0496, 0.0322, 0.0406], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0107, 0.0144, 0.0113, 0.0100, 0.0109, 0.0099, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4608e-05, 8.2891e-05, 1.1370e-04, 8.6678e-05, 7.8214e-05, 8.0433e-05, + 7.3747e-05, 8.3856e-05], device='cuda:1') +2023-03-26 21:29:17,296 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7536, 1.5616, 2.2080, 1.8710, 1.8720, 4.2037, 1.7143, 1.7568], + device='cuda:1'), covar=tensor([0.0915, 0.1829, 0.1197, 0.1014, 0.1585, 0.0185, 0.1500, 0.1838], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0081, 0.0074, 0.0077, 0.0091, 0.0080, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:29:20,371 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7095, 1.6504, 1.4081, 1.6734, 2.0832, 1.9220, 1.6286, 1.5271], + device='cuda:1'), covar=tensor([0.0298, 0.0336, 0.0598, 0.0339, 0.0198, 0.0442, 0.0317, 0.0404], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0107, 0.0144, 0.0113, 0.0101, 0.0109, 0.0099, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4685e-05, 8.2998e-05, 1.1389e-04, 8.6798e-05, 7.8305e-05, 8.0577e-05, + 7.3806e-05, 8.3992e-05], device='cuda:1') +2023-03-26 21:29:24,375 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1861, 1.9997, 1.8243, 1.7808, 1.8962, 1.9334, 1.9804, 2.5811], + device='cuda:1'), covar=tensor([0.3692, 0.4188, 0.3185, 0.3695, 0.3523, 0.2459, 0.3446, 0.1673], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0228, 0.0276, 0.0252, 0.0220, 0.0252, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:29:25,507 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=98046.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:29:28,301 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.027e+02 1.543e+02 1.878e+02 2.128e+02 3.672e+02, threshold=3.757e+02, percent-clipped=1.0 +2023-03-26 21:29:28,985 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=98051.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:29:41,544 INFO [finetune.py:976] (1/7) Epoch 18, batch 700, loss[loss=0.1776, simple_loss=0.257, pruned_loss=0.04913, over 4804.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2525, pruned_loss=0.05621, over 930293.75 frames. ], batch size: 39, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:29:45,870 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=98078.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:30:03,505 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.72 vs. limit=5.0 +2023-03-26 21:30:05,698 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98106.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:30:15,236 INFO [finetune.py:976] (1/7) Epoch 18, batch 750, loss[loss=0.1637, simple_loss=0.2406, pruned_loss=0.04338, over 4815.00 frames. ], tot_loss[loss=0.1829, simple_loss=0.2534, pruned_loss=0.05619, over 937435.44 frames. ], batch size: 33, lr: 3.36e-03, grad_scale: 64.0 +2023-03-26 21:30:30,968 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.65 vs. limit=5.0 +2023-03-26 21:30:34,724 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.151e+02 1.501e+02 1.785e+02 2.309e+02 4.193e+02, threshold=3.569e+02, percent-clipped=2.0 +2023-03-26 21:30:46,066 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98167.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:30:48,362 INFO [finetune.py:976] (1/7) Epoch 18, batch 800, loss[loss=0.1668, simple_loss=0.2396, pruned_loss=0.04702, over 4787.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2524, pruned_loss=0.05538, over 940573.48 frames. ], batch size: 29, lr: 3.36e-03, grad_scale: 32.0 +2023-03-26 21:30:50,406 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 21:30:56,328 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0343, 1.9476, 1.8483, 1.9990, 1.5269, 4.6193, 1.8477, 2.2491], + device='cuda:1'), covar=tensor([0.3183, 0.2345, 0.1961, 0.2227, 0.1597, 0.0126, 0.2235, 0.1161], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0123, 0.0113, 0.0096, 0.0096, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:31:15,627 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98211.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:31:22,189 INFO [finetune.py:976] (1/7) Epoch 18, batch 850, loss[loss=0.172, simple_loss=0.2388, pruned_loss=0.05265, over 4933.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.25, pruned_loss=0.05468, over 944279.85 frames. ], batch size: 33, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:31:45,826 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98244.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:31:55,530 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.155e+02 1.511e+02 1.794e+02 2.111e+02 3.360e+02, threshold=3.589e+02, percent-clipped=0.0 +2023-03-26 21:32:06,592 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=98259.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:32:15,533 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8336, 1.6896, 1.4937, 1.8591, 2.1625, 1.8980, 1.4524, 1.5181], + device='cuda:1'), covar=tensor([0.1978, 0.1922, 0.1801, 0.1468, 0.1610, 0.1190, 0.2362, 0.1775], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0209, 0.0212, 0.0191, 0.0242, 0.0187, 0.0215, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:32:25,141 INFO [finetune.py:976] (1/7) Epoch 18, batch 900, loss[loss=0.152, simple_loss=0.2225, pruned_loss=0.0408, over 4834.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2481, pruned_loss=0.05455, over 945258.28 frames. ], batch size: 39, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:33:02,857 INFO [finetune.py:976] (1/7) Epoch 18, batch 950, loss[loss=0.2202, simple_loss=0.2684, pruned_loss=0.08602, over 4824.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2471, pruned_loss=0.0547, over 947806.17 frames. ], batch size: 30, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:33:08,457 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98330.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 21:33:23,082 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.033e+02 1.511e+02 1.760e+02 2.160e+02 3.441e+02, threshold=3.521e+02, percent-clipped=0.0 +2023-03-26 21:33:37,315 INFO [finetune.py:976] (1/7) Epoch 18, batch 1000, loss[loss=0.177, simple_loss=0.2649, pruned_loss=0.04457, over 4817.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2479, pruned_loss=0.05495, over 948829.86 frames. ], batch size: 51, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:33:42,166 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=98378.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:34:02,791 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98410.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:34:06,240 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2922, 1.3440, 1.4311, 1.5059, 1.4937, 2.9432, 1.2940, 1.4583], + device='cuda:1'), covar=tensor([0.1011, 0.1937, 0.1199, 0.1034, 0.1696, 0.0258, 0.1626, 0.1917], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0077, 0.0091, 0.0080, 0.0084, 0.0078], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:34:14,798 INFO [finetune.py:976] (1/7) Epoch 18, batch 1050, loss[loss=0.2139, simple_loss=0.2926, pruned_loss=0.06763, over 4899.00 frames. ], tot_loss[loss=0.1815, simple_loss=0.2517, pruned_loss=0.05563, over 949934.97 frames. ], batch size: 43, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:34:26,453 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9350, 1.8330, 1.6419, 1.4113, 1.9534, 1.6701, 1.8516, 1.9366], + device='cuda:1'), covar=tensor([0.1445, 0.1890, 0.3046, 0.2619, 0.2711, 0.1848, 0.2846, 0.1803], + device='cuda:1'), in_proj_covar=tensor([0.0184, 0.0187, 0.0235, 0.0254, 0.0245, 0.0203, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:34:36,628 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.123e+02 1.558e+02 1.856e+02 2.287e+02 5.753e+02, threshold=3.713e+02, percent-clipped=4.0 +2023-03-26 21:34:44,673 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98462.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:34:51,512 INFO [finetune.py:976] (1/7) Epoch 18, batch 1100, loss[loss=0.2466, simple_loss=0.2957, pruned_loss=0.09877, over 4919.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2527, pruned_loss=0.05592, over 952745.69 frames. ], batch size: 38, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:34:51,647 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98471.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:35:08,483 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.8467, 0.8551, 0.8489, 1.0387, 1.0509, 1.0055, 0.8958, 0.8741], + device='cuda:1'), covar=tensor([0.0446, 0.0290, 0.0583, 0.0251, 0.0267, 0.0386, 0.0295, 0.0342], + device='cuda:1'), in_proj_covar=tensor([0.0097, 0.0108, 0.0144, 0.0113, 0.0101, 0.0109, 0.0099, 0.0111], + device='cuda:1'), out_proj_covar=tensor([7.5148e-05, 8.3451e-05, 1.1378e-04, 8.6666e-05, 7.8786e-05, 8.0849e-05, + 7.4099e-05, 8.4446e-05], device='cuda:1') +2023-03-26 21:35:14,494 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8629, 1.3033, 1.8585, 1.8761, 1.6491, 1.5646, 1.7741, 1.7033], + device='cuda:1'), covar=tensor([0.3989, 0.4028, 0.3289, 0.3380, 0.4764, 0.3836, 0.4190, 0.3135], + device='cuda:1'), in_proj_covar=tensor([0.0248, 0.0240, 0.0259, 0.0274, 0.0273, 0.0248, 0.0283, 0.0241], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:35:24,231 INFO [finetune.py:976] (1/7) Epoch 18, batch 1150, loss[loss=0.1713, simple_loss=0.2457, pruned_loss=0.04851, over 4736.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.2549, pruned_loss=0.05665, over 954421.73 frames. ], batch size: 27, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:35:39,053 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98543.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:35:39,636 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98544.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:35:43,755 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.197e+02 1.665e+02 1.912e+02 2.323e+02 4.830e+02, threshold=3.825e+02, percent-clipped=3.0 +2023-03-26 21:35:57,261 INFO [finetune.py:976] (1/7) Epoch 18, batch 1200, loss[loss=0.1708, simple_loss=0.2388, pruned_loss=0.05143, over 4844.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2536, pruned_loss=0.0566, over 955663.56 frames. ], batch size: 49, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:35:59,143 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2720, 1.8477, 2.1207, 2.1521, 1.9042, 1.9169, 2.0590, 1.9971], + device='cuda:1'), covar=tensor([0.4353, 0.4224, 0.3366, 0.4280, 0.5081, 0.4206, 0.4865, 0.3305], + device='cuda:1'), in_proj_covar=tensor([0.0249, 0.0241, 0.0259, 0.0275, 0.0274, 0.0248, 0.0284, 0.0241], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:36:12,051 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=98592.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:36:19,568 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98604.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:36:31,240 INFO [finetune.py:976] (1/7) Epoch 18, batch 1250, loss[loss=0.1797, simple_loss=0.2482, pruned_loss=0.05558, over 4898.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.251, pruned_loss=0.0557, over 956994.49 frames. ], batch size: 35, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:37:01,600 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.074e+02 1.465e+02 1.816e+02 2.333e+02 4.053e+02, threshold=3.632e+02, percent-clipped=1.0 +2023-03-26 21:37:29,823 INFO [finetune.py:976] (1/7) Epoch 18, batch 1300, loss[loss=0.1803, simple_loss=0.2421, pruned_loss=0.05923, over 4222.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2474, pruned_loss=0.05473, over 957310.00 frames. ], batch size: 66, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:37:54,841 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.4464, 2.9941, 2.6393, 1.4171, 2.9405, 2.4188, 2.3796, 2.7685], + device='cuda:1'), covar=tensor([0.0832, 0.0906, 0.1865, 0.2284, 0.1703, 0.2178, 0.2145, 0.1047], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0196, 0.0201, 0.0183, 0.0214, 0.0208, 0.0223, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:38:07,461 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5088, 2.6122, 2.4217, 1.9684, 2.6475, 2.7821, 2.8555, 2.3776], + device='cuda:1'), covar=tensor([0.0645, 0.0605, 0.0785, 0.0855, 0.0565, 0.0683, 0.0609, 0.0980], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0135, 0.0141, 0.0122, 0.0125, 0.0140, 0.0141, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:38:11,397 INFO [finetune.py:976] (1/7) Epoch 18, batch 1350, loss[loss=0.2453, simple_loss=0.307, pruned_loss=0.09181, over 4758.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.247, pruned_loss=0.05485, over 955481.91 frames. ], batch size: 54, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:38:16,262 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7098, 1.5999, 1.5921, 1.5972, 1.0863, 3.0184, 1.1366, 1.6041], + device='cuda:1'), covar=tensor([0.3332, 0.2497, 0.2141, 0.2383, 0.1863, 0.0259, 0.2640, 0.1300], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0121, 0.0122, 0.0113, 0.0096, 0.0096, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:38:31,462 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.112e+02 1.676e+02 1.899e+02 2.271e+02 3.691e+02, threshold=3.798e+02, percent-clipped=1.0 +2023-03-26 21:38:38,222 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98762.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:38:40,596 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98766.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:38:40,623 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3119, 1.3769, 1.8403, 1.6494, 1.4986, 3.2268, 1.2881, 1.5001], + device='cuda:1'), covar=tensor([0.1023, 0.1743, 0.1295, 0.0979, 0.1554, 0.0219, 0.1494, 0.1725], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0077, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:38:44,610 INFO [finetune.py:976] (1/7) Epoch 18, batch 1400, loss[loss=0.2241, simple_loss=0.2944, pruned_loss=0.07683, over 4901.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.25, pruned_loss=0.05615, over 955071.87 frames. ], batch size: 43, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:38:46,496 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-26 21:39:04,082 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9417, 1.6376, 2.2262, 1.6367, 2.0871, 2.1525, 1.6290, 2.2738], + device='cuda:1'), covar=tensor([0.1111, 0.1813, 0.1358, 0.1861, 0.0726, 0.1268, 0.2390, 0.0792], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0203, 0.0192, 0.0191, 0.0176, 0.0213, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:39:04,840 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4071, 1.4371, 1.2181, 1.4721, 1.7860, 1.6045, 1.4226, 1.2788], + device='cuda:1'), covar=tensor([0.0333, 0.0273, 0.0592, 0.0267, 0.0186, 0.0530, 0.0387, 0.0387], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0107, 0.0143, 0.0111, 0.0100, 0.0108, 0.0098, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4502e-05, 8.2511e-05, 1.1313e-04, 8.5763e-05, 7.7652e-05, 8.0067e-05, + 7.3422e-05, 8.3809e-05], device='cuda:1') +2023-03-26 21:39:10,802 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=98810.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:39:12,640 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5694, 1.6706, 1.4001, 1.6672, 1.9878, 1.8879, 1.6937, 1.5094], + device='cuda:1'), covar=tensor([0.0364, 0.0318, 0.0597, 0.0300, 0.0207, 0.0424, 0.0285, 0.0365], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0107, 0.0144, 0.0112, 0.0100, 0.0108, 0.0098, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4592e-05, 8.2613e-05, 1.1332e-04, 8.5820e-05, 7.7717e-05, 8.0128e-05, + 7.3544e-05, 8.3886e-05], device='cuda:1') +2023-03-26 21:39:17,933 INFO [finetune.py:976] (1/7) Epoch 18, batch 1450, loss[loss=0.185, simple_loss=0.2509, pruned_loss=0.05954, over 4763.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2519, pruned_loss=0.05643, over 955771.91 frames. ], batch size: 28, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:39:19,756 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6374, 2.4481, 1.8889, 2.5467, 2.3768, 2.1454, 2.8885, 2.5880], + device='cuda:1'), covar=tensor([0.1189, 0.2209, 0.3213, 0.2652, 0.2763, 0.1742, 0.3666, 0.1866], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0187, 0.0234, 0.0254, 0.0245, 0.0202, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:39:40,442 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.160e+02 1.618e+02 1.947e+02 2.343e+02 3.750e+02, threshold=3.895e+02, percent-clipped=0.0 +2023-03-26 21:39:52,508 INFO [finetune.py:976] (1/7) Epoch 18, batch 1500, loss[loss=0.1765, simple_loss=0.2541, pruned_loss=0.04946, over 4807.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2529, pruned_loss=0.05666, over 955984.41 frames. ], batch size: 40, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:39:53,291 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7393, 1.6412, 1.4858, 1.7852, 2.2488, 1.7867, 1.5778, 1.4723], + device='cuda:1'), covar=tensor([0.1885, 0.1915, 0.1687, 0.1421, 0.1535, 0.1202, 0.2244, 0.1745], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0209, 0.0212, 0.0191, 0.0241, 0.0186, 0.0215, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:39:59,595 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3310, 1.3951, 1.5168, 0.8386, 1.4685, 1.7515, 1.7365, 1.3263], + device='cuda:1'), covar=tensor([0.1101, 0.0727, 0.0539, 0.0602, 0.0543, 0.0595, 0.0392, 0.0704], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0150, 0.0123, 0.0126, 0.0130, 0.0128, 0.0142, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.0693e-05, 1.0893e-04, 8.8221e-05, 8.9495e-05, 9.1716e-05, 9.1744e-05, + 1.0202e-04, 1.0564e-04], device='cuda:1') +2023-03-26 21:40:06,463 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98890.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:40:10,033 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98895.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:40:12,871 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98899.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:40:26,173 INFO [finetune.py:976] (1/7) Epoch 18, batch 1550, loss[loss=0.2227, simple_loss=0.2962, pruned_loss=0.07456, over 4882.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.253, pruned_loss=0.05659, over 956208.88 frames. ], batch size: 35, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:40:47,930 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.111e+02 1.516e+02 1.782e+02 2.221e+02 4.511e+02, threshold=3.564e+02, percent-clipped=1.0 +2023-03-26 21:40:48,050 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98951.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:40:51,135 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98956.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:41:00,138 INFO [finetune.py:976] (1/7) Epoch 18, batch 1600, loss[loss=0.1968, simple_loss=0.2573, pruned_loss=0.06817, over 4817.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.251, pruned_loss=0.05626, over 956924.09 frames. ], batch size: 41, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:41:00,250 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98971.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:41:33,949 INFO [finetune.py:976] (1/7) Epoch 18, batch 1650, loss[loss=0.1562, simple_loss=0.2314, pruned_loss=0.04053, over 4815.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2491, pruned_loss=0.05556, over 956103.76 frames. ], batch size: 41, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:41:41,745 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99032.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:42:02,528 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.819e+01 1.633e+02 2.015e+02 2.343e+02 3.841e+02, threshold=4.030e+02, percent-clipped=3.0 +2023-03-26 21:42:02,644 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99051.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:42:17,586 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99066.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:42:25,674 INFO [finetune.py:976] (1/7) Epoch 18, batch 1700, loss[loss=0.1778, simple_loss=0.2437, pruned_loss=0.05598, over 4794.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2472, pruned_loss=0.05496, over 956736.93 frames. ], batch size: 29, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:42:54,402 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4961, 1.4194, 1.3694, 1.4075, 1.2727, 3.4746, 1.4078, 1.8254], + device='cuda:1'), covar=tensor([0.4315, 0.3192, 0.2499, 0.3037, 0.1694, 0.0270, 0.2865, 0.1310], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0122, 0.0113, 0.0095, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:43:01,002 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99112.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:43:02,158 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=99114.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:43:10,652 INFO [finetune.py:976] (1/7) Epoch 18, batch 1750, loss[loss=0.2451, simple_loss=0.311, pruned_loss=0.08959, over 4827.00 frames. ], tot_loss[loss=0.181, simple_loss=0.2499, pruned_loss=0.05611, over 956570.39 frames. ], batch size: 39, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:43:31,987 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9896, 1.0982, 2.0101, 1.9117, 1.7612, 1.6862, 1.7813, 1.9712], + device='cuda:1'), covar=tensor([0.3373, 0.3455, 0.3157, 0.3320, 0.4406, 0.3665, 0.3905, 0.2883], + device='cuda:1'), in_proj_covar=tensor([0.0248, 0.0240, 0.0259, 0.0275, 0.0274, 0.0248, 0.0284, 0.0241], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:43:38,844 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.106e+01 1.643e+02 1.917e+02 2.422e+02 4.876e+02, threshold=3.835e+02, percent-clipped=2.0 +2023-03-26 21:43:51,913 INFO [finetune.py:976] (1/7) Epoch 18, batch 1800, loss[loss=0.1839, simple_loss=0.2522, pruned_loss=0.05784, over 4752.00 frames. ], tot_loss[loss=0.1833, simple_loss=0.2528, pruned_loss=0.05696, over 955126.99 frames. ], batch size: 54, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:44:10,871 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99199.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:44:25,712 INFO [finetune.py:976] (1/7) Epoch 18, batch 1850, loss[loss=0.1899, simple_loss=0.2619, pruned_loss=0.05897, over 4809.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2549, pruned_loss=0.05836, over 955346.07 frames. ], batch size: 39, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:44:30,695 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99229.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:44:42,407 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99246.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:44:43,016 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=99247.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:44:45,838 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.071e+02 1.618e+02 1.939e+02 2.302e+02 3.831e+02, threshold=3.878e+02, percent-clipped=0.0 +2023-03-26 21:44:45,921 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99251.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:44:57,649 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99268.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:44:59,348 INFO [finetune.py:976] (1/7) Epoch 18, batch 1900, loss[loss=0.1728, simple_loss=0.2475, pruned_loss=0.04906, over 4814.00 frames. ], tot_loss[loss=0.1856, simple_loss=0.2553, pruned_loss=0.058, over 955788.74 frames. ], batch size: 45, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:45:11,542 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99290.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:45:24,441 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.45 vs. limit=5.0 +2023-03-26 21:45:29,059 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0218, 1.8219, 2.2888, 1.6861, 2.2417, 2.3654, 1.7499, 2.4694], + device='cuda:1'), covar=tensor([0.1243, 0.1944, 0.1443, 0.1725, 0.0879, 0.1337, 0.2599, 0.0771], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0204, 0.0193, 0.0191, 0.0176, 0.0214, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:45:33,104 INFO [finetune.py:976] (1/7) Epoch 18, batch 1950, loss[loss=0.2025, simple_loss=0.2766, pruned_loss=0.0642, over 4905.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2519, pruned_loss=0.05607, over 954220.59 frames. ], batch size: 36, lr: 3.35e-03, grad_scale: 32.0 +2023-03-26 21:45:35,059 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99324.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:45:36,825 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99327.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:45:38,095 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6297, 1.5846, 1.5309, 1.5641, 1.1530, 3.4099, 1.4190, 1.8845], + device='cuda:1'), covar=tensor([0.3405, 0.2493, 0.2193, 0.2391, 0.1761, 0.0238, 0.2673, 0.1243], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0123, 0.0114, 0.0096, 0.0096, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:45:38,121 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99329.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:45:52,709 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.452e+02 1.660e+02 1.987e+02 4.820e+02, threshold=3.320e+02, percent-clipped=1.0 +2023-03-26 21:46:06,274 INFO [finetune.py:976] (1/7) Epoch 18, batch 2000, loss[loss=0.1779, simple_loss=0.2486, pruned_loss=0.05362, over 4891.00 frames. ], tot_loss[loss=0.1814, simple_loss=0.2503, pruned_loss=0.05623, over 955763.47 frames. ], batch size: 35, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:46:10,528 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1052, 1.4196, 1.0192, 1.9853, 2.3563, 1.9083, 1.7194, 1.9195], + device='cuda:1'), covar=tensor([0.1244, 0.1907, 0.1862, 0.1029, 0.1715, 0.1713, 0.1288, 0.1778], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0096, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:46:15,410 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99385.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:46:26,148 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4513, 1.4229, 1.8942, 1.7054, 1.4993, 3.3922, 1.3007, 1.5547], + device='cuda:1'), covar=tensor([0.0942, 0.1825, 0.1117, 0.0964, 0.1600, 0.0220, 0.1559, 0.1709], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 21:46:29,042 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8000, 1.1965, 0.8555, 1.4818, 2.1511, 1.0637, 1.4374, 1.5051], + device='cuda:1'), covar=tensor([0.1464, 0.2137, 0.1876, 0.1221, 0.1848, 0.1956, 0.1538, 0.1933], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0096, 0.0111, 0.0093, 0.0120, 0.0094, 0.0099, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 21:46:30,194 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99407.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:46:39,588 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99420.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:46:40,078 INFO [finetune.py:976] (1/7) Epoch 18, batch 2050, loss[loss=0.1757, simple_loss=0.2453, pruned_loss=0.05301, over 4908.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2471, pruned_loss=0.05483, over 956293.17 frames. ], batch size: 35, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:46:59,827 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.109e+02 1.434e+02 1.742e+02 2.145e+02 5.049e+02, threshold=3.484e+02, percent-clipped=5.0 +2023-03-26 21:46:59,912 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5387, 2.9171, 2.8240, 1.3225, 3.0385, 2.3183, 0.8877, 1.9985], + device='cuda:1'), covar=tensor([0.1990, 0.2530, 0.1706, 0.3571, 0.1388, 0.1085, 0.4016, 0.1674], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0178, 0.0160, 0.0130, 0.0161, 0.0124, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 21:47:17,710 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99468.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:47:19,439 INFO [finetune.py:976] (1/7) Epoch 18, batch 2100, loss[loss=0.1847, simple_loss=0.245, pruned_loss=0.06225, over 4871.00 frames. ], tot_loss[loss=0.1787, simple_loss=0.2473, pruned_loss=0.05505, over 955153.11 frames. ], batch size: 34, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:47:31,508 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99481.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:47:58,125 INFO [finetune.py:976] (1/7) Epoch 18, batch 2150, loss[loss=0.1762, simple_loss=0.2553, pruned_loss=0.04856, over 4822.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2515, pruned_loss=0.05654, over 955529.30 frames. ], batch size: 51, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:48:08,418 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99529.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 21:48:28,283 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99546.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:48:35,643 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.054e+02 1.546e+02 1.925e+02 2.366e+02 3.688e+02, threshold=3.850e+02, percent-clipped=2.0 +2023-03-26 21:48:35,732 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99551.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:48:51,480 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99568.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:48:53,717 INFO [finetune.py:976] (1/7) Epoch 18, batch 2200, loss[loss=0.1484, simple_loss=0.2265, pruned_loss=0.03521, over 4806.00 frames. ], tot_loss[loss=0.1858, simple_loss=0.2551, pruned_loss=0.05824, over 954867.34 frames. ], batch size: 25, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:49:03,262 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99585.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:49:05,756 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99589.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:49:06,389 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0134, 1.9638, 1.6141, 1.9289, 1.8303, 1.7819, 1.8920, 2.5840], + device='cuda:1'), covar=tensor([0.4193, 0.4981, 0.3671, 0.4720, 0.4754, 0.2851, 0.4526, 0.1935], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0261, 0.0228, 0.0275, 0.0251, 0.0219, 0.0251, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:49:08,744 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=99594.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:49:08,770 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.0198, 4.3300, 4.5143, 4.8683, 4.7820, 4.5377, 5.1503, 1.4850], + device='cuda:1'), covar=tensor([0.0741, 0.0820, 0.0743, 0.0839, 0.1162, 0.1491, 0.0508, 0.6134], + device='cuda:1'), in_proj_covar=tensor([0.0345, 0.0242, 0.0276, 0.0290, 0.0331, 0.0278, 0.0298, 0.0292], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:49:11,757 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=99599.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:49:27,187 INFO [finetune.py:976] (1/7) Epoch 18, batch 2250, loss[loss=0.2408, simple_loss=0.3042, pruned_loss=0.08866, over 4895.00 frames. ], tot_loss[loss=0.1863, simple_loss=0.2561, pruned_loss=0.05825, over 955472.21 frames. ], batch size: 36, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:49:29,590 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99624.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:49:31,397 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99627.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:49:33,061 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99629.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:49:46,352 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99650.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:49:46,818 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.914e+01 1.516e+02 1.824e+02 2.092e+02 3.162e+02, threshold=3.647e+02, percent-clipped=0.0 +2023-03-26 21:49:50,498 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2862, 2.1853, 1.6715, 2.1006, 2.1952, 1.9275, 2.5473, 2.2962], + device='cuda:1'), covar=tensor([0.1329, 0.2215, 0.3321, 0.2936, 0.2680, 0.1762, 0.3236, 0.1819], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0188, 0.0234, 0.0253, 0.0244, 0.0202, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:50:00,832 INFO [finetune.py:976] (1/7) Epoch 18, batch 2300, loss[loss=0.1554, simple_loss=0.2361, pruned_loss=0.03731, over 4775.00 frames. ], tot_loss[loss=0.1845, simple_loss=0.2545, pruned_loss=0.05723, over 953735.05 frames. ], batch size: 26, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:50:03,800 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=99675.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:50:06,798 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99680.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:50:06,829 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99680.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:50:24,112 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99707.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:50:34,080 INFO [finetune.py:976] (1/7) Epoch 18, batch 2350, loss[loss=0.19, simple_loss=0.2537, pruned_loss=0.06309, over 4928.00 frames. ], tot_loss[loss=0.1821, simple_loss=0.2514, pruned_loss=0.05639, over 951219.57 frames. ], batch size: 33, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:50:47,879 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99741.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:50:53,524 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-26 21:50:53,857 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8640, 1.2987, 0.8501, 1.7595, 2.1270, 1.5047, 1.6691, 1.7536], + device='cuda:1'), covar=tensor([0.1403, 0.2020, 0.2053, 0.1110, 0.1960, 0.1993, 0.1313, 0.1892], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0096, 0.0110, 0.0092, 0.0120, 0.0094, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:50:54,364 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.551e+02 1.845e+02 2.144e+02 4.060e+02, threshold=3.690e+02, percent-clipped=1.0 +2023-03-26 21:50:56,922 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=99755.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:51:08,061 INFO [finetune.py:976] (1/7) Epoch 18, batch 2400, loss[loss=0.1454, simple_loss=0.221, pruned_loss=0.03484, over 4755.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.2466, pruned_loss=0.05435, over 952945.45 frames. ], batch size: 28, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:51:11,641 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99776.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:51:24,887 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 21:51:35,068 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2240, 1.3588, 1.5076, 1.0513, 1.2565, 1.4896, 1.3341, 1.5940], + device='cuda:1'), covar=tensor([0.1279, 0.2021, 0.1216, 0.1487, 0.0880, 0.1168, 0.2952, 0.0795], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0203, 0.0191, 0.0188, 0.0175, 0.0211, 0.0216, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:51:37,906 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4192, 2.3968, 2.4226, 1.8840, 2.2838, 2.6029, 2.6291, 2.0968], + device='cuda:1'), covar=tensor([0.0508, 0.0533, 0.0658, 0.0793, 0.1050, 0.0536, 0.0503, 0.0985], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0134, 0.0140, 0.0121, 0.0123, 0.0138, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:51:41,396 INFO [finetune.py:976] (1/7) Epoch 18, batch 2450, loss[loss=0.1857, simple_loss=0.2576, pruned_loss=0.05691, over 4909.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2432, pruned_loss=0.05285, over 954247.41 frames. ], batch size: 36, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:51:42,721 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8869, 1.2744, 0.8657, 1.7313, 2.1031, 1.3245, 1.5515, 1.7295], + device='cuda:1'), covar=tensor([0.1406, 0.2065, 0.2043, 0.1125, 0.1880, 0.2036, 0.1432, 0.1874], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0110, 0.0092, 0.0119, 0.0093, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 21:51:43,688 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99824.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 21:52:01,733 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.091e+01 1.409e+02 1.705e+02 2.080e+02 4.896e+02, threshold=3.409e+02, percent-clipped=2.0 +2023-03-26 21:52:14,325 INFO [finetune.py:976] (1/7) Epoch 18, batch 2500, loss[loss=0.192, simple_loss=0.2667, pruned_loss=0.0586, over 4817.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2452, pruned_loss=0.05376, over 951596.63 frames. ], batch size: 40, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:52:26,295 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99885.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:52:32,698 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0165, 2.7432, 2.8303, 2.7896, 2.6536, 2.6693, 3.1502, 1.0331], + device='cuda:1'), covar=tensor([0.2086, 0.2264, 0.2006, 0.2298, 0.3147, 0.3307, 0.2138, 0.7923], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0243, 0.0277, 0.0291, 0.0332, 0.0281, 0.0298, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:52:50,111 INFO [finetune.py:976] (1/7) Epoch 18, batch 2550, loss[loss=0.178, simple_loss=0.2584, pruned_loss=0.04883, over 4913.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2498, pruned_loss=0.0552, over 952117.03 frames. ], batch size: 35, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:52:52,503 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99924.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:52:52,539 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99924.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:52:58,895 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=99933.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:53:06,650 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99945.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:53:10,619 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.138e+02 1.647e+02 1.933e+02 2.438e+02 4.501e+02, threshold=3.867e+02, percent-clipped=7.0 +2023-03-26 21:53:29,758 INFO [finetune.py:976] (1/7) Epoch 18, batch 2600, loss[loss=0.1896, simple_loss=0.2588, pruned_loss=0.06024, over 4811.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2512, pruned_loss=0.05517, over 952515.28 frames. ], batch size: 51, lr: 3.34e-03, grad_scale: 32.0 +2023-03-26 21:53:30,433 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=99972.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:53:40,396 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99980.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:54:12,886 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100007.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 21:54:24,615 INFO [finetune.py:976] (1/7) Epoch 18, batch 2650, loss[loss=0.1244, simple_loss=0.197, pruned_loss=0.02583, over 4756.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2514, pruned_loss=0.05502, over 951314.10 frames. ], batch size: 23, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 21:54:29,386 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=100028.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:54:35,197 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100036.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:54:46,209 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.050e+02 1.530e+02 1.912e+02 2.295e+02 4.144e+02, threshold=3.823e+02, percent-clipped=1.0 +2023-03-26 21:54:56,617 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100068.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 21:54:58,304 INFO [finetune.py:976] (1/7) Epoch 18, batch 2700, loss[loss=0.1751, simple_loss=0.2491, pruned_loss=0.05056, over 4890.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2508, pruned_loss=0.05451, over 952663.13 frames. ], batch size: 32, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 21:55:01,427 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100076.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:55:31,862 INFO [finetune.py:976] (1/7) Epoch 18, batch 2750, loss[loss=0.1761, simple_loss=0.2345, pruned_loss=0.05888, over 4870.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.2478, pruned_loss=0.05394, over 953197.64 frames. ], batch size: 31, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 21:55:33,700 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=100124.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:55:33,729 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100124.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 21:55:52,997 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.121e+02 1.510e+02 1.766e+02 2.096e+02 4.575e+02, threshold=3.532e+02, percent-clipped=1.0 +2023-03-26 21:55:56,155 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2340, 1.9828, 1.7560, 1.9136, 1.8616, 1.8670, 1.9052, 2.6532], + device='cuda:1'), covar=tensor([0.3860, 0.4449, 0.3459, 0.3922, 0.4178, 0.2601, 0.3978, 0.1837], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0259, 0.0226, 0.0272, 0.0249, 0.0218, 0.0249, 0.0230], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:56:05,347 INFO [finetune.py:976] (1/7) Epoch 18, batch 2800, loss[loss=0.1669, simple_loss=0.2297, pruned_loss=0.05209, over 4049.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2447, pruned_loss=0.05305, over 954062.80 frames. ], batch size: 17, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 21:56:06,013 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=100172.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:56:11,402 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6473, 2.4583, 2.1020, 1.0108, 2.2545, 2.0137, 1.9266, 2.2205], + device='cuda:1'), covar=tensor([0.0852, 0.0845, 0.1733, 0.2160, 0.1568, 0.2228, 0.2120, 0.1016], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0194, 0.0199, 0.0183, 0.0212, 0.0208, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:56:35,085 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-26 21:56:38,931 INFO [finetune.py:976] (1/7) Epoch 18, batch 2850, loss[loss=0.1568, simple_loss=0.2327, pruned_loss=0.04039, over 4760.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2438, pruned_loss=0.05312, over 953265.68 frames. ], batch size: 27, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 21:56:40,886 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100224.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:56:49,829 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100238.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:56:54,587 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100245.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:56:59,195 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.104e+02 1.635e+02 1.899e+02 2.309e+02 4.393e+02, threshold=3.799e+02, percent-clipped=4.0 +2023-03-26 21:57:09,521 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-26 21:57:11,715 INFO [finetune.py:976] (1/7) Epoch 18, batch 2900, loss[loss=0.158, simple_loss=0.2381, pruned_loss=0.03891, over 4819.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2484, pruned_loss=0.05485, over 953414.41 frames. ], batch size: 45, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 21:57:12,862 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=100272.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:57:26,156 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=100293.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:57:30,865 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100299.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:57:37,407 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2531, 2.0075, 2.1297, 1.5099, 2.0743, 2.2054, 2.1562, 1.6957], + device='cuda:1'), covar=tensor([0.0500, 0.0653, 0.0693, 0.0910, 0.0654, 0.0656, 0.0634, 0.1106], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0134, 0.0141, 0.0121, 0.0123, 0.0138, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:57:45,540 INFO [finetune.py:976] (1/7) Epoch 18, batch 2950, loss[loss=0.1653, simple_loss=0.228, pruned_loss=0.05127, over 4301.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2513, pruned_loss=0.05503, over 954283.55 frames. ], batch size: 65, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 21:57:55,265 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100336.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:58:06,321 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.110e+02 1.490e+02 1.822e+02 2.174e+02 4.072e+02, threshold=3.643e+02, percent-clipped=2.0 +2023-03-26 21:58:13,578 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100363.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 21:58:14,891 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.47 vs. limit=5.0 +2023-03-26 21:58:18,818 INFO [finetune.py:976] (1/7) Epoch 18, batch 3000, loss[loss=0.177, simple_loss=0.2687, pruned_loss=0.04268, over 4848.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2515, pruned_loss=0.05497, over 954220.18 frames. ], batch size: 44, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 21:58:18,818 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 21:58:24,301 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8182, 1.0620, 1.9121, 1.7856, 1.6974, 1.5867, 1.6425, 1.7317], + device='cuda:1'), covar=tensor([0.3957, 0.4153, 0.3313, 0.3549, 0.4767, 0.3783, 0.4559, 0.3080], + device='cuda:1'), in_proj_covar=tensor([0.0248, 0.0238, 0.0259, 0.0274, 0.0273, 0.0247, 0.0282, 0.0240], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 21:58:31,198 INFO [finetune.py:1010] (1/7) Epoch 18, validation: loss=0.1568, simple_loss=0.2261, pruned_loss=0.04375, over 2265189.00 frames. +2023-03-26 21:58:31,198 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 21:58:44,386 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=100384.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:59:06,992 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100402.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 21:59:29,743 INFO [finetune.py:976] (1/7) Epoch 18, batch 3050, loss[loss=0.1729, simple_loss=0.2542, pruned_loss=0.04576, over 4898.00 frames. ], tot_loss[loss=0.1803, simple_loss=0.2514, pruned_loss=0.05463, over 954722.18 frames. ], batch size: 43, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 21:59:30,577 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-03-26 21:59:53,773 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.049e+02 1.572e+02 1.917e+02 2.276e+02 3.597e+02, threshold=3.833e+02, percent-clipped=0.0 +2023-03-26 22:00:01,161 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100463.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 22:00:07,403 INFO [finetune.py:976] (1/7) Epoch 18, batch 3100, loss[loss=0.1665, simple_loss=0.2322, pruned_loss=0.05038, over 4754.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2496, pruned_loss=0.05414, over 956139.54 frames. ], batch size: 28, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 22:00:36,107 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8007, 1.6242, 1.5181, 1.9013, 2.0505, 1.9133, 1.3243, 1.5056], + device='cuda:1'), covar=tensor([0.2140, 0.2079, 0.1962, 0.1642, 0.1595, 0.1159, 0.2537, 0.1960], + device='cuda:1'), in_proj_covar=tensor([0.0241, 0.0207, 0.0211, 0.0191, 0.0240, 0.0185, 0.0214, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:00:40,542 INFO [finetune.py:976] (1/7) Epoch 18, batch 3150, loss[loss=0.1675, simple_loss=0.2461, pruned_loss=0.0445, over 4795.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2475, pruned_loss=0.0538, over 956238.90 frames. ], batch size: 29, lr: 3.34e-03, grad_scale: 16.0 +2023-03-26 22:00:52,364 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 22:01:00,943 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.466e+01 1.523e+02 1.835e+02 2.195e+02 4.344e+02, threshold=3.670e+02, percent-clipped=3.0 +2023-03-26 22:01:01,896 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.89 vs. limit=5.0 +2023-03-26 22:01:12,892 INFO [finetune.py:976] (1/7) Epoch 18, batch 3200, loss[loss=0.1793, simple_loss=0.2451, pruned_loss=0.05677, over 4769.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2463, pruned_loss=0.05406, over 957667.53 frames. ], batch size: 28, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:01:28,941 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100594.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:01:46,315 INFO [finetune.py:976] (1/7) Epoch 18, batch 3250, loss[loss=0.2276, simple_loss=0.2852, pruned_loss=0.08499, over 4866.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2469, pruned_loss=0.05476, over 956094.68 frames. ], batch size: 44, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:02:08,114 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.084e+02 1.502e+02 1.862e+02 2.235e+02 4.464e+02, threshold=3.723e+02, percent-clipped=3.0 +2023-03-26 22:02:14,934 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100663.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:02:20,158 INFO [finetune.py:976] (1/7) Epoch 18, batch 3300, loss[loss=0.2072, simple_loss=0.2831, pruned_loss=0.06562, over 4815.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2509, pruned_loss=0.05578, over 956201.14 frames. ], batch size: 38, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:02:23,937 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-26 22:02:36,887 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1523, 2.0369, 1.6447, 2.0955, 2.0644, 1.7767, 2.3688, 2.1608], + device='cuda:1'), covar=tensor([0.1351, 0.2222, 0.3097, 0.2784, 0.2695, 0.1782, 0.3303, 0.1820], + device='cuda:1'), in_proj_covar=tensor([0.0183, 0.0187, 0.0233, 0.0252, 0.0244, 0.0202, 0.0213, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:02:47,024 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=100711.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:02:52,990 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 22:02:53,405 INFO [finetune.py:976] (1/7) Epoch 18, batch 3350, loss[loss=0.1452, simple_loss=0.2208, pruned_loss=0.0348, over 4742.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2518, pruned_loss=0.05572, over 955839.70 frames. ], batch size: 27, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:02:59,531 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100730.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 22:03:14,085 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.901e+01 1.577e+02 1.865e+02 2.249e+02 4.268e+02, threshold=3.731e+02, percent-clipped=3.0 +2023-03-26 22:03:18,276 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100758.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 22:03:22,796 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 22:03:26,649 INFO [finetune.py:976] (1/7) Epoch 18, batch 3400, loss[loss=0.143, simple_loss=0.2254, pruned_loss=0.03033, over 4752.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2526, pruned_loss=0.05611, over 953821.68 frames. ], batch size: 27, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:03:40,302 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100791.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 22:04:07,224 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.60 vs. limit=2.0 +2023-03-26 22:04:13,287 INFO [finetune.py:976] (1/7) Epoch 18, batch 3450, loss[loss=0.1937, simple_loss=0.2568, pruned_loss=0.0653, over 4891.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2527, pruned_loss=0.05557, over 954251.17 frames. ], batch size: 35, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:04:14,630 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1784, 2.8414, 2.5394, 1.4675, 2.8100, 2.3160, 2.2388, 2.5786], + device='cuda:1'), covar=tensor([0.0860, 0.0760, 0.1892, 0.2112, 0.1556, 0.2059, 0.2090, 0.1119], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0193, 0.0200, 0.0183, 0.0212, 0.0207, 0.0223, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:04:49,164 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.055e+02 1.551e+02 1.762e+02 2.136e+02 3.810e+02, threshold=3.524e+02, percent-clipped=1.0 +2023-03-26 22:05:04,969 INFO [finetune.py:976] (1/7) Epoch 18, batch 3500, loss[loss=0.166, simple_loss=0.2329, pruned_loss=0.04958, over 4824.00 frames. ], tot_loss[loss=0.1803, simple_loss=0.2503, pruned_loss=0.05513, over 956177.74 frames. ], batch size: 39, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:05:07,067 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-26 22:05:21,048 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100894.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:05:38,763 INFO [finetune.py:976] (1/7) Epoch 18, batch 3550, loss[loss=0.1757, simple_loss=0.2391, pruned_loss=0.05613, over 4902.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2484, pruned_loss=0.05488, over 956294.77 frames. ], batch size: 35, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:05:52,400 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=100942.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:05:54,850 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-26 22:05:59,322 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.557e+01 1.553e+02 1.835e+02 2.348e+02 4.609e+02, threshold=3.670e+02, percent-clipped=4.0 +2023-03-26 22:06:01,248 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1355, 2.0695, 2.1506, 1.4290, 2.1433, 2.3012, 2.2064, 1.7651], + device='cuda:1'), covar=tensor([0.0590, 0.0597, 0.0671, 0.0900, 0.0643, 0.0602, 0.0577, 0.1050], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0135, 0.0140, 0.0120, 0.0123, 0.0138, 0.0139, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:06:12,132 INFO [finetune.py:976] (1/7) Epoch 18, batch 3600, loss[loss=0.1431, simple_loss=0.2125, pruned_loss=0.03685, over 4811.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2458, pruned_loss=0.05422, over 956858.14 frames. ], batch size: 25, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:06:13,602 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.96 vs. limit=5.0 +2023-03-26 22:06:46,075 INFO [finetune.py:976] (1/7) Epoch 18, batch 3650, loss[loss=0.1962, simple_loss=0.2715, pruned_loss=0.0604, over 4845.00 frames. ], tot_loss[loss=0.179, simple_loss=0.2477, pruned_loss=0.0551, over 957129.33 frames. ], batch size: 49, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:07:06,790 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.559e+02 1.860e+02 2.177e+02 4.070e+02, threshold=3.719e+02, percent-clipped=1.0 +2023-03-26 22:07:06,941 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6694, 1.5313, 1.3532, 1.7221, 1.9968, 1.7441, 1.3163, 1.4104], + device='cuda:1'), covar=tensor([0.2107, 0.1919, 0.1826, 0.1508, 0.1528, 0.1180, 0.2437, 0.1872], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0210, 0.0214, 0.0194, 0.0242, 0.0187, 0.0217, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:07:10,529 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=101058.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:07:17,330 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.79 vs. limit=5.0 +2023-03-26 22:07:18,906 INFO [finetune.py:976] (1/7) Epoch 18, batch 3700, loss[loss=0.1829, simple_loss=0.2711, pruned_loss=0.0473, over 4817.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2517, pruned_loss=0.05674, over 955649.16 frames. ], batch size: 45, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:07:28,500 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=101086.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:07:31,620 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-26 22:07:37,432 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9357, 4.1465, 4.0061, 2.3371, 4.3778, 3.3744, 1.3476, 3.0745], + device='cuda:1'), covar=tensor([0.2110, 0.2367, 0.1427, 0.3180, 0.0989, 0.0858, 0.4029, 0.1462], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0177, 0.0161, 0.0130, 0.0160, 0.0125, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 22:07:42,148 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=101106.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:07:43,275 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101107.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:07:51,543 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6460, 1.6708, 1.4522, 1.6721, 2.0791, 1.8505, 1.6339, 1.4601], + device='cuda:1'), covar=tensor([0.0284, 0.0303, 0.0537, 0.0312, 0.0179, 0.0513, 0.0345, 0.0406], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0107, 0.0143, 0.0110, 0.0100, 0.0108, 0.0098, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4201e-05, 8.2454e-05, 1.1282e-04, 8.4549e-05, 7.8267e-05, 8.0103e-05, + 7.3415e-05, 8.4123e-05], device='cuda:1') +2023-03-26 22:07:52,639 INFO [finetune.py:976] (1/7) Epoch 18, batch 3750, loss[loss=0.1286, simple_loss=0.2032, pruned_loss=0.02703, over 4744.00 frames. ], tot_loss[loss=0.1853, simple_loss=0.2546, pruned_loss=0.05798, over 958428.71 frames. ], batch size: 23, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:08:12,837 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.641e+02 1.814e+02 2.131e+02 4.110e+02, threshold=3.627e+02, percent-clipped=2.0 +2023-03-26 22:08:24,433 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=101168.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:08:26,163 INFO [finetune.py:976] (1/7) Epoch 18, batch 3800, loss[loss=0.1607, simple_loss=0.2341, pruned_loss=0.04367, over 4752.00 frames. ], tot_loss[loss=0.1867, simple_loss=0.2562, pruned_loss=0.05854, over 957448.13 frames. ], batch size: 27, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:08:59,889 INFO [finetune.py:976] (1/7) Epoch 18, batch 3850, loss[loss=0.1782, simple_loss=0.2445, pruned_loss=0.05595, over 4812.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.254, pruned_loss=0.05713, over 956321.59 frames. ], batch size: 41, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:09:30,993 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.034e+02 1.525e+02 1.862e+02 2.180e+02 4.556e+02, threshold=3.724e+02, percent-clipped=2.0 +2023-03-26 22:09:31,148 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2037, 2.0976, 1.7631, 2.0781, 1.9715, 1.9458, 2.0170, 2.7585], + device='cuda:1'), covar=tensor([0.3425, 0.3946, 0.3120, 0.3672, 0.4114, 0.2288, 0.3586, 0.1611], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0229, 0.0274, 0.0252, 0.0220, 0.0251, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:09:57,716 INFO [finetune.py:976] (1/7) Epoch 18, batch 3900, loss[loss=0.1556, simple_loss=0.2281, pruned_loss=0.04159, over 4835.00 frames. ], tot_loss[loss=0.1822, simple_loss=0.2513, pruned_loss=0.05654, over 956888.29 frames. ], batch size: 25, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:10:41,970 INFO [finetune.py:976] (1/7) Epoch 18, batch 3950, loss[loss=0.1296, simple_loss=0.1932, pruned_loss=0.03294, over 4742.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2472, pruned_loss=0.05459, over 957414.89 frames. ], batch size: 23, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:11:02,251 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.024e+02 1.470e+02 1.754e+02 2.083e+02 3.090e+02, threshold=3.508e+02, percent-clipped=0.0 +2023-03-26 22:11:08,465 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.84 vs. limit=5.0 +2023-03-26 22:11:15,363 INFO [finetune.py:976] (1/7) Epoch 18, batch 4000, loss[loss=0.1823, simple_loss=0.2585, pruned_loss=0.05306, over 4900.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2471, pruned_loss=0.0547, over 957423.97 frames. ], batch size: 35, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:11:26,010 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=101386.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:11:49,438 INFO [finetune.py:976] (1/7) Epoch 18, batch 4050, loss[loss=0.1676, simple_loss=0.2533, pruned_loss=0.04095, over 4916.00 frames. ], tot_loss[loss=0.1821, simple_loss=0.2516, pruned_loss=0.05625, over 958747.82 frames. ], batch size: 38, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:11:57,753 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.20 vs. limit=5.0 +2023-03-26 22:11:58,803 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=101434.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:12:10,024 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.148e+02 1.670e+02 2.040e+02 2.363e+02 9.256e+02, threshold=4.080e+02, percent-clipped=2.0 +2023-03-26 22:12:17,267 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=101463.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:12:22,996 INFO [finetune.py:976] (1/7) Epoch 18, batch 4100, loss[loss=0.1898, simple_loss=0.26, pruned_loss=0.05976, over 4902.00 frames. ], tot_loss[loss=0.1846, simple_loss=0.2547, pruned_loss=0.05725, over 958521.12 frames. ], batch size: 36, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:12:53,501 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.53 vs. limit=5.0 +2023-03-26 22:12:56,254 INFO [finetune.py:976] (1/7) Epoch 18, batch 4150, loss[loss=0.1926, simple_loss=0.2552, pruned_loss=0.065, over 4844.00 frames. ], tot_loss[loss=0.1843, simple_loss=0.2543, pruned_loss=0.05719, over 957765.05 frames. ], batch size: 30, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:13:16,877 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.507e+01 1.508e+02 1.851e+02 2.208e+02 3.984e+02, threshold=3.702e+02, percent-clipped=0.0 +2023-03-26 22:13:29,471 INFO [finetune.py:976] (1/7) Epoch 18, batch 4200, loss[loss=0.1705, simple_loss=0.2378, pruned_loss=0.05156, over 4817.00 frames. ], tot_loss[loss=0.1837, simple_loss=0.2534, pruned_loss=0.05702, over 954826.76 frames. ], batch size: 33, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:13:30,271 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.82 vs. limit=5.0 +2023-03-26 22:13:49,575 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 22:14:03,051 INFO [finetune.py:976] (1/7) Epoch 18, batch 4250, loss[loss=0.155, simple_loss=0.2387, pruned_loss=0.03572, over 4833.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2494, pruned_loss=0.05486, over 955266.57 frames. ], batch size: 49, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:14:20,728 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0782, 1.0044, 1.0325, 1.1494, 1.2222, 1.1788, 1.0517, 0.9891], + device='cuda:1'), covar=tensor([0.0361, 0.0289, 0.0596, 0.0321, 0.0286, 0.0428, 0.0325, 0.0430], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0106, 0.0141, 0.0109, 0.0099, 0.0107, 0.0097, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.3066e-05, 8.1547e-05, 1.1144e-04, 8.3799e-05, 7.7115e-05, 7.9102e-05, + 7.2576e-05, 8.2733e-05], device='cuda:1') +2023-03-26 22:14:24,253 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.150e+02 1.580e+02 1.774e+02 2.219e+02 3.425e+02, threshold=3.547e+02, percent-clipped=0.0 +2023-03-26 22:14:38,479 INFO [finetune.py:976] (1/7) Epoch 18, batch 4300, loss[loss=0.1591, simple_loss=0.2248, pruned_loss=0.04674, over 4825.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2476, pruned_loss=0.05482, over 955341.84 frames. ], batch size: 30, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:14:54,319 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101684.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:15:33,151 INFO [finetune.py:976] (1/7) Epoch 18, batch 4350, loss[loss=0.1921, simple_loss=0.2538, pruned_loss=0.06513, over 4825.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2445, pruned_loss=0.0532, over 956526.14 frames. ], batch size: 30, lr: 3.33e-03, grad_scale: 16.0 +2023-03-26 22:16:05,135 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=101745.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:16:08,064 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6898, 2.4435, 2.1521, 0.9699, 2.3483, 2.0283, 1.9461, 2.2719], + device='cuda:1'), covar=tensor([0.0864, 0.0848, 0.1493, 0.2179, 0.1420, 0.2472, 0.2134, 0.0996], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0192, 0.0199, 0.0183, 0.0212, 0.0207, 0.0222, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:16:10,238 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.272e+01 1.483e+02 1.686e+02 2.087e+02 3.591e+02, threshold=3.373e+02, percent-clipped=1.0 +2023-03-26 22:16:16,942 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=101763.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:16:21,249 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3957, 2.3298, 1.8769, 2.4723, 2.2876, 2.0469, 2.7729, 2.4654], + device='cuda:1'), covar=tensor([0.1281, 0.2223, 0.2989, 0.2754, 0.2509, 0.1670, 0.3267, 0.1726], + device='cuda:1'), in_proj_covar=tensor([0.0184, 0.0188, 0.0235, 0.0253, 0.0245, 0.0202, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:16:21,723 INFO [finetune.py:976] (1/7) Epoch 18, batch 4400, loss[loss=0.1536, simple_loss=0.2243, pruned_loss=0.04143, over 4761.00 frames. ], tot_loss[loss=0.177, simple_loss=0.246, pruned_loss=0.05403, over 954092.80 frames. ], batch size: 54, lr: 3.32e-03, grad_scale: 16.0 +2023-03-26 22:16:49,621 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=101811.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:16:55,630 INFO [finetune.py:976] (1/7) Epoch 18, batch 4450, loss[loss=0.1603, simple_loss=0.2359, pruned_loss=0.04237, over 4758.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.25, pruned_loss=0.05487, over 954835.16 frames. ], batch size: 59, lr: 3.32e-03, grad_scale: 16.0 +2023-03-26 22:17:16,756 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.107e+02 1.606e+02 1.893e+02 2.313e+02 4.401e+02, threshold=3.785e+02, percent-clipped=7.0 +2023-03-26 22:17:19,717 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8292, 1.7352, 1.5322, 1.9589, 2.2833, 1.9492, 1.5390, 1.4666], + device='cuda:1'), covar=tensor([0.2271, 0.2071, 0.1951, 0.1650, 0.1693, 0.1220, 0.2410, 0.2077], + device='cuda:1'), in_proj_covar=tensor([0.0243, 0.0210, 0.0214, 0.0193, 0.0242, 0.0187, 0.0216, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:17:29,390 INFO [finetune.py:976] (1/7) Epoch 18, batch 4500, loss[loss=0.2075, simple_loss=0.264, pruned_loss=0.07548, over 4751.00 frames. ], tot_loss[loss=0.1828, simple_loss=0.253, pruned_loss=0.05623, over 956548.59 frames. ], batch size: 27, lr: 3.32e-03, grad_scale: 16.0 +2023-03-26 22:17:43,733 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6612, 2.4728, 2.0150, 1.0560, 2.1634, 1.9621, 1.8992, 2.2484], + device='cuda:1'), covar=tensor([0.0882, 0.0898, 0.1830, 0.2112, 0.1609, 0.2268, 0.2176, 0.0931], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0194, 0.0201, 0.0184, 0.0214, 0.0209, 0.0225, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:18:02,648 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101920.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:18:03,131 INFO [finetune.py:976] (1/7) Epoch 18, batch 4550, loss[loss=0.206, simple_loss=0.2717, pruned_loss=0.07021, over 4872.00 frames. ], tot_loss[loss=0.1836, simple_loss=0.2541, pruned_loss=0.05655, over 956223.19 frames. ], batch size: 34, lr: 3.32e-03, grad_scale: 16.0 +2023-03-26 22:18:13,340 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.70 vs. limit=5.0 +2023-03-26 22:18:23,150 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0182, 3.4580, 3.7065, 3.8561, 3.8107, 3.5070, 4.0967, 1.4012], + device='cuda:1'), covar=tensor([0.0808, 0.0930, 0.0802, 0.1043, 0.1121, 0.1526, 0.0751, 0.5264], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0242, 0.0276, 0.0290, 0.0331, 0.0281, 0.0301, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:18:24,137 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.522e+02 1.794e+02 2.336e+02 4.256e+02, threshold=3.587e+02, percent-clipped=2.0 +2023-03-26 22:18:36,747 INFO [finetune.py:976] (1/7) Epoch 18, batch 4600, loss[loss=0.1707, simple_loss=0.2452, pruned_loss=0.04811, over 4897.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2539, pruned_loss=0.05652, over 956307.72 frames. ], batch size: 35, lr: 3.32e-03, grad_scale: 16.0 +2023-03-26 22:18:42,882 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=101981.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 22:18:47,483 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101987.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:19:11,217 INFO [finetune.py:976] (1/7) Epoch 18, batch 4650, loss[loss=0.1888, simple_loss=0.249, pruned_loss=0.06425, over 4929.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2513, pruned_loss=0.0559, over 956642.81 frames. ], batch size: 33, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:19:23,878 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=102040.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:19:29,225 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=102048.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:19:31,550 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.060e+02 1.564e+02 1.867e+02 2.217e+02 4.281e+02, threshold=3.734e+02, percent-clipped=4.0 +2023-03-26 22:19:45,055 INFO [finetune.py:976] (1/7) Epoch 18, batch 4700, loss[loss=0.1785, simple_loss=0.243, pruned_loss=0.05697, over 4825.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2476, pruned_loss=0.05465, over 957185.86 frames. ], batch size: 39, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:19:46,966 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3391, 2.1390, 1.7564, 2.2300, 2.2843, 2.0206, 2.5432, 2.2921], + device='cuda:1'), covar=tensor([0.1292, 0.2095, 0.3064, 0.2528, 0.2381, 0.1636, 0.3083, 0.1797], + device='cuda:1'), in_proj_covar=tensor([0.0184, 0.0189, 0.0235, 0.0254, 0.0246, 0.0203, 0.0216, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:19:59,987 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1619, 2.0665, 1.7889, 1.9676, 1.9687, 1.9947, 2.0183, 2.7701], + device='cuda:1'), covar=tensor([0.3666, 0.4501, 0.3166, 0.3978, 0.3954, 0.2346, 0.3764, 0.1542], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0261, 0.0227, 0.0273, 0.0250, 0.0219, 0.0250, 0.0231], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:20:13,425 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-26 22:20:31,466 INFO [finetune.py:976] (1/7) Epoch 18, batch 4750, loss[loss=0.1828, simple_loss=0.2544, pruned_loss=0.05565, over 4908.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2451, pruned_loss=0.05403, over 955619.09 frames. ], batch size: 43, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:20:43,079 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4982, 2.3156, 1.9417, 0.9206, 2.1646, 1.9000, 1.8214, 2.0623], + device='cuda:1'), covar=tensor([0.0747, 0.0818, 0.1351, 0.1903, 0.1209, 0.2110, 0.1916, 0.0866], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0194, 0.0201, 0.0184, 0.0214, 0.0209, 0.0224, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:20:56,172 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.538e+02 1.904e+02 2.326e+02 4.380e+02, threshold=3.807e+02, percent-clipped=2.0 +2023-03-26 22:20:56,311 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6469, 1.7394, 1.7447, 0.9962, 1.8278, 2.1147, 2.0298, 1.5751], + device='cuda:1'), covar=tensor([0.0888, 0.0631, 0.0425, 0.0556, 0.0404, 0.0527, 0.0305, 0.0605], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0152, 0.0124, 0.0127, 0.0132, 0.0130, 0.0142, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.1250e-05, 1.0986e-04, 8.8416e-05, 9.0141e-05, 9.2817e-05, 9.2899e-05, + 1.0207e-04, 1.0666e-04], device='cuda:1') +2023-03-26 22:21:23,316 INFO [finetune.py:976] (1/7) Epoch 18, batch 4800, loss[loss=0.2166, simple_loss=0.2827, pruned_loss=0.07526, over 4773.00 frames. ], tot_loss[loss=0.1793, simple_loss=0.2482, pruned_loss=0.05517, over 954373.01 frames. ], batch size: 28, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:21:30,761 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0482, 1.5407, 2.0447, 1.9455, 1.8049, 1.7578, 1.9797, 1.8451], + device='cuda:1'), covar=tensor([0.3677, 0.4222, 0.3378, 0.3810, 0.5100, 0.3934, 0.4759, 0.3149], + device='cuda:1'), in_proj_covar=tensor([0.0249, 0.0240, 0.0258, 0.0275, 0.0275, 0.0248, 0.0283, 0.0240], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:21:51,142 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6568, 1.4841, 1.0706, 0.2705, 1.1923, 1.4287, 1.4032, 1.3872], + device='cuda:1'), covar=tensor([0.0990, 0.0932, 0.1450, 0.2035, 0.1504, 0.2507, 0.2333, 0.0867], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0194, 0.0201, 0.0184, 0.0214, 0.0209, 0.0224, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:22:00,411 INFO [finetune.py:976] (1/7) Epoch 18, batch 4850, loss[loss=0.193, simple_loss=0.2702, pruned_loss=0.05792, over 4916.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2519, pruned_loss=0.0559, over 953769.97 frames. ], batch size: 38, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:22:05,228 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1976, 2.8881, 2.7638, 1.2409, 3.0033, 2.2140, 0.6854, 1.8613], + device='cuda:1'), covar=tensor([0.2568, 0.2582, 0.1897, 0.3664, 0.1496, 0.1177, 0.4368, 0.1741], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0161, 0.0130, 0.0162, 0.0125, 0.0149, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 22:22:12,739 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-26 22:22:20,215 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.897e+01 1.478e+02 1.803e+02 2.234e+02 3.533e+02, threshold=3.606e+02, percent-clipped=0.0 +2023-03-26 22:22:33,608 INFO [finetune.py:976] (1/7) Epoch 18, batch 4900, loss[loss=0.1882, simple_loss=0.2454, pruned_loss=0.06544, over 4694.00 frames. ], tot_loss[loss=0.1844, simple_loss=0.2546, pruned_loss=0.05713, over 954364.14 frames. ], batch size: 23, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:22:37,663 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=102276.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 22:23:06,928 INFO [finetune.py:976] (1/7) Epoch 18, batch 4950, loss[loss=0.1464, simple_loss=0.2316, pruned_loss=0.03056, over 4776.00 frames. ], tot_loss[loss=0.1827, simple_loss=0.2534, pruned_loss=0.05603, over 953828.67 frames. ], batch size: 29, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:23:19,549 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=102340.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:23:21,352 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=102343.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:23:26,706 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.970e+01 1.449e+02 1.847e+02 2.188e+02 4.191e+02, threshold=3.694e+02, percent-clipped=1.0 +2023-03-26 22:23:40,084 INFO [finetune.py:976] (1/7) Epoch 18, batch 5000, loss[loss=0.1927, simple_loss=0.2537, pruned_loss=0.06584, over 4887.00 frames. ], tot_loss[loss=0.1817, simple_loss=0.2519, pruned_loss=0.05569, over 954903.60 frames. ], batch size: 35, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:23:51,807 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=102388.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:24:13,295 INFO [finetune.py:976] (1/7) Epoch 18, batch 5050, loss[loss=0.1599, simple_loss=0.232, pruned_loss=0.04388, over 4748.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2485, pruned_loss=0.0544, over 955497.23 frames. ], batch size: 27, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:24:33,970 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.223e+01 1.582e+02 1.966e+02 2.354e+02 3.513e+02, threshold=3.932e+02, percent-clipped=0.0 +2023-03-26 22:24:42,569 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.68 vs. limit=5.0 +2023-03-26 22:24:43,102 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0878, 1.4611, 2.0904, 2.0152, 1.8904, 1.8519, 1.9189, 1.9377], + device='cuda:1'), covar=tensor([0.3757, 0.4227, 0.3452, 0.3691, 0.4659, 0.3752, 0.4651, 0.3070], + device='cuda:1'), in_proj_covar=tensor([0.0248, 0.0240, 0.0258, 0.0275, 0.0274, 0.0247, 0.0283, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:24:46,889 INFO [finetune.py:976] (1/7) Epoch 18, batch 5100, loss[loss=0.1278, simple_loss=0.2003, pruned_loss=0.0277, over 4789.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2451, pruned_loss=0.05313, over 956222.90 frames. ], batch size: 28, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:24:52,361 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3215, 1.2826, 1.1557, 1.2803, 1.6405, 1.4879, 1.3935, 1.1659], + device='cuda:1'), covar=tensor([0.0363, 0.0343, 0.0634, 0.0359, 0.0224, 0.0441, 0.0321, 0.0467], + device='cuda:1'), in_proj_covar=tensor([0.0095, 0.0106, 0.0142, 0.0110, 0.0099, 0.0108, 0.0098, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.3529e-05, 8.2128e-05, 1.1222e-04, 8.4281e-05, 7.7087e-05, 7.9413e-05, + 7.3413e-05, 8.3363e-05], device='cuda:1') +2023-03-26 22:24:54,717 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 22:25:20,567 INFO [finetune.py:976] (1/7) Epoch 18, batch 5150, loss[loss=0.1501, simple_loss=0.2276, pruned_loss=0.03628, over 4772.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2464, pruned_loss=0.05395, over 956157.20 frames. ], batch size: 28, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:25:46,748 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6611, 1.7621, 1.8673, 1.1096, 1.8384, 2.0638, 2.0652, 1.6291], + device='cuda:1'), covar=tensor([0.0851, 0.0629, 0.0493, 0.0553, 0.0566, 0.0606, 0.0351, 0.0610], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0151, 0.0124, 0.0126, 0.0131, 0.0129, 0.0142, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.0957e-05, 1.0937e-04, 8.8577e-05, 8.9567e-05, 9.2235e-05, 9.2423e-05, + 1.0179e-04, 1.0627e-04], device='cuda:1') +2023-03-26 22:25:53,892 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.631e+02 1.989e+02 2.441e+02 4.766e+02, threshold=3.977e+02, percent-clipped=3.0 +2023-03-26 22:26:14,502 INFO [finetune.py:976] (1/7) Epoch 18, batch 5200, loss[loss=0.2075, simple_loss=0.2778, pruned_loss=0.06858, over 4204.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2515, pruned_loss=0.05605, over 956344.51 frames. ], batch size: 65, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:26:22,142 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=102576.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 22:26:40,352 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 22:26:44,025 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2791, 1.4464, 0.8784, 2.0374, 2.6003, 1.8689, 1.7798, 1.8840], + device='cuda:1'), covar=tensor([0.1344, 0.2059, 0.1912, 0.1093, 0.1596, 0.1820, 0.1384, 0.1948], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0119, 0.0093, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 22:26:56,624 INFO [finetune.py:976] (1/7) Epoch 18, batch 5250, loss[loss=0.2012, simple_loss=0.275, pruned_loss=0.06364, over 4215.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2529, pruned_loss=0.05604, over 954756.63 frames. ], batch size: 65, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:26:58,547 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=102624.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 22:27:11,554 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=102643.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:27:17,760 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.182e+02 1.670e+02 1.981e+02 2.217e+02 4.217e+02, threshold=3.962e+02, percent-clipped=1.0 +2023-03-26 22:27:29,390 INFO [finetune.py:976] (1/7) Epoch 18, batch 5300, loss[loss=0.1993, simple_loss=0.2807, pruned_loss=0.05897, over 4918.00 frames. ], tot_loss[loss=0.1831, simple_loss=0.2539, pruned_loss=0.05618, over 954750.24 frames. ], batch size: 33, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:27:44,025 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=102691.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:27:54,724 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-03-26 22:28:03,083 INFO [finetune.py:976] (1/7) Epoch 18, batch 5350, loss[loss=0.1543, simple_loss=0.2248, pruned_loss=0.04188, over 4376.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.253, pruned_loss=0.05538, over 955155.13 frames. ], batch size: 65, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:28:11,344 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5785, 2.8239, 2.5934, 1.9222, 2.7682, 2.8765, 3.0009, 2.3844], + device='cuda:1'), covar=tensor([0.0602, 0.0508, 0.0686, 0.0784, 0.0519, 0.0639, 0.0526, 0.0937], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0134, 0.0140, 0.0120, 0.0124, 0.0138, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:28:19,021 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8428, 3.4005, 3.5398, 3.7366, 3.5954, 3.3849, 3.9337, 1.1682], + device='cuda:1'), covar=tensor([0.0950, 0.0925, 0.0994, 0.1016, 0.1446, 0.1688, 0.0959, 0.5604], + device='cuda:1'), in_proj_covar=tensor([0.0355, 0.0247, 0.0281, 0.0294, 0.0336, 0.0285, 0.0306, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:28:25,303 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.195e+02 1.536e+02 1.872e+02 2.228e+02 4.473e+02, threshold=3.745e+02, percent-clipped=1.0 +2023-03-26 22:28:36,912 INFO [finetune.py:976] (1/7) Epoch 18, batch 5400, loss[loss=0.1571, simple_loss=0.2285, pruned_loss=0.04289, over 4822.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2508, pruned_loss=0.05533, over 955657.30 frames. ], batch size: 30, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:28:40,219 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.67 vs. limit=5.0 +2023-03-26 22:29:07,391 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.95 vs. limit=5.0 +2023-03-26 22:29:10,819 INFO [finetune.py:976] (1/7) Epoch 18, batch 5450, loss[loss=0.1984, simple_loss=0.2597, pruned_loss=0.0685, over 4796.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2474, pruned_loss=0.05456, over 955370.27 frames. ], batch size: 51, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:29:30,965 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=102851.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:29:31,447 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.551e+01 1.511e+02 1.793e+02 2.102e+02 5.113e+02, threshold=3.586e+02, percent-clipped=1.0 +2023-03-26 22:29:33,413 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-26 22:29:44,409 INFO [finetune.py:976] (1/7) Epoch 18, batch 5500, loss[loss=0.1148, simple_loss=0.1866, pruned_loss=0.02149, over 4772.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2441, pruned_loss=0.05312, over 955805.00 frames. ], batch size: 28, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:29:54,084 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=102887.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:30:00,542 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 22:30:02,350 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-26 22:30:02,929 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0869, 1.7815, 2.3761, 1.4929, 2.3245, 2.3678, 1.7061, 2.5197], + device='cuda:1'), covar=tensor([0.1176, 0.1970, 0.1418, 0.2097, 0.0824, 0.1381, 0.2714, 0.0819], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0205, 0.0191, 0.0191, 0.0176, 0.0215, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:30:12,696 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=102912.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:30:18,101 INFO [finetune.py:976] (1/7) Epoch 18, batch 5550, loss[loss=0.193, simple_loss=0.2708, pruned_loss=0.05763, over 4863.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.247, pruned_loss=0.05435, over 953819.62 frames. ], batch size: 31, lr: 3.32e-03, grad_scale: 32.0 +2023-03-26 22:30:36,209 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=102948.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:30:39,503 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.093e+02 1.559e+02 1.902e+02 2.213e+02 4.520e+02, threshold=3.805e+02, percent-clipped=3.0 +2023-03-26 22:30:50,053 INFO [finetune.py:976] (1/7) Epoch 18, batch 5600, loss[loss=0.181, simple_loss=0.2631, pruned_loss=0.04948, over 4907.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2514, pruned_loss=0.05519, over 954491.79 frames. ], batch size: 43, lr: 3.32e-03, grad_scale: 16.0 +2023-03-26 22:30:56,421 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 22:30:57,107 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-26 22:30:57,537 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8568, 1.7463, 2.3108, 1.4068, 2.2002, 2.1672, 1.6266, 2.4031], + device='cuda:1'), covar=tensor([0.1340, 0.2231, 0.1519, 0.2196, 0.0919, 0.1642, 0.2966, 0.0896], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0204, 0.0191, 0.0190, 0.0176, 0.0215, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:31:42,253 INFO [finetune.py:976] (1/7) Epoch 18, batch 5650, loss[loss=0.1923, simple_loss=0.2655, pruned_loss=0.05954, over 4912.00 frames. ], tot_loss[loss=0.1824, simple_loss=0.2539, pruned_loss=0.05549, over 954060.41 frames. ], batch size: 37, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:32:02,934 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-26 22:32:09,152 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.800e+01 1.506e+02 1.835e+02 2.191e+02 3.638e+02, threshold=3.670e+02, percent-clipped=0.0 +2023-03-26 22:32:09,238 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8625, 2.1786, 1.1875, 2.5387, 3.0945, 2.3740, 2.4990, 2.4540], + device='cuda:1'), covar=tensor([0.1209, 0.1678, 0.2066, 0.0992, 0.1459, 0.1580, 0.1156, 0.1793], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0120, 0.0094, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 22:32:19,851 INFO [finetune.py:976] (1/7) Epoch 18, batch 5700, loss[loss=0.185, simple_loss=0.2444, pruned_loss=0.06282, over 4264.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2498, pruned_loss=0.05518, over 934955.65 frames. ], batch size: 18, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:32:48,075 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103098.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:32:48,575 INFO [finetune.py:976] (1/7) Epoch 19, batch 0, loss[loss=0.202, simple_loss=0.2675, pruned_loss=0.06823, over 4779.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2675, pruned_loss=0.06823, over 4779.00 frames. ], batch size: 51, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:32:48,575 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 22:33:03,099 INFO [finetune.py:1010] (1/7) Epoch 19, validation: loss=0.1586, simple_loss=0.2282, pruned_loss=0.04454, over 2265189.00 frames. +2023-03-26 22:33:03,100 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 22:33:25,836 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103132.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:33:32,557 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-26 22:33:38,061 INFO [finetune.py:976] (1/7) Epoch 19, batch 50, loss[loss=0.1428, simple_loss=0.2165, pruned_loss=0.03459, over 4817.00 frames. ], tot_loss[loss=0.1837, simple_loss=0.2543, pruned_loss=0.05656, over 216983.21 frames. ], batch size: 33, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:33:40,498 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.779e+01 1.456e+02 1.782e+02 2.150e+02 3.860e+02, threshold=3.565e+02, percent-clipped=1.0 +2023-03-26 22:33:41,942 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 22:33:44,736 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103159.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:34:07,242 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103193.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:34:11,663 INFO [finetune.py:976] (1/7) Epoch 19, batch 100, loss[loss=0.1667, simple_loss=0.2262, pruned_loss=0.05365, over 4840.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2453, pruned_loss=0.05309, over 382709.52 frames. ], batch size: 47, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:34:17,531 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103207.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:34:40,704 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103243.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:34:45,882 INFO [finetune.py:976] (1/7) Epoch 19, batch 150, loss[loss=0.148, simple_loss=0.2173, pruned_loss=0.03934, over 4896.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2419, pruned_loss=0.05191, over 512269.71 frames. ], batch size: 35, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:34:48,708 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.032e+02 1.463e+02 1.787e+02 2.269e+02 3.542e+02, threshold=3.573e+02, percent-clipped=0.0 +2023-03-26 22:35:08,281 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5089, 1.4870, 1.6367, 1.6960, 1.5952, 3.2520, 1.3193, 1.5013], + device='cuda:1'), covar=tensor([0.0940, 0.1808, 0.1073, 0.0970, 0.1485, 0.0252, 0.1507, 0.1716], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 22:35:19,727 INFO [finetune.py:976] (1/7) Epoch 19, batch 200, loss[loss=0.1762, simple_loss=0.2501, pruned_loss=0.05112, over 4817.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2426, pruned_loss=0.05284, over 611591.93 frames. ], batch size: 45, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:35:22,748 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6655, 2.5797, 2.1225, 2.6915, 2.5642, 2.3941, 2.9048, 2.6749], + device='cuda:1'), covar=tensor([0.1335, 0.1799, 0.2731, 0.2099, 0.2129, 0.1580, 0.2816, 0.1573], + device='cuda:1'), in_proj_covar=tensor([0.0185, 0.0188, 0.0235, 0.0254, 0.0246, 0.0203, 0.0214, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:35:42,743 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7379, 1.4337, 0.9865, 1.6280, 2.0928, 1.3343, 1.5482, 1.7672], + device='cuda:1'), covar=tensor([0.1448, 0.1911, 0.1864, 0.1162, 0.1937, 0.1956, 0.1428, 0.1803], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 22:35:53,176 INFO [finetune.py:976] (1/7) Epoch 19, batch 250, loss[loss=0.1221, simple_loss=0.1886, pruned_loss=0.02779, over 4819.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.245, pruned_loss=0.05318, over 687416.51 frames. ], batch size: 25, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:35:56,516 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.389e+01 1.572e+02 1.886e+02 2.263e+02 4.128e+02, threshold=3.772e+02, percent-clipped=1.0 +2023-03-26 22:36:05,016 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103366.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 22:36:06,878 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3520, 1.2686, 1.4164, 1.5209, 1.4850, 2.8649, 1.2498, 1.3583], + device='cuda:1'), covar=tensor([0.1108, 0.2121, 0.1410, 0.1140, 0.1827, 0.0352, 0.1827, 0.2187], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0077, 0.0091, 0.0079, 0.0084, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 22:36:16,978 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.78 vs. limit=5.0 +2023-03-26 22:36:25,723 INFO [finetune.py:976] (1/7) Epoch 19, batch 300, loss[loss=0.1276, simple_loss=0.1962, pruned_loss=0.02948, over 4792.00 frames. ], tot_loss[loss=0.179, simple_loss=0.2491, pruned_loss=0.05449, over 746676.29 frames. ], batch size: 25, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:37:02,190 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103427.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:37:21,222 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4164, 1.2382, 1.2573, 1.3673, 1.6261, 1.5457, 1.3392, 1.1903], + device='cuda:1'), covar=tensor([0.0360, 0.0344, 0.0603, 0.0298, 0.0263, 0.0444, 0.0405, 0.0419], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0105, 0.0139, 0.0108, 0.0097, 0.0106, 0.0097, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.2818e-05, 8.1198e-05, 1.0983e-04, 8.3082e-05, 7.5469e-05, 7.8029e-05, + 7.2140e-05, 8.2289e-05], device='cuda:1') +2023-03-26 22:37:21,695 INFO [finetune.py:976] (1/7) Epoch 19, batch 350, loss[loss=0.1815, simple_loss=0.2609, pruned_loss=0.05111, over 4788.00 frames. ], tot_loss[loss=0.1815, simple_loss=0.2516, pruned_loss=0.05566, over 792905.84 frames. ], batch size: 29, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:37:28,082 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.109e+02 1.554e+02 1.898e+02 2.403e+02 5.343e+02, threshold=3.796e+02, percent-clipped=4.0 +2023-03-26 22:37:29,298 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103454.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:38:03,044 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103488.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:38:04,301 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6540, 1.7334, 1.8057, 0.9742, 1.8471, 2.1350, 2.0616, 1.5166], + device='cuda:1'), covar=tensor([0.0917, 0.0617, 0.0443, 0.0550, 0.0378, 0.0501, 0.0299, 0.0845], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0151, 0.0124, 0.0127, 0.0131, 0.0129, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.1611e-05, 1.0959e-04, 8.8976e-05, 9.0237e-05, 9.2332e-05, 9.2663e-05, + 1.0223e-04, 1.0671e-04], device='cuda:1') +2023-03-26 22:38:05,514 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5299, 1.4410, 1.3628, 1.6243, 1.8148, 1.7296, 1.5497, 1.3354], + device='cuda:1'), covar=tensor([0.0292, 0.0289, 0.0565, 0.0235, 0.0224, 0.0402, 0.0300, 0.0374], + device='cuda:1'), in_proj_covar=tensor([0.0094, 0.0105, 0.0139, 0.0108, 0.0097, 0.0106, 0.0097, 0.0108], + device='cuda:1'), out_proj_covar=tensor([7.2698e-05, 8.1159e-05, 1.0978e-04, 8.3015e-05, 7.5424e-05, 7.7888e-05, + 7.2049e-05, 8.2335e-05], device='cuda:1') +2023-03-26 22:38:10,178 INFO [finetune.py:976] (1/7) Epoch 19, batch 400, loss[loss=0.14, simple_loss=0.2135, pruned_loss=0.03321, over 4855.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2531, pruned_loss=0.05607, over 827912.66 frames. ], batch size: 31, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:38:15,634 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103507.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:38:39,532 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103543.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:38:39,563 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0487, 0.9624, 0.9891, 0.3691, 0.9267, 1.1911, 1.1759, 0.9941], + device='cuda:1'), covar=tensor([0.0914, 0.0616, 0.0591, 0.0601, 0.0552, 0.0643, 0.0412, 0.0725], + device='cuda:1'), in_proj_covar=tensor([0.0126, 0.0152, 0.0125, 0.0127, 0.0132, 0.0130, 0.0143, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.2081e-05, 1.1010e-04, 8.9257e-05, 9.0604e-05, 9.2836e-05, 9.3144e-05, + 1.0273e-04, 1.0725e-04], device='cuda:1') +2023-03-26 22:38:43,097 INFO [finetune.py:976] (1/7) Epoch 19, batch 450, loss[loss=0.1759, simple_loss=0.2437, pruned_loss=0.05407, over 4886.00 frames. ], tot_loss[loss=0.18, simple_loss=0.2509, pruned_loss=0.05458, over 858443.80 frames. ], batch size: 32, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:38:45,992 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.312e+01 1.502e+02 1.696e+02 2.061e+02 2.854e+02, threshold=3.392e+02, percent-clipped=0.0 +2023-03-26 22:38:47,237 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=103555.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:39:14,763 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0135, 1.8002, 2.4266, 1.5736, 2.3015, 2.3624, 1.7044, 2.4783], + device='cuda:1'), covar=tensor([0.1397, 0.2062, 0.1596, 0.2222, 0.0882, 0.1478, 0.2770, 0.0926], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0204, 0.0190, 0.0189, 0.0174, 0.0213, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:39:19,608 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=103591.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:39:24,372 INFO [finetune.py:976] (1/7) Epoch 19, batch 500, loss[loss=0.1398, simple_loss=0.2153, pruned_loss=0.03215, over 4747.00 frames. ], tot_loss[loss=0.179, simple_loss=0.2488, pruned_loss=0.05455, over 879926.48 frames. ], batch size: 27, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:39:25,214 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8998, 1.4313, 1.1041, 1.8651, 2.1894, 1.6694, 1.5988, 1.8199], + device='cuda:1'), covar=tensor([0.1197, 0.1678, 0.1679, 0.0918, 0.1750, 0.1836, 0.1226, 0.1528], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 22:39:37,412 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-26 22:39:56,012 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6524, 1.5347, 1.1077, 0.3055, 1.2460, 1.5058, 1.4658, 1.4250], + device='cuda:1'), covar=tensor([0.1119, 0.0876, 0.1433, 0.2048, 0.1458, 0.2551, 0.2340, 0.0930], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0192, 0.0198, 0.0181, 0.0209, 0.0205, 0.0220, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:39:57,690 INFO [finetune.py:976] (1/7) Epoch 19, batch 550, loss[loss=0.1592, simple_loss=0.236, pruned_loss=0.0412, over 4802.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.247, pruned_loss=0.05409, over 896545.39 frames. ], batch size: 51, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:40:00,581 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.093e+02 1.543e+02 1.834e+02 2.179e+02 4.966e+02, threshold=3.668e+02, percent-clipped=2.0 +2023-03-26 22:40:31,341 INFO [finetune.py:976] (1/7) Epoch 19, batch 600, loss[loss=0.152, simple_loss=0.2238, pruned_loss=0.04014, over 4790.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.248, pruned_loss=0.05438, over 908925.33 frames. ], batch size: 29, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:40:47,252 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103722.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 22:40:55,598 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8211, 3.7268, 3.5663, 1.8405, 3.8447, 2.7998, 0.6905, 2.5478], + device='cuda:1'), covar=tensor([0.2168, 0.2063, 0.1333, 0.3167, 0.0980, 0.1029, 0.4467, 0.1494], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0176, 0.0160, 0.0128, 0.0160, 0.0123, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 22:40:58,726 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103740.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:41:04,044 INFO [finetune.py:976] (1/7) Epoch 19, batch 650, loss[loss=0.1916, simple_loss=0.2644, pruned_loss=0.05937, over 4904.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.2509, pruned_loss=0.05498, over 917754.58 frames. ], batch size: 35, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:41:06,467 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.131e+02 1.539e+02 1.795e+02 2.169e+02 3.837e+02, threshold=3.591e+02, percent-clipped=1.0 +2023-03-26 22:41:07,190 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103754.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:41:15,363 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103765.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:41:27,821 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6691, 1.6910, 1.7213, 1.1062, 1.9011, 2.1560, 2.0483, 1.5357], + device='cuda:1'), covar=tensor([0.1048, 0.0728, 0.0565, 0.0574, 0.0397, 0.0498, 0.0345, 0.0913], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0150, 0.0123, 0.0125, 0.0129, 0.0127, 0.0141, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.0470e-05, 1.0846e-04, 8.8051e-05, 8.9154e-05, 9.1102e-05, 9.1261e-05, + 1.0104e-04, 1.0581e-04], device='cuda:1') +2023-03-26 22:41:30,828 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103788.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:41:37,406 INFO [finetune.py:976] (1/7) Epoch 19, batch 700, loss[loss=0.18, simple_loss=0.2465, pruned_loss=0.05673, over 4901.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.2515, pruned_loss=0.05469, over 928405.81 frames. ], batch size: 35, lr: 3.31e-03, grad_scale: 16.0 +2023-03-26 22:41:38,889 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4603, 1.5153, 1.3449, 1.5967, 1.8057, 1.7375, 1.5251, 1.3910], + device='cuda:1'), covar=tensor([0.0331, 0.0301, 0.0580, 0.0276, 0.0222, 0.0476, 0.0331, 0.0383], + device='cuda:1'), in_proj_covar=tensor([0.0095, 0.0107, 0.0142, 0.0109, 0.0098, 0.0107, 0.0098, 0.0109], + device='cuda:1'), out_proj_covar=tensor([7.3711e-05, 8.2320e-05, 1.1145e-04, 8.3979e-05, 7.6509e-05, 7.9319e-05, + 7.3234e-05, 8.3418e-05], device='cuda:1') +2023-03-26 22:41:38,895 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103801.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:41:39,435 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=103802.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:42:02,012 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103826.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:42:13,877 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=103836.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:42:25,794 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9366, 1.7444, 2.2317, 3.7687, 2.4417, 2.6565, 1.0290, 3.1520], + device='cuda:1'), covar=tensor([0.1667, 0.1448, 0.1443, 0.0566, 0.0789, 0.1966, 0.1861, 0.0434], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0117, 0.0134, 0.0165, 0.0100, 0.0137, 0.0125, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 22:42:26,321 INFO [finetune.py:976] (1/7) Epoch 19, batch 750, loss[loss=0.152, simple_loss=0.2359, pruned_loss=0.03404, over 4835.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.254, pruned_loss=0.05549, over 936497.93 frames. ], batch size: 47, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:42:33,271 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.573e+02 1.871e+02 2.192e+02 5.260e+02, threshold=3.742e+02, percent-clipped=2.0 +2023-03-26 22:42:45,432 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6461, 1.2362, 0.7567, 1.5582, 2.0325, 1.3316, 1.5040, 1.5378], + device='cuda:1'), covar=tensor([0.1539, 0.2014, 0.1971, 0.1132, 0.1874, 0.1963, 0.1353, 0.1992], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0120, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-26 22:43:03,204 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103876.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:43:15,762 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-03-26 22:43:17,414 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1426, 1.9822, 1.7943, 1.8903, 1.8233, 1.8471, 1.9285, 2.5557], + device='cuda:1'), covar=tensor([0.3357, 0.4272, 0.3133, 0.3815, 0.3939, 0.2335, 0.3685, 0.1583], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0261, 0.0230, 0.0276, 0.0252, 0.0220, 0.0252, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:43:22,082 INFO [finetune.py:976] (1/7) Epoch 19, batch 800, loss[loss=0.1877, simple_loss=0.2576, pruned_loss=0.05892, over 4793.00 frames. ], tot_loss[loss=0.1815, simple_loss=0.2533, pruned_loss=0.05487, over 939552.12 frames. ], batch size: 51, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:43:48,641 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103937.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:43:55,832 INFO [finetune.py:976] (1/7) Epoch 19, batch 850, loss[loss=0.1583, simple_loss=0.2365, pruned_loss=0.04006, over 4775.00 frames. ], tot_loss[loss=0.1793, simple_loss=0.2507, pruned_loss=0.05398, over 945050.99 frames. ], batch size: 26, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:43:58,235 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.100e+02 1.499e+02 1.798e+02 2.103e+02 3.961e+02, threshold=3.597e+02, percent-clipped=1.0 +2023-03-26 22:44:25,959 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103990.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:44:30,241 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103997.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:44:31,321 INFO [finetune.py:976] (1/7) Epoch 19, batch 900, loss[loss=0.1516, simple_loss=0.2145, pruned_loss=0.04436, over 4712.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2479, pruned_loss=0.05318, over 946546.75 frames. ], batch size: 23, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:44:46,728 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104022.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:44:53,214 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-03-26 22:45:05,988 INFO [finetune.py:976] (1/7) Epoch 19, batch 950, loss[loss=0.1637, simple_loss=0.2381, pruned_loss=0.04469, over 4685.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2456, pruned_loss=0.05283, over 949284.23 frames. ], batch size: 23, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:45:06,106 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=104049.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:45:07,328 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104051.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:45:08,392 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.542e+01 1.532e+02 1.748e+02 2.079e+02 4.067e+02, threshold=3.497e+02, percent-clipped=1.0 +2023-03-26 22:45:11,555 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104058.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:45:18,738 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=104070.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 22:45:36,942 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104096.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:45:38,695 INFO [finetune.py:976] (1/7) Epoch 19, batch 1000, loss[loss=0.183, simple_loss=0.2543, pruned_loss=0.05589, over 4870.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2474, pruned_loss=0.05371, over 952557.10 frames. ], batch size: 34, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:45:45,414 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104110.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:45:52,085 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104121.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:46:12,343 INFO [finetune.py:976] (1/7) Epoch 19, batch 1050, loss[loss=0.2052, simple_loss=0.2834, pruned_loss=0.06353, over 4910.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2507, pruned_loss=0.05438, over 954493.02 frames. ], batch size: 43, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:46:14,761 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.111e+02 1.591e+02 1.940e+02 2.273e+02 3.456e+02, threshold=3.881e+02, percent-clipped=0.0 +2023-03-26 22:46:53,935 INFO [finetune.py:976] (1/7) Epoch 19, batch 1100, loss[loss=0.1837, simple_loss=0.2589, pruned_loss=0.05419, over 4745.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2507, pruned_loss=0.05424, over 952627.87 frames. ], batch size: 59, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:46:54,086 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0104, 1.8287, 1.6424, 1.7202, 1.6849, 1.6302, 1.7855, 2.3307], + device='cuda:1'), covar=tensor([0.3116, 0.3535, 0.2771, 0.3100, 0.3512, 0.2073, 0.3454, 0.1540], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0263, 0.0232, 0.0277, 0.0253, 0.0221, 0.0253, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:47:16,902 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104232.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 22:47:39,155 INFO [finetune.py:976] (1/7) Epoch 19, batch 1150, loss[loss=0.2216, simple_loss=0.2939, pruned_loss=0.07464, over 4891.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2528, pruned_loss=0.05542, over 953155.18 frames. ], batch size: 37, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:47:47,030 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.055e+02 1.710e+02 1.992e+02 2.366e+02 4.129e+02, threshold=3.984e+02, percent-clipped=1.0 +2023-03-26 22:47:50,723 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-03-26 22:48:25,526 INFO [finetune.py:976] (1/7) Epoch 19, batch 1200, loss[loss=0.1991, simple_loss=0.2575, pruned_loss=0.07034, over 4818.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2508, pruned_loss=0.05472, over 953346.14 frames. ], batch size: 39, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:48:45,108 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-03-26 22:48:46,779 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2364, 1.7956, 2.1880, 2.1649, 1.9180, 1.9375, 2.0967, 2.0130], + device='cuda:1'), covar=tensor([0.4045, 0.4377, 0.3552, 0.4172, 0.5380, 0.4397, 0.5462, 0.3181], + device='cuda:1'), in_proj_covar=tensor([0.0252, 0.0242, 0.0261, 0.0278, 0.0276, 0.0251, 0.0286, 0.0242], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:49:05,634 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104346.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:49:06,328 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-26 22:49:07,373 INFO [finetune.py:976] (1/7) Epoch 19, batch 1250, loss[loss=0.1791, simple_loss=0.2585, pruned_loss=0.04984, over 4867.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2488, pruned_loss=0.05415, over 952823.67 frames. ], batch size: 31, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:49:10,326 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.979e+01 1.472e+02 1.754e+02 2.218e+02 4.171e+02, threshold=3.509e+02, percent-clipped=1.0 +2023-03-26 22:49:10,411 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104353.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:49:21,799 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6995, 1.6914, 1.6074, 1.7273, 1.1464, 3.3104, 1.4900, 1.8275], + device='cuda:1'), covar=tensor([0.3199, 0.2321, 0.1951, 0.2170, 0.1710, 0.0225, 0.2480, 0.1137], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0123, 0.0114, 0.0095, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 22:49:39,161 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104396.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:49:41,418 INFO [finetune.py:976] (1/7) Epoch 19, batch 1300, loss[loss=0.2076, simple_loss=0.2712, pruned_loss=0.07198, over 4841.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2452, pruned_loss=0.05299, over 954503.59 frames. ], batch size: 47, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:49:43,439 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0581, 1.9157, 1.6348, 1.8594, 2.0447, 1.7832, 2.1857, 2.0417], + device='cuda:1'), covar=tensor([0.1300, 0.1866, 0.2622, 0.2289, 0.2237, 0.1490, 0.2958, 0.1579], + device='cuda:1'), in_proj_covar=tensor([0.0185, 0.0188, 0.0235, 0.0253, 0.0244, 0.0203, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:49:45,742 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104405.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:49:56,435 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104421.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:49:57,089 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1350, 2.1670, 1.9984, 2.3652, 2.6460, 2.3520, 2.1287, 1.6745], + device='cuda:1'), covar=tensor([0.2404, 0.1882, 0.1851, 0.1505, 0.2004, 0.1170, 0.2190, 0.2053], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0211, 0.0215, 0.0194, 0.0244, 0.0188, 0.0217, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:50:10,348 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=104444.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:50:14,762 INFO [finetune.py:976] (1/7) Epoch 19, batch 1350, loss[loss=0.2533, simple_loss=0.3037, pruned_loss=0.1015, over 4902.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2452, pruned_loss=0.05359, over 953361.28 frames. ], batch size: 43, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:50:17,642 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.092e+02 1.617e+02 1.931e+02 2.310e+02 3.973e+02, threshold=3.863e+02, percent-clipped=4.0 +2023-03-26 22:50:29,050 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=104469.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:50:48,508 INFO [finetune.py:976] (1/7) Epoch 19, batch 1400, loss[loss=0.1749, simple_loss=0.224, pruned_loss=0.06293, over 4107.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.2476, pruned_loss=0.05394, over 953263.14 frames. ], batch size: 17, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:51:10,565 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104532.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 22:51:17,924 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.66 vs. limit=5.0 +2023-03-26 22:51:21,268 INFO [finetune.py:976] (1/7) Epoch 19, batch 1450, loss[loss=0.1987, simple_loss=0.2803, pruned_loss=0.05849, over 4868.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2502, pruned_loss=0.05465, over 952453.68 frames. ], batch size: 34, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:51:24,643 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.038e+02 1.650e+02 1.913e+02 2.290e+02 4.485e+02, threshold=3.826e+02, percent-clipped=3.0 +2023-03-26 22:51:32,074 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 22:51:34,725 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0392, 1.8354, 1.6145, 1.8359, 1.7360, 1.7437, 1.7862, 2.4392], + device='cuda:1'), covar=tensor([0.3579, 0.3947, 0.3153, 0.3683, 0.4231, 0.2329, 0.4029, 0.1691], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0231, 0.0276, 0.0252, 0.0222, 0.0253, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:51:42,883 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=104580.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:52:02,825 INFO [finetune.py:976] (1/7) Epoch 19, batch 1500, loss[loss=0.1938, simple_loss=0.2675, pruned_loss=0.06007, over 4852.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2515, pruned_loss=0.05519, over 953876.89 frames. ], batch size: 31, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:52:08,994 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-03-26 22:52:13,085 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8073, 1.0805, 1.8092, 1.7254, 1.5824, 1.5090, 1.6347, 1.6973], + device='cuda:1'), covar=tensor([0.2995, 0.3205, 0.2735, 0.3081, 0.4171, 0.3237, 0.3482, 0.2506], + device='cuda:1'), in_proj_covar=tensor([0.0251, 0.0242, 0.0260, 0.0278, 0.0276, 0.0250, 0.0286, 0.0242], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:52:17,818 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1245, 1.9571, 1.6019, 1.8111, 1.8380, 1.8335, 1.9083, 2.5788], + device='cuda:1'), covar=tensor([0.3657, 0.4221, 0.3336, 0.3873, 0.4018, 0.2420, 0.3873, 0.1648], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0261, 0.0230, 0.0275, 0.0251, 0.0221, 0.0252, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:52:30,349 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.04 vs. limit=5.0 +2023-03-26 22:52:33,850 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104646.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:52:35,532 INFO [finetune.py:976] (1/7) Epoch 19, batch 1550, loss[loss=0.1647, simple_loss=0.2151, pruned_loss=0.05711, over 4179.00 frames. ], tot_loss[loss=0.1818, simple_loss=0.2521, pruned_loss=0.05575, over 954894.03 frames. ], batch size: 18, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:52:40,198 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.878e+01 1.494e+02 1.864e+02 2.283e+02 3.386e+02, threshold=3.728e+02, percent-clipped=0.0 +2023-03-26 22:52:40,300 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104653.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:53:26,171 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=104694.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:53:29,188 INFO [finetune.py:976] (1/7) Epoch 19, batch 1600, loss[loss=0.1762, simple_loss=0.2392, pruned_loss=0.05656, over 4856.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2486, pruned_loss=0.05427, over 955545.65 frames. ], batch size: 44, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:53:35,244 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=104701.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:53:38,265 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104705.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:53:46,634 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1962, 1.3921, 1.4840, 0.7277, 1.4064, 1.6622, 1.6997, 1.3951], + device='cuda:1'), covar=tensor([0.0816, 0.0511, 0.0511, 0.0497, 0.0457, 0.0647, 0.0318, 0.0684], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0151, 0.0125, 0.0126, 0.0132, 0.0129, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.1612e-05, 1.0971e-04, 8.9305e-05, 8.9674e-05, 9.2851e-05, 9.2611e-05, + 1.0181e-04, 1.0675e-04], device='cuda:1') +2023-03-26 22:54:01,106 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6520, 1.4608, 1.8867, 1.9544, 1.6401, 3.5379, 1.4383, 1.6347], + device='cuda:1'), covar=tensor([0.0912, 0.1937, 0.1117, 0.0890, 0.1597, 0.0240, 0.1457, 0.1782], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0091, 0.0080, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 22:54:11,339 INFO [finetune.py:976] (1/7) Epoch 19, batch 1650, loss[loss=0.1953, simple_loss=0.2529, pruned_loss=0.06885, over 4860.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2466, pruned_loss=0.05372, over 955975.23 frames. ], batch size: 49, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:54:13,774 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.017e+02 1.513e+02 1.754e+02 2.121e+02 3.523e+02, threshold=3.508e+02, percent-clipped=0.0 +2023-03-26 22:54:13,846 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=104753.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 22:54:19,811 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8904, 1.9133, 1.6871, 1.7534, 1.3181, 4.0805, 1.6177, 1.9883], + device='cuda:1'), covar=tensor([0.2915, 0.2169, 0.1873, 0.2126, 0.1536, 0.0161, 0.2609, 0.1170], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0123, 0.0114, 0.0096, 0.0096, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 22:54:33,281 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6509, 1.8031, 1.7893, 0.9154, 1.9254, 2.1321, 2.0704, 1.6322], + device='cuda:1'), covar=tensor([0.0974, 0.0623, 0.0548, 0.0589, 0.0397, 0.0483, 0.0305, 0.0687], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0151, 0.0125, 0.0126, 0.0132, 0.0129, 0.0141, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.1495e-05, 1.0960e-04, 8.9196e-05, 8.9379e-05, 9.2834e-05, 9.2410e-05, + 1.0153e-04, 1.0662e-04], device='cuda:1') +2023-03-26 22:54:36,312 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3106, 1.4109, 1.9052, 1.6689, 1.6095, 3.3241, 1.2887, 1.5807], + device='cuda:1'), covar=tensor([0.1016, 0.1804, 0.1299, 0.0981, 0.1463, 0.0242, 0.1529, 0.1630], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0091, 0.0080, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 22:54:41,779 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7375, 1.5659, 1.3944, 1.2933, 1.5169, 1.4252, 1.4831, 2.0102], + device='cuda:1'), covar=tensor([0.3210, 0.3241, 0.2862, 0.2825, 0.3195, 0.2125, 0.3088, 0.1574], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0260, 0.0230, 0.0274, 0.0250, 0.0220, 0.0251, 0.0231], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:54:44,699 INFO [finetune.py:976] (1/7) Epoch 19, batch 1700, loss[loss=0.1767, simple_loss=0.234, pruned_loss=0.05968, over 4188.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2443, pruned_loss=0.05292, over 955312.83 frames. ], batch size: 18, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:55:17,898 INFO [finetune.py:976] (1/7) Epoch 19, batch 1750, loss[loss=0.1421, simple_loss=0.2217, pruned_loss=0.0313, over 4769.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2464, pruned_loss=0.05419, over 955230.56 frames. ], batch size: 27, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:55:20,303 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.630e+02 1.888e+02 2.368e+02 5.925e+02, threshold=3.776e+02, percent-clipped=5.0 +2023-03-26 22:55:48,648 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5019, 1.3194, 1.4256, 0.8336, 1.4505, 1.4366, 1.4639, 1.2577], + device='cuda:1'), covar=tensor([0.0592, 0.0908, 0.0746, 0.0979, 0.0895, 0.0754, 0.0679, 0.1312], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0132, 0.0137, 0.0118, 0.0123, 0.0135, 0.0137, 0.0159], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:55:51,602 INFO [finetune.py:976] (1/7) Epoch 19, batch 1800, loss[loss=0.1784, simple_loss=0.253, pruned_loss=0.05191, over 4805.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2489, pruned_loss=0.05499, over 953166.36 frames. ], batch size: 51, lr: 3.30e-03, grad_scale: 16.0 +2023-03-26 22:55:59,090 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-03-26 22:55:59,675 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.64 vs. limit=5.0 +2023-03-26 22:56:13,540 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4018, 1.5607, 1.6912, 0.8642, 1.6182, 1.8425, 1.8188, 1.4993], + device='cuda:1'), covar=tensor([0.0967, 0.0716, 0.0567, 0.0586, 0.0653, 0.0863, 0.0458, 0.0779], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0152, 0.0125, 0.0126, 0.0132, 0.0130, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.1746e-05, 1.1018e-04, 8.9406e-05, 8.9728e-05, 9.3072e-05, 9.2862e-05, + 1.0180e-04, 1.0728e-04], device='cuda:1') +2023-03-26 22:56:19,921 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 22:56:25,139 INFO [finetune.py:976] (1/7) Epoch 19, batch 1850, loss[loss=0.1915, simple_loss=0.2441, pruned_loss=0.06946, over 4174.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2493, pruned_loss=0.05492, over 950594.15 frames. ], batch size: 17, lr: 3.30e-03, grad_scale: 32.0 +2023-03-26 22:56:27,537 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.864e+01 1.560e+02 1.785e+02 2.312e+02 4.235e+02, threshold=3.569e+02, percent-clipped=1.0 +2023-03-26 22:57:00,525 INFO [finetune.py:976] (1/7) Epoch 19, batch 1900, loss[loss=0.1445, simple_loss=0.2234, pruned_loss=0.03277, over 4792.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2513, pruned_loss=0.05505, over 952327.13 frames. ], batch size: 25, lr: 3.30e-03, grad_scale: 32.0 +2023-03-26 22:57:22,448 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4858, 2.6049, 2.3151, 1.8137, 2.3790, 2.6700, 2.7145, 2.1617], + device='cuda:1'), covar=tensor([0.0640, 0.0591, 0.0791, 0.0903, 0.0813, 0.0649, 0.0626, 0.1045], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0132, 0.0137, 0.0118, 0.0123, 0.0136, 0.0137, 0.0159], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:57:27,679 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1116, 2.0398, 1.7165, 2.0048, 2.1104, 1.8740, 2.3957, 2.1972], + device='cuda:1'), covar=tensor([0.1260, 0.1939, 0.2895, 0.2373, 0.2318, 0.1559, 0.2840, 0.1578], + device='cuda:1'), in_proj_covar=tensor([0.0185, 0.0188, 0.0236, 0.0253, 0.0246, 0.0203, 0.0214, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:57:42,257 INFO [finetune.py:976] (1/7) Epoch 19, batch 1950, loss[loss=0.1468, simple_loss=0.2189, pruned_loss=0.03738, over 4765.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2503, pruned_loss=0.05424, over 954403.72 frames. ], batch size: 27, lr: 3.30e-03, grad_scale: 32.0 +2023-03-26 22:57:44,664 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.372e+01 1.430e+02 1.759e+02 2.099e+02 5.293e+02, threshold=3.517e+02, percent-clipped=3.0 +2023-03-26 22:57:51,472 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4376, 2.1517, 3.0018, 1.8578, 2.6796, 2.9353, 1.9660, 3.0032], + device='cuda:1'), covar=tensor([0.1511, 0.2191, 0.1464, 0.2344, 0.0886, 0.1451, 0.2920, 0.0912], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0203, 0.0190, 0.0187, 0.0174, 0.0212, 0.0216, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:58:31,188 INFO [finetune.py:976] (1/7) Epoch 19, batch 2000, loss[loss=0.2003, simple_loss=0.2549, pruned_loss=0.07279, over 4939.00 frames. ], tot_loss[loss=0.178, simple_loss=0.248, pruned_loss=0.05404, over 955911.42 frames. ], batch size: 33, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 22:59:17,047 INFO [finetune.py:976] (1/7) Epoch 19, batch 2050, loss[loss=0.1506, simple_loss=0.2183, pruned_loss=0.04148, over 4832.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2459, pruned_loss=0.05324, over 956187.09 frames. ], batch size: 30, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 22:59:18,999 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.5076, 3.1107, 2.7076, 1.5888, 2.9494, 2.4315, 2.3062, 2.5278], + device='cuda:1'), covar=tensor([0.0791, 0.0811, 0.2147, 0.2116, 0.1581, 0.2088, 0.2101, 0.1239], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0195, 0.0202, 0.0184, 0.0213, 0.0209, 0.0224, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 22:59:19,897 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.050e+02 1.557e+02 1.803e+02 2.317e+02 4.729e+02, threshold=3.605e+02, percent-clipped=3.0 +2023-03-26 22:59:50,002 INFO [finetune.py:976] (1/7) Epoch 19, batch 2100, loss[loss=0.1734, simple_loss=0.2539, pruned_loss=0.04643, over 4815.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2453, pruned_loss=0.0531, over 956031.29 frames. ], batch size: 38, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 22:59:53,866 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1389, 2.0249, 1.7669, 2.0735, 1.9025, 1.9229, 2.0198, 2.6475], + device='cuda:1'), covar=tensor([0.3874, 0.4391, 0.3275, 0.3791, 0.4167, 0.2455, 0.3770, 0.1721], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0231, 0.0277, 0.0254, 0.0222, 0.0253, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:00:23,801 INFO [finetune.py:976] (1/7) Epoch 19, batch 2150, loss[loss=0.2344, simple_loss=0.3201, pruned_loss=0.0744, over 4856.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.249, pruned_loss=0.05468, over 953137.96 frames. ], batch size: 44, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:00:26,648 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.009e+02 1.573e+02 1.937e+02 2.406e+02 5.182e+02, threshold=3.875e+02, percent-clipped=4.0 +2023-03-26 23:00:48,174 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9507, 1.8385, 1.3618, 1.3353, 2.2611, 2.4253, 2.0364, 1.8176], + device='cuda:1'), covar=tensor([0.0412, 0.0396, 0.0804, 0.0437, 0.0287, 0.0532, 0.0328, 0.0462], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0108, 0.0143, 0.0111, 0.0100, 0.0110, 0.0100, 0.0111], + device='cuda:1'), out_proj_covar=tensor([7.4539e-05, 8.3254e-05, 1.1255e-04, 8.5302e-05, 7.7630e-05, 8.1255e-05, + 7.4785e-05, 8.4587e-05], device='cuda:1') +2023-03-26 23:00:57,385 INFO [finetune.py:976] (1/7) Epoch 19, batch 2200, loss[loss=0.2051, simple_loss=0.2684, pruned_loss=0.07093, over 4830.00 frames. ], tot_loss[loss=0.18, simple_loss=0.2507, pruned_loss=0.0546, over 954195.11 frames. ], batch size: 30, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:01:01,216 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-03-26 23:01:26,010 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4154, 2.3193, 1.8825, 2.4475, 2.2969, 2.0094, 2.9093, 2.4397], + device='cuda:1'), covar=tensor([0.1422, 0.2458, 0.3220, 0.2820, 0.2770, 0.1829, 0.3420, 0.1850], + device='cuda:1'), in_proj_covar=tensor([0.0186, 0.0188, 0.0237, 0.0254, 0.0248, 0.0204, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:01:30,610 INFO [finetune.py:976] (1/7) Epoch 19, batch 2250, loss[loss=0.1641, simple_loss=0.2442, pruned_loss=0.04197, over 4763.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2523, pruned_loss=0.0546, over 955363.01 frames. ], batch size: 27, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:01:33,460 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.177e+02 1.629e+02 1.918e+02 2.372e+02 6.301e+02, threshold=3.835e+02, percent-clipped=3.0 +2023-03-26 23:01:42,469 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0644, 1.9016, 1.6128, 1.9066, 1.7612, 1.7966, 1.8440, 2.5763], + device='cuda:1'), covar=tensor([0.3740, 0.4509, 0.3333, 0.3943, 0.4423, 0.2422, 0.3851, 0.1690], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0231, 0.0277, 0.0253, 0.0222, 0.0253, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:01:56,199 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=105388.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:02:03,219 INFO [finetune.py:976] (1/7) Epoch 19, batch 2300, loss[loss=0.177, simple_loss=0.2366, pruned_loss=0.05867, over 4220.00 frames. ], tot_loss[loss=0.1803, simple_loss=0.2525, pruned_loss=0.05404, over 954348.80 frames. ], batch size: 18, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:02:19,814 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8654, 2.1290, 1.6531, 1.7579, 2.3924, 2.4035, 2.0502, 1.9756], + device='cuda:1'), covar=tensor([0.0423, 0.0321, 0.0597, 0.0358, 0.0234, 0.0626, 0.0366, 0.0390], + device='cuda:1'), in_proj_covar=tensor([0.0095, 0.0107, 0.0141, 0.0110, 0.0098, 0.0109, 0.0099, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.3621e-05, 8.2217e-05, 1.1110e-04, 8.4207e-05, 7.6640e-05, 8.0434e-05, + 7.4000e-05, 8.3613e-05], device='cuda:1') +2023-03-26 23:02:26,690 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6665, 1.6095, 1.4989, 1.6920, 1.1140, 3.6418, 1.4450, 1.8221], + device='cuda:1'), covar=tensor([0.3192, 0.2527, 0.2182, 0.2172, 0.1834, 0.0184, 0.2484, 0.1274], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0123, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 23:02:45,812 INFO [finetune.py:976] (1/7) Epoch 19, batch 2350, loss[loss=0.1413, simple_loss=0.2196, pruned_loss=0.03154, over 4722.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2491, pruned_loss=0.0528, over 954770.96 frames. ], batch size: 23, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:02:45,928 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=105449.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 23:02:48,225 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.033e+02 1.493e+02 1.723e+02 2.054e+02 4.367e+02, threshold=3.447e+02, percent-clipped=1.0 +2023-03-26 23:03:03,541 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9930, 5.0795, 4.8153, 2.9108, 5.1945, 3.7733, 1.0462, 3.6560], + device='cuda:1'), covar=tensor([0.2304, 0.2047, 0.1257, 0.2986, 0.0720, 0.0997, 0.4830, 0.1222], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0177, 0.0160, 0.0129, 0.0160, 0.0123, 0.0147, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 23:03:09,576 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5031, 1.4761, 1.8517, 2.9196, 1.9199, 2.2633, 0.8032, 2.4921], + device='cuda:1'), covar=tensor([0.1700, 0.1349, 0.1174, 0.0615, 0.0817, 0.1339, 0.1756, 0.0447], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0115, 0.0132, 0.0162, 0.0099, 0.0134, 0.0122, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:03:10,222 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=105486.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:03:19,408 INFO [finetune.py:976] (1/7) Epoch 19, batch 2400, loss[loss=0.1837, simple_loss=0.2411, pruned_loss=0.0631, over 4908.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2466, pruned_loss=0.05245, over 955624.00 frames. ], batch size: 32, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:03:23,367 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=105502.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:04:13,220 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8211, 1.6870, 1.5243, 1.8131, 2.1574, 1.8442, 1.5549, 1.5215], + device='cuda:1'), covar=tensor([0.1890, 0.1824, 0.1742, 0.1518, 0.1626, 0.1134, 0.2427, 0.1729], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0209, 0.0212, 0.0192, 0.0242, 0.0186, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:04:14,405 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=105547.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:04:15,985 INFO [finetune.py:976] (1/7) Epoch 19, batch 2450, loss[loss=0.1858, simple_loss=0.2445, pruned_loss=0.06352, over 4928.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2445, pruned_loss=0.05245, over 957321.50 frames. ], batch size: 38, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:04:18,403 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.415e+01 1.457e+02 1.742e+02 2.171e+02 6.143e+02, threshold=3.484e+02, percent-clipped=3.0 +2023-03-26 23:04:20,385 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2102, 2.1458, 2.8047, 1.5818, 2.4226, 2.6388, 1.8844, 2.7095], + device='cuda:1'), covar=tensor([0.1290, 0.1709, 0.1267, 0.2122, 0.0834, 0.1232, 0.2636, 0.0802], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0204, 0.0190, 0.0187, 0.0174, 0.0212, 0.0216, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:04:22,717 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9458, 1.8660, 1.7837, 2.0934, 2.3700, 2.0939, 1.7946, 1.6476], + device='cuda:1'), covar=tensor([0.2381, 0.2102, 0.2052, 0.1695, 0.1739, 0.1223, 0.2455, 0.2077], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0210, 0.0213, 0.0193, 0.0242, 0.0187, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:04:25,635 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=105563.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:04:49,911 INFO [finetune.py:976] (1/7) Epoch 19, batch 2500, loss[loss=0.1713, simple_loss=0.2499, pruned_loss=0.04632, over 4759.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2473, pruned_loss=0.05356, over 957852.97 frames. ], batch size: 27, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:05:01,615 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-26 23:05:23,461 INFO [finetune.py:976] (1/7) Epoch 19, batch 2550, loss[loss=0.1581, simple_loss=0.2409, pruned_loss=0.03765, over 4802.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2506, pruned_loss=0.05427, over 956821.08 frames. ], batch size: 45, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:05:26,385 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.162e+02 1.575e+02 1.924e+02 2.412e+02 4.379e+02, threshold=3.848e+02, percent-clipped=4.0 +2023-03-26 23:05:56,905 INFO [finetune.py:976] (1/7) Epoch 19, batch 2600, loss[loss=0.1948, simple_loss=0.2735, pruned_loss=0.05806, over 4818.00 frames. ], tot_loss[loss=0.1826, simple_loss=0.2535, pruned_loss=0.05587, over 954900.75 frames. ], batch size: 51, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:06:04,019 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6725, 1.2142, 0.8759, 1.5077, 2.0573, 1.0436, 1.4121, 1.4409], + device='cuda:1'), covar=tensor([0.1471, 0.2035, 0.1770, 0.1096, 0.1876, 0.1865, 0.1442, 0.1991], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0093, 0.0109, 0.0090, 0.0118, 0.0091, 0.0097, 0.0087], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:06:08,110 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0800, 1.8977, 2.0825, 1.6741, 2.0553, 2.2667, 2.1904, 1.4564], + device='cuda:1'), covar=tensor([0.0624, 0.0726, 0.0680, 0.1021, 0.0809, 0.0574, 0.0586, 0.1592], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0133, 0.0138, 0.0119, 0.0123, 0.0136, 0.0138, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:06:12,864 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.7875, 1.4705, 1.4140, 0.8548, 1.5711, 1.7468, 1.7294, 1.3867], + device='cuda:1'), covar=tensor([0.0714, 0.0632, 0.0552, 0.0517, 0.0528, 0.0514, 0.0330, 0.0593], + device='cuda:1'), in_proj_covar=tensor([0.0125, 0.0152, 0.0125, 0.0127, 0.0132, 0.0130, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.1512e-05, 1.1024e-04, 8.9815e-05, 8.9903e-05, 9.3322e-05, 9.3072e-05, + 1.0224e-04, 1.0707e-04], device='cuda:1') +2023-03-26 23:06:27,121 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=105744.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 23:06:30,086 INFO [finetune.py:976] (1/7) Epoch 19, batch 2650, loss[loss=0.1734, simple_loss=0.2527, pruned_loss=0.04703, over 4885.00 frames. ], tot_loss[loss=0.1841, simple_loss=0.2553, pruned_loss=0.05646, over 953744.68 frames. ], batch size: 43, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:06:32,907 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.372e+01 1.545e+02 1.903e+02 2.181e+02 3.189e+02, threshold=3.806e+02, percent-clipped=0.0 +2023-03-26 23:06:59,543 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=105792.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:07:03,678 INFO [finetune.py:976] (1/7) Epoch 19, batch 2700, loss[loss=0.142, simple_loss=0.2134, pruned_loss=0.03525, over 4724.00 frames. ], tot_loss[loss=0.1819, simple_loss=0.2527, pruned_loss=0.05552, over 954204.87 frames. ], batch size: 54, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:07:03,799 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=105799.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:07:33,447 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=105842.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:07:37,643 INFO [finetune.py:976] (1/7) Epoch 19, batch 2750, loss[loss=0.1676, simple_loss=0.2331, pruned_loss=0.051, over 4815.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2501, pruned_loss=0.0545, over 955063.20 frames. ], batch size: 41, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:07:37,898 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-26 23:07:40,098 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.803e+01 1.395e+02 1.671e+02 1.966e+02 3.086e+02, threshold=3.343e+02, percent-clipped=0.0 +2023-03-26 23:07:40,225 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=105853.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:07:40,854 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6989, 1.6357, 1.3742, 1.7227, 1.9517, 1.9225, 1.5494, 1.3956], + device='cuda:1'), covar=tensor([0.0324, 0.0318, 0.0666, 0.0297, 0.0225, 0.0431, 0.0372, 0.0467], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0108, 0.0143, 0.0111, 0.0100, 0.0110, 0.0100, 0.0111], + device='cuda:1'), out_proj_covar=tensor([7.4789e-05, 8.2893e-05, 1.1296e-04, 8.4947e-05, 7.7837e-05, 8.1422e-05, + 7.4968e-05, 8.4984e-05], device='cuda:1') +2023-03-26 23:07:43,731 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=105858.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:07:45,011 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=105860.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:08:22,328 INFO [finetune.py:976] (1/7) Epoch 19, batch 2800, loss[loss=0.1355, simple_loss=0.2083, pruned_loss=0.03135, over 4892.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2453, pruned_loss=0.05267, over 954570.47 frames. ], batch size: 35, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:08:30,416 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-26 23:08:50,270 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6644, 1.6452, 1.5340, 1.6220, 1.1980, 3.5492, 1.4851, 1.8244], + device='cuda:1'), covar=tensor([0.3386, 0.2552, 0.2309, 0.2479, 0.1840, 0.0212, 0.2654, 0.1299], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0114, 0.0119, 0.0122, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 23:09:03,615 INFO [finetune.py:976] (1/7) Epoch 19, batch 2850, loss[loss=0.1672, simple_loss=0.222, pruned_loss=0.05623, over 4149.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2451, pruned_loss=0.05324, over 953551.38 frames. ], batch size: 17, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:09:10,695 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.025e+01 1.444e+02 1.775e+02 2.176e+02 4.047e+02, threshold=3.549e+02, percent-clipped=5.0 +2023-03-26 23:09:27,455 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5851, 3.3307, 3.1336, 1.4012, 3.4752, 2.5600, 0.7983, 2.2828], + device='cuda:1'), covar=tensor([0.2527, 0.2484, 0.1701, 0.3778, 0.1293, 0.1085, 0.4504, 0.1607], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0177, 0.0161, 0.0130, 0.0162, 0.0123, 0.0148, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 23:09:47,547 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7880, 4.3889, 4.1359, 1.9685, 4.4125, 3.4519, 1.1168, 3.0869], + device='cuda:1'), covar=tensor([0.2596, 0.1789, 0.1355, 0.3251, 0.0909, 0.0817, 0.3995, 0.1299], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0177, 0.0161, 0.0129, 0.0161, 0.0123, 0.0147, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 23:09:49,344 INFO [finetune.py:976] (1/7) Epoch 19, batch 2900, loss[loss=0.1734, simple_loss=0.2254, pruned_loss=0.06068, over 4193.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2469, pruned_loss=0.05357, over 952510.66 frames. ], batch size: 18, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:10:12,615 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4058, 2.2239, 2.7522, 1.6180, 2.4740, 2.7198, 2.0135, 2.7158], + device='cuda:1'), covar=tensor([0.1355, 0.1939, 0.1534, 0.2345, 0.1021, 0.1344, 0.2479, 0.0922], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0205, 0.0191, 0.0189, 0.0174, 0.0214, 0.0217, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:10:20,990 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106044.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:10:24,380 INFO [finetune.py:976] (1/7) Epoch 19, batch 2950, loss[loss=0.1771, simple_loss=0.2481, pruned_loss=0.05307, over 4813.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2493, pruned_loss=0.05369, over 954580.57 frames. ], batch size: 25, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:10:27,319 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.008e+02 1.599e+02 1.865e+02 2.251e+02 4.962e+02, threshold=3.729e+02, percent-clipped=1.0 +2023-03-26 23:10:43,290 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106077.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:10:45,759 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-26 23:10:53,295 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=106092.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:10:57,518 INFO [finetune.py:976] (1/7) Epoch 19, batch 3000, loss[loss=0.1472, simple_loss=0.2359, pruned_loss=0.02923, over 4820.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2499, pruned_loss=0.05364, over 952970.25 frames. ], batch size: 38, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:10:57,518 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 23:11:02,104 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1955, 1.4701, 1.5326, 0.7726, 1.5290, 1.6645, 1.7481, 1.3932], + device='cuda:1'), covar=tensor([0.0955, 0.0613, 0.0615, 0.0604, 0.0607, 0.0655, 0.0434, 0.0633], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0151, 0.0125, 0.0125, 0.0132, 0.0128, 0.0141, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.0712e-05, 1.0916e-04, 8.9115e-05, 8.9033e-05, 9.2727e-05, 9.1976e-05, + 1.0137e-04, 1.0618e-04], device='cuda:1') +2023-03-26 23:11:08,360 INFO [finetune.py:1010] (1/7) Epoch 19, validation: loss=0.1576, simple_loss=0.2259, pruned_loss=0.04462, over 2265189.00 frames. +2023-03-26 23:11:08,360 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 23:11:28,969 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6101, 3.5625, 3.3920, 1.6206, 3.6698, 2.7407, 0.8344, 2.4837], + device='cuda:1'), covar=tensor([0.2601, 0.1839, 0.1579, 0.3462, 0.1178, 0.1129, 0.4503, 0.1595], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0177, 0.0160, 0.0129, 0.0160, 0.0123, 0.0147, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 23:11:43,395 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106138.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:11:46,230 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106142.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:11:50,365 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106148.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:11:50,870 INFO [finetune.py:976] (1/7) Epoch 19, batch 3050, loss[loss=0.1526, simple_loss=0.2262, pruned_loss=0.03949, over 4757.00 frames. ], tot_loss[loss=0.1787, simple_loss=0.2508, pruned_loss=0.05328, over 953802.72 frames. ], batch size: 28, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:11:53,804 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.036e+02 1.577e+02 1.927e+02 2.196e+02 3.458e+02, threshold=3.854e+02, percent-clipped=0.0 +2023-03-26 23:11:55,078 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106155.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:11:57,418 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106158.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:12:18,599 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=106190.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:12:24,007 INFO [finetune.py:976] (1/7) Epoch 19, batch 3100, loss[loss=0.1563, simple_loss=0.2212, pruned_loss=0.04572, over 4862.00 frames. ], tot_loss[loss=0.1781, simple_loss=0.2497, pruned_loss=0.05327, over 955932.55 frames. ], batch size: 31, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:12:29,243 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=106206.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:12:57,681 INFO [finetune.py:976] (1/7) Epoch 19, batch 3150, loss[loss=0.181, simple_loss=0.2524, pruned_loss=0.05483, over 4821.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.247, pruned_loss=0.0529, over 957934.73 frames. ], batch size: 41, lr: 3.29e-03, grad_scale: 32.0 +2023-03-26 23:13:00,114 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.828e+01 1.675e+02 1.879e+02 2.192e+02 3.916e+02, threshold=3.758e+02, percent-clipped=1.0 +2023-03-26 23:13:04,424 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6735, 1.6639, 1.5983, 1.6831, 1.2610, 3.4645, 1.4608, 1.8066], + device='cuda:1'), covar=tensor([0.3082, 0.2327, 0.2064, 0.2157, 0.1589, 0.0181, 0.2478, 0.1193], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0114, 0.0119, 0.0122, 0.0113, 0.0095, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 23:13:17,032 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106268.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:13:36,067 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2317, 2.2499, 1.9162, 2.3361, 2.1019, 2.1673, 2.1433, 2.9829], + device='cuda:1'), covar=tensor([0.3888, 0.4919, 0.3492, 0.4228, 0.4862, 0.2521, 0.4288, 0.1607], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0260, 0.0230, 0.0275, 0.0251, 0.0221, 0.0251, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:13:36,620 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9544, 1.6075, 2.1444, 1.3987, 2.0950, 2.1646, 1.5315, 2.2328], + device='cuda:1'), covar=tensor([0.1203, 0.1967, 0.1307, 0.1960, 0.0743, 0.1417, 0.2786, 0.0714], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0204, 0.0189, 0.0187, 0.0173, 0.0212, 0.0216, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:13:41,053 INFO [finetune.py:976] (1/7) Epoch 19, batch 3200, loss[loss=0.159, simple_loss=0.23, pruned_loss=0.04398, over 4727.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2437, pruned_loss=0.05208, over 957107.50 frames. ], batch size: 54, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:14:01,439 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106329.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:14:05,118 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4472, 2.4228, 1.9478, 2.5921, 2.3644, 2.0944, 2.8960, 2.5072], + device='cuda:1'), covar=tensor([0.1330, 0.2274, 0.2708, 0.2485, 0.2501, 0.1582, 0.2904, 0.1703], + device='cuda:1'), in_proj_covar=tensor([0.0184, 0.0188, 0.0235, 0.0253, 0.0247, 0.0203, 0.0215, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:14:05,697 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106336.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:14:16,902 INFO [finetune.py:976] (1/7) Epoch 19, batch 3250, loss[loss=0.174, simple_loss=0.2465, pruned_loss=0.05074, over 4814.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2434, pruned_loss=0.0524, over 952472.42 frames. ], batch size: 45, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:14:24,769 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.520e+02 1.839e+02 2.222e+02 4.428e+02, threshold=3.677e+02, percent-clipped=2.0 +2023-03-26 23:14:36,279 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8793, 1.2701, 0.8766, 1.5530, 2.2014, 1.1877, 1.3726, 1.5710], + device='cuda:1'), covar=tensor([0.1519, 0.2121, 0.1998, 0.1278, 0.1810, 0.1994, 0.1567, 0.2049], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0091, 0.0119, 0.0092, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:15:03,609 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0897, 2.1229, 1.9074, 2.3315, 2.6114, 2.1447, 1.9785, 1.6277], + device='cuda:1'), covar=tensor([0.2134, 0.1915, 0.1788, 0.1433, 0.1746, 0.1110, 0.2107, 0.1844], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0209, 0.0212, 0.0192, 0.0242, 0.0187, 0.0215, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:15:08,346 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5625, 3.9401, 4.1931, 4.4272, 4.3353, 4.0423, 4.6908, 1.3927], + device='cuda:1'), covar=tensor([0.0806, 0.0877, 0.0910, 0.1147, 0.1247, 0.1680, 0.0612, 0.5831], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0242, 0.0278, 0.0289, 0.0330, 0.0281, 0.0301, 0.0294], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:15:08,402 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106397.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:15:10,071 INFO [finetune.py:976] (1/7) Epoch 19, batch 3300, loss[loss=0.2294, simple_loss=0.2992, pruned_loss=0.07983, over 4237.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2467, pruned_loss=0.05332, over 952501.41 frames. ], batch size: 66, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:15:17,763 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0301, 1.8137, 1.9713, 0.8181, 2.3637, 2.5220, 2.1862, 1.8006], + device='cuda:1'), covar=tensor([0.0910, 0.0754, 0.0506, 0.0772, 0.0475, 0.0684, 0.0463, 0.0742], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0151, 0.0125, 0.0125, 0.0131, 0.0129, 0.0141, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.1033e-05, 1.0943e-04, 8.9173e-05, 8.8815e-05, 9.2403e-05, 9.2479e-05, + 1.0088e-04, 1.0598e-04], device='cuda:1') +2023-03-26 23:15:41,453 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106433.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:15:51,034 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106448.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:15:51,546 INFO [finetune.py:976] (1/7) Epoch 19, batch 3350, loss[loss=0.1419, simple_loss=0.2235, pruned_loss=0.03018, over 4838.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2475, pruned_loss=0.0533, over 951589.27 frames. ], batch size: 49, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:15:54,457 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.049e+02 1.616e+02 1.883e+02 2.222e+02 4.657e+02, threshold=3.766e+02, percent-clipped=2.0 +2023-03-26 23:15:55,179 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106454.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:15:55,759 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106455.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:16:04,797 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-26 23:16:31,242 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=106496.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:16:31,291 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6549, 1.5671, 1.5221, 1.5520, 1.1584, 3.4687, 1.4830, 1.8028], + device='cuda:1'), covar=tensor([0.3222, 0.2380, 0.2168, 0.2302, 0.1781, 0.0213, 0.2693, 0.1280], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0123, 0.0114, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 23:16:33,553 INFO [finetune.py:976] (1/7) Epoch 19, batch 3400, loss[loss=0.1627, simple_loss=0.2258, pruned_loss=0.0498, over 4684.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.249, pruned_loss=0.05373, over 952651.50 frames. ], batch size: 23, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:16:36,071 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=106503.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:16:44,445 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106515.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:16:58,600 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.82 vs. limit=5.0 +2023-03-26 23:16:59,090 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3921, 3.3194, 3.1924, 1.4057, 3.5474, 2.5620, 0.6888, 2.3109], + device='cuda:1'), covar=tensor([0.2629, 0.1903, 0.1685, 0.3453, 0.1133, 0.1140, 0.4336, 0.1457], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0177, 0.0160, 0.0129, 0.0160, 0.0122, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 23:17:06,742 INFO [finetune.py:976] (1/7) Epoch 19, batch 3450, loss[loss=0.1698, simple_loss=0.235, pruned_loss=0.05232, over 4879.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2491, pruned_loss=0.05367, over 955180.31 frames. ], batch size: 35, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:17:09,626 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.792e+01 1.525e+02 1.781e+02 2.060e+02 3.433e+02, threshold=3.562e+02, percent-clipped=0.0 +2023-03-26 23:17:18,256 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.72 vs. limit=5.0 +2023-03-26 23:17:40,386 INFO [finetune.py:976] (1/7) Epoch 19, batch 3500, loss[loss=0.1876, simple_loss=0.2581, pruned_loss=0.05856, over 4852.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2477, pruned_loss=0.05358, over 955994.21 frames. ], batch size: 44, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:17:57,680 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106624.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:18:14,144 INFO [finetune.py:976] (1/7) Epoch 19, batch 3550, loss[loss=0.1663, simple_loss=0.2332, pruned_loss=0.04966, over 4757.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2452, pruned_loss=0.05293, over 956598.23 frames. ], batch size: 27, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:18:16,539 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.026e+02 1.567e+02 1.861e+02 2.307e+02 3.604e+02, threshold=3.722e+02, percent-clipped=2.0 +2023-03-26 23:18:51,973 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106692.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:18:56,155 INFO [finetune.py:976] (1/7) Epoch 19, batch 3600, loss[loss=0.2173, simple_loss=0.2832, pruned_loss=0.07569, over 4938.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.2436, pruned_loss=0.05297, over 957061.46 frames. ], batch size: 38, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:19:00,397 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3799, 2.1299, 2.1220, 1.5182, 2.1880, 2.2284, 2.3787, 1.7281], + device='cuda:1'), covar=tensor([0.0500, 0.0653, 0.0682, 0.0860, 0.0607, 0.0703, 0.0519, 0.1145], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0133, 0.0138, 0.0119, 0.0123, 0.0137, 0.0138, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:19:19,205 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106733.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:19:29,860 INFO [finetune.py:976] (1/7) Epoch 19, batch 3650, loss[loss=0.17, simple_loss=0.251, pruned_loss=0.04452, over 4836.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2463, pruned_loss=0.05381, over 956253.86 frames. ], batch size: 49, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:19:34,982 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.178e+02 1.609e+02 2.013e+02 2.438e+02 4.457e+02, threshold=4.025e+02, percent-clipped=1.0 +2023-03-26 23:20:03,523 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=106781.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:20:23,369 INFO [finetune.py:976] (1/7) Epoch 19, batch 3700, loss[loss=0.1863, simple_loss=0.263, pruned_loss=0.05478, over 4818.00 frames. ], tot_loss[loss=0.1781, simple_loss=0.2487, pruned_loss=0.05377, over 956393.44 frames. ], batch size: 40, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:20:32,848 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106810.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:20:48,974 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6012, 1.5064, 2.3041, 3.5274, 2.2415, 2.3571, 1.0434, 2.8969], + device='cuda:1'), covar=tensor([0.1836, 0.1542, 0.1317, 0.0490, 0.0876, 0.1611, 0.1881, 0.0446], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0115, 0.0133, 0.0163, 0.0099, 0.0135, 0.0123, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:20:59,510 INFO [finetune.py:976] (1/7) Epoch 19, batch 3750, loss[loss=0.1831, simple_loss=0.2552, pruned_loss=0.05553, over 4811.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2511, pruned_loss=0.05513, over 956858.93 frames. ], batch size: 38, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:21:06,524 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.083e+02 1.604e+02 1.833e+02 2.350e+02 4.465e+02, threshold=3.666e+02, percent-clipped=2.0 +2023-03-26 23:21:48,082 INFO [finetune.py:976] (1/7) Epoch 19, batch 3800, loss[loss=0.2123, simple_loss=0.2867, pruned_loss=0.06895, over 4725.00 frames. ], tot_loss[loss=0.1828, simple_loss=0.2534, pruned_loss=0.05608, over 956675.40 frames. ], batch size: 54, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:22:01,229 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.70 vs. limit=5.0 +2023-03-26 23:22:07,698 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106924.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:22:08,301 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5074, 1.4178, 1.5227, 0.7721, 1.5513, 1.4773, 1.4321, 1.3932], + device='cuda:1'), covar=tensor([0.0605, 0.0803, 0.0649, 0.0983, 0.0830, 0.0714, 0.0688, 0.1243], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0134, 0.0139, 0.0120, 0.0124, 0.0138, 0.0139, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:22:24,705 INFO [finetune.py:976] (1/7) Epoch 19, batch 3850, loss[loss=0.1769, simple_loss=0.2507, pruned_loss=0.0516, over 4815.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2514, pruned_loss=0.05519, over 956478.10 frames. ], batch size: 38, lr: 3.28e-03, grad_scale: 64.0 +2023-03-26 23:22:27,157 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.119e+02 1.623e+02 1.818e+02 2.255e+02 6.115e+02, threshold=3.637e+02, percent-clipped=1.0 +2023-03-26 23:22:39,183 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=106972.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:22:52,766 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106992.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:22:57,315 INFO [finetune.py:976] (1/7) Epoch 19, batch 3900, loss[loss=0.1558, simple_loss=0.2217, pruned_loss=0.04491, over 4791.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2479, pruned_loss=0.05408, over 955900.05 frames. ], batch size: 29, lr: 3.28e-03, grad_scale: 64.0 +2023-03-26 23:23:12,935 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5033, 1.4933, 1.8630, 1.8531, 1.5583, 3.2494, 1.3128, 1.5694], + device='cuda:1'), covar=tensor([0.0925, 0.1720, 0.1086, 0.0855, 0.1547, 0.0255, 0.1480, 0.1717], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0092, 0.0081, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 23:23:24,072 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=107040.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:23:29,943 INFO [finetune.py:976] (1/7) Epoch 19, batch 3950, loss[loss=0.1619, simple_loss=0.2375, pruned_loss=0.04311, over 4913.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2448, pruned_loss=0.05275, over 956671.37 frames. ], batch size: 35, lr: 3.28e-03, grad_scale: 64.0 +2023-03-26 23:23:34,603 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0012, 1.3585, 2.0696, 2.0117, 1.8201, 1.7700, 1.8986, 1.9099], + device='cuda:1'), covar=tensor([0.3663, 0.3762, 0.3069, 0.3403, 0.4207, 0.3549, 0.4146, 0.2867], + device='cuda:1'), in_proj_covar=tensor([0.0250, 0.0239, 0.0260, 0.0278, 0.0275, 0.0250, 0.0285, 0.0241], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:23:35,024 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.301e+01 1.519e+02 1.802e+02 2.250e+02 5.271e+02, threshold=3.605e+02, percent-clipped=1.0 +2023-03-26 23:24:12,974 INFO [finetune.py:976] (1/7) Epoch 19, batch 4000, loss[loss=0.2435, simple_loss=0.3153, pruned_loss=0.08583, over 4837.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2449, pruned_loss=0.05323, over 955002.30 frames. ], batch size: 49, lr: 3.28e-03, grad_scale: 64.0 +2023-03-26 23:24:21,380 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=107110.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:24:41,098 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107140.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 23:24:46,974 INFO [finetune.py:976] (1/7) Epoch 19, batch 4050, loss[loss=0.1768, simple_loss=0.2524, pruned_loss=0.05058, over 4814.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.249, pruned_loss=0.05472, over 955615.48 frames. ], batch size: 51, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:24:48,836 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107152.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:24:49,865 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.013e+02 1.579e+02 1.895e+02 2.231e+02 3.900e+02, threshold=3.790e+02, percent-clipped=1.0 +2023-03-26 23:24:52,894 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=107158.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:25:40,275 INFO [finetune.py:976] (1/7) Epoch 19, batch 4100, loss[loss=0.1769, simple_loss=0.2501, pruned_loss=0.05183, over 4897.00 frames. ], tot_loss[loss=0.181, simple_loss=0.2511, pruned_loss=0.05543, over 954568.49 frames. ], batch size: 43, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:25:41,799 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107201.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 23:25:50,482 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107213.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 23:26:13,432 INFO [finetune.py:976] (1/7) Epoch 19, batch 4150, loss[loss=0.1752, simple_loss=0.2386, pruned_loss=0.05589, over 4875.00 frames. ], tot_loss[loss=0.1827, simple_loss=0.2527, pruned_loss=0.05639, over 954805.11 frames. ], batch size: 34, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:26:21,809 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.077e+02 1.570e+02 1.970e+02 2.461e+02 5.293e+02, threshold=3.939e+02, percent-clipped=1.0 +2023-03-26 23:26:24,386 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5990, 1.5714, 1.4823, 1.6532, 0.9964, 3.2952, 1.2568, 1.7267], + device='cuda:1'), covar=tensor([0.3254, 0.2432, 0.2132, 0.2341, 0.1882, 0.0192, 0.2717, 0.1238], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0123, 0.0114, 0.0095, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 23:26:56,743 INFO [finetune.py:976] (1/7) Epoch 19, batch 4200, loss[loss=0.1779, simple_loss=0.2602, pruned_loss=0.0478, over 4291.00 frames. ], tot_loss[loss=0.182, simple_loss=0.2528, pruned_loss=0.05565, over 954607.45 frames. ], batch size: 66, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:27:05,833 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6717, 1.7043, 2.2376, 1.9113, 1.8304, 4.2627, 1.5292, 1.7643], + device='cuda:1'), covar=tensor([0.0930, 0.1709, 0.1116, 0.0955, 0.1524, 0.0178, 0.1453, 0.1739], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0081, 0.0073, 0.0076, 0.0090, 0.0079, 0.0084, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-26 23:27:19,889 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7833, 1.6893, 2.3763, 3.1054, 2.1343, 2.4004, 1.4473, 2.4611], + device='cuda:1'), covar=tensor([0.1411, 0.1244, 0.0937, 0.0579, 0.0731, 0.1721, 0.1460, 0.0566], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0114, 0.0132, 0.0163, 0.0099, 0.0135, 0.0122, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:27:29,942 INFO [finetune.py:976] (1/7) Epoch 19, batch 4250, loss[loss=0.1716, simple_loss=0.2369, pruned_loss=0.05313, over 4796.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2498, pruned_loss=0.05448, over 954760.40 frames. ], batch size: 25, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:27:33,459 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.100e+02 1.503e+02 1.795e+02 2.146e+02 3.676e+02, threshold=3.590e+02, percent-clipped=0.0 +2023-03-26 23:27:40,893 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.10 vs. limit=5.0 +2023-03-26 23:28:03,400 INFO [finetune.py:976] (1/7) Epoch 19, batch 4300, loss[loss=0.1863, simple_loss=0.2513, pruned_loss=0.06065, over 4787.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2479, pruned_loss=0.0545, over 955509.81 frames. ], batch size: 29, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:28:27,195 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6065, 1.5279, 1.5953, 0.8460, 1.6117, 1.5563, 1.5309, 1.4400], + device='cuda:1'), covar=tensor([0.0573, 0.0715, 0.0649, 0.0963, 0.0909, 0.0724, 0.0678, 0.1259], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0135, 0.0140, 0.0121, 0.0125, 0.0139, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:28:36,189 INFO [finetune.py:976] (1/7) Epoch 19, batch 4350, loss[loss=0.2282, simple_loss=0.2906, pruned_loss=0.08286, over 4809.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2445, pruned_loss=0.05325, over 955145.43 frames. ], batch size: 40, lr: 3.28e-03, grad_scale: 32.0 +2023-03-26 23:28:40,179 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.110e+02 1.532e+02 1.813e+02 2.231e+02 3.395e+02, threshold=3.625e+02, percent-clipped=1.0 +2023-03-26 23:29:21,308 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=107496.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:29:23,018 INFO [finetune.py:976] (1/7) Epoch 19, batch 4400, loss[loss=0.2203, simple_loss=0.2655, pruned_loss=0.08755, over 3930.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2447, pruned_loss=0.05323, over 955082.17 frames. ], batch size: 17, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:29:28,392 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0623, 1.8922, 1.5526, 0.6125, 1.5631, 1.6661, 1.5131, 1.7008], + device='cuda:1'), covar=tensor([0.1051, 0.0826, 0.1501, 0.2053, 0.1450, 0.2541, 0.2419, 0.0963], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0194, 0.0201, 0.0183, 0.0211, 0.0208, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:29:29,602 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=107508.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:29:31,352 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107510.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:29:42,070 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6594, 1.5150, 1.1861, 0.3908, 1.2836, 1.4599, 1.4114, 1.4195], + device='cuda:1'), covar=tensor([0.0925, 0.0761, 0.1338, 0.1753, 0.1375, 0.2365, 0.2277, 0.0893], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0194, 0.0201, 0.0183, 0.0211, 0.0208, 0.0224, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:29:42,711 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.20 vs. limit=5.0 +2023-03-26 23:29:42,809 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-26 23:29:56,826 INFO [finetune.py:976] (1/7) Epoch 19, batch 4450, loss[loss=0.209, simple_loss=0.2869, pruned_loss=0.06554, over 4858.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2484, pruned_loss=0.05421, over 954828.34 frames. ], batch size: 49, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:29:59,905 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.983e+01 1.601e+02 1.972e+02 2.467e+02 3.942e+02, threshold=3.944e+02, percent-clipped=4.0 +2023-03-26 23:30:01,674 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5711, 1.4868, 1.3469, 1.4593, 1.7537, 1.7178, 1.4485, 1.3450], + device='cuda:1'), covar=tensor([0.0321, 0.0314, 0.0619, 0.0315, 0.0229, 0.0476, 0.0372, 0.0403], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0106, 0.0143, 0.0111, 0.0098, 0.0109, 0.0099, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4449e-05, 8.1592e-05, 1.1266e-04, 8.5020e-05, 7.6604e-05, 8.0974e-05, + 7.3505e-05, 8.4015e-05], device='cuda:1') +2023-03-26 23:30:07,510 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0364, 1.0002, 0.9119, 1.1444, 1.2096, 1.1783, 0.9760, 0.9088], + device='cuda:1'), covar=tensor([0.0370, 0.0316, 0.0693, 0.0314, 0.0288, 0.0436, 0.0352, 0.0412], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0106, 0.0143, 0.0111, 0.0098, 0.0109, 0.0099, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4462e-05, 8.1601e-05, 1.1271e-04, 8.5051e-05, 7.6649e-05, 8.0973e-05, + 7.3571e-05, 8.4017e-05], device='cuda:1') +2023-03-26 23:30:12,271 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107571.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:30:15,797 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2642, 2.2195, 1.7108, 2.2742, 2.1612, 1.8810, 2.5503, 2.2743], + device='cuda:1'), covar=tensor([0.1398, 0.1993, 0.2976, 0.2503, 0.2559, 0.1674, 0.3076, 0.1751], + device='cuda:1'), in_proj_covar=tensor([0.0185, 0.0188, 0.0236, 0.0254, 0.0247, 0.0203, 0.0216, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:30:41,969 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-26 23:30:42,901 INFO [finetune.py:976] (1/7) Epoch 19, batch 4500, loss[loss=0.1842, simple_loss=0.2511, pruned_loss=0.05862, over 4890.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2507, pruned_loss=0.05477, over 954995.33 frames. ], batch size: 32, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:30:46,690 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6820, 3.9101, 3.6487, 2.0437, 3.9974, 2.9487, 0.7641, 2.7641], + device='cuda:1'), covar=tensor([0.2399, 0.1560, 0.1365, 0.2822, 0.0972, 0.0968, 0.4265, 0.1278], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0176, 0.0160, 0.0130, 0.0160, 0.0123, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 23:31:25,198 INFO [finetune.py:976] (1/7) Epoch 19, batch 4550, loss[loss=0.1654, simple_loss=0.2366, pruned_loss=0.04712, over 4746.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2512, pruned_loss=0.05509, over 952380.85 frames. ], batch size: 27, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:31:28,200 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.023e+02 1.552e+02 1.832e+02 2.186e+02 5.352e+02, threshold=3.664e+02, percent-clipped=1.0 +2023-03-26 23:31:31,358 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107659.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:32:04,069 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1788, 2.1245, 1.6831, 2.0444, 1.9426, 1.9015, 1.9635, 2.7653], + device='cuda:1'), covar=tensor([0.3984, 0.4187, 0.3566, 0.4097, 0.4321, 0.2625, 0.4287, 0.1821], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0231, 0.0276, 0.0252, 0.0222, 0.0253, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:32:12,110 INFO [finetune.py:976] (1/7) Epoch 19, batch 4600, loss[loss=0.1455, simple_loss=0.1984, pruned_loss=0.0463, over 4248.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2511, pruned_loss=0.05519, over 951572.65 frames. ], batch size: 18, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:32:26,301 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107720.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:32:37,556 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107737.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:32:45,689 INFO [finetune.py:976] (1/7) Epoch 19, batch 4650, loss[loss=0.1731, simple_loss=0.2411, pruned_loss=0.05256, over 4713.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2484, pruned_loss=0.05444, over 952688.40 frames. ], batch size: 59, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:32:48,738 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.855e+01 1.504e+02 1.713e+02 2.086e+02 4.043e+02, threshold=3.426e+02, percent-clipped=2.0 +2023-03-26 23:33:17,160 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=107796.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:33:18,395 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107798.0, num_to_drop=1, layers_to_drop={2} +2023-03-26 23:33:19,347 INFO [finetune.py:976] (1/7) Epoch 19, batch 4700, loss[loss=0.1743, simple_loss=0.2414, pruned_loss=0.05363, over 4843.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2445, pruned_loss=0.05277, over 952837.29 frames. ], batch size: 30, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:33:25,057 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=107808.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:33:33,514 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-26 23:33:49,095 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=107844.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:33:54,086 INFO [finetune.py:976] (1/7) Epoch 19, batch 4750, loss[loss=0.2083, simple_loss=0.2633, pruned_loss=0.07669, over 4130.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2432, pruned_loss=0.05267, over 953030.67 frames. ], batch size: 65, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:33:57,616 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.048e+02 1.438e+02 1.688e+02 2.143e+02 3.806e+02, threshold=3.376e+02, percent-clipped=2.0 +2023-03-26 23:33:58,340 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8383, 1.7249, 2.3557, 1.5273, 2.1111, 2.2366, 1.6295, 2.3806], + device='cuda:1'), covar=tensor([0.1434, 0.1973, 0.1364, 0.2012, 0.0993, 0.1417, 0.2878, 0.0826], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0207, 0.0192, 0.0191, 0.0176, 0.0215, 0.0220, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:33:58,881 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=107856.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:34:04,987 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=107866.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:34:17,632 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2920, 2.2612, 2.2093, 1.5104, 2.2229, 2.3552, 2.3648, 1.7788], + device='cuda:1'), covar=tensor([0.0615, 0.0603, 0.0666, 0.0898, 0.0666, 0.0632, 0.0529, 0.1221], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0134, 0.0140, 0.0121, 0.0125, 0.0139, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:34:37,230 INFO [finetune.py:976] (1/7) Epoch 19, batch 4800, loss[loss=0.1675, simple_loss=0.2421, pruned_loss=0.04641, over 4822.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2446, pruned_loss=0.05276, over 953047.48 frames. ], batch size: 51, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:35:05,280 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-26 23:35:10,747 INFO [finetune.py:976] (1/7) Epoch 19, batch 4850, loss[loss=0.1455, simple_loss=0.2052, pruned_loss=0.04288, over 4381.00 frames. ], tot_loss[loss=0.1793, simple_loss=0.2493, pruned_loss=0.05462, over 954506.57 frames. ], batch size: 19, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:35:13,753 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.163e+02 1.610e+02 1.895e+02 2.225e+02 4.035e+02, threshold=3.790e+02, percent-clipped=2.0 +2023-03-26 23:35:45,813 INFO [finetune.py:976] (1/7) Epoch 19, batch 4900, loss[loss=0.1922, simple_loss=0.2576, pruned_loss=0.0634, over 4793.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2504, pruned_loss=0.05526, over 955142.84 frames. ], batch size: 25, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:35:57,375 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=108015.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:36:07,144 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-26 23:36:28,166 INFO [finetune.py:976] (1/7) Epoch 19, batch 4950, loss[loss=0.1824, simple_loss=0.2463, pruned_loss=0.05925, over 4234.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2504, pruned_loss=0.05474, over 953564.64 frames. ], batch size: 65, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:36:31,628 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.136e+02 1.572e+02 1.807e+02 2.323e+02 4.539e+02, threshold=3.614e+02, percent-clipped=1.0 +2023-03-26 23:37:04,009 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=108093.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:37:10,955 INFO [finetune.py:976] (1/7) Epoch 19, batch 5000, loss[loss=0.1849, simple_loss=0.2514, pruned_loss=0.05921, over 4753.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.2484, pruned_loss=0.05373, over 954575.43 frames. ], batch size: 26, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:37:54,037 INFO [finetune.py:976] (1/7) Epoch 19, batch 5050, loss[loss=0.1772, simple_loss=0.2454, pruned_loss=0.05448, over 4918.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2473, pruned_loss=0.05386, over 954683.49 frames. ], batch size: 37, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:37:57,572 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.150e+02 1.577e+02 1.863e+02 2.132e+02 3.762e+02, threshold=3.725e+02, percent-clipped=1.0 +2023-03-26 23:38:05,858 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=108166.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:38:10,045 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0254, 0.9669, 0.8188, 0.9580, 1.1758, 1.1685, 1.0131, 0.8823], + device='cuda:1'), covar=tensor([0.0343, 0.0361, 0.0882, 0.0413, 0.0284, 0.0447, 0.0352, 0.0457], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0107, 0.0144, 0.0111, 0.0099, 0.0110, 0.0099, 0.0111], + device='cuda:1'), out_proj_covar=tensor([7.4721e-05, 8.2550e-05, 1.1326e-04, 8.5455e-05, 7.7151e-05, 8.1460e-05, + 7.3894e-05, 8.4802e-05], device='cuda:1') +2023-03-26 23:38:27,769 INFO [finetune.py:976] (1/7) Epoch 19, batch 5100, loss[loss=0.17, simple_loss=0.2362, pruned_loss=0.05191, over 4821.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.245, pruned_loss=0.0529, over 955576.46 frames. ], batch size: 30, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:38:38,020 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=108214.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:39:00,757 INFO [finetune.py:976] (1/7) Epoch 19, batch 5150, loss[loss=0.1331, simple_loss=0.2084, pruned_loss=0.02886, over 4815.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2452, pruned_loss=0.05272, over 956813.50 frames. ], batch size: 25, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:39:04,795 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.278e+01 1.451e+02 1.873e+02 2.231e+02 4.201e+02, threshold=3.747e+02, percent-clipped=0.0 +2023-03-26 23:39:39,542 INFO [finetune.py:976] (1/7) Epoch 19, batch 5200, loss[loss=0.201, simple_loss=0.2582, pruned_loss=0.07189, over 4350.00 frames. ], tot_loss[loss=0.1806, simple_loss=0.2509, pruned_loss=0.05518, over 954816.05 frames. ], batch size: 19, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:39:54,419 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=108315.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:40:01,024 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=108325.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:40:16,476 INFO [finetune.py:976] (1/7) Epoch 19, batch 5250, loss[loss=0.1736, simple_loss=0.2448, pruned_loss=0.05125, over 4756.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2523, pruned_loss=0.0555, over 955837.29 frames. ], batch size: 54, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:40:19,991 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.128e+02 1.566e+02 1.984e+02 2.332e+02 4.295e+02, threshold=3.968e+02, percent-clipped=2.0 +2023-03-26 23:40:26,549 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=108363.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:40:42,042 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=108386.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:40:46,350 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=108393.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:40:49,957 INFO [finetune.py:976] (1/7) Epoch 19, batch 5300, loss[loss=0.1754, simple_loss=0.2502, pruned_loss=0.05026, over 4821.00 frames. ], tot_loss[loss=0.1814, simple_loss=0.2521, pruned_loss=0.0553, over 955045.39 frames. ], batch size: 38, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:40:56,571 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0620, 1.8179, 2.3453, 3.7558, 2.6097, 2.6708, 0.8707, 3.2443], + device='cuda:1'), covar=tensor([0.1648, 0.1551, 0.1501, 0.0663, 0.0826, 0.2043, 0.2107, 0.0447], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0164, 0.0100, 0.0136, 0.0124, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:41:22,521 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=108432.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:41:32,025 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=108441.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:41:36,859 INFO [finetune.py:976] (1/7) Epoch 19, batch 5350, loss[loss=0.2511, simple_loss=0.3114, pruned_loss=0.09543, over 4804.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2522, pruned_loss=0.05515, over 953663.91 frames. ], batch size: 40, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:41:39,474 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.73 vs. limit=2.0 +2023-03-26 23:41:39,875 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.546e+01 1.453e+02 1.815e+02 2.266e+02 3.194e+02, threshold=3.630e+02, percent-clipped=0.0 +2023-03-26 23:42:03,425 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-26 23:42:09,079 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=108493.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 23:42:12,598 INFO [finetune.py:976] (1/7) Epoch 19, batch 5400, loss[loss=0.1734, simple_loss=0.2528, pruned_loss=0.04703, over 4779.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2494, pruned_loss=0.05445, over 953776.83 frames. ], batch size: 28, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:42:58,083 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5776, 3.4860, 3.3460, 1.4943, 3.5998, 2.7032, 0.8124, 2.4966], + device='cuda:1'), covar=tensor([0.2723, 0.2176, 0.1599, 0.3474, 0.1170, 0.1088, 0.4418, 0.1494], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0176, 0.0160, 0.0129, 0.0159, 0.0122, 0.0146, 0.0122], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 23:42:58,626 INFO [finetune.py:976] (1/7) Epoch 19, batch 5450, loss[loss=0.1886, simple_loss=0.2512, pruned_loss=0.06306, over 4808.00 frames. ], tot_loss[loss=0.176, simple_loss=0.246, pruned_loss=0.05299, over 954842.14 frames. ], batch size: 51, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:43:01,646 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.000e+02 1.562e+02 1.816e+02 2.216e+02 4.232e+02, threshold=3.632e+02, percent-clipped=2.0 +2023-03-26 23:43:05,420 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1272, 1.8420, 2.2959, 1.5939, 2.1385, 2.2752, 1.6811, 2.4734], + device='cuda:1'), covar=tensor([0.1142, 0.2004, 0.1366, 0.2002, 0.0904, 0.1481, 0.2833, 0.0855], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0204, 0.0191, 0.0189, 0.0174, 0.0213, 0.0216, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:43:17,254 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=108577.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:43:18,412 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6081, 1.6119, 2.1530, 3.2949, 2.1447, 2.4126, 1.2267, 2.6694], + device='cuda:1'), covar=tensor([0.1886, 0.1400, 0.1322, 0.0495, 0.0893, 0.1252, 0.1787, 0.0527], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0134, 0.0164, 0.0101, 0.0137, 0.0124, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:43:19,088 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0629, 1.5345, 2.1177, 1.9852, 1.8447, 1.8025, 1.9880, 1.9617], + device='cuda:1'), covar=tensor([0.3500, 0.3582, 0.3250, 0.3519, 0.4589, 0.3701, 0.4320, 0.3020], + device='cuda:1'), in_proj_covar=tensor([0.0251, 0.0239, 0.0260, 0.0279, 0.0276, 0.0251, 0.0286, 0.0242], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:43:31,857 INFO [finetune.py:976] (1/7) Epoch 19, batch 5500, loss[loss=0.1507, simple_loss=0.2149, pruned_loss=0.04326, over 4769.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2428, pruned_loss=0.05207, over 954535.82 frames. ], batch size: 28, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:43:58,826 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=108638.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:44:05,685 INFO [finetune.py:976] (1/7) Epoch 19, batch 5550, loss[loss=0.195, simple_loss=0.2668, pruned_loss=0.06164, over 4801.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2429, pruned_loss=0.05176, over 951104.72 frames. ], batch size: 45, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:44:08,703 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.553e+02 1.822e+02 2.201e+02 3.552e+02, threshold=3.643e+02, percent-clipped=0.0 +2023-03-26 23:44:27,070 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=108681.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:44:37,477 INFO [finetune.py:976] (1/7) Epoch 19, batch 5600, loss[loss=0.1675, simple_loss=0.2419, pruned_loss=0.04652, over 4862.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2469, pruned_loss=0.05292, over 952930.67 frames. ], batch size: 31, lr: 3.27e-03, grad_scale: 32.0 +2023-03-26 23:44:38,823 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.61 vs. limit=5.0 +2023-03-26 23:45:09,466 INFO [finetune.py:976] (1/7) Epoch 19, batch 5650, loss[loss=0.1657, simple_loss=0.2522, pruned_loss=0.03955, over 4926.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.25, pruned_loss=0.05339, over 954622.27 frames. ], batch size: 38, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:45:12,317 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.055e+02 1.584e+02 1.878e+02 2.184e+02 3.636e+02, threshold=3.756e+02, percent-clipped=0.0 +2023-03-26 23:45:32,949 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=108788.0, num_to_drop=1, layers_to_drop={3} +2023-03-26 23:45:34,788 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=108791.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:45:39,454 INFO [finetune.py:976] (1/7) Epoch 19, batch 5700, loss[loss=0.1687, simple_loss=0.2239, pruned_loss=0.05673, over 4179.00 frames. ], tot_loss[loss=0.177, simple_loss=0.247, pruned_loss=0.05355, over 933899.18 frames. ], batch size: 18, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:45:44,094 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.59 vs. limit=2.0 +2023-03-26 23:45:51,627 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4407, 2.2608, 1.9660, 2.2560, 2.3105, 2.1338, 2.5844, 2.4230], + device='cuda:1'), covar=tensor([0.1298, 0.2101, 0.3118, 0.2441, 0.2552, 0.1651, 0.2542, 0.1890], + device='cuda:1'), in_proj_covar=tensor([0.0185, 0.0188, 0.0235, 0.0254, 0.0247, 0.0204, 0.0216, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:46:08,106 INFO [finetune.py:976] (1/7) Epoch 20, batch 0, loss[loss=0.1834, simple_loss=0.2561, pruned_loss=0.05539, over 4665.00 frames. ], tot_loss[loss=0.1834, simple_loss=0.2561, pruned_loss=0.05539, over 4665.00 frames. ], batch size: 59, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:46:08,106 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-26 23:46:21,823 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8885, 1.3133, 0.9714, 1.6667, 2.1168, 1.2515, 1.6626, 1.5876], + device='cuda:1'), covar=tensor([0.1264, 0.1725, 0.1677, 0.0995, 0.1757, 0.1821, 0.1131, 0.1789], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0111, 0.0091, 0.0120, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:46:23,200 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2160, 1.9468, 1.8419, 1.7896, 1.8845, 1.9581, 1.9320, 2.6012], + device='cuda:1'), covar=tensor([0.3586, 0.4812, 0.3394, 0.3695, 0.4043, 0.2673, 0.3855, 0.1700], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0261, 0.0230, 0.0275, 0.0251, 0.0221, 0.0252, 0.0232], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:46:24,545 INFO [finetune.py:1010] (1/7) Epoch 20, validation: loss=0.158, simple_loss=0.2276, pruned_loss=0.04423, over 2265189.00 frames. +2023-03-26 23:46:24,545 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-26 23:46:52,595 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=108852.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:46:58,208 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.806e+01 1.424e+02 1.737e+02 2.098e+02 5.389e+02, threshold=3.475e+02, percent-clipped=2.0 +2023-03-26 23:47:01,486 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-26 23:47:17,654 INFO [finetune.py:976] (1/7) Epoch 20, batch 50, loss[loss=0.1834, simple_loss=0.2534, pruned_loss=0.05667, over 4818.00 frames. ], tot_loss[loss=0.1835, simple_loss=0.2541, pruned_loss=0.05645, over 216435.33 frames. ], batch size: 38, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:47:31,379 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0079, 4.4134, 4.3537, 2.1471, 4.5540, 3.4184, 0.6953, 3.1846], + device='cuda:1'), covar=tensor([0.2238, 0.1871, 0.1227, 0.3079, 0.0732, 0.0870, 0.4598, 0.1270], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0173, 0.0158, 0.0128, 0.0157, 0.0121, 0.0145, 0.0121], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-26 23:47:57,360 INFO [finetune.py:976] (1/7) Epoch 20, batch 100, loss[loss=0.1485, simple_loss=0.2147, pruned_loss=0.04111, over 4719.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.2462, pruned_loss=0.05484, over 380580.66 frames. ], batch size: 54, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:48:06,347 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=108933.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:48:23,147 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.009e+02 1.362e+02 1.754e+02 2.070e+02 5.157e+02, threshold=3.508e+02, percent-clipped=1.0 +2023-03-26 23:48:23,913 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6768, 1.5323, 1.9247, 1.2549, 1.6423, 1.8303, 1.5006, 1.9562], + device='cuda:1'), covar=tensor([0.1176, 0.2084, 0.1185, 0.1703, 0.0918, 0.1311, 0.2710, 0.0798], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0203, 0.0190, 0.0188, 0.0173, 0.0211, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:48:38,579 INFO [finetune.py:976] (1/7) Epoch 20, batch 150, loss[loss=0.178, simple_loss=0.2473, pruned_loss=0.0544, over 4914.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2429, pruned_loss=0.0545, over 508892.41 frames. ], batch size: 36, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:48:41,558 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=108981.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:48:42,198 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4903, 1.3632, 1.3338, 1.4103, 1.6901, 1.5712, 1.4156, 1.3272], + device='cuda:1'), covar=tensor([0.0294, 0.0306, 0.0629, 0.0316, 0.0213, 0.0474, 0.0330, 0.0395], + device='cuda:1'), in_proj_covar=tensor([0.0096, 0.0106, 0.0142, 0.0110, 0.0098, 0.0109, 0.0098, 0.0110], + device='cuda:1'), out_proj_covar=tensor([7.4134e-05, 8.1646e-05, 1.1191e-04, 8.4489e-05, 7.6606e-05, 8.0729e-05, + 7.3179e-05, 8.3893e-05], device='cuda:1') +2023-03-26 23:49:01,586 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7303, 1.2696, 0.8662, 1.7023, 1.9913, 1.5912, 1.6127, 1.6478], + device='cuda:1'), covar=tensor([0.1475, 0.2017, 0.2062, 0.1081, 0.2079, 0.1847, 0.1306, 0.1902], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0091, 0.0120, 0.0092, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:49:11,420 INFO [finetune.py:976] (1/7) Epoch 20, batch 200, loss[loss=0.1435, simple_loss=0.2207, pruned_loss=0.03317, over 4756.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2423, pruned_loss=0.05388, over 610648.05 frames. ], batch size: 28, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:49:11,520 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8521, 1.2878, 0.9260, 1.7319, 2.1505, 1.5358, 1.6974, 1.6812], + device='cuda:1'), covar=tensor([0.1521, 0.2120, 0.1963, 0.1210, 0.1989, 0.1784, 0.1410, 0.1959], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0111, 0.0091, 0.0120, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:49:13,181 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=109029.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:49:16,079 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8371, 1.2342, 0.9001, 1.6202, 2.0973, 1.3926, 1.6044, 1.5822], + device='cuda:1'), covar=tensor([0.1501, 0.2206, 0.2004, 0.1252, 0.1988, 0.1932, 0.1452, 0.2072], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0111, 0.0091, 0.0120, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:49:19,012 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109037.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:49:29,131 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.011e+02 1.519e+02 1.780e+02 2.129e+02 3.450e+02, threshold=3.561e+02, percent-clipped=0.0 +2023-03-26 23:49:44,471 INFO [finetune.py:976] (1/7) Epoch 20, batch 250, loss[loss=0.2268, simple_loss=0.3024, pruned_loss=0.07563, over 4811.00 frames. ], tot_loss[loss=0.1781, simple_loss=0.2466, pruned_loss=0.05482, over 688780.47 frames. ], batch size: 45, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:49:52,145 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109088.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:49:58,699 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109098.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:50:17,243 INFO [finetune.py:976] (1/7) Epoch 20, batch 300, loss[loss=0.1788, simple_loss=0.2547, pruned_loss=0.05145, over 4918.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2489, pruned_loss=0.05505, over 747522.26 frames. ], batch size: 38, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:50:23,611 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=109136.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:50:31,197 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109147.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:50:35,398 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.121e+02 1.518e+02 1.856e+02 2.256e+02 3.204e+02, threshold=3.712e+02, percent-clipped=0.0 +2023-03-26 23:50:47,510 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8577, 1.6770, 1.5347, 1.9503, 2.0574, 1.9310, 1.3508, 1.5265], + device='cuda:1'), covar=tensor([0.2085, 0.1916, 0.1836, 0.1703, 0.1604, 0.1155, 0.2528, 0.1884], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0210, 0.0211, 0.0194, 0.0244, 0.0188, 0.0216, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:50:50,164 INFO [finetune.py:976] (1/7) Epoch 20, batch 350, loss[loss=0.1615, simple_loss=0.2405, pruned_loss=0.04128, over 4847.00 frames. ], tot_loss[loss=0.1816, simple_loss=0.2514, pruned_loss=0.05588, over 793408.23 frames. ], batch size: 49, lr: 3.26e-03, grad_scale: 64.0 +2023-03-26 23:51:01,459 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109194.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:51:25,272 INFO [finetune.py:976] (1/7) Epoch 20, batch 400, loss[loss=0.1659, simple_loss=0.2376, pruned_loss=0.04708, over 4791.00 frames. ], tot_loss[loss=0.1813, simple_loss=0.2517, pruned_loss=0.05545, over 829514.79 frames. ], batch size: 29, lr: 3.26e-03, grad_scale: 64.0 +2023-03-26 23:51:34,321 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109233.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:51:53,582 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5517, 1.5433, 2.0355, 2.9888, 1.9879, 2.3169, 1.2298, 2.4990], + device='cuda:1'), covar=tensor([0.1783, 0.1356, 0.1228, 0.0701, 0.0874, 0.1199, 0.1607, 0.0536], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0134, 0.0165, 0.0101, 0.0136, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:52:03,601 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.047e+02 1.657e+02 1.900e+02 2.185e+02 4.941e+02, threshold=3.801e+02, percent-clipped=3.0 +2023-03-26 23:52:03,751 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8148, 1.7097, 1.4744, 1.8569, 2.2319, 1.9450, 1.4552, 1.4717], + device='cuda:1'), covar=tensor([0.2116, 0.2004, 0.1942, 0.1722, 0.1708, 0.1128, 0.2576, 0.2007], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0211, 0.0211, 0.0194, 0.0244, 0.0188, 0.0216, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-26 23:52:04,369 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109255.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:52:04,432 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.53 vs. limit=2.0 +2023-03-26 23:52:17,241 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7410, 1.1566, 0.8499, 1.6153, 2.0760, 1.4824, 1.5270, 1.6736], + device='cuda:1'), covar=tensor([0.1419, 0.1966, 0.1875, 0.1128, 0.1868, 0.1999, 0.1327, 0.1741], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0091, 0.0119, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-26 23:52:26,282 INFO [finetune.py:976] (1/7) Epoch 20, batch 450, loss[loss=0.1799, simple_loss=0.2435, pruned_loss=0.05816, over 4887.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2498, pruned_loss=0.0548, over 855947.22 frames. ], batch size: 32, lr: 3.26e-03, grad_scale: 64.0 +2023-03-26 23:52:29,295 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=109281.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:52:38,178 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109294.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:52:54,642 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109318.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:53:02,102 INFO [finetune.py:976] (1/7) Epoch 20, batch 500, loss[loss=0.2446, simple_loss=0.2861, pruned_loss=0.1016, over 4712.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2478, pruned_loss=0.0543, over 877947.05 frames. ], batch size: 23, lr: 3.26e-03, grad_scale: 64.0 +2023-03-26 23:53:23,181 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.83 vs. limit=5.0 +2023-03-26 23:53:34,824 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.087e+02 1.492e+02 1.802e+02 2.178e+02 4.247e+02, threshold=3.605e+02, percent-clipped=3.0 +2023-03-26 23:53:39,384 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109355.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:53:47,670 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3882, 1.4079, 1.4221, 0.8061, 1.4790, 1.6250, 1.7037, 1.3649], + device='cuda:1'), covar=tensor([0.0817, 0.0529, 0.0483, 0.0435, 0.0444, 0.0586, 0.0288, 0.0536], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0149, 0.0124, 0.0124, 0.0130, 0.0128, 0.0142, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.0243e-05, 1.0815e-04, 8.8500e-05, 8.7808e-05, 9.1689e-05, 9.1520e-05, + 1.0162e-04, 1.0523e-04], device='cuda:1') +2023-03-26 23:53:52,983 INFO [finetune.py:976] (1/7) Epoch 20, batch 550, loss[loss=0.1663, simple_loss=0.2232, pruned_loss=0.05463, over 4903.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2455, pruned_loss=0.05355, over 896840.98 frames. ], batch size: 32, lr: 3.26e-03, grad_scale: 64.0 +2023-03-26 23:53:54,338 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109379.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:54:03,684 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109393.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:54:26,238 INFO [finetune.py:976] (1/7) Epoch 20, batch 600, loss[loss=0.1806, simple_loss=0.2619, pruned_loss=0.04963, over 4913.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2466, pruned_loss=0.05434, over 908914.37 frames. ], batch size: 43, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:54:39,778 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109447.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:54:44,547 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.417e+01 1.548e+02 1.729e+02 2.159e+02 3.434e+02, threshold=3.458e+02, percent-clipped=0.0 +2023-03-26 23:54:56,531 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-03-26 23:54:59,322 INFO [finetune.py:976] (1/7) Epoch 20, batch 650, loss[loss=0.2224, simple_loss=0.2879, pruned_loss=0.07846, over 4815.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2492, pruned_loss=0.05496, over 919000.31 frames. ], batch size: 39, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:55:10,858 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=109495.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:55:33,023 INFO [finetune.py:976] (1/7) Epoch 20, batch 700, loss[loss=0.1759, simple_loss=0.2439, pruned_loss=0.054, over 4810.00 frames. ], tot_loss[loss=0.18, simple_loss=0.2503, pruned_loss=0.05487, over 926334.31 frames. ], batch size: 39, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:55:47,501 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109550.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:55:51,348 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.095e+02 1.485e+02 1.783e+02 2.085e+02 4.380e+02, threshold=3.566e+02, percent-clipped=3.0 +2023-03-26 23:56:06,061 INFO [finetune.py:976] (1/7) Epoch 20, batch 750, loss[loss=0.1509, simple_loss=0.2305, pruned_loss=0.03564, over 4820.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2509, pruned_loss=0.05411, over 934318.76 frames. ], batch size: 25, lr: 3.26e-03, grad_scale: 32.0 +2023-03-26 23:56:07,735 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109579.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:56:39,568 INFO [finetune.py:976] (1/7) Epoch 20, batch 800, loss[loss=0.1754, simple_loss=0.2489, pruned_loss=0.05096, over 4820.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2502, pruned_loss=0.05427, over 937385.48 frames. ], batch size: 41, lr: 3.25e-03, grad_scale: 32.0 +2023-03-26 23:56:42,613 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-26 23:56:50,330 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109640.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:56:56,827 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109650.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:56:59,766 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.018e+02 1.497e+02 1.774e+02 2.103e+02 3.199e+02, threshold=3.548e+02, percent-clipped=0.0 +2023-03-26 23:57:03,313 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109659.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:57:23,375 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109674.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 23:57:25,121 INFO [finetune.py:976] (1/7) Epoch 20, batch 850, loss[loss=0.2439, simple_loss=0.2949, pruned_loss=0.09646, over 4768.00 frames. ], tot_loss[loss=0.1784, simple_loss=0.2486, pruned_loss=0.05404, over 943433.50 frames. ], batch size: 26, lr: 3.25e-03, grad_scale: 32.0 +2023-03-26 23:57:40,049 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109693.0, num_to_drop=1, layers_to_drop={0} +2023-03-26 23:58:06,042 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109720.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:58:10,728 INFO [finetune.py:976] (1/7) Epoch 20, batch 900, loss[loss=0.1396, simple_loss=0.2111, pruned_loss=0.03411, over 4851.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2459, pruned_loss=0.05335, over 946226.36 frames. ], batch size: 47, lr: 3.25e-03, grad_scale: 32.0 +2023-03-26 23:58:22,308 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=109741.0, num_to_drop=1, layers_to_drop={1} +2023-03-26 23:58:40,322 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.017e+02 1.530e+02 1.823e+02 2.181e+02 3.809e+02, threshold=3.647e+02, percent-clipped=1.0 +2023-03-26 23:59:03,843 INFO [finetune.py:976] (1/7) Epoch 20, batch 950, loss[loss=0.2357, simple_loss=0.3018, pruned_loss=0.08482, over 4743.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2447, pruned_loss=0.05304, over 949129.32 frames. ], batch size: 59, lr: 3.25e-03, grad_scale: 32.0 +2023-03-26 23:59:36,851 INFO [finetune.py:976] (1/7) Epoch 20, batch 1000, loss[loss=0.2126, simple_loss=0.2772, pruned_loss=0.07402, over 4922.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2457, pruned_loss=0.0536, over 950888.90 frames. ], batch size: 38, lr: 3.25e-03, grad_scale: 32.0 +2023-03-26 23:59:38,840 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.74 vs. limit=2.0 +2023-03-26 23:59:52,406 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109850.0, num_to_drop=0, layers_to_drop=set() +2023-03-26 23:59:55,343 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.032e+02 1.630e+02 1.951e+02 2.313e+02 5.473e+02, threshold=3.903e+02, percent-clipped=2.0 +2023-03-27 00:00:00,715 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5963, 3.6356, 3.4133, 1.4660, 3.7424, 2.8930, 1.2287, 2.4954], + device='cuda:1'), covar=tensor([0.2518, 0.1887, 0.1495, 0.3745, 0.1069, 0.0935, 0.3888, 0.1438], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0177, 0.0160, 0.0131, 0.0161, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 00:00:08,426 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1174, 1.8882, 1.8075, 1.9761, 1.6201, 4.7664, 1.8976, 2.2563], + device='cuda:1'), covar=tensor([0.2937, 0.2222, 0.2024, 0.2215, 0.1517, 0.0126, 0.2245, 0.1134], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0120, 0.0122, 0.0114, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 00:00:10,589 INFO [finetune.py:976] (1/7) Epoch 20, batch 1050, loss[loss=0.1997, simple_loss=0.2749, pruned_loss=0.06222, over 4751.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2495, pruned_loss=0.05438, over 950999.63 frames. ], batch size: 28, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:00:10,791 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-03-27 00:00:24,789 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=109898.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:00:43,715 INFO [finetune.py:976] (1/7) Epoch 20, batch 1100, loss[loss=0.2218, simple_loss=0.284, pruned_loss=0.07978, over 4868.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2509, pruned_loss=0.05505, over 949209.67 frames. ], batch size: 31, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:00:49,670 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109935.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:00:59,278 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-27 00:00:59,753 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109950.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:01:02,698 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.100e+02 1.532e+02 1.890e+02 2.271e+02 3.423e+02, threshold=3.780e+02, percent-clipped=0.0 +2023-03-27 00:01:15,520 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109974.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:01:17,221 INFO [finetune.py:976] (1/7) Epoch 20, batch 1150, loss[loss=0.1896, simple_loss=0.2543, pruned_loss=0.06244, over 4771.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2519, pruned_loss=0.05524, over 951882.63 frames. ], batch size: 27, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:01:31,934 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=109998.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:01:32,008 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1669, 1.3576, 1.4260, 0.6916, 1.3333, 1.6006, 1.6455, 1.3213], + device='cuda:1'), covar=tensor([0.0890, 0.0602, 0.0564, 0.0546, 0.0514, 0.0618, 0.0364, 0.0741], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0149, 0.0124, 0.0124, 0.0130, 0.0128, 0.0141, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.0143e-05, 1.0813e-04, 8.8282e-05, 8.7655e-05, 9.1458e-05, 9.1814e-05, + 1.0129e-04, 1.0529e-04], device='cuda:1') +2023-03-27 00:01:40,361 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5851, 1.4184, 1.3094, 1.4346, 1.7468, 1.7646, 1.4761, 1.3416], + device='cuda:1'), covar=tensor([0.0357, 0.0351, 0.0678, 0.0341, 0.0278, 0.0478, 0.0375, 0.0438], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0108, 0.0145, 0.0112, 0.0100, 0.0111, 0.0100, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.5785e-05, 8.3129e-05, 1.1455e-04, 8.6021e-05, 7.8142e-05, 8.2445e-05, + 7.4904e-05, 8.5218e-05], device='cuda:1') +2023-03-27 00:01:44,406 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110015.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:01:45,658 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110017.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:01:48,679 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=110022.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:01:52,168 INFO [finetune.py:976] (1/7) Epoch 20, batch 1200, loss[loss=0.1652, simple_loss=0.2285, pruned_loss=0.05093, over 4920.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2508, pruned_loss=0.05474, over 953861.16 frames. ], batch size: 33, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:01:53,537 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8673, 1.7980, 1.5639, 1.9642, 2.3038, 1.9706, 1.5846, 1.5025], + device='cuda:1'), covar=tensor([0.2064, 0.1872, 0.1847, 0.1471, 0.1581, 0.1179, 0.2357, 0.1902], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0211, 0.0194, 0.0245, 0.0188, 0.0216, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:02:04,537 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110044.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:02:04,672 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-03-27 00:02:11,656 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.682e+01 1.527e+02 1.776e+02 2.166e+02 4.163e+02, threshold=3.552e+02, percent-clipped=2.0 +2023-03-27 00:02:32,491 INFO [finetune.py:976] (1/7) Epoch 20, batch 1250, loss[loss=0.1461, simple_loss=0.2121, pruned_loss=0.04004, over 4910.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.2477, pruned_loss=0.05381, over 953467.39 frames. ], batch size: 36, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:02:33,243 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110078.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:03:02,206 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110105.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:03:09,988 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110110.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:03:13,079 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110115.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:03:21,579 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9495, 1.6913, 2.3302, 1.4605, 2.1317, 2.2066, 1.5873, 2.3373], + device='cuda:1'), covar=tensor([0.1234, 0.2019, 0.1346, 0.1897, 0.0838, 0.1252, 0.2829, 0.0782], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0206, 0.0192, 0.0191, 0.0176, 0.0215, 0.0219, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:03:23,867 INFO [finetune.py:976] (1/7) Epoch 20, batch 1300, loss[loss=0.1855, simple_loss=0.2553, pruned_loss=0.05788, over 4815.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2452, pruned_loss=0.05298, over 953348.29 frames. ], batch size: 41, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:03:45,426 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.072e+02 1.532e+02 1.884e+02 2.318e+02 4.682e+02, threshold=3.767e+02, percent-clipped=2.0 +2023-03-27 00:04:01,793 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7002, 1.6220, 1.6959, 1.5855, 1.2663, 2.7437, 1.2528, 1.6833], + device='cuda:1'), covar=tensor([0.2930, 0.2295, 0.1813, 0.2133, 0.1523, 0.0318, 0.2463, 0.1125], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0115, 0.0119, 0.0122, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 00:04:04,754 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110171.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:04:12,705 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110176.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:04:13,180 INFO [finetune.py:976] (1/7) Epoch 20, batch 1350, loss[loss=0.1755, simple_loss=0.2527, pruned_loss=0.04918, over 4809.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2441, pruned_loss=0.05267, over 952219.91 frames. ], batch size: 38, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:05:15,850 INFO [finetune.py:976] (1/7) Epoch 20, batch 1400, loss[loss=0.2008, simple_loss=0.2692, pruned_loss=0.06623, over 4907.00 frames. ], tot_loss[loss=0.1773, simple_loss=0.2475, pruned_loss=0.05361, over 952418.52 frames. ], batch size: 35, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:05:26,720 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110235.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:05:48,218 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.073e+02 1.545e+02 1.824e+02 2.176e+02 3.637e+02, threshold=3.648e+02, percent-clipped=0.0 +2023-03-27 00:06:02,045 INFO [finetune.py:976] (1/7) Epoch 20, batch 1450, loss[loss=0.1508, simple_loss=0.2349, pruned_loss=0.03334, over 4832.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2492, pruned_loss=0.05336, over 951171.41 frames. ], batch size: 30, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:06:06,235 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=110283.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:06:06,278 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110283.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:06:20,402 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110303.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:06:21,574 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7922, 1.2921, 0.8537, 1.6398, 2.1341, 1.5397, 1.6577, 1.7186], + device='cuda:1'), covar=tensor([0.1436, 0.2100, 0.1982, 0.1134, 0.1913, 0.2100, 0.1349, 0.1947], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0091, 0.0119, 0.0092, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:06:24,483 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6310, 3.5787, 3.4366, 1.7349, 3.7018, 2.7282, 0.7459, 2.4598], + device='cuda:1'), covar=tensor([0.2338, 0.1810, 0.1619, 0.3255, 0.1023, 0.1124, 0.4449, 0.1457], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0179, 0.0162, 0.0131, 0.0163, 0.0124, 0.0148, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 00:06:28,585 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110315.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:06:30,442 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110318.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:06:35,827 INFO [finetune.py:976] (1/7) Epoch 20, batch 1500, loss[loss=0.1543, simple_loss=0.2308, pruned_loss=0.03886, over 4907.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2495, pruned_loss=0.05281, over 953542.12 frames. ], batch size: 36, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:06:47,651 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110344.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:06:55,109 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.117e+02 1.590e+02 1.939e+02 2.244e+02 3.777e+02, threshold=3.878e+02, percent-clipped=2.0 +2023-03-27 00:07:00,487 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=110363.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:07:01,142 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110364.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:07:07,073 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110373.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:07:09,448 INFO [finetune.py:976] (1/7) Epoch 20, batch 1550, loss[loss=0.1466, simple_loss=0.218, pruned_loss=0.03756, over 4930.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2501, pruned_loss=0.05298, over 954132.77 frames. ], batch size: 33, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:07:10,799 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110379.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:07:13,056 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7962, 1.2539, 0.8900, 1.6193, 2.1014, 1.4971, 1.5367, 1.6435], + device='cuda:1'), covar=tensor([0.1546, 0.2124, 0.2117, 0.1247, 0.2074, 0.2082, 0.1501, 0.2036], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0091, 0.0119, 0.0092, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:07:25,532 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110400.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:07:43,127 INFO [finetune.py:976] (1/7) Epoch 20, batch 1600, loss[loss=0.1725, simple_loss=0.2371, pruned_loss=0.05392, over 4861.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2478, pruned_loss=0.05272, over 953078.38 frames. ], batch size: 44, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:08:13,511 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.758e+01 1.502e+02 1.787e+02 2.306e+02 4.709e+02, threshold=3.574e+02, percent-clipped=2.0 +2023-03-27 00:08:29,698 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110466.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:08:33,148 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110471.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:08:40,392 INFO [finetune.py:976] (1/7) Epoch 20, batch 1650, loss[loss=0.1682, simple_loss=0.2359, pruned_loss=0.05018, over 4830.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2453, pruned_loss=0.05223, over 953670.54 frames. ], batch size: 30, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:08:52,649 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6074, 1.5879, 1.3300, 1.6147, 2.0336, 1.9293, 1.6331, 1.4881], + device='cuda:1'), covar=tensor([0.0312, 0.0338, 0.0630, 0.0305, 0.0184, 0.0555, 0.0325, 0.0395], + device='cuda:1'), in_proj_covar=tensor([0.0097, 0.0107, 0.0144, 0.0111, 0.0100, 0.0111, 0.0100, 0.0111], + device='cuda:1'), out_proj_covar=tensor([7.5287e-05, 8.2659e-05, 1.1362e-04, 8.5584e-05, 7.7930e-05, 8.2311e-05, + 7.4095e-05, 8.4972e-05], device='cuda:1') +2023-03-27 00:09:24,056 INFO [finetune.py:976] (1/7) Epoch 20, batch 1700, loss[loss=0.1875, simple_loss=0.2601, pruned_loss=0.05741, over 4936.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2436, pruned_loss=0.0518, over 954490.73 frames. ], batch size: 38, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:09:36,234 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3680, 2.4557, 2.2904, 1.8350, 2.3697, 2.6819, 2.7005, 2.1168], + device='cuda:1'), covar=tensor([0.0638, 0.0583, 0.0732, 0.0872, 0.0739, 0.0657, 0.0539, 0.1007], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0134, 0.0139, 0.0119, 0.0123, 0.0138, 0.0139, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:09:42,518 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.071e+02 1.500e+02 1.773e+02 2.246e+02 3.830e+02, threshold=3.546e+02, percent-clipped=2.0 +2023-03-27 00:09:57,622 INFO [finetune.py:976] (1/7) Epoch 20, batch 1750, loss[loss=0.1333, simple_loss=0.2006, pruned_loss=0.033, over 4089.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2461, pruned_loss=0.0528, over 955405.24 frames. ], batch size: 17, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:10:08,601 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7350, 1.7799, 2.2818, 1.9751, 1.9825, 4.3452, 1.7287, 1.8696], + device='cuda:1'), covar=tensor([0.1174, 0.2151, 0.1325, 0.1189, 0.1636, 0.0220, 0.1773, 0.2170], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 00:10:20,385 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110611.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:10:30,987 INFO [finetune.py:976] (1/7) Epoch 20, batch 1800, loss[loss=0.1985, simple_loss=0.2778, pruned_loss=0.05966, over 4892.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2469, pruned_loss=0.05228, over 955402.99 frames. ], batch size: 35, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:10:38,897 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110639.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:10:39,019 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-27 00:10:55,666 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.008e+02 1.630e+02 1.887e+02 2.220e+02 3.285e+02, threshold=3.774e+02, percent-clipped=0.0 +2023-03-27 00:11:01,587 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110659.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:11:03,357 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3862, 2.2536, 1.7422, 2.4490, 2.2413, 1.9989, 2.6751, 2.3631], + device='cuda:1'), covar=tensor([0.1408, 0.2189, 0.3431, 0.2564, 0.2681, 0.1899, 0.2815, 0.1834], + device='cuda:1'), in_proj_covar=tensor([0.0186, 0.0189, 0.0235, 0.0254, 0.0249, 0.0205, 0.0216, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:11:11,031 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110672.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:11:11,623 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110673.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:11:12,217 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110674.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:11:14,002 INFO [finetune.py:976] (1/7) Epoch 20, batch 1850, loss[loss=0.1673, simple_loss=0.2225, pruned_loss=0.05603, over 3832.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2479, pruned_loss=0.05278, over 953100.34 frames. ], batch size: 16, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:11:29,603 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110700.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:11:44,147 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=110721.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:11:47,772 INFO [finetune.py:976] (1/7) Epoch 20, batch 1900, loss[loss=0.1932, simple_loss=0.2693, pruned_loss=0.05859, over 4924.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2493, pruned_loss=0.05333, over 952671.48 frames. ], batch size: 41, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:11:47,881 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9162, 1.9877, 2.4563, 2.1420, 2.0540, 4.6097, 1.9635, 2.1441], + device='cuda:1'), covar=tensor([0.0886, 0.1579, 0.0974, 0.0902, 0.1435, 0.0199, 0.1304, 0.1558], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 00:11:54,345 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110736.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:12:01,609 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=110748.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:12:06,288 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.024e+02 1.472e+02 1.826e+02 2.198e+02 3.929e+02, threshold=3.651e+02, percent-clipped=1.0 +2023-03-27 00:12:14,060 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110766.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:12:17,569 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110771.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:12:21,622 INFO [finetune.py:976] (1/7) Epoch 20, batch 1950, loss[loss=0.1925, simple_loss=0.2525, pruned_loss=0.06626, over 4718.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2484, pruned_loss=0.05272, over 954961.61 frames. ], batch size: 59, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:12:34,793 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110797.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 00:12:46,182 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=110814.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:12:49,722 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=110819.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:12:55,561 INFO [finetune.py:976] (1/7) Epoch 20, batch 2000, loss[loss=0.2144, simple_loss=0.2712, pruned_loss=0.07882, over 4824.00 frames. ], tot_loss[loss=0.1746, simple_loss=0.2458, pruned_loss=0.05168, over 957215.69 frames. ], batch size: 30, lr: 3.25e-03, grad_scale: 32.0 +2023-03-27 00:13:13,700 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-27 00:13:17,583 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.033e+02 1.482e+02 1.740e+02 2.016e+02 2.901e+02, threshold=3.480e+02, percent-clipped=0.0 +2023-03-27 00:13:30,387 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110868.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:13:38,673 INFO [finetune.py:976] (1/7) Epoch 20, batch 2050, loss[loss=0.1596, simple_loss=0.2374, pruned_loss=0.04089, over 4872.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2432, pruned_loss=0.05099, over 955849.77 frames. ], batch size: 31, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:13:54,501 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6524, 0.7320, 1.6975, 1.5872, 1.4902, 1.4081, 1.5357, 1.5549], + device='cuda:1'), covar=tensor([0.3239, 0.3417, 0.2962, 0.3191, 0.3975, 0.3105, 0.3697, 0.2758], + device='cuda:1'), in_proj_covar=tensor([0.0254, 0.0242, 0.0263, 0.0282, 0.0280, 0.0255, 0.0290, 0.0244], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:14:28,244 INFO [finetune.py:976] (1/7) Epoch 20, batch 2100, loss[loss=0.1616, simple_loss=0.2239, pruned_loss=0.04962, over 4713.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2442, pruned_loss=0.05195, over 953638.35 frames. ], batch size: 23, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:14:29,606 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110929.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:14:39,411 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1464, 2.0284, 1.6317, 2.0020, 2.0490, 1.8099, 2.3369, 2.1588], + device='cuda:1'), covar=tensor([0.1350, 0.2075, 0.3094, 0.2576, 0.2526, 0.1715, 0.3045, 0.1807], + device='cuda:1'), in_proj_covar=tensor([0.0186, 0.0190, 0.0236, 0.0254, 0.0249, 0.0205, 0.0216, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:14:39,985 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110939.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:14:50,636 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.458e+01 1.520e+02 1.862e+02 2.217e+02 3.516e+02, threshold=3.725e+02, percent-clipped=1.0 +2023-03-27 00:14:52,569 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110958.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:14:53,162 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110959.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:14:58,413 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110967.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:15:03,705 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110974.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:15:05,918 INFO [finetune.py:976] (1/7) Epoch 20, batch 2150, loss[loss=0.2171, simple_loss=0.2703, pruned_loss=0.08189, over 4768.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2474, pruned_loss=0.05325, over 953667.85 frames. ], batch size: 54, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:15:12,589 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=110987.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:15:25,800 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=111007.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:15:28,985 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-27 00:15:33,649 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111019.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:15:35,879 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=111022.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:15:38,858 INFO [finetune.py:976] (1/7) Epoch 20, batch 2200, loss[loss=0.2214, simple_loss=0.2877, pruned_loss=0.07752, over 4826.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2505, pruned_loss=0.0542, over 953824.54 frames. ], batch size: 39, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:16:00,234 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.032e+02 1.523e+02 1.854e+02 2.195e+02 4.707e+02, threshold=3.708e+02, percent-clipped=2.0 +2023-03-27 00:16:20,242 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5965, 2.4204, 1.9042, 0.9407, 2.0520, 1.9866, 1.8542, 2.1527], + device='cuda:1'), covar=tensor([0.0771, 0.0709, 0.1790, 0.2207, 0.1453, 0.2262, 0.2154, 0.0969], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0192, 0.0201, 0.0183, 0.0211, 0.0208, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:16:23,019 INFO [finetune.py:976] (1/7) Epoch 20, batch 2250, loss[loss=0.2052, simple_loss=0.2818, pruned_loss=0.06433, over 4820.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2519, pruned_loss=0.05483, over 953600.69 frames. ], batch size: 51, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:16:29,617 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6721, 1.6819, 1.4511, 1.8152, 2.1055, 1.8568, 1.5344, 1.3602], + device='cuda:1'), covar=tensor([0.2289, 0.2020, 0.2021, 0.1681, 0.1917, 0.1242, 0.2567, 0.2078], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0211, 0.0194, 0.0243, 0.0188, 0.0216, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:16:33,689 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=111092.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 00:16:51,321 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.12 vs. limit=5.0 +2023-03-27 00:16:56,459 INFO [finetune.py:976] (1/7) Epoch 20, batch 2300, loss[loss=0.1994, simple_loss=0.2747, pruned_loss=0.06204, over 4850.00 frames. ], tot_loss[loss=0.1801, simple_loss=0.2515, pruned_loss=0.05435, over 952104.68 frames. ], batch size: 49, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:17:15,905 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.510e+02 1.791e+02 2.077e+02 4.254e+02, threshold=3.582e+02, percent-clipped=2.0 +2023-03-27 00:17:30,234 INFO [finetune.py:976] (1/7) Epoch 20, batch 2350, loss[loss=0.1878, simple_loss=0.2556, pruned_loss=0.06006, over 4820.00 frames. ], tot_loss[loss=0.1773, simple_loss=0.2482, pruned_loss=0.05325, over 950600.97 frames. ], batch size: 39, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:18:01,642 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=111224.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:18:02,304 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111225.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:18:03,428 INFO [finetune.py:976] (1/7) Epoch 20, batch 2400, loss[loss=0.1838, simple_loss=0.2487, pruned_loss=0.05941, over 4802.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2463, pruned_loss=0.05333, over 952325.48 frames. ], batch size: 25, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:18:22,740 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.826e+01 1.548e+02 1.775e+02 2.063e+02 3.363e+02, threshold=3.550e+02, percent-clipped=0.0 +2023-03-27 00:18:30,958 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111267.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:18:36,922 INFO [finetune.py:976] (1/7) Epoch 20, batch 2450, loss[loss=0.147, simple_loss=0.2202, pruned_loss=0.03692, over 4821.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2438, pruned_loss=0.05209, over 953010.62 frames. ], batch size: 25, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:18:50,035 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111286.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:19:21,758 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=111314.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:19:22,348 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=111315.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:19:33,325 INFO [finetune.py:976] (1/7) Epoch 20, batch 2500, loss[loss=0.1722, simple_loss=0.2477, pruned_loss=0.04832, over 4836.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.247, pruned_loss=0.05392, over 953763.33 frames. ], batch size: 39, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:19:48,180 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5930, 1.5833, 1.8406, 1.2354, 1.6235, 1.9424, 1.5361, 1.9658], + device='cuda:1'), covar=tensor([0.1076, 0.1828, 0.1113, 0.1400, 0.0796, 0.0880, 0.2376, 0.0706], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0204, 0.0190, 0.0188, 0.0173, 0.0212, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:20:03,161 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.143e+02 1.591e+02 1.841e+02 2.119e+02 4.112e+02, threshold=3.683e+02, percent-clipped=1.0 +2023-03-27 00:20:17,414 INFO [finetune.py:976] (1/7) Epoch 20, batch 2550, loss[loss=0.1959, simple_loss=0.2738, pruned_loss=0.05896, over 4906.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2486, pruned_loss=0.05423, over 951741.83 frames. ], batch size: 43, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:20:27,521 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111392.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 00:20:43,048 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3715, 3.7415, 4.0307, 4.2269, 4.1251, 3.8668, 4.4688, 1.4168], + device='cuda:1'), covar=tensor([0.0886, 0.0993, 0.0883, 0.1189, 0.1364, 0.1690, 0.0729, 0.5881], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0246, 0.0278, 0.0292, 0.0334, 0.0283, 0.0303, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:20:51,266 INFO [finetune.py:976] (1/7) Epoch 20, batch 2600, loss[loss=0.1444, simple_loss=0.1955, pruned_loss=0.0466, over 4080.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2501, pruned_loss=0.05467, over 952686.04 frames. ], batch size: 17, lr: 3.24e-03, grad_scale: 64.0 +2023-03-27 00:20:51,547 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.35 vs. limit=5.0 +2023-03-27 00:20:52,686 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.48 vs. limit=5.0 +2023-03-27 00:20:59,723 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=111440.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:21:10,161 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.233e+02 1.591e+02 1.878e+02 2.220e+02 5.233e+02, threshold=3.757e+02, percent-clipped=1.0 +2023-03-27 00:21:20,883 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111468.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:21:31,397 INFO [finetune.py:976] (1/7) Epoch 20, batch 2650, loss[loss=0.24, simple_loss=0.3185, pruned_loss=0.08077, over 4808.00 frames. ], tot_loss[loss=0.1805, simple_loss=0.2511, pruned_loss=0.05493, over 953155.21 frames. ], batch size: 45, lr: 3.24e-03, grad_scale: 64.0 +2023-03-27 00:22:06,255 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111524.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:22:07,975 INFO [finetune.py:976] (1/7) Epoch 20, batch 2700, loss[loss=0.2046, simple_loss=0.2806, pruned_loss=0.06427, over 4885.00 frames. ], tot_loss[loss=0.1813, simple_loss=0.2518, pruned_loss=0.0554, over 950972.95 frames. ], batch size: 43, lr: 3.24e-03, grad_scale: 64.0 +2023-03-27 00:22:09,760 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111529.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:22:20,473 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.96 vs. limit=5.0 +2023-03-27 00:22:27,297 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.466e+01 1.581e+02 1.832e+02 2.307e+02 3.346e+02, threshold=3.664e+02, percent-clipped=0.0 +2023-03-27 00:22:38,121 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=111572.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:22:41,154 INFO [finetune.py:976] (1/7) Epoch 20, batch 2750, loss[loss=0.1598, simple_loss=0.2303, pruned_loss=0.04464, over 4903.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2498, pruned_loss=0.05489, over 952994.68 frames. ], batch size: 35, lr: 3.24e-03, grad_scale: 64.0 +2023-03-27 00:22:44,064 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=111581.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:22:46,507 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7172, 2.6218, 2.4550, 2.1211, 2.7539, 2.8753, 3.0841, 2.4007], + device='cuda:1'), covar=tensor([0.0596, 0.0647, 0.0742, 0.0811, 0.0642, 0.0626, 0.0524, 0.0981], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0136, 0.0140, 0.0120, 0.0124, 0.0139, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:23:06,506 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111614.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:23:14,366 INFO [finetune.py:976] (1/7) Epoch 20, batch 2800, loss[loss=0.1864, simple_loss=0.2546, pruned_loss=0.05905, over 4842.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.2469, pruned_loss=0.05429, over 953782.70 frames. ], batch size: 44, lr: 3.24e-03, grad_scale: 64.0 +2023-03-27 00:23:32,755 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.029e+01 1.565e+02 1.748e+02 2.177e+02 3.583e+02, threshold=3.496e+02, percent-clipped=0.0 +2023-03-27 00:23:38,053 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=111662.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:23:48,042 INFO [finetune.py:976] (1/7) Epoch 20, batch 2850, loss[loss=0.2075, simple_loss=0.2655, pruned_loss=0.07474, over 4873.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2451, pruned_loss=0.05364, over 955145.36 frames. ], batch size: 34, lr: 3.24e-03, grad_scale: 64.0 +2023-03-27 00:24:41,627 INFO [finetune.py:976] (1/7) Epoch 20, batch 2900, loss[loss=0.2008, simple_loss=0.2896, pruned_loss=0.05595, over 4826.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2479, pruned_loss=0.05449, over 955632.73 frames. ], batch size: 33, lr: 3.24e-03, grad_scale: 64.0 +2023-03-27 00:24:41,771 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3291, 2.3739, 1.8535, 2.4871, 2.3737, 2.0184, 2.8576, 2.4266], + device='cuda:1'), covar=tensor([0.1354, 0.2250, 0.3060, 0.2751, 0.2531, 0.1661, 0.2844, 0.1784], + device='cuda:1'), in_proj_covar=tensor([0.0184, 0.0188, 0.0233, 0.0252, 0.0246, 0.0203, 0.0214, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:24:49,102 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7611, 1.6158, 1.6200, 1.7203, 1.3891, 3.6605, 1.4511, 1.9407], + device='cuda:1'), covar=tensor([0.3381, 0.2569, 0.2173, 0.2440, 0.1701, 0.0193, 0.2690, 0.1255], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0120, 0.0124, 0.0114, 0.0097, 0.0096, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 00:25:12,980 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.101e+02 1.542e+02 1.820e+02 2.254e+02 5.949e+02, threshold=3.641e+02, percent-clipped=1.0 +2023-03-27 00:25:28,523 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4423, 2.3148, 1.8589, 2.3077, 2.2945, 2.0202, 2.6334, 2.4692], + device='cuda:1'), covar=tensor([0.1344, 0.2194, 0.3163, 0.2837, 0.2703, 0.1806, 0.3246, 0.1819], + device='cuda:1'), in_proj_covar=tensor([0.0184, 0.0188, 0.0233, 0.0252, 0.0246, 0.0203, 0.0213, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:25:31,431 INFO [finetune.py:976] (1/7) Epoch 20, batch 2950, loss[loss=0.2, simple_loss=0.268, pruned_loss=0.06597, over 4792.00 frames. ], tot_loss[loss=0.1803, simple_loss=0.2511, pruned_loss=0.0548, over 955815.88 frames. ], batch size: 25, lr: 3.24e-03, grad_scale: 64.0 +2023-03-27 00:25:40,059 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111791.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 00:26:03,392 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=111824.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:26:05,121 INFO [finetune.py:976] (1/7) Epoch 20, batch 3000, loss[loss=0.1822, simple_loss=0.258, pruned_loss=0.05318, over 4824.00 frames. ], tot_loss[loss=0.1808, simple_loss=0.2519, pruned_loss=0.05484, over 953875.27 frames. ], batch size: 30, lr: 3.24e-03, grad_scale: 64.0 +2023-03-27 00:26:05,121 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 00:26:20,304 INFO [finetune.py:1010] (1/7) Epoch 20, validation: loss=0.1563, simple_loss=0.2257, pruned_loss=0.04344, over 2265189.00 frames. +2023-03-27 00:26:20,305 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 00:26:31,084 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4580, 1.1523, 0.7423, 1.3152, 1.9117, 0.8167, 1.2935, 1.3983], + device='cuda:1'), covar=tensor([0.1485, 0.1954, 0.1802, 0.1191, 0.1918, 0.1842, 0.1379, 0.1837], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0091, 0.0119, 0.0093, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:26:31,802 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.55 vs. limit=5.0 +2023-03-27 00:26:38,926 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6977, 1.6228, 2.0022, 1.3316, 1.8891, 1.9823, 1.4815, 2.2070], + device='cuda:1'), covar=tensor([0.1468, 0.2129, 0.1501, 0.2167, 0.0960, 0.1614, 0.2914, 0.0873], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0206, 0.0193, 0.0191, 0.0176, 0.0214, 0.0220, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:26:40,747 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111852.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 00:26:48,436 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.202e+02 1.618e+02 1.875e+02 2.230e+02 3.575e+02, threshold=3.749e+02, percent-clipped=0.0 +2023-03-27 00:27:07,143 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7887, 1.6687, 2.2758, 2.9970, 2.1245, 2.2761, 1.5906, 2.4646], + device='cuda:1'), covar=tensor([0.1407, 0.1182, 0.0946, 0.0543, 0.0721, 0.1985, 0.1203, 0.0489], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0162, 0.0100, 0.0134, 0.0123, 0.0098], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:27:11,151 INFO [finetune.py:976] (1/7) Epoch 20, batch 3050, loss[loss=0.1638, simple_loss=0.2284, pruned_loss=0.04962, over 4824.00 frames. ], tot_loss[loss=0.1812, simple_loss=0.2529, pruned_loss=0.05482, over 955448.42 frames. ], batch size: 25, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:27:18,696 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111881.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:27:37,345 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.2245, 4.4322, 4.7525, 5.0564, 4.9296, 4.6522, 5.3708, 1.5979], + device='cuda:1'), covar=tensor([0.0754, 0.0772, 0.0710, 0.0839, 0.1175, 0.1592, 0.0501, 0.5930], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0279, 0.0293, 0.0334, 0.0284, 0.0305, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:28:14,418 INFO [finetune.py:976] (1/7) Epoch 20, batch 3100, loss[loss=0.1626, simple_loss=0.2328, pruned_loss=0.04621, over 4789.00 frames. ], tot_loss[loss=0.179, simple_loss=0.2503, pruned_loss=0.05386, over 957183.83 frames. ], batch size: 29, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:28:15,692 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=111929.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:28:33,002 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.135e+02 1.507e+02 1.737e+02 2.301e+02 5.151e+02, threshold=3.474e+02, percent-clipped=3.0 +2023-03-27 00:28:40,283 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5622, 1.5515, 1.6617, 0.7838, 1.7047, 1.9151, 1.8210, 1.4417], + device='cuda:1'), covar=tensor([0.1129, 0.0816, 0.0454, 0.0731, 0.0480, 0.0606, 0.0400, 0.0770], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0149, 0.0124, 0.0124, 0.0129, 0.0128, 0.0140, 0.0147], + device='cuda:1'), out_proj_covar=tensor([9.0218e-05, 1.0817e-04, 8.8433e-05, 8.7493e-05, 9.0917e-05, 9.1848e-05, + 1.0056e-04, 1.0525e-04], device='cuda:1') +2023-03-27 00:28:47,144 INFO [finetune.py:976] (1/7) Epoch 20, batch 3150, loss[loss=0.1469, simple_loss=0.2217, pruned_loss=0.0361, over 4833.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2472, pruned_loss=0.05349, over 955385.69 frames. ], batch size: 33, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:28:57,253 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5302, 2.4482, 2.1144, 1.0238, 2.2400, 1.9521, 1.8002, 2.1956], + device='cuda:1'), covar=tensor([0.0976, 0.0728, 0.1760, 0.2198, 0.1545, 0.2330, 0.2456, 0.1050], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0192, 0.0200, 0.0183, 0.0212, 0.0207, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:29:07,811 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.33 vs. limit=5.0 +2023-03-27 00:29:23,072 INFO [finetune.py:976] (1/7) Epoch 20, batch 3200, loss[loss=0.2233, simple_loss=0.2827, pruned_loss=0.08191, over 4743.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2435, pruned_loss=0.05185, over 954962.20 frames. ], batch size: 59, lr: 3.24e-03, grad_scale: 32.0 +2023-03-27 00:29:43,188 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2603, 2.1218, 2.0237, 2.2854, 2.9150, 2.2418, 2.3650, 1.7401], + device='cuda:1'), covar=tensor([0.2530, 0.2266, 0.2216, 0.1910, 0.1910, 0.1399, 0.2247, 0.2361], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0211, 0.0213, 0.0194, 0.0244, 0.0189, 0.0218, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:29:53,391 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0260, 1.3803, 0.7064, 1.8164, 2.3192, 1.8115, 1.5630, 1.8015], + device='cuda:1'), covar=tensor([0.1761, 0.2617, 0.2642, 0.1507, 0.2149, 0.2587, 0.1898, 0.2465], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0091, 0.0119, 0.0092, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:29:56,972 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.705e+01 1.565e+02 1.739e+02 2.228e+02 3.922e+02, threshold=3.479e+02, percent-clipped=1.0 +2023-03-27 00:30:06,300 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=112066.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:30:13,834 INFO [finetune.py:976] (1/7) Epoch 20, batch 3250, loss[loss=0.1456, simple_loss=0.2279, pruned_loss=0.03166, over 4751.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2444, pruned_loss=0.05232, over 952749.29 frames. ], batch size: 27, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:30:15,744 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9818, 1.8003, 1.5941, 1.6805, 1.7417, 1.7542, 1.8241, 2.4787], + device='cuda:1'), covar=tensor([0.3501, 0.4177, 0.2917, 0.3454, 0.3897, 0.2343, 0.3548, 0.1555], + device='cuda:1'), in_proj_covar=tensor([0.0285, 0.0261, 0.0230, 0.0275, 0.0251, 0.0221, 0.0250, 0.0231], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:30:48,108 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3555, 2.2793, 2.4189, 1.0721, 2.7078, 2.9689, 2.5633, 2.0216], + device='cuda:1'), covar=tensor([0.1166, 0.0854, 0.0548, 0.0789, 0.0766, 0.0625, 0.0476, 0.0921], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0151, 0.0125, 0.0125, 0.0131, 0.0129, 0.0142, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.0881e-05, 1.0893e-04, 8.9349e-05, 8.8287e-05, 9.2029e-05, 9.2592e-05, + 1.0155e-04, 1.0634e-04], device='cuda:1') +2023-03-27 00:30:54,490 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=112124.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:30:56,192 INFO [finetune.py:976] (1/7) Epoch 20, batch 3300, loss[loss=0.199, simple_loss=0.2682, pruned_loss=0.06492, over 4764.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2477, pruned_loss=0.05358, over 951310.48 frames. ], batch size: 28, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:30:56,312 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=112127.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:31:09,775 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112147.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 00:31:16,114 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.319e+01 1.653e+02 1.982e+02 2.472e+02 3.934e+02, threshold=3.965e+02, percent-clipped=2.0 +2023-03-27 00:31:26,466 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=112172.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:31:29,977 INFO [finetune.py:976] (1/7) Epoch 20, batch 3350, loss[loss=0.2027, simple_loss=0.2583, pruned_loss=0.07355, over 4924.00 frames. ], tot_loss[loss=0.1793, simple_loss=0.2499, pruned_loss=0.05434, over 952283.88 frames. ], batch size: 33, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:32:11,815 INFO [finetune.py:976] (1/7) Epoch 20, batch 3400, loss[loss=0.2159, simple_loss=0.2807, pruned_loss=0.07558, over 4909.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2505, pruned_loss=0.05456, over 952344.82 frames. ], batch size: 33, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:32:28,440 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9163, 2.6141, 2.5153, 1.3756, 2.5693, 2.0609, 1.9753, 2.3882], + device='cuda:1'), covar=tensor([0.1162, 0.0810, 0.1667, 0.1969, 0.1717, 0.2432, 0.2074, 0.1083], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0192, 0.0200, 0.0182, 0.0211, 0.0207, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:32:29,020 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5910, 1.4883, 2.0062, 3.1515, 2.0643, 2.2952, 1.0618, 2.6366], + device='cuda:1'), covar=tensor([0.1725, 0.1427, 0.1267, 0.0607, 0.0857, 0.1267, 0.1767, 0.0463], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0163, 0.0100, 0.0134, 0.0123, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:32:31,199 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.110e+02 1.676e+02 1.964e+02 2.392e+02 4.564e+02, threshold=3.928e+02, percent-clipped=2.0 +2023-03-27 00:32:31,350 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9169, 1.9404, 1.7311, 2.1323, 2.3226, 2.1241, 1.7218, 1.5772], + device='cuda:1'), covar=tensor([0.2371, 0.2020, 0.2137, 0.1829, 0.1819, 0.1184, 0.2401, 0.2177], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0211, 0.0213, 0.0195, 0.0244, 0.0189, 0.0218, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:32:34,358 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4338, 1.4210, 1.2291, 1.4336, 1.7639, 1.6642, 1.4947, 1.2797], + device='cuda:1'), covar=tensor([0.0335, 0.0299, 0.0633, 0.0282, 0.0206, 0.0434, 0.0325, 0.0398], + device='cuda:1'), in_proj_covar=tensor([0.0097, 0.0108, 0.0145, 0.0112, 0.0101, 0.0112, 0.0100, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.5499e-05, 8.2920e-05, 1.1431e-04, 8.5745e-05, 7.8383e-05, 8.2965e-05, + 7.4686e-05, 8.6228e-05], device='cuda:1') +2023-03-27 00:32:44,378 INFO [finetune.py:976] (1/7) Epoch 20, batch 3450, loss[loss=0.1454, simple_loss=0.2275, pruned_loss=0.03171, over 4811.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2507, pruned_loss=0.05445, over 953871.59 frames. ], batch size: 40, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:33:19,430 INFO [finetune.py:976] (1/7) Epoch 20, batch 3500, loss[loss=0.1752, simple_loss=0.2424, pruned_loss=0.05406, over 4854.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2484, pruned_loss=0.05384, over 953439.34 frames. ], batch size: 49, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:33:29,339 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4746, 3.4248, 3.1953, 1.5780, 3.4443, 2.6112, 0.7840, 2.4655], + device='cuda:1'), covar=tensor([0.2488, 0.1887, 0.1797, 0.3576, 0.1371, 0.1125, 0.4563, 0.1606], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0177, 0.0160, 0.0129, 0.0162, 0.0122, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 00:33:56,443 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.856e+01 1.454e+02 1.799e+02 2.150e+02 5.052e+02, threshold=3.598e+02, percent-clipped=2.0 +2023-03-27 00:34:18,957 INFO [finetune.py:976] (1/7) Epoch 20, batch 3550, loss[loss=0.1511, simple_loss=0.2352, pruned_loss=0.03354, over 4833.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2447, pruned_loss=0.05257, over 952226.35 frames. ], batch size: 40, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:34:28,788 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3568, 1.3473, 1.6060, 1.6065, 1.5060, 2.9716, 1.2748, 1.5213], + device='cuda:1'), covar=tensor([0.0987, 0.1911, 0.1080, 0.0970, 0.1724, 0.0313, 0.1588, 0.1882], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0075, 0.0077, 0.0092, 0.0081, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 00:34:36,592 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=112389.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:35:17,935 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112422.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:35:20,890 INFO [finetune.py:976] (1/7) Epoch 20, batch 3600, loss[loss=0.1352, simple_loss=0.2095, pruned_loss=0.03046, over 4864.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2432, pruned_loss=0.05206, over 953513.68 frames. ], batch size: 31, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:35:32,159 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.70 vs. limit=2.0 +2023-03-27 00:35:37,949 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=112447.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 00:35:39,829 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=112450.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:35:44,343 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.087e+02 1.527e+02 1.833e+02 2.137e+02 4.874e+02, threshold=3.667e+02, percent-clipped=1.0 +2023-03-27 00:36:08,332 INFO [finetune.py:976] (1/7) Epoch 20, batch 3650, loss[loss=0.1437, simple_loss=0.2103, pruned_loss=0.03851, over 4761.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2448, pruned_loss=0.05257, over 954237.82 frames. ], batch size: 26, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:36:19,884 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0724, 1.9715, 2.1943, 1.5490, 2.1061, 2.3223, 1.7946, 2.4605], + device='cuda:1'), covar=tensor([0.1254, 0.1867, 0.1467, 0.1906, 0.0947, 0.1350, 0.2864, 0.0785], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0205, 0.0191, 0.0190, 0.0175, 0.0212, 0.0219, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:36:20,406 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=112495.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 00:36:41,250 INFO [finetune.py:976] (1/7) Epoch 20, batch 3700, loss[loss=0.151, simple_loss=0.2258, pruned_loss=0.03812, over 4770.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2474, pruned_loss=0.05317, over 953725.90 frames. ], batch size: 28, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:37:01,299 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.006e+02 1.568e+02 1.941e+02 2.323e+02 3.962e+02, threshold=3.882e+02, percent-clipped=1.0 +2023-03-27 00:37:05,587 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=112562.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:37:15,914 INFO [finetune.py:976] (1/7) Epoch 20, batch 3750, loss[loss=0.1923, simple_loss=0.2709, pruned_loss=0.05688, over 4811.00 frames. ], tot_loss[loss=0.1793, simple_loss=0.2501, pruned_loss=0.0543, over 956492.62 frames. ], batch size: 40, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:37:16,030 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=112577.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:37:39,043 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-27 00:37:50,094 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=112617.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:37:54,225 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=112623.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:37:57,417 INFO [finetune.py:976] (1/7) Epoch 20, batch 3800, loss[loss=0.1681, simple_loss=0.2501, pruned_loss=0.04302, over 4789.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2503, pruned_loss=0.05379, over 955449.03 frames. ], batch size: 29, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:38:04,240 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=112638.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 00:38:06,655 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3695, 1.2761, 1.5284, 2.4230, 1.5612, 2.2579, 0.8817, 2.0978], + device='cuda:1'), covar=tensor([0.1986, 0.1787, 0.1523, 0.1153, 0.1111, 0.1355, 0.1876, 0.0772], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0164, 0.0100, 0.0135, 0.0124, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:38:16,043 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.507e+01 1.515e+02 1.800e+02 2.233e+02 3.828e+02, threshold=3.600e+02, percent-clipped=0.0 +2023-03-27 00:38:16,846 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.09 vs. limit=5.0 +2023-03-27 00:38:30,668 INFO [finetune.py:976] (1/7) Epoch 20, batch 3850, loss[loss=0.203, simple_loss=0.2765, pruned_loss=0.06476, over 4920.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2484, pruned_loss=0.05287, over 956314.16 frames. ], batch size: 42, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:38:31,383 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=112678.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 00:38:39,040 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3808, 2.9415, 2.8240, 1.3846, 3.0211, 2.2389, 0.7279, 1.9465], + device='cuda:1'), covar=tensor([0.2129, 0.2098, 0.1706, 0.3444, 0.1445, 0.1239, 0.4097, 0.1666], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0161, 0.0130, 0.0162, 0.0124, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 00:38:59,757 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=112722.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:39:03,154 INFO [finetune.py:976] (1/7) Epoch 20, batch 3900, loss[loss=0.1712, simple_loss=0.2432, pruned_loss=0.0496, over 4861.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2467, pruned_loss=0.05304, over 957280.83 frames. ], batch size: 34, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:39:15,042 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112745.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:39:21,552 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.403e+01 1.612e+02 1.959e+02 2.410e+02 4.123e+02, threshold=3.918e+02, percent-clipped=1.0 +2023-03-27 00:39:29,352 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3883, 1.4138, 1.7548, 1.6422, 1.5135, 3.3221, 1.3592, 1.6178], + device='cuda:1'), covar=tensor([0.0963, 0.1712, 0.1069, 0.0988, 0.1606, 0.0245, 0.1448, 0.1656], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0075, 0.0077, 0.0092, 0.0081, 0.0086, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 00:39:32,077 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=112770.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:39:36,267 INFO [finetune.py:976] (1/7) Epoch 20, batch 3950, loss[loss=0.1585, simple_loss=0.2256, pruned_loss=0.04566, over 4803.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.244, pruned_loss=0.0522, over 958645.70 frames. ], batch size: 29, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:39:40,503 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.13 vs. limit=5.0 +2023-03-27 00:39:43,488 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-27 00:40:09,283 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.81 vs. limit=5.0 +2023-03-27 00:40:19,039 INFO [finetune.py:976] (1/7) Epoch 20, batch 4000, loss[loss=0.1984, simple_loss=0.2619, pruned_loss=0.0674, over 4824.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2428, pruned_loss=0.0517, over 960252.56 frames. ], batch size: 51, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:40:22,653 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2399, 2.0858, 1.7494, 2.0231, 2.1656, 1.9285, 2.4571, 2.2503], + device='cuda:1'), covar=tensor([0.1340, 0.2147, 0.3029, 0.2558, 0.2566, 0.1667, 0.2787, 0.1695], + device='cuda:1'), in_proj_covar=tensor([0.0185, 0.0188, 0.0235, 0.0252, 0.0246, 0.0203, 0.0215, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:40:48,437 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.088e+02 1.638e+02 1.904e+02 2.320e+02 3.891e+02, threshold=3.808e+02, percent-clipped=0.0 +2023-03-27 00:40:49,225 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1029, 1.7407, 2.1030, 2.1037, 1.8662, 1.8456, 2.0866, 1.9668], + device='cuda:1'), covar=tensor([0.4099, 0.4035, 0.3103, 0.3888, 0.4713, 0.3800, 0.4563, 0.2997], + device='cuda:1'), in_proj_covar=tensor([0.0252, 0.0241, 0.0262, 0.0280, 0.0278, 0.0253, 0.0288, 0.0243], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:40:55,068 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6615, 1.5800, 1.5717, 1.6431, 1.3281, 4.2267, 1.6088, 1.8958], + device='cuda:1'), covar=tensor([0.3304, 0.2582, 0.2114, 0.2334, 0.1650, 0.0111, 0.2632, 0.1236], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0123, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 00:41:04,948 INFO [finetune.py:976] (1/7) Epoch 20, batch 4050, loss[loss=0.2821, simple_loss=0.3146, pruned_loss=0.1248, over 4051.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2472, pruned_loss=0.05402, over 959712.86 frames. ], batch size: 65, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:41:36,670 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3289, 1.3879, 1.6199, 1.0949, 1.3518, 1.5241, 1.3512, 1.6404], + device='cuda:1'), covar=tensor([0.1236, 0.2135, 0.1270, 0.1625, 0.1056, 0.1253, 0.3019, 0.0860], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0205, 0.0191, 0.0190, 0.0174, 0.0213, 0.0219, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:41:41,242 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112918.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:41:43,107 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=112921.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:41:47,212 INFO [finetune.py:976] (1/7) Epoch 20, batch 4100, loss[loss=0.1735, simple_loss=0.2581, pruned_loss=0.04447, over 4815.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2504, pruned_loss=0.05461, over 959981.14 frames. ], batch size: 38, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:41:51,358 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112933.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 00:42:06,080 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.570e+02 1.869e+02 2.355e+02 4.214e+02, threshold=3.739e+02, percent-clipped=0.0 +2023-03-27 00:42:09,473 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-03-27 00:42:14,620 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2535, 1.3500, 1.5142, 1.0395, 1.3385, 1.4729, 1.2876, 1.5911], + device='cuda:1'), covar=tensor([0.1215, 0.2100, 0.1440, 0.1683, 0.0930, 0.1235, 0.2958, 0.0876], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0204, 0.0190, 0.0189, 0.0173, 0.0212, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:42:17,469 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112973.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 00:42:19,818 INFO [finetune.py:976] (1/7) Epoch 20, batch 4150, loss[loss=0.1848, simple_loss=0.259, pruned_loss=0.05532, over 4819.00 frames. ], tot_loss[loss=0.18, simple_loss=0.2508, pruned_loss=0.0546, over 958221.95 frames. ], batch size: 30, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:42:23,996 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=112982.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:42:42,193 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5670, 1.4414, 1.2736, 1.5880, 1.4847, 1.5992, 0.9244, 1.3395], + device='cuda:1'), covar=tensor([0.2094, 0.1947, 0.1836, 0.1480, 0.1568, 0.1163, 0.2356, 0.1714], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0209, 0.0211, 0.0194, 0.0242, 0.0187, 0.0216, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:43:03,967 INFO [finetune.py:976] (1/7) Epoch 20, batch 4200, loss[loss=0.1816, simple_loss=0.2539, pruned_loss=0.05468, over 4811.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2516, pruned_loss=0.05488, over 956259.23 frames. ], batch size: 38, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:43:16,380 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113045.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:43:17,501 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9057, 1.7732, 1.9158, 1.3036, 1.8734, 2.0196, 2.0246, 1.5447], + device='cuda:1'), covar=tensor([0.0507, 0.0638, 0.0596, 0.0814, 0.0727, 0.0549, 0.0466, 0.1089], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0135, 0.0138, 0.0119, 0.0124, 0.0138, 0.0139, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:43:23,432 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.027e+02 1.578e+02 1.900e+02 2.258e+02 5.235e+02, threshold=3.800e+02, percent-clipped=4.0 +2023-03-27 00:43:36,991 INFO [finetune.py:976] (1/7) Epoch 20, batch 4250, loss[loss=0.193, simple_loss=0.2535, pruned_loss=0.06618, over 4905.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.2499, pruned_loss=0.05492, over 953358.62 frames. ], batch size: 32, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:43:47,715 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=113093.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:43:49,045 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0949, 1.9246, 1.7293, 1.8117, 1.8668, 1.8652, 1.9298, 2.5907], + device='cuda:1'), covar=tensor([0.3706, 0.4214, 0.3124, 0.3743, 0.3850, 0.2531, 0.3807, 0.1619], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0231, 0.0277, 0.0253, 0.0222, 0.0252, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:44:10,286 INFO [finetune.py:976] (1/7) Epoch 20, batch 4300, loss[loss=0.1415, simple_loss=0.204, pruned_loss=0.0395, over 4827.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2472, pruned_loss=0.05441, over 952914.30 frames. ], batch size: 25, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:44:19,119 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7672, 3.3828, 3.5594, 3.4471, 3.3415, 3.2176, 3.8130, 1.3069], + device='cuda:1'), covar=tensor([0.1373, 0.1685, 0.1755, 0.1936, 0.2197, 0.2767, 0.1537, 0.7422], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0243, 0.0278, 0.0292, 0.0331, 0.0283, 0.0304, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:44:20,380 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6041, 2.3307, 1.7995, 0.8211, 1.9112, 2.0960, 1.9663, 2.0739], + device='cuda:1'), covar=tensor([0.0720, 0.0714, 0.1451, 0.2133, 0.1401, 0.2382, 0.1916, 0.0825], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0191, 0.0199, 0.0182, 0.0210, 0.0208, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:44:30,855 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.679e+01 1.396e+02 1.741e+02 2.023e+02 4.034e+02, threshold=3.482e+02, percent-clipped=1.0 +2023-03-27 00:44:35,351 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-03-27 00:44:43,635 INFO [finetune.py:976] (1/7) Epoch 20, batch 4350, loss[loss=0.1118, simple_loss=0.185, pruned_loss=0.0193, over 4812.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2429, pruned_loss=0.05239, over 955580.78 frames. ], batch size: 25, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:45:12,192 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113218.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:45:17,616 INFO [finetune.py:976] (1/7) Epoch 20, batch 4400, loss[loss=0.1484, simple_loss=0.207, pruned_loss=0.04495, over 4122.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2437, pruned_loss=0.0526, over 955272.50 frames. ], batch size: 17, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:45:21,852 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113233.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:45:22,466 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=113234.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:45:49,232 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.729e+01 1.603e+02 1.846e+02 2.173e+02 5.642e+02, threshold=3.692e+02, percent-clipped=4.0 +2023-03-27 00:46:00,743 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=113266.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:46:08,354 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113273.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 00:46:09,611 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5100, 2.4653, 2.4721, 1.8083, 2.4744, 2.6198, 2.6889, 2.0598], + device='cuda:1'), covar=tensor([0.0498, 0.0491, 0.0550, 0.0761, 0.0682, 0.0541, 0.0463, 0.0885], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0134, 0.0138, 0.0119, 0.0123, 0.0137, 0.0139, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:46:10,701 INFO [finetune.py:976] (1/7) Epoch 20, batch 4450, loss[loss=0.1506, simple_loss=0.2301, pruned_loss=0.03558, over 4811.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2479, pruned_loss=0.05354, over 956470.36 frames. ], batch size: 25, lr: 3.23e-03, grad_scale: 32.0 +2023-03-27 00:46:10,767 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=113277.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:46:13,204 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=113281.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:46:23,195 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=113295.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:46:50,258 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=113321.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:46:53,840 INFO [finetune.py:976] (1/7) Epoch 20, batch 4500, loss[loss=0.185, simple_loss=0.2599, pruned_loss=0.05508, over 4894.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2502, pruned_loss=0.05463, over 955320.49 frames. ], batch size: 35, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:47:13,422 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.100e+02 1.617e+02 2.027e+02 2.372e+02 5.258e+02, threshold=4.055e+02, percent-clipped=2.0 +2023-03-27 00:47:27,581 INFO [finetune.py:976] (1/7) Epoch 20, batch 4550, loss[loss=0.2303, simple_loss=0.3108, pruned_loss=0.07492, over 4838.00 frames. ], tot_loss[loss=0.1813, simple_loss=0.2524, pruned_loss=0.05505, over 957202.65 frames. ], batch size: 47, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:47:42,880 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4806, 1.4190, 1.9012, 2.9988, 1.9613, 2.2295, 1.0087, 2.5386], + device='cuda:1'), covar=tensor([0.1904, 0.1418, 0.1332, 0.0516, 0.0918, 0.1203, 0.1829, 0.0515], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0134, 0.0165, 0.0101, 0.0136, 0.0124, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:47:44,935 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7935, 2.5803, 2.3544, 2.8848, 2.5900, 2.4962, 2.5361, 3.3766], + device='cuda:1'), covar=tensor([0.3453, 0.4285, 0.3293, 0.3932, 0.3851, 0.2403, 0.3882, 0.1586], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0231, 0.0277, 0.0253, 0.0222, 0.0252, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:47:47,434 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3010, 1.4212, 1.4986, 0.8258, 1.4751, 1.7448, 1.7777, 1.3102], + device='cuda:1'), covar=tensor([0.1072, 0.0654, 0.0513, 0.0562, 0.0483, 0.0547, 0.0304, 0.0744], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0149, 0.0124, 0.0123, 0.0129, 0.0128, 0.0140, 0.0147], + device='cuda:1'), out_proj_covar=tensor([8.9582e-05, 1.0771e-04, 8.8539e-05, 8.6679e-05, 9.0437e-05, 9.1342e-05, + 1.0022e-04, 1.0523e-04], device='cuda:1') +2023-03-27 00:47:49,213 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2746, 2.8701, 2.7533, 1.0234, 3.0103, 2.2707, 0.6989, 1.9159], + device='cuda:1'), covar=tensor([0.2443, 0.2139, 0.1802, 0.3561, 0.1450, 0.1137, 0.4055, 0.1568], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0161, 0.0129, 0.0162, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 00:48:01,685 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2748, 2.2026, 1.8148, 2.1488, 2.0485, 2.0698, 2.1331, 2.7734], + device='cuda:1'), covar=tensor([0.3523, 0.4383, 0.3042, 0.3855, 0.3704, 0.2386, 0.3612, 0.1774], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0262, 0.0230, 0.0276, 0.0252, 0.0221, 0.0251, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:48:03,324 INFO [finetune.py:976] (1/7) Epoch 20, batch 4600, loss[loss=0.1708, simple_loss=0.2381, pruned_loss=0.05168, over 4882.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2511, pruned_loss=0.0543, over 956139.70 frames. ], batch size: 32, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:48:16,377 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.73 vs. limit=2.0 +2023-03-27 00:48:31,337 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.757e+01 1.614e+02 1.832e+02 2.145e+02 3.668e+02, threshold=3.664e+02, percent-clipped=0.0 +2023-03-27 00:48:35,693 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.43 vs. limit=5.0 +2023-03-27 00:48:43,890 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0146, 1.4984, 2.1341, 2.0372, 1.8169, 1.7812, 2.0164, 1.9017], + device='cuda:1'), covar=tensor([0.4145, 0.3855, 0.3179, 0.3535, 0.4499, 0.3450, 0.3915, 0.2996], + device='cuda:1'), in_proj_covar=tensor([0.0254, 0.0242, 0.0263, 0.0281, 0.0279, 0.0254, 0.0289, 0.0245], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:48:45,584 INFO [finetune.py:976] (1/7) Epoch 20, batch 4650, loss[loss=0.1648, simple_loss=0.2296, pruned_loss=0.05, over 4870.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2497, pruned_loss=0.05466, over 955652.65 frames. ], batch size: 31, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:49:12,600 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2317, 2.2149, 2.2050, 1.6856, 2.2067, 2.3655, 2.4140, 1.8425], + device='cuda:1'), covar=tensor([0.0577, 0.0572, 0.0705, 0.0820, 0.0707, 0.0630, 0.0501, 0.1086], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0135, 0.0139, 0.0120, 0.0124, 0.0138, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:49:19,010 INFO [finetune.py:976] (1/7) Epoch 20, batch 4700, loss[loss=0.1791, simple_loss=0.2474, pruned_loss=0.05535, over 4925.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2462, pruned_loss=0.05354, over 956471.31 frames. ], batch size: 37, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:49:34,887 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5115, 1.1190, 0.7426, 1.3591, 1.9386, 0.7747, 1.2867, 1.4519], + device='cuda:1'), covar=tensor([0.1540, 0.2105, 0.1797, 0.1211, 0.2013, 0.1919, 0.1497, 0.1912], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0109, 0.0091, 0.0118, 0.0092, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:49:37,210 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.713e+01 1.518e+02 1.819e+02 2.172e+02 5.123e+02, threshold=3.639e+02, percent-clipped=2.0 +2023-03-27 00:49:51,489 INFO [finetune.py:976] (1/7) Epoch 20, batch 4750, loss[loss=0.1656, simple_loss=0.2383, pruned_loss=0.04648, over 4818.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2435, pruned_loss=0.05259, over 956820.26 frames. ], batch size: 39, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:49:51,602 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113577.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:50:00,071 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=113590.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:50:07,964 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-27 00:50:18,373 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=113617.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:50:23,693 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=113625.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:50:24,845 INFO [finetune.py:976] (1/7) Epoch 20, batch 4800, loss[loss=0.1894, simple_loss=0.2669, pruned_loss=0.05593, over 4903.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2446, pruned_loss=0.05254, over 954162.43 frames. ], batch size: 36, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:50:36,268 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6569, 1.9317, 1.5543, 1.6062, 2.2767, 2.1040, 1.9665, 1.7886], + device='cuda:1'), covar=tensor([0.0540, 0.0368, 0.0626, 0.0391, 0.0308, 0.0674, 0.0338, 0.0450], + device='cuda:1'), in_proj_covar=tensor([0.0097, 0.0107, 0.0144, 0.0111, 0.0100, 0.0111, 0.0100, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.5312e-05, 8.2496e-05, 1.1349e-04, 8.5284e-05, 7.8005e-05, 8.2257e-05, + 7.4679e-05, 8.5645e-05], device='cuda:1') +2023-03-27 00:50:37,469 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=113646.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:50:43,817 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.144e+02 1.583e+02 1.918e+02 2.188e+02 4.674e+02, threshold=3.836e+02, percent-clipped=2.0 +2023-03-27 00:51:01,329 INFO [finetune.py:976] (1/7) Epoch 20, batch 4850, loss[loss=0.256, simple_loss=0.3126, pruned_loss=0.09969, over 4262.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2478, pruned_loss=0.05357, over 953500.53 frames. ], batch size: 66, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:51:02,045 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=113678.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:51:34,447 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=113707.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:51:51,615 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0 +2023-03-27 00:51:57,944 INFO [finetune.py:976] (1/7) Epoch 20, batch 4900, loss[loss=0.1809, simple_loss=0.2527, pruned_loss=0.05456, over 4744.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.2485, pruned_loss=0.05363, over 951974.69 frames. ], batch size: 27, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:52:20,123 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.029e+02 1.649e+02 2.053e+02 2.498e+02 4.513e+02, threshold=4.105e+02, percent-clipped=3.0 +2023-03-27 00:52:34,813 INFO [finetune.py:976] (1/7) Epoch 20, batch 4950, loss[loss=0.1561, simple_loss=0.2342, pruned_loss=0.03899, over 4780.00 frames. ], tot_loss[loss=0.1806, simple_loss=0.2512, pruned_loss=0.05501, over 953202.68 frames. ], batch size: 26, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:53:07,553 INFO [finetune.py:976] (1/7) Epoch 20, batch 5000, loss[loss=0.183, simple_loss=0.2522, pruned_loss=0.05695, over 4851.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2495, pruned_loss=0.05411, over 954647.70 frames. ], batch size: 44, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:53:26,538 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.026e+02 1.516e+02 1.779e+02 2.023e+02 5.156e+02, threshold=3.559e+02, percent-clipped=2.0 +2023-03-27 00:53:39,282 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0780, 1.7976, 2.1831, 1.3998, 2.0705, 2.2127, 1.6567, 2.3939], + device='cuda:1'), covar=tensor([0.1240, 0.1904, 0.1451, 0.2056, 0.0909, 0.1421, 0.2705, 0.0778], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0206, 0.0190, 0.0189, 0.0175, 0.0213, 0.0219, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:53:42,068 INFO [finetune.py:976] (1/7) Epoch 20, batch 5050, loss[loss=0.1538, simple_loss=0.2216, pruned_loss=0.04296, over 4819.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2458, pruned_loss=0.05266, over 956978.79 frames. ], batch size: 25, lr: 3.22e-03, grad_scale: 64.0 +2023-03-27 00:53:51,006 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113890.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:54:08,485 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6558, 1.6963, 1.3863, 1.6179, 2.0674, 1.9427, 1.6705, 1.4598], + device='cuda:1'), covar=tensor([0.0323, 0.0280, 0.0577, 0.0285, 0.0175, 0.0426, 0.0337, 0.0387], + device='cuda:1'), in_proj_covar=tensor([0.0097, 0.0107, 0.0143, 0.0111, 0.0100, 0.0111, 0.0100, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.4825e-05, 8.2160e-05, 1.1248e-04, 8.4814e-05, 7.7456e-05, 8.1871e-05, + 7.4082e-05, 8.5356e-05], device='cuda:1') +2023-03-27 00:54:14,769 INFO [finetune.py:976] (1/7) Epoch 20, batch 5100, loss[loss=0.1773, simple_loss=0.2448, pruned_loss=0.0549, over 4795.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2428, pruned_loss=0.05171, over 957310.29 frames. ], batch size: 29, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:54:23,010 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=113938.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:54:35,763 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.073e+02 1.567e+02 1.826e+02 2.193e+02 3.507e+02, threshold=3.652e+02, percent-clipped=0.0 +2023-03-27 00:54:38,926 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5017, 1.4255, 2.0187, 1.8147, 1.5973, 3.4914, 1.3430, 1.5377], + device='cuda:1'), covar=tensor([0.0970, 0.1746, 0.1080, 0.0882, 0.1465, 0.0245, 0.1482, 0.1690], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0082, 0.0075, 0.0077, 0.0092, 0.0081, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 00:54:40,128 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6422, 1.2447, 0.7860, 1.5630, 2.0920, 1.4440, 1.3175, 1.7178], + device='cuda:1'), covar=tensor([0.2092, 0.2872, 0.2633, 0.1680, 0.2301, 0.2791, 0.2181, 0.2702], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0091, 0.0118, 0.0093, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:54:46,071 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=113973.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:54:48,407 INFO [finetune.py:976] (1/7) Epoch 20, batch 5150, loss[loss=0.1775, simple_loss=0.2635, pruned_loss=0.04572, over 4718.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2435, pruned_loss=0.05226, over 957213.29 frames. ], batch size: 59, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:55:07,780 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=114002.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:55:07,865 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8092, 1.3545, 1.9000, 1.7981, 1.6462, 1.5964, 1.7079, 1.7493], + device='cuda:1'), covar=tensor([0.3972, 0.3849, 0.3081, 0.3708, 0.4522, 0.3784, 0.4475, 0.2959], + device='cuda:1'), in_proj_covar=tensor([0.0254, 0.0243, 0.0263, 0.0282, 0.0279, 0.0256, 0.0290, 0.0245], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:55:23,295 INFO [finetune.py:976] (1/7) Epoch 20, batch 5200, loss[loss=0.1661, simple_loss=0.24, pruned_loss=0.04615, over 4754.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2471, pruned_loss=0.05344, over 953503.44 frames. ], batch size: 28, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:55:43,820 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.106e+02 1.581e+02 1.814e+02 2.243e+02 3.815e+02, threshold=3.628e+02, percent-clipped=1.0 +2023-03-27 00:55:56,417 INFO [finetune.py:976] (1/7) Epoch 20, batch 5250, loss[loss=0.1401, simple_loss=0.2044, pruned_loss=0.03793, over 4719.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2482, pruned_loss=0.05342, over 951424.42 frames. ], batch size: 23, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:56:07,010 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6942, 2.4380, 2.0251, 1.1145, 2.3755, 2.0145, 1.6244, 2.3329], + device='cuda:1'), covar=tensor([0.0771, 0.0871, 0.1615, 0.1966, 0.1260, 0.2017, 0.2214, 0.0954], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0191, 0.0200, 0.0182, 0.0210, 0.0209, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:56:11,905 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-27 00:56:36,233 INFO [finetune.py:976] (1/7) Epoch 20, batch 5300, loss[loss=0.1527, simple_loss=0.2339, pruned_loss=0.03574, over 4747.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2488, pruned_loss=0.05375, over 952317.03 frames. ], batch size: 26, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:56:36,352 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=114127.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:57:12,284 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3377, 2.1494, 1.6152, 2.1506, 2.1026, 1.8157, 2.4155, 2.2617], + device='cuda:1'), covar=tensor([0.1248, 0.2056, 0.3236, 0.2822, 0.2739, 0.1891, 0.3419, 0.1805], + device='cuda:1'), in_proj_covar=tensor([0.0185, 0.0187, 0.0233, 0.0253, 0.0246, 0.0203, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:57:13,340 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.589e+01 1.501e+02 1.834e+02 2.338e+02 4.244e+02, threshold=3.669e+02, percent-clipped=1.0 +2023-03-27 00:57:30,109 INFO [finetune.py:976] (1/7) Epoch 20, batch 5350, loss[loss=0.1607, simple_loss=0.2293, pruned_loss=0.04607, over 4751.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2489, pruned_loss=0.0532, over 950014.56 frames. ], batch size: 27, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:57:37,349 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=114188.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:57:46,924 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1036, 1.9785, 1.6257, 1.8559, 2.0055, 1.6978, 2.2101, 2.0849], + device='cuda:1'), covar=tensor([0.1292, 0.1974, 0.2941, 0.2581, 0.2676, 0.1837, 0.2936, 0.1721], + device='cuda:1'), in_proj_covar=tensor([0.0187, 0.0188, 0.0235, 0.0255, 0.0248, 0.0205, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:57:57,844 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6771, 3.3981, 3.3303, 1.4978, 3.5697, 2.6318, 0.7967, 2.3684], + device='cuda:1'), covar=tensor([0.2071, 0.2064, 0.1571, 0.3804, 0.1273, 0.1070, 0.4595, 0.1627], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0177, 0.0160, 0.0130, 0.0161, 0.0123, 0.0148, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 00:58:03,283 INFO [finetune.py:976] (1/7) Epoch 20, batch 5400, loss[loss=0.1697, simple_loss=0.2336, pruned_loss=0.05287, over 4897.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2464, pruned_loss=0.05252, over 950148.06 frames. ], batch size: 35, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:58:11,136 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-27 00:58:19,314 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6080, 1.0801, 0.7220, 1.4275, 2.0029, 0.7847, 1.3041, 1.4991], + device='cuda:1'), covar=tensor([0.1447, 0.1969, 0.1715, 0.1107, 0.1849, 0.1957, 0.1395, 0.1776], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0091, 0.0119, 0.0093, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 00:58:23,337 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.275e+01 1.521e+02 1.786e+02 2.103e+02 5.074e+02, threshold=3.573e+02, percent-clipped=1.0 +2023-03-27 00:58:27,147 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2710, 1.8843, 2.3921, 1.6618, 2.1872, 2.4231, 1.7193, 2.6314], + device='cuda:1'), covar=tensor([0.1199, 0.2107, 0.1368, 0.1962, 0.0963, 0.1389, 0.2785, 0.0830], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0206, 0.0190, 0.0189, 0.0174, 0.0213, 0.0219, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:58:33,637 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=114273.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:58:35,983 INFO [finetune.py:976] (1/7) Epoch 20, batch 5450, loss[loss=0.1507, simple_loss=0.2301, pruned_loss=0.03565, over 4901.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2451, pruned_loss=0.05282, over 951634.99 frames. ], batch size: 32, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:58:51,604 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=114302.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:58:53,296 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3480, 3.7504, 3.9555, 4.1796, 4.1286, 3.8553, 4.4621, 1.3449], + device='cuda:1'), covar=tensor([0.0827, 0.0817, 0.0865, 0.1081, 0.1171, 0.1556, 0.0693, 0.5516], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0242, 0.0278, 0.0291, 0.0330, 0.0284, 0.0303, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:59:05,065 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=114321.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:59:08,648 INFO [finetune.py:976] (1/7) Epoch 20, batch 5500, loss[loss=0.1589, simple_loss=0.2153, pruned_loss=0.05122, over 4196.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2438, pruned_loss=0.05244, over 954174.33 frames. ], batch size: 18, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:59:16,044 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3690, 2.2139, 1.8683, 0.9269, 2.0814, 1.8650, 1.6817, 2.0561], + device='cuda:1'), covar=tensor([0.0932, 0.0713, 0.1782, 0.2001, 0.1307, 0.2378, 0.2209, 0.0994], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0191, 0.0199, 0.0182, 0.0209, 0.0209, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 00:59:23,094 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=114350.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 00:59:27,726 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.653e+01 1.524e+02 1.755e+02 2.254e+02 4.886e+02, threshold=3.510e+02, percent-clipped=3.0 +2023-03-27 00:59:42,370 INFO [finetune.py:976] (1/7) Epoch 20, batch 5550, loss[loss=0.1596, simple_loss=0.2413, pruned_loss=0.039, over 4799.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2458, pruned_loss=0.0532, over 954761.03 frames. ], batch size: 45, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 00:59:49,156 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7816, 3.8157, 3.7296, 1.8038, 3.9434, 3.0938, 0.8195, 2.7633], + device='cuda:1'), covar=tensor([0.2533, 0.2244, 0.1477, 0.3552, 0.1097, 0.0893, 0.4727, 0.1502], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0177, 0.0160, 0.0130, 0.0161, 0.0123, 0.0148, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:00:00,261 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-27 01:00:14,060 INFO [finetune.py:976] (1/7) Epoch 20, batch 5600, loss[loss=0.239, simple_loss=0.3027, pruned_loss=0.08764, over 4062.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.2496, pruned_loss=0.05438, over 952561.40 frames. ], batch size: 65, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 01:00:14,902 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-27 01:00:31,887 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.005e+02 1.587e+02 1.911e+02 2.256e+02 4.682e+02, threshold=3.822e+02, percent-clipped=4.0 +2023-03-27 01:00:43,493 INFO [finetune.py:976] (1/7) Epoch 20, batch 5650, loss[loss=0.1601, simple_loss=0.2324, pruned_loss=0.04388, over 4828.00 frames. ], tot_loss[loss=0.1798, simple_loss=0.251, pruned_loss=0.05431, over 954686.65 frames. ], batch size: 30, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 01:00:43,570 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6062, 1.5127, 2.0203, 1.9995, 1.6712, 3.5000, 1.5369, 1.6248], + device='cuda:1'), covar=tensor([0.1088, 0.1991, 0.1129, 0.1044, 0.1666, 0.0312, 0.1683, 0.1928], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0077, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 01:00:46,982 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=114483.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:00:50,088 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.57 vs. limit=2.0 +2023-03-27 01:00:53,549 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9487, 1.2826, 2.0613, 1.9304, 1.7743, 1.7202, 1.8120, 1.9095], + device='cuda:1'), covar=tensor([0.3932, 0.3998, 0.3343, 0.3774, 0.5115, 0.3818, 0.4557, 0.3003], + device='cuda:1'), in_proj_covar=tensor([0.0254, 0.0242, 0.0262, 0.0281, 0.0279, 0.0255, 0.0290, 0.0245], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:00:56,317 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4974, 1.0728, 0.8283, 1.3518, 1.7686, 0.8685, 1.2789, 1.3967], + device='cuda:1'), covar=tensor([0.1312, 0.1803, 0.1600, 0.1033, 0.1879, 0.1995, 0.1226, 0.1740], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0091, 0.0120, 0.0093, 0.0097, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 01:00:57,508 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1103, 1.8906, 2.6458, 3.8213, 2.6436, 2.7606, 1.4834, 3.0800], + device='cuda:1'), covar=tensor([0.1625, 0.1340, 0.1178, 0.0522, 0.0764, 0.1292, 0.1744, 0.0502], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0163, 0.0100, 0.0135, 0.0123, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 01:01:13,265 INFO [finetune.py:976] (1/7) Epoch 20, batch 5700, loss[loss=0.1149, simple_loss=0.1818, pruned_loss=0.02406, over 3994.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2476, pruned_loss=0.05332, over 942994.38 frames. ], batch size: 17, lr: 3.22e-03, grad_scale: 32.0 +2023-03-27 01:01:24,539 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9239, 4.1157, 4.0529, 2.3342, 4.1604, 3.2486, 1.3728, 3.0373], + device='cuda:1'), covar=tensor([0.2139, 0.1799, 0.1304, 0.2900, 0.1033, 0.0896, 0.3917, 0.1360], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0177, 0.0160, 0.0129, 0.0161, 0.0123, 0.0148, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:01:39,094 INFO [finetune.py:976] (1/7) Epoch 21, batch 0, loss[loss=0.134, simple_loss=0.218, pruned_loss=0.02501, over 4769.00 frames. ], tot_loss[loss=0.134, simple_loss=0.218, pruned_loss=0.02501, over 4769.00 frames. ], batch size: 29, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:01:39,094 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 01:01:46,405 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8722, 3.4781, 3.6157, 3.7890, 3.6116, 3.4631, 3.9383, 1.3126], + device='cuda:1'), covar=tensor([0.0847, 0.0782, 0.0778, 0.0796, 0.1481, 0.1606, 0.0769, 0.5246], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0241, 0.0277, 0.0289, 0.0330, 0.0283, 0.0301, 0.0296], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:01:52,342 INFO [finetune.py:1010] (1/7) Epoch 21, validation: loss=0.1598, simple_loss=0.2277, pruned_loss=0.0459, over 2265189.00 frames. +2023-03-27 01:01:52,343 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 01:01:56,941 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.573e+01 1.356e+02 1.658e+02 2.014e+02 3.472e+02, threshold=3.316e+02, percent-clipped=0.0 +2023-03-27 01:02:29,069 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2427, 1.9274, 2.2780, 2.2214, 1.9403, 1.9457, 2.2064, 2.1647], + device='cuda:1'), covar=tensor([0.4110, 0.4106, 0.3160, 0.4192, 0.4926, 0.4065, 0.4826, 0.3048], + device='cuda:1'), in_proj_covar=tensor([0.0254, 0.0243, 0.0263, 0.0281, 0.0280, 0.0255, 0.0290, 0.0245], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:02:47,783 INFO [finetune.py:976] (1/7) Epoch 21, batch 50, loss[loss=0.1732, simple_loss=0.2577, pruned_loss=0.0444, over 4922.00 frames. ], tot_loss[loss=0.178, simple_loss=0.2485, pruned_loss=0.0538, over 215891.53 frames. ], batch size: 33, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:03:04,149 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0370, 2.0331, 1.6876, 1.9074, 1.9189, 1.8472, 1.8865, 2.6134], + device='cuda:1'), covar=tensor([0.3800, 0.3905, 0.3168, 0.3846, 0.3813, 0.2378, 0.3768, 0.1629], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0262, 0.0231, 0.0276, 0.0252, 0.0222, 0.0252, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:03:12,725 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.58 vs. limit=2.0 +2023-03-27 01:03:21,578 INFO [finetune.py:976] (1/7) Epoch 21, batch 100, loss[loss=0.1488, simple_loss=0.2232, pruned_loss=0.03717, over 4725.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2427, pruned_loss=0.05182, over 380864.09 frames. ], batch size: 54, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:03:23,372 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.865e+01 1.560e+02 1.971e+02 2.354e+02 5.080e+02, threshold=3.943e+02, percent-clipped=2.0 +2023-03-27 01:03:23,570 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-27 01:03:54,252 INFO [finetune.py:976] (1/7) Epoch 21, batch 150, loss[loss=0.1672, simple_loss=0.2329, pruned_loss=0.05076, over 4871.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2414, pruned_loss=0.05253, over 510157.35 frames. ], batch size: 31, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:04:02,521 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=114716.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:04:03,307 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-03-27 01:04:16,921 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6039, 2.4027, 2.0517, 1.0711, 2.2512, 2.0416, 1.8474, 2.2329], + device='cuda:1'), covar=tensor([0.0888, 0.0829, 0.1760, 0.2231, 0.1454, 0.2141, 0.2211, 0.1121], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0191, 0.0198, 0.0182, 0.0209, 0.0209, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:04:26,907 INFO [finetune.py:976] (1/7) Epoch 21, batch 200, loss[loss=0.1862, simple_loss=0.2674, pruned_loss=0.05247, over 4778.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2409, pruned_loss=0.05198, over 609014.97 frames. ], batch size: 54, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:04:29,189 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.002e+02 1.562e+02 1.886e+02 2.300e+02 5.249e+02, threshold=3.772e+02, percent-clipped=1.0 +2023-03-27 01:04:42,953 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=114777.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:04:46,578 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=114783.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:05:00,771 INFO [finetune.py:976] (1/7) Epoch 21, batch 250, loss[loss=0.189, simple_loss=0.2533, pruned_loss=0.06236, over 4890.00 frames. ], tot_loss[loss=0.1781, simple_loss=0.2469, pruned_loss=0.05465, over 685166.70 frames. ], batch size: 32, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:05:16,376 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7251, 2.8963, 2.5244, 1.9788, 2.7591, 2.8693, 3.1190, 2.4184], + device='cuda:1'), covar=tensor([0.0577, 0.0484, 0.0693, 0.0840, 0.0554, 0.0649, 0.0451, 0.0912], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0135, 0.0139, 0.0120, 0.0124, 0.0138, 0.0139, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:05:19,224 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=114831.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:05:33,145 INFO [finetune.py:976] (1/7) Epoch 21, batch 300, loss[loss=0.1917, simple_loss=0.2742, pruned_loss=0.05459, over 4813.00 frames. ], tot_loss[loss=0.18, simple_loss=0.25, pruned_loss=0.05497, over 745612.95 frames. ], batch size: 41, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:05:36,361 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.093e+02 1.500e+02 1.787e+02 2.137e+02 3.935e+02, threshold=3.575e+02, percent-clipped=3.0 +2023-03-27 01:05:37,093 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4286, 2.5459, 2.3676, 1.8794, 2.4428, 2.7444, 2.8636, 2.0754], + device='cuda:1'), covar=tensor([0.0583, 0.0624, 0.0786, 0.0943, 0.1012, 0.0580, 0.0529, 0.1066], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0136, 0.0140, 0.0121, 0.0125, 0.0139, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:05:46,355 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=114871.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:05:47,588 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0234, 1.8816, 1.7901, 2.0830, 2.3537, 2.1290, 1.6190, 1.7646], + device='cuda:1'), covar=tensor([0.2269, 0.2090, 0.2093, 0.1724, 0.1625, 0.1225, 0.2536, 0.2020], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0211, 0.0214, 0.0195, 0.0243, 0.0188, 0.0218, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:06:06,528 INFO [finetune.py:976] (1/7) Epoch 21, batch 350, loss[loss=0.1383, simple_loss=0.2121, pruned_loss=0.03225, over 4221.00 frames. ], tot_loss[loss=0.1823, simple_loss=0.2523, pruned_loss=0.05613, over 789859.30 frames. ], batch size: 18, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:06:26,471 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=114932.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:06:40,072 INFO [finetune.py:976] (1/7) Epoch 21, batch 400, loss[loss=0.1468, simple_loss=0.2357, pruned_loss=0.02892, over 4772.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.251, pruned_loss=0.05488, over 826089.04 frames. ], batch size: 26, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:06:41,872 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.016e+02 1.627e+02 1.950e+02 2.383e+02 4.205e+02, threshold=3.900e+02, percent-clipped=3.0 +2023-03-27 01:07:20,652 INFO [finetune.py:976] (1/7) Epoch 21, batch 450, loss[loss=0.1757, simple_loss=0.2329, pruned_loss=0.05926, over 4822.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2503, pruned_loss=0.05438, over 854543.04 frames. ], batch size: 41, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:07:50,287 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4507, 2.5889, 2.3613, 1.7989, 2.3501, 2.6537, 2.7741, 2.1711], + device='cuda:1'), covar=tensor([0.0652, 0.0630, 0.0774, 0.0901, 0.0943, 0.0713, 0.0578, 0.1134], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0138, 0.0141, 0.0122, 0.0126, 0.0141, 0.0142, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:08:11,192 INFO [finetune.py:976] (1/7) Epoch 21, batch 500, loss[loss=0.1615, simple_loss=0.2338, pruned_loss=0.04454, over 4799.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2472, pruned_loss=0.05312, over 875920.41 frames. ], batch size: 29, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:08:13,018 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.154e+02 1.445e+02 1.728e+02 2.124e+02 2.919e+02, threshold=3.456e+02, percent-clipped=0.0 +2023-03-27 01:08:24,206 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=115072.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:08:33,765 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=115086.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:08:45,010 INFO [finetune.py:976] (1/7) Epoch 21, batch 550, loss[loss=0.2023, simple_loss=0.264, pruned_loss=0.07032, over 4901.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2451, pruned_loss=0.05253, over 895133.43 frames. ], batch size: 32, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:08:52,883 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1613, 2.2255, 1.8187, 2.1889, 2.7886, 2.3261, 2.1188, 1.6640], + device='cuda:1'), covar=tensor([0.2323, 0.1829, 0.1917, 0.1790, 0.1670, 0.1063, 0.1963, 0.1892], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0210, 0.0212, 0.0194, 0.0242, 0.0187, 0.0217, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:09:14,153 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=115147.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:09:18,263 INFO [finetune.py:976] (1/7) Epoch 21, batch 600, loss[loss=0.1972, simple_loss=0.2665, pruned_loss=0.06389, over 4867.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2466, pruned_loss=0.05342, over 908588.84 frames. ], batch size: 34, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:09:19,078 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-27 01:09:20,110 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.558e+02 1.835e+02 2.263e+02 4.639e+02, threshold=3.670e+02, percent-clipped=5.0 +2023-03-27 01:09:45,870 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-27 01:09:51,869 INFO [finetune.py:976] (1/7) Epoch 21, batch 650, loss[loss=0.1998, simple_loss=0.2682, pruned_loss=0.06569, over 4861.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2497, pruned_loss=0.05424, over 919391.47 frames. ], batch size: 44, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:09:54,353 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1397, 1.2817, 1.5201, 1.4049, 1.4257, 2.4907, 1.2309, 1.4503], + device='cuda:1'), covar=tensor([0.0984, 0.1843, 0.0988, 0.0935, 0.1724, 0.0377, 0.1566, 0.1757], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0076, 0.0092, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 01:10:07,421 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=115227.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:10:25,138 INFO [finetune.py:976] (1/7) Epoch 21, batch 700, loss[loss=0.2211, simple_loss=0.2751, pruned_loss=0.0835, over 4895.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.2506, pruned_loss=0.05393, over 927619.75 frames. ], batch size: 43, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:10:26,913 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.090e+02 1.664e+02 1.957e+02 2.283e+02 3.730e+02, threshold=3.914e+02, percent-clipped=1.0 +2023-03-27 01:10:32,655 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.58 vs. limit=5.0 +2023-03-27 01:10:58,917 INFO [finetune.py:976] (1/7) Epoch 21, batch 750, loss[loss=0.1842, simple_loss=0.2664, pruned_loss=0.05095, over 4813.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2512, pruned_loss=0.05379, over 934082.58 frames. ], batch size: 33, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:11:31,743 INFO [finetune.py:976] (1/7) Epoch 21, batch 800, loss[loss=0.1968, simple_loss=0.267, pruned_loss=0.0633, over 4805.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2511, pruned_loss=0.05401, over 938292.00 frames. ], batch size: 45, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:11:33,561 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.112e+02 1.426e+02 1.720e+02 2.044e+02 3.360e+02, threshold=3.441e+02, percent-clipped=0.0 +2023-03-27 01:11:41,498 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=115370.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:11:42,120 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=115371.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:11:42,697 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=115372.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:11:48,706 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5290, 1.9349, 1.5060, 1.6047, 2.2841, 2.1284, 1.9272, 1.8772], + device='cuda:1'), covar=tensor([0.0578, 0.0335, 0.0591, 0.0355, 0.0301, 0.0687, 0.0386, 0.0390], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0109, 0.0146, 0.0113, 0.0101, 0.0112, 0.0101, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.6653e-05, 8.3669e-05, 1.1514e-04, 8.6645e-05, 7.8521e-05, 8.3039e-05, + 7.5312e-05, 8.6898e-05], device='cuda:1') +2023-03-27 01:12:04,583 INFO [finetune.py:976] (1/7) Epoch 21, batch 850, loss[loss=0.1666, simple_loss=0.2362, pruned_loss=0.04849, over 4830.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.2486, pruned_loss=0.0536, over 943233.28 frames. ], batch size: 33, lr: 3.21e-03, grad_scale: 32.0 +2023-03-27 01:12:10,832 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0 +2023-03-27 01:12:13,711 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4564, 2.2762, 1.7598, 0.8819, 1.8824, 2.0287, 1.9000, 2.0579], + device='cuda:1'), covar=tensor([0.0852, 0.0745, 0.1559, 0.1893, 0.1357, 0.2199, 0.2017, 0.0888], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0190, 0.0198, 0.0181, 0.0209, 0.0208, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:12:14,850 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=115420.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:12:16,189 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.45 vs. limit=5.0 +2023-03-27 01:12:24,437 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=115431.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:12:25,033 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=115432.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:12:32,209 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6605, 2.4328, 2.0022, 1.0032, 2.1408, 2.0488, 1.8513, 2.1942], + device='cuda:1'), covar=tensor([0.0750, 0.0698, 0.1450, 0.1837, 0.1213, 0.2180, 0.2042, 0.0864], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0190, 0.0198, 0.0181, 0.0209, 0.0209, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:12:41,967 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=115442.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:12:54,665 INFO [finetune.py:976] (1/7) Epoch 21, batch 900, loss[loss=0.1543, simple_loss=0.2237, pruned_loss=0.04247, over 4890.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2454, pruned_loss=0.05235, over 946525.36 frames. ], batch size: 35, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:13:00,780 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.046e+02 1.494e+02 1.776e+02 2.140e+02 4.219e+02, threshold=3.551e+02, percent-clipped=3.0 +2023-03-27 01:13:37,348 INFO [finetune.py:976] (1/7) Epoch 21, batch 950, loss[loss=0.2109, simple_loss=0.265, pruned_loss=0.07838, over 4824.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2452, pruned_loss=0.05311, over 947392.85 frames. ], batch size: 30, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:13:42,949 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-03-27 01:13:51,946 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=115527.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:14:11,295 INFO [finetune.py:976] (1/7) Epoch 21, batch 1000, loss[loss=0.22, simple_loss=0.2857, pruned_loss=0.07715, over 4746.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2458, pruned_loss=0.05279, over 948012.71 frames. ], batch size: 54, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:14:13,112 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.049e+02 1.550e+02 1.849e+02 2.159e+02 3.452e+02, threshold=3.698e+02, percent-clipped=0.0 +2023-03-27 01:14:24,462 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=115575.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:14:44,023 INFO [finetune.py:976] (1/7) Epoch 21, batch 1050, loss[loss=0.1498, simple_loss=0.2176, pruned_loss=0.04099, over 4437.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2493, pruned_loss=0.05354, over 950433.09 frames. ], batch size: 19, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:15:16,669 INFO [finetune.py:976] (1/7) Epoch 21, batch 1100, loss[loss=0.2406, simple_loss=0.2933, pruned_loss=0.09391, over 4929.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2505, pruned_loss=0.05336, over 951790.47 frames. ], batch size: 42, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:15:19,444 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.043e+02 1.582e+02 1.822e+02 2.328e+02 4.675e+02, threshold=3.643e+02, percent-clipped=4.0 +2023-03-27 01:15:20,303 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-27 01:15:50,437 INFO [finetune.py:976] (1/7) Epoch 21, batch 1150, loss[loss=0.1873, simple_loss=0.2581, pruned_loss=0.05818, over 4868.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2501, pruned_loss=0.05247, over 953378.17 frames. ], batch size: 34, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:16:05,310 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=115726.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:16:05,911 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=115727.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:16:14,924 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=115742.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:16:15,570 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4234, 2.4801, 2.5680, 1.3086, 2.8919, 3.1177, 2.7031, 2.2814], + device='cuda:1'), covar=tensor([0.0879, 0.0783, 0.0456, 0.0751, 0.0667, 0.0714, 0.0472, 0.0817], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0149, 0.0125, 0.0123, 0.0130, 0.0129, 0.0142, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.9982e-05, 1.0796e-04, 8.9652e-05, 8.7279e-05, 9.1453e-05, 9.1930e-05, + 1.0173e-04, 1.0609e-04], device='cuda:1') +2023-03-27 01:16:16,249 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-27 01:16:24,029 INFO [finetune.py:976] (1/7) Epoch 21, batch 1200, loss[loss=0.148, simple_loss=0.2074, pruned_loss=0.04434, over 3969.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2486, pruned_loss=0.05167, over 954321.94 frames. ], batch size: 17, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:16:25,822 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.936e+01 1.465e+02 1.737e+02 2.048e+02 4.574e+02, threshold=3.475e+02, percent-clipped=2.0 +2023-03-27 01:16:47,357 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=115790.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:16:56,822 INFO [finetune.py:976] (1/7) Epoch 21, batch 1250, loss[loss=0.1492, simple_loss=0.2275, pruned_loss=0.0354, over 4650.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2465, pruned_loss=0.05123, over 956295.99 frames. ], batch size: 23, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:17:13,344 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5305, 3.4376, 3.2187, 1.3515, 3.5562, 2.7202, 0.9525, 2.3911], + device='cuda:1'), covar=tensor([0.2272, 0.2005, 0.1712, 0.3574, 0.1201, 0.0996, 0.4213, 0.1548], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0178, 0.0159, 0.0129, 0.0161, 0.0123, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:17:29,532 INFO [finetune.py:976] (1/7) Epoch 21, batch 1300, loss[loss=0.1284, simple_loss=0.1992, pruned_loss=0.02879, over 4915.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.243, pruned_loss=0.0506, over 955492.90 frames. ], batch size: 35, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:17:32,371 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.105e+02 1.563e+02 1.759e+02 2.181e+02 4.124e+02, threshold=3.519e+02, percent-clipped=1.0 +2023-03-27 01:17:52,270 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8874, 1.7169, 1.5494, 1.2756, 1.6953, 1.6686, 1.7019, 2.2298], + device='cuda:1'), covar=tensor([0.3520, 0.3577, 0.3169, 0.3448, 0.3474, 0.2251, 0.3153, 0.1761], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0233, 0.0277, 0.0253, 0.0223, 0.0253, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:17:53,438 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7246, 1.6809, 2.2173, 3.6740, 2.4813, 2.4767, 1.8254, 3.0012], + device='cuda:1'), covar=tensor([0.1823, 0.1467, 0.1432, 0.0529, 0.0780, 0.1382, 0.1317, 0.0455], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0163, 0.0100, 0.0136, 0.0124, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 01:18:11,971 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.65 vs. limit=5.0 +2023-03-27 01:18:12,435 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=115892.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 01:18:23,980 INFO [finetune.py:976] (1/7) Epoch 21, batch 1350, loss[loss=0.1983, simple_loss=0.2643, pruned_loss=0.06612, over 4837.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2439, pruned_loss=0.05119, over 956423.42 frames. ], batch size: 33, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:18:56,327 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7780, 1.6936, 1.5882, 1.7944, 1.4779, 3.7190, 1.5005, 2.0715], + device='cuda:1'), covar=tensor([0.3142, 0.2335, 0.2091, 0.2289, 0.1548, 0.0169, 0.2391, 0.1111], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0115, 0.0120, 0.0122, 0.0113, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 01:19:00,569 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=115953.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 01:19:01,046 INFO [finetune.py:976] (1/7) Epoch 21, batch 1400, loss[loss=0.1922, simple_loss=0.2636, pruned_loss=0.0604, over 4915.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2455, pruned_loss=0.05165, over 955302.98 frames. ], batch size: 37, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:19:02,861 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.014e+02 1.541e+02 1.818e+02 2.071e+02 3.575e+02, threshold=3.635e+02, percent-clipped=1.0 +2023-03-27 01:19:09,280 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-03-27 01:19:27,716 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9520, 1.7078, 1.6193, 1.2711, 1.7488, 1.7492, 1.7777, 2.2932], + device='cuda:1'), covar=tensor([0.3563, 0.3671, 0.3044, 0.3491, 0.3500, 0.2067, 0.3101, 0.1756], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0261, 0.0231, 0.0275, 0.0251, 0.0221, 0.0251, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:19:35,505 INFO [finetune.py:976] (1/7) Epoch 21, batch 1450, loss[loss=0.1919, simple_loss=0.2679, pruned_loss=0.05797, over 4909.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2487, pruned_loss=0.05283, over 957190.78 frames. ], batch size: 37, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:19:43,010 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116014.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:19:51,745 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116026.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:19:52,337 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116027.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:20:00,692 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3073, 1.5375, 1.5163, 0.8316, 1.5287, 1.8026, 1.7848, 1.4180], + device='cuda:1'), covar=tensor([0.0994, 0.0580, 0.0586, 0.0530, 0.0510, 0.0583, 0.0362, 0.0703], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0149, 0.0125, 0.0123, 0.0130, 0.0128, 0.0141, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.9918e-05, 1.0780e-04, 8.9473e-05, 8.7318e-05, 9.1146e-05, 9.1593e-05, + 1.0139e-04, 1.0594e-04], device='cuda:1') +2023-03-27 01:20:09,068 INFO [finetune.py:976] (1/7) Epoch 21, batch 1500, loss[loss=0.1841, simple_loss=0.2511, pruned_loss=0.05851, over 4848.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2507, pruned_loss=0.05417, over 955048.42 frames. ], batch size: 44, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:20:10,889 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.231e+02 1.719e+02 2.058e+02 2.312e+02 4.180e+02, threshold=4.116e+02, percent-clipped=2.0 +2023-03-27 01:20:11,730 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.85 vs. limit=5.0 +2023-03-27 01:20:23,169 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=116074.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:20:23,793 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=116075.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:20:23,864 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116075.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 01:20:31,500 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6971, 0.7492, 1.8130, 1.6555, 1.5627, 1.4867, 1.5392, 1.7093], + device='cuda:1'), covar=tensor([0.3586, 0.3793, 0.3153, 0.3287, 0.4363, 0.3552, 0.4180, 0.3011], + device='cuda:1'), in_proj_covar=tensor([0.0253, 0.0242, 0.0263, 0.0281, 0.0280, 0.0256, 0.0290, 0.0245], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:20:32,630 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2271, 1.3579, 1.5314, 1.5655, 1.5251, 2.8874, 1.2871, 1.5108], + device='cuda:1'), covar=tensor([0.1009, 0.1793, 0.1086, 0.0939, 0.1599, 0.0290, 0.1482, 0.1680], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0075, 0.0077, 0.0092, 0.0081, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 01:20:36,124 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1208, 1.9381, 1.6635, 1.8446, 1.8510, 1.7796, 1.9228, 2.6595], + device='cuda:1'), covar=tensor([0.3556, 0.4029, 0.3298, 0.3464, 0.3784, 0.2407, 0.3429, 0.1550], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0261, 0.0232, 0.0275, 0.0252, 0.0222, 0.0252, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:20:40,954 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4420, 2.6562, 2.4042, 1.8397, 2.3288, 2.6547, 2.6775, 2.1810], + device='cuda:1'), covar=tensor([0.0577, 0.0526, 0.0759, 0.0889, 0.0933, 0.0763, 0.0557, 0.1021], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0135, 0.0139, 0.0120, 0.0125, 0.0138, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:20:42,654 INFO [finetune.py:976] (1/7) Epoch 21, batch 1550, loss[loss=0.1873, simple_loss=0.2582, pruned_loss=0.05819, over 4778.00 frames. ], tot_loss[loss=0.1813, simple_loss=0.2523, pruned_loss=0.05515, over 953728.01 frames. ], batch size: 51, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:20:43,362 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116105.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:20:48,776 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5471, 2.6802, 2.4579, 1.8899, 2.5072, 2.6751, 2.8491, 2.2867], + device='cuda:1'), covar=tensor([0.0580, 0.0582, 0.0700, 0.0813, 0.0724, 0.0745, 0.0529, 0.0988], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0135, 0.0139, 0.0119, 0.0125, 0.0138, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:20:50,858 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116116.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:21:07,210 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2771, 2.1119, 1.7189, 0.8531, 1.9145, 1.8026, 1.6703, 1.9633], + device='cuda:1'), covar=tensor([0.0757, 0.0664, 0.1500, 0.1913, 0.1225, 0.2108, 0.2115, 0.0814], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0191, 0.0197, 0.0181, 0.0209, 0.0208, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:21:15,951 INFO [finetune.py:976] (1/7) Epoch 21, batch 1600, loss[loss=0.1592, simple_loss=0.2354, pruned_loss=0.04145, over 4824.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2502, pruned_loss=0.05482, over 953273.53 frames. ], batch size: 33, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:21:17,776 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.061e+01 1.555e+02 1.841e+02 2.223e+02 4.654e+02, threshold=3.683e+02, percent-clipped=1.0 +2023-03-27 01:21:20,323 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3838, 2.5730, 2.2144, 1.7324, 2.3008, 2.5661, 2.6953, 2.1551], + device='cuda:1'), covar=tensor([0.0621, 0.0570, 0.0886, 0.0912, 0.1006, 0.0757, 0.0605, 0.1116], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0135, 0.0138, 0.0119, 0.0124, 0.0137, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:21:23,367 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116166.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:21:29,518 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5845, 3.7981, 3.5402, 1.6589, 3.8412, 2.8610, 0.8460, 2.6704], + device='cuda:1'), covar=tensor([0.2619, 0.2082, 0.1779, 0.3767, 0.1141, 0.1096, 0.5013, 0.1593], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0160, 0.0130, 0.0162, 0.0124, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:21:31,973 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116177.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:21:49,856 INFO [finetune.py:976] (1/7) Epoch 21, batch 1650, loss[loss=0.1468, simple_loss=0.2057, pruned_loss=0.04397, over 4916.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2469, pruned_loss=0.05342, over 955809.38 frames. ], batch size: 43, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:22:12,327 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7251, 1.6668, 1.6188, 1.7332, 1.2070, 3.6720, 1.4997, 2.0352], + device='cuda:1'), covar=tensor([0.3206, 0.2414, 0.2041, 0.2255, 0.1676, 0.0171, 0.2390, 0.1099], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0115, 0.0120, 0.0122, 0.0112, 0.0096, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 01:22:19,471 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116248.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 01:22:23,485 INFO [finetune.py:976] (1/7) Epoch 21, batch 1700, loss[loss=0.1475, simple_loss=0.2181, pruned_loss=0.03849, over 4812.00 frames. ], tot_loss[loss=0.1746, simple_loss=0.2444, pruned_loss=0.0524, over 957165.54 frames. ], batch size: 25, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:22:25,327 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.524e+01 1.502e+02 1.774e+02 2.116e+02 3.203e+02, threshold=3.548e+02, percent-clipped=0.0 +2023-03-27 01:22:59,247 INFO [finetune.py:976] (1/7) Epoch 21, batch 1750, loss[loss=0.136, simple_loss=0.2113, pruned_loss=0.0303, over 4766.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2467, pruned_loss=0.05315, over 957147.65 frames. ], batch size: 28, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:23:19,675 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116323.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:23:51,745 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4267, 3.8195, 4.0462, 4.2093, 4.2255, 3.9591, 4.4782, 1.8149], + device='cuda:1'), covar=tensor([0.0806, 0.1038, 0.1053, 0.1170, 0.1186, 0.1461, 0.0770, 0.5217], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0240, 0.0278, 0.0290, 0.0331, 0.0284, 0.0301, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:23:58,755 INFO [finetune.py:976] (1/7) Epoch 21, batch 1800, loss[loss=0.1794, simple_loss=0.2584, pruned_loss=0.05021, over 4821.00 frames. ], tot_loss[loss=0.1788, simple_loss=0.2499, pruned_loss=0.05387, over 955621.02 frames. ], batch size: 38, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:23:59,540 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9482, 1.0716, 1.8561, 1.8566, 1.6884, 1.6648, 1.7152, 1.8068], + device='cuda:1'), covar=tensor([0.3756, 0.3981, 0.3455, 0.3568, 0.4804, 0.3881, 0.4180, 0.3052], + device='cuda:1'), in_proj_covar=tensor([0.0254, 0.0241, 0.0262, 0.0281, 0.0280, 0.0256, 0.0290, 0.0245], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:24:00,588 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.248e+02 1.627e+02 1.938e+02 2.363e+02 5.057e+02, threshold=3.876e+02, percent-clipped=3.0 +2023-03-27 01:24:02,499 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7005, 2.5164, 2.1240, 1.1346, 2.2835, 1.9705, 1.8746, 2.2312], + device='cuda:1'), covar=tensor([0.0761, 0.0833, 0.1685, 0.2150, 0.1435, 0.2254, 0.2090, 0.1062], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0192, 0.0199, 0.0183, 0.0210, 0.0209, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:24:08,546 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116370.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 01:24:09,812 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3100, 2.2150, 2.7704, 1.6512, 2.3323, 2.5481, 1.8565, 2.8617], + device='cuda:1'), covar=tensor([0.1504, 0.1998, 0.1588, 0.2340, 0.1052, 0.1631, 0.3014, 0.0840], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0202, 0.0189, 0.0188, 0.0173, 0.0212, 0.0217, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:24:12,248 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8527, 1.8750, 1.5981, 2.0870, 2.4337, 2.0695, 1.7576, 1.5293], + device='cuda:1'), covar=tensor([0.2250, 0.1962, 0.1985, 0.1617, 0.1590, 0.1184, 0.2254, 0.1995], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0211, 0.0213, 0.0195, 0.0244, 0.0189, 0.0219, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:24:18,613 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116384.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:24:25,694 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4929, 3.4164, 3.2709, 1.6031, 3.5210, 2.5563, 0.8435, 2.2987], + device='cuda:1'), covar=tensor([0.2199, 0.1952, 0.1604, 0.3190, 0.1215, 0.1050, 0.4381, 0.1475], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0177, 0.0158, 0.0129, 0.0161, 0.0123, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:24:31,788 INFO [finetune.py:976] (1/7) Epoch 21, batch 1850, loss[loss=0.195, simple_loss=0.2621, pruned_loss=0.06392, over 4865.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2521, pruned_loss=0.05463, over 955300.44 frames. ], batch size: 34, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:24:34,457 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-27 01:24:50,474 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2438, 2.1295, 1.7660, 2.0338, 2.1265, 1.8743, 2.4146, 2.1782], + device='cuda:1'), covar=tensor([0.1377, 0.1962, 0.3026, 0.2486, 0.2556, 0.1782, 0.2807, 0.1837], + device='cuda:1'), in_proj_covar=tensor([0.0185, 0.0188, 0.0235, 0.0253, 0.0248, 0.0203, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:25:00,244 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116446.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:25:04,916 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2819, 2.2338, 2.1922, 1.6489, 2.2278, 2.3932, 2.4240, 1.7859], + device='cuda:1'), covar=tensor([0.0586, 0.0522, 0.0644, 0.0835, 0.0618, 0.0625, 0.0457, 0.1108], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0137, 0.0140, 0.0121, 0.0126, 0.0139, 0.0141, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:25:05,404 INFO [finetune.py:976] (1/7) Epoch 21, batch 1900, loss[loss=0.1509, simple_loss=0.2323, pruned_loss=0.03477, over 4868.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2529, pruned_loss=0.0547, over 955400.95 frames. ], batch size: 34, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:25:07,229 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.037e+02 1.572e+02 1.880e+02 2.123e+02 3.861e+02, threshold=3.760e+02, percent-clipped=0.0 +2023-03-27 01:25:10,211 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116461.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:25:16,978 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116472.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:25:38,738 INFO [finetune.py:976] (1/7) Epoch 21, batch 1950, loss[loss=0.1268, simple_loss=0.2084, pruned_loss=0.02262, over 4706.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2503, pruned_loss=0.05314, over 956921.92 frames. ], batch size: 23, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:25:39,474 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116505.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:25:40,700 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116507.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 01:25:45,372 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5424, 1.4955, 2.3984, 1.9502, 1.8910, 4.0462, 1.4125, 1.8374], + device='cuda:1'), covar=tensor([0.0975, 0.1839, 0.1130, 0.0969, 0.1577, 0.0214, 0.1575, 0.1760], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0076, 0.0092, 0.0080, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 01:26:07,964 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116548.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 01:26:11,460 INFO [finetune.py:976] (1/7) Epoch 21, batch 2000, loss[loss=0.2031, simple_loss=0.2633, pruned_loss=0.0714, over 4901.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2471, pruned_loss=0.05194, over 958171.43 frames. ], batch size: 43, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:26:13,786 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 7.123e+01 1.433e+02 1.690e+02 2.067e+02 3.885e+02, threshold=3.380e+02, percent-clipped=2.0 +2023-03-27 01:26:19,757 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116566.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:26:39,350 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=116596.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 01:26:39,409 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5071, 1.7390, 1.7005, 0.9251, 1.8390, 2.0412, 1.9297, 1.4968], + device='cuda:1'), covar=tensor([0.0959, 0.0669, 0.0529, 0.0553, 0.0421, 0.0664, 0.0361, 0.0801], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0149, 0.0125, 0.0124, 0.0130, 0.0128, 0.0142, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.9837e-05, 1.0803e-04, 8.9403e-05, 8.7394e-05, 9.1482e-05, 9.1759e-05, + 1.0159e-04, 1.0590e-04], device='cuda:1') +2023-03-27 01:26:44,684 INFO [finetune.py:976] (1/7) Epoch 21, batch 2050, loss[loss=0.1892, simple_loss=0.2548, pruned_loss=0.06178, over 4892.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2437, pruned_loss=0.05098, over 957327.17 frames. ], batch size: 32, lr: 3.20e-03, grad_scale: 64.0 +2023-03-27 01:26:54,228 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116618.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:26:56,682 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9688, 2.6898, 2.4717, 1.3844, 2.7754, 2.1427, 2.0303, 2.4247], + device='cuda:1'), covar=tensor([0.1077, 0.0905, 0.1841, 0.2157, 0.1371, 0.2368, 0.2280, 0.1143], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0191, 0.0197, 0.0181, 0.0207, 0.0207, 0.0221, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:27:11,116 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.10 vs. limit=2.0 +2023-03-27 01:27:18,449 INFO [finetune.py:976] (1/7) Epoch 21, batch 2100, loss[loss=0.1659, simple_loss=0.2349, pruned_loss=0.04842, over 4768.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2438, pruned_loss=0.0513, over 957851.21 frames. ], batch size: 29, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:27:20,846 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.015e+02 1.617e+02 1.783e+02 2.198e+02 6.495e+02, threshold=3.567e+02, percent-clipped=4.0 +2023-03-27 01:27:29,058 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116670.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 01:27:34,425 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116679.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:27:34,465 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116679.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:27:51,925 INFO [finetune.py:976] (1/7) Epoch 21, batch 2150, loss[loss=0.191, simple_loss=0.259, pruned_loss=0.06143, over 4840.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2468, pruned_loss=0.05271, over 956442.92 frames. ], batch size: 33, lr: 3.20e-03, grad_scale: 32.0 +2023-03-27 01:27:52,031 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116704.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:28:00,964 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=116718.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:28:26,697 INFO [finetune.py:976] (1/7) Epoch 21, batch 2200, loss[loss=0.2108, simple_loss=0.2809, pruned_loss=0.0704, over 4862.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.2488, pruned_loss=0.05342, over 956900.00 frames. ], batch size: 44, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:28:30,714 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.107e+02 1.636e+02 2.054e+02 2.505e+02 6.138e+02, threshold=4.108e+02, percent-clipped=5.0 +2023-03-27 01:28:32,678 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116761.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:28:34,566 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-27 01:28:39,668 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116765.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:28:44,484 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116772.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:29:19,238 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116802.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 01:29:19,264 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116802.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:29:20,383 INFO [finetune.py:976] (1/7) Epoch 21, batch 2250, loss[loss=0.2049, simple_loss=0.2781, pruned_loss=0.06585, over 4888.00 frames. ], tot_loss[loss=0.1788, simple_loss=0.25, pruned_loss=0.05374, over 957623.58 frames. ], batch size: 43, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:29:28,967 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=116809.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:29:39,990 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=116820.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:29:52,785 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.86 vs. limit=5.0 +2023-03-27 01:29:53,295 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1787, 2.0591, 1.7797, 2.1228, 1.9383, 1.9410, 1.9949, 2.6951], + device='cuda:1'), covar=tensor([0.3965, 0.4362, 0.3393, 0.3828, 0.4184, 0.2580, 0.4076, 0.1786], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0232, 0.0277, 0.0253, 0.0223, 0.0253, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:30:01,467 INFO [finetune.py:976] (1/7) Epoch 21, batch 2300, loss[loss=0.184, simple_loss=0.2509, pruned_loss=0.0585, over 4809.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.2491, pruned_loss=0.05312, over 958370.39 frames. ], batch size: 41, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:30:04,886 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.160e+02 1.477e+02 1.740e+02 2.172e+02 4.454e+02, threshold=3.479e+02, percent-clipped=1.0 +2023-03-27 01:30:06,808 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116861.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:30:08,075 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116863.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:30:12,956 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-27 01:30:25,265 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-27 01:30:34,136 INFO [finetune.py:976] (1/7) Epoch 21, batch 2350, loss[loss=0.1425, simple_loss=0.2121, pruned_loss=0.03642, over 4815.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.248, pruned_loss=0.05357, over 956500.60 frames. ], batch size: 38, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:30:40,651 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.59 vs. limit=5.0 +2023-03-27 01:31:07,402 INFO [finetune.py:976] (1/7) Epoch 21, batch 2400, loss[loss=0.1818, simple_loss=0.2473, pruned_loss=0.0581, over 4764.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2446, pruned_loss=0.05242, over 955934.25 frames. ], batch size: 54, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:31:09,766 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.097e+02 1.566e+02 1.863e+02 2.219e+02 3.648e+02, threshold=3.726e+02, percent-clipped=1.0 +2023-03-27 01:31:21,632 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116974.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:31:25,171 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116979.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:31:40,995 INFO [finetune.py:976] (1/7) Epoch 21, batch 2450, loss[loss=0.1909, simple_loss=0.2555, pruned_loss=0.06313, over 4805.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2427, pruned_loss=0.05204, over 957319.24 frames. ], batch size: 45, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:31:42,055 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-27 01:31:50,910 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8469, 3.6459, 3.7751, 1.9324, 3.9645, 2.9883, 1.2913, 2.7518], + device='cuda:1'), covar=tensor([0.2722, 0.2108, 0.1230, 0.2840, 0.0876, 0.0876, 0.3491, 0.1407], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0178, 0.0158, 0.0130, 0.0161, 0.0123, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:31:57,424 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=117027.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:32:14,703 INFO [finetune.py:976] (1/7) Epoch 21, batch 2500, loss[loss=0.2081, simple_loss=0.2713, pruned_loss=0.07247, over 4395.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2451, pruned_loss=0.05297, over 955454.87 frames. ], batch size: 19, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:32:17,112 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.025e+02 1.510e+02 1.797e+02 2.340e+02 3.968e+02, threshold=3.593e+02, percent-clipped=1.0 +2023-03-27 01:32:18,384 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=117060.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:32:46,784 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117102.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:32:47,553 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-27 01:32:47,902 INFO [finetune.py:976] (1/7) Epoch 21, batch 2550, loss[loss=0.1479, simple_loss=0.2212, pruned_loss=0.03729, over 4307.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2496, pruned_loss=0.0548, over 955018.61 frames. ], batch size: 19, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:33:02,850 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4186, 1.3417, 1.4181, 0.7114, 1.3603, 1.4218, 1.3666, 1.2672], + device='cuda:1'), covar=tensor([0.0458, 0.0590, 0.0548, 0.0787, 0.0899, 0.0613, 0.0534, 0.1011], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0134, 0.0138, 0.0119, 0.0124, 0.0138, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:33:19,400 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=117150.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:33:21,775 INFO [finetune.py:976] (1/7) Epoch 21, batch 2600, loss[loss=0.2139, simple_loss=0.2975, pruned_loss=0.0651, over 4861.00 frames. ], tot_loss[loss=0.1804, simple_loss=0.2506, pruned_loss=0.05506, over 952328.15 frames. ], batch size: 44, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:33:24,219 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.102e+02 1.533e+02 1.830e+02 2.226e+02 4.351e+02, threshold=3.661e+02, percent-clipped=3.0 +2023-03-27 01:33:24,297 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=117158.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:33:26,115 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117161.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:33:38,459 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-03-27 01:34:06,271 INFO [finetune.py:976] (1/7) Epoch 21, batch 2650, loss[loss=0.1782, simple_loss=0.2597, pruned_loss=0.04834, over 4817.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2513, pruned_loss=0.05502, over 953848.68 frames. ], batch size: 38, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:34:09,385 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=117209.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:34:26,094 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=117221.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:35:03,766 INFO [finetune.py:976] (1/7) Epoch 21, batch 2700, loss[loss=0.2048, simple_loss=0.2616, pruned_loss=0.074, over 4864.00 frames. ], tot_loss[loss=0.1795, simple_loss=0.2503, pruned_loss=0.05433, over 956311.59 frames. ], batch size: 31, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:35:06,200 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.072e+02 1.522e+02 1.732e+02 2.127e+02 4.053e+02, threshold=3.464e+02, percent-clipped=3.0 +2023-03-27 01:35:23,782 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117274.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:35:30,664 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=117282.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:35:41,279 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2047, 2.1055, 1.7364, 1.8881, 1.9727, 1.9368, 2.0058, 2.6729], + device='cuda:1'), covar=tensor([0.3303, 0.3859, 0.3015, 0.3652, 0.3513, 0.2319, 0.3479, 0.1642], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0261, 0.0232, 0.0276, 0.0252, 0.0222, 0.0251, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:35:45,280 INFO [finetune.py:976] (1/7) Epoch 21, batch 2750, loss[loss=0.1392, simple_loss=0.2243, pruned_loss=0.02707, over 4760.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2477, pruned_loss=0.05324, over 958015.76 frames. ], batch size: 27, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:35:54,487 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7999, 2.8577, 2.6134, 1.8785, 2.7609, 2.8787, 2.9289, 2.5110], + device='cuda:1'), covar=tensor([0.0581, 0.0627, 0.0771, 0.0950, 0.0543, 0.0794, 0.0645, 0.1048], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0136, 0.0140, 0.0121, 0.0125, 0.0139, 0.0141, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:35:56,252 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=117322.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:36:01,561 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=117329.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:36:18,643 INFO [finetune.py:976] (1/7) Epoch 21, batch 2800, loss[loss=0.1809, simple_loss=0.2422, pruned_loss=0.05982, over 4749.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.244, pruned_loss=0.05234, over 956057.49 frames. ], batch size: 54, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:36:21,563 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.020e+02 1.495e+02 1.752e+02 2.115e+02 2.888e+02, threshold=3.503e+02, percent-clipped=0.0 +2023-03-27 01:36:22,888 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117360.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:36:32,671 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-27 01:36:42,939 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=117390.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:36:44,809 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6449, 1.5735, 1.0618, 0.2847, 1.2560, 1.4964, 1.5399, 1.4602], + device='cuda:1'), covar=tensor([0.0836, 0.0746, 0.1254, 0.1962, 0.1363, 0.2319, 0.2052, 0.0844], + device='cuda:1'), in_proj_covar=tensor([0.0168, 0.0189, 0.0196, 0.0181, 0.0207, 0.0206, 0.0219, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:36:49,648 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6049, 1.5521, 2.1703, 3.2372, 2.2161, 2.3305, 1.2562, 2.7698], + device='cuda:1'), covar=tensor([0.1860, 0.1494, 0.1290, 0.0722, 0.0816, 0.1403, 0.1715, 0.0507], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0163, 0.0101, 0.0136, 0.0124, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 01:36:52,478 INFO [finetune.py:976] (1/7) Epoch 21, batch 2850, loss[loss=0.1368, simple_loss=0.2054, pruned_loss=0.03407, over 4774.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2424, pruned_loss=0.05192, over 955636.95 frames. ], batch size: 27, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:36:52,584 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6291, 1.5040, 1.4680, 1.5683, 0.9705, 3.4873, 1.3126, 1.7711], + device='cuda:1'), covar=tensor([0.3177, 0.2443, 0.2137, 0.2351, 0.1923, 0.0188, 0.2604, 0.1227], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0115, 0.0120, 0.0122, 0.0113, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 01:36:54,970 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=117408.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:37:25,543 INFO [finetune.py:976] (1/7) Epoch 21, batch 2900, loss[loss=0.1831, simple_loss=0.2555, pruned_loss=0.05532, over 4824.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2454, pruned_loss=0.0532, over 956828.97 frames. ], batch size: 30, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:37:28,393 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.016e+02 1.554e+02 1.875e+02 2.295e+02 6.888e+02, threshold=3.749e+02, percent-clipped=2.0 +2023-03-27 01:37:28,494 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117458.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:37:52,811 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9914, 1.8844, 1.9782, 1.3234, 1.9091, 2.0452, 2.0349, 1.6059], + device='cuda:1'), covar=tensor([0.0620, 0.0705, 0.0683, 0.0852, 0.0731, 0.0705, 0.0590, 0.1142], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0136, 0.0140, 0.0121, 0.0126, 0.0139, 0.0141, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:37:59,199 INFO [finetune.py:976] (1/7) Epoch 21, batch 2950, loss[loss=0.2468, simple_loss=0.3063, pruned_loss=0.09363, over 4226.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2488, pruned_loss=0.05392, over 957206.57 frames. ], batch size: 65, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:38:00,488 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=117506.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:38:32,247 INFO [finetune.py:976] (1/7) Epoch 21, batch 3000, loss[loss=0.1673, simple_loss=0.2458, pruned_loss=0.04442, over 4839.00 frames. ], tot_loss[loss=0.1811, simple_loss=0.2518, pruned_loss=0.05521, over 958487.65 frames. ], batch size: 44, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:38:32,247 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 01:38:34,103 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8938, 3.4335, 3.5746, 3.7869, 3.6371, 3.4889, 3.9435, 1.3648], + device='cuda:1'), covar=tensor([0.0965, 0.0903, 0.0991, 0.0940, 0.1497, 0.1663, 0.0828, 0.5325], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0240, 0.0278, 0.0290, 0.0331, 0.0282, 0.0301, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:38:35,404 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0787, 1.2094, 2.1944, 1.9992, 1.9593, 1.8035, 1.8275, 2.0382], + device='cuda:1'), covar=tensor([0.3433, 0.3615, 0.3255, 0.3609, 0.4810, 0.3357, 0.4265, 0.2918], + device='cuda:1'), in_proj_covar=tensor([0.0255, 0.0243, 0.0264, 0.0283, 0.0281, 0.0257, 0.0291, 0.0245], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:38:38,789 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6159, 3.5091, 3.3798, 1.4945, 3.6556, 2.8267, 0.8490, 2.3999], + device='cuda:1'), covar=tensor([0.2209, 0.1871, 0.1539, 0.3545, 0.1108, 0.0960, 0.4172, 0.1684], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0180, 0.0159, 0.0131, 0.0162, 0.0124, 0.0149, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:38:38,975 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2330, 1.9842, 1.8840, 1.9035, 1.9437, 1.9728, 1.9841, 2.6169], + device='cuda:1'), covar=tensor([0.3771, 0.4429, 0.3249, 0.3792, 0.4041, 0.2491, 0.3836, 0.1858], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0262, 0.0232, 0.0276, 0.0252, 0.0222, 0.0251, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:38:42,800 INFO [finetune.py:1010] (1/7) Epoch 21, validation: loss=0.1567, simple_loss=0.2253, pruned_loss=0.04408, over 2265189.00 frames. +2023-03-27 01:38:42,800 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 01:38:45,676 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.002e+02 1.534e+02 1.924e+02 2.362e+02 3.621e+02, threshold=3.849e+02, percent-clipped=0.0 +2023-03-27 01:38:50,641 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9256, 4.7914, 4.5586, 2.5624, 4.9083, 3.9061, 1.1309, 3.3133], + device='cuda:1'), covar=tensor([0.2122, 0.1461, 0.1273, 0.2728, 0.0772, 0.0722, 0.4132, 0.1335], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0180, 0.0159, 0.0131, 0.0162, 0.0124, 0.0148, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:39:00,137 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=117577.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:39:17,642 INFO [finetune.py:976] (1/7) Epoch 21, batch 3050, loss[loss=0.1492, simple_loss=0.2391, pruned_loss=0.02967, over 4904.00 frames. ], tot_loss[loss=0.1809, simple_loss=0.2525, pruned_loss=0.05464, over 958733.27 frames. ], batch size: 37, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:40:13,671 INFO [finetune.py:976] (1/7) Epoch 21, batch 3100, loss[loss=0.1241, simple_loss=0.1989, pruned_loss=0.0246, over 4826.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2496, pruned_loss=0.05371, over 958034.29 frames. ], batch size: 39, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:40:19,667 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.886e+01 1.482e+02 1.759e+02 2.208e+02 4.258e+02, threshold=3.518e+02, percent-clipped=1.0 +2023-03-27 01:40:46,221 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1021, 1.8410, 2.3287, 1.4314, 2.0631, 2.2974, 1.5996, 2.4476], + device='cuda:1'), covar=tensor([0.1182, 0.1773, 0.1266, 0.2018, 0.0872, 0.1263, 0.2875, 0.0732], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0205, 0.0191, 0.0190, 0.0174, 0.0214, 0.0218, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:40:46,800 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=117685.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:40:58,317 INFO [finetune.py:976] (1/7) Epoch 21, batch 3150, loss[loss=0.2012, simple_loss=0.2745, pruned_loss=0.0639, over 4917.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2468, pruned_loss=0.05329, over 956546.18 frames. ], batch size: 46, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:41:31,632 INFO [finetune.py:976] (1/7) Epoch 21, batch 3200, loss[loss=0.1721, simple_loss=0.2398, pruned_loss=0.05222, over 4878.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2438, pruned_loss=0.05204, over 956487.07 frames. ], batch size: 31, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:41:34,036 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.069e+02 1.569e+02 1.801e+02 2.101e+02 4.822e+02, threshold=3.602e+02, percent-clipped=2.0 +2023-03-27 01:42:02,036 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8196, 3.8982, 3.6633, 1.7630, 4.0435, 3.0960, 0.9341, 2.7184], + device='cuda:1'), covar=tensor([0.2007, 0.2063, 0.1376, 0.3430, 0.0942, 0.0913, 0.4441, 0.1396], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0178, 0.0157, 0.0130, 0.0160, 0.0122, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:42:05,171 INFO [finetune.py:976] (1/7) Epoch 21, batch 3250, loss[loss=0.2153, simple_loss=0.2716, pruned_loss=0.07957, over 4919.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2433, pruned_loss=0.05203, over 956390.33 frames. ], batch size: 36, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:42:31,252 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7070, 1.0091, 1.7453, 1.6997, 1.4807, 1.4602, 1.6650, 1.6485], + device='cuda:1'), covar=tensor([0.3252, 0.3537, 0.2808, 0.2976, 0.4231, 0.3305, 0.3373, 0.2654], + device='cuda:1'), in_proj_covar=tensor([0.0254, 0.0243, 0.0263, 0.0283, 0.0281, 0.0257, 0.0290, 0.0245], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:42:33,050 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7680, 1.7469, 1.5950, 1.7000, 1.3896, 4.3344, 1.5986, 2.0728], + device='cuda:1'), covar=tensor([0.3291, 0.2555, 0.2126, 0.2523, 0.1706, 0.0159, 0.2560, 0.1247], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0122, 0.0124, 0.0114, 0.0097, 0.0096, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 01:42:38,404 INFO [finetune.py:976] (1/7) Epoch 21, batch 3300, loss[loss=0.199, simple_loss=0.2537, pruned_loss=0.07215, over 3827.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2464, pruned_loss=0.05289, over 955742.91 frames. ], batch size: 16, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:42:40,846 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.148e+02 1.638e+02 1.917e+02 2.241e+02 9.038e+02, threshold=3.833e+02, percent-clipped=2.0 +2023-03-27 01:42:54,833 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117877.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:43:11,570 INFO [finetune.py:976] (1/7) Epoch 21, batch 3350, loss[loss=0.1449, simple_loss=0.2183, pruned_loss=0.03578, over 4176.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2494, pruned_loss=0.05387, over 955613.51 frames. ], batch size: 18, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:43:22,937 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4443, 2.5203, 2.3740, 1.7096, 2.2998, 2.6821, 2.5873, 2.1568], + device='cuda:1'), covar=tensor([0.0577, 0.0574, 0.0715, 0.0877, 0.1020, 0.0659, 0.0612, 0.1043], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0136, 0.0139, 0.0120, 0.0125, 0.0138, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:43:25,774 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=117925.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:43:45,043 INFO [finetune.py:976] (1/7) Epoch 21, batch 3400, loss[loss=0.181, simple_loss=0.2625, pruned_loss=0.04974, over 4899.00 frames. ], tot_loss[loss=0.1788, simple_loss=0.2503, pruned_loss=0.05368, over 955220.41 frames. ], batch size: 37, lr: 3.19e-03, grad_scale: 32.0 +2023-03-27 01:43:47,450 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.189e+02 1.619e+02 1.880e+02 2.233e+02 5.629e+02, threshold=3.761e+02, percent-clipped=2.0 +2023-03-27 01:44:05,854 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117985.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:44:19,699 INFO [finetune.py:976] (1/7) Epoch 21, batch 3450, loss[loss=0.2125, simple_loss=0.2688, pruned_loss=0.07809, over 4922.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2505, pruned_loss=0.0536, over 956558.67 frames. ], batch size: 42, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:44:39,326 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=118033.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:44:44,120 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=118040.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:44:59,761 INFO [finetune.py:976] (1/7) Epoch 21, batch 3500, loss[loss=0.1618, simple_loss=0.2318, pruned_loss=0.04586, over 4931.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2484, pruned_loss=0.05319, over 956432.54 frames. ], batch size: 33, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:45:02,215 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.127e+01 1.501e+02 1.833e+02 2.184e+02 3.839e+02, threshold=3.666e+02, percent-clipped=2.0 +2023-03-27 01:45:12,216 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5387, 1.7350, 1.6849, 0.9410, 1.9619, 2.1316, 1.9812, 1.6014], + device='cuda:1'), covar=tensor([0.1002, 0.0616, 0.0535, 0.0618, 0.0388, 0.0603, 0.0383, 0.0681], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0150, 0.0126, 0.0124, 0.0131, 0.0129, 0.0142, 0.0148], + device='cuda:1'), out_proj_covar=tensor([9.0317e-05, 1.0823e-04, 9.0428e-05, 8.7432e-05, 9.2007e-05, 9.2059e-05, + 1.0189e-04, 1.0648e-04], device='cuda:1') +2023-03-27 01:45:27,160 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6144, 3.6905, 3.5106, 1.6611, 3.7718, 2.9130, 1.1498, 2.5697], + device='cuda:1'), covar=tensor([0.2227, 0.1740, 0.1345, 0.2989, 0.1005, 0.0869, 0.3566, 0.1339], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0177, 0.0157, 0.0129, 0.0160, 0.0122, 0.0146, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 01:45:52,328 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=118101.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:45:57,919 INFO [finetune.py:976] (1/7) Epoch 21, batch 3550, loss[loss=0.1338, simple_loss=0.2066, pruned_loss=0.03047, over 4915.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2468, pruned_loss=0.05329, over 957254.71 frames. ], batch size: 43, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:46:24,811 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3131, 2.2074, 1.7412, 1.9931, 2.2174, 1.9952, 2.4974, 2.2882], + device='cuda:1'), covar=tensor([0.1268, 0.1835, 0.2977, 0.2434, 0.2386, 0.1704, 0.2524, 0.1753], + device='cuda:1'), in_proj_covar=tensor([0.0187, 0.0188, 0.0235, 0.0253, 0.0247, 0.0204, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:46:30,089 INFO [finetune.py:976] (1/7) Epoch 21, batch 3600, loss[loss=0.2112, simple_loss=0.2794, pruned_loss=0.07149, over 4730.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2441, pruned_loss=0.05284, over 955403.51 frames. ], batch size: 59, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:46:33,039 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.102e+02 1.558e+02 1.902e+02 2.180e+02 3.976e+02, threshold=3.804e+02, percent-clipped=2.0 +2023-03-27 01:46:39,829 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=118168.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:46:44,694 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4139, 2.3023, 2.2729, 1.6455, 2.1746, 2.4678, 2.4620, 1.9161], + device='cuda:1'), covar=tensor([0.0497, 0.0566, 0.0654, 0.0880, 0.0724, 0.0552, 0.0475, 0.1072], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0135, 0.0138, 0.0120, 0.0124, 0.0137, 0.0139, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:47:03,624 INFO [finetune.py:976] (1/7) Epoch 21, batch 3650, loss[loss=0.2057, simple_loss=0.2737, pruned_loss=0.06884, over 4759.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2477, pruned_loss=0.05436, over 955713.09 frames. ], batch size: 59, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:47:19,799 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=118229.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 01:47:36,721 INFO [finetune.py:976] (1/7) Epoch 21, batch 3700, loss[loss=0.192, simple_loss=0.2682, pruned_loss=0.05791, over 4865.00 frames. ], tot_loss[loss=0.1807, simple_loss=0.2512, pruned_loss=0.05516, over 955014.53 frames. ], batch size: 31, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:47:39,056 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.015e+02 1.607e+02 1.941e+02 2.377e+02 3.454e+02, threshold=3.882e+02, percent-clipped=0.0 +2023-03-27 01:47:46,808 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=118269.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:48:00,446 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=118289.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:48:10,297 INFO [finetune.py:976] (1/7) Epoch 21, batch 3750, loss[loss=0.2137, simple_loss=0.2785, pruned_loss=0.07444, over 4855.00 frames. ], tot_loss[loss=0.181, simple_loss=0.252, pruned_loss=0.05505, over 954716.89 frames. ], batch size: 31, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:48:24,530 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9105, 1.3602, 0.9507, 1.7637, 2.3003, 1.6813, 1.7552, 1.7165], + device='cuda:1'), covar=tensor([0.1391, 0.2070, 0.1918, 0.1208, 0.1782, 0.1840, 0.1473, 0.1971], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0111, 0.0092, 0.0120, 0.0093, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 01:48:26,351 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=118329.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:48:26,967 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=118330.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:48:40,932 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=118350.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:48:43,713 INFO [finetune.py:976] (1/7) Epoch 21, batch 3800, loss[loss=0.1422, simple_loss=0.2066, pruned_loss=0.03883, over 4721.00 frames. ], tot_loss[loss=0.1825, simple_loss=0.2534, pruned_loss=0.05583, over 952442.13 frames. ], batch size: 23, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:48:46,094 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.364e+01 1.561e+02 1.815e+02 2.293e+02 4.441e+02, threshold=3.631e+02, percent-clipped=1.0 +2023-03-27 01:49:06,631 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=118390.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:49:11,212 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118396.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:49:17,068 INFO [finetune.py:976] (1/7) Epoch 21, batch 3850, loss[loss=0.219, simple_loss=0.2721, pruned_loss=0.08294, over 4782.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.2515, pruned_loss=0.0545, over 952623.08 frames. ], batch size: 29, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:49:50,288 INFO [finetune.py:976] (1/7) Epoch 21, batch 3900, loss[loss=0.1527, simple_loss=0.2162, pruned_loss=0.04457, over 4914.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2485, pruned_loss=0.05335, over 952045.05 frames. ], batch size: 36, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:49:52,685 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.530e+02 1.819e+02 2.327e+02 4.856e+02, threshold=3.639e+02, percent-clipped=2.0 +2023-03-27 01:50:03,439 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-03-27 01:50:25,015 INFO [finetune.py:976] (1/7) Epoch 21, batch 3950, loss[loss=0.1783, simple_loss=0.2448, pruned_loss=0.05587, over 4903.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2448, pruned_loss=0.05191, over 953532.83 frames. ], batch size: 32, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:50:39,350 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.2844, 1.5196, 1.5735, 0.8074, 1.5480, 1.8328, 1.8515, 1.4790], + device='cuda:1'), covar=tensor([0.1004, 0.0647, 0.0542, 0.0574, 0.0554, 0.0538, 0.0370, 0.0769], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0151, 0.0127, 0.0124, 0.0131, 0.0129, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([9.0495e-05, 1.0882e-04, 9.0761e-05, 8.7562e-05, 9.2534e-05, 9.2078e-05, + 1.0178e-04, 1.0664e-04], device='cuda:1') +2023-03-27 01:50:41,209 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1623, 1.7022, 2.3132, 1.5870, 2.1560, 2.3069, 1.6589, 2.3967], + device='cuda:1'), covar=tensor([0.1149, 0.2053, 0.1569, 0.2085, 0.0775, 0.1298, 0.2859, 0.0789], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0206, 0.0192, 0.0191, 0.0176, 0.0215, 0.0219, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:50:42,576 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.59 vs. limit=5.0 +2023-03-27 01:50:43,602 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118524.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 01:51:19,653 INFO [finetune.py:976] (1/7) Epoch 21, batch 4000, loss[loss=0.1599, simple_loss=0.2233, pruned_loss=0.04828, over 4230.00 frames. ], tot_loss[loss=0.1733, simple_loss=0.2434, pruned_loss=0.05156, over 951684.26 frames. ], batch size: 18, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:51:26,592 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.742e+01 1.543e+02 1.809e+02 2.201e+02 4.154e+02, threshold=3.618e+02, percent-clipped=2.0 +2023-03-27 01:51:56,640 INFO [finetune.py:976] (1/7) Epoch 21, batch 4050, loss[loss=0.1692, simple_loss=0.2495, pruned_loss=0.04442, over 4765.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.247, pruned_loss=0.0534, over 949831.21 frames. ], batch size: 28, lr: 3.18e-03, grad_scale: 32.0 +2023-03-27 01:52:07,461 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.83 vs. limit=5.0 +2023-03-27 01:52:11,486 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118625.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:52:15,695 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=118631.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:52:19,358 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-03-27 01:52:24,051 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7144, 1.2692, 0.8697, 1.6544, 2.1921, 1.1271, 1.4666, 1.5677], + device='cuda:1'), covar=tensor([0.1546, 0.2188, 0.1990, 0.1284, 0.1999, 0.2072, 0.1626, 0.2091], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0110, 0.0092, 0.0119, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 01:52:24,636 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118645.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:52:30,074 INFO [finetune.py:976] (1/7) Epoch 21, batch 4100, loss[loss=0.1645, simple_loss=0.2301, pruned_loss=0.04946, over 4773.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2482, pruned_loss=0.05338, over 950413.05 frames. ], batch size: 26, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:52:33,507 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.828e+01 1.601e+02 1.824e+02 2.338e+02 3.980e+02, threshold=3.647e+02, percent-clipped=1.0 +2023-03-27 01:52:51,561 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118685.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:52:56,323 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=118692.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:52:58,681 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118696.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:53:03,515 INFO [finetune.py:976] (1/7) Epoch 21, batch 4150, loss[loss=0.1784, simple_loss=0.2532, pruned_loss=0.05186, over 4910.00 frames. ], tot_loss[loss=0.1802, simple_loss=0.251, pruned_loss=0.05472, over 951731.16 frames. ], batch size: 37, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:53:31,226 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=118744.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:53:37,401 INFO [finetune.py:976] (1/7) Epoch 21, batch 4200, loss[loss=0.1636, simple_loss=0.2382, pruned_loss=0.04449, over 4817.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.25, pruned_loss=0.05325, over 953916.90 frames. ], batch size: 39, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:53:39,820 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.932e+01 1.492e+02 1.761e+02 2.066e+02 3.902e+02, threshold=3.521e+02, percent-clipped=1.0 +2023-03-27 01:53:40,022 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.52 vs. limit=5.0 +2023-03-27 01:53:45,823 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-27 01:53:47,362 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6646, 1.4780, 1.8569, 3.0021, 2.0325, 2.2115, 0.9590, 2.5389], + device='cuda:1'), covar=tensor([0.1643, 0.1406, 0.1325, 0.0671, 0.0804, 0.1190, 0.1771, 0.0536], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0163, 0.0101, 0.0137, 0.0125, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 01:54:11,363 INFO [finetune.py:976] (1/7) Epoch 21, batch 4250, loss[loss=0.1741, simple_loss=0.2504, pruned_loss=0.0489, over 4908.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2475, pruned_loss=0.05252, over 951579.25 frames. ], batch size: 43, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:54:25,964 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118824.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:54:42,830 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2483, 2.1448, 1.7370, 2.2318, 2.1823, 1.9291, 2.5301, 2.2685], + device='cuda:1'), covar=tensor([0.1351, 0.2144, 0.3209, 0.2510, 0.2679, 0.1695, 0.2954, 0.1757], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0189, 0.0237, 0.0254, 0.0248, 0.0206, 0.0216, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:54:45,138 INFO [finetune.py:976] (1/7) Epoch 21, batch 4300, loss[loss=0.139, simple_loss=0.2089, pruned_loss=0.03455, over 4807.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2452, pruned_loss=0.05174, over 953060.97 frames. ], batch size: 51, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:54:47,580 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.108e+02 1.563e+02 1.855e+02 2.179e+02 3.656e+02, threshold=3.709e+02, percent-clipped=1.0 +2023-03-27 01:54:57,550 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=118872.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:55:14,853 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9865, 1.8910, 1.7595, 2.2149, 2.3531, 2.1847, 1.6859, 1.6103], + device='cuda:1'), covar=tensor([0.2215, 0.1990, 0.1911, 0.1447, 0.1722, 0.1124, 0.2383, 0.1967], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0214, 0.0196, 0.0244, 0.0189, 0.0218, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:55:18,872 INFO [finetune.py:976] (1/7) Epoch 21, batch 4350, loss[loss=0.159, simple_loss=0.2293, pruned_loss=0.04431, over 4909.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2421, pruned_loss=0.05107, over 953955.57 frames. ], batch size: 35, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:55:20,347 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.68 vs. limit=5.0 +2023-03-27 01:55:33,249 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118925.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:55:48,468 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118945.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:55:59,297 INFO [finetune.py:976] (1/7) Epoch 21, batch 4400, loss[loss=0.1303, simple_loss=0.2049, pruned_loss=0.02779, over 4770.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2436, pruned_loss=0.05235, over 953832.06 frames. ], batch size: 26, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:56:01,711 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.753e+01 1.466e+02 1.745e+02 2.136e+02 3.634e+02, threshold=3.490e+02, percent-clipped=0.0 +2023-03-27 01:56:17,886 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=118973.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:56:17,939 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4416, 1.3331, 1.3290, 1.3171, 0.8882, 2.3422, 0.7937, 1.2509], + device='cuda:1'), covar=tensor([0.3301, 0.2653, 0.2203, 0.2498, 0.1842, 0.0335, 0.2878, 0.1318], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0117, 0.0121, 0.0124, 0.0114, 0.0097, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 01:56:31,289 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118985.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:56:36,933 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118987.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:56:40,525 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=118993.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:56:51,290 INFO [finetune.py:976] (1/7) Epoch 21, batch 4450, loss[loss=0.1977, simple_loss=0.2632, pruned_loss=0.06603, over 4867.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2471, pruned_loss=0.05356, over 952726.36 frames. ], batch size: 34, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:57:11,406 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=119033.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:57:16,764 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4886, 1.6011, 1.4724, 0.8566, 1.6514, 1.8529, 1.8630, 1.3990], + device='cuda:1'), covar=tensor([0.1014, 0.0570, 0.0520, 0.0501, 0.0376, 0.0500, 0.0309, 0.0704], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0149, 0.0126, 0.0123, 0.0130, 0.0128, 0.0141, 0.0147], + device='cuda:1'), out_proj_covar=tensor([8.9306e-05, 1.0747e-04, 8.9880e-05, 8.6632e-05, 9.1639e-05, 9.1562e-05, + 1.0119e-04, 1.0565e-04], device='cuda:1') +2023-03-27 01:57:20,390 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-27 01:57:25,001 INFO [finetune.py:976] (1/7) Epoch 21, batch 4500, loss[loss=0.1757, simple_loss=0.2551, pruned_loss=0.04813, over 4852.00 frames. ], tot_loss[loss=0.179, simple_loss=0.2498, pruned_loss=0.05409, over 952321.35 frames. ], batch size: 44, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:57:27,416 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.079e+02 1.548e+02 1.910e+02 2.429e+02 4.520e+02, threshold=3.820e+02, percent-clipped=3.0 +2023-03-27 01:57:37,162 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1314, 2.1380, 1.8097, 2.2131, 2.0970, 2.0581, 2.0596, 2.6700], + device='cuda:1'), covar=tensor([0.3801, 0.5021, 0.3534, 0.4112, 0.4572, 0.2563, 0.4459, 0.1773], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0262, 0.0232, 0.0277, 0.0254, 0.0224, 0.0253, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:57:38,337 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119075.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:57:49,668 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6570, 1.4705, 1.0332, 0.2711, 1.2634, 1.4660, 1.3586, 1.3899], + device='cuda:1'), covar=tensor([0.1026, 0.0857, 0.1573, 0.2116, 0.1510, 0.2548, 0.2574, 0.0978], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0191, 0.0199, 0.0183, 0.0210, 0.0209, 0.0223, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:57:51,475 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119093.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:57:58,450 INFO [finetune.py:976] (1/7) Epoch 21, batch 4550, loss[loss=0.1633, simple_loss=0.2414, pruned_loss=0.04256, over 4785.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2512, pruned_loss=0.0543, over 952964.23 frames. ], batch size: 51, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:58:19,744 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119136.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 01:58:31,550 INFO [finetune.py:976] (1/7) Epoch 21, batch 4600, loss[loss=0.2186, simple_loss=0.2773, pruned_loss=0.07995, over 4821.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2497, pruned_loss=0.05347, over 953210.09 frames. ], batch size: 30, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:58:31,664 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119154.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 01:58:34,457 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.081e+02 1.653e+02 1.869e+02 2.317e+02 3.451e+02, threshold=3.738e+02, percent-clipped=0.0 +2023-03-27 01:58:59,853 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0513, 1.9088, 1.6167, 1.5876, 1.7889, 1.7813, 1.8351, 2.4966], + device='cuda:1'), covar=tensor([0.3246, 0.3415, 0.2935, 0.3348, 0.3462, 0.2099, 0.3080, 0.1538], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0232, 0.0277, 0.0254, 0.0224, 0.0253, 0.0236], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:59:05,261 INFO [finetune.py:976] (1/7) Epoch 21, batch 4650, loss[loss=0.1488, simple_loss=0.2197, pruned_loss=0.03893, over 4868.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2472, pruned_loss=0.05285, over 954214.49 frames. ], batch size: 31, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:59:05,384 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5317, 2.4837, 2.0221, 0.9526, 2.2042, 2.0093, 1.8868, 2.2310], + device='cuda:1'), covar=tensor([0.0889, 0.0757, 0.1652, 0.2227, 0.1467, 0.2243, 0.2109, 0.0969], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0192, 0.0200, 0.0183, 0.0211, 0.0210, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 01:59:37,813 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7734, 1.2130, 0.7705, 1.6831, 2.2910, 1.4831, 1.3819, 1.4711], + device='cuda:1'), covar=tensor([0.1569, 0.2470, 0.2310, 0.1382, 0.1822, 0.2001, 0.1831, 0.2335], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0111, 0.0092, 0.0120, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 01:59:37,963 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.74 vs. limit=5.0 +2023-03-27 01:59:38,313 INFO [finetune.py:976] (1/7) Epoch 21, batch 4700, loss[loss=0.1544, simple_loss=0.2135, pruned_loss=0.04762, over 4773.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2433, pruned_loss=0.05129, over 955617.54 frames. ], batch size: 28, lr: 3.18e-03, grad_scale: 64.0 +2023-03-27 01:59:40,726 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.102e+02 1.521e+02 1.791e+02 2.382e+02 6.096e+02, threshold=3.583e+02, percent-clipped=7.0 +2023-03-27 01:59:59,485 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=119287.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:00:05,290 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.3104, 2.9258, 3.0709, 3.2525, 3.1049, 2.9391, 3.3582, 0.9469], + device='cuda:1'), covar=tensor([0.1217, 0.1058, 0.1063, 0.1204, 0.1784, 0.1843, 0.1127, 0.6068], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0244, 0.0280, 0.0293, 0.0335, 0.0285, 0.0304, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:00:11,568 INFO [finetune.py:976] (1/7) Epoch 21, batch 4750, loss[loss=0.1605, simple_loss=0.2289, pruned_loss=0.04605, over 4872.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2422, pruned_loss=0.05184, over 954785.02 frames. ], batch size: 31, lr: 3.17e-03, grad_scale: 64.0 +2023-03-27 02:00:31,332 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=119335.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:00:36,548 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8558, 2.0799, 1.7924, 1.9161, 2.5701, 2.6265, 2.0378, 2.0256], + device='cuda:1'), covar=tensor([0.0397, 0.0338, 0.0613, 0.0339, 0.0253, 0.0397, 0.0410, 0.0387], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0106, 0.0143, 0.0111, 0.0099, 0.0110, 0.0100, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.5983e-05, 8.1465e-05, 1.1248e-04, 8.5469e-05, 7.6758e-05, 8.1520e-05, + 7.4604e-05, 8.5532e-05], device='cuda:1') +2023-03-27 02:00:44,656 INFO [finetune.py:976] (1/7) Epoch 21, batch 4800, loss[loss=0.1552, simple_loss=0.2326, pruned_loss=0.03894, over 4934.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2447, pruned_loss=0.05214, over 956342.75 frames. ], batch size: 33, lr: 3.17e-03, grad_scale: 64.0 +2023-03-27 02:00:47,502 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.052e+02 1.488e+02 1.781e+02 2.195e+02 3.360e+02, threshold=3.562e+02, percent-clipped=0.0 +2023-03-27 02:01:22,405 INFO [finetune.py:976] (1/7) Epoch 21, batch 4850, loss[loss=0.1383, simple_loss=0.208, pruned_loss=0.03427, over 4786.00 frames. ], tot_loss[loss=0.1777, simple_loss=0.2485, pruned_loss=0.05347, over 952530.94 frames. ], batch size: 29, lr: 3.17e-03, grad_scale: 64.0 +2023-03-27 02:01:55,523 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=119431.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:02:15,264 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=119449.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 02:02:19,177 INFO [finetune.py:976] (1/7) Epoch 21, batch 4900, loss[loss=0.2126, simple_loss=0.2759, pruned_loss=0.07467, over 4851.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2497, pruned_loss=0.05371, over 952608.19 frames. ], batch size: 44, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:02:25,214 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.094e+02 1.684e+02 1.937e+02 2.365e+02 4.201e+02, threshold=3.874e+02, percent-clipped=2.0 +2023-03-27 02:02:55,580 INFO [finetune.py:976] (1/7) Epoch 21, batch 4950, loss[loss=0.1345, simple_loss=0.2122, pruned_loss=0.02838, over 4748.00 frames. ], tot_loss[loss=0.1782, simple_loss=0.2496, pruned_loss=0.05341, over 952196.53 frames. ], batch size: 23, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:03:02,768 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119514.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:03:29,004 INFO [finetune.py:976] (1/7) Epoch 21, batch 5000, loss[loss=0.1686, simple_loss=0.2387, pruned_loss=0.04925, over 4746.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2473, pruned_loss=0.0526, over 953051.38 frames. ], batch size: 27, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:03:30,186 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1834, 2.1103, 2.1948, 1.5469, 2.1972, 2.4279, 2.3631, 1.6797], + device='cuda:1'), covar=tensor([0.0644, 0.0732, 0.0683, 0.0924, 0.0672, 0.0617, 0.0590, 0.1339], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0136, 0.0139, 0.0120, 0.0126, 0.0138, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:03:32,979 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.061e+02 1.554e+02 1.853e+02 2.138e+02 3.358e+02, threshold=3.705e+02, percent-clipped=0.0 +2023-03-27 02:03:43,305 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119575.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:04:02,226 INFO [finetune.py:976] (1/7) Epoch 21, batch 5050, loss[loss=0.1577, simple_loss=0.2339, pruned_loss=0.04079, over 4836.00 frames. ], tot_loss[loss=0.1733, simple_loss=0.2439, pruned_loss=0.05133, over 954477.82 frames. ], batch size: 33, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:04:31,735 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5239, 3.8755, 4.1622, 4.4029, 4.3064, 4.0493, 4.6117, 1.4381], + device='cuda:1'), covar=tensor([0.0818, 0.0925, 0.0787, 0.0935, 0.1201, 0.1490, 0.0733, 0.5893], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0245, 0.0283, 0.0295, 0.0337, 0.0286, 0.0306, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:04:35,255 INFO [finetune.py:976] (1/7) Epoch 21, batch 5100, loss[loss=0.146, simple_loss=0.2139, pruned_loss=0.0391, over 4820.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2418, pruned_loss=0.05069, over 955748.96 frames. ], batch size: 51, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:04:39,201 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.052e+02 1.479e+02 1.750e+02 2.173e+02 3.976e+02, threshold=3.500e+02, percent-clipped=1.0 +2023-03-27 02:04:43,509 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119665.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:05:08,859 INFO [finetune.py:976] (1/7) Epoch 21, batch 5150, loss[loss=0.1711, simple_loss=0.2427, pruned_loss=0.04978, over 4765.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2425, pruned_loss=0.05112, over 955889.90 frames. ], batch size: 28, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:05:24,657 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119726.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:05:27,582 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=119731.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:05:38,883 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=119749.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 02:05:42,231 INFO [finetune.py:976] (1/7) Epoch 21, batch 5200, loss[loss=0.2254, simple_loss=0.3004, pruned_loss=0.07522, over 4849.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2469, pruned_loss=0.05263, over 955036.62 frames. ], batch size: 44, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:05:45,722 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.011e+02 1.656e+02 1.877e+02 2.270e+02 4.720e+02, threshold=3.754e+02, percent-clipped=1.0 +2023-03-27 02:05:59,325 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=119779.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:06:10,788 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=119797.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:06:15,130 INFO [finetune.py:976] (1/7) Epoch 21, batch 5250, loss[loss=0.2001, simple_loss=0.2632, pruned_loss=0.06854, over 4897.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2482, pruned_loss=0.05244, over 955446.90 frames. ], batch size: 35, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:06:53,711 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8181, 1.7867, 2.1806, 3.3544, 2.3453, 2.3691, 1.1122, 2.7362], + device='cuda:1'), covar=tensor([0.1641, 0.1324, 0.1363, 0.0641, 0.0794, 0.1384, 0.1885, 0.0554], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0134, 0.0166, 0.0102, 0.0139, 0.0126, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 02:06:59,459 INFO [finetune.py:976] (1/7) Epoch 21, batch 5300, loss[loss=0.1876, simple_loss=0.2432, pruned_loss=0.06596, over 4185.00 frames. ], tot_loss[loss=0.1789, simple_loss=0.2506, pruned_loss=0.05362, over 955051.69 frames. ], batch size: 65, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:07:07,164 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.978e+01 1.512e+02 1.747e+02 2.069e+02 4.039e+02, threshold=3.495e+02, percent-clipped=1.0 +2023-03-27 02:07:18,529 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=119870.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:07:54,831 INFO [finetune.py:976] (1/7) Epoch 21, batch 5350, loss[loss=0.2136, simple_loss=0.279, pruned_loss=0.07407, over 4888.00 frames. ], tot_loss[loss=0.1792, simple_loss=0.2512, pruned_loss=0.05364, over 955993.67 frames. ], batch size: 43, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:08:00,323 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0754, 1.8920, 1.6870, 1.7727, 1.8007, 1.8156, 1.7938, 2.5334], + device='cuda:1'), covar=tensor([0.3428, 0.3730, 0.3093, 0.3543, 0.3815, 0.2271, 0.3636, 0.1620], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0261, 0.0231, 0.0276, 0.0252, 0.0222, 0.0252, 0.0233], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:08:24,169 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3261, 1.2648, 1.2132, 1.3751, 1.5699, 1.4812, 1.2985, 1.2105], + device='cuda:1'), covar=tensor([0.0330, 0.0283, 0.0587, 0.0262, 0.0211, 0.0456, 0.0325, 0.0408], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0107, 0.0144, 0.0112, 0.0099, 0.0111, 0.0101, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.6238e-05, 8.2077e-05, 1.1342e-04, 8.5878e-05, 7.7382e-05, 8.2362e-05, + 7.5034e-05, 8.6290e-05], device='cuda:1') +2023-03-27 02:08:28,093 INFO [finetune.py:976] (1/7) Epoch 21, batch 5400, loss[loss=0.16, simple_loss=0.2253, pruned_loss=0.04738, over 4117.00 frames. ], tot_loss[loss=0.1773, simple_loss=0.2488, pruned_loss=0.0529, over 956598.33 frames. ], batch size: 65, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:08:31,173 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.038e+02 1.474e+02 1.662e+02 2.015e+02 3.492e+02, threshold=3.324e+02, percent-clipped=0.0 +2023-03-27 02:08:36,069 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8181, 1.6778, 1.5434, 1.9439, 1.9659, 1.9480, 1.3690, 1.5173], + device='cuda:1'), covar=tensor([0.2286, 0.2095, 0.1981, 0.1652, 0.1626, 0.1197, 0.2461, 0.2066], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0211, 0.0214, 0.0195, 0.0244, 0.0189, 0.0218, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:09:02,904 INFO [finetune.py:976] (1/7) Epoch 21, batch 5450, loss[loss=0.1795, simple_loss=0.2561, pruned_loss=0.05151, over 4902.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2451, pruned_loss=0.05193, over 957460.31 frames. ], batch size: 37, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:09:13,779 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=120021.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:09:36,195 INFO [finetune.py:976] (1/7) Epoch 21, batch 5500, loss[loss=0.1799, simple_loss=0.2528, pruned_loss=0.05345, over 4894.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2423, pruned_loss=0.05127, over 957503.83 frames. ], batch size: 32, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:09:39,679 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.953e+01 1.422e+02 1.683e+02 2.099e+02 3.794e+02, threshold=3.366e+02, percent-clipped=2.0 +2023-03-27 02:09:49,989 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3026, 3.7489, 3.9292, 4.1666, 4.0434, 3.8516, 4.4045, 1.4403], + device='cuda:1'), covar=tensor([0.0847, 0.0942, 0.0863, 0.1050, 0.1283, 0.1577, 0.0767, 0.5975], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0244, 0.0282, 0.0292, 0.0335, 0.0285, 0.0304, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:10:09,942 INFO [finetune.py:976] (1/7) Epoch 21, batch 5550, loss[loss=0.1628, simple_loss=0.2297, pruned_loss=0.048, over 4829.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2427, pruned_loss=0.05101, over 957424.44 frames. ], batch size: 25, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:10:19,064 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8629, 1.5865, 2.0137, 1.4038, 1.9411, 2.0884, 1.5079, 2.1443], + device='cuda:1'), covar=tensor([0.1093, 0.1723, 0.1336, 0.1687, 0.0778, 0.1155, 0.2465, 0.0771], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0205, 0.0192, 0.0190, 0.0175, 0.0215, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:10:34,177 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2807, 1.2534, 1.1307, 1.3355, 1.5852, 1.3756, 1.2650, 1.1205], + device='cuda:1'), covar=tensor([0.0403, 0.0337, 0.0681, 0.0322, 0.0244, 0.0639, 0.0320, 0.0463], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0108, 0.0145, 0.0112, 0.0100, 0.0112, 0.0101, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.6695e-05, 8.2567e-05, 1.1414e-04, 8.6330e-05, 7.7973e-05, 8.3108e-05, + 7.5475e-05, 8.6595e-05], device='cuda:1') +2023-03-27 02:10:42,284 INFO [finetune.py:976] (1/7) Epoch 21, batch 5600, loss[loss=0.2095, simple_loss=0.2922, pruned_loss=0.0634, over 4837.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2478, pruned_loss=0.05231, over 955778.71 frames. ], batch size: 49, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:10:45,200 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.030e+02 1.550e+02 1.831e+02 2.203e+02 3.727e+02, threshold=3.662e+02, percent-clipped=1.0 +2023-03-27 02:10:52,179 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=120170.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:11:09,870 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4357, 2.0778, 2.8993, 1.7151, 2.4796, 2.7087, 1.9648, 2.8832], + device='cuda:1'), covar=tensor([0.1412, 0.2123, 0.1642, 0.2334, 0.1015, 0.1511, 0.2706, 0.0894], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0206, 0.0192, 0.0191, 0.0175, 0.0216, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:11:12,065 INFO [finetune.py:976] (1/7) Epoch 21, batch 5650, loss[loss=0.1562, simple_loss=0.2283, pruned_loss=0.0421, over 4838.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.249, pruned_loss=0.05221, over 954829.87 frames. ], batch size: 49, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:11:19,262 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1444, 1.2994, 1.3371, 0.6393, 1.2541, 1.5101, 1.5882, 1.2770], + device='cuda:1'), covar=tensor([0.0906, 0.0530, 0.0546, 0.0556, 0.0543, 0.0575, 0.0309, 0.0649], + device='cuda:1'), in_proj_covar=tensor([0.0124, 0.0151, 0.0128, 0.0125, 0.0133, 0.0130, 0.0143, 0.0150], + device='cuda:1'), out_proj_covar=tensor([9.0582e-05, 1.0928e-04, 9.1321e-05, 8.8039e-05, 9.3225e-05, 9.2656e-05, + 1.0250e-04, 1.0726e-04], device='cuda:1') +2023-03-27 02:11:20,664 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=120218.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:11:31,339 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6551, 2.4888, 2.2026, 2.6847, 2.5332, 2.3122, 2.8043, 2.6311], + device='cuda:1'), covar=tensor([0.1073, 0.1815, 0.2647, 0.2159, 0.2224, 0.1466, 0.2680, 0.1481], + device='cuda:1'), in_proj_covar=tensor([0.0185, 0.0187, 0.0233, 0.0251, 0.0245, 0.0202, 0.0213, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:11:41,964 INFO [finetune.py:976] (1/7) Epoch 21, batch 5700, loss[loss=0.1749, simple_loss=0.2227, pruned_loss=0.06356, over 4266.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2436, pruned_loss=0.05081, over 934752.05 frames. ], batch size: 18, lr: 3.17e-03, grad_scale: 32.0 +2023-03-27 02:11:42,658 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4521, 1.3745, 1.4875, 0.7056, 1.4576, 1.4999, 1.4808, 1.2892], + device='cuda:1'), covar=tensor([0.0578, 0.0740, 0.0656, 0.0919, 0.1041, 0.0706, 0.0633, 0.1182], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0137, 0.0139, 0.0121, 0.0126, 0.0139, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:11:44,946 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.012e+02 1.458e+02 1.705e+02 2.128e+02 3.595e+02, threshold=3.409e+02, percent-clipped=0.0 +2023-03-27 02:11:50,440 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9489, 1.7845, 1.6595, 2.0217, 2.0590, 2.0006, 1.4435, 1.6456], + device='cuda:1'), covar=tensor([0.1947, 0.1800, 0.1837, 0.1375, 0.1553, 0.1052, 0.2476, 0.1843], + device='cuda:1'), in_proj_covar=tensor([0.0243, 0.0209, 0.0212, 0.0194, 0.0243, 0.0188, 0.0217, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:12:12,191 INFO [finetune.py:976] (1/7) Epoch 22, batch 0, loss[loss=0.1685, simple_loss=0.2417, pruned_loss=0.04766, over 4912.00 frames. ], tot_loss[loss=0.1685, simple_loss=0.2417, pruned_loss=0.04766, over 4912.00 frames. ], batch size: 38, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:12:12,192 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 02:12:27,778 INFO [finetune.py:1010] (1/7) Epoch 22, validation: loss=0.1597, simple_loss=0.228, pruned_loss=0.04574, over 2265189.00 frames. +2023-03-27 02:12:27,778 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 02:12:35,249 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2090, 2.1595, 2.1483, 1.5099, 2.1223, 2.2721, 2.3556, 1.7525], + device='cuda:1'), covar=tensor([0.0604, 0.0667, 0.0764, 0.0872, 0.0757, 0.0762, 0.0609, 0.1252], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0137, 0.0139, 0.0121, 0.0125, 0.0139, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:13:15,373 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=120321.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:13:25,894 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.08 vs. limit=2.0 +2023-03-27 02:13:27,846 INFO [finetune.py:976] (1/7) Epoch 22, batch 50, loss[loss=0.1645, simple_loss=0.2393, pruned_loss=0.04482, over 4884.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2482, pruned_loss=0.05025, over 216884.57 frames. ], batch size: 35, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:13:45,823 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8033, 0.9950, 1.8408, 1.8052, 1.6041, 1.5557, 1.7174, 1.7203], + device='cuda:1'), covar=tensor([0.3436, 0.3571, 0.2749, 0.3038, 0.4193, 0.3246, 0.3675, 0.2757], + device='cuda:1'), in_proj_covar=tensor([0.0256, 0.0243, 0.0264, 0.0284, 0.0282, 0.0259, 0.0292, 0.0247], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:13:48,728 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.043e+02 1.631e+02 1.972e+02 2.363e+02 4.295e+02, threshold=3.943e+02, percent-clipped=3.0 +2023-03-27 02:13:55,424 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=120369.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:14:04,333 INFO [finetune.py:976] (1/7) Epoch 22, batch 100, loss[loss=0.1763, simple_loss=0.2426, pruned_loss=0.05505, over 4889.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2424, pruned_loss=0.04986, over 381101.55 frames. ], batch size: 35, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:14:14,886 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2867, 2.3104, 1.8285, 2.4017, 2.2237, 2.1267, 2.1555, 3.0713], + device='cuda:1'), covar=tensor([0.4206, 0.4406, 0.3773, 0.4573, 0.4471, 0.2829, 0.4339, 0.2046], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0263, 0.0233, 0.0277, 0.0254, 0.0224, 0.0254, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:14:21,103 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.52 vs. limit=5.0 +2023-03-27 02:14:37,018 INFO [finetune.py:976] (1/7) Epoch 22, batch 150, loss[loss=0.1737, simple_loss=0.236, pruned_loss=0.05569, over 4307.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2383, pruned_loss=0.04956, over 507428.31 frames. ], batch size: 65, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:14:55,336 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.054e+02 1.445e+02 1.737e+02 2.098e+02 4.550e+02, threshold=3.473e+02, percent-clipped=2.0 +2023-03-27 02:15:09,391 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-27 02:15:10,227 INFO [finetune.py:976] (1/7) Epoch 22, batch 200, loss[loss=0.152, simple_loss=0.2217, pruned_loss=0.04115, over 4914.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2371, pruned_loss=0.04937, over 605829.24 frames. ], batch size: 36, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:15:42,763 INFO [finetune.py:976] (1/7) Epoch 22, batch 250, loss[loss=0.1994, simple_loss=0.2611, pruned_loss=0.06881, over 4913.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2382, pruned_loss=0.04956, over 681887.89 frames. ], batch size: 35, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:15:57,934 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-27 02:16:01,906 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.094e+02 1.519e+02 1.843e+02 2.180e+02 3.548e+02, threshold=3.686e+02, percent-clipped=1.0 +2023-03-27 02:16:16,407 INFO [finetune.py:976] (1/7) Epoch 22, batch 300, loss[loss=0.1802, simple_loss=0.245, pruned_loss=0.0577, over 4910.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2438, pruned_loss=0.05117, over 743927.32 frames. ], batch size: 43, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:16:42,747 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0900, 1.8704, 1.4542, 0.5554, 1.5811, 1.7211, 1.5841, 1.8060], + device='cuda:1'), covar=tensor([0.0852, 0.0803, 0.1417, 0.1938, 0.1210, 0.2110, 0.2156, 0.0792], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0191, 0.0198, 0.0183, 0.0209, 0.0207, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:16:50,473 INFO [finetune.py:976] (1/7) Epoch 22, batch 350, loss[loss=0.2036, simple_loss=0.2873, pruned_loss=0.05996, over 4923.00 frames. ], tot_loss[loss=0.1768, simple_loss=0.2481, pruned_loss=0.0528, over 790822.01 frames. ], batch size: 38, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:17:09,291 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.647e+02 1.881e+02 2.280e+02 4.594e+02, threshold=3.762e+02, percent-clipped=3.0 +2023-03-27 02:17:19,808 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=120676.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:17:23,371 INFO [finetune.py:976] (1/7) Epoch 22, batch 400, loss[loss=0.1601, simple_loss=0.2223, pruned_loss=0.04894, over 4216.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.25, pruned_loss=0.05334, over 828376.79 frames. ], batch size: 18, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:17:41,250 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7642, 1.8328, 1.5568, 1.9566, 2.3201, 2.0850, 1.7805, 1.4799], + device='cuda:1'), covar=tensor([0.2282, 0.1853, 0.1878, 0.1538, 0.1789, 0.1230, 0.2283, 0.1892], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0209, 0.0213, 0.0195, 0.0243, 0.0189, 0.0217, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:17:57,079 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9832, 1.8965, 1.6678, 1.7337, 1.8067, 1.7584, 1.8507, 2.4801], + device='cuda:1'), covar=tensor([0.3518, 0.3775, 0.3128, 0.3708, 0.3641, 0.2276, 0.3392, 0.1620], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0264, 0.0233, 0.0278, 0.0255, 0.0224, 0.0254, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:18:12,317 INFO [finetune.py:976] (1/7) Epoch 22, batch 450, loss[loss=0.2141, simple_loss=0.2743, pruned_loss=0.07694, over 4756.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2487, pruned_loss=0.05312, over 856554.08 frames. ], batch size: 54, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:18:20,746 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=120737.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:18:23,801 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6186, 1.5214, 1.4545, 1.5520, 0.9810, 2.8739, 1.1329, 1.5100], + device='cuda:1'), covar=tensor([0.3288, 0.2446, 0.2126, 0.2396, 0.1828, 0.0248, 0.2525, 0.1224], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0124, 0.0114, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 02:18:30,454 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5056, 1.4034, 1.9237, 2.9417, 1.9726, 2.2015, 0.9243, 2.5067], + device='cuda:1'), covar=tensor([0.1750, 0.1450, 0.1247, 0.0577, 0.0832, 0.1357, 0.1795, 0.0459], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0117, 0.0135, 0.0165, 0.0102, 0.0138, 0.0126, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 02:18:43,318 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.117e+02 1.581e+02 1.866e+02 2.243e+02 3.725e+02, threshold=3.731e+02, percent-clipped=0.0 +2023-03-27 02:18:50,970 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0621, 4.9021, 4.6569, 2.9599, 4.9695, 3.9065, 1.2870, 3.5240], + device='cuda:1'), covar=tensor([0.2395, 0.1643, 0.1597, 0.3034, 0.0650, 0.0851, 0.4398, 0.1499], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0178, 0.0158, 0.0129, 0.0159, 0.0122, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 02:19:05,608 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7824, 2.5164, 2.2461, 1.1251, 2.4229, 2.0766, 1.9328, 2.2842], + device='cuda:1'), covar=tensor([0.0896, 0.0859, 0.1476, 0.2236, 0.1495, 0.2432, 0.2334, 0.1038], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0190, 0.0197, 0.0182, 0.0208, 0.0206, 0.0222, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:19:07,337 INFO [finetune.py:976] (1/7) Epoch 22, batch 500, loss[loss=0.1521, simple_loss=0.2227, pruned_loss=0.04078, over 4910.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2454, pruned_loss=0.05236, over 878611.04 frames. ], batch size: 37, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:19:13,677 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-03-27 02:19:40,280 INFO [finetune.py:976] (1/7) Epoch 22, batch 550, loss[loss=0.1598, simple_loss=0.2271, pruned_loss=0.04625, over 4756.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2424, pruned_loss=0.0513, over 894702.52 frames. ], batch size: 27, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:19:49,973 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1312, 3.5914, 3.7651, 3.9739, 3.9108, 3.6635, 4.2181, 1.3546], + device='cuda:1'), covar=tensor([0.0838, 0.0876, 0.0917, 0.1074, 0.1253, 0.1529, 0.0756, 0.5857], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0241, 0.0277, 0.0290, 0.0330, 0.0281, 0.0301, 0.0297], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:19:58,123 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.063e+02 1.544e+02 1.778e+02 2.172e+02 3.720e+02, threshold=3.555e+02, percent-clipped=0.0 +2023-03-27 02:20:13,118 INFO [finetune.py:976] (1/7) Epoch 22, batch 600, loss[loss=0.1777, simple_loss=0.2419, pruned_loss=0.05678, over 4837.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2439, pruned_loss=0.05225, over 908921.03 frames. ], batch size: 30, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:20:46,528 INFO [finetune.py:976] (1/7) Epoch 22, batch 650, loss[loss=0.1976, simple_loss=0.2712, pruned_loss=0.06204, over 4808.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2478, pruned_loss=0.05367, over 920404.98 frames. ], batch size: 39, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:21:04,770 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.113e+02 1.580e+02 1.877e+02 2.237e+02 3.344e+02, threshold=3.754e+02, percent-clipped=0.0 +2023-03-27 02:21:20,031 INFO [finetune.py:976] (1/7) Epoch 22, batch 700, loss[loss=0.1554, simple_loss=0.2239, pruned_loss=0.04349, over 4829.00 frames. ], tot_loss[loss=0.1791, simple_loss=0.2498, pruned_loss=0.05419, over 929136.76 frames. ], batch size: 30, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:21:32,892 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=121003.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:21:43,180 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3452, 1.1765, 1.5304, 2.3595, 1.6342, 2.2455, 0.9136, 2.1375], + device='cuda:1'), covar=tensor([0.2120, 0.2143, 0.1579, 0.1135, 0.1115, 0.1135, 0.1876, 0.0805], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0117, 0.0135, 0.0166, 0.0102, 0.0138, 0.0126, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 02:21:53,220 INFO [finetune.py:976] (1/7) Epoch 22, batch 750, loss[loss=0.1097, simple_loss=0.169, pruned_loss=0.02521, over 3994.00 frames. ], tot_loss[loss=0.1796, simple_loss=0.2505, pruned_loss=0.0544, over 934121.52 frames. ], batch size: 17, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:21:53,300 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121032.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:22:09,806 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.017e+02 1.617e+02 1.906e+02 2.410e+02 4.829e+02, threshold=3.812e+02, percent-clipped=5.0 +2023-03-27 02:22:14,816 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=121064.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:22:26,768 INFO [finetune.py:976] (1/7) Epoch 22, batch 800, loss[loss=0.1395, simple_loss=0.2242, pruned_loss=0.0274, over 4778.00 frames. ], tot_loss[loss=0.1786, simple_loss=0.2498, pruned_loss=0.05374, over 937049.04 frames. ], batch size: 29, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:22:28,702 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9628, 4.6774, 4.4777, 2.3298, 4.7738, 3.5415, 0.9266, 3.2719], + device='cuda:1'), covar=tensor([0.2185, 0.1576, 0.1258, 0.3095, 0.0663, 0.0866, 0.4561, 0.1305], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0177, 0.0158, 0.0130, 0.0159, 0.0122, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 02:22:45,042 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 02:22:47,258 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2620, 1.4995, 1.7642, 1.5770, 1.5348, 3.0158, 1.4948, 1.5414], + device='cuda:1'), covar=tensor([0.0997, 0.1745, 0.0959, 0.0937, 0.1530, 0.0301, 0.1368, 0.1754], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0081, 0.0073, 0.0076, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 02:23:10,263 INFO [finetune.py:976] (1/7) Epoch 22, batch 850, loss[loss=0.2045, simple_loss=0.2685, pruned_loss=0.07026, over 4250.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2473, pruned_loss=0.05226, over 940022.98 frames. ], batch size: 66, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:23:28,702 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.269e+01 1.509e+02 1.830e+02 2.168e+02 4.982e+02, threshold=3.659e+02, percent-clipped=1.0 +2023-03-27 02:23:37,199 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=121164.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:23:48,693 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7074, 1.5981, 1.5790, 1.6083, 1.2244, 3.2424, 1.3441, 1.7371], + device='cuda:1'), covar=tensor([0.3254, 0.2457, 0.2071, 0.2336, 0.1694, 0.0220, 0.2590, 0.1247], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0124, 0.0114, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 02:23:58,238 INFO [finetune.py:976] (1/7) Epoch 22, batch 900, loss[loss=0.1604, simple_loss=0.2285, pruned_loss=0.04613, over 4822.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2453, pruned_loss=0.0518, over 945219.88 frames. ], batch size: 30, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:24:09,545 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3243, 1.4150, 1.6206, 0.9128, 1.6037, 1.7072, 1.7890, 1.3743], + device='cuda:1'), covar=tensor([0.0955, 0.0836, 0.0503, 0.0536, 0.0484, 0.0873, 0.0391, 0.0789], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0150, 0.0125, 0.0123, 0.0130, 0.0129, 0.0141, 0.0147], + device='cuda:1'), out_proj_covar=tensor([8.9136e-05, 1.0803e-04, 8.9656e-05, 8.6466e-05, 9.1737e-05, 9.1740e-05, + 1.0085e-04, 1.0542e-04], device='cuda:1') +2023-03-27 02:24:24,197 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-27 02:24:38,323 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=121225.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:24:42,941 INFO [finetune.py:976] (1/7) Epoch 22, batch 950, loss[loss=0.1743, simple_loss=0.2407, pruned_loss=0.05394, over 4897.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2434, pruned_loss=0.05128, over 949140.27 frames. ], batch size: 46, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:24:59,302 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.960e+01 1.496e+02 1.804e+02 2.225e+02 4.174e+02, threshold=3.608e+02, percent-clipped=1.0 +2023-03-27 02:25:09,454 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2349, 2.1470, 1.7368, 2.0811, 2.2438, 1.8969, 2.4262, 2.2439], + device='cuda:1'), covar=tensor([0.1438, 0.1988, 0.3055, 0.2480, 0.2528, 0.1742, 0.2927, 0.1769], + device='cuda:1'), in_proj_covar=tensor([0.0186, 0.0187, 0.0234, 0.0253, 0.0246, 0.0203, 0.0213, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:25:16,227 INFO [finetune.py:976] (1/7) Epoch 22, batch 1000, loss[loss=0.2109, simple_loss=0.2697, pruned_loss=0.07602, over 4914.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2458, pruned_loss=0.05241, over 951141.94 frames. ], batch size: 35, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:25:33,746 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=121311.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 02:25:42,361 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4524, 2.3656, 1.8395, 2.6307, 2.4787, 2.0557, 2.8812, 2.4940], + device='cuda:1'), covar=tensor([0.1254, 0.2220, 0.2921, 0.2476, 0.2459, 0.1658, 0.3335, 0.1805], + device='cuda:1'), in_proj_covar=tensor([0.0186, 0.0188, 0.0235, 0.0254, 0.0247, 0.0204, 0.0214, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:25:49,276 INFO [finetune.py:976] (1/7) Epoch 22, batch 1050, loss[loss=0.1765, simple_loss=0.2505, pruned_loss=0.05121, over 4868.00 frames. ], tot_loss[loss=0.1779, simple_loss=0.2487, pruned_loss=0.05353, over 951525.30 frames. ], batch size: 31, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:25:49,380 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=121332.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:25:51,802 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0777, 0.9331, 0.9729, 0.3846, 0.9260, 1.1616, 1.1473, 1.0257], + device='cuda:1'), covar=tensor([0.0923, 0.0611, 0.0555, 0.0562, 0.0535, 0.0649, 0.0404, 0.0675], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0150, 0.0127, 0.0124, 0.0132, 0.0130, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.9814e-05, 1.0860e-04, 9.0406e-05, 8.7095e-05, 9.2787e-05, 9.2661e-05, + 1.0186e-04, 1.0648e-04], device='cuda:1') +2023-03-27 02:26:05,530 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.073e+02 1.613e+02 2.056e+02 2.633e+02 6.948e+02, threshold=4.113e+02, percent-clipped=5.0 +2023-03-27 02:26:05,612 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121359.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:26:09,271 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7767, 2.6358, 2.4133, 1.5097, 2.5221, 1.9890, 1.8880, 2.3323], + device='cuda:1'), covar=tensor([0.1327, 0.0791, 0.1987, 0.2055, 0.1805, 0.2502, 0.2362, 0.1353], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0192, 0.0199, 0.0183, 0.0210, 0.0209, 0.0223, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:26:12,287 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5931, 3.8260, 3.5504, 1.7959, 3.9621, 3.0036, 0.8059, 2.6640], + device='cuda:1'), covar=tensor([0.2500, 0.1835, 0.1537, 0.3323, 0.0919, 0.0930, 0.4465, 0.1444], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0177, 0.0158, 0.0129, 0.0159, 0.0122, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 02:26:13,506 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=121372.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 02:26:19,236 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=121380.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:26:20,880 INFO [finetune.py:976] (1/7) Epoch 22, batch 1100, loss[loss=0.1556, simple_loss=0.2289, pruned_loss=0.04117, over 4753.00 frames. ], tot_loss[loss=0.1797, simple_loss=0.2508, pruned_loss=0.05429, over 951908.50 frames. ], batch size: 27, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:26:31,164 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2161, 1.9432, 2.6618, 4.3313, 2.9464, 2.8081, 1.0452, 3.6906], + device='cuda:1'), covar=tensor([0.1623, 0.1479, 0.1349, 0.0502, 0.0724, 0.1530, 0.1887, 0.0344], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0118, 0.0136, 0.0167, 0.0102, 0.0139, 0.0127, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 02:26:36,793 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=121407.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:26:53,106 INFO [finetune.py:976] (1/7) Epoch 22, batch 1150, loss[loss=0.1571, simple_loss=0.2377, pruned_loss=0.03827, over 4731.00 frames. ], tot_loss[loss=0.1799, simple_loss=0.2514, pruned_loss=0.05427, over 952628.73 frames. ], batch size: 54, lr: 3.16e-03, grad_scale: 32.0 +2023-03-27 02:27:02,646 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0910, 1.9637, 1.4705, 0.6605, 1.5798, 1.6584, 1.5535, 1.8122], + device='cuda:1'), covar=tensor([0.0885, 0.0696, 0.1419, 0.1911, 0.1194, 0.2428, 0.2303, 0.0798], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0192, 0.0199, 0.0183, 0.0210, 0.0209, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:27:10,453 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.055e+01 1.683e+02 1.964e+02 2.424e+02 3.625e+02, threshold=3.928e+02, percent-clipped=0.0 +2023-03-27 02:27:15,981 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=121468.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:27:20,146 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3368, 2.9453, 2.8640, 1.2365, 3.0845, 2.2981, 0.6852, 1.9392], + device='cuda:1'), covar=tensor([0.2312, 0.1969, 0.1723, 0.3330, 0.1287, 0.1046, 0.3980, 0.1628], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0159, 0.0130, 0.0159, 0.0122, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 02:27:25,671 INFO [finetune.py:976] (1/7) Epoch 22, batch 1200, loss[loss=0.1732, simple_loss=0.2474, pruned_loss=0.04948, over 4926.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2497, pruned_loss=0.05344, over 953637.06 frames. ], batch size: 33, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:27:44,334 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3987, 2.3155, 1.9753, 2.4630, 2.2404, 2.2372, 2.2179, 3.1870], + device='cuda:1'), covar=tensor([0.3813, 0.4930, 0.3355, 0.4204, 0.4578, 0.2681, 0.4286, 0.1621], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0262, 0.0232, 0.0277, 0.0255, 0.0225, 0.0254, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:27:49,704 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121520.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:27:57,407 INFO [finetune.py:976] (1/7) Epoch 22, batch 1250, loss[loss=0.1266, simple_loss=0.1989, pruned_loss=0.02714, over 4907.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2472, pruned_loss=0.0527, over 953414.83 frames. ], batch size: 32, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:28:26,676 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.015e+02 1.487e+02 1.790e+02 2.203e+02 4.512e+02, threshold=3.581e+02, percent-clipped=1.0 +2023-03-27 02:28:40,600 INFO [finetune.py:976] (1/7) Epoch 22, batch 1300, loss[loss=0.1646, simple_loss=0.2317, pruned_loss=0.04871, over 4915.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2438, pruned_loss=0.05164, over 953837.69 frames. ], batch size: 43, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:29:08,212 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1419, 2.1310, 1.8274, 2.1004, 1.9671, 1.9873, 1.9638, 2.7355], + device='cuda:1'), covar=tensor([0.3865, 0.4224, 0.3302, 0.3959, 0.4271, 0.2541, 0.3896, 0.1612], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0263, 0.0233, 0.0278, 0.0256, 0.0225, 0.0254, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:29:39,001 INFO [finetune.py:976] (1/7) Epoch 22, batch 1350, loss[loss=0.1974, simple_loss=0.2601, pruned_loss=0.06734, over 4835.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2439, pruned_loss=0.05154, over 955219.20 frames. ], batch size: 47, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:30:02,099 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.135e+02 1.546e+02 1.913e+02 2.289e+02 6.231e+02, threshold=3.826e+02, percent-clipped=1.0 +2023-03-27 02:30:02,194 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=121659.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:30:07,016 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121667.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 02:30:11,398 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-27 02:30:16,102 INFO [finetune.py:976] (1/7) Epoch 22, batch 1400, loss[loss=0.1679, simple_loss=0.2566, pruned_loss=0.03957, over 4850.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2461, pruned_loss=0.05166, over 956335.29 frames. ], batch size: 44, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:30:16,843 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7824, 1.7952, 1.5633, 1.9704, 2.3013, 1.9834, 1.6291, 1.4347], + device='cuda:1'), covar=tensor([0.2222, 0.1926, 0.1973, 0.1593, 0.1731, 0.1186, 0.2300, 0.2003], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0211, 0.0215, 0.0196, 0.0243, 0.0189, 0.0218, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:30:34,336 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=121707.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:30:49,371 INFO [finetune.py:976] (1/7) Epoch 22, batch 1450, loss[loss=0.169, simple_loss=0.2446, pruned_loss=0.04674, over 4768.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2471, pruned_loss=0.05131, over 956370.44 frames. ], batch size: 54, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:31:08,590 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.158e+02 1.585e+02 1.838e+02 2.176e+02 4.319e+02, threshold=3.676e+02, percent-clipped=1.0 +2023-03-27 02:31:11,094 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121763.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:31:22,460 INFO [finetune.py:976] (1/7) Epoch 22, batch 1500, loss[loss=0.1748, simple_loss=0.2508, pruned_loss=0.04943, over 4891.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2483, pruned_loss=0.05223, over 954345.92 frames. ], batch size: 35, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:31:43,930 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-27 02:31:49,161 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=121820.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:31:56,350 INFO [finetune.py:976] (1/7) Epoch 22, batch 1550, loss[loss=0.1328, simple_loss=0.2023, pruned_loss=0.03167, over 4723.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2488, pruned_loss=0.05253, over 953159.50 frames. ], batch size: 23, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:32:15,624 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.105e+02 1.581e+02 1.856e+02 2.152e+02 3.350e+02, threshold=3.712e+02, percent-clipped=0.0 +2023-03-27 02:32:21,172 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=121868.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:32:29,536 INFO [finetune.py:976] (1/7) Epoch 22, batch 1600, loss[loss=0.1851, simple_loss=0.2527, pruned_loss=0.05873, over 4675.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.2461, pruned_loss=0.0518, over 952852.95 frames. ], batch size: 23, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:32:51,273 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0553, 1.9615, 1.5937, 1.7319, 2.0485, 1.7439, 2.2550, 2.0738], + device='cuda:1'), covar=tensor([0.1356, 0.1863, 0.2950, 0.2453, 0.2450, 0.1647, 0.2729, 0.1608], + device='cuda:1'), in_proj_covar=tensor([0.0187, 0.0188, 0.0235, 0.0254, 0.0248, 0.0204, 0.0215, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:33:02,672 INFO [finetune.py:976] (1/7) Epoch 22, batch 1650, loss[loss=0.1216, simple_loss=0.2057, pruned_loss=0.01869, over 4791.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2444, pruned_loss=0.05165, over 953846.62 frames. ], batch size: 51, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:33:13,730 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=121950.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:33:22,466 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.023e+02 1.553e+02 1.837e+02 2.107e+02 3.976e+02, threshold=3.675e+02, percent-clipped=1.0 +2023-03-27 02:33:33,821 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=121967.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 02:33:46,459 INFO [finetune.py:976] (1/7) Epoch 22, batch 1700, loss[loss=0.1846, simple_loss=0.261, pruned_loss=0.05413, over 4852.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2418, pruned_loss=0.05034, over 955499.07 frames. ], batch size: 49, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:34:13,900 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122011.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:34:16,838 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=122015.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 02:34:36,866 INFO [finetune.py:976] (1/7) Epoch 22, batch 1750, loss[loss=0.1826, simple_loss=0.2734, pruned_loss=0.04587, over 4814.00 frames. ], tot_loss[loss=0.1722, simple_loss=0.2433, pruned_loss=0.05057, over 955182.25 frames. ], batch size: 41, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:35:06,507 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.061e+02 1.629e+02 1.889e+02 2.189e+02 5.095e+02, threshold=3.778e+02, percent-clipped=2.0 +2023-03-27 02:35:10,492 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122063.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:35:22,845 INFO [finetune.py:976] (1/7) Epoch 22, batch 1800, loss[loss=0.1951, simple_loss=0.2635, pruned_loss=0.06337, over 4883.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2447, pruned_loss=0.05107, over 951646.43 frames. ], batch size: 32, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:35:34,080 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8903, 1.6329, 2.1354, 1.4721, 1.9362, 2.1457, 1.5227, 2.3090], + device='cuda:1'), covar=tensor([0.1264, 0.2120, 0.1373, 0.1917, 0.1022, 0.1291, 0.2796, 0.0772], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0206, 0.0192, 0.0190, 0.0175, 0.0213, 0.0217, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:35:41,156 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=122111.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:35:46,330 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122117.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:35:53,455 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.63 vs. limit=2.0 +2023-03-27 02:35:56,270 INFO [finetune.py:976] (1/7) Epoch 22, batch 1850, loss[loss=0.227, simple_loss=0.2844, pruned_loss=0.08481, over 4903.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2469, pruned_loss=0.05186, over 954288.75 frames. ], batch size: 37, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:36:12,755 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.044e+01 1.577e+02 1.909e+02 2.256e+02 5.766e+02, threshold=3.818e+02, percent-clipped=1.0 +2023-03-27 02:36:27,297 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122178.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 02:36:29,612 INFO [finetune.py:976] (1/7) Epoch 22, batch 1900, loss[loss=0.1684, simple_loss=0.2451, pruned_loss=0.04585, over 4735.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2478, pruned_loss=0.05166, over 954576.97 frames. ], batch size: 54, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:36:37,615 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122195.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:37:01,818 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5320, 1.4292, 2.0101, 2.9655, 2.0077, 2.1750, 0.9841, 2.5256], + device='cuda:1'), covar=tensor([0.1710, 0.1443, 0.1184, 0.0612, 0.0828, 0.1359, 0.1813, 0.0527], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0163, 0.0100, 0.0136, 0.0124, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 02:37:03,559 INFO [finetune.py:976] (1/7) Epoch 22, batch 1950, loss[loss=0.1796, simple_loss=0.2603, pruned_loss=0.04944, over 4920.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2478, pruned_loss=0.05153, over 955409.91 frames. ], batch size: 46, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:37:17,114 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-27 02:37:18,290 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122256.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:37:19,977 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.844e+01 1.567e+02 1.786e+02 2.225e+02 4.203e+02, threshold=3.573e+02, percent-clipped=2.0 +2023-03-27 02:37:36,886 INFO [finetune.py:976] (1/7) Epoch 22, batch 2000, loss[loss=0.1703, simple_loss=0.2438, pruned_loss=0.04846, over 4919.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2448, pruned_loss=0.05049, over 957540.57 frames. ], batch size: 37, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:37:37,714 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-27 02:37:49,744 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6146, 1.5259, 1.0804, 0.2715, 1.2344, 1.4653, 1.5172, 1.4799], + device='cuda:1'), covar=tensor([0.0955, 0.0886, 0.1492, 0.2222, 0.1440, 0.2545, 0.2447, 0.0973], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0193, 0.0200, 0.0183, 0.0210, 0.0209, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:37:51,519 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122306.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:37:59,438 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122319.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:38:04,444 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5707, 1.4406, 1.3213, 1.5941, 1.6323, 1.6138, 1.0354, 1.3421], + device='cuda:1'), covar=tensor([0.2345, 0.2250, 0.2191, 0.1785, 0.1662, 0.1368, 0.2658, 0.2081], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0210, 0.0214, 0.0196, 0.0243, 0.0190, 0.0218, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:38:10,018 INFO [finetune.py:976] (1/7) Epoch 22, batch 2050, loss[loss=0.1391, simple_loss=0.2121, pruned_loss=0.03305, over 4931.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2415, pruned_loss=0.04991, over 954291.99 frames. ], batch size: 33, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:38:26,818 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.990e+01 1.410e+02 1.787e+02 2.088e+02 3.673e+02, threshold=3.574e+02, percent-clipped=1.0 +2023-03-27 02:38:42,502 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122380.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:38:44,602 INFO [finetune.py:976] (1/7) Epoch 22, batch 2100, loss[loss=0.2004, simple_loss=0.2591, pruned_loss=0.07079, over 4915.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.241, pruned_loss=0.0497, over 953285.31 frames. ], batch size: 37, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:39:07,473 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6889, 1.3994, 1.3910, 1.4763, 1.8601, 1.7611, 1.4911, 1.3342], + device='cuda:1'), covar=tensor([0.0332, 0.0366, 0.0579, 0.0322, 0.0221, 0.0456, 0.0387, 0.0448], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0106, 0.0143, 0.0111, 0.0099, 0.0111, 0.0100, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.6321e-05, 8.1587e-05, 1.1204e-04, 8.4846e-05, 7.6793e-05, 8.2317e-05, + 7.4362e-05, 8.5572e-05], device='cuda:1') +2023-03-27 02:39:25,543 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3496, 1.4281, 1.5815, 1.1716, 1.3470, 1.4902, 1.4128, 1.6915], + device='cuda:1'), covar=tensor([0.1201, 0.2107, 0.1335, 0.1491, 0.0942, 0.1225, 0.2831, 0.0846], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0207, 0.0193, 0.0191, 0.0176, 0.0214, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:39:28,192 INFO [finetune.py:976] (1/7) Epoch 22, batch 2150, loss[loss=0.2318, simple_loss=0.2951, pruned_loss=0.08428, over 4805.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2437, pruned_loss=0.05085, over 953870.03 frames. ], batch size: 41, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:39:35,246 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7599, 1.2689, 0.8481, 1.5904, 2.1675, 1.1911, 1.5294, 1.6129], + device='cuda:1'), covar=tensor([0.1406, 0.2016, 0.1875, 0.1144, 0.1690, 0.1785, 0.1382, 0.1857], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0091, 0.0119, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 02:39:52,212 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7392, 3.7438, 3.5472, 1.7161, 3.8870, 2.8929, 0.8844, 2.5394], + device='cuda:1'), covar=tensor([0.2500, 0.2457, 0.1544, 0.3667, 0.1054, 0.1047, 0.4765, 0.1662], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0178, 0.0160, 0.0131, 0.0161, 0.0123, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 02:40:03,768 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.048e+02 1.591e+02 1.910e+02 2.352e+02 5.051e+02, threshold=3.819e+02, percent-clipped=3.0 +2023-03-27 02:40:13,671 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122468.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:40:16,628 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122473.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 02:40:26,433 INFO [finetune.py:976] (1/7) Epoch 22, batch 2200, loss[loss=0.1887, simple_loss=0.2609, pruned_loss=0.05825, over 4905.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2458, pruned_loss=0.05114, over 955633.33 frames. ], batch size: 43, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:40:26,515 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.0361, 4.4600, 4.6406, 4.8682, 4.7558, 4.5131, 5.1937, 1.5671], + device='cuda:1'), covar=tensor([0.0775, 0.0739, 0.0709, 0.0871, 0.1268, 0.1722, 0.0545, 0.5785], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0280, 0.0291, 0.0334, 0.0285, 0.0304, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:40:27,246 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-27 02:40:34,097 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3041, 2.1733, 1.8155, 2.1752, 2.2447, 1.9791, 2.4448, 2.2525], + device='cuda:1'), covar=tensor([0.1340, 0.1882, 0.3125, 0.2353, 0.2537, 0.1716, 0.2618, 0.1728], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0189, 0.0236, 0.0254, 0.0248, 0.0204, 0.0214, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:40:42,884 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1912, 3.6311, 3.8200, 4.0010, 3.9589, 3.7222, 4.2484, 1.4225], + device='cuda:1'), covar=tensor([0.0747, 0.0823, 0.0853, 0.0915, 0.1121, 0.1550, 0.0689, 0.5508], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0246, 0.0280, 0.0292, 0.0334, 0.0285, 0.0305, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:40:47,769 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0902, 1.9792, 1.7322, 1.8943, 1.8599, 1.8057, 1.8871, 2.6403], + device='cuda:1'), covar=tensor([0.3374, 0.4032, 0.2983, 0.3590, 0.3794, 0.2242, 0.3606, 0.1558], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0233, 0.0276, 0.0255, 0.0225, 0.0253, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:40:51,911 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.3521, 4.6312, 4.8654, 5.1774, 5.0885, 4.7199, 5.4533, 1.7434], + device='cuda:1'), covar=tensor([0.0716, 0.0779, 0.0789, 0.0775, 0.1125, 0.1781, 0.0521, 0.6006], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0245, 0.0280, 0.0291, 0.0334, 0.0285, 0.0305, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:40:56,767 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122529.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:40:58,943 INFO [finetune.py:976] (1/7) Epoch 22, batch 2250, loss[loss=0.2007, simple_loss=0.2737, pruned_loss=0.06386, over 4898.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.2468, pruned_loss=0.05141, over 955329.49 frames. ], batch size: 37, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:41:12,977 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122551.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:41:17,776 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.977e+01 1.468e+02 1.830e+02 2.095e+02 3.153e+02, threshold=3.659e+02, percent-clipped=0.0 +2023-03-27 02:41:20,319 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3499, 2.5055, 2.2830, 1.6451, 2.2288, 2.5130, 2.6860, 2.1691], + device='cuda:1'), covar=tensor([0.0650, 0.0574, 0.0770, 0.0917, 0.1124, 0.0721, 0.0577, 0.1006], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0141, 0.0121, 0.0126, 0.0139, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:41:31,700 INFO [finetune.py:976] (1/7) Epoch 22, batch 2300, loss[loss=0.1427, simple_loss=0.2207, pruned_loss=0.03238, over 4743.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2469, pruned_loss=0.05067, over 955856.08 frames. ], batch size: 27, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:41:40,538 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2755, 3.6873, 3.9074, 4.0759, 4.0487, 3.7595, 4.3526, 1.4071], + device='cuda:1'), covar=tensor([0.0754, 0.0807, 0.0909, 0.0890, 0.1136, 0.1648, 0.0654, 0.5911], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0245, 0.0279, 0.0291, 0.0334, 0.0284, 0.0304, 0.0298], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:41:49,493 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122606.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:42:05,213 INFO [finetune.py:976] (1/7) Epoch 22, batch 2350, loss[loss=0.1528, simple_loss=0.23, pruned_loss=0.03783, over 4815.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2456, pruned_loss=0.05017, over 956263.52 frames. ], batch size: 39, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:42:21,478 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=122654.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:42:24,441 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.047e+02 1.417e+02 1.625e+02 2.018e+02 3.172e+02, threshold=3.250e+02, percent-clipped=0.0 +2023-03-27 02:42:34,173 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122675.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:42:38,345 INFO [finetune.py:976] (1/7) Epoch 22, batch 2400, loss[loss=0.1707, simple_loss=0.2314, pruned_loss=0.055, over 4869.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2433, pruned_loss=0.04999, over 956104.88 frames. ], batch size: 34, lr: 3.15e-03, grad_scale: 64.0 +2023-03-27 02:43:11,471 INFO [finetune.py:976] (1/7) Epoch 22, batch 2450, loss[loss=0.1653, simple_loss=0.2372, pruned_loss=0.04671, over 4830.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2412, pruned_loss=0.04997, over 955042.80 frames. ], batch size: 33, lr: 3.14e-03, grad_scale: 64.0 +2023-03-27 02:43:31,104 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.081e+02 1.588e+02 1.841e+02 2.130e+02 2.968e+02, threshold=3.682e+02, percent-clipped=0.0 +2023-03-27 02:43:34,996 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.73 vs. limit=5.0 +2023-03-27 02:43:39,674 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122773.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 02:43:45,021 INFO [finetune.py:976] (1/7) Epoch 22, batch 2500, loss[loss=0.1752, simple_loss=0.2555, pruned_loss=0.04741, over 4840.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2431, pruned_loss=0.05129, over 956142.89 frames. ], batch size: 49, lr: 3.14e-03, grad_scale: 64.0 +2023-03-27 02:44:10,752 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-03-27 02:44:21,537 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=122821.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:44:23,359 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122824.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:44:28,112 INFO [finetune.py:976] (1/7) Epoch 22, batch 2550, loss[loss=0.1466, simple_loss=0.2313, pruned_loss=0.03094, over 4933.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2476, pruned_loss=0.05245, over 957909.64 frames. ], batch size: 38, lr: 3.14e-03, grad_scale: 64.0 +2023-03-27 02:44:42,554 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122851.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:44:53,792 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.164e+02 1.566e+02 1.889e+02 2.331e+02 3.878e+02, threshold=3.777e+02, percent-clipped=1.0 +2023-03-27 02:45:07,320 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122872.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:45:22,082 INFO [finetune.py:976] (1/7) Epoch 22, batch 2600, loss[loss=0.1241, simple_loss=0.2067, pruned_loss=0.02075, over 4767.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2483, pruned_loss=0.05227, over 957118.98 frames. ], batch size: 28, lr: 3.14e-03, grad_scale: 64.0 +2023-03-27 02:45:40,402 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=122899.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:46:02,682 INFO [finetune.py:976] (1/7) Epoch 22, batch 2650, loss[loss=0.1561, simple_loss=0.2246, pruned_loss=0.04382, over 4712.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.25, pruned_loss=0.05285, over 956975.86 frames. ], batch size: 23, lr: 3.14e-03, grad_scale: 64.0 +2023-03-27 02:46:03,415 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122933.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:46:19,588 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.898e+01 1.493e+02 1.785e+02 2.135e+02 3.458e+02, threshold=3.570e+02, percent-clipped=0.0 +2023-03-27 02:46:31,724 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122975.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:46:35,902 INFO [finetune.py:976] (1/7) Epoch 22, batch 2700, loss[loss=0.1614, simple_loss=0.2451, pruned_loss=0.03891, over 4899.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2486, pruned_loss=0.05166, over 957327.99 frames. ], batch size: 35, lr: 3.14e-03, grad_scale: 64.0 +2023-03-27 02:46:47,126 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9205, 1.7021, 2.0485, 1.4678, 1.8662, 2.1137, 1.5825, 2.2924], + device='cuda:1'), covar=tensor([0.1209, 0.2001, 0.1418, 0.1770, 0.0946, 0.1328, 0.2886, 0.0783], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0208, 0.0193, 0.0191, 0.0177, 0.0215, 0.0219, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:47:04,314 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=123023.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:47:09,737 INFO [finetune.py:976] (1/7) Epoch 22, batch 2750, loss[loss=0.1479, simple_loss=0.2175, pruned_loss=0.03914, over 4925.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2462, pruned_loss=0.05144, over 956679.58 frames. ], batch size: 38, lr: 3.14e-03, grad_scale: 64.0 +2023-03-27 02:47:24,428 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4314, 1.3978, 1.7474, 1.7437, 1.6367, 3.1630, 1.3984, 1.5872], + device='cuda:1'), covar=tensor([0.0951, 0.1709, 0.1073, 0.0887, 0.1497, 0.0274, 0.1414, 0.1646], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0092, 0.0081, 0.0086, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 02:47:26,634 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.465e+02 1.704e+02 2.045e+02 3.535e+02, threshold=3.409e+02, percent-clipped=0.0 +2023-03-27 02:47:42,979 INFO [finetune.py:976] (1/7) Epoch 22, batch 2800, loss[loss=0.1822, simple_loss=0.2408, pruned_loss=0.06175, over 4847.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2434, pruned_loss=0.0508, over 956604.36 frames. ], batch size: 47, lr: 3.14e-03, grad_scale: 64.0 +2023-03-27 02:47:49,727 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6226, 1.5199, 1.4988, 1.5253, 1.1554, 3.0503, 1.1757, 1.5693], + device='cuda:1'), covar=tensor([0.3408, 0.2614, 0.2269, 0.2537, 0.1885, 0.0255, 0.2645, 0.1356], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0123, 0.0114, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 02:47:54,539 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1668, 4.7508, 4.5037, 2.4388, 4.8328, 3.6813, 1.0164, 3.5011], + device='cuda:1'), covar=tensor([0.2371, 0.2036, 0.1296, 0.3258, 0.0789, 0.0875, 0.4600, 0.1369], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0176, 0.0158, 0.0129, 0.0160, 0.0122, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 02:48:10,887 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=123124.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:48:16,590 INFO [finetune.py:976] (1/7) Epoch 22, batch 2850, loss[loss=0.1623, simple_loss=0.2287, pruned_loss=0.04796, over 4790.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2409, pruned_loss=0.04961, over 958316.90 frames. ], batch size: 29, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:48:33,487 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.071e+02 1.606e+02 1.919e+02 2.375e+02 6.875e+02, threshold=3.839e+02, percent-clipped=7.0 +2023-03-27 02:48:42,229 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=123172.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:48:49,677 INFO [finetune.py:976] (1/7) Epoch 22, batch 2900, loss[loss=0.1755, simple_loss=0.2453, pruned_loss=0.05282, over 4762.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2443, pruned_loss=0.05068, over 957971.09 frames. ], batch size: 27, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:49:00,951 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7358, 1.4918, 2.1989, 3.4556, 2.4306, 2.3042, 1.1776, 2.9453], + device='cuda:1'), covar=tensor([0.1616, 0.1408, 0.1277, 0.0601, 0.0745, 0.1529, 0.1715, 0.0400], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0135, 0.0165, 0.0101, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 02:49:12,998 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1624, 2.1733, 1.6297, 2.2093, 2.0835, 1.8692, 2.4845, 2.2256], + device='cuda:1'), covar=tensor([0.1386, 0.2196, 0.3147, 0.2617, 0.2637, 0.1681, 0.3261, 0.1782], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0189, 0.0237, 0.0255, 0.0249, 0.0205, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:49:22,629 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=123228.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:49:25,027 INFO [finetune.py:976] (1/7) Epoch 22, batch 2950, loss[loss=0.1495, simple_loss=0.2312, pruned_loss=0.03388, over 4787.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2489, pruned_loss=0.05243, over 957199.44 frames. ], batch size: 29, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:49:42,387 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.515e+01 1.556e+02 1.822e+02 2.269e+02 3.192e+02, threshold=3.643e+02, percent-clipped=0.0 +2023-03-27 02:49:47,296 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7220, 1.6210, 1.3968, 1.7858, 2.0233, 1.7585, 1.3038, 1.3693], + device='cuda:1'), covar=tensor([0.1978, 0.1902, 0.1827, 0.1472, 0.1490, 0.1141, 0.2355, 0.1814], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0213, 0.0196, 0.0242, 0.0189, 0.0217, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:49:59,812 INFO [finetune.py:976] (1/7) Epoch 22, batch 3000, loss[loss=0.2029, simple_loss=0.269, pruned_loss=0.06837, over 4877.00 frames. ], tot_loss[loss=0.1771, simple_loss=0.2492, pruned_loss=0.05256, over 957024.80 frames. ], batch size: 43, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:49:59,812 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 02:50:03,400 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7059, 1.6061, 1.6083, 1.6411, 1.0712, 3.0145, 1.2270, 1.6500], + device='cuda:1'), covar=tensor([0.3220, 0.2301, 0.2032, 0.2255, 0.1755, 0.0256, 0.2526, 0.1176], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0122, 0.0124, 0.0114, 0.0096, 0.0095, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 02:50:15,178 INFO [finetune.py:1010] (1/7) Epoch 22, validation: loss=0.1575, simple_loss=0.2256, pruned_loss=0.04471, over 2265189.00 frames. +2023-03-27 02:50:15,179 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 02:50:47,932 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=123314.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:51:04,455 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-27 02:51:07,175 INFO [finetune.py:976] (1/7) Epoch 22, batch 3050, loss[loss=0.1625, simple_loss=0.2404, pruned_loss=0.04235, over 4788.00 frames. ], tot_loss[loss=0.1781, simple_loss=0.2503, pruned_loss=0.05297, over 956875.17 frames. ], batch size: 29, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:51:27,199 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.922e+01 1.647e+02 1.960e+02 2.377e+02 4.726e+02, threshold=3.920e+02, percent-clipped=6.0 +2023-03-27 02:51:36,584 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=123375.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:51:41,107 INFO [finetune.py:976] (1/7) Epoch 22, batch 3100, loss[loss=0.1984, simple_loss=0.2666, pruned_loss=0.06513, over 4931.00 frames. ], tot_loss[loss=0.1768, simple_loss=0.2485, pruned_loss=0.05257, over 956687.39 frames. ], batch size: 33, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:52:02,578 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4967, 2.6753, 2.2557, 1.6928, 2.3262, 2.6753, 2.6993, 2.2736], + device='cuda:1'), covar=tensor([0.0581, 0.0540, 0.0799, 0.0898, 0.1122, 0.0643, 0.0547, 0.0951], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0134, 0.0138, 0.0119, 0.0124, 0.0137, 0.0138, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:52:12,217 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=123428.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:52:14,576 INFO [finetune.py:976] (1/7) Epoch 22, batch 3150, loss[loss=0.156, simple_loss=0.23, pruned_loss=0.04095, over 4862.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2471, pruned_loss=0.05268, over 957207.19 frames. ], batch size: 34, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:52:17,217 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-03-27 02:52:34,405 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.134e+02 1.485e+02 1.808e+02 2.093e+02 3.688e+02, threshold=3.617e+02, percent-clipped=0.0 +2023-03-27 02:52:41,903 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9218, 1.3094, 1.9642, 1.9490, 1.7280, 1.6923, 1.8557, 1.8615], + device='cuda:1'), covar=tensor([0.3313, 0.3451, 0.2850, 0.3026, 0.3972, 0.3292, 0.3644, 0.2662], + device='cuda:1'), in_proj_covar=tensor([0.0258, 0.0243, 0.0265, 0.0285, 0.0284, 0.0259, 0.0294, 0.0247], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:52:47,863 INFO [finetune.py:976] (1/7) Epoch 22, batch 3200, loss[loss=0.1662, simple_loss=0.2294, pruned_loss=0.05152, over 4871.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2438, pruned_loss=0.05148, over 957595.53 frames. ], batch size: 34, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:52:52,367 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=123489.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:53:18,964 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=123528.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:53:21,321 INFO [finetune.py:976] (1/7) Epoch 22, batch 3250, loss[loss=0.1605, simple_loss=0.239, pruned_loss=0.041, over 4784.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2431, pruned_loss=0.05118, over 956072.64 frames. ], batch size: 26, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:53:24,504 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0254, 1.5451, 2.0778, 2.0066, 1.8071, 1.7526, 1.9696, 1.9398], + device='cuda:1'), covar=tensor([0.3684, 0.3796, 0.2956, 0.3478, 0.4437, 0.3645, 0.4259, 0.2843], + device='cuda:1'), in_proj_covar=tensor([0.0258, 0.0243, 0.0265, 0.0285, 0.0284, 0.0259, 0.0294, 0.0247], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:53:41,211 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.285e+01 1.537e+02 1.740e+02 2.121e+02 3.629e+02, threshold=3.481e+02, percent-clipped=1.0 +2023-03-27 02:53:51,149 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=123576.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:53:54,732 INFO [finetune.py:976] (1/7) Epoch 22, batch 3300, loss[loss=0.2, simple_loss=0.2806, pruned_loss=0.05971, over 4817.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2463, pruned_loss=0.05192, over 955384.73 frames. ], batch size: 33, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:54:06,909 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7328, 1.5536, 2.0124, 1.4244, 1.8659, 2.0299, 1.4762, 2.1866], + device='cuda:1'), covar=tensor([0.1416, 0.2074, 0.1489, 0.1844, 0.0912, 0.1400, 0.2911, 0.0834], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0207, 0.0192, 0.0190, 0.0175, 0.0214, 0.0217, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:54:28,300 INFO [finetune.py:976] (1/7) Epoch 22, batch 3350, loss[loss=0.1931, simple_loss=0.2658, pruned_loss=0.06018, over 4851.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2484, pruned_loss=0.05224, over 957362.64 frames. ], batch size: 44, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:54:35,080 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8950, 1.7841, 2.3676, 1.6058, 1.9947, 2.2412, 1.6761, 2.3406], + device='cuda:1'), covar=tensor([0.1133, 0.1797, 0.1296, 0.1767, 0.0792, 0.1151, 0.2521, 0.0710], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0207, 0.0191, 0.0189, 0.0175, 0.0213, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:54:47,671 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.106e+02 1.542e+02 1.809e+02 2.055e+02 5.285e+02, threshold=3.617e+02, percent-clipped=2.0 +2023-03-27 02:54:54,264 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=123670.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:55:01,512 INFO [finetune.py:976] (1/7) Epoch 22, batch 3400, loss[loss=0.2258, simple_loss=0.2819, pruned_loss=0.08482, over 4234.00 frames. ], tot_loss[loss=0.1773, simple_loss=0.2493, pruned_loss=0.05266, over 956698.78 frames. ], batch size: 65, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:55:17,041 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0096, 2.6266, 2.5257, 1.4936, 2.6216, 2.1761, 2.0578, 2.4498], + device='cuda:1'), covar=tensor([0.1200, 0.0910, 0.1640, 0.2035, 0.1704, 0.2070, 0.2200, 0.1177], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0191, 0.0198, 0.0182, 0.0210, 0.0206, 0.0223, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:55:54,856 INFO [finetune.py:976] (1/7) Epoch 22, batch 3450, loss[loss=0.1489, simple_loss=0.2196, pruned_loss=0.03911, over 4793.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2488, pruned_loss=0.05219, over 955983.59 frames. ], batch size: 29, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:56:07,067 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-03-27 02:56:26,819 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.145e+02 1.545e+02 1.898e+02 2.347e+02 3.548e+02, threshold=3.797e+02, percent-clipped=0.0 +2023-03-27 02:56:42,907 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0955, 3.5898, 3.7356, 3.9316, 3.8960, 3.6575, 4.1808, 1.2166], + device='cuda:1'), covar=tensor([0.0735, 0.0831, 0.0885, 0.0964, 0.1211, 0.1496, 0.0744, 0.5790], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0281, 0.0291, 0.0335, 0.0285, 0.0305, 0.0301], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:56:45,230 INFO [finetune.py:976] (1/7) Epoch 22, batch 3500, loss[loss=0.1666, simple_loss=0.2203, pruned_loss=0.05643, over 4904.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2469, pruned_loss=0.05201, over 957723.72 frames. ], batch size: 35, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:56:46,494 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=123784.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:56:50,996 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-27 02:57:01,649 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9043, 1.6390, 2.0892, 1.4164, 1.8964, 2.0992, 1.5728, 2.2524], + device='cuda:1'), covar=tensor([0.0973, 0.1754, 0.1194, 0.1555, 0.0712, 0.1059, 0.2553, 0.0559], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0206, 0.0191, 0.0189, 0.0174, 0.0213, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 02:57:18,475 INFO [finetune.py:976] (1/7) Epoch 22, batch 3550, loss[loss=0.139, simple_loss=0.207, pruned_loss=0.03551, over 4807.00 frames. ], tot_loss[loss=0.1733, simple_loss=0.2438, pruned_loss=0.05142, over 955955.17 frames. ], batch size: 25, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:57:27,295 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.02 vs. limit=5.0 +2023-03-27 02:57:36,085 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.884e+01 1.429e+02 1.766e+02 2.143e+02 3.754e+02, threshold=3.531e+02, percent-clipped=0.0 +2023-03-27 02:57:45,449 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=123872.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:57:52,354 INFO [finetune.py:976] (1/7) Epoch 22, batch 3600, loss[loss=0.1906, simple_loss=0.2581, pruned_loss=0.06149, over 4733.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2425, pruned_loss=0.05109, over 957296.59 frames. ], batch size: 59, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:58:00,477 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.08 vs. limit=5.0 +2023-03-27 02:58:25,719 INFO [finetune.py:976] (1/7) Epoch 22, batch 3650, loss[loss=0.2077, simple_loss=0.2851, pruned_loss=0.06511, over 4834.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2451, pruned_loss=0.05213, over 955401.18 frames. ], batch size: 40, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:58:26,482 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=123933.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:58:35,642 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.53 vs. limit=5.0 +2023-03-27 02:58:43,171 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.106e+01 1.598e+02 1.925e+02 2.525e+02 5.508e+02, threshold=3.851e+02, percent-clipped=5.0 +2023-03-27 02:58:49,848 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=123970.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:58:56,148 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7450, 1.7263, 1.5937, 1.6997, 1.4609, 4.1566, 1.5865, 1.9418], + device='cuda:1'), covar=tensor([0.3380, 0.2528, 0.2162, 0.2308, 0.1638, 0.0149, 0.2613, 0.1306], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0122, 0.0124, 0.0114, 0.0096, 0.0095, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 02:58:59,577 INFO [finetune.py:976] (1/7) Epoch 22, batch 3700, loss[loss=0.1213, simple_loss=0.2008, pruned_loss=0.02089, over 4794.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2471, pruned_loss=0.05228, over 955311.29 frames. ], batch size: 29, lr: 3.14e-03, grad_scale: 32.0 +2023-03-27 02:59:23,314 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=124018.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:59:34,052 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7923, 1.2964, 0.9786, 1.6882, 2.0300, 1.4236, 1.5982, 1.6869], + device='cuda:1'), covar=tensor([0.1361, 0.2066, 0.1850, 0.1084, 0.1882, 0.1909, 0.1390, 0.1851], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0111, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 02:59:34,574 INFO [finetune.py:976] (1/7) Epoch 22, batch 3750, loss[loss=0.1493, simple_loss=0.2274, pruned_loss=0.03559, over 4907.00 frames. ], tot_loss[loss=0.1768, simple_loss=0.2483, pruned_loss=0.05259, over 953493.52 frames. ], batch size: 37, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 02:59:37,123 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124036.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:59:51,700 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.170e+02 1.541e+02 1.786e+02 2.095e+02 2.976e+02, threshold=3.572e+02, percent-clipped=0.0 +2023-03-27 02:59:53,036 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124062.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 02:59:53,767 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.67 vs. limit=2.0 +2023-03-27 03:00:06,948 INFO [finetune.py:976] (1/7) Epoch 22, batch 3800, loss[loss=0.1731, simple_loss=0.2418, pruned_loss=0.05227, over 4693.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2488, pruned_loss=0.05257, over 951452.80 frames. ], batch size: 23, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:00:08,710 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124084.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:00:17,178 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124097.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:00:39,623 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124123.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:00:51,384 INFO [finetune.py:976] (1/7) Epoch 22, batch 3850, loss[loss=0.1374, simple_loss=0.216, pruned_loss=0.0294, over 4757.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.247, pruned_loss=0.0516, over 952178.80 frames. ], batch size: 27, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:00:51,454 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=124132.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:00:54,264 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4172, 1.6556, 0.7025, 2.2518, 2.5684, 1.7086, 2.0633, 2.1479], + device='cuda:1'), covar=tensor([0.1213, 0.1904, 0.2123, 0.1020, 0.1667, 0.1764, 0.1211, 0.1835], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0111, 0.0092, 0.0120, 0.0094, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:01:02,133 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.71 vs. limit=2.0 +2023-03-27 03:01:04,463 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4759, 1.0307, 0.7970, 1.3426, 1.9554, 0.6831, 1.2447, 1.4077], + device='cuda:1'), covar=tensor([0.1531, 0.2143, 0.1683, 0.1190, 0.1883, 0.1822, 0.1470, 0.1922], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0111, 0.0092, 0.0120, 0.0094, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:01:17,438 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-27 03:01:20,791 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.486e+02 1.746e+02 2.060e+02 3.550e+02, threshold=3.491e+02, percent-clipped=0.0 +2023-03-27 03:01:21,599 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.08 vs. limit=5.0 +2023-03-27 03:01:46,862 INFO [finetune.py:976] (1/7) Epoch 22, batch 3900, loss[loss=0.1904, simple_loss=0.2566, pruned_loss=0.06207, over 4781.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2451, pruned_loss=0.0511, over 951556.78 frames. ], batch size: 26, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:01:54,932 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8997, 1.6949, 1.5438, 1.3581, 1.6508, 1.6667, 1.6577, 2.2055], + device='cuda:1'), covar=tensor([0.3457, 0.3614, 0.2984, 0.3271, 0.3493, 0.2168, 0.3398, 0.1654], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0263, 0.0234, 0.0276, 0.0255, 0.0225, 0.0253, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:02:05,915 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4226, 1.3037, 1.7132, 2.4489, 1.6556, 2.2478, 0.8589, 2.0831], + device='cuda:1'), covar=tensor([0.1663, 0.1416, 0.1094, 0.0706, 0.0919, 0.1094, 0.1582, 0.0618], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0114, 0.0132, 0.0163, 0.0100, 0.0135, 0.0124, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 03:02:17,914 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124228.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:02:20,285 INFO [finetune.py:976] (1/7) Epoch 22, batch 3950, loss[loss=0.1649, simple_loss=0.239, pruned_loss=0.04542, over 4751.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2434, pruned_loss=0.05063, over 953714.06 frames. ], batch size: 27, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:02:39,959 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.148e+02 1.559e+02 1.897e+02 2.284e+02 3.853e+02, threshold=3.794e+02, percent-clipped=3.0 +2023-03-27 03:02:47,328 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124272.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:02:53,667 INFO [finetune.py:976] (1/7) Epoch 22, batch 4000, loss[loss=0.1835, simple_loss=0.2629, pruned_loss=0.05205, over 4907.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2409, pruned_loss=0.04979, over 953090.90 frames. ], batch size: 37, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:03:26,976 INFO [finetune.py:976] (1/7) Epoch 22, batch 4050, loss[loss=0.2265, simple_loss=0.3023, pruned_loss=0.07541, over 4911.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.244, pruned_loss=0.05094, over 954598.68 frames. ], batch size: 36, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:03:27,707 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124333.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:03:39,269 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124348.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 03:03:46,865 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.585e+02 1.909e+02 2.213e+02 3.315e+02, threshold=3.818e+02, percent-clipped=0.0 +2023-03-27 03:03:51,210 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2819, 1.3617, 1.6121, 1.4747, 1.4740, 2.9856, 1.3079, 1.4652], + device='cuda:1'), covar=tensor([0.1053, 0.1820, 0.1136, 0.1011, 0.1675, 0.0334, 0.1554, 0.1880], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0092, 0.0081, 0.0086, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 03:04:00,186 INFO [finetune.py:976] (1/7) Epoch 22, batch 4100, loss[loss=0.1915, simple_loss=0.2706, pruned_loss=0.05618, over 4802.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2462, pruned_loss=0.05184, over 952092.96 frames. ], batch size: 45, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:04:07,287 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124392.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:04:19,533 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124409.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 03:04:24,894 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124418.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:04:33,281 INFO [finetune.py:976] (1/7) Epoch 22, batch 4150, loss[loss=0.1484, simple_loss=0.2154, pruned_loss=0.04072, over 4815.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2466, pruned_loss=0.05197, over 950143.87 frames. ], batch size: 25, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:04:33,385 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124432.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:04:53,595 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.819e+01 1.555e+02 1.746e+02 2.178e+02 5.076e+02, threshold=3.492e+02, percent-clipped=2.0 +2023-03-27 03:05:06,939 INFO [finetune.py:976] (1/7) Epoch 22, batch 4200, loss[loss=0.1528, simple_loss=0.2225, pruned_loss=0.04162, over 4808.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2469, pruned_loss=0.05154, over 950884.70 frames. ], batch size: 41, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:05:09,017 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-27 03:05:14,212 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124493.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:05:40,280 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124528.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:05:42,630 INFO [finetune.py:976] (1/7) Epoch 22, batch 4250, loss[loss=0.1308, simple_loss=0.2043, pruned_loss=0.02863, over 4722.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.245, pruned_loss=0.05097, over 951636.51 frames. ], batch size: 23, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:06:02,063 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.200e+01 1.490e+02 1.704e+02 2.072e+02 3.423e+02, threshold=3.408e+02, percent-clipped=0.0 +2023-03-27 03:06:12,354 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=124576.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:06:17,861 INFO [finetune.py:976] (1/7) Epoch 22, batch 4300, loss[loss=0.1692, simple_loss=0.2439, pruned_loss=0.04728, over 4787.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2437, pruned_loss=0.0513, over 954133.01 frames. ], batch size: 29, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:06:32,212 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9999, 1.3585, 0.8786, 1.8825, 2.3638, 1.8003, 1.5976, 1.8045], + device='cuda:1'), covar=tensor([0.1222, 0.1921, 0.1833, 0.1062, 0.1713, 0.1899, 0.1390, 0.1790], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0111, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:07:10,853 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124628.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:07:17,350 INFO [finetune.py:976] (1/7) Epoch 22, batch 4350, loss[loss=0.1856, simple_loss=0.2595, pruned_loss=0.05579, over 4857.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2407, pruned_loss=0.05037, over 952940.32 frames. ], batch size: 49, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:07:21,114 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124638.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:07:48,793 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.554e+01 1.414e+02 1.739e+02 2.148e+02 3.782e+02, threshold=3.479e+02, percent-clipped=2.0 +2023-03-27 03:07:50,767 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124663.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:08:06,294 INFO [finetune.py:976] (1/7) Epoch 22, batch 4400, loss[loss=0.1606, simple_loss=0.2332, pruned_loss=0.04402, over 4779.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2404, pruned_loss=0.05004, over 952433.38 frames. ], batch size: 26, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:08:12,497 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124692.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:08:17,311 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124699.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:08:20,762 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124704.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 03:08:31,157 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124718.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:08:35,344 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124724.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:08:40,098 INFO [finetune.py:976] (1/7) Epoch 22, batch 4450, loss[loss=0.1831, simple_loss=0.2647, pruned_loss=0.05075, over 4747.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2447, pruned_loss=0.0515, over 953551.35 frames. ], batch size: 28, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:08:44,499 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7708, 1.3671, 1.9750, 1.8475, 1.6803, 1.5875, 1.8256, 1.7634], + device='cuda:1'), covar=tensor([0.4224, 0.3889, 0.2841, 0.3468, 0.4528, 0.3571, 0.4120, 0.2963], + device='cuda:1'), in_proj_covar=tensor([0.0258, 0.0243, 0.0263, 0.0285, 0.0283, 0.0260, 0.0293, 0.0246], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:08:45,008 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=124740.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:08:58,177 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.229e+02 1.653e+02 1.910e+02 2.206e+02 5.425e+02, threshold=3.820e+02, percent-clipped=3.0 +2023-03-27 03:09:02,352 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=124766.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:09:13,853 INFO [finetune.py:976] (1/7) Epoch 22, batch 4500, loss[loss=0.2324, simple_loss=0.3002, pruned_loss=0.0823, over 4756.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2467, pruned_loss=0.05254, over 952489.99 frames. ], batch size: 59, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:09:17,605 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124788.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:09:19,562 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.66 vs. limit=5.0 +2023-03-27 03:09:38,653 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124820.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:09:46,154 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.6878, 3.2034, 3.4233, 3.4778, 3.4869, 3.2561, 3.7159, 1.5252], + device='cuda:1'), covar=tensor([0.0824, 0.0832, 0.0809, 0.1008, 0.1254, 0.1375, 0.0976, 0.4857], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0280, 0.0291, 0.0335, 0.0284, 0.0305, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:09:47,291 INFO [finetune.py:976] (1/7) Epoch 22, batch 4550, loss[loss=0.1856, simple_loss=0.2603, pruned_loss=0.05544, over 4824.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2483, pruned_loss=0.053, over 951906.40 frames. ], batch size: 41, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:10:04,841 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.024e+02 1.469e+02 1.742e+02 1.998e+02 4.285e+02, threshold=3.484e+02, percent-clipped=1.0 +2023-03-27 03:10:19,587 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124881.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:10:20,090 INFO [finetune.py:976] (1/7) Epoch 22, batch 4600, loss[loss=0.1909, simple_loss=0.2602, pruned_loss=0.0608, over 4899.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2476, pruned_loss=0.05246, over 951045.98 frames. ], batch size: 35, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:10:48,843 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0198, 1.8752, 1.5893, 1.7454, 1.7701, 1.7617, 1.8050, 2.4844], + device='cuda:1'), covar=tensor([0.3425, 0.4046, 0.3112, 0.3453, 0.3661, 0.2342, 0.3440, 0.1648], + device='cuda:1'), in_proj_covar=tensor([0.0286, 0.0261, 0.0233, 0.0275, 0.0254, 0.0223, 0.0252, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:10:50,461 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124928.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:10:53,306 INFO [finetune.py:976] (1/7) Epoch 22, batch 4650, loss[loss=0.2022, simple_loss=0.2522, pruned_loss=0.07615, over 4242.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2448, pruned_loss=0.05169, over 951890.10 frames. ], batch size: 65, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:11:10,703 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.046e+02 1.467e+02 1.675e+02 2.008e+02 3.769e+02, threshold=3.350e+02, percent-clipped=1.0 +2023-03-27 03:11:21,851 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=124976.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:11:24,378 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1588, 2.0260, 1.8253, 2.0002, 1.8933, 1.9244, 1.9337, 2.6799], + device='cuda:1'), covar=tensor([0.3598, 0.4113, 0.3204, 0.3644, 0.4246, 0.2325, 0.3689, 0.1722], + device='cuda:1'), in_proj_covar=tensor([0.0287, 0.0262, 0.0234, 0.0275, 0.0254, 0.0224, 0.0252, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:11:24,418 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.03 vs. limit=5.0 +2023-03-27 03:11:26,384 INFO [finetune.py:976] (1/7) Epoch 22, batch 4700, loss[loss=0.1711, simple_loss=0.2289, pruned_loss=0.05666, over 4822.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2419, pruned_loss=0.05063, over 951978.66 frames. ], batch size: 38, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:11:36,918 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124994.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:11:43,180 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125004.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 03:11:52,768 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125019.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:12:07,772 INFO [finetune.py:976] (1/7) Epoch 22, batch 4750, loss[loss=0.2101, simple_loss=0.2782, pruned_loss=0.07096, over 4814.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2408, pruned_loss=0.05046, over 953596.78 frames. ], batch size: 39, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:12:30,811 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=125052.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 03:12:39,937 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.109e+02 1.467e+02 1.710e+02 2.084e+02 3.277e+02, threshold=3.420e+02, percent-clipped=0.0 +2023-03-27 03:13:08,203 INFO [finetune.py:976] (1/7) Epoch 22, batch 4800, loss[loss=0.1988, simple_loss=0.2794, pruned_loss=0.05911, over 4836.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2429, pruned_loss=0.05084, over 952866.75 frames. ], batch size: 49, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:13:16,552 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125088.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:13:42,362 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125128.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:13:45,182 INFO [finetune.py:976] (1/7) Epoch 22, batch 4850, loss[loss=0.1722, simple_loss=0.2544, pruned_loss=0.04507, over 4743.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2456, pruned_loss=0.05064, over 954867.08 frames. ], batch size: 54, lr: 3.13e-03, grad_scale: 64.0 +2023-03-27 03:13:47,687 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=125136.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:14:00,642 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5673, 1.4048, 1.4187, 1.4775, 0.9077, 2.9856, 1.0932, 1.5825], + device='cuda:1'), covar=tensor([0.3316, 0.2558, 0.2283, 0.2586, 0.1993, 0.0291, 0.2668, 0.1255], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0115, 0.0121, 0.0123, 0.0113, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 03:14:04,216 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.151e+02 1.665e+02 1.915e+02 2.224e+02 3.844e+02, threshold=3.831e+02, percent-clipped=1.0 +2023-03-27 03:14:10,493 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-03-27 03:14:12,860 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3719, 2.3605, 2.1983, 2.5667, 3.0245, 2.4188, 2.5199, 1.8418], + device='cuda:1'), covar=tensor([0.2320, 0.1993, 0.1835, 0.1650, 0.1661, 0.1134, 0.1933, 0.2002], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0210, 0.0212, 0.0196, 0.0243, 0.0188, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:14:14,550 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125176.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:14:18,597 INFO [finetune.py:976] (1/7) Epoch 22, batch 4900, loss[loss=0.1373, simple_loss=0.2199, pruned_loss=0.02742, over 4837.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2468, pruned_loss=0.05112, over 954041.62 frames. ], batch size: 49, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:14:23,439 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125189.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:14:52,064 INFO [finetune.py:976] (1/7) Epoch 22, batch 4950, loss[loss=0.2129, simple_loss=0.2767, pruned_loss=0.07454, over 4812.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2482, pruned_loss=0.05174, over 955570.68 frames. ], batch size: 30, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:15:12,017 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.098e+02 1.500e+02 1.887e+02 2.169e+02 5.445e+02, threshold=3.774e+02, percent-clipped=5.0 +2023-03-27 03:15:25,140 INFO [finetune.py:976] (1/7) Epoch 22, batch 5000, loss[loss=0.1416, simple_loss=0.2059, pruned_loss=0.03869, over 4895.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2465, pruned_loss=0.05126, over 952818.00 frames. ], batch size: 32, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:15:33,513 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125294.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:15:50,231 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125319.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:15:58,103 INFO [finetune.py:976] (1/7) Epoch 22, batch 5050, loss[loss=0.1148, simple_loss=0.1912, pruned_loss=0.01923, over 4693.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2444, pruned_loss=0.05074, over 953527.72 frames. ], batch size: 23, lr: 3.13e-03, grad_scale: 32.0 +2023-03-27 03:16:05,255 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=125342.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:16:15,322 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125356.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:16:18,273 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.708e+01 1.493e+02 1.877e+02 2.229e+02 3.838e+02, threshold=3.755e+02, percent-clipped=1.0 +2023-03-27 03:16:22,417 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=125367.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:16:27,962 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125376.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:16:31,543 INFO [finetune.py:976] (1/7) Epoch 22, batch 5100, loss[loss=0.1644, simple_loss=0.2477, pruned_loss=0.04054, over 4925.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2415, pruned_loss=0.04962, over 954006.41 frames. ], batch size: 46, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:16:55,556 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125417.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 03:16:58,434 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125421.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:17:05,015 INFO [finetune.py:976] (1/7) Epoch 22, batch 5150, loss[loss=0.1972, simple_loss=0.2829, pruned_loss=0.05575, over 4830.00 frames. ], tot_loss[loss=0.1722, simple_loss=0.2428, pruned_loss=0.05077, over 955216.87 frames. ], batch size: 39, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:17:13,030 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125437.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:17:23,259 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7314, 1.3391, 2.0345, 1.2584, 1.8635, 1.8790, 1.2268, 1.9075], + device='cuda:1'), covar=tensor([0.1438, 0.2420, 0.1293, 0.1980, 0.0935, 0.1707, 0.3130, 0.1243], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0190, 0.0190, 0.0173, 0.0214, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:17:26,818 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6833, 1.4243, 2.1261, 3.3842, 2.2386, 2.4882, 1.1213, 2.7779], + device='cuda:1'), covar=tensor([0.1740, 0.1493, 0.1343, 0.0537, 0.0818, 0.1427, 0.1731, 0.0501], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0164, 0.0101, 0.0136, 0.0124, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 03:17:30,473 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-27 03:17:33,157 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.242e+02 1.608e+02 1.768e+02 2.290e+02 4.207e+02, threshold=3.536e+02, percent-clipped=1.0 +2023-03-27 03:17:36,283 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5413, 3.4396, 3.2144, 1.7313, 3.5698, 2.6731, 0.8493, 2.4400], + device='cuda:1'), covar=tensor([0.2993, 0.2096, 0.1730, 0.3544, 0.1127, 0.1053, 0.4519, 0.1658], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0179, 0.0162, 0.0130, 0.0162, 0.0124, 0.0150, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 03:17:45,313 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125476.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:17:46,510 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5564, 1.0877, 0.8321, 1.4329, 1.9374, 1.1250, 1.3439, 1.4803], + device='cuda:1'), covar=tensor([0.1506, 0.2207, 0.2063, 0.1254, 0.2084, 0.2284, 0.1491, 0.1859], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0096, 0.0111, 0.0093, 0.0121, 0.0095, 0.0100, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:17:48,789 INFO [finetune.py:976] (1/7) Epoch 22, batch 5200, loss[loss=0.1604, simple_loss=0.2453, pruned_loss=0.03774, over 4763.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2465, pruned_loss=0.05169, over 956874.87 frames. ], batch size: 54, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:17:48,939 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125482.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:17:54,880 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125484.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:18:39,396 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=125524.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:18:44,719 INFO [finetune.py:976] (1/7) Epoch 22, batch 5250, loss[loss=0.1909, simple_loss=0.2614, pruned_loss=0.06019, over 4763.00 frames. ], tot_loss[loss=0.177, simple_loss=0.2491, pruned_loss=0.05245, over 956029.35 frames. ], batch size: 54, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:18:51,000 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-27 03:19:03,922 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.075e+02 1.572e+02 1.864e+02 2.118e+02 3.675e+02, threshold=3.728e+02, percent-clipped=1.0 +2023-03-27 03:19:18,601 INFO [finetune.py:976] (1/7) Epoch 22, batch 5300, loss[loss=0.2305, simple_loss=0.2843, pruned_loss=0.08841, over 4903.00 frames. ], tot_loss[loss=0.1794, simple_loss=0.2514, pruned_loss=0.05364, over 955673.68 frames. ], batch size: 37, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:19:22,409 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4738, 2.4369, 2.0242, 2.5593, 2.3050, 2.2687, 2.2437, 3.3094], + device='cuda:1'), covar=tensor([0.3873, 0.4826, 0.3545, 0.3896, 0.4277, 0.2628, 0.4416, 0.1604], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0261, 0.0233, 0.0276, 0.0254, 0.0224, 0.0252, 0.0234], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:19:32,530 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125603.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:19:52,427 INFO [finetune.py:976] (1/7) Epoch 22, batch 5350, loss[loss=0.1555, simple_loss=0.2295, pruned_loss=0.04074, over 4896.00 frames. ], tot_loss[loss=0.1781, simple_loss=0.2502, pruned_loss=0.05301, over 955356.21 frames. ], batch size: 36, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:19:54,402 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9218, 1.7970, 1.6974, 2.1020, 2.4875, 2.0161, 1.7361, 1.5668], + device='cuda:1'), covar=tensor([0.2357, 0.2134, 0.2028, 0.1634, 0.1504, 0.1250, 0.2297, 0.1985], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0211, 0.0212, 0.0197, 0.0243, 0.0189, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:20:10,863 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.759e+01 1.403e+02 1.743e+02 2.180e+02 4.274e+02, threshold=3.486e+02, percent-clipped=1.0 +2023-03-27 03:20:13,354 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125664.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:20:25,089 INFO [finetune.py:976] (1/7) Epoch 22, batch 5400, loss[loss=0.1978, simple_loss=0.2687, pruned_loss=0.06346, over 4721.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2475, pruned_loss=0.05218, over 955247.59 frames. ], batch size: 59, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:20:44,657 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125712.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 03:20:53,960 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125725.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:20:58,061 INFO [finetune.py:976] (1/7) Epoch 22, batch 5450, loss[loss=0.1522, simple_loss=0.2337, pruned_loss=0.03537, over 4897.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2446, pruned_loss=0.05133, over 956195.52 frames. ], batch size: 35, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:20:58,129 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125732.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:21:17,000 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.003e+02 1.470e+02 1.717e+02 2.000e+02 3.602e+02, threshold=3.434e+02, percent-clipped=1.0 +2023-03-27 03:21:27,711 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125777.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:21:31,103 INFO [finetune.py:976] (1/7) Epoch 22, batch 5500, loss[loss=0.1701, simple_loss=0.2432, pruned_loss=0.04853, over 4907.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2421, pruned_loss=0.05084, over 955195.55 frames. ], batch size: 46, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:21:32,419 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125784.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:21:33,655 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125786.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:22:04,442 INFO [finetune.py:976] (1/7) Epoch 22, batch 5550, loss[loss=0.1931, simple_loss=0.2794, pruned_loss=0.0534, over 4850.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2431, pruned_loss=0.05095, over 955318.02 frames. ], batch size: 44, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:22:04,489 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=125832.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:22:32,019 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.163e+02 1.619e+02 1.949e+02 2.379e+02 4.295e+02, threshold=3.899e+02, percent-clipped=6.0 +2023-03-27 03:22:48,802 INFO [finetune.py:976] (1/7) Epoch 22, batch 5600, loss[loss=0.1751, simple_loss=0.245, pruned_loss=0.05263, over 4811.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2478, pruned_loss=0.0522, over 953652.27 frames. ], batch size: 41, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:22:52,402 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0826, 1.0443, 0.9788, 0.5113, 0.9056, 1.1402, 1.1719, 0.9940], + device='cuda:1'), covar=tensor([0.0861, 0.0500, 0.0550, 0.0470, 0.0533, 0.0610, 0.0376, 0.0593], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0150, 0.0128, 0.0123, 0.0131, 0.0131, 0.0141, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.9931e-05, 1.0840e-04, 9.1649e-05, 8.6331e-05, 9.1925e-05, 9.3299e-05, + 1.0103e-04, 1.0704e-04], device='cuda:1') +2023-03-27 03:23:20,374 INFO [finetune.py:976] (1/7) Epoch 22, batch 5650, loss[loss=0.1925, simple_loss=0.2707, pruned_loss=0.05715, over 4842.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2502, pruned_loss=0.05246, over 954289.49 frames. ], batch size: 47, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:23:39,011 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5354, 1.3987, 0.9526, 0.2877, 1.2410, 1.4098, 1.4653, 1.3514], + device='cuda:1'), covar=tensor([0.0792, 0.0736, 0.1182, 0.1533, 0.1166, 0.1688, 0.1788, 0.0707], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0193, 0.0200, 0.0183, 0.0210, 0.0209, 0.0225, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:23:46,869 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.8994, 1.7614, 1.8822, 1.1455, 2.0847, 2.0497, 2.1787, 1.6709], + device='cuda:1'), covar=tensor([0.0824, 0.0651, 0.0491, 0.0542, 0.0398, 0.0681, 0.0311, 0.0701], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0150, 0.0128, 0.0122, 0.0130, 0.0130, 0.0141, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.9809e-05, 1.0814e-04, 9.1474e-05, 8.6205e-05, 9.1436e-05, 9.2787e-05, + 1.0087e-04, 1.0677e-04], device='cuda:1') +2023-03-27 03:23:47,406 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125959.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:23:48,544 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.919e+01 1.591e+02 1.865e+02 2.221e+02 3.612e+02, threshold=3.730e+02, percent-clipped=0.0 +2023-03-27 03:24:11,012 INFO [finetune.py:976] (1/7) Epoch 22, batch 5700, loss[loss=0.1525, simple_loss=0.2178, pruned_loss=0.04363, over 4308.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2457, pruned_loss=0.05161, over 936366.05 frames. ], batch size: 18, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:24:20,535 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125998.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:24:21,139 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1039, 2.0131, 2.4056, 1.6699, 2.0979, 2.4870, 1.9519, 2.5419], + device='cuda:1'), covar=tensor([0.1256, 0.1753, 0.1367, 0.1738, 0.0950, 0.1131, 0.2458, 0.0782], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0206, 0.0191, 0.0190, 0.0174, 0.0215, 0.0216, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:24:21,235 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.58 vs. limit=2.0 +2023-03-27 03:24:37,871 INFO [finetune.py:976] (1/7) Epoch 23, batch 0, loss[loss=0.2238, simple_loss=0.2863, pruned_loss=0.0806, over 4841.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2863, pruned_loss=0.0806, over 4841.00 frames. ], batch size: 49, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:24:37,871 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 03:24:43,644 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8415, 1.7643, 2.1822, 2.8220, 1.9558, 2.3049, 1.3649, 2.3571], + device='cuda:1'), covar=tensor([0.1213, 0.0948, 0.0811, 0.0532, 0.0793, 0.1244, 0.1267, 0.0456], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0163, 0.0100, 0.0135, 0.0123, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 03:24:44,652 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0145, 1.7894, 2.0351, 1.3506, 1.9355, 2.0192, 2.0768, 1.6371], + device='cuda:1'), covar=tensor([0.0514, 0.0664, 0.0589, 0.0849, 0.0788, 0.0623, 0.0516, 0.1088], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0134, 0.0138, 0.0119, 0.0123, 0.0137, 0.0137, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:24:53,039 INFO [finetune.py:1010] (1/7) Epoch 23, validation: loss=0.1587, simple_loss=0.2268, pruned_loss=0.04533, over 2265189.00 frames. +2023-03-27 03:24:53,039 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 03:24:59,386 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126012.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 03:25:12,072 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126032.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:25:29,460 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.47 vs. limit=5.0 +2023-03-27 03:25:30,342 INFO [finetune.py:976] (1/7) Epoch 23, batch 50, loss[loss=0.2057, simple_loss=0.2693, pruned_loss=0.071, over 4744.00 frames. ], tot_loss[loss=0.1785, simple_loss=0.2491, pruned_loss=0.05391, over 214380.02 frames. ], batch size: 59, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:25:30,459 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126059.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:25:31,918 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=126060.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:25:32,453 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.266e+01 1.545e+02 1.923e+02 2.325e+02 3.929e+02, threshold=3.846e+02, percent-clipped=1.0 +2023-03-27 03:25:38,629 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1536, 1.9680, 1.6946, 1.7833, 2.0775, 1.7882, 2.2705, 2.1180], + device='cuda:1'), covar=tensor([0.1338, 0.1891, 0.2882, 0.2376, 0.2480, 0.1683, 0.2657, 0.1666], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0189, 0.0235, 0.0254, 0.0248, 0.0205, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:25:42,850 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126077.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:25:44,662 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=126080.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:25:45,300 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=126081.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:26:03,745 INFO [finetune.py:976] (1/7) Epoch 23, batch 100, loss[loss=0.1323, simple_loss=0.2121, pruned_loss=0.02624, over 4782.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2415, pruned_loss=0.05046, over 380403.50 frames. ], batch size: 29, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:26:11,469 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7378, 1.1631, 0.7902, 1.6671, 2.0810, 1.6560, 1.5272, 1.6444], + device='cuda:1'), covar=tensor([0.1354, 0.1923, 0.1921, 0.1058, 0.1976, 0.1986, 0.1299, 0.1743], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0095, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:26:14,894 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=126125.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:26:19,894 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2033, 2.8433, 2.5936, 1.4593, 2.6472, 2.2507, 2.0961, 2.5225], + device='cuda:1'), covar=tensor([0.0881, 0.0880, 0.1730, 0.2072, 0.1852, 0.2270, 0.2146, 0.1080], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0194, 0.0201, 0.0184, 0.0211, 0.0210, 0.0226, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:26:37,114 INFO [finetune.py:976] (1/7) Epoch 23, batch 150, loss[loss=0.1806, simple_loss=0.2354, pruned_loss=0.0629, over 4817.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.2375, pruned_loss=0.04943, over 508766.36 frames. ], batch size: 33, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:26:38,293 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.556e+02 1.791e+02 2.255e+02 5.687e+02, threshold=3.583e+02, percent-clipped=3.0 +2023-03-27 03:27:10,672 INFO [finetune.py:976] (1/7) Epoch 23, batch 200, loss[loss=0.1579, simple_loss=0.2356, pruned_loss=0.04007, over 4908.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2373, pruned_loss=0.04971, over 607039.15 frames. ], batch size: 35, lr: 3.12e-03, grad_scale: 32.0 +2023-03-27 03:28:05,167 INFO [finetune.py:976] (1/7) Epoch 23, batch 250, loss[loss=0.2196, simple_loss=0.2957, pruned_loss=0.07175, over 4809.00 frames. ], tot_loss[loss=0.171, simple_loss=0.2409, pruned_loss=0.05056, over 684819.92 frames. ], batch size: 51, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:28:05,275 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126259.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:28:06,383 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.143e+02 1.625e+02 1.849e+02 2.180e+02 4.181e+02, threshold=3.697e+02, percent-clipped=1.0 +2023-03-27 03:28:37,017 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=126307.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:28:40,576 INFO [finetune.py:976] (1/7) Epoch 23, batch 300, loss[loss=0.1278, simple_loss=0.2088, pruned_loss=0.02334, over 4800.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2436, pruned_loss=0.05116, over 743067.11 frames. ], batch size: 25, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:29:00,169 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.16 vs. limit=5.0 +2023-03-27 03:29:20,739 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=126354.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:29:24,663 INFO [finetune.py:976] (1/7) Epoch 23, batch 350, loss[loss=0.1274, simple_loss=0.2107, pruned_loss=0.02205, over 4811.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.246, pruned_loss=0.05129, over 790619.81 frames. ], batch size: 25, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:29:25,830 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.521e+01 1.514e+02 1.805e+02 2.248e+02 3.946e+02, threshold=3.610e+02, percent-clipped=1.0 +2023-03-27 03:29:39,001 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=126380.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:29:39,583 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126381.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:29:45,065 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.47 vs. limit=5.0 +2023-03-27 03:29:56,088 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8564, 1.3842, 1.9148, 1.8566, 1.6477, 1.6103, 1.7961, 1.8008], + device='cuda:1'), covar=tensor([0.3679, 0.3989, 0.3221, 0.3573, 0.4726, 0.3824, 0.4360, 0.2989], + device='cuda:1'), in_proj_covar=tensor([0.0259, 0.0244, 0.0264, 0.0287, 0.0284, 0.0262, 0.0293, 0.0246], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:29:59,416 INFO [finetune.py:976] (1/7) Epoch 23, batch 400, loss[loss=0.1579, simple_loss=0.2286, pruned_loss=0.04362, over 4026.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2463, pruned_loss=0.0508, over 824061.48 frames. ], batch size: 17, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:30:21,836 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=126429.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:30:25,025 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-27 03:30:29,119 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3624, 1.3316, 1.2322, 1.4354, 1.6259, 1.5399, 1.3231, 1.2179], + device='cuda:1'), covar=tensor([0.0337, 0.0291, 0.0577, 0.0263, 0.0201, 0.0388, 0.0324, 0.0363], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0106, 0.0142, 0.0111, 0.0098, 0.0111, 0.0101, 0.0111], + device='cuda:1'), out_proj_covar=tensor([7.6345e-05, 8.1182e-05, 1.1140e-04, 8.5128e-05, 7.6349e-05, 8.1906e-05, + 7.5054e-05, 8.4533e-05], device='cuda:1') +2023-03-27 03:30:29,725 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126441.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:30:35,179 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=126450.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:30:40,551 INFO [finetune.py:976] (1/7) Epoch 23, batch 450, loss[loss=0.1838, simple_loss=0.243, pruned_loss=0.06228, over 4666.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2462, pruned_loss=0.051, over 854437.98 frames. ], batch size: 23, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:30:42,257 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.347e+01 1.414e+02 1.639e+02 2.017e+02 3.767e+02, threshold=3.277e+02, percent-clipped=3.0 +2023-03-27 03:30:42,540 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-27 03:30:45,265 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8161, 1.7180, 1.6160, 1.9557, 2.3510, 1.9606, 1.5719, 1.4969], + device='cuda:1'), covar=tensor([0.2358, 0.2031, 0.1957, 0.1674, 0.1625, 0.1188, 0.2462, 0.1983], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0211, 0.0213, 0.0197, 0.0244, 0.0189, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:31:13,829 INFO [finetune.py:976] (1/7) Epoch 23, batch 500, loss[loss=0.1757, simple_loss=0.2425, pruned_loss=0.05445, over 4935.00 frames. ], tot_loss[loss=0.1746, simple_loss=0.2456, pruned_loss=0.0518, over 877423.51 frames. ], batch size: 38, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:31:15,650 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126511.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:31:47,405 INFO [finetune.py:976] (1/7) Epoch 23, batch 550, loss[loss=0.1555, simple_loss=0.2103, pruned_loss=0.05032, over 4137.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2434, pruned_loss=0.05138, over 895766.42 frames. ], batch size: 18, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:31:48,617 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.316e+01 1.588e+02 1.845e+02 2.407e+02 4.330e+02, threshold=3.691e+02, percent-clipped=4.0 +2023-03-27 03:32:12,897 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-27 03:32:21,216 INFO [finetune.py:976] (1/7) Epoch 23, batch 600, loss[loss=0.1482, simple_loss=0.2185, pruned_loss=0.03888, over 4825.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2446, pruned_loss=0.05197, over 906101.12 frames. ], batch size: 33, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:32:31,432 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-03-27 03:32:44,166 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.56 vs. limit=5.0 +2023-03-27 03:33:02,510 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126654.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:33:03,815 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2117, 2.0951, 1.6419, 2.1940, 2.0814, 1.8307, 2.4594, 2.2103], + device='cuda:1'), covar=tensor([0.1302, 0.2213, 0.3084, 0.2609, 0.2642, 0.1635, 0.3226, 0.1691], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0190, 0.0237, 0.0256, 0.0251, 0.0207, 0.0217, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:33:05,530 INFO [finetune.py:976] (1/7) Epoch 23, batch 650, loss[loss=0.1761, simple_loss=0.2461, pruned_loss=0.05304, over 4858.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2478, pruned_loss=0.05255, over 918011.95 frames. ], batch size: 31, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:33:06,760 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.083e+02 1.579e+02 1.902e+02 2.270e+02 1.001e+03, threshold=3.804e+02, percent-clipped=1.0 +2023-03-27 03:33:06,990 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 03:33:13,322 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 03:33:42,517 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=126702.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:33:47,143 INFO [finetune.py:976] (1/7) Epoch 23, batch 700, loss[loss=0.1659, simple_loss=0.2351, pruned_loss=0.04828, over 4787.00 frames. ], tot_loss[loss=0.1781, simple_loss=0.2498, pruned_loss=0.05319, over 927437.74 frames. ], batch size: 26, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:34:11,542 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=126736.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:34:29,141 INFO [finetune.py:976] (1/7) Epoch 23, batch 750, loss[loss=0.1611, simple_loss=0.2421, pruned_loss=0.04006, over 4769.00 frames. ], tot_loss[loss=0.1773, simple_loss=0.2497, pruned_loss=0.05249, over 934724.04 frames. ], batch size: 28, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:34:30,819 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.063e+01 1.442e+02 1.710e+02 2.057e+02 3.398e+02, threshold=3.419e+02, percent-clipped=0.0 +2023-03-27 03:34:43,544 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-27 03:34:44,558 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0472, 1.5124, 0.7615, 1.9036, 2.3798, 1.6759, 1.6338, 1.8305], + device='cuda:1'), covar=tensor([0.1957, 0.2746, 0.2631, 0.1524, 0.2280, 0.2656, 0.2095, 0.2803], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:34:56,253 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6599, 1.5333, 2.1225, 3.3585, 2.1985, 2.4064, 1.2734, 2.7497], + device='cuda:1'), covar=tensor([0.1759, 0.1437, 0.1364, 0.0598, 0.0832, 0.1462, 0.1714, 0.0514], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0163, 0.0100, 0.0136, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 03:35:00,638 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=126806.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:35:02,394 INFO [finetune.py:976] (1/7) Epoch 23, batch 800, loss[loss=0.13, simple_loss=0.2163, pruned_loss=0.02189, over 4886.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2482, pruned_loss=0.05153, over 939761.70 frames. ], batch size: 32, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:35:07,275 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=126816.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:35:44,511 INFO [finetune.py:976] (1/7) Epoch 23, batch 850, loss[loss=0.1677, simple_loss=0.2402, pruned_loss=0.04757, over 4714.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2451, pruned_loss=0.0502, over 943733.31 frames. ], batch size: 23, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:35:45,682 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.329e+01 1.478e+02 1.768e+02 2.024e+02 3.574e+02, threshold=3.536e+02, percent-clipped=1.0 +2023-03-27 03:35:56,138 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126877.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:36:18,422 INFO [finetune.py:976] (1/7) Epoch 23, batch 900, loss[loss=0.1845, simple_loss=0.2483, pruned_loss=0.06029, over 4820.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2423, pruned_loss=0.04944, over 945810.16 frames. ], batch size: 38, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:36:52,033 INFO [finetune.py:976] (1/7) Epoch 23, batch 950, loss[loss=0.171, simple_loss=0.2476, pruned_loss=0.04723, over 4808.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2415, pruned_loss=0.04971, over 947162.66 frames. ], batch size: 38, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:36:53,226 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.057e+02 1.486e+02 1.775e+02 2.132e+02 3.351e+02, threshold=3.551e+02, percent-clipped=0.0 +2023-03-27 03:37:26,100 INFO [finetune.py:976] (1/7) Epoch 23, batch 1000, loss[loss=0.1774, simple_loss=0.26, pruned_loss=0.04735, over 4820.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2437, pruned_loss=0.05063, over 949323.60 frames. ], batch size: 38, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:37:43,501 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=127036.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:37:51,211 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6864, 1.5972, 1.4342, 1.7639, 2.0178, 1.7426, 1.3289, 1.3527], + device='cuda:1'), covar=tensor([0.2057, 0.1943, 0.1873, 0.1555, 0.1575, 0.1179, 0.2361, 0.1844], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0211, 0.0214, 0.0198, 0.0245, 0.0190, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:38:01,446 INFO [finetune.py:976] (1/7) Epoch 23, batch 1050, loss[loss=0.1884, simple_loss=0.2754, pruned_loss=0.05075, over 4837.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2477, pruned_loss=0.05172, over 950786.27 frames. ], batch size: 47, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:38:02,652 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.005e+02 1.597e+02 1.830e+02 2.248e+02 5.450e+02, threshold=3.660e+02, percent-clipped=4.0 +2023-03-27 03:38:27,251 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=127084.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:38:42,636 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=127106.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:38:44,898 INFO [finetune.py:976] (1/7) Epoch 23, batch 1100, loss[loss=0.1193, simple_loss=0.1912, pruned_loss=0.02371, over 4682.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2474, pruned_loss=0.05116, over 952410.66 frames. ], batch size: 23, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:39:08,664 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.62 vs. limit=5.0 +2023-03-27 03:39:14,543 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=127154.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:39:21,916 INFO [finetune.py:976] (1/7) Epoch 23, batch 1150, loss[loss=0.1525, simple_loss=0.2223, pruned_loss=0.04134, over 4824.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.249, pruned_loss=0.05189, over 953280.53 frames. ], batch size: 30, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:39:23,618 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.031e+02 1.547e+02 1.781e+02 2.032e+02 3.739e+02, threshold=3.562e+02, percent-clipped=1.0 +2023-03-27 03:39:33,514 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.15 vs. limit=5.0 +2023-03-27 03:39:35,834 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127172.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:40:04,150 INFO [finetune.py:976] (1/7) Epoch 23, batch 1200, loss[loss=0.1744, simple_loss=0.247, pruned_loss=0.05088, over 4798.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2473, pruned_loss=0.05158, over 953033.67 frames. ], batch size: 51, lr: 3.11e-03, grad_scale: 64.0 +2023-03-27 03:40:10,142 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6496, 1.6580, 1.7085, 0.9716, 1.8732, 2.1515, 2.0330, 1.4531], + device='cuda:1'), covar=tensor([0.0940, 0.0693, 0.0598, 0.0618, 0.0449, 0.0557, 0.0337, 0.0815], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0148, 0.0127, 0.0122, 0.0130, 0.0129, 0.0141, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.9516e-05, 1.0709e-04, 9.0898e-05, 8.5929e-05, 9.1560e-05, 9.1630e-05, + 1.0076e-04, 1.0597e-04], device='cuda:1') +2023-03-27 03:40:33,004 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127249.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:40:33,625 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4168, 2.3508, 2.2495, 1.7660, 2.3017, 2.6186, 2.5742, 2.0369], + device='cuda:1'), covar=tensor([0.0657, 0.0714, 0.0871, 0.0896, 0.0931, 0.0788, 0.0654, 0.1269], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0135, 0.0138, 0.0120, 0.0125, 0.0137, 0.0138, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:40:36,127 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-27 03:40:44,043 INFO [finetune.py:976] (1/7) Epoch 23, batch 1250, loss[loss=0.1565, simple_loss=0.2204, pruned_loss=0.04631, over 4722.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.244, pruned_loss=0.05048, over 953016.37 frames. ], batch size: 23, lr: 3.11e-03, grad_scale: 64.0 +2023-03-27 03:40:45,214 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.489e+02 1.742e+02 2.248e+02 3.707e+02, threshold=3.484e+02, percent-clipped=1.0 +2023-03-27 03:41:21,264 INFO [finetune.py:976] (1/7) Epoch 23, batch 1300, loss[loss=0.1414, simple_loss=0.2169, pruned_loss=0.03289, over 4825.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2411, pruned_loss=0.04958, over 955201.98 frames. ], batch size: 40, lr: 3.11e-03, grad_scale: 64.0 +2023-03-27 03:41:22,022 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127310.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:41:54,394 INFO [finetune.py:976] (1/7) Epoch 23, batch 1350, loss[loss=0.2163, simple_loss=0.2771, pruned_loss=0.07777, over 4910.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2413, pruned_loss=0.04969, over 954537.07 frames. ], batch size: 36, lr: 3.11e-03, grad_scale: 64.0 +2023-03-27 03:41:55,607 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.500e+02 1.797e+02 2.250e+02 4.549e+02, threshold=3.594e+02, percent-clipped=1.0 +2023-03-27 03:42:27,772 INFO [finetune.py:976] (1/7) Epoch 23, batch 1400, loss[loss=0.1672, simple_loss=0.2492, pruned_loss=0.04259, over 4904.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2441, pruned_loss=0.05023, over 954301.58 frames. ], batch size: 43, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:42:40,737 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127427.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 03:42:50,489 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8134, 1.8234, 2.1701, 2.0699, 1.9521, 3.5397, 1.6488, 1.8965], + device='cuda:1'), covar=tensor([0.0855, 0.1452, 0.0910, 0.0800, 0.1261, 0.0311, 0.1296, 0.1492], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0081, 0.0074, 0.0076, 0.0092, 0.0081, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 03:42:58,106 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7039, 1.6994, 1.4571, 1.9173, 2.2483, 1.8648, 1.5709, 1.3891], + device='cuda:1'), covar=tensor([0.2152, 0.1889, 0.1939, 0.1495, 0.1652, 0.1236, 0.2329, 0.1944], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0213, 0.0196, 0.0244, 0.0189, 0.0217, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:43:01,001 INFO [finetune.py:976] (1/7) Epoch 23, batch 1450, loss[loss=0.1589, simple_loss=0.2486, pruned_loss=0.03458, over 4875.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2469, pruned_loss=0.05167, over 952954.18 frames. ], batch size: 34, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:43:03,312 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.817e+01 1.596e+02 1.913e+02 2.194e+02 3.811e+02, threshold=3.827e+02, percent-clipped=1.0 +2023-03-27 03:43:09,876 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=127472.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:43:10,005 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.70 vs. limit=2.0 +2023-03-27 03:43:25,059 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127488.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 03:43:46,709 INFO [finetune.py:976] (1/7) Epoch 23, batch 1500, loss[loss=0.14, simple_loss=0.2128, pruned_loss=0.03365, over 4692.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2478, pruned_loss=0.05172, over 955234.40 frames. ], batch size: 23, lr: 3.11e-03, grad_scale: 32.0 +2023-03-27 03:43:52,736 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8544, 1.7553, 2.2462, 1.4607, 2.0812, 2.1565, 1.5724, 2.2248], + device='cuda:1'), covar=tensor([0.1222, 0.2053, 0.1272, 0.1966, 0.0833, 0.1327, 0.2698, 0.0918], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0205, 0.0191, 0.0189, 0.0173, 0.0213, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:43:53,898 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=127520.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:43:56,260 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4834, 1.0879, 0.7906, 1.3168, 1.9086, 0.6999, 1.2239, 1.3183], + device='cuda:1'), covar=tensor([0.1585, 0.2186, 0.1759, 0.1221, 0.2001, 0.2039, 0.1582, 0.2070], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0092, 0.0119, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:43:59,947 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127529.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:44:20,425 INFO [finetune.py:976] (1/7) Epoch 23, batch 1550, loss[loss=0.1619, simple_loss=0.2439, pruned_loss=0.03996, over 4892.00 frames. ], tot_loss[loss=0.1775, simple_loss=0.2497, pruned_loss=0.05263, over 954716.23 frames. ], batch size: 43, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:44:22,241 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.110e+02 1.462e+02 1.755e+02 2.123e+02 3.197e+02, threshold=3.511e+02, percent-clipped=0.0 +2023-03-27 03:44:48,017 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127590.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:45:04,793 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127605.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:45:07,683 INFO [finetune.py:976] (1/7) Epoch 23, batch 1600, loss[loss=0.1859, simple_loss=0.2441, pruned_loss=0.06381, over 4815.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2467, pruned_loss=0.05187, over 955542.45 frames. ], batch size: 41, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:45:34,516 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0 +2023-03-27 03:45:40,959 INFO [finetune.py:976] (1/7) Epoch 23, batch 1650, loss[loss=0.158, simple_loss=0.2295, pruned_loss=0.04321, over 4936.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2433, pruned_loss=0.0509, over 955978.01 frames. ], batch size: 38, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:45:43,334 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.066e+02 1.567e+02 1.812e+02 2.212e+02 4.212e+02, threshold=3.624e+02, percent-clipped=4.0 +2023-03-27 03:45:43,475 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127662.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:46:26,410 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7056, 3.8911, 3.6433, 1.6558, 3.8880, 3.0415, 0.9166, 2.7633], + device='cuda:1'), covar=tensor([0.2446, 0.2434, 0.1418, 0.3405, 0.1180, 0.0973, 0.4279, 0.1481], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0178, 0.0159, 0.0128, 0.0161, 0.0124, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 03:46:26,934 INFO [finetune.py:976] (1/7) Epoch 23, batch 1700, loss[loss=0.2118, simple_loss=0.2687, pruned_loss=0.07743, over 4821.00 frames. ], tot_loss[loss=0.171, simple_loss=0.2416, pruned_loss=0.05023, over 956299.50 frames. ], batch size: 38, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:46:36,931 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127723.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:47:00,092 INFO [finetune.py:976] (1/7) Epoch 23, batch 1750, loss[loss=0.1977, simple_loss=0.2651, pruned_loss=0.06514, over 4926.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2431, pruned_loss=0.05087, over 955704.36 frames. ], batch size: 38, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:47:01,901 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.127e+02 1.511e+02 1.783e+02 2.157e+02 3.427e+02, threshold=3.565e+02, percent-clipped=0.0 +2023-03-27 03:47:06,851 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4050, 2.3708, 2.0679, 2.5995, 2.9060, 2.4757, 2.2569, 1.8850], + device='cuda:1'), covar=tensor([0.2204, 0.1788, 0.1793, 0.1526, 0.1684, 0.1020, 0.2051, 0.1848], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0210, 0.0213, 0.0196, 0.0244, 0.0189, 0.0217, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:47:12,608 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0496, 2.6880, 2.8380, 2.9443, 2.8740, 2.7426, 3.0885, 1.1596], + device='cuda:1'), covar=tensor([0.1036, 0.0965, 0.1027, 0.1033, 0.1461, 0.1662, 0.1300, 0.5257], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0247, 0.0281, 0.0294, 0.0337, 0.0288, 0.0306, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:47:13,900 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127779.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:47:16,793 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127783.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 03:47:22,127 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9063, 1.5669, 1.4226, 1.4565, 2.0547, 2.1131, 1.7317, 1.6015], + device='cuda:1'), covar=tensor([0.0384, 0.0436, 0.0703, 0.0442, 0.0245, 0.0473, 0.0363, 0.0453], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0105, 0.0143, 0.0111, 0.0098, 0.0111, 0.0100, 0.0111], + device='cuda:1'), out_proj_covar=tensor([7.6405e-05, 8.0848e-05, 1.1179e-04, 8.4827e-05, 7.6365e-05, 8.1909e-05, + 7.4560e-05, 8.4357e-05], device='cuda:1') +2023-03-27 03:47:33,985 INFO [finetune.py:976] (1/7) Epoch 23, batch 1800, loss[loss=0.1375, simple_loss=0.2123, pruned_loss=0.03136, over 4749.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2458, pruned_loss=0.05175, over 956018.28 frames. ], batch size: 27, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:47:54,752 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127840.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:48:01,836 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127850.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:48:07,744 INFO [finetune.py:976] (1/7) Epoch 23, batch 1850, loss[loss=0.1855, simple_loss=0.2702, pruned_loss=0.05045, over 4809.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2475, pruned_loss=0.05209, over 955709.11 frames. ], batch size: 45, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:48:09,575 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.256e+01 1.480e+02 1.674e+02 2.095e+02 4.093e+02, threshold=3.347e+02, percent-clipped=1.0 +2023-03-27 03:48:12,660 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8426, 3.3861, 3.5608, 3.7239, 3.5808, 3.3481, 3.9099, 1.2151], + device='cuda:1'), covar=tensor([0.0977, 0.0945, 0.0886, 0.1129, 0.1552, 0.1812, 0.0895, 0.5923], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0248, 0.0283, 0.0296, 0.0339, 0.0290, 0.0307, 0.0305], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:48:16,136 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127872.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:48:19,742 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-27 03:48:24,925 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127885.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:48:40,232 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=127905.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:48:42,624 INFO [finetune.py:976] (1/7) Epoch 23, batch 1900, loss[loss=0.1573, simple_loss=0.2337, pruned_loss=0.0404, over 4841.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2482, pruned_loss=0.05236, over 955290.53 frames. ], batch size: 49, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:48:43,940 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127911.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:48:58,804 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127933.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:49:11,954 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=127953.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:49:16,014 INFO [finetune.py:976] (1/7) Epoch 23, batch 1950, loss[loss=0.1534, simple_loss=0.2046, pruned_loss=0.05113, over 4011.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.247, pruned_loss=0.05189, over 953154.90 frames. ], batch size: 17, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:49:17,847 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.653e+01 1.391e+02 1.764e+02 2.084e+02 3.552e+02, threshold=3.528e+02, percent-clipped=1.0 +2023-03-27 03:49:51,365 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7895, 3.7488, 3.7678, 2.0758, 4.0029, 2.9824, 1.2885, 2.8799], + device='cuda:1'), covar=tensor([0.2727, 0.2042, 0.1251, 0.2751, 0.0911, 0.0916, 0.3588, 0.1297], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0159, 0.0128, 0.0161, 0.0124, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 03:49:53,065 INFO [finetune.py:976] (1/7) Epoch 23, batch 2000, loss[loss=0.1438, simple_loss=0.2086, pruned_loss=0.03953, over 4709.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2453, pruned_loss=0.05166, over 954355.02 frames. ], batch size: 23, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:50:03,202 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128018.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:50:19,606 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8123, 1.2419, 0.8172, 1.6631, 2.0540, 1.4469, 1.5336, 1.5616], + device='cuda:1'), covar=tensor([0.1423, 0.2115, 0.1972, 0.1146, 0.2024, 0.1846, 0.1473, 0.2065], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0092, 0.0119, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:50:38,970 INFO [finetune.py:976] (1/7) Epoch 23, batch 2050, loss[loss=0.1686, simple_loss=0.2409, pruned_loss=0.04817, over 4898.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2425, pruned_loss=0.05062, over 954656.28 frames. ], batch size: 32, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:50:41,269 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.050e+02 1.483e+02 1.749e+02 2.101e+02 3.191e+02, threshold=3.498e+02, percent-clipped=0.0 +2023-03-27 03:50:54,688 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128083.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 03:51:13,972 INFO [finetune.py:976] (1/7) Epoch 23, batch 2100, loss[loss=0.2115, simple_loss=0.2758, pruned_loss=0.07355, over 4871.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2426, pruned_loss=0.05101, over 955054.11 frames. ], batch size: 34, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:51:40,711 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=128131.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 03:51:43,605 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128135.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:52:00,528 INFO [finetune.py:976] (1/7) Epoch 23, batch 2150, loss[loss=0.204, simple_loss=0.2757, pruned_loss=0.06614, over 4933.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2449, pruned_loss=0.05139, over 955736.06 frames. ], batch size: 33, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:52:02,380 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.009e+02 1.558e+02 1.811e+02 2.178e+02 3.611e+02, threshold=3.622e+02, percent-clipped=2.0 +2023-03-27 03:52:17,020 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128185.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:52:31,846 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128206.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:52:33,600 INFO [finetune.py:976] (1/7) Epoch 23, batch 2200, loss[loss=0.151, simple_loss=0.2324, pruned_loss=0.03481, over 4827.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2465, pruned_loss=0.05143, over 955614.41 frames. ], batch size: 47, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:52:46,630 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128228.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:52:47,954 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9999, 1.9213, 1.6061, 1.6955, 1.8329, 1.7907, 1.8947, 2.4784], + device='cuda:1'), covar=tensor([0.3875, 0.4317, 0.3397, 0.3893, 0.3988, 0.2489, 0.3478, 0.1866], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0235, 0.0275, 0.0256, 0.0227, 0.0254, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:52:49,641 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=128233.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:53:07,264 INFO [finetune.py:976] (1/7) Epoch 23, batch 2250, loss[loss=0.19, simple_loss=0.2649, pruned_loss=0.05753, over 4888.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2477, pruned_loss=0.05151, over 956684.79 frames. ], batch size: 35, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:53:09,084 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.465e+01 1.491e+02 1.772e+02 2.217e+02 3.841e+02, threshold=3.544e+02, percent-clipped=1.0 +2023-03-27 03:53:28,170 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7652, 1.7922, 1.6803, 1.8173, 1.6353, 4.5532, 1.7242, 2.0561], + device='cuda:1'), covar=tensor([0.3370, 0.2481, 0.2094, 0.2290, 0.1511, 0.0124, 0.2351, 0.1234], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0121, 0.0124, 0.0114, 0.0096, 0.0095, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 03:53:40,844 INFO [finetune.py:976] (1/7) Epoch 23, batch 2300, loss[loss=0.2053, simple_loss=0.2743, pruned_loss=0.06811, over 4896.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2485, pruned_loss=0.05127, over 954591.06 frames. ], batch size: 36, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:53:47,298 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128318.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:53:47,912 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7469, 1.9199, 1.5831, 1.7663, 2.2559, 2.2684, 1.9459, 1.8449], + device='cuda:1'), covar=tensor([0.0537, 0.0358, 0.0644, 0.0344, 0.0358, 0.0642, 0.0357, 0.0400], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0106, 0.0144, 0.0111, 0.0100, 0.0111, 0.0101, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.7315e-05, 8.1618e-05, 1.1282e-04, 8.5458e-05, 7.7366e-05, 8.1967e-05, + 7.5318e-05, 8.5303e-05], device='cuda:1') +2023-03-27 03:53:48,502 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128320.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:54:13,558 INFO [finetune.py:976] (1/7) Epoch 23, batch 2350, loss[loss=0.2177, simple_loss=0.2814, pruned_loss=0.07703, over 4814.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2464, pruned_loss=0.05099, over 954296.76 frames. ], batch size: 39, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:54:15,918 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.030e+02 1.510e+02 1.720e+02 2.103e+02 3.385e+02, threshold=3.440e+02, percent-clipped=0.0 +2023-03-27 03:54:18,449 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=128366.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:54:28,984 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128381.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:54:46,896 INFO [finetune.py:976] (1/7) Epoch 23, batch 2400, loss[loss=0.172, simple_loss=0.2405, pruned_loss=0.05172, over 4914.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2448, pruned_loss=0.05124, over 956213.17 frames. ], batch size: 43, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:54:49,474 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128413.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:55:03,841 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6531, 1.4766, 1.9773, 3.0641, 2.0262, 2.2187, 0.9923, 2.5478], + device='cuda:1'), covar=tensor([0.1854, 0.1658, 0.1458, 0.0878, 0.0961, 0.1752, 0.2014, 0.0677], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0134, 0.0164, 0.0100, 0.0137, 0.0124, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 03:55:06,801 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128435.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:55:35,543 INFO [finetune.py:976] (1/7) Epoch 23, batch 2450, loss[loss=0.1361, simple_loss=0.2135, pruned_loss=0.02931, over 4856.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2425, pruned_loss=0.0507, over 953719.72 frames. ], batch size: 49, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:55:41,390 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.054e+01 1.492e+02 1.779e+02 2.209e+02 4.084e+02, threshold=3.557e+02, percent-clipped=2.0 +2023-03-27 03:55:49,845 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128474.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:55:55,742 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=128483.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:55:57,006 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6567, 1.2915, 0.8903, 1.5459, 2.0342, 1.2372, 1.4255, 1.4768], + device='cuda:1'), covar=tensor([0.1498, 0.2024, 0.1866, 0.1155, 0.1903, 0.1945, 0.1396, 0.1941], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0111, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:56:10,249 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128506.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:56:12,479 INFO [finetune.py:976] (1/7) Epoch 23, batch 2500, loss[loss=0.175, simple_loss=0.2447, pruned_loss=0.05262, over 4830.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2435, pruned_loss=0.05128, over 953943.75 frames. ], batch size: 33, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:56:24,796 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4544, 1.2862, 1.2654, 1.3540, 1.6539, 1.6167, 1.4120, 1.2474], + device='cuda:1'), covar=tensor([0.0349, 0.0351, 0.0628, 0.0328, 0.0242, 0.0458, 0.0333, 0.0426], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0107, 0.0144, 0.0111, 0.0100, 0.0111, 0.0101, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.7326e-05, 8.1845e-05, 1.1291e-04, 8.5483e-05, 7.7395e-05, 8.1959e-05, + 7.5416e-05, 8.5430e-05], device='cuda:1') +2023-03-27 03:56:25,599 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.93 vs. limit=5.0 +2023-03-27 03:56:25,957 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128528.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:56:28,300 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7018, 1.1852, 0.9502, 1.5764, 1.9755, 1.4289, 1.4182, 1.4897], + device='cuda:1'), covar=tensor([0.1460, 0.2040, 0.1827, 0.1155, 0.1953, 0.2002, 0.1448, 0.1929], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0111, 0.0093, 0.0120, 0.0095, 0.0100, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 03:56:33,673 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7191, 1.6799, 1.6204, 1.6941, 1.5630, 4.1917, 1.6071, 1.9866], + device='cuda:1'), covar=tensor([0.3226, 0.2521, 0.2110, 0.2270, 0.1502, 0.0186, 0.2358, 0.1233], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0121, 0.0124, 0.0113, 0.0096, 0.0095, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 03:56:48,370 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=128554.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:56:55,781 INFO [finetune.py:976] (1/7) Epoch 23, batch 2550, loss[loss=0.211, simple_loss=0.2847, pruned_loss=0.0687, over 4830.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2478, pruned_loss=0.05264, over 954667.74 frames. ], batch size: 33, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:56:58,589 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.110e+02 1.563e+02 1.837e+02 2.202e+02 4.665e+02, threshold=3.674e+02, percent-clipped=3.0 +2023-03-27 03:57:11,626 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=128576.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:57:19,456 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8191, 1.6539, 1.4681, 1.3186, 1.6463, 1.5628, 1.6263, 2.2057], + device='cuda:1'), covar=tensor([0.3984, 0.3646, 0.3276, 0.3429, 0.3506, 0.2372, 0.3236, 0.1767], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0262, 0.0235, 0.0276, 0.0255, 0.0227, 0.0254, 0.0236], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:57:33,476 INFO [finetune.py:976] (1/7) Epoch 23, batch 2600, loss[loss=0.1649, simple_loss=0.2409, pruned_loss=0.04451, over 4905.00 frames. ], tot_loss[loss=0.1768, simple_loss=0.2488, pruned_loss=0.05243, over 954495.66 frames. ], batch size: 43, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:58:07,053 INFO [finetune.py:976] (1/7) Epoch 23, batch 2650, loss[loss=0.1698, simple_loss=0.2476, pruned_loss=0.04603, over 4751.00 frames. ], tot_loss[loss=0.1776, simple_loss=0.2501, pruned_loss=0.05253, over 954849.02 frames. ], batch size: 26, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:58:08,890 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.089e+02 1.579e+02 1.820e+02 2.183e+02 3.562e+02, threshold=3.640e+02, percent-clipped=0.0 +2023-03-27 03:58:18,295 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7519, 3.9793, 3.6205, 2.0549, 4.1625, 3.1878, 1.0682, 2.7582], + device='cuda:1'), covar=tensor([0.2179, 0.1875, 0.1600, 0.3120, 0.0939, 0.0980, 0.4168, 0.1510], + device='cuda:1'), in_proj_covar=tensor([0.0155, 0.0180, 0.0161, 0.0130, 0.0162, 0.0124, 0.0150, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 03:58:18,911 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128676.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:58:21,536 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-27 03:58:26,707 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7337, 1.5465, 1.9673, 1.4180, 1.7296, 1.9008, 1.5285, 2.0076], + device='cuda:1'), covar=tensor([0.0912, 0.1608, 0.1018, 0.1396, 0.0697, 0.1133, 0.2226, 0.0674], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0192, 0.0189, 0.0173, 0.0213, 0.0215, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 03:58:40,901 INFO [finetune.py:976] (1/7) Epoch 23, batch 2700, loss[loss=0.1666, simple_loss=0.2307, pruned_loss=0.05128, over 4932.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2481, pruned_loss=0.05163, over 953360.26 frames. ], batch size: 33, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:59:01,816 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.61 vs. limit=2.0 +2023-03-27 03:59:14,673 INFO [finetune.py:976] (1/7) Epoch 23, batch 2750, loss[loss=0.1622, simple_loss=0.2352, pruned_loss=0.04458, over 4778.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.246, pruned_loss=0.05119, over 952902.94 frames. ], batch size: 26, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 03:59:16,465 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.314e+01 1.516e+02 1.813e+02 2.146e+02 3.615e+02, threshold=3.627e+02, percent-clipped=0.0 +2023-03-27 03:59:21,317 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128769.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:59:29,510 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128781.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 03:59:48,512 INFO [finetune.py:976] (1/7) Epoch 23, batch 2800, loss[loss=0.1329, simple_loss=0.2087, pruned_loss=0.02854, over 4868.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2426, pruned_loss=0.04993, over 954331.55 frames. ], batch size: 31, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 04:00:10,896 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128842.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:00:13,471 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-27 04:00:22,070 INFO [finetune.py:976] (1/7) Epoch 23, batch 2850, loss[loss=0.1968, simple_loss=0.2736, pruned_loss=0.06001, over 4825.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2401, pruned_loss=0.04938, over 953077.65 frames. ], batch size: 39, lr: 3.10e-03, grad_scale: 32.0 +2023-03-27 04:00:23,886 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.021e+01 1.432e+02 1.754e+02 2.169e+02 4.729e+02, threshold=3.508e+02, percent-clipped=1.0 +2023-03-27 04:00:30,413 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7728, 1.7511, 1.6611, 1.7663, 1.4724, 4.3693, 1.7635, 2.2296], + device='cuda:1'), covar=tensor([0.3061, 0.2425, 0.1982, 0.2301, 0.1560, 0.0105, 0.2252, 0.1069], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0121, 0.0124, 0.0113, 0.0096, 0.0095, 0.0096], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:00:34,672 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3270, 2.9160, 2.6918, 1.1842, 3.0101, 2.2403, 0.6894, 1.8229], + device='cuda:1'), covar=tensor([0.2658, 0.2981, 0.2136, 0.3996, 0.1402, 0.1292, 0.4452, 0.2032], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0160, 0.0129, 0.0161, 0.0124, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 04:00:52,199 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3918, 1.3686, 1.5843, 1.5143, 1.4872, 2.9681, 1.2734, 1.4536], + device='cuda:1'), covar=tensor([0.0906, 0.1733, 0.1077, 0.0969, 0.1611, 0.0281, 0.1481, 0.1779], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0081, 0.0074, 0.0076, 0.0092, 0.0081, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:00:55,206 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128890.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 04:01:08,141 INFO [finetune.py:976] (1/7) Epoch 23, batch 2900, loss[loss=0.1819, simple_loss=0.2578, pruned_loss=0.053, over 4927.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.243, pruned_loss=0.05024, over 954349.49 frames. ], batch size: 33, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:01:14,477 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.35 vs. limit=5.0 +2023-03-27 04:01:36,755 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128951.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:01:41,988 INFO [finetune.py:976] (1/7) Epoch 23, batch 2950, loss[loss=0.1842, simple_loss=0.2545, pruned_loss=0.05698, over 4886.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2464, pruned_loss=0.05153, over 950975.98 frames. ], batch size: 35, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:01:43,785 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.806e+01 1.599e+02 2.012e+02 2.314e+02 4.261e+02, threshold=4.024e+02, percent-clipped=1.0 +2023-03-27 04:01:52,363 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128976.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:02:31,486 INFO [finetune.py:976] (1/7) Epoch 23, batch 3000, loss[loss=0.1595, simple_loss=0.2272, pruned_loss=0.04591, over 4409.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2478, pruned_loss=0.05216, over 953024.67 frames. ], batch size: 19, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:02:31,486 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 04:02:39,083 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5053, 1.3531, 1.2961, 1.4073, 1.6906, 1.6113, 1.4006, 1.2535], + device='cuda:1'), covar=tensor([0.0378, 0.0329, 0.0607, 0.0339, 0.0267, 0.0486, 0.0349, 0.0442], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0106, 0.0144, 0.0111, 0.0099, 0.0111, 0.0101, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.7401e-05, 8.1668e-05, 1.1308e-04, 8.5108e-05, 7.7247e-05, 8.1928e-05, + 7.5444e-05, 8.5130e-05], device='cuda:1') +2023-03-27 04:02:42,304 INFO [finetune.py:1010] (1/7) Epoch 23, validation: loss=0.1567, simple_loss=0.225, pruned_loss=0.04424, over 2265189.00 frames. +2023-03-27 04:02:42,304 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 04:02:51,934 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=129024.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:03:14,563 INFO [finetune.py:976] (1/7) Epoch 23, batch 3050, loss[loss=0.1409, simple_loss=0.2056, pruned_loss=0.03813, over 3892.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2486, pruned_loss=0.05162, over 954168.26 frames. ], batch size: 17, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:03:16,829 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.176e+02 1.606e+02 1.899e+02 2.236e+02 5.313e+02, threshold=3.798e+02, percent-clipped=3.0 +2023-03-27 04:03:22,107 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=129069.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:03:47,842 INFO [finetune.py:976] (1/7) Epoch 23, batch 3100, loss[loss=0.1661, simple_loss=0.2512, pruned_loss=0.04045, over 4821.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2475, pruned_loss=0.05158, over 955625.57 frames. ], batch size: 38, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:03:53,277 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=129117.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:04:06,393 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=129137.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:04:20,604 INFO [finetune.py:976] (1/7) Epoch 23, batch 3150, loss[loss=0.1561, simple_loss=0.2266, pruned_loss=0.04282, over 4906.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2458, pruned_loss=0.05165, over 955491.59 frames. ], batch size: 37, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:04:22,455 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.200e+02 1.503e+02 1.751e+02 2.296e+02 3.694e+02, threshold=3.502e+02, percent-clipped=0.0 +2023-03-27 04:05:01,880 INFO [finetune.py:976] (1/7) Epoch 23, batch 3200, loss[loss=0.1199, simple_loss=0.1993, pruned_loss=0.02025, over 4753.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2412, pruned_loss=0.04957, over 955879.68 frames. ], batch size: 27, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:05:26,801 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=129246.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 04:05:30,397 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9517, 1.7796, 1.6009, 1.4131, 1.9425, 1.6934, 1.8549, 1.9247], + device='cuda:1'), covar=tensor([0.1382, 0.1994, 0.2963, 0.2544, 0.2600, 0.1694, 0.2700, 0.1887], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0188, 0.0234, 0.0253, 0.0248, 0.0204, 0.0213, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:05:37,439 INFO [finetune.py:976] (1/7) Epoch 23, batch 3250, loss[loss=0.1907, simple_loss=0.2554, pruned_loss=0.06298, over 4872.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2421, pruned_loss=0.05081, over 955386.96 frames. ], batch size: 34, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:05:39,767 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.902e+01 1.487e+02 1.735e+02 2.015e+02 4.622e+02, threshold=3.470e+02, percent-clipped=1.0 +2023-03-27 04:06:02,581 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-27 04:06:05,535 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1626, 1.8887, 2.0385, 1.5200, 2.0354, 2.1358, 2.2286, 1.6834], + device='cuda:1'), covar=tensor([0.0592, 0.0718, 0.0779, 0.0892, 0.0729, 0.0727, 0.0570, 0.1225], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0135, 0.0138, 0.0119, 0.0124, 0.0137, 0.0137, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:06:15,313 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-27 04:06:22,328 INFO [finetune.py:976] (1/7) Epoch 23, batch 3300, loss[loss=0.1722, simple_loss=0.2639, pruned_loss=0.04027, over 4922.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2444, pruned_loss=0.05122, over 953105.72 frames. ], batch size: 42, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:06:39,470 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6830, 1.5872, 1.9351, 1.2783, 1.8892, 1.8885, 1.5292, 2.0955], + device='cuda:1'), covar=tensor([0.1110, 0.2055, 0.1266, 0.1665, 0.0747, 0.1214, 0.2696, 0.0815], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0208, 0.0194, 0.0191, 0.0175, 0.0216, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:06:50,880 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7823, 1.6712, 1.4729, 1.4736, 1.8503, 1.5537, 1.8073, 1.7905], + device='cuda:1'), covar=tensor([0.1432, 0.1743, 0.2832, 0.2268, 0.2329, 0.1678, 0.2806, 0.1615], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0189, 0.0234, 0.0253, 0.0248, 0.0205, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:06:56,062 INFO [finetune.py:976] (1/7) Epoch 23, batch 3350, loss[loss=0.2088, simple_loss=0.2768, pruned_loss=0.07036, over 4789.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2477, pruned_loss=0.05265, over 954401.78 frames. ], batch size: 51, lr: 3.09e-03, grad_scale: 32.0 +2023-03-27 04:06:57,831 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.264e+01 1.616e+02 1.807e+02 2.144e+02 4.365e+02, threshold=3.613e+02, percent-clipped=1.0 +2023-03-27 04:07:00,382 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6068, 1.5756, 2.1665, 1.8486, 1.7912, 3.7944, 1.4010, 1.6731], + device='cuda:1'), covar=tensor([0.0941, 0.1709, 0.1399, 0.0967, 0.1483, 0.0223, 0.1460, 0.1724], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0092, 0.0082, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:07:12,851 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.42 vs. limit=5.0 +2023-03-27 04:07:47,795 INFO [finetune.py:976] (1/7) Epoch 23, batch 3400, loss[loss=0.2306, simple_loss=0.2862, pruned_loss=0.08753, over 4872.00 frames. ], tot_loss[loss=0.1778, simple_loss=0.2493, pruned_loss=0.05318, over 954242.92 frames. ], batch size: 34, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:07:50,371 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=129413.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:08:07,361 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=129437.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:08:19,490 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6032, 1.5327, 1.4688, 1.5619, 1.1126, 3.3956, 1.4105, 1.8167], + device='cuda:1'), covar=tensor([0.3235, 0.2536, 0.2205, 0.2397, 0.1810, 0.0214, 0.2525, 0.1219], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0124, 0.0113, 0.0095, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:08:21,220 INFO [finetune.py:976] (1/7) Epoch 23, batch 3450, loss[loss=0.1733, simple_loss=0.2488, pruned_loss=0.04892, over 4850.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2485, pruned_loss=0.05235, over 953818.71 frames. ], batch size: 49, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:08:21,349 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6062, 1.4594, 1.2972, 1.4943, 1.9440, 1.8724, 1.6081, 1.3648], + device='cuda:1'), covar=tensor([0.0451, 0.0412, 0.0883, 0.0423, 0.0285, 0.0529, 0.0390, 0.0550], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0107, 0.0145, 0.0111, 0.0100, 0.0111, 0.0102, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.7637e-05, 8.2042e-05, 1.1356e-04, 8.5527e-05, 7.7478e-05, 8.2343e-05, + 7.5919e-05, 8.5626e-05], device='cuda:1') +2023-03-27 04:08:23,471 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.549e+02 1.981e+02 2.372e+02 4.494e+02, threshold=3.962e+02, percent-clipped=6.0 +2023-03-27 04:08:31,791 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=129474.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 04:08:39,422 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=129485.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:08:54,911 INFO [finetune.py:976] (1/7) Epoch 23, batch 3500, loss[loss=0.1634, simple_loss=0.2481, pruned_loss=0.03933, over 4812.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2467, pruned_loss=0.05151, over 954122.21 frames. ], batch size: 40, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:09:20,953 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=129546.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 04:09:28,786 INFO [finetune.py:976] (1/7) Epoch 23, batch 3550, loss[loss=0.1533, simple_loss=0.2293, pruned_loss=0.03859, over 4825.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2435, pruned_loss=0.05065, over 954602.30 frames. ], batch size: 38, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:09:30,571 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.427e+01 1.468e+02 1.701e+02 2.000e+02 4.470e+02, threshold=3.402e+02, percent-clipped=1.0 +2023-03-27 04:09:52,427 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=129594.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:10:11,046 INFO [finetune.py:976] (1/7) Epoch 23, batch 3600, loss[loss=0.1542, simple_loss=0.2164, pruned_loss=0.04595, over 4889.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2413, pruned_loss=0.05031, over 955080.86 frames. ], batch size: 32, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:10:12,957 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=129612.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:10:18,881 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0872, 2.0042, 1.6723, 1.8097, 1.8202, 1.8067, 1.9086, 2.6081], + device='cuda:1'), covar=tensor([0.3356, 0.3836, 0.3033, 0.3456, 0.3478, 0.2351, 0.3354, 0.1514], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0262, 0.0235, 0.0276, 0.0255, 0.0227, 0.0254, 0.0236], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:10:44,782 INFO [finetune.py:976] (1/7) Epoch 23, batch 3650, loss[loss=0.1429, simple_loss=0.2277, pruned_loss=0.02907, over 4716.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.243, pruned_loss=0.05137, over 954190.98 frames. ], batch size: 23, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:10:46,575 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.022e+02 1.535e+02 1.802e+02 2.202e+02 3.404e+02, threshold=3.605e+02, percent-clipped=1.0 +2023-03-27 04:10:53,480 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=129673.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:10:54,637 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-27 04:11:23,352 INFO [finetune.py:976] (1/7) Epoch 23, batch 3700, loss[loss=0.2176, simple_loss=0.2918, pruned_loss=0.07173, over 4862.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2467, pruned_loss=0.052, over 954258.36 frames. ], batch size: 34, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:12:00,257 INFO [finetune.py:976] (1/7) Epoch 23, batch 3750, loss[loss=0.153, simple_loss=0.2279, pruned_loss=0.03904, over 4768.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2481, pruned_loss=0.05211, over 954300.21 frames. ], batch size: 26, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:12:02,070 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.237e+01 1.564e+02 1.874e+02 2.284e+02 3.839e+02, threshold=3.748e+02, percent-clipped=2.0 +2023-03-27 04:12:02,203 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0229, 2.8747, 2.5659, 1.3958, 2.7063, 2.0930, 2.1362, 2.4723], + device='cuda:1'), covar=tensor([0.1160, 0.0722, 0.1739, 0.2143, 0.1631, 0.2179, 0.2202, 0.1018], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0192, 0.0200, 0.0182, 0.0211, 0.0209, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:12:06,379 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=129769.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 04:12:14,991 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-03-27 04:12:20,325 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=129791.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:12:35,161 INFO [finetune.py:976] (1/7) Epoch 23, batch 3800, loss[loss=0.1497, simple_loss=0.2152, pruned_loss=0.04207, over 4690.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2487, pruned_loss=0.05251, over 953308.85 frames. ], batch size: 23, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:12:54,962 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-27 04:13:16,860 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=129852.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:13:21,830 INFO [finetune.py:976] (1/7) Epoch 23, batch 3850, loss[loss=0.1854, simple_loss=0.2486, pruned_loss=0.06117, over 4832.00 frames. ], tot_loss[loss=0.175, simple_loss=0.247, pruned_loss=0.05154, over 953219.31 frames. ], batch size: 30, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:13:24,155 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.288e+01 1.597e+02 1.881e+02 2.160e+02 3.613e+02, threshold=3.763e+02, percent-clipped=0.0 +2023-03-27 04:13:55,057 INFO [finetune.py:976] (1/7) Epoch 23, batch 3900, loss[loss=0.1267, simple_loss=0.21, pruned_loss=0.02173, over 4776.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2439, pruned_loss=0.05066, over 953054.76 frames. ], batch size: 29, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:14:00,427 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9761, 1.3880, 2.0370, 2.0018, 1.7978, 1.7351, 1.9495, 1.8919], + device='cuda:1'), covar=tensor([0.3983, 0.4094, 0.3316, 0.3869, 0.4738, 0.4117, 0.4744, 0.3166], + device='cuda:1'), in_proj_covar=tensor([0.0257, 0.0242, 0.0263, 0.0285, 0.0284, 0.0261, 0.0293, 0.0246], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:14:03,999 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9386, 4.7230, 4.5133, 2.6697, 4.7502, 3.6519, 1.1904, 3.2607], + device='cuda:1'), covar=tensor([0.2483, 0.1714, 0.1393, 0.2926, 0.0918, 0.0946, 0.4312, 0.1452], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0160, 0.0129, 0.0161, 0.0123, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 04:14:27,719 INFO [finetune.py:976] (1/7) Epoch 23, batch 3950, loss[loss=0.145, simple_loss=0.2184, pruned_loss=0.03581, over 4739.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2407, pruned_loss=0.04953, over 956155.60 frames. ], batch size: 27, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:14:29,947 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.481e+02 1.820e+02 2.101e+02 4.779e+02, threshold=3.640e+02, percent-clipped=1.0 +2023-03-27 04:14:34,551 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=129968.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:14:50,432 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.64 vs. limit=5.0 +2023-03-27 04:14:56,374 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130000.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:14:58,099 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130002.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:15:02,811 INFO [finetune.py:976] (1/7) Epoch 23, batch 4000, loss[loss=0.1955, simple_loss=0.2678, pruned_loss=0.06161, over 4927.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2416, pruned_loss=0.05092, over 956238.82 frames. ], batch size: 33, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:15:45,414 INFO [finetune.py:976] (1/7) Epoch 23, batch 4050, loss[loss=0.1825, simple_loss=0.2608, pruned_loss=0.05209, over 4759.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2457, pruned_loss=0.05254, over 956569.71 frames. ], batch size: 54, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:15:47,211 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130061.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:15:47,689 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.063e+02 1.665e+02 1.960e+02 2.479e+02 5.275e+02, threshold=3.921e+02, percent-clipped=4.0 +2023-03-27 04:15:48,445 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130063.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 04:15:53,622 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130069.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:16:19,200 INFO [finetune.py:976] (1/7) Epoch 23, batch 4100, loss[loss=0.1647, simple_loss=0.2298, pruned_loss=0.04986, over 4711.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2473, pruned_loss=0.05261, over 956624.03 frames. ], batch size: 23, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:16:26,578 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=130117.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 04:16:55,003 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130147.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:17:02,631 INFO [finetune.py:976] (1/7) Epoch 23, batch 4150, loss[loss=0.1376, simple_loss=0.2161, pruned_loss=0.02959, over 4891.00 frames. ], tot_loss[loss=0.1781, simple_loss=0.2496, pruned_loss=0.05323, over 956007.72 frames. ], batch size: 32, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:17:04,904 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.036e+02 1.505e+02 1.863e+02 2.291e+02 4.324e+02, threshold=3.726e+02, percent-clipped=3.0 +2023-03-27 04:17:36,688 INFO [finetune.py:976] (1/7) Epoch 23, batch 4200, loss[loss=0.1489, simple_loss=0.2341, pruned_loss=0.03187, over 4792.00 frames. ], tot_loss[loss=0.1774, simple_loss=0.2497, pruned_loss=0.05256, over 954958.98 frames. ], batch size: 51, lr: 3.09e-03, grad_scale: 64.0 +2023-03-27 04:18:24,042 INFO [finetune.py:976] (1/7) Epoch 23, batch 4250, loss[loss=0.1683, simple_loss=0.2362, pruned_loss=0.05016, over 4904.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2478, pruned_loss=0.05199, over 955698.32 frames. ], batch size: 36, lr: 3.08e-03, grad_scale: 64.0 +2023-03-27 04:18:25,853 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.027e+02 1.516e+02 1.759e+02 2.094e+02 3.793e+02, threshold=3.518e+02, percent-clipped=1.0 +2023-03-27 04:18:30,089 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130268.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:18:57,489 INFO [finetune.py:976] (1/7) Epoch 23, batch 4300, loss[loss=0.1482, simple_loss=0.2235, pruned_loss=0.03647, over 4892.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2448, pruned_loss=0.05117, over 957237.75 frames. ], batch size: 32, lr: 3.08e-03, grad_scale: 64.0 +2023-03-27 04:19:02,809 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=130316.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:19:17,313 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5020, 2.4107, 2.3264, 1.7050, 2.0846, 2.4549, 2.5350, 2.0343], + device='cuda:1'), covar=tensor([0.0534, 0.0576, 0.0776, 0.0892, 0.1465, 0.0759, 0.0533, 0.1050], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0135, 0.0138, 0.0119, 0.0125, 0.0138, 0.0137, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:19:29,228 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130356.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:19:30,469 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130358.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 04:19:31,029 INFO [finetune.py:976] (1/7) Epoch 23, batch 4350, loss[loss=0.1745, simple_loss=0.2436, pruned_loss=0.05271, over 4743.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.242, pruned_loss=0.05017, over 956751.37 frames. ], batch size: 54, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:19:31,737 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7266, 1.6301, 1.6234, 1.6767, 1.3305, 3.6845, 1.6199, 1.9298], + device='cuda:1'), covar=tensor([0.3319, 0.2552, 0.2183, 0.2410, 0.1740, 0.0198, 0.2447, 0.1272], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0123, 0.0114, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:19:32,364 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1609, 1.3241, 1.4791, 1.0465, 1.0978, 1.4075, 1.3189, 1.5614], + device='cuda:1'), covar=tensor([0.1078, 0.1664, 0.1063, 0.1220, 0.0833, 0.1003, 0.2319, 0.0695], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0191, 0.0190, 0.0173, 0.0213, 0.0215, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:19:33,423 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.012e+02 1.417e+02 1.746e+02 2.112e+02 4.412e+02, threshold=3.492e+02, percent-clipped=1.0 +2023-03-27 04:19:35,900 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8254, 0.9585, 1.7026, 1.6780, 1.5362, 1.4492, 1.6141, 1.7128], + device='cuda:1'), covar=tensor([0.3787, 0.3612, 0.3386, 0.3578, 0.4721, 0.4001, 0.4117, 0.3226], + device='cuda:1'), in_proj_covar=tensor([0.0256, 0.0242, 0.0262, 0.0285, 0.0285, 0.0261, 0.0292, 0.0246], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:19:44,168 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0449, 1.8098, 1.8893, 1.2953, 1.9171, 1.9242, 1.9693, 1.5764], + device='cuda:1'), covar=tensor([0.0554, 0.0769, 0.0783, 0.0935, 0.0758, 0.0746, 0.0636, 0.1222], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0135, 0.0138, 0.0119, 0.0125, 0.0138, 0.0137, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:19:48,203 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130384.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:20:04,340 INFO [finetune.py:976] (1/7) Epoch 23, batch 4400, loss[loss=0.1607, simple_loss=0.2436, pruned_loss=0.03894, over 4806.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2439, pruned_loss=0.05168, over 955358.63 frames. ], batch size: 45, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:20:19,511 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130432.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:20:24,167 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130438.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:20:30,571 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130445.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:20:31,759 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130447.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:20:39,894 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-03-27 04:20:46,860 INFO [finetune.py:976] (1/7) Epoch 23, batch 4450, loss[loss=0.2443, simple_loss=0.3179, pruned_loss=0.08534, over 4803.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.246, pruned_loss=0.05187, over 953789.45 frames. ], batch size: 45, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:20:46,995 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1621, 2.8111, 2.6201, 1.3644, 2.6618, 2.2395, 2.1190, 2.5662], + device='cuda:1'), covar=tensor([0.1005, 0.0801, 0.1659, 0.2142, 0.1705, 0.2136, 0.2257, 0.1079], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0192, 0.0201, 0.0182, 0.0211, 0.0209, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:20:49,236 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.124e+02 1.491e+02 1.813e+02 2.246e+02 3.707e+02, threshold=3.626e+02, percent-clipped=3.0 +2023-03-27 04:21:10,560 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130493.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:21:11,728 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=130495.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:21:14,692 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130499.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:21:20,674 INFO [finetune.py:976] (1/7) Epoch 23, batch 4500, loss[loss=0.2443, simple_loss=0.3049, pruned_loss=0.09185, over 4841.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2474, pruned_loss=0.05194, over 954342.44 frames. ], batch size: 44, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:21:53,293 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130548.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:22:04,071 INFO [finetune.py:976] (1/7) Epoch 23, batch 4550, loss[loss=0.2333, simple_loss=0.3069, pruned_loss=0.0799, over 4727.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2491, pruned_loss=0.0524, over 954875.90 frames. ], batch size: 59, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:22:06,500 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.768e+01 1.546e+02 1.777e+02 2.233e+02 3.779e+02, threshold=3.553e+02, percent-clipped=2.0 +2023-03-27 04:22:07,819 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130565.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:22:37,457 INFO [finetune.py:976] (1/7) Epoch 23, batch 4600, loss[loss=0.1698, simple_loss=0.2273, pruned_loss=0.05619, over 4888.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2475, pruned_loss=0.05155, over 954530.48 frames. ], batch size: 35, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:22:37,566 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130609.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 04:22:47,777 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130626.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:23:10,855 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130656.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:23:17,057 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130658.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:23:17,583 INFO [finetune.py:976] (1/7) Epoch 23, batch 4650, loss[loss=0.153, simple_loss=0.2174, pruned_loss=0.0443, over 4737.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2446, pruned_loss=0.05082, over 955068.52 frames. ], batch size: 28, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:23:19,983 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.906e+01 1.515e+02 1.768e+02 2.232e+02 6.495e+02, threshold=3.536e+02, percent-clipped=3.0 +2023-03-27 04:23:54,690 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=130704.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:23:55,881 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=130706.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:23:58,153 INFO [finetune.py:976] (1/7) Epoch 23, batch 4700, loss[loss=0.1355, simple_loss=0.2099, pruned_loss=0.03056, over 4906.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2417, pruned_loss=0.05034, over 956191.79 frames. ], batch size: 43, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:24:00,092 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6905, 1.4762, 1.0634, 0.2252, 1.2633, 1.5252, 1.5421, 1.4613], + device='cuda:1'), covar=tensor([0.0852, 0.0934, 0.1512, 0.2116, 0.1515, 0.2586, 0.2312, 0.0879], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0192, 0.0201, 0.0182, 0.0211, 0.0209, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:24:18,049 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130740.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:24:31,365 INFO [finetune.py:976] (1/7) Epoch 23, batch 4750, loss[loss=0.1795, simple_loss=0.2481, pruned_loss=0.0554, over 4752.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2398, pruned_loss=0.05, over 954139.55 frames. ], batch size: 54, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:24:34,231 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.253e+02 1.528e+02 1.803e+02 2.150e+02 3.686e+02, threshold=3.606e+02, percent-clipped=2.0 +2023-03-27 04:24:49,884 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130788.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:24:54,004 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130794.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:25:04,663 INFO [finetune.py:976] (1/7) Epoch 23, batch 4800, loss[loss=0.1903, simple_loss=0.2664, pruned_loss=0.0571, over 4909.00 frames. ], tot_loss[loss=0.1733, simple_loss=0.2438, pruned_loss=0.0514, over 953449.34 frames. ], batch size: 43, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:25:10,729 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6035, 0.7515, 1.6726, 1.5962, 1.4785, 1.4403, 1.5137, 1.6459], + device='cuda:1'), covar=tensor([0.3549, 0.3601, 0.2985, 0.3167, 0.4052, 0.3406, 0.3807, 0.2650], + device='cuda:1'), in_proj_covar=tensor([0.0257, 0.0241, 0.0261, 0.0285, 0.0284, 0.0260, 0.0292, 0.0245], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:25:13,150 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.56 vs. limit=2.0 +2023-03-27 04:25:37,285 INFO [finetune.py:976] (1/7) Epoch 23, batch 4850, loss[loss=0.1712, simple_loss=0.2282, pruned_loss=0.05709, over 4167.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2465, pruned_loss=0.0522, over 952066.41 frames. ], batch size: 18, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:25:37,418 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7100, 1.7747, 1.5041, 1.9758, 2.3525, 1.9400, 1.6718, 1.4000], + device='cuda:1'), covar=tensor([0.2309, 0.1967, 0.1967, 0.1580, 0.1709, 0.1219, 0.2313, 0.2102], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0213, 0.0196, 0.0244, 0.0189, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:25:40,092 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.117e+02 1.613e+02 1.947e+02 2.336e+02 6.046e+02, threshold=3.894e+02, percent-clipped=4.0 +2023-03-27 04:25:43,969 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0149, 1.7527, 2.0479, 2.0796, 1.7856, 1.8137, 1.9909, 1.9115], + device='cuda:1'), covar=tensor([0.4324, 0.3998, 0.3245, 0.4006, 0.4968, 0.4123, 0.4624, 0.3176], + device='cuda:1'), in_proj_covar=tensor([0.0257, 0.0242, 0.0262, 0.0285, 0.0285, 0.0261, 0.0292, 0.0246], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:25:57,075 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-27 04:26:15,654 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130904.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:26:19,115 INFO [finetune.py:976] (1/7) Epoch 23, batch 4900, loss[loss=0.1963, simple_loss=0.2644, pruned_loss=0.06411, over 4835.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2474, pruned_loss=0.05236, over 952876.32 frames. ], batch size: 47, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:26:28,441 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130921.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:26:37,515 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4993, 2.2584, 1.7952, 0.7912, 1.9291, 1.9710, 1.8312, 2.0209], + device='cuda:1'), covar=tensor([0.0792, 0.0828, 0.1620, 0.2135, 0.1375, 0.2337, 0.2351, 0.0956], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0192, 0.0201, 0.0182, 0.0210, 0.0210, 0.0225, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:26:52,302 INFO [finetune.py:976] (1/7) Epoch 23, batch 4950, loss[loss=0.1682, simple_loss=0.2551, pruned_loss=0.04061, over 4820.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2485, pruned_loss=0.05218, over 953993.77 frames. ], batch size: 38, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:26:57,593 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.290e+01 1.586e+02 1.789e+02 2.374e+02 3.586e+02, threshold=3.578e+02, percent-clipped=0.0 +2023-03-27 04:27:36,348 INFO [finetune.py:976] (1/7) Epoch 23, batch 5000, loss[loss=0.1794, simple_loss=0.2417, pruned_loss=0.05853, over 4825.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2471, pruned_loss=0.05171, over 956633.50 frames. ], batch size: 38, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:27:57,571 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131040.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:28:09,927 INFO [finetune.py:976] (1/7) Epoch 23, batch 5050, loss[loss=0.1576, simple_loss=0.2301, pruned_loss=0.0426, over 4828.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2444, pruned_loss=0.05094, over 956872.55 frames. ], batch size: 33, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:28:12,371 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.026e+02 1.381e+02 1.770e+02 2.059e+02 4.416e+02, threshold=3.539e+02, percent-clipped=4.0 +2023-03-27 04:28:24,012 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-03-27 04:28:41,480 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=131088.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:28:41,524 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131088.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:28:45,153 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131094.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:28:57,921 INFO [finetune.py:976] (1/7) Epoch 23, batch 5100, loss[loss=0.1731, simple_loss=0.241, pruned_loss=0.0526, over 4887.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2413, pruned_loss=0.05025, over 956072.72 frames. ], batch size: 32, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:29:11,440 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.3002, 2.9931, 3.0676, 3.2347, 3.0848, 2.9280, 3.3489, 0.9961], + device='cuda:1'), covar=tensor([0.1167, 0.1052, 0.1123, 0.1220, 0.1808, 0.2102, 0.1257, 0.6066], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0246, 0.0281, 0.0294, 0.0340, 0.0287, 0.0306, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:29:17,264 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=131136.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:29:20,883 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=131142.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:29:29,398 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8211, 3.3767, 3.5177, 3.6516, 3.5788, 3.4032, 3.9094, 1.2064], + device='cuda:1'), covar=tensor([0.1035, 0.1016, 0.1009, 0.1359, 0.1528, 0.1876, 0.0921, 0.6231], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0246, 0.0280, 0.0294, 0.0340, 0.0287, 0.0306, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:29:31,085 INFO [finetune.py:976] (1/7) Epoch 23, batch 5150, loss[loss=0.175, simple_loss=0.2499, pruned_loss=0.05008, over 4809.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2419, pruned_loss=0.05076, over 955202.49 frames. ], batch size: 41, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:29:34,466 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.214e+01 1.572e+02 1.903e+02 2.241e+02 4.010e+02, threshold=3.805e+02, percent-clipped=1.0 +2023-03-27 04:29:41,273 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131174.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:29:49,478 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8793, 1.9208, 1.9026, 1.8728, 1.7526, 3.6796, 1.9424, 2.3999], + device='cuda:1'), covar=tensor([0.2684, 0.1955, 0.1645, 0.1934, 0.1304, 0.0265, 0.2483, 0.0889], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0123, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:30:01,315 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131204.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:30:04,268 INFO [finetune.py:976] (1/7) Epoch 23, batch 5200, loss[loss=0.2109, simple_loss=0.2841, pruned_loss=0.06881, over 4760.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2445, pruned_loss=0.05112, over 954957.31 frames. ], batch size: 54, lr: 3.08e-03, grad_scale: 32.0 +2023-03-27 04:30:12,620 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131221.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:30:22,997 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131235.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:30:33,245 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=131252.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:30:37,393 INFO [finetune.py:976] (1/7) Epoch 23, batch 5250, loss[loss=0.1644, simple_loss=0.2372, pruned_loss=0.04579, over 4216.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2471, pruned_loss=0.05166, over 953580.48 frames. ], batch size: 65, lr: 3.08e-03, grad_scale: 16.0 +2023-03-27 04:30:40,885 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.130e+02 1.531e+02 1.792e+02 2.239e+02 3.281e+02, threshold=3.585e+02, percent-clipped=0.0 +2023-03-27 04:30:44,379 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=131269.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:31:12,544 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0101, 1.4089, 0.9247, 1.8924, 2.3505, 1.8061, 1.7466, 1.8380], + device='cuda:1'), covar=tensor([0.1518, 0.2104, 0.1978, 0.1239, 0.1939, 0.1874, 0.1384, 0.1984], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0119, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 04:31:21,562 INFO [finetune.py:976] (1/7) Epoch 23, batch 5300, loss[loss=0.1367, simple_loss=0.2117, pruned_loss=0.03083, over 4775.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2485, pruned_loss=0.05204, over 954743.10 frames. ], batch size: 26, lr: 3.08e-03, grad_scale: 16.0 +2023-03-27 04:31:42,533 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8166, 1.3196, 0.8292, 1.6251, 2.1109, 1.5549, 1.6223, 1.6662], + device='cuda:1'), covar=tensor([0.1425, 0.2074, 0.1977, 0.1223, 0.1892, 0.1850, 0.1443, 0.1993], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0119, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 04:31:54,370 INFO [finetune.py:976] (1/7) Epoch 23, batch 5350, loss[loss=0.1706, simple_loss=0.2368, pruned_loss=0.05222, over 4729.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2482, pruned_loss=0.05168, over 956583.06 frames. ], batch size: 23, lr: 3.08e-03, grad_scale: 16.0 +2023-03-27 04:31:57,384 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.087e+02 1.529e+02 1.830e+02 2.196e+02 3.219e+02, threshold=3.659e+02, percent-clipped=0.0 +2023-03-27 04:32:35,784 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8775, 2.0710, 1.7664, 1.8460, 2.4936, 2.5868, 2.0618, 1.9984], + device='cuda:1'), covar=tensor([0.0446, 0.0337, 0.0638, 0.0325, 0.0305, 0.0461, 0.0376, 0.0446], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0106, 0.0143, 0.0111, 0.0099, 0.0111, 0.0101, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.6887e-05, 8.1110e-05, 1.1217e-04, 8.4783e-05, 7.7205e-05, 8.1885e-05, + 7.4936e-05, 8.5049e-05], device='cuda:1') +2023-03-27 04:32:38,097 INFO [finetune.py:976] (1/7) Epoch 23, batch 5400, loss[loss=0.181, simple_loss=0.2445, pruned_loss=0.05875, over 4937.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2457, pruned_loss=0.05151, over 953128.04 frames. ], batch size: 38, lr: 3.08e-03, grad_scale: 16.0 +2023-03-27 04:32:38,212 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131409.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:32:40,024 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0448, 1.8513, 2.3127, 1.4984, 2.0243, 2.3103, 1.7295, 2.4012], + device='cuda:1'), covar=tensor([0.1213, 0.1897, 0.1734, 0.2026, 0.1054, 0.1503, 0.2708, 0.0901], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0205, 0.0191, 0.0190, 0.0172, 0.0213, 0.0215, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:32:42,394 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5059, 3.3661, 3.2757, 1.4561, 3.4769, 2.6853, 0.8726, 2.3119], + device='cuda:1'), covar=tensor([0.2288, 0.2028, 0.1595, 0.3414, 0.1144, 0.1023, 0.4217, 0.1523], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0177, 0.0160, 0.0129, 0.0160, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 04:32:42,432 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131416.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:33:11,759 INFO [finetune.py:976] (1/7) Epoch 23, batch 5450, loss[loss=0.1868, simple_loss=0.2587, pruned_loss=0.05743, over 4808.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2433, pruned_loss=0.05117, over 953289.37 frames. ], batch size: 38, lr: 3.08e-03, grad_scale: 16.0 +2023-03-27 04:33:14,785 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.094e+02 1.514e+02 1.875e+02 2.409e+02 5.439e+02, threshold=3.749e+02, percent-clipped=4.0 +2023-03-27 04:33:18,565 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131470.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:33:23,307 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131477.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:33:33,328 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131492.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:33:51,830 INFO [finetune.py:976] (1/7) Epoch 23, batch 5500, loss[loss=0.1476, simple_loss=0.2203, pruned_loss=0.03746, over 4825.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.241, pruned_loss=0.05016, over 954440.50 frames. ], batch size: 38, lr: 3.08e-03, grad_scale: 16.0 +2023-03-27 04:34:12,684 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=131530.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 04:34:29,179 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131553.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:34:33,174 INFO [finetune.py:976] (1/7) Epoch 23, batch 5550, loss[loss=0.2301, simple_loss=0.2939, pruned_loss=0.0832, over 4828.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.243, pruned_loss=0.05109, over 952533.20 frames. ], batch size: 33, lr: 3.08e-03, grad_scale: 16.0 +2023-03-27 04:34:36,709 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.823e+01 1.422e+02 1.728e+02 2.186e+02 5.215e+02, threshold=3.457e+02, percent-clipped=2.0 +2023-03-27 04:35:04,719 INFO [finetune.py:976] (1/7) Epoch 23, batch 5600, loss[loss=0.2008, simple_loss=0.2765, pruned_loss=0.06249, over 4864.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2466, pruned_loss=0.05178, over 953200.46 frames. ], batch size: 31, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:35:29,480 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8652, 1.8196, 1.5914, 1.9062, 2.3706, 1.9945, 1.5331, 1.5283], + device='cuda:1'), covar=tensor([0.2117, 0.1819, 0.1783, 0.1550, 0.1475, 0.1127, 0.2287, 0.1789], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0209, 0.0212, 0.0194, 0.0242, 0.0189, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:35:34,623 INFO [finetune.py:976] (1/7) Epoch 23, batch 5650, loss[loss=0.1292, simple_loss=0.2036, pruned_loss=0.0274, over 4772.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.248, pruned_loss=0.05129, over 953348.31 frames. ], batch size: 28, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:35:37,860 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.955e+01 1.494e+02 1.801e+02 2.339e+02 4.576e+02, threshold=3.601e+02, percent-clipped=4.0 +2023-03-27 04:36:04,505 INFO [finetune.py:976] (1/7) Epoch 23, batch 5700, loss[loss=0.1616, simple_loss=0.2322, pruned_loss=0.04552, over 4275.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2433, pruned_loss=0.0498, over 937639.68 frames. ], batch size: 18, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:36:40,058 INFO [finetune.py:976] (1/7) Epoch 24, batch 0, loss[loss=0.1549, simple_loss=0.2368, pruned_loss=0.03655, over 4781.00 frames. ], tot_loss[loss=0.1549, simple_loss=0.2368, pruned_loss=0.03655, over 4781.00 frames. ], batch size: 28, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:36:40,058 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 04:36:42,907 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0599, 1.8638, 1.6464, 1.7238, 1.7938, 1.8283, 1.7880, 2.5044], + device='cuda:1'), covar=tensor([0.4201, 0.4564, 0.3419, 0.4065, 0.4305, 0.2536, 0.3846, 0.1909], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0263, 0.0234, 0.0275, 0.0256, 0.0226, 0.0253, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:36:49,217 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1611, 1.9279, 1.8127, 1.8326, 1.9242, 1.9954, 1.9266, 2.6199], + device='cuda:1'), covar=tensor([0.3879, 0.4966, 0.3541, 0.3977, 0.4141, 0.2512, 0.3843, 0.1847], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0263, 0.0234, 0.0275, 0.0256, 0.0226, 0.0253, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:36:50,762 INFO [finetune.py:1010] (1/7) Epoch 24, validation: loss=0.1594, simple_loss=0.227, pruned_loss=0.04592, over 2265189.00 frames. +2023-03-27 04:36:50,762 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 04:37:06,605 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-27 04:37:07,462 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 7.206e+01 1.398e+02 1.674e+02 2.004e+02 3.219e+02, threshold=3.348e+02, percent-clipped=0.0 +2023-03-27 04:37:08,156 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=131765.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:37:12,915 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=131772.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:37:14,332 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.73 vs. limit=5.0 +2023-03-27 04:37:22,868 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8384, 1.9152, 1.7188, 1.7694, 2.3193, 2.3355, 2.0537, 1.8422], + device='cuda:1'), covar=tensor([0.0448, 0.0366, 0.0587, 0.0366, 0.0295, 0.0520, 0.0354, 0.0453], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0105, 0.0142, 0.0109, 0.0098, 0.0110, 0.0100, 0.0111], + device='cuda:1'), out_proj_covar=tensor([7.6409e-05, 8.0382e-05, 1.1084e-04, 8.3785e-05, 7.6317e-05, 8.1508e-05, + 7.4082e-05, 8.4062e-05], device='cuda:1') +2023-03-27 04:37:25,289 INFO [finetune.py:976] (1/7) Epoch 24, batch 50, loss[loss=0.1501, simple_loss=0.2217, pruned_loss=0.03921, over 4862.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2461, pruned_loss=0.05183, over 216920.39 frames. ], batch size: 34, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:37:51,395 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7439, 3.9323, 3.7575, 1.9038, 4.0648, 3.0142, 0.9183, 2.6311], + device='cuda:1'), covar=tensor([0.1994, 0.1904, 0.1341, 0.2996, 0.0952, 0.0949, 0.3924, 0.1475], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0161, 0.0129, 0.0161, 0.0123, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 04:38:02,583 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131830.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:38:07,275 INFO [finetune.py:976] (1/7) Epoch 24, batch 100, loss[loss=0.1651, simple_loss=0.2392, pruned_loss=0.04552, over 4908.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2419, pruned_loss=0.04948, over 381831.34 frames. ], batch size: 37, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:38:15,489 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=131848.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:38:20,990 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131857.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:38:25,097 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.662e+01 1.465e+02 1.761e+02 2.142e+02 3.724e+02, threshold=3.523e+02, percent-clipped=1.0 +2023-03-27 04:38:34,588 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=131878.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:38:40,511 INFO [finetune.py:976] (1/7) Epoch 24, batch 150, loss[loss=0.165, simple_loss=0.2298, pruned_loss=0.05009, over 4754.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2384, pruned_loss=0.04884, over 508449.05 frames. ], batch size: 26, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:38:45,055 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3750, 1.4242, 1.5030, 1.5870, 1.5838, 2.7852, 1.3978, 1.5295], + device='cuda:1'), covar=tensor([0.0959, 0.1731, 0.1157, 0.0888, 0.1542, 0.0309, 0.1424, 0.1751], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0081, 0.0073, 0.0076, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:39:10,826 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131918.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:39:29,542 INFO [finetune.py:976] (1/7) Epoch 24, batch 200, loss[loss=0.1694, simple_loss=0.2435, pruned_loss=0.04768, over 4836.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.2373, pruned_loss=0.04896, over 608936.55 frames. ], batch size: 30, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:39:51,182 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.027e+02 1.517e+02 1.799e+02 2.123e+02 6.232e+02, threshold=3.598e+02, percent-clipped=3.0 +2023-03-27 04:39:52,097 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.52 vs. limit=5.0 +2023-03-27 04:40:06,649 INFO [finetune.py:976] (1/7) Epoch 24, batch 250, loss[loss=0.2175, simple_loss=0.2981, pruned_loss=0.06844, over 4834.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2415, pruned_loss=0.05047, over 684855.45 frames. ], batch size: 49, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:40:10,322 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-03-27 04:40:36,969 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132031.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:40:41,433 INFO [finetune.py:976] (1/7) Epoch 24, batch 300, loss[loss=0.172, simple_loss=0.2386, pruned_loss=0.0527, over 4926.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2446, pruned_loss=0.05108, over 744134.50 frames. ], batch size: 33, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:40:52,214 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.59 vs. limit=5.0 +2023-03-27 04:40:53,239 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132054.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:40:59,163 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.071e+02 1.635e+02 1.887e+02 2.261e+02 6.512e+02, threshold=3.774e+02, percent-clipped=2.0 +2023-03-27 04:40:59,869 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=132065.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:41:04,176 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=132072.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:41:08,891 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6113, 1.4718, 1.5146, 1.5484, 1.1110, 3.5325, 1.3281, 1.7640], + device='cuda:1'), covar=tensor([0.3206, 0.2577, 0.2144, 0.2431, 0.1789, 0.0172, 0.2622, 0.1277], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0123, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:41:14,119 INFO [finetune.py:976] (1/7) Epoch 24, batch 350, loss[loss=0.1837, simple_loss=0.2649, pruned_loss=0.05123, over 4784.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2474, pruned_loss=0.05222, over 791559.85 frames. ], batch size: 51, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:41:17,752 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132092.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:41:33,450 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=132113.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:41:34,745 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132115.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 04:41:42,166 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=132120.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:41:42,203 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0584, 1.3461, 0.8154, 1.8761, 2.3462, 1.8422, 1.7157, 1.7754], + device='cuda:1'), covar=tensor([0.1359, 0.2082, 0.1961, 0.1108, 0.1800, 0.1768, 0.1343, 0.1981], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0110, 0.0092, 0.0119, 0.0094, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 04:41:56,058 INFO [finetune.py:976] (1/7) Epoch 24, batch 400, loss[loss=0.2023, simple_loss=0.271, pruned_loss=0.06677, over 4235.00 frames. ], tot_loss[loss=0.1766, simple_loss=0.2486, pruned_loss=0.05233, over 828263.57 frames. ], batch size: 65, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:42:03,885 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=132148.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:42:15,409 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.594e+02 1.883e+02 2.349e+02 4.739e+02, threshold=3.766e+02, percent-clipped=1.0 +2023-03-27 04:42:29,849 INFO [finetune.py:976] (1/7) Epoch 24, batch 450, loss[loss=0.1942, simple_loss=0.2517, pruned_loss=0.06835, over 4791.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2474, pruned_loss=0.05158, over 857007.39 frames. ], batch size: 51, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:42:33,469 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6851, 1.4733, 1.0766, 0.2726, 1.2945, 1.4848, 1.4181, 1.4899], + device='cuda:1'), covar=tensor([0.0948, 0.0879, 0.1512, 0.2033, 0.1476, 0.2516, 0.2374, 0.0890], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0193, 0.0201, 0.0182, 0.0211, 0.0211, 0.0225, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:42:36,329 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=132196.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:42:55,013 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=132213.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:43:02,541 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4905, 1.4880, 1.5147, 0.7572, 1.7057, 1.7041, 1.7776, 1.3727], + device='cuda:1'), covar=tensor([0.0882, 0.0646, 0.0592, 0.0536, 0.0467, 0.0621, 0.0353, 0.0752], + device='cuda:1'), in_proj_covar=tensor([0.0120, 0.0146, 0.0124, 0.0120, 0.0129, 0.0128, 0.0138, 0.0147], + device='cuda:1'), out_proj_covar=tensor([8.7761e-05, 1.0549e-04, 8.8653e-05, 8.4635e-05, 9.0829e-05, 9.0907e-05, + 9.8689e-05, 1.0488e-04], device='cuda:1') +2023-03-27 04:43:13,241 INFO [finetune.py:976] (1/7) Epoch 24, batch 500, loss[loss=0.1525, simple_loss=0.2145, pruned_loss=0.04524, over 4752.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2449, pruned_loss=0.05114, over 880062.11 frames. ], batch size: 54, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:43:32,460 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 7.902e+01 1.519e+02 1.809e+02 2.135e+02 3.897e+02, threshold=3.617e+02, percent-clipped=1.0 +2023-03-27 04:43:36,956 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1892, 2.0146, 1.7822, 1.9380, 1.9037, 1.8932, 1.9553, 2.6794], + device='cuda:1'), covar=tensor([0.3377, 0.4183, 0.3084, 0.3686, 0.4014, 0.2345, 0.3689, 0.1547], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0232, 0.0274, 0.0256, 0.0225, 0.0252, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:43:46,922 INFO [finetune.py:976] (1/7) Epoch 24, batch 550, loss[loss=0.1821, simple_loss=0.2543, pruned_loss=0.05494, over 4771.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2427, pruned_loss=0.05059, over 896810.16 frames. ], batch size: 54, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:44:26,902 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-27 04:44:30,200 INFO [finetune.py:976] (1/7) Epoch 24, batch 600, loss[loss=0.2135, simple_loss=0.2877, pruned_loss=0.06964, over 4863.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2439, pruned_loss=0.05156, over 909876.70 frames. ], batch size: 44, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:44:58,205 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.737e+01 1.576e+02 1.859e+02 2.330e+02 3.343e+02, threshold=3.718e+02, percent-clipped=0.0 +2023-03-27 04:45:12,585 INFO [finetune.py:976] (1/7) Epoch 24, batch 650, loss[loss=0.2191, simple_loss=0.2824, pruned_loss=0.07794, over 4817.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2471, pruned_loss=0.05261, over 920130.16 frames. ], batch size: 40, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:45:12,656 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=132387.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:45:28,112 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=132410.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 04:45:43,331 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.57 vs. limit=2.0 +2023-03-27 04:45:46,140 INFO [finetune.py:976] (1/7) Epoch 24, batch 700, loss[loss=0.1468, simple_loss=0.2112, pruned_loss=0.04121, over 4822.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2475, pruned_loss=0.05223, over 928735.16 frames. ], batch size: 25, lr: 3.07e-03, grad_scale: 16.0 +2023-03-27 04:46:03,853 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.085e+02 1.592e+02 1.911e+02 2.262e+02 4.191e+02, threshold=3.822e+02, percent-clipped=2.0 +2023-03-27 04:46:19,325 INFO [finetune.py:976] (1/7) Epoch 24, batch 750, loss[loss=0.2242, simple_loss=0.3035, pruned_loss=0.07241, over 4757.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2489, pruned_loss=0.05278, over 934983.98 frames. ], batch size: 54, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:46:36,490 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=132513.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:46:58,682 INFO [finetune.py:976] (1/7) Epoch 24, batch 800, loss[loss=0.1735, simple_loss=0.241, pruned_loss=0.05297, over 4782.00 frames. ], tot_loss[loss=0.176, simple_loss=0.2483, pruned_loss=0.05184, over 939739.35 frames. ], batch size: 29, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:47:10,465 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4683, 1.5088, 2.2556, 1.8212, 1.8919, 4.3179, 1.4390, 1.8211], + device='cuda:1'), covar=tensor([0.1082, 0.1828, 0.1202, 0.0993, 0.1566, 0.0222, 0.1609, 0.1822], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0081, 0.0073, 0.0076, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:47:17,150 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=132561.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:47:19,959 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.775e+01 1.480e+02 1.729e+02 2.075e+02 4.531e+02, threshold=3.459e+02, percent-clipped=1.0 +2023-03-27 04:47:31,181 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8281, 1.6584, 2.3019, 3.1552, 2.2170, 2.4017, 1.2664, 2.6026], + device='cuda:1'), covar=tensor([0.1437, 0.1122, 0.0981, 0.0495, 0.0653, 0.2062, 0.1511, 0.0467], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0131, 0.0161, 0.0099, 0.0135, 0.0123, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 04:47:35,955 INFO [finetune.py:976] (1/7) Epoch 24, batch 850, loss[loss=0.1205, simple_loss=0.1956, pruned_loss=0.02273, over 3995.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2459, pruned_loss=0.05116, over 943681.39 frames. ], batch size: 17, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:47:54,081 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9737, 2.7005, 2.2823, 1.4363, 2.5599, 2.4301, 2.2115, 2.6012], + device='cuda:1'), covar=tensor([0.0670, 0.0622, 0.1368, 0.1639, 0.1066, 0.1656, 0.1796, 0.0686], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0192, 0.0200, 0.0180, 0.0209, 0.0209, 0.0224, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:48:18,613 INFO [finetune.py:976] (1/7) Epoch 24, batch 900, loss[loss=0.1664, simple_loss=0.2384, pruned_loss=0.04716, over 4908.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2436, pruned_loss=0.05048, over 947025.83 frames. ], batch size: 37, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:48:32,509 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5302, 2.3519, 1.9307, 0.9541, 2.0597, 2.0071, 1.8591, 2.1912], + device='cuda:1'), covar=tensor([0.0855, 0.0836, 0.1740, 0.1970, 0.1389, 0.2265, 0.2279, 0.0907], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0191, 0.0200, 0.0180, 0.0209, 0.0208, 0.0224, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:48:33,114 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132660.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:48:35,407 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.013e+02 1.417e+02 1.718e+02 2.002e+02 3.598e+02, threshold=3.436e+02, percent-clipped=1.0 +2023-03-27 04:48:52,530 INFO [finetune.py:976] (1/7) Epoch 24, batch 950, loss[loss=0.1834, simple_loss=0.2552, pruned_loss=0.05583, over 4931.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2411, pruned_loss=0.04958, over 949147.48 frames. ], batch size: 33, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:48:52,611 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=132687.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:49:07,082 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=132710.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 04:49:13,782 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-03-27 04:49:14,308 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132721.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:49:15,510 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8856, 1.6024, 2.3874, 1.4887, 1.9990, 2.3234, 1.5237, 2.3522], + device='cuda:1'), covar=tensor([0.1433, 0.2178, 0.1389, 0.2025, 0.0909, 0.1365, 0.2874, 0.0866], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0207, 0.0191, 0.0191, 0.0173, 0.0214, 0.0216, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:49:26,440 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=132735.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:49:28,064 INFO [finetune.py:976] (1/7) Epoch 24, batch 1000, loss[loss=0.2329, simple_loss=0.3005, pruned_loss=0.08265, over 4819.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2438, pruned_loss=0.05069, over 951419.80 frames. ], batch size: 39, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:49:38,679 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132746.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:49:50,803 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=132758.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:49:54,880 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.132e+02 1.613e+02 1.803e+02 2.355e+02 4.590e+02, threshold=3.605e+02, percent-clipped=3.0 +2023-03-27 04:50:17,563 INFO [finetune.py:976] (1/7) Epoch 24, batch 1050, loss[loss=0.1945, simple_loss=0.2584, pruned_loss=0.06528, over 4749.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2449, pruned_loss=0.05069, over 950850.87 frames. ], batch size: 26, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:50:31,466 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132807.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:50:35,094 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2457, 2.2528, 1.9308, 2.2594, 2.1284, 2.1581, 2.1498, 2.9686], + device='cuda:1'), covar=tensor([0.3663, 0.4437, 0.3181, 0.4117, 0.4176, 0.2331, 0.4121, 0.1613], + device='cuda:1'), in_proj_covar=tensor([0.0292, 0.0264, 0.0236, 0.0278, 0.0259, 0.0228, 0.0255, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:50:51,430 INFO [finetune.py:976] (1/7) Epoch 24, batch 1100, loss[loss=0.189, simple_loss=0.2748, pruned_loss=0.0516, over 4816.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2467, pruned_loss=0.05108, over 953404.03 frames. ], batch size: 39, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:51:08,747 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.005e+02 1.623e+02 1.902e+02 2.304e+02 4.937e+02, threshold=3.804e+02, percent-clipped=2.0 +2023-03-27 04:51:24,184 INFO [finetune.py:976] (1/7) Epoch 24, batch 1150, loss[loss=0.1676, simple_loss=0.2476, pruned_loss=0.04377, over 4891.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2481, pruned_loss=0.05181, over 954882.66 frames. ], batch size: 43, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:51:57,324 INFO [finetune.py:976] (1/7) Epoch 24, batch 1200, loss[loss=0.173, simple_loss=0.2425, pruned_loss=0.05173, over 4905.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2463, pruned_loss=0.0511, over 956032.43 frames. ], batch size: 37, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:52:24,716 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.055e+01 1.470e+02 1.716e+02 2.148e+02 3.548e+02, threshold=3.432e+02, percent-clipped=0.0 +2023-03-27 04:52:32,085 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1774, 1.8011, 2.2214, 2.2204, 1.9361, 1.9497, 2.1661, 2.1036], + device='cuda:1'), covar=tensor([0.4095, 0.4540, 0.3184, 0.3955, 0.5218, 0.4335, 0.4819, 0.3126], + device='cuda:1'), in_proj_covar=tensor([0.0260, 0.0244, 0.0263, 0.0288, 0.0287, 0.0263, 0.0296, 0.0247], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:52:36,778 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9223, 4.6877, 4.5212, 2.8240, 4.7216, 3.6262, 0.9443, 3.1186], + device='cuda:1'), covar=tensor([0.2086, 0.1398, 0.1275, 0.2457, 0.0786, 0.0861, 0.4315, 0.1478], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0179, 0.0162, 0.0129, 0.0161, 0.0124, 0.0149, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 04:52:40,280 INFO [finetune.py:976] (1/7) Epoch 24, batch 1250, loss[loss=0.2103, simple_loss=0.2767, pruned_loss=0.0719, over 4263.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2442, pruned_loss=0.05093, over 956582.73 frames. ], batch size: 65, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:52:49,683 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 04:52:59,659 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=133016.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:53:15,461 INFO [finetune.py:976] (1/7) Epoch 24, batch 1300, loss[loss=0.2088, simple_loss=0.2658, pruned_loss=0.07593, over 4233.00 frames. ], tot_loss[loss=0.169, simple_loss=0.2398, pruned_loss=0.04911, over 955128.53 frames. ], batch size: 65, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:53:23,662 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.62 vs. limit=5.0 +2023-03-27 04:53:42,213 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.345e+01 1.496e+02 1.852e+02 2.141e+02 4.041e+02, threshold=3.705e+02, percent-clipped=2.0 +2023-03-27 04:53:47,898 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4137, 2.4722, 2.5109, 1.6934, 2.1726, 2.7458, 2.6868, 2.1302], + device='cuda:1'), covar=tensor([0.0643, 0.0665, 0.0681, 0.0894, 0.1340, 0.0664, 0.0559, 0.1130], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0135, 0.0138, 0.0119, 0.0126, 0.0137, 0.0137, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:53:57,207 INFO [finetune.py:976] (1/7) Epoch 24, batch 1350, loss[loss=0.169, simple_loss=0.2495, pruned_loss=0.04424, over 4827.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2402, pruned_loss=0.0497, over 953262.62 frames. ], batch size: 33, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:54:07,460 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=133102.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:54:16,244 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6470, 1.5610, 1.1149, 0.3230, 1.3515, 1.4890, 1.4263, 1.4760], + device='cuda:1'), covar=tensor([0.0877, 0.0864, 0.1386, 0.1870, 0.1319, 0.2038, 0.2478, 0.0849], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0191, 0.0200, 0.0182, 0.0210, 0.0209, 0.0224, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:54:31,058 INFO [finetune.py:976] (1/7) Epoch 24, batch 1400, loss[loss=0.3044, simple_loss=0.3474, pruned_loss=0.1307, over 4262.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2428, pruned_loss=0.05065, over 952128.87 frames. ], batch size: 65, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:54:33,567 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5392, 1.0801, 0.8841, 1.3341, 1.9676, 1.0869, 1.2290, 1.2170], + device='cuda:1'), covar=tensor([0.1695, 0.2543, 0.2067, 0.1435, 0.2145, 0.2201, 0.1839, 0.2444], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0119, 0.0094, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 04:54:59,477 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.355e+01 1.624e+02 1.914e+02 2.315e+02 3.947e+02, threshold=3.828e+02, percent-clipped=1.0 +2023-03-27 04:55:19,623 INFO [finetune.py:976] (1/7) Epoch 24, batch 1450, loss[loss=0.1548, simple_loss=0.2215, pruned_loss=0.04403, over 4708.00 frames. ], tot_loss[loss=0.1746, simple_loss=0.2461, pruned_loss=0.0515, over 954398.34 frames. ], batch size: 23, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:55:56,690 INFO [finetune.py:976] (1/7) Epoch 24, batch 1500, loss[loss=0.1866, simple_loss=0.2478, pruned_loss=0.06264, over 4134.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2468, pruned_loss=0.05178, over 952928.91 frames. ], batch size: 18, lr: 3.06e-03, grad_scale: 16.0 +2023-03-27 04:56:15,023 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.176e+02 1.579e+02 1.864e+02 2.355e+02 5.095e+02, threshold=3.727e+02, percent-clipped=2.0 +2023-03-27 04:56:30,467 INFO [finetune.py:976] (1/7) Epoch 24, batch 1550, loss[loss=0.1742, simple_loss=0.2457, pruned_loss=0.05139, over 4836.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2473, pruned_loss=0.05179, over 953783.01 frames. ], batch size: 30, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 04:56:50,671 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=133316.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:57:04,264 INFO [finetune.py:976] (1/7) Epoch 24, batch 1600, loss[loss=0.1626, simple_loss=0.2337, pruned_loss=0.04576, over 4912.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2465, pruned_loss=0.05209, over 953656.42 frames. ], batch size: 37, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 04:57:08,020 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0679, 2.0304, 2.2700, 1.5138, 2.0796, 2.2330, 2.2534, 1.7828], + device='cuda:1'), covar=tensor([0.0562, 0.0635, 0.0580, 0.0798, 0.0805, 0.0614, 0.0516, 0.1059], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0136, 0.0139, 0.0119, 0.0127, 0.0138, 0.0138, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:57:12,168 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8412, 3.9474, 3.7670, 1.9299, 4.0537, 3.0958, 1.0073, 2.8037], + device='cuda:1'), covar=tensor([0.2190, 0.2065, 0.1460, 0.3362, 0.1065, 0.1047, 0.4385, 0.1587], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0179, 0.0162, 0.0130, 0.0162, 0.0124, 0.0149, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 04:57:28,469 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.451e+02 1.796e+02 2.043e+02 3.402e+02, threshold=3.593e+02, percent-clipped=0.0 +2023-03-27 04:57:28,563 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=133364.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:57:38,829 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=133374.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:57:40,052 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=133376.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:57:46,648 INFO [finetune.py:976] (1/7) Epoch 24, batch 1650, loss[loss=0.1678, simple_loss=0.2348, pruned_loss=0.05043, over 4824.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2442, pruned_loss=0.05131, over 954755.09 frames. ], batch size: 39, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 04:57:55,361 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0963, 1.9760, 1.7207, 1.7018, 1.8336, 1.8089, 1.8854, 2.5635], + device='cuda:1'), covar=tensor([0.3455, 0.3574, 0.2938, 0.3274, 0.3645, 0.2217, 0.3258, 0.1562], + device='cuda:1'), in_proj_covar=tensor([0.0291, 0.0264, 0.0235, 0.0278, 0.0258, 0.0228, 0.0255, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:57:56,869 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=133402.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:58:05,759 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1934, 1.8478, 2.8848, 1.6022, 2.3007, 2.4997, 1.7118, 2.5907], + device='cuda:1'), covar=tensor([0.1389, 0.2245, 0.1078, 0.2048, 0.1058, 0.1558, 0.2850, 0.1024], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0208, 0.0192, 0.0190, 0.0173, 0.0214, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:58:19,417 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=133435.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 04:58:20,504 INFO [finetune.py:976] (1/7) Epoch 24, batch 1700, loss[loss=0.203, simple_loss=0.2776, pruned_loss=0.06418, over 4724.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2415, pruned_loss=0.05072, over 955699.87 frames. ], batch size: 59, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 04:58:20,619 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=133437.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:58:30,781 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-27 04:58:31,232 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=133450.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:58:41,297 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4342, 1.4883, 1.5121, 0.8514, 1.6103, 1.7753, 1.7852, 1.3822], + device='cuda:1'), covar=tensor([0.1088, 0.0623, 0.0617, 0.0600, 0.0488, 0.0612, 0.0363, 0.0740], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0127, 0.0122, 0.0131, 0.0130, 0.0141, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.9097e-05, 1.0685e-04, 9.0800e-05, 8.5814e-05, 9.2015e-05, 9.2501e-05, + 1.0079e-04, 1.0593e-04], device='cuda:1') +2023-03-27 04:58:48,614 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.499e+01 1.457e+02 1.770e+02 2.219e+02 3.253e+02, threshold=3.541e+02, percent-clipped=0.0 +2023-03-27 04:58:55,863 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=133474.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:59:04,124 INFO [finetune.py:976] (1/7) Epoch 24, batch 1750, loss[loss=0.1428, simple_loss=0.2128, pruned_loss=0.03642, over 4906.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2448, pruned_loss=0.05197, over 955442.19 frames. ], batch size: 36, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 04:59:27,344 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7846, 1.7617, 1.5477, 1.9399, 2.5034, 1.8787, 1.7259, 1.4311], + device='cuda:1'), covar=tensor([0.2321, 0.2014, 0.1928, 0.1659, 0.1546, 0.1307, 0.2274, 0.1981], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0214, 0.0196, 0.0245, 0.0191, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 04:59:36,809 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=133535.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 04:59:37,911 INFO [finetune.py:976] (1/7) Epoch 24, batch 1800, loss[loss=0.1391, simple_loss=0.21, pruned_loss=0.03408, over 4723.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2475, pruned_loss=0.05207, over 955767.30 frames. ], batch size: 23, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 04:59:47,571 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5041, 1.3613, 1.8192, 1.7666, 1.5105, 3.1947, 1.3824, 1.4689], + device='cuda:1'), covar=tensor([0.1021, 0.1874, 0.1085, 0.0957, 0.1680, 0.0272, 0.1534, 0.1844], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0092, 0.0081, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 04:59:57,746 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.575e+02 1.839e+02 2.282e+02 3.463e+02, threshold=3.677e+02, percent-clipped=0.0 +2023-03-27 05:00:23,457 INFO [finetune.py:976] (1/7) Epoch 24, batch 1850, loss[loss=0.1595, simple_loss=0.2394, pruned_loss=0.03982, over 4746.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2486, pruned_loss=0.05286, over 955171.66 frames. ], batch size: 28, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 05:01:04,053 INFO [finetune.py:976] (1/7) Epoch 24, batch 1900, loss[loss=0.167, simple_loss=0.2509, pruned_loss=0.04158, over 4784.00 frames. ], tot_loss[loss=0.1772, simple_loss=0.2491, pruned_loss=0.0527, over 954893.54 frames. ], batch size: 51, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 05:01:14,194 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=133652.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:01:21,802 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.050e+02 1.540e+02 1.881e+02 2.277e+02 3.366e+02, threshold=3.762e+02, percent-clipped=0.0 +2023-03-27 05:01:37,658 INFO [finetune.py:976] (1/7) Epoch 24, batch 1950, loss[loss=0.1507, simple_loss=0.2353, pruned_loss=0.03309, over 4791.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2471, pruned_loss=0.05151, over 954922.46 frames. ], batch size: 29, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 05:01:55,006 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=133713.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:02:06,299 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=133730.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 05:02:07,529 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=133732.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:02:10,916 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7623, 1.7799, 2.2985, 1.8889, 2.0777, 4.3169, 1.8850, 1.9445], + device='cuda:1'), covar=tensor([0.0896, 0.1653, 0.1088, 0.0947, 0.1371, 0.0173, 0.1304, 0.1617], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0074, 0.0077, 0.0092, 0.0081, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 05:02:11,408 INFO [finetune.py:976] (1/7) Epoch 24, batch 2000, loss[loss=0.1503, simple_loss=0.2154, pruned_loss=0.04265, over 4789.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2424, pruned_loss=0.04986, over 953010.32 frames. ], batch size: 29, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 05:02:28,712 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.381e+01 1.373e+02 1.735e+02 2.243e+02 3.912e+02, threshold=3.469e+02, percent-clipped=2.0 +2023-03-27 05:02:31,824 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9364, 2.0130, 1.6701, 2.1645, 2.5087, 2.1270, 1.8504, 1.5342], + device='cuda:1'), covar=tensor([0.2196, 0.1735, 0.1886, 0.1544, 0.1604, 0.1118, 0.2122, 0.2050], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0209, 0.0213, 0.0195, 0.0243, 0.0190, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:02:48,895 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3987, 2.2674, 2.8500, 1.5746, 2.3142, 2.6280, 2.0128, 2.8500], + device='cuda:1'), covar=tensor([0.1407, 0.1925, 0.1749, 0.2495, 0.1012, 0.1629, 0.2835, 0.0918], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0210, 0.0194, 0.0193, 0.0175, 0.0216, 0.0219, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:02:54,158 INFO [finetune.py:976] (1/7) Epoch 24, batch 2050, loss[loss=0.1798, simple_loss=0.2562, pruned_loss=0.05166, over 4928.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.2396, pruned_loss=0.04902, over 954094.33 frames. ], batch size: 38, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 05:03:01,178 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7408, 2.7280, 2.4130, 2.9660, 2.6195, 2.7113, 2.6392, 3.6455], + device='cuda:1'), covar=tensor([0.3109, 0.4098, 0.2914, 0.3522, 0.3569, 0.2131, 0.3650, 0.1237], + device='cuda:1'), in_proj_covar=tensor([0.0292, 0.0264, 0.0235, 0.0277, 0.0259, 0.0228, 0.0255, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:03:06,223 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-27 05:03:23,215 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=133830.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:03:26,778 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6690, 1.4733, 2.0921, 3.2893, 2.2550, 2.4210, 1.2096, 2.7139], + device='cuda:1'), covar=tensor([0.1549, 0.1383, 0.1241, 0.0611, 0.0761, 0.1638, 0.1547, 0.0452], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0134, 0.0164, 0.0101, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 05:03:27,929 INFO [finetune.py:976] (1/7) Epoch 24, batch 2100, loss[loss=0.2641, simple_loss=0.3157, pruned_loss=0.1063, over 4185.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2397, pruned_loss=0.04971, over 955030.37 frames. ], batch size: 65, lr: 3.06e-03, grad_scale: 32.0 +2023-03-27 05:03:30,879 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6771, 3.8367, 3.6563, 1.8604, 3.9227, 2.9611, 0.9129, 2.7033], + device='cuda:1'), covar=tensor([0.2460, 0.2172, 0.1477, 0.3401, 0.0958, 0.1010, 0.4484, 0.1481], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0161, 0.0129, 0.0160, 0.0123, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 05:03:47,589 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.827e+01 1.541e+02 1.860e+02 2.293e+02 6.118e+02, threshold=3.720e+02, percent-clipped=3.0 +2023-03-27 05:04:11,227 INFO [finetune.py:976] (1/7) Epoch 24, batch 2150, loss[loss=0.1847, simple_loss=0.2623, pruned_loss=0.05356, over 4853.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2434, pruned_loss=0.05108, over 955938.07 frames. ], batch size: 49, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:04:25,505 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3299, 2.9650, 2.8480, 1.2925, 3.0929, 2.2657, 0.5754, 2.0162], + device='cuda:1'), covar=tensor([0.2058, 0.2170, 0.1631, 0.3493, 0.1312, 0.1194, 0.4182, 0.1558], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0161, 0.0129, 0.0160, 0.0123, 0.0148, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 05:04:44,943 INFO [finetune.py:976] (1/7) Epoch 24, batch 2200, loss[loss=0.222, simple_loss=0.2804, pruned_loss=0.0818, over 4888.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2447, pruned_loss=0.05072, over 954502.45 frames. ], batch size: 32, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:05:02,713 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.027e+02 1.464e+02 1.824e+02 2.239e+02 3.694e+02, threshold=3.648e+02, percent-clipped=0.0 +2023-03-27 05:05:24,731 INFO [finetune.py:976] (1/7) Epoch 24, batch 2250, loss[loss=0.1752, simple_loss=0.2468, pruned_loss=0.05185, over 4894.00 frames. ], tot_loss[loss=0.1756, simple_loss=0.2474, pruned_loss=0.05188, over 954287.68 frames. ], batch size: 32, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:05:24,805 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6977, 3.9374, 3.6938, 2.0125, 4.0338, 2.9945, 0.8018, 2.8009], + device='cuda:1'), covar=tensor([0.2301, 0.1723, 0.1469, 0.3081, 0.0901, 0.0921, 0.4408, 0.1346], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0178, 0.0161, 0.0129, 0.0160, 0.0123, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 05:05:24,850 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7793, 1.7586, 1.6538, 1.7531, 1.3553, 3.5295, 1.4369, 1.8101], + device='cuda:1'), covar=tensor([0.3141, 0.2350, 0.2079, 0.2347, 0.1580, 0.0207, 0.2524, 0.1260], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0123, 0.0113, 0.0097, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 05:05:34,955 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=133997.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:05:49,821 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=134008.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:06:07,542 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134030.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 05:06:09,759 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134032.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:06:12,678 INFO [finetune.py:976] (1/7) Epoch 24, batch 2300, loss[loss=0.1767, simple_loss=0.2451, pruned_loss=0.0542, over 4904.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2475, pruned_loss=0.05134, over 953024.88 frames. ], batch size: 43, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:06:27,047 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=134058.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:06:31,059 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.533e+02 1.723e+02 2.089e+02 4.293e+02, threshold=3.445e+02, percent-clipped=2.0 +2023-03-27 05:06:40,130 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=134078.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 05:06:41,359 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=134080.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:06:46,524 INFO [finetune.py:976] (1/7) Epoch 24, batch 2350, loss[loss=0.1687, simple_loss=0.233, pruned_loss=0.05222, over 4897.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2462, pruned_loss=0.0512, over 954302.69 frames. ], batch size: 36, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:07:14,953 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134130.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:07:19,176 INFO [finetune.py:976] (1/7) Epoch 24, batch 2400, loss[loss=0.1524, simple_loss=0.2279, pruned_loss=0.03849, over 4923.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2438, pruned_loss=0.05095, over 954390.90 frames. ], batch size: 46, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:07:38,327 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.553e+01 1.434e+02 1.789e+02 2.166e+02 3.942e+02, threshold=3.577e+02, percent-clipped=1.0 +2023-03-27 05:07:47,957 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=134178.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:07:55,475 INFO [finetune.py:976] (1/7) Epoch 24, batch 2450, loss[loss=0.1936, simple_loss=0.2622, pruned_loss=0.06253, over 4794.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2403, pruned_loss=0.04951, over 954649.58 frames. ], batch size: 51, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:08:01,886 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.34 vs. limit=5.0 +2023-03-27 05:08:30,424 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9479, 4.2239, 4.5023, 4.7273, 4.6807, 4.3518, 5.0200, 1.4676], + device='cuda:1'), covar=tensor([0.0738, 0.0953, 0.0918, 0.0936, 0.1168, 0.1964, 0.0620, 0.6441], + device='cuda:1'), in_proj_covar=tensor([0.0347, 0.0247, 0.0280, 0.0293, 0.0335, 0.0286, 0.0307, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:08:36,910 INFO [finetune.py:976] (1/7) Epoch 24, batch 2500, loss[loss=0.1715, simple_loss=0.2514, pruned_loss=0.04581, over 4859.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2416, pruned_loss=0.05007, over 955395.60 frames. ], batch size: 44, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:08:55,715 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.396e+01 1.499e+02 1.865e+02 2.171e+02 5.575e+02, threshold=3.730e+02, percent-clipped=1.0 +2023-03-27 05:09:20,384 INFO [finetune.py:976] (1/7) Epoch 24, batch 2550, loss[loss=0.1512, simple_loss=0.2386, pruned_loss=0.0319, over 4755.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2454, pruned_loss=0.05116, over 953912.80 frames. ], batch size: 54, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:09:32,213 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=134304.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:09:34,652 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134308.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:09:35,283 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=134309.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:09:52,962 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4716, 1.4322, 2.0828, 2.8971, 1.9651, 2.1391, 1.3131, 2.4821], + device='cuda:1'), covar=tensor([0.1696, 0.1369, 0.1043, 0.0533, 0.0818, 0.1405, 0.1445, 0.0457], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0162, 0.0100, 0.0136, 0.0123, 0.0099], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 05:09:53,623 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1019, 2.0235, 1.6962, 1.9246, 1.9104, 1.9592, 1.9941, 2.7174], + device='cuda:1'), covar=tensor([0.3539, 0.4133, 0.3185, 0.3846, 0.4044, 0.2342, 0.3842, 0.1591], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0262, 0.0233, 0.0275, 0.0256, 0.0226, 0.0253, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:09:54,099 INFO [finetune.py:976] (1/7) Epoch 24, batch 2600, loss[loss=0.2003, simple_loss=0.2724, pruned_loss=0.06407, over 4810.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.246, pruned_loss=0.05113, over 953108.01 frames. ], batch size: 40, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:10:01,383 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1333, 2.0467, 1.6863, 2.0490, 1.9379, 1.9645, 1.9983, 2.7481], + device='cuda:1'), covar=tensor([0.3307, 0.3927, 0.3051, 0.3612, 0.4018, 0.2144, 0.3301, 0.1447], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0262, 0.0233, 0.0276, 0.0256, 0.0226, 0.0254, 0.0236], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:10:04,822 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=134353.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:10:07,200 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=134356.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:10:12,062 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.001e+02 1.471e+02 1.806e+02 2.184e+02 4.519e+02, threshold=3.612e+02, percent-clipped=2.0 +2023-03-27 05:10:13,313 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=134365.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:10:16,846 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=134370.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:10:29,856 INFO [finetune.py:976] (1/7) Epoch 24, batch 2650, loss[loss=0.1479, simple_loss=0.2313, pruned_loss=0.03221, over 4771.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.247, pruned_loss=0.05095, over 953586.86 frames. ], batch size: 26, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:10:38,658 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.55 vs. limit=2.0 +2023-03-27 05:11:21,024 INFO [finetune.py:976] (1/7) Epoch 24, batch 2700, loss[loss=0.1792, simple_loss=0.2423, pruned_loss=0.05807, over 4773.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2451, pruned_loss=0.05007, over 952358.03 frames. ], batch size: 26, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:11:39,163 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.916e+01 1.416e+02 1.758e+02 2.159e+02 3.599e+02, threshold=3.516e+02, percent-clipped=0.0 +2023-03-27 05:11:54,599 INFO [finetune.py:976] (1/7) Epoch 24, batch 2750, loss[loss=0.1529, simple_loss=0.2344, pruned_loss=0.03572, over 4910.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2435, pruned_loss=0.04994, over 951470.84 frames. ], batch size: 36, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:12:01,828 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4832, 1.5754, 1.6872, 0.8861, 1.7026, 1.9232, 1.8293, 1.5145], + device='cuda:1'), covar=tensor([0.0922, 0.0656, 0.0499, 0.0535, 0.0497, 0.0693, 0.0383, 0.0602], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0125, 0.0121, 0.0130, 0.0128, 0.0140, 0.0147], + device='cuda:1'), out_proj_covar=tensor([8.8195e-05, 1.0571e-04, 8.9538e-05, 8.4912e-05, 9.1069e-05, 9.1291e-05, + 1.0003e-04, 1.0497e-04], device='cuda:1') +2023-03-27 05:12:13,129 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-27 05:12:27,873 INFO [finetune.py:976] (1/7) Epoch 24, batch 2800, loss[loss=0.1852, simple_loss=0.2468, pruned_loss=0.06178, over 4912.00 frames. ], tot_loss[loss=0.1693, simple_loss=0.2406, pruned_loss=0.04903, over 952113.49 frames. ], batch size: 32, lr: 3.05e-03, grad_scale: 32.0 +2023-03-27 05:12:46,115 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.301e+01 1.456e+02 1.823e+02 2.196e+02 4.309e+02, threshold=3.645e+02, percent-clipped=3.0 +2023-03-27 05:12:46,242 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2557, 2.1421, 1.7425, 0.9164, 1.8911, 1.8078, 1.6777, 1.9589], + device='cuda:1'), covar=tensor([0.0928, 0.0732, 0.1503, 0.1859, 0.1370, 0.2128, 0.2116, 0.0879], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0190, 0.0198, 0.0180, 0.0209, 0.0208, 0.0223, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:13:01,598 INFO [finetune.py:976] (1/7) Epoch 24, batch 2850, loss[loss=0.1712, simple_loss=0.2371, pruned_loss=0.05267, over 4829.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.24, pruned_loss=0.04963, over 948934.71 frames. ], batch size: 33, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:13:03,550 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5704, 1.5336, 1.6961, 0.8985, 1.8509, 2.0160, 1.9093, 1.5122], + device='cuda:1'), covar=tensor([0.1113, 0.0875, 0.0562, 0.0684, 0.0470, 0.0768, 0.0413, 0.0873], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0126, 0.0121, 0.0130, 0.0129, 0.0140, 0.0147], + device='cuda:1'), out_proj_covar=tensor([8.8536e-05, 1.0619e-04, 8.9814e-05, 8.5219e-05, 9.1282e-05, 9.1586e-05, + 1.0027e-04, 1.0542e-04], device='cuda:1') +2023-03-27 05:13:45,377 INFO [finetune.py:976] (1/7) Epoch 24, batch 2900, loss[loss=0.1656, simple_loss=0.2431, pruned_loss=0.04409, over 4911.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2441, pruned_loss=0.05086, over 951350.15 frames. ], batch size: 37, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:13:50,543 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-27 05:13:56,299 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134653.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:14:00,539 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=134660.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:14:03,944 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.005e+02 1.559e+02 1.783e+02 2.063e+02 3.902e+02, threshold=3.566e+02, percent-clipped=1.0 +2023-03-27 05:14:04,026 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=134665.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:14:20,897 INFO [finetune.py:976] (1/7) Epoch 24, batch 2950, loss[loss=0.1598, simple_loss=0.2411, pruned_loss=0.03918, over 4854.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.246, pruned_loss=0.0509, over 952597.47 frames. ], batch size: 44, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:14:37,442 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=134701.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:14:52,790 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3921, 2.3477, 1.8872, 2.5076, 2.3315, 2.0122, 2.8124, 2.4430], + device='cuda:1'), covar=tensor([0.1283, 0.2358, 0.2988, 0.2762, 0.2499, 0.1697, 0.3185, 0.1700], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0190, 0.0236, 0.0255, 0.0251, 0.0207, 0.0214, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:15:02,624 INFO [finetune.py:976] (1/7) Epoch 24, batch 3000, loss[loss=0.1453, simple_loss=0.2142, pruned_loss=0.03823, over 4766.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2469, pruned_loss=0.05148, over 951653.62 frames. ], batch size: 26, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:15:02,624 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 05:15:13,333 INFO [finetune.py:1010] (1/7) Epoch 24, validation: loss=0.1561, simple_loss=0.2251, pruned_loss=0.0436, over 2265189.00 frames. +2023-03-27 05:15:13,334 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 05:15:15,764 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4667, 1.4679, 2.1100, 3.1688, 2.0309, 2.2084, 1.2240, 2.6330], + device='cuda:1'), covar=tensor([0.1838, 0.1460, 0.1196, 0.0609, 0.0851, 0.1437, 0.1643, 0.0520], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0115, 0.0133, 0.0163, 0.0101, 0.0136, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 05:15:31,258 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.188e+02 1.527e+02 1.858e+02 2.241e+02 5.364e+02, threshold=3.716e+02, percent-clipped=3.0 +2023-03-27 05:15:36,148 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3377, 2.2370, 1.7491, 2.2779, 2.2084, 1.9557, 2.5063, 2.3629], + device='cuda:1'), covar=tensor([0.1306, 0.2158, 0.3082, 0.2510, 0.2759, 0.1697, 0.3476, 0.1679], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0190, 0.0236, 0.0255, 0.0251, 0.0207, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:15:48,110 INFO [finetune.py:976] (1/7) Epoch 24, batch 3050, loss[loss=0.1813, simple_loss=0.255, pruned_loss=0.05378, over 4779.00 frames. ], tot_loss[loss=0.1765, simple_loss=0.2488, pruned_loss=0.05209, over 953795.86 frames. ], batch size: 51, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:15:58,189 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.44 vs. limit=5.0 +2023-03-27 05:16:39,365 INFO [finetune.py:976] (1/7) Epoch 24, batch 3100, loss[loss=0.1655, simple_loss=0.2368, pruned_loss=0.04711, over 4732.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.247, pruned_loss=0.05102, over 955125.13 frames. ], batch size: 59, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:16:56,665 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=134862.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:16:58,376 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.835e+01 1.385e+02 1.668e+02 2.115e+02 5.080e+02, threshold=3.336e+02, percent-clipped=1.0 +2023-03-27 05:17:12,755 INFO [finetune.py:976] (1/7) Epoch 24, batch 3150, loss[loss=0.1659, simple_loss=0.2316, pruned_loss=0.05009, over 4920.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2446, pruned_loss=0.05062, over 953058.78 frames. ], batch size: 37, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:17:30,524 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.7257, 1.4020, 1.5176, 0.7888, 1.7208, 1.7243, 1.7114, 1.4000], + device='cuda:1'), covar=tensor([0.0919, 0.0880, 0.0578, 0.0600, 0.0488, 0.0741, 0.0443, 0.0775], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0125, 0.0121, 0.0130, 0.0129, 0.0140, 0.0147], + device='cuda:1'), out_proj_covar=tensor([8.8256e-05, 1.0569e-04, 8.9446e-05, 8.4897e-05, 9.1159e-05, 9.1530e-05, + 9.9788e-05, 1.0510e-04], device='cuda:1') +2023-03-27 05:17:32,347 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2532, 1.3515, 1.5958, 1.0466, 1.3501, 1.4744, 1.3270, 1.6056], + device='cuda:1'), covar=tensor([0.1285, 0.2086, 0.1310, 0.1517, 0.0963, 0.1299, 0.2892, 0.0910], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0208, 0.0192, 0.0190, 0.0173, 0.0214, 0.0216, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:17:37,238 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=134923.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:17:46,591 INFO [finetune.py:976] (1/7) Epoch 24, batch 3200, loss[loss=0.1653, simple_loss=0.226, pruned_loss=0.05226, over 4911.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2417, pruned_loss=0.04973, over 953838.33 frames. ], batch size: 43, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:18:03,021 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134960.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:18:04,197 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2830, 2.9450, 3.0583, 3.1884, 3.0887, 2.8891, 3.3578, 0.9561], + device='cuda:1'), covar=tensor([0.1299, 0.1016, 0.1272, 0.1450, 0.1766, 0.2068, 0.1118, 0.5680], + device='cuda:1'), in_proj_covar=tensor([0.0346, 0.0246, 0.0280, 0.0292, 0.0335, 0.0286, 0.0306, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:18:05,932 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.665e+01 1.513e+02 1.793e+02 2.252e+02 3.579e+02, threshold=3.586e+02, percent-clipped=1.0 +2023-03-27 05:18:06,036 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134965.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:18:22,480 INFO [finetune.py:976] (1/7) Epoch 24, batch 3250, loss[loss=0.1505, simple_loss=0.215, pruned_loss=0.04303, over 3919.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2422, pruned_loss=0.05032, over 952171.72 frames. ], batch size: 17, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:18:45,537 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=135008.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:18:46,188 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9853, 1.9878, 2.1622, 1.4041, 1.9072, 2.1382, 2.2381, 1.5945], + device='cuda:1'), covar=tensor([0.0732, 0.0725, 0.0728, 0.0977, 0.0846, 0.0750, 0.0614, 0.1326], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0136, 0.0139, 0.0119, 0.0127, 0.0138, 0.0138, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:18:48,601 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=135013.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:18:50,455 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4801, 1.3123, 1.3699, 1.3149, 0.7591, 2.3629, 0.7640, 1.1768], + device='cuda:1'), covar=tensor([0.3437, 0.2696, 0.2227, 0.2581, 0.2139, 0.0362, 0.2945, 0.1498], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0124, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 05:19:03,035 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-27 05:19:04,066 INFO [finetune.py:976] (1/7) Epoch 24, batch 3300, loss[loss=0.1789, simple_loss=0.2571, pruned_loss=0.05033, over 4928.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2457, pruned_loss=0.05092, over 952510.67 frames. ], batch size: 38, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:19:23,522 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.113e+02 1.563e+02 1.899e+02 2.275e+02 5.700e+02, threshold=3.799e+02, percent-clipped=2.0 +2023-03-27 05:19:44,199 INFO [finetune.py:976] (1/7) Epoch 24, batch 3350, loss[loss=0.1876, simple_loss=0.2534, pruned_loss=0.06096, over 4281.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2481, pruned_loss=0.0513, over 954502.63 frames. ], batch size: 65, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:20:21,443 INFO [finetune.py:976] (1/7) Epoch 24, batch 3400, loss[loss=0.1567, simple_loss=0.2251, pruned_loss=0.04412, over 4746.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2495, pruned_loss=0.05213, over 956198.60 frames. ], batch size: 27, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:20:38,234 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1282, 2.0572, 1.7598, 2.0219, 1.9126, 1.9488, 1.9446, 2.7281], + device='cuda:1'), covar=tensor([0.3450, 0.3974, 0.3063, 0.3602, 0.3965, 0.2348, 0.3604, 0.1583], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0262, 0.0234, 0.0275, 0.0256, 0.0226, 0.0253, 0.0235], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:20:40,369 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.077e+02 1.584e+02 1.828e+02 2.150e+02 3.792e+02, threshold=3.656e+02, percent-clipped=0.0 +2023-03-27 05:20:44,637 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=135171.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:20:53,791 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8244, 1.7086, 1.5519, 1.8383, 2.3621, 1.8833, 1.6227, 1.4596], + device='cuda:1'), covar=tensor([0.2117, 0.1990, 0.2011, 0.1729, 0.1634, 0.1262, 0.2419, 0.2096], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0213, 0.0216, 0.0199, 0.0246, 0.0192, 0.0218, 0.0207], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:20:54,274 INFO [finetune.py:976] (1/7) Epoch 24, batch 3450, loss[loss=0.1684, simple_loss=0.2436, pruned_loss=0.04664, over 4834.00 frames. ], tot_loss[loss=0.1762, simple_loss=0.2495, pruned_loss=0.05146, over 956556.21 frames. ], batch size: 47, lr: 3.05e-03, grad_scale: 16.0 +2023-03-27 05:21:27,780 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=135218.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:21:40,716 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=135232.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:21:40,796 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-27 05:21:47,100 INFO [finetune.py:976] (1/7) Epoch 24, batch 3500, loss[loss=0.1788, simple_loss=0.2405, pruned_loss=0.05852, over 4913.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2463, pruned_loss=0.05027, over 957264.58 frames. ], batch size: 43, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:21:57,983 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8978, 1.7908, 1.6106, 1.9186, 2.4847, 2.0800, 1.6924, 1.5038], + device='cuda:1'), covar=tensor([0.2054, 0.1999, 0.1818, 0.1692, 0.1493, 0.1085, 0.2196, 0.1835], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0212, 0.0215, 0.0198, 0.0245, 0.0191, 0.0217, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:22:06,082 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.678e+01 1.506e+02 1.714e+02 2.011e+02 3.544e+02, threshold=3.428e+02, percent-clipped=0.0 +2023-03-27 05:22:20,455 INFO [finetune.py:976] (1/7) Epoch 24, batch 3550, loss[loss=0.1351, simple_loss=0.2131, pruned_loss=0.02858, over 4790.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.243, pruned_loss=0.0492, over 955285.07 frames. ], batch size: 29, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:22:54,386 INFO [finetune.py:976] (1/7) Epoch 24, batch 3600, loss[loss=0.1501, simple_loss=0.2315, pruned_loss=0.03434, over 4751.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2403, pruned_loss=0.0478, over 956452.10 frames. ], batch size: 27, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:23:12,313 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8164, 1.9948, 1.5282, 1.7208, 2.2961, 2.2638, 1.8996, 1.8103], + device='cuda:1'), covar=tensor([0.0437, 0.0378, 0.0692, 0.0414, 0.0314, 0.0606, 0.0400, 0.0474], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0106, 0.0145, 0.0112, 0.0101, 0.0114, 0.0103, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.7485e-05, 8.1502e-05, 1.1366e-04, 8.6100e-05, 7.8172e-05, 8.4619e-05, + 7.6476e-05, 8.5815e-05], device='cuda:1') +2023-03-27 05:23:12,795 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.699e+01 1.474e+02 1.759e+02 2.084e+02 3.295e+02, threshold=3.517e+02, percent-clipped=0.0 +2023-03-27 05:23:25,968 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-27 05:23:28,231 INFO [finetune.py:976] (1/7) Epoch 24, batch 3650, loss[loss=0.2021, simple_loss=0.263, pruned_loss=0.07056, over 4750.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2424, pruned_loss=0.0493, over 957439.15 frames. ], batch size: 27, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:24:00,319 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5198, 3.4602, 3.1849, 1.5874, 3.5572, 2.6567, 0.8897, 2.2481], + device='cuda:1'), covar=tensor([0.2483, 0.2219, 0.1719, 0.3518, 0.1157, 0.1050, 0.4283, 0.1674], + device='cuda:1'), in_proj_covar=tensor([0.0154, 0.0181, 0.0162, 0.0130, 0.0162, 0.0124, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 05:24:11,249 INFO [finetune.py:976] (1/7) Epoch 24, batch 3700, loss[loss=0.2212, simple_loss=0.2981, pruned_loss=0.07211, over 4843.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2474, pruned_loss=0.05083, over 957182.39 frames. ], batch size: 44, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:24:28,521 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.118e+01 1.614e+02 1.999e+02 2.429e+02 5.138e+02, threshold=3.998e+02, percent-clipped=6.0 +2023-03-27 05:24:43,336 INFO [finetune.py:976] (1/7) Epoch 24, batch 3750, loss[loss=0.169, simple_loss=0.2424, pruned_loss=0.04783, over 4775.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2487, pruned_loss=0.05152, over 955943.31 frames. ], batch size: 28, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:25:09,909 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9143, 1.6314, 2.2132, 1.4917, 2.0585, 2.2009, 1.6101, 2.2583], + device='cuda:1'), covar=tensor([0.1372, 0.2040, 0.1468, 0.1895, 0.1001, 0.1343, 0.2627, 0.0969], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0208, 0.0193, 0.0191, 0.0175, 0.0215, 0.0218, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:25:12,915 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=135518.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:25:18,852 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=135527.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:25:18,861 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3503, 1.3041, 1.6050, 2.4395, 1.6484, 2.1675, 0.8847, 2.0914], + device='cuda:1'), covar=tensor([0.1798, 0.1432, 0.1194, 0.0650, 0.0925, 0.1225, 0.1573, 0.0604], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0115, 0.0133, 0.0163, 0.0101, 0.0136, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 05:25:26,760 INFO [finetune.py:976] (1/7) Epoch 24, batch 3800, loss[loss=0.1866, simple_loss=0.255, pruned_loss=0.0591, over 4832.00 frames. ], tot_loss[loss=0.1767, simple_loss=0.2494, pruned_loss=0.05195, over 956776.56 frames. ], batch size: 30, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:25:32,989 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0213, 0.9879, 0.9332, 1.1006, 1.1850, 1.0942, 1.0052, 0.9487], + device='cuda:1'), covar=tensor([0.0394, 0.0314, 0.0698, 0.0343, 0.0329, 0.0470, 0.0392, 0.0459], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0108, 0.0148, 0.0114, 0.0102, 0.0116, 0.0104, 0.0115], + device='cuda:1'), out_proj_covar=tensor([7.8407e-05, 8.2709e-05, 1.1543e-04, 8.7079e-05, 7.9461e-05, 8.5885e-05, + 7.7439e-05, 8.7083e-05], device='cuda:1') +2023-03-27 05:25:36,770 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-03-27 05:25:44,706 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.088e+02 1.524e+02 1.815e+02 2.221e+02 4.659e+02, threshold=3.630e+02, percent-clipped=3.0 +2023-03-27 05:25:45,381 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=135566.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:26:00,453 INFO [finetune.py:976] (1/7) Epoch 24, batch 3850, loss[loss=0.1485, simple_loss=0.2319, pruned_loss=0.03257, over 4856.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2478, pruned_loss=0.05165, over 956326.80 frames. ], batch size: 44, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:26:17,559 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1273, 1.5668, 2.1329, 2.1577, 1.9718, 1.9581, 2.0434, 2.0744], + device='cuda:1'), covar=tensor([0.3618, 0.3651, 0.3159, 0.3452, 0.4766, 0.3672, 0.4622, 0.2763], + device='cuda:1'), in_proj_covar=tensor([0.0261, 0.0244, 0.0265, 0.0290, 0.0289, 0.0266, 0.0296, 0.0249], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:26:38,623 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=135630.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:26:45,795 INFO [finetune.py:976] (1/7) Epoch 24, batch 3900, loss[loss=0.198, simple_loss=0.252, pruned_loss=0.07202, over 4847.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.244, pruned_loss=0.05015, over 956126.62 frames. ], batch size: 49, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:27:08,436 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7054, 0.7852, 1.8148, 1.7335, 1.6404, 1.6032, 1.6346, 1.7320], + device='cuda:1'), covar=tensor([0.3716, 0.3891, 0.3213, 0.3275, 0.4318, 0.3507, 0.4145, 0.3083], + device='cuda:1'), in_proj_covar=tensor([0.0262, 0.0245, 0.0266, 0.0291, 0.0290, 0.0267, 0.0297, 0.0250], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:27:10,711 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.672e+01 1.400e+02 1.667e+02 1.961e+02 4.314e+02, threshold=3.334e+02, percent-clipped=1.0 +2023-03-27 05:27:26,038 INFO [finetune.py:976] (1/7) Epoch 24, batch 3950, loss[loss=0.1668, simple_loss=0.2416, pruned_loss=0.04606, over 4778.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2405, pruned_loss=0.04879, over 954843.02 frames. ], batch size: 29, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:27:29,113 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=135691.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:27:39,368 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3085, 2.9335, 2.7916, 1.2964, 3.0004, 2.1931, 0.7499, 1.9174], + device='cuda:1'), covar=tensor([0.2673, 0.2597, 0.1919, 0.3831, 0.1469, 0.1220, 0.4571, 0.1810], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0180, 0.0162, 0.0130, 0.0162, 0.0124, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 05:27:57,308 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4457, 1.3703, 1.6399, 2.5060, 1.6803, 2.1924, 0.9874, 2.1618], + device='cuda:1'), covar=tensor([0.1692, 0.1322, 0.1072, 0.0685, 0.0850, 0.1000, 0.1392, 0.0596], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0114, 0.0132, 0.0162, 0.0100, 0.0135, 0.0123, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 05:27:58,422 INFO [finetune.py:976] (1/7) Epoch 24, batch 4000, loss[loss=0.1964, simple_loss=0.2709, pruned_loss=0.06099, over 4823.00 frames. ], tot_loss[loss=0.17, simple_loss=0.241, pruned_loss=0.04951, over 957111.62 frames. ], batch size: 39, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:28:16,421 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.536e+01 1.548e+02 1.897e+02 2.315e+02 3.943e+02, threshold=3.793e+02, percent-clipped=5.0 +2023-03-27 05:28:31,232 INFO [finetune.py:976] (1/7) Epoch 24, batch 4050, loss[loss=0.144, simple_loss=0.2099, pruned_loss=0.03903, over 4679.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2446, pruned_loss=0.05144, over 956504.58 frames. ], batch size: 23, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:28:59,137 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=135827.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:29:09,952 INFO [finetune.py:976] (1/7) Epoch 24, batch 4100, loss[loss=0.1426, simple_loss=0.2152, pruned_loss=0.03498, over 4676.00 frames. ], tot_loss[loss=0.1733, simple_loss=0.2452, pruned_loss=0.05071, over 954145.91 frames. ], batch size: 23, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:29:16,202 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7229, 1.7358, 1.5563, 1.7295, 1.3944, 4.3192, 1.6259, 1.9576], + device='cuda:1'), covar=tensor([0.3505, 0.2516, 0.2199, 0.2589, 0.1716, 0.0144, 0.2481, 0.1255], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0123, 0.0114, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 05:29:32,644 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.978e+01 1.562e+02 1.866e+02 2.353e+02 4.250e+02, threshold=3.731e+02, percent-clipped=2.0 +2023-03-27 05:29:39,221 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=135875.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:29:41,595 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4795, 1.5355, 1.2812, 1.4630, 1.7739, 1.7963, 1.5233, 1.4250], + device='cuda:1'), covar=tensor([0.0330, 0.0302, 0.0644, 0.0327, 0.0239, 0.0423, 0.0325, 0.0404], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0106, 0.0145, 0.0111, 0.0100, 0.0113, 0.0101, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.6813e-05, 8.0870e-05, 1.1299e-04, 8.5002e-05, 7.7594e-05, 8.3720e-05, + 7.5262e-05, 8.5053e-05], device='cuda:1') +2023-03-27 05:29:46,948 INFO [finetune.py:976] (1/7) Epoch 24, batch 4150, loss[loss=0.1854, simple_loss=0.2641, pruned_loss=0.05339, over 4807.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2459, pruned_loss=0.05078, over 951861.62 frames. ], batch size: 51, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:30:30,442 INFO [finetune.py:976] (1/7) Epoch 24, batch 4200, loss[loss=0.1407, simple_loss=0.2217, pruned_loss=0.02989, over 4877.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2467, pruned_loss=0.05082, over 952188.36 frames. ], batch size: 43, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:30:39,953 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-03-27 05:30:47,789 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-27 05:30:49,318 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.053e+01 1.587e+02 1.796e+02 2.438e+02 3.967e+02, threshold=3.591e+02, percent-clipped=1.0 +2023-03-27 05:30:57,345 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-27 05:31:00,642 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=135982.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:31:03,041 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=135986.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:31:03,578 INFO [finetune.py:976] (1/7) Epoch 24, batch 4250, loss[loss=0.2022, simple_loss=0.2638, pruned_loss=0.07025, over 4928.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2447, pruned_loss=0.05057, over 953751.00 frames. ], batch size: 38, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:31:08,462 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=135994.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:31:45,376 INFO [finetune.py:976] (1/7) Epoch 24, batch 4300, loss[loss=0.144, simple_loss=0.2266, pruned_loss=0.03072, over 4837.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2422, pruned_loss=0.04977, over 953000.08 frames. ], batch size: 30, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:31:49,201 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136043.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:31:49,257 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.66 vs. limit=2.0 +2023-03-27 05:32:02,579 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136055.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:32:04,972 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-27 05:32:14,137 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.579e+01 1.491e+02 1.827e+02 2.181e+02 5.621e+02, threshold=3.653e+02, percent-clipped=1.0 +2023-03-27 05:32:20,553 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-27 05:32:31,267 INFO [finetune.py:976] (1/7) Epoch 24, batch 4350, loss[loss=0.1523, simple_loss=0.2286, pruned_loss=0.03805, over 4758.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2399, pruned_loss=0.04919, over 951475.08 frames. ], batch size: 27, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:32:47,245 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3679, 1.4662, 1.7336, 1.6884, 1.5745, 3.2140, 1.4026, 1.5698], + device='cuda:1'), covar=tensor([0.0960, 0.1720, 0.0973, 0.0922, 0.1552, 0.0227, 0.1427, 0.1665], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0081, 0.0072, 0.0076, 0.0091, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 05:33:04,533 INFO [finetune.py:976] (1/7) Epoch 24, batch 4400, loss[loss=0.1716, simple_loss=0.2507, pruned_loss=0.04628, over 4784.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.241, pruned_loss=0.04974, over 951556.59 frames. ], batch size: 29, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:33:08,147 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=136142.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:33:19,379 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3125, 1.8504, 2.2983, 2.2512, 1.9749, 2.0257, 2.2033, 2.0770], + device='cuda:1'), covar=tensor([0.3887, 0.4231, 0.3301, 0.3987, 0.5219, 0.3940, 0.4838, 0.3105], + device='cuda:1'), in_proj_covar=tensor([0.0261, 0.0244, 0.0264, 0.0289, 0.0289, 0.0266, 0.0296, 0.0248], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:33:23,889 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.065e+02 1.540e+02 1.819e+02 2.170e+02 3.954e+02, threshold=3.638e+02, percent-clipped=3.0 +2023-03-27 05:33:37,771 INFO [finetune.py:976] (1/7) Epoch 24, batch 4450, loss[loss=0.1769, simple_loss=0.2401, pruned_loss=0.05686, over 4218.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2446, pruned_loss=0.05053, over 953243.89 frames. ], batch size: 65, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:33:39,859 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.82 vs. limit=5.0 +2023-03-27 05:33:45,534 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1596, 3.5954, 3.8522, 4.0406, 3.9267, 3.7019, 4.2343, 1.3204], + device='cuda:1'), covar=tensor([0.0840, 0.0949, 0.0992, 0.1141, 0.1436, 0.1751, 0.0893, 0.5849], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0248, 0.0284, 0.0296, 0.0339, 0.0288, 0.0308, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:33:48,796 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136203.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:34:04,802 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5051, 3.9344, 4.1501, 4.3428, 4.2592, 4.0194, 4.6212, 1.3751], + device='cuda:1'), covar=tensor([0.0792, 0.0840, 0.0894, 0.1101, 0.1450, 0.1579, 0.0724, 0.5972], + device='cuda:1'), in_proj_covar=tensor([0.0348, 0.0247, 0.0283, 0.0295, 0.0338, 0.0287, 0.0307, 0.0301], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:34:13,599 INFO [finetune.py:976] (1/7) Epoch 24, batch 4500, loss[loss=0.2134, simple_loss=0.2762, pruned_loss=0.0753, over 4835.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2469, pruned_loss=0.05172, over 951897.14 frames. ], batch size: 44, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:34:39,505 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.886e+01 1.509e+02 1.852e+02 2.239e+02 3.856e+02, threshold=3.704e+02, percent-clipped=1.0 +2023-03-27 05:34:39,588 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9654, 4.2622, 4.1048, 2.3749, 4.4468, 3.3571, 0.7061, 3.1343], + device='cuda:1'), covar=tensor([0.2316, 0.1404, 0.1312, 0.2754, 0.0711, 0.0815, 0.4412, 0.1178], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0178, 0.0160, 0.0129, 0.0160, 0.0123, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 05:34:42,999 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3464, 2.1048, 2.7411, 4.4441, 3.1480, 3.0315, 1.1082, 3.6883], + device='cuda:1'), covar=tensor([0.1622, 0.1271, 0.1323, 0.0385, 0.0605, 0.1172, 0.1892, 0.0325], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0115, 0.0132, 0.0163, 0.0100, 0.0135, 0.0123, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 05:34:54,280 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136286.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:34:54,762 INFO [finetune.py:976] (1/7) Epoch 24, batch 4550, loss[loss=0.1421, simple_loss=0.2063, pruned_loss=0.03896, over 4239.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2461, pruned_loss=0.051, over 952365.44 frames. ], batch size: 18, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:35:22,699 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4530, 1.4318, 1.8333, 1.7038, 1.5703, 3.0736, 1.2983, 1.5992], + device='cuda:1'), covar=tensor([0.0958, 0.1702, 0.1191, 0.0928, 0.1493, 0.0260, 0.1441, 0.1637], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0092, 0.0080, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 05:35:28,208 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=136334.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:35:28,912 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=136335.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:35:30,006 INFO [finetune.py:976] (1/7) Epoch 24, batch 4600, loss[loss=0.2241, simple_loss=0.2771, pruned_loss=0.08556, over 4872.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2463, pruned_loss=0.05117, over 950845.75 frames. ], batch size: 34, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:35:35,220 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136338.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:35:45,768 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136350.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:35:56,270 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.033e+02 1.624e+02 1.856e+02 2.259e+02 4.732e+02, threshold=3.713e+02, percent-clipped=2.0 +2023-03-27 05:36:11,519 INFO [finetune.py:976] (1/7) Epoch 24, batch 4650, loss[loss=0.1769, simple_loss=0.2488, pruned_loss=0.05247, over 4837.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2446, pruned_loss=0.05088, over 952174.36 frames. ], batch size: 33, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:36:17,115 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136396.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:36:45,425 INFO [finetune.py:976] (1/7) Epoch 24, batch 4700, loss[loss=0.1506, simple_loss=0.2199, pruned_loss=0.04063, over 4824.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2411, pruned_loss=0.04933, over 953656.17 frames. ], batch size: 39, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:36:55,623 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-27 05:37:13,733 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.054e+02 1.479e+02 1.754e+02 2.064e+02 3.231e+02, threshold=3.507e+02, percent-clipped=0.0 +2023-03-27 05:37:22,840 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5415, 1.4368, 1.4337, 1.5013, 1.0006, 2.8028, 1.0795, 1.4893], + device='cuda:1'), covar=tensor([0.3144, 0.2372, 0.2017, 0.2230, 0.1775, 0.0261, 0.2792, 0.1216], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0117, 0.0122, 0.0124, 0.0114, 0.0097, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 05:37:38,205 INFO [finetune.py:976] (1/7) Epoch 24, batch 4750, loss[loss=0.1574, simple_loss=0.2285, pruned_loss=0.04314, over 4907.00 frames. ], tot_loss[loss=0.1667, simple_loss=0.2379, pruned_loss=0.04773, over 954396.17 frames. ], batch size: 35, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:37:44,921 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136498.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:38:10,343 INFO [finetune.py:976] (1/7) Epoch 24, batch 4800, loss[loss=0.2028, simple_loss=0.2815, pruned_loss=0.06212, over 4815.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2418, pruned_loss=0.0491, over 956034.62 frames. ], batch size: 39, lr: 3.04e-03, grad_scale: 16.0 +2023-03-27 05:38:24,507 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.47 vs. limit=5.0 +2023-03-27 05:38:28,976 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.123e+02 1.582e+02 2.021e+02 2.347e+02 5.093e+02, threshold=4.042e+02, percent-clipped=3.0 +2023-03-27 05:38:37,443 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3961, 1.3491, 1.4517, 0.8027, 1.5422, 1.4776, 1.4680, 1.2468], + device='cuda:1'), covar=tensor([0.0638, 0.0794, 0.0718, 0.1017, 0.0891, 0.0763, 0.0677, 0.1307], + device='cuda:1'), in_proj_covar=tensor([0.0129, 0.0135, 0.0138, 0.0118, 0.0125, 0.0137, 0.0136, 0.0159], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:38:44,074 INFO [finetune.py:976] (1/7) Epoch 24, batch 4850, loss[loss=0.225, simple_loss=0.2933, pruned_loss=0.07832, over 4836.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2443, pruned_loss=0.04929, over 954332.42 frames. ], batch size: 47, lr: 3.04e-03, grad_scale: 32.0 +2023-03-27 05:39:17,546 INFO [finetune.py:976] (1/7) Epoch 24, batch 4900, loss[loss=0.1731, simple_loss=0.2565, pruned_loss=0.04487, over 4915.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2459, pruned_loss=0.04994, over 954399.82 frames. ], batch size: 37, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:39:18,258 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136638.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:39:22,222 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=136643.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:39:26,540 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136650.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:39:42,309 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.069e+02 1.600e+02 1.925e+02 2.438e+02 3.559e+02, threshold=3.849e+02, percent-clipped=0.0 +2023-03-27 05:39:51,728 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6834, 3.4384, 3.3566, 1.8043, 3.6862, 2.7948, 0.9373, 2.4363], + device='cuda:1'), covar=tensor([0.2759, 0.1616, 0.1671, 0.3030, 0.1090, 0.0972, 0.4157, 0.1471], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0178, 0.0161, 0.0129, 0.0161, 0.0123, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 05:39:59,909 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=136686.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:40:00,476 INFO [finetune.py:976] (1/7) Epoch 24, batch 4950, loss[loss=0.2109, simple_loss=0.2728, pruned_loss=0.07455, over 4877.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2464, pruned_loss=0.05029, over 955237.64 frames. ], batch size: 35, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:40:03,956 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136691.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:40:08,762 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=136698.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:40:09,635 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-27 05:40:12,474 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136704.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:40:31,645 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9752, 1.9164, 1.6646, 2.1187, 2.4794, 2.0884, 1.7857, 1.5547], + device='cuda:1'), covar=tensor([0.2140, 0.1986, 0.1974, 0.1552, 0.1552, 0.1222, 0.2305, 0.1901], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0212, 0.0216, 0.0197, 0.0245, 0.0191, 0.0218, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:40:33,840 INFO [finetune.py:976] (1/7) Epoch 24, batch 5000, loss[loss=0.1576, simple_loss=0.2256, pruned_loss=0.04478, over 4776.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.245, pruned_loss=0.04965, over 956331.39 frames. ], batch size: 25, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:41:02,609 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.076e+02 1.520e+02 1.782e+02 2.173e+02 3.913e+02, threshold=3.563e+02, percent-clipped=1.0 +2023-03-27 05:41:17,086 INFO [finetune.py:976] (1/7) Epoch 24, batch 5050, loss[loss=0.1633, simple_loss=0.2331, pruned_loss=0.04668, over 4912.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2425, pruned_loss=0.04948, over 955205.07 frames. ], batch size: 36, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:41:25,180 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136798.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:41:30,676 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=136806.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:41:31,883 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4614, 1.3572, 1.3572, 1.3559, 0.9964, 2.3496, 0.7726, 1.1910], + device='cuda:1'), covar=tensor([0.3770, 0.2746, 0.2359, 0.2737, 0.1901, 0.0373, 0.2786, 0.1413], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0123, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 05:41:49,839 INFO [finetune.py:976] (1/7) Epoch 24, batch 5100, loss[loss=0.1751, simple_loss=0.246, pruned_loss=0.05207, over 4933.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2401, pruned_loss=0.049, over 953743.62 frames. ], batch size: 38, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:41:56,301 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=136846.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:42:11,831 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.036e+02 1.576e+02 1.921e+02 2.257e+02 4.191e+02, threshold=3.841e+02, percent-clipped=1.0 +2023-03-27 05:42:13,173 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136867.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:42:35,176 INFO [finetune.py:976] (1/7) Epoch 24, batch 5150, loss[loss=0.1749, simple_loss=0.2406, pruned_loss=0.05461, over 4901.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2401, pruned_loss=0.04979, over 955090.58 frames. ], batch size: 32, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:43:16,542 INFO [finetune.py:976] (1/7) Epoch 24, batch 5200, loss[loss=0.1787, simple_loss=0.2645, pruned_loss=0.04649, over 4800.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2435, pruned_loss=0.05106, over 951278.56 frames. ], batch size: 51, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:43:35,514 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.133e+02 1.647e+02 1.940e+02 2.397e+02 3.428e+02, threshold=3.879e+02, percent-clipped=0.0 +2023-03-27 05:43:48,851 INFO [finetune.py:976] (1/7) Epoch 24, batch 5250, loss[loss=0.1728, simple_loss=0.2543, pruned_loss=0.04564, over 4801.00 frames. ], tot_loss[loss=0.1761, simple_loss=0.2474, pruned_loss=0.05236, over 952995.43 frames. ], batch size: 45, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:43:51,885 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136991.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:43:57,223 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136999.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:44:22,622 INFO [finetune.py:976] (1/7) Epoch 24, batch 5300, loss[loss=0.2018, simple_loss=0.2771, pruned_loss=0.0633, over 4852.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2481, pruned_loss=0.05226, over 952027.42 frames. ], batch size: 44, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:44:23,939 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=137039.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:44:42,412 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.600e+02 1.832e+02 2.198e+02 3.821e+02, threshold=3.665e+02, percent-clipped=0.0 +2023-03-27 05:45:04,673 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7329, 1.2126, 0.7583, 1.4861, 2.1367, 1.1078, 1.3741, 1.4531], + device='cuda:1'), covar=tensor([0.1622, 0.2203, 0.1994, 0.1332, 0.2002, 0.2048, 0.1664, 0.2143], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0093, 0.0110, 0.0092, 0.0119, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 05:45:05,280 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7681, 1.3859, 0.9450, 1.5244, 2.0899, 1.5172, 1.5912, 1.6798], + device='cuda:1'), covar=tensor([0.1496, 0.1914, 0.1776, 0.1196, 0.1995, 0.1909, 0.1379, 0.1883], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0093, 0.0110, 0.0092, 0.0119, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 05:45:05,791 INFO [finetune.py:976] (1/7) Epoch 24, batch 5350, loss[loss=0.1565, simple_loss=0.2325, pruned_loss=0.0403, over 4904.00 frames. ], tot_loss[loss=0.1763, simple_loss=0.2485, pruned_loss=0.05203, over 953381.05 frames. ], batch size: 36, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:45:38,836 INFO [finetune.py:976] (1/7) Epoch 24, batch 5400, loss[loss=0.2117, simple_loss=0.2765, pruned_loss=0.07351, over 4792.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2457, pruned_loss=0.05118, over 952011.41 frames. ], batch size: 45, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:45:57,157 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=137162.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:45:58,904 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.720e+01 1.449e+02 1.770e+02 2.209e+02 4.288e+02, threshold=3.541e+02, percent-clipped=1.0 +2023-03-27 05:46:22,823 INFO [finetune.py:976] (1/7) Epoch 24, batch 5450, loss[loss=0.1629, simple_loss=0.2275, pruned_loss=0.04917, over 4827.00 frames. ], tot_loss[loss=0.172, simple_loss=0.243, pruned_loss=0.05052, over 952006.11 frames. ], batch size: 51, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:46:31,917 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2766, 2.1741, 1.9412, 2.3016, 2.7573, 2.2632, 2.2935, 1.8755], + device='cuda:1'), covar=tensor([0.1657, 0.1553, 0.1559, 0.1246, 0.1538, 0.0974, 0.1838, 0.1484], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0210, 0.0213, 0.0194, 0.0242, 0.0188, 0.0214, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:46:37,305 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6582, 2.3056, 2.9083, 1.6811, 2.4315, 3.0594, 2.2280, 3.0363], + device='cuda:1'), covar=tensor([0.1345, 0.2072, 0.1516, 0.2194, 0.1056, 0.1331, 0.2735, 0.0802], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0206, 0.0191, 0.0189, 0.0174, 0.0213, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:46:51,360 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-03-27 05:46:55,994 INFO [finetune.py:976] (1/7) Epoch 24, batch 5500, loss[loss=0.1604, simple_loss=0.234, pruned_loss=0.04335, over 4750.00 frames. ], tot_loss[loss=0.169, simple_loss=0.2392, pruned_loss=0.04939, over 953334.87 frames. ], batch size: 27, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:47:13,426 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.031e+02 1.493e+02 1.897e+02 2.213e+02 3.719e+02, threshold=3.794e+02, percent-clipped=2.0 +2023-03-27 05:47:14,034 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1025, 0.9513, 0.9343, 0.2777, 0.9813, 1.1843, 1.2304, 0.9851], + device='cuda:1'), covar=tensor([0.0772, 0.0766, 0.0606, 0.0510, 0.0526, 0.0637, 0.0420, 0.0701], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0150, 0.0128, 0.0122, 0.0132, 0.0130, 0.0142, 0.0150], + device='cuda:1'), out_proj_covar=tensor([8.9828e-05, 1.0763e-04, 9.1275e-05, 8.6044e-05, 9.2627e-05, 9.2786e-05, + 1.0159e-04, 1.0695e-04], device='cuda:1') +2023-03-27 05:47:34,585 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-03-27 05:47:36,857 INFO [finetune.py:976] (1/7) Epoch 24, batch 5550, loss[loss=0.1507, simple_loss=0.2281, pruned_loss=0.03663, over 4779.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.241, pruned_loss=0.04993, over 952975.11 frames. ], batch size: 29, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:47:44,082 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2614, 2.1796, 1.7944, 2.2674, 2.1447, 1.9151, 2.4563, 2.3387], + device='cuda:1'), covar=tensor([0.1340, 0.2079, 0.2999, 0.2372, 0.2789, 0.1780, 0.2950, 0.1748], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0189, 0.0236, 0.0254, 0.0250, 0.0206, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:47:47,667 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=137299.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:48:14,910 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3279, 1.4023, 1.4915, 0.7413, 1.5146, 1.7324, 1.7752, 1.3333], + device='cuda:1'), covar=tensor([0.0883, 0.0604, 0.0584, 0.0521, 0.0499, 0.0576, 0.0336, 0.0722], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0150, 0.0128, 0.0122, 0.0132, 0.0131, 0.0142, 0.0150], + device='cuda:1'), out_proj_covar=tensor([8.9924e-05, 1.0787e-04, 9.1458e-05, 8.6144e-05, 9.2513e-05, 9.2882e-05, + 1.0159e-04, 1.0693e-04], device='cuda:1') +2023-03-27 05:48:20,577 INFO [finetune.py:976] (1/7) Epoch 24, batch 5600, loss[loss=0.1679, simple_loss=0.2495, pruned_loss=0.04311, over 4839.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2439, pruned_loss=0.05038, over 953716.10 frames. ], batch size: 47, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:48:26,395 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=137347.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:48:32,121 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.56 vs. limit=2.0 +2023-03-27 05:48:37,702 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.121e+02 1.584e+02 1.843e+02 2.259e+02 3.753e+02, threshold=3.686e+02, percent-clipped=0.0 +2023-03-27 05:48:51,115 INFO [finetune.py:976] (1/7) Epoch 24, batch 5650, loss[loss=0.2173, simple_loss=0.288, pruned_loss=0.07327, over 4857.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2462, pruned_loss=0.0506, over 952860.40 frames. ], batch size: 49, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:48:55,375 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-27 05:49:20,946 INFO [finetune.py:976] (1/7) Epoch 24, batch 5700, loss[loss=0.1675, simple_loss=0.2257, pruned_loss=0.0547, over 4090.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.2429, pruned_loss=0.04985, over 938737.08 frames. ], batch size: 17, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:49:31,091 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9741, 1.8014, 1.6990, 2.0609, 2.1366, 2.0143, 1.4783, 1.6678], + device='cuda:1'), covar=tensor([0.1886, 0.1773, 0.1669, 0.1301, 0.1396, 0.0979, 0.2258, 0.1818], + device='cuda:1'), in_proj_covar=tensor([0.0242, 0.0210, 0.0213, 0.0195, 0.0243, 0.0189, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:49:35,755 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=137462.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:49:52,032 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.393e+01 1.414e+02 1.686e+02 2.138e+02 3.465e+02, threshold=3.373e+02, percent-clipped=0.0 +2023-03-27 05:49:52,048 INFO [finetune.py:976] (1/7) Epoch 25, batch 0, loss[loss=0.1949, simple_loss=0.2755, pruned_loss=0.05717, over 4817.00 frames. ], tot_loss[loss=0.1949, simple_loss=0.2755, pruned_loss=0.05717, over 4817.00 frames. ], batch size: 38, lr: 3.03e-03, grad_scale: 32.0 +2023-03-27 05:49:52,049 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 05:49:58,650 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6835, 3.6198, 3.3694, 1.5077, 3.6000, 2.6811, 0.6514, 2.4431], + device='cuda:1'), covar=tensor([0.1872, 0.2263, 0.1720, 0.3563, 0.1193, 0.1119, 0.4102, 0.1606], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0177, 0.0160, 0.0129, 0.0161, 0.0123, 0.0147, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 05:50:06,682 INFO [finetune.py:1010] (1/7) Epoch 25, validation: loss=0.1587, simple_loss=0.2267, pruned_loss=0.04536, over 2265189.00 frames. +2023-03-27 05:50:06,683 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 05:50:46,518 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=137510.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:50:50,024 INFO [finetune.py:976] (1/7) Epoch 25, batch 50, loss[loss=0.1831, simple_loss=0.2579, pruned_loss=0.0542, over 4898.00 frames. ], tot_loss[loss=0.1815, simple_loss=0.2524, pruned_loss=0.05526, over 216004.29 frames. ], batch size: 37, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:50:51,849 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6703, 1.6833, 1.5960, 1.7097, 1.3393, 3.3628, 1.4266, 1.7972], + device='cuda:1'), covar=tensor([0.3091, 0.2410, 0.2049, 0.2194, 0.1513, 0.0252, 0.2424, 0.1209], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0123, 0.0113, 0.0095, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 05:50:53,709 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.26 vs. limit=5.0 +2023-03-27 05:51:06,076 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6787, 1.2585, 0.9414, 1.6126, 2.0553, 1.3010, 1.4690, 1.6352], + device='cuda:1'), covar=tensor([0.1362, 0.1834, 0.1748, 0.1119, 0.1801, 0.1895, 0.1289, 0.1726], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0093, 0.0109, 0.0092, 0.0119, 0.0092, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 05:51:09,918 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2932, 1.1922, 1.4452, 2.3189, 1.5241, 2.1185, 0.7811, 2.0510], + device='cuda:1'), covar=tensor([0.2099, 0.2025, 0.1551, 0.1213, 0.1194, 0.1415, 0.1968, 0.0771], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0163, 0.0101, 0.0136, 0.0124, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 05:51:12,267 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1701, 1.2968, 1.3803, 0.6407, 1.2972, 1.5714, 1.6186, 1.3009], + device='cuda:1'), covar=tensor([0.0938, 0.0627, 0.0542, 0.0498, 0.0496, 0.0650, 0.0326, 0.0758], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0150, 0.0127, 0.0123, 0.0131, 0.0130, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.9609e-05, 1.0756e-04, 9.0843e-05, 8.6264e-05, 9.2124e-05, 9.2654e-05, + 1.0144e-04, 1.0627e-04], device='cuda:1') +2023-03-27 05:51:25,237 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.123e+02 1.580e+02 1.851e+02 2.170e+02 4.183e+02, threshold=3.702e+02, percent-clipped=2.0 +2023-03-27 05:51:25,253 INFO [finetune.py:976] (1/7) Epoch 25, batch 100, loss[loss=0.1419, simple_loss=0.2248, pruned_loss=0.02947, over 4906.00 frames. ], tot_loss[loss=0.1753, simple_loss=0.2455, pruned_loss=0.05256, over 381837.69 frames. ], batch size: 32, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:51:59,264 INFO [finetune.py:976] (1/7) Epoch 25, batch 150, loss[loss=0.1923, simple_loss=0.2472, pruned_loss=0.06868, over 4895.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2405, pruned_loss=0.05059, over 510796.99 frames. ], batch size: 32, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:52:09,243 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.11 vs. limit=2.0 +2023-03-27 05:52:33,558 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.113e+02 1.544e+02 1.791e+02 2.141e+02 4.771e+02, threshold=3.582e+02, percent-clipped=2.0 +2023-03-27 05:52:33,574 INFO [finetune.py:976] (1/7) Epoch 25, batch 200, loss[loss=0.1948, simple_loss=0.266, pruned_loss=0.06183, over 4919.00 frames. ], tot_loss[loss=0.1685, simple_loss=0.2378, pruned_loss=0.04961, over 610378.61 frames. ], batch size: 37, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:52:49,589 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-27 05:53:27,000 INFO [finetune.py:976] (1/7) Epoch 25, batch 250, loss[loss=0.2035, simple_loss=0.28, pruned_loss=0.06349, over 4805.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2408, pruned_loss=0.05008, over 686982.23 frames. ], batch size: 41, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:53:31,850 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0318, 2.0441, 1.6994, 1.9598, 1.8335, 1.8774, 1.8936, 2.5918], + device='cuda:1'), covar=tensor([0.3419, 0.3702, 0.3108, 0.3482, 0.3706, 0.2272, 0.3571, 0.1516], + device='cuda:1'), in_proj_covar=tensor([0.0291, 0.0265, 0.0236, 0.0277, 0.0260, 0.0229, 0.0257, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:54:00,389 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.518e+01 1.603e+02 1.998e+02 2.287e+02 4.515e+02, threshold=3.995e+02, percent-clipped=2.0 +2023-03-27 05:54:00,405 INFO [finetune.py:976] (1/7) Epoch 25, batch 300, loss[loss=0.1634, simple_loss=0.2374, pruned_loss=0.04469, over 4791.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2435, pruned_loss=0.05014, over 746952.44 frames. ], batch size: 29, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:54:01,146 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7253, 1.0936, 1.8667, 1.8003, 1.6130, 1.5372, 1.7188, 1.7513], + device='cuda:1'), covar=tensor([0.3448, 0.3487, 0.2657, 0.2990, 0.3733, 0.3391, 0.3584, 0.2544], + device='cuda:1'), in_proj_covar=tensor([0.0262, 0.0246, 0.0266, 0.0291, 0.0291, 0.0267, 0.0297, 0.0249], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:54:33,830 INFO [finetune.py:976] (1/7) Epoch 25, batch 350, loss[loss=0.186, simple_loss=0.267, pruned_loss=0.05251, over 4818.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2463, pruned_loss=0.05064, over 794157.21 frames. ], batch size: 39, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:55:00,507 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-27 05:55:07,123 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.080e+02 1.541e+02 1.822e+02 2.129e+02 2.910e+02, threshold=3.644e+02, percent-clipped=0.0 +2023-03-27 05:55:07,139 INFO [finetune.py:976] (1/7) Epoch 25, batch 400, loss[loss=0.162, simple_loss=0.2409, pruned_loss=0.04156, over 4827.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2486, pruned_loss=0.05161, over 829686.32 frames. ], batch size: 30, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:55:31,420 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=137892.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:55:33,187 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=137895.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:55:53,571 INFO [finetune.py:976] (1/7) Epoch 25, batch 450, loss[loss=0.1343, simple_loss=0.1965, pruned_loss=0.03604, over 4240.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2469, pruned_loss=0.05074, over 859504.34 frames. ], batch size: 18, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:55:54,837 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2113, 1.7463, 1.2509, 1.9813, 2.4892, 1.8103, 1.9414, 1.9752], + device='cuda:1'), covar=tensor([0.1219, 0.1751, 0.1599, 0.1020, 0.1492, 0.1561, 0.1237, 0.1769], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0093, 0.0109, 0.0092, 0.0118, 0.0092, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 05:56:19,215 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=137953.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:56:20,980 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=137956.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:56:26,869 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.118e+02 1.489e+02 1.801e+02 2.267e+02 5.324e+02, threshold=3.602e+02, percent-clipped=3.0 +2023-03-27 05:56:26,885 INFO [finetune.py:976] (1/7) Epoch 25, batch 500, loss[loss=0.1783, simple_loss=0.2468, pruned_loss=0.05495, over 4816.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2447, pruned_loss=0.05039, over 879418.70 frames. ], batch size: 41, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:56:57,126 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-27 05:56:58,241 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9033, 1.8013, 1.6311, 1.8544, 2.1249, 2.0835, 1.8055, 1.6670], + device='cuda:1'), covar=tensor([0.0373, 0.0383, 0.0644, 0.0363, 0.0255, 0.0543, 0.0346, 0.0458], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0106, 0.0145, 0.0112, 0.0101, 0.0114, 0.0102, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.7783e-05, 8.1419e-05, 1.1367e-04, 8.5842e-05, 7.8596e-05, 8.3952e-05, + 7.5641e-05, 8.5857e-05], device='cuda:1') +2023-03-27 05:57:01,546 INFO [finetune.py:976] (1/7) Epoch 25, batch 550, loss[loss=0.1605, simple_loss=0.2265, pruned_loss=0.04724, over 4866.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2411, pruned_loss=0.04909, over 896933.21 frames. ], batch size: 31, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:57:34,656 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.805e+01 1.427e+02 1.740e+02 1.994e+02 3.808e+02, threshold=3.480e+02, percent-clipped=1.0 +2023-03-27 05:57:34,672 INFO [finetune.py:976] (1/7) Epoch 25, batch 600, loss[loss=0.1796, simple_loss=0.2526, pruned_loss=0.05333, over 4917.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.241, pruned_loss=0.04928, over 910900.23 frames. ], batch size: 36, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:58:07,363 INFO [finetune.py:976] (1/7) Epoch 25, batch 650, loss[loss=0.1823, simple_loss=0.2544, pruned_loss=0.05506, over 4753.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2436, pruned_loss=0.05002, over 921330.42 frames. ], batch size: 27, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:58:08,576 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9385, 1.5142, 0.8116, 1.7392, 2.2694, 1.4783, 1.7968, 1.7623], + device='cuda:1'), covar=tensor([0.1554, 0.2026, 0.2136, 0.1280, 0.1843, 0.1859, 0.1464, 0.2127], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0093, 0.0119, 0.0093, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 05:58:59,111 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.045e+02 1.613e+02 1.882e+02 2.325e+02 5.163e+02, threshold=3.765e+02, percent-clipped=4.0 +2023-03-27 05:58:59,127 INFO [finetune.py:976] (1/7) Epoch 25, batch 700, loss[loss=0.1243, simple_loss=0.1895, pruned_loss=0.02955, over 4702.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2465, pruned_loss=0.05116, over 928306.74 frames. ], batch size: 23, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:58:59,877 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8875, 1.3917, 1.9467, 1.8935, 1.7020, 1.6529, 1.8357, 1.8030], + device='cuda:1'), covar=tensor([0.3839, 0.3826, 0.2990, 0.3401, 0.4206, 0.3625, 0.3939, 0.2806], + device='cuda:1'), in_proj_covar=tensor([0.0262, 0.0246, 0.0266, 0.0291, 0.0291, 0.0268, 0.0297, 0.0248], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:59:16,069 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-27 05:59:32,527 INFO [finetune.py:976] (1/7) Epoch 25, batch 750, loss[loss=0.1547, simple_loss=0.2288, pruned_loss=0.04025, over 4759.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2474, pruned_loss=0.05142, over 935854.30 frames. ], batch size: 59, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 05:59:41,549 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7878, 1.4912, 1.8986, 1.2992, 1.8850, 1.8918, 1.4409, 2.0723], + device='cuda:1'), covar=tensor([0.0915, 0.1967, 0.1104, 0.1584, 0.0635, 0.1019, 0.2639, 0.0621], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0191, 0.0189, 0.0174, 0.0212, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 05:59:53,522 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138248.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 05:59:55,849 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138251.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:00:05,193 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.386e+01 1.513e+02 1.803e+02 2.270e+02 6.862e+02, threshold=3.605e+02, percent-clipped=3.0 +2023-03-27 06:00:05,209 INFO [finetune.py:976] (1/7) Epoch 25, batch 800, loss[loss=0.1614, simple_loss=0.247, pruned_loss=0.03786, over 4906.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2478, pruned_loss=0.05135, over 937894.63 frames. ], batch size: 43, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 06:00:08,368 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=138270.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:00:38,347 INFO [finetune.py:976] (1/7) Epoch 25, batch 850, loss[loss=0.1488, simple_loss=0.2176, pruned_loss=0.04, over 4693.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2454, pruned_loss=0.05014, over 942310.65 frames. ], batch size: 23, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 06:00:44,520 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=138322.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:00:53,878 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6491, 1.4766, 2.1927, 3.4447, 2.3579, 2.4904, 0.6546, 2.8719], + device='cuda:1'), covar=tensor([0.1782, 0.1465, 0.1376, 0.0559, 0.0795, 0.1237, 0.2150, 0.0434], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0164, 0.0102, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 06:00:54,544 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=138331.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:00:58,290 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.10 vs. limit=5.0 +2023-03-27 06:01:24,851 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.023e+02 1.388e+02 1.706e+02 2.091e+02 3.563e+02, threshold=3.412e+02, percent-clipped=0.0 +2023-03-27 06:01:24,867 INFO [finetune.py:976] (1/7) Epoch 25, batch 900, loss[loss=0.1225, simple_loss=0.1931, pruned_loss=0.02592, over 4104.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2423, pruned_loss=0.04896, over 944805.69 frames. ], batch size: 18, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 06:01:36,363 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=138383.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:01:39,580 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-03-27 06:01:41,968 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.42 vs. limit=5.0 +2023-03-27 06:01:57,678 INFO [finetune.py:976] (1/7) Epoch 25, batch 950, loss[loss=0.0954, simple_loss=0.1629, pruned_loss=0.01397, over 4238.00 frames. ], tot_loss[loss=0.169, simple_loss=0.2403, pruned_loss=0.04882, over 946649.50 frames. ], batch size: 18, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 06:02:30,846 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.004e+02 1.447e+02 1.781e+02 2.298e+02 4.571e+02, threshold=3.563e+02, percent-clipped=3.0 +2023-03-27 06:02:30,862 INFO [finetune.py:976] (1/7) Epoch 25, batch 1000, loss[loss=0.152, simple_loss=0.2324, pruned_loss=0.03581, over 4815.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.2404, pruned_loss=0.04861, over 948386.29 frames. ], batch size: 51, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 06:02:56,199 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5016, 2.2293, 2.8731, 1.7457, 2.5343, 2.8237, 2.0497, 2.8133], + device='cuda:1'), covar=tensor([0.1336, 0.1843, 0.1350, 0.2118, 0.0910, 0.1438, 0.2592, 0.0897], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0207, 0.0191, 0.0189, 0.0174, 0.0213, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:03:00,325 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5270, 3.9474, 4.1440, 4.3618, 4.3311, 4.1014, 4.6206, 1.5286], + device='cuda:1'), covar=tensor([0.0752, 0.0798, 0.0792, 0.0912, 0.1094, 0.1403, 0.0595, 0.5685], + device='cuda:1'), in_proj_covar=tensor([0.0345, 0.0245, 0.0280, 0.0292, 0.0335, 0.0284, 0.0304, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:03:03,759 INFO [finetune.py:976] (1/7) Epoch 25, batch 1050, loss[loss=0.1475, simple_loss=0.2263, pruned_loss=0.03433, over 4756.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.244, pruned_loss=0.04974, over 949135.22 frames. ], batch size: 27, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 06:03:15,703 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5541, 1.5012, 1.4331, 1.5722, 1.1934, 3.2936, 1.2948, 1.6480], + device='cuda:1'), covar=tensor([0.3385, 0.2596, 0.2184, 0.2482, 0.1694, 0.0213, 0.2656, 0.1304], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0123, 0.0112, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 06:03:25,361 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=138548.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:03:27,133 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=138551.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:03:39,079 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.570e+02 1.874e+02 2.177e+02 7.699e+02, threshold=3.747e+02, percent-clipped=3.0 +2023-03-27 06:03:39,095 INFO [finetune.py:976] (1/7) Epoch 25, batch 1100, loss[loss=0.1701, simple_loss=0.2467, pruned_loss=0.0468, over 4856.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2467, pruned_loss=0.05098, over 949622.96 frames. ], batch size: 44, lr: 3.02e-03, grad_scale: 32.0 +2023-03-27 06:04:15,223 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-27 06:04:17,868 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=138596.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:04:19,633 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=138599.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:04:29,870 INFO [finetune.py:976] (1/7) Epoch 25, batch 1150, loss[loss=0.1539, simple_loss=0.2345, pruned_loss=0.03668, over 4815.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.248, pruned_loss=0.05139, over 951642.38 frames. ], batch size: 38, lr: 3.02e-03, grad_scale: 64.0 +2023-03-27 06:04:39,030 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138626.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:04:52,176 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6195, 1.5476, 1.5336, 1.5596, 1.0505, 3.0038, 1.1689, 1.6429], + device='cuda:1'), covar=tensor([0.3350, 0.2546, 0.2089, 0.2428, 0.1870, 0.0263, 0.2618, 0.1275], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0124, 0.0113, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 06:04:58,857 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4717, 1.6124, 1.2657, 1.5516, 1.8231, 1.7883, 1.5198, 1.3410], + device='cuda:1'), covar=tensor([0.0401, 0.0315, 0.0732, 0.0338, 0.0227, 0.0461, 0.0411, 0.0502], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0108, 0.0147, 0.0113, 0.0102, 0.0115, 0.0103, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8626e-05, 8.2718e-05, 1.1515e-04, 8.6753e-05, 7.9317e-05, 8.4797e-05, + 7.6653e-05, 8.6964e-05], device='cuda:1') +2023-03-27 06:05:03,419 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.832e+01 1.520e+02 1.733e+02 2.222e+02 3.582e+02, threshold=3.466e+02, percent-clipped=0.0 +2023-03-27 06:05:03,435 INFO [finetune.py:976] (1/7) Epoch 25, batch 1200, loss[loss=0.1513, simple_loss=0.2294, pruned_loss=0.03662, over 4896.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2468, pruned_loss=0.05075, over 953420.19 frames. ], batch size: 43, lr: 3.02e-03, grad_scale: 64.0 +2023-03-27 06:05:13,801 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138678.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:05:15,614 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=138681.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:05:18,699 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=138686.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:05:25,916 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=138697.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:05:28,980 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7699, 1.7070, 1.4489, 1.7087, 2.1022, 1.9936, 1.7191, 1.5126], + device='cuda:1'), covar=tensor([0.0350, 0.0320, 0.0636, 0.0322, 0.0193, 0.0486, 0.0318, 0.0440], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0108, 0.0147, 0.0113, 0.0102, 0.0115, 0.0103, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8745e-05, 8.2774e-05, 1.1523e-04, 8.6788e-05, 7.9350e-05, 8.4945e-05, + 7.6567e-05, 8.6954e-05], device='cuda:1') +2023-03-27 06:05:37,221 INFO [finetune.py:976] (1/7) Epoch 25, batch 1250, loss[loss=0.1833, simple_loss=0.2433, pruned_loss=0.06168, over 4257.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2436, pruned_loss=0.04996, over 954514.07 frames. ], batch size: 65, lr: 3.02e-03, grad_scale: 64.0 +2023-03-27 06:05:55,586 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=138742.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:05:59,136 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=138747.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:06:05,732 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=138758.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:06:11,262 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3886, 2.2371, 1.8654, 2.2579, 2.2956, 2.0618, 2.5502, 2.3601], + device='cuda:1'), covar=tensor([0.1342, 0.2202, 0.3101, 0.2618, 0.2755, 0.1903, 0.2953, 0.1856], + device='cuda:1'), in_proj_covar=tensor([0.0187, 0.0189, 0.0233, 0.0252, 0.0248, 0.0204, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:06:12,323 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.092e+02 1.434e+02 1.700e+02 2.124e+02 3.591e+02, threshold=3.400e+02, percent-clipped=1.0 +2023-03-27 06:06:12,339 INFO [finetune.py:976] (1/7) Epoch 25, batch 1300, loss[loss=0.1723, simple_loss=0.236, pruned_loss=0.05428, over 4827.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2399, pruned_loss=0.04869, over 954706.31 frames. ], batch size: 51, lr: 3.02e-03, grad_scale: 64.0 +2023-03-27 06:06:23,666 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0485, 3.5298, 3.6791, 3.7887, 3.8547, 3.6496, 4.1462, 1.5949], + device='cuda:1'), covar=tensor([0.0844, 0.0958, 0.0974, 0.1216, 0.1173, 0.1498, 0.0779, 0.5422], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0247, 0.0283, 0.0296, 0.0338, 0.0286, 0.0306, 0.0301], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:06:49,007 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0175, 1.7249, 2.3316, 1.5237, 2.2316, 2.2604, 1.6035, 2.4800], + device='cuda:1'), covar=tensor([0.1204, 0.2115, 0.1228, 0.1880, 0.0756, 0.1306, 0.2946, 0.0723], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0206, 0.0190, 0.0189, 0.0173, 0.0213, 0.0215, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:06:49,179 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-27 06:06:57,378 INFO [finetune.py:976] (1/7) Epoch 25, batch 1350, loss[loss=0.1814, simple_loss=0.255, pruned_loss=0.05393, over 4866.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.2393, pruned_loss=0.04796, over 956816.95 frames. ], batch size: 34, lr: 3.02e-03, grad_scale: 64.0 +2023-03-27 06:07:03,531 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8511, 1.3940, 1.8287, 1.8161, 1.5983, 1.5484, 1.7836, 1.7482], + device='cuda:1'), covar=tensor([0.3679, 0.3653, 0.3144, 0.3333, 0.4743, 0.3630, 0.4178, 0.2934], + device='cuda:1'), in_proj_covar=tensor([0.0262, 0.0245, 0.0265, 0.0291, 0.0291, 0.0267, 0.0296, 0.0248], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:07:18,249 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5232, 1.7105, 1.3960, 1.4056, 2.0510, 1.9430, 1.8342, 1.6930], + device='cuda:1'), covar=tensor([0.0521, 0.0411, 0.0704, 0.0398, 0.0331, 0.0771, 0.0392, 0.0424], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0108, 0.0148, 0.0113, 0.0102, 0.0115, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8843e-05, 8.2958e-05, 1.1548e-04, 8.6827e-05, 7.9490e-05, 8.5117e-05, + 7.6887e-05, 8.7011e-05], device='cuda:1') +2023-03-27 06:07:31,274 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.011e+02 1.573e+02 1.999e+02 2.321e+02 4.595e+02, threshold=3.999e+02, percent-clipped=3.0 +2023-03-27 06:07:31,290 INFO [finetune.py:976] (1/7) Epoch 25, batch 1400, loss[loss=0.1395, simple_loss=0.2249, pruned_loss=0.027, over 4783.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2417, pruned_loss=0.04821, over 956639.89 frames. ], batch size: 29, lr: 3.02e-03, grad_scale: 64.0 +2023-03-27 06:07:49,659 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9305, 1.7573, 1.5316, 1.3602, 1.7452, 1.7254, 1.6888, 2.1997], + device='cuda:1'), covar=tensor([0.3949, 0.3968, 0.3361, 0.3712, 0.3935, 0.2427, 0.3451, 0.2068], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0263, 0.0234, 0.0274, 0.0258, 0.0228, 0.0255, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:08:01,088 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=138909.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:08:04,578 INFO [finetune.py:976] (1/7) Epoch 25, batch 1450, loss[loss=0.1841, simple_loss=0.2688, pruned_loss=0.04971, over 4885.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2435, pruned_loss=0.04845, over 954765.23 frames. ], batch size: 35, lr: 3.01e-03, grad_scale: 64.0 +2023-03-27 06:08:06,364 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1566, 1.5226, 2.5967, 4.1115, 2.7973, 2.9134, 1.3133, 3.6751], + device='cuda:1'), covar=tensor([0.1902, 0.2101, 0.1545, 0.0870, 0.0904, 0.1429, 0.2054, 0.0460], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0134, 0.0165, 0.0102, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 06:08:11,750 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=138926.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:08:38,071 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.768e+01 1.472e+02 1.793e+02 2.201e+02 3.947e+02, threshold=3.587e+02, percent-clipped=0.0 +2023-03-27 06:08:38,087 INFO [finetune.py:976] (1/7) Epoch 25, batch 1500, loss[loss=0.2066, simple_loss=0.2647, pruned_loss=0.07423, over 4844.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2456, pruned_loss=0.04927, over 955842.12 frames. ], batch size: 47, lr: 3.01e-03, grad_scale: 64.0 +2023-03-27 06:08:41,069 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2400, 2.1432, 2.2687, 1.5455, 2.3002, 2.4491, 2.3131, 1.7773], + device='cuda:1'), covar=tensor([0.0584, 0.0666, 0.0700, 0.0816, 0.0668, 0.0659, 0.0619, 0.1313], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0138, 0.0141, 0.0120, 0.0127, 0.0140, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:08:41,668 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=138970.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 06:08:44,019 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=138974.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:08:46,497 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=138978.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:09:23,187 INFO [finetune.py:976] (1/7) Epoch 25, batch 1550, loss[loss=0.1758, simple_loss=0.2514, pruned_loss=0.05012, over 4823.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2462, pruned_loss=0.04969, over 953363.41 frames. ], batch size: 30, lr: 3.01e-03, grad_scale: 64.0 +2023-03-27 06:09:34,900 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=139026.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:09:46,775 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139037.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:09:50,279 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139042.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:09:57,457 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139053.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:10:02,896 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1233, 1.5010, 2.1197, 2.1249, 1.8706, 1.8285, 2.0626, 2.0132], + device='cuda:1'), covar=tensor([0.3389, 0.3384, 0.2987, 0.3180, 0.4536, 0.3730, 0.3824, 0.2991], + device='cuda:1'), in_proj_covar=tensor([0.0261, 0.0245, 0.0265, 0.0290, 0.0291, 0.0267, 0.0297, 0.0248], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:10:05,107 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.522e+01 1.375e+02 1.683e+02 2.077e+02 3.862e+02, threshold=3.366e+02, percent-clipped=3.0 +2023-03-27 06:10:05,123 INFO [finetune.py:976] (1/7) Epoch 25, batch 1600, loss[loss=0.1844, simple_loss=0.2581, pruned_loss=0.05536, over 4916.00 frames. ], tot_loss[loss=0.1722, simple_loss=0.2448, pruned_loss=0.04977, over 955232.52 frames. ], batch size: 36, lr: 3.01e-03, grad_scale: 64.0 +2023-03-27 06:10:38,951 INFO [finetune.py:976] (1/7) Epoch 25, batch 1650, loss[loss=0.1662, simple_loss=0.2234, pruned_loss=0.05452, over 4702.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2422, pruned_loss=0.04913, over 953866.96 frames. ], batch size: 23, lr: 3.01e-03, grad_scale: 64.0 +2023-03-27 06:11:12,573 INFO [finetune.py:976] (1/7) Epoch 25, batch 1700, loss[loss=0.2246, simple_loss=0.2913, pruned_loss=0.07893, over 4185.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2414, pruned_loss=0.04968, over 953459.12 frames. ], batch size: 65, lr: 3.01e-03, grad_scale: 32.0 +2023-03-27 06:11:13,175 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.831e+01 1.435e+02 1.759e+02 2.193e+02 3.727e+02, threshold=3.518e+02, percent-clipped=3.0 +2023-03-27 06:11:21,690 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4631, 1.3159, 1.4913, 0.8110, 1.4714, 1.4906, 1.4684, 1.2631], + device='cuda:1'), covar=tensor([0.0637, 0.0970, 0.0743, 0.0984, 0.0929, 0.0812, 0.0719, 0.1417], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0140, 0.0120, 0.0127, 0.0139, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:11:33,712 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1388, 2.0237, 1.7557, 1.9357, 1.8933, 1.8923, 1.9276, 2.6704], + device='cuda:1'), covar=tensor([0.3340, 0.3944, 0.3198, 0.3681, 0.3880, 0.2309, 0.3630, 0.1584], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0263, 0.0234, 0.0274, 0.0258, 0.0228, 0.0255, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:11:56,394 INFO [finetune.py:976] (1/7) Epoch 25, batch 1750, loss[loss=0.2019, simple_loss=0.2771, pruned_loss=0.0633, over 4911.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2444, pruned_loss=0.05134, over 954443.56 frames. ], batch size: 36, lr: 3.01e-03, grad_scale: 32.0 +2023-03-27 06:12:20,712 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139238.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:12:23,608 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0205, 1.9829, 2.1354, 1.5664, 1.9616, 2.2262, 2.1711, 1.6718], + device='cuda:1'), covar=tensor([0.0494, 0.0557, 0.0579, 0.0769, 0.0956, 0.0526, 0.0487, 0.1148], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0140, 0.0120, 0.0127, 0.0139, 0.0139, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:12:39,404 INFO [finetune.py:976] (1/7) Epoch 25, batch 1800, loss[loss=0.1884, simple_loss=0.262, pruned_loss=0.05741, over 4904.00 frames. ], tot_loss[loss=0.1751, simple_loss=0.2466, pruned_loss=0.05179, over 953852.77 frames. ], batch size: 37, lr: 3.01e-03, grad_scale: 32.0 +2023-03-27 06:12:39,475 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139265.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 06:12:39,955 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.974e+01 1.553e+02 1.833e+02 2.131e+02 4.022e+02, threshold=3.667e+02, percent-clipped=3.0 +2023-03-27 06:12:44,232 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4258, 2.4102, 1.9198, 0.9643, 2.1289, 1.8608, 1.7436, 2.2038], + device='cuda:1'), covar=tensor([0.0921, 0.0674, 0.1674, 0.1994, 0.1261, 0.2359, 0.2252, 0.0929], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0192, 0.0201, 0.0181, 0.0209, 0.0210, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:12:45,500 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1936, 2.1346, 1.6947, 2.1145, 2.1488, 1.8841, 2.4602, 2.2044], + device='cuda:1'), covar=tensor([0.1267, 0.1868, 0.2896, 0.2266, 0.2385, 0.1630, 0.2682, 0.1533], + device='cuda:1'), in_proj_covar=tensor([0.0187, 0.0189, 0.0233, 0.0252, 0.0248, 0.0204, 0.0213, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:12:46,674 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139276.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:12:55,168 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139289.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:13:02,260 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139299.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:13:13,307 INFO [finetune.py:976] (1/7) Epoch 25, batch 1850, loss[loss=0.1914, simple_loss=0.2633, pruned_loss=0.05973, over 4815.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2476, pruned_loss=0.05213, over 954781.04 frames. ], batch size: 33, lr: 3.01e-03, grad_scale: 32.0 +2023-03-27 06:13:27,855 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139337.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:13:27,900 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139337.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 06:13:30,856 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139342.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:13:36,154 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139350.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:13:38,380 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139353.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:13:46,132 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.50 vs. limit=2.0 +2023-03-27 06:13:46,544 INFO [finetune.py:976] (1/7) Epoch 25, batch 1900, loss[loss=0.1697, simple_loss=0.2502, pruned_loss=0.04457, over 4862.00 frames. ], tot_loss[loss=0.1783, simple_loss=0.2502, pruned_loss=0.05325, over 952468.71 frames. ], batch size: 44, lr: 3.01e-03, grad_scale: 32.0 +2023-03-27 06:13:47,140 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.007e+02 1.553e+02 1.835e+02 2.150e+02 3.557e+02, threshold=3.671e+02, percent-clipped=0.0 +2023-03-27 06:13:57,337 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6212, 1.3941, 1.3469, 0.8702, 1.5151, 1.7023, 1.5858, 1.2989], + device='cuda:1'), covar=tensor([0.1049, 0.1036, 0.0704, 0.0627, 0.0582, 0.0711, 0.0508, 0.0950], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0128, 0.0122, 0.0130, 0.0129, 0.0141, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.9125e-05, 1.0626e-04, 9.1222e-05, 8.5930e-05, 9.1374e-05, 9.1886e-05, + 1.0069e-04, 1.0553e-04], device='cuda:1') +2023-03-27 06:13:59,662 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=139385.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:14:03,126 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=139390.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:14:05,032 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4115, 2.7079, 2.2664, 1.7494, 2.4571, 2.7491, 2.6839, 2.2069], + device='cuda:1'), covar=tensor([0.0606, 0.0520, 0.0731, 0.0814, 0.0736, 0.0662, 0.0596, 0.1025], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0137, 0.0140, 0.0120, 0.0126, 0.0139, 0.0139, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:14:10,499 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=139401.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:14:19,883 INFO [finetune.py:976] (1/7) Epoch 25, batch 1950, loss[loss=0.1377, simple_loss=0.2173, pruned_loss=0.02905, over 4813.00 frames. ], tot_loss[loss=0.1764, simple_loss=0.2482, pruned_loss=0.05228, over 952578.74 frames. ], batch size: 38, lr: 3.01e-03, grad_scale: 32.0 +2023-03-27 06:15:11,616 INFO [finetune.py:976] (1/7) Epoch 25, batch 2000, loss[loss=0.1819, simple_loss=0.248, pruned_loss=0.05796, over 4739.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2465, pruned_loss=0.05197, over 954814.85 frames. ], batch size: 54, lr: 3.01e-03, grad_scale: 32.0 +2023-03-27 06:15:12,708 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.253e+01 1.362e+02 1.721e+02 2.187e+02 3.038e+02, threshold=3.442e+02, percent-clipped=0.0 +2023-03-27 06:15:45,236 INFO [finetune.py:976] (1/7) Epoch 25, batch 2050, loss[loss=0.205, simple_loss=0.266, pruned_loss=0.07199, over 4934.00 frames. ], tot_loss[loss=0.172, simple_loss=0.243, pruned_loss=0.05051, over 955199.36 frames. ], batch size: 38, lr: 3.01e-03, grad_scale: 32.0 +2023-03-27 06:15:50,218 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-27 06:16:14,494 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5474, 2.8120, 2.4689, 1.8373, 2.7328, 2.9119, 2.9048, 2.4683], + device='cuda:1'), covar=tensor([0.0628, 0.0580, 0.0837, 0.0897, 0.0631, 0.0756, 0.0608, 0.1036], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0140, 0.0120, 0.0126, 0.0139, 0.0139, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:16:18,436 INFO [finetune.py:976] (1/7) Epoch 25, batch 2100, loss[loss=0.1911, simple_loss=0.2439, pruned_loss=0.06917, over 4892.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.2419, pruned_loss=0.05035, over 952229.37 frames. ], batch size: 32, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:16:18,536 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139565.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 06:16:20,121 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.352e+01 1.449e+02 1.714e+02 2.109e+02 3.824e+02, threshold=3.428e+02, percent-clipped=2.0 +2023-03-27 06:16:22,740 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139571.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:16:30,370 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139582.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:16:38,028 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139594.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:16:50,720 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=139613.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:16:52,382 INFO [finetune.py:976] (1/7) Epoch 25, batch 2150, loss[loss=0.1438, simple_loss=0.2098, pruned_loss=0.03886, over 4763.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2437, pruned_loss=0.05076, over 953382.68 frames. ], batch size: 27, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:17:09,456 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139632.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 06:17:09,507 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139632.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:17:21,675 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139643.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:17:27,567 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139645.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:17:43,824 INFO [finetune.py:976] (1/7) Epoch 25, batch 2200, loss[loss=0.2036, simple_loss=0.2762, pruned_loss=0.06548, over 4803.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2451, pruned_loss=0.05087, over 953663.59 frames. ], batch size: 41, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:17:45,447 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.088e+02 1.512e+02 1.789e+02 2.111e+02 3.462e+02, threshold=3.578e+02, percent-clipped=1.0 +2023-03-27 06:18:16,521 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3046, 3.7390, 3.9290, 4.1456, 4.1057, 3.7975, 4.4108, 1.2960], + device='cuda:1'), covar=tensor([0.0744, 0.0820, 0.0900, 0.0996, 0.1154, 0.1666, 0.0616, 0.6090], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0251, 0.0286, 0.0299, 0.0343, 0.0290, 0.0310, 0.0307], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:18:17,070 INFO [finetune.py:976] (1/7) Epoch 25, batch 2250, loss[loss=0.1692, simple_loss=0.2503, pruned_loss=0.04402, over 4840.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.246, pruned_loss=0.05077, over 955444.49 frames. ], batch size: 44, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:18:38,448 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.49 vs. limit=5.0 +2023-03-27 06:18:48,522 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2704, 2.2191, 1.9134, 2.1514, 2.0690, 2.0438, 2.1133, 2.7995], + device='cuda:1'), covar=tensor([0.3740, 0.3903, 0.3374, 0.3408, 0.3718, 0.2429, 0.3351, 0.1713], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0264, 0.0235, 0.0276, 0.0259, 0.0229, 0.0256, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:18:48,702 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-27 06:18:50,822 INFO [finetune.py:976] (1/7) Epoch 25, batch 2300, loss[loss=0.1906, simple_loss=0.264, pruned_loss=0.05861, over 4810.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2468, pruned_loss=0.05069, over 952762.66 frames. ], batch size: 41, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:18:52,009 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.130e+02 1.533e+02 1.822e+02 2.118e+02 3.916e+02, threshold=3.645e+02, percent-clipped=1.0 +2023-03-27 06:18:53,809 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139769.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:18:53,844 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7209, 1.0019, 1.7880, 1.7354, 1.5765, 1.5007, 1.6777, 1.6959], + device='cuda:1'), covar=tensor([0.3845, 0.3634, 0.3066, 0.3427, 0.4271, 0.3696, 0.3851, 0.2834], + device='cuda:1'), in_proj_covar=tensor([0.0262, 0.0245, 0.0265, 0.0291, 0.0291, 0.0267, 0.0296, 0.0249], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:19:06,794 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139789.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:19:06,817 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2920, 2.1931, 1.7342, 2.3593, 2.1625, 1.9713, 2.5532, 2.3495], + device='cuda:1'), covar=tensor([0.1318, 0.2178, 0.2859, 0.2639, 0.2490, 0.1605, 0.2917, 0.1772], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0191, 0.0237, 0.0256, 0.0251, 0.0206, 0.0216, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:19:15,043 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8745, 1.3854, 0.8542, 1.6695, 2.1537, 1.5631, 1.6661, 1.7636], + device='cuda:1'), covar=tensor([0.1427, 0.1934, 0.1942, 0.1175, 0.1928, 0.1813, 0.1411, 0.1845], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0109, 0.0092, 0.0119, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 06:19:16,886 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139804.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:19:18,107 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139806.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:19:23,931 INFO [finetune.py:976] (1/7) Epoch 25, batch 2350, loss[loss=0.2024, simple_loss=0.2609, pruned_loss=0.07191, over 4823.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2445, pruned_loss=0.05001, over 954116.63 frames. ], batch size: 30, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:19:35,053 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139830.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:19:44,204 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2265, 3.7602, 3.8956, 4.0793, 4.0363, 3.7888, 4.3736, 1.5020], + device='cuda:1'), covar=tensor([0.0865, 0.0855, 0.0963, 0.1077, 0.1175, 0.1509, 0.0733, 0.5544], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0250, 0.0285, 0.0298, 0.0341, 0.0288, 0.0309, 0.0305], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:19:54,878 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139850.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:19:54,887 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139850.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:20:07,500 INFO [finetune.py:976] (1/7) Epoch 25, batch 2400, loss[loss=0.1568, simple_loss=0.231, pruned_loss=0.0413, over 4740.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2417, pruned_loss=0.04942, over 954725.55 frames. ], batch size: 59, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:20:07,633 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139865.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:20:10,609 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.173e+01 1.419e+02 1.768e+02 2.081e+02 3.267e+02, threshold=3.536e+02, percent-clipped=0.0 +2023-03-27 06:20:10,739 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139867.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:20:22,870 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-27 06:20:35,935 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139894.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:20:47,286 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139911.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:20:50,036 INFO [finetune.py:976] (1/7) Epoch 25, batch 2450, loss[loss=0.2222, simple_loss=0.2826, pruned_loss=0.08091, over 4804.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2393, pruned_loss=0.0488, over 955769.95 frames. ], batch size: 40, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:20:57,352 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139927.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:21:01,383 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139932.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 06:21:05,445 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139938.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:21:07,908 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-27 06:21:08,384 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=139942.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:21:10,175 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139945.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:21:15,591 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1047, 2.2475, 1.8077, 2.2302, 2.1107, 2.0871, 2.1655, 2.9003], + device='cuda:1'), covar=tensor([0.3826, 0.4304, 0.3435, 0.4154, 0.4214, 0.2516, 0.3906, 0.1753], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0262, 0.0234, 0.0275, 0.0258, 0.0229, 0.0255, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:21:22,674 INFO [finetune.py:976] (1/7) Epoch 25, batch 2500, loss[loss=0.1853, simple_loss=0.2642, pruned_loss=0.05323, over 4822.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2411, pruned_loss=0.05021, over 954784.39 frames. ], batch size: 45, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:21:24,362 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.029e+02 1.523e+02 1.884e+02 2.422e+02 3.755e+02, threshold=3.768e+02, percent-clipped=3.0 +2023-03-27 06:21:24,565 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-27 06:21:32,355 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=139980.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:21:40,971 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5944, 1.4929, 1.4309, 1.5368, 1.2456, 3.4579, 1.3172, 1.8256], + device='cuda:1'), covar=tensor([0.3328, 0.2504, 0.2220, 0.2438, 0.1588, 0.0191, 0.2548, 0.1187], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0123, 0.0112, 0.0096, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 06:21:42,124 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=139993.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:21:57,547 INFO [finetune.py:976] (1/7) Epoch 25, batch 2550, loss[loss=0.2445, simple_loss=0.3112, pruned_loss=0.08891, over 4181.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.2457, pruned_loss=0.05193, over 953738.42 frames. ], batch size: 65, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:22:09,997 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4280, 2.4046, 2.0273, 1.1324, 2.1407, 1.8778, 1.7199, 2.2012], + device='cuda:1'), covar=tensor([0.1142, 0.0720, 0.1758, 0.2051, 0.1665, 0.2356, 0.2272, 0.1036], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0191, 0.0199, 0.0180, 0.0209, 0.0210, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:22:36,077 INFO [finetune.py:976] (1/7) Epoch 25, batch 2600, loss[loss=0.1582, simple_loss=0.2298, pruned_loss=0.04331, over 4782.00 frames. ], tot_loss[loss=0.1755, simple_loss=0.2472, pruned_loss=0.05197, over 954300.60 frames. ], batch size: 29, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:22:42,048 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.019e+02 1.547e+02 1.969e+02 2.320e+02 4.703e+02, threshold=3.938e+02, percent-clipped=1.0 +2023-03-27 06:23:14,131 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140103.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:23:19,354 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140110.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:23:22,259 INFO [finetune.py:976] (1/7) Epoch 25, batch 2650, loss[loss=0.1696, simple_loss=0.2465, pruned_loss=0.04641, over 4787.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2481, pruned_loss=0.05179, over 954770.26 frames. ], batch size: 29, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:23:28,790 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140125.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:23:41,811 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140145.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:23:45,115 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7697, 3.8582, 3.5944, 1.8876, 3.9898, 2.9813, 1.0999, 2.7521], + device='cuda:1'), covar=tensor([0.2504, 0.1863, 0.1450, 0.3184, 0.1001, 0.0948, 0.3951, 0.1346], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0179, 0.0161, 0.0131, 0.0161, 0.0124, 0.0149, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 06:23:51,798 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140160.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:23:53,487 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140162.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:23:55,119 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140164.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:23:55,632 INFO [finetune.py:976] (1/7) Epoch 25, batch 2700, loss[loss=0.1784, simple_loss=0.2576, pruned_loss=0.0496, over 4817.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2461, pruned_loss=0.05039, over 956488.54 frames. ], batch size: 40, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:23:56,850 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.003e+02 1.476e+02 1.708e+02 2.136e+02 4.297e+02, threshold=3.417e+02, percent-clipped=1.0 +2023-03-27 06:23:59,423 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140171.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:24:22,855 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140206.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:24:28,674 INFO [finetune.py:976] (1/7) Epoch 25, batch 2750, loss[loss=0.1533, simple_loss=0.2193, pruned_loss=0.04362, over 4783.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2433, pruned_loss=0.04945, over 955450.15 frames. ], batch size: 25, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:24:36,562 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140227.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:24:43,686 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140238.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:24:58,270 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140259.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:25:01,787 INFO [finetune.py:976] (1/7) Epoch 25, batch 2800, loss[loss=0.1559, simple_loss=0.2124, pruned_loss=0.04967, over 4273.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2405, pruned_loss=0.04862, over 954669.68 frames. ], batch size: 18, lr: 3.01e-03, grad_scale: 16.0 +2023-03-27 06:25:02,939 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.021e+02 1.479e+02 1.751e+02 2.221e+02 3.486e+02, threshold=3.502e+02, percent-clipped=1.0 +2023-03-27 06:25:06,561 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-27 06:25:10,750 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140275.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:25:22,267 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140286.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:25:53,836 INFO [finetune.py:976] (1/7) Epoch 25, batch 2850, loss[loss=0.1407, simple_loss=0.2128, pruned_loss=0.03427, over 4760.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2392, pruned_loss=0.04844, over 953743.65 frames. ], batch size: 23, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:25:56,990 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140320.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:26:27,553 INFO [finetune.py:976] (1/7) Epoch 25, batch 2900, loss[loss=0.1916, simple_loss=0.2745, pruned_loss=0.05436, over 4285.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2419, pruned_loss=0.04955, over 951727.32 frames. ], batch size: 65, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:26:28,760 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.105e+02 1.583e+02 1.866e+02 2.190e+02 4.311e+02, threshold=3.732e+02, percent-clipped=1.0 +2023-03-27 06:26:41,988 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140387.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:27:01,453 INFO [finetune.py:976] (1/7) Epoch 25, batch 2950, loss[loss=0.1726, simple_loss=0.246, pruned_loss=0.04965, over 4790.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2451, pruned_loss=0.05092, over 951310.93 frames. ], batch size: 51, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:27:07,556 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140425.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:27:14,043 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9025, 1.3480, 0.8544, 1.6451, 2.1665, 1.3362, 1.6215, 1.6723], + device='cuda:1'), covar=tensor([0.1421, 0.2130, 0.1943, 0.1205, 0.1966, 0.2094, 0.1415, 0.1991], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0109, 0.0092, 0.0119, 0.0093, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 06:27:20,073 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0659, 1.9647, 1.8043, 2.2302, 2.5458, 2.0976, 2.0506, 1.6208], + device='cuda:1'), covar=tensor([0.2228, 0.2031, 0.1970, 0.1619, 0.1761, 0.1204, 0.2081, 0.1959], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0212, 0.0215, 0.0198, 0.0245, 0.0191, 0.0217, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:27:21,216 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140445.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:27:23,042 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140448.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:27:30,138 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140459.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:27:30,778 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140460.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:27:32,482 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140462.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:27:32,517 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1153, 1.9919, 1.5412, 0.6587, 1.7086, 1.7479, 1.6202, 1.8527], + device='cuda:1'), covar=tensor([0.0893, 0.0778, 0.1462, 0.1977, 0.1307, 0.2625, 0.2366, 0.0878], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0192, 0.0200, 0.0181, 0.0210, 0.0211, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:27:34,685 INFO [finetune.py:976] (1/7) Epoch 25, batch 3000, loss[loss=0.2317, simple_loss=0.2883, pruned_loss=0.08751, over 4334.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2469, pruned_loss=0.05151, over 950035.47 frames. ], batch size: 19, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:27:34,685 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 06:27:48,790 INFO [finetune.py:1010] (1/7) Epoch 25, validation: loss=0.1571, simple_loss=0.2254, pruned_loss=0.04443, over 2265189.00 frames. +2023-03-27 06:27:48,790 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 06:27:49,499 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140466.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:27:50,499 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.023e+02 1.568e+02 1.888e+02 2.214e+02 4.503e+02, threshold=3.776e+02, percent-clipped=3.0 +2023-03-27 06:27:59,537 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140473.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:28:14,371 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140489.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:28:20,344 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140493.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:28:29,170 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140506.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:28:30,340 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140508.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:28:31,000 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140509.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:28:31,579 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140510.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:28:33,997 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7662, 3.7475, 3.6243, 1.5586, 3.9330, 2.9303, 0.9706, 2.8179], + device='cuda:1'), covar=tensor([0.2137, 0.1789, 0.1553, 0.3837, 0.0944, 0.0873, 0.4505, 0.1319], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0178, 0.0162, 0.0130, 0.0161, 0.0124, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 06:28:34,546 INFO [finetune.py:976] (1/7) Epoch 25, batch 3050, loss[loss=0.1441, simple_loss=0.2185, pruned_loss=0.03483, over 4732.00 frames. ], tot_loss[loss=0.1754, simple_loss=0.2475, pruned_loss=0.05161, over 952370.64 frames. ], batch size: 23, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:28:58,040 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7863, 1.2245, 1.8521, 1.8210, 1.6054, 1.5560, 1.7494, 1.7775], + device='cuda:1'), covar=tensor([0.3895, 0.3798, 0.3125, 0.3411, 0.4505, 0.3754, 0.4186, 0.2919], + device='cuda:1'), in_proj_covar=tensor([0.0264, 0.0246, 0.0266, 0.0292, 0.0292, 0.0268, 0.0298, 0.0249], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:28:58,632 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140550.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:29:00,939 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140554.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:29:03,319 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3560, 2.4115, 2.0303, 2.4924, 2.2998, 2.3268, 2.2498, 3.1924], + device='cuda:1'), covar=tensor([0.3922, 0.4656, 0.3368, 0.4121, 0.4344, 0.2470, 0.4414, 0.1585], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0264, 0.0236, 0.0277, 0.0260, 0.0230, 0.0257, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:29:08,052 INFO [finetune.py:976] (1/7) Epoch 25, batch 3100, loss[loss=0.1295, simple_loss=0.2058, pruned_loss=0.02657, over 4796.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2453, pruned_loss=0.05088, over 953618.89 frames. ], batch size: 29, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:29:09,242 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.313e+01 1.495e+02 1.767e+02 2.180e+02 4.499e+02, threshold=3.535e+02, percent-clipped=1.0 +2023-03-27 06:29:12,209 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140570.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:29:15,091 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.8612, 4.2765, 4.4518, 4.6807, 4.6331, 4.3588, 5.0078, 1.5772], + device='cuda:1'), covar=tensor([0.0819, 0.0861, 0.0881, 0.0997, 0.1233, 0.1524, 0.0621, 0.6257], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0249, 0.0284, 0.0298, 0.0339, 0.0288, 0.0309, 0.0304], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:29:42,051 INFO [finetune.py:976] (1/7) Epoch 25, batch 3150, loss[loss=0.1517, simple_loss=0.2206, pruned_loss=0.04135, over 4734.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2432, pruned_loss=0.05018, over 952759.20 frames. ], batch size: 59, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:29:42,119 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140615.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:30:15,048 INFO [finetune.py:976] (1/7) Epoch 25, batch 3200, loss[loss=0.1476, simple_loss=0.2276, pruned_loss=0.03377, over 4832.00 frames. ], tot_loss[loss=0.169, simple_loss=0.2399, pruned_loss=0.04907, over 953611.97 frames. ], batch size: 30, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:30:16,220 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.929e+01 1.486e+02 1.750e+02 2.144e+02 4.466e+02, threshold=3.500e+02, percent-clipped=2.0 +2023-03-27 06:30:40,177 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8619, 1.8306, 1.5512, 1.8038, 2.2345, 2.2585, 1.8010, 1.5985], + device='cuda:1'), covar=tensor([0.0267, 0.0301, 0.0623, 0.0308, 0.0190, 0.0325, 0.0318, 0.0401], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0106, 0.0144, 0.0110, 0.0099, 0.0113, 0.0102, 0.0111], + device='cuda:1'), out_proj_covar=tensor([7.6708e-05, 8.0870e-05, 1.1224e-04, 8.4132e-05, 7.6810e-05, 8.3152e-05, + 7.5465e-05, 8.4219e-05], device='cuda:1') +2023-03-27 06:31:06,559 INFO [finetune.py:976] (1/7) Epoch 25, batch 3250, loss[loss=0.1827, simple_loss=0.2652, pruned_loss=0.05007, over 4831.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2418, pruned_loss=0.05015, over 953290.40 frames. ], batch size: 49, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:31:09,743 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-27 06:31:11,486 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140723.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:31:12,225 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-27 06:31:20,946 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4394, 1.2868, 1.1698, 1.4326, 1.5096, 1.5201, 0.9382, 1.2115], + device='cuda:1'), covar=tensor([0.2273, 0.2218, 0.2083, 0.1835, 0.1631, 0.1294, 0.2658, 0.2041], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0210, 0.0214, 0.0197, 0.0244, 0.0190, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:31:25,160 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140743.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:31:35,355 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140759.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:31:39,361 INFO [finetune.py:976] (1/7) Epoch 25, batch 3300, loss[loss=0.1719, simple_loss=0.2544, pruned_loss=0.04468, over 4909.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2455, pruned_loss=0.05131, over 954397.21 frames. ], batch size: 36, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:31:40,535 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140766.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:31:41,064 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.042e+02 1.629e+02 1.945e+02 2.397e+02 4.021e+02, threshold=3.889e+02, percent-clipped=5.0 +2023-03-27 06:31:53,077 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140784.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:32:08,086 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140807.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:32:09,343 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140809.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:32:12,393 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140814.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:32:12,930 INFO [finetune.py:976] (1/7) Epoch 25, batch 3350, loss[loss=0.2063, simple_loss=0.2768, pruned_loss=0.0679, over 4825.00 frames. ], tot_loss[loss=0.1769, simple_loss=0.2488, pruned_loss=0.0525, over 956715.84 frames. ], batch size: 30, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:32:33,951 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140845.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:32:46,606 INFO [finetune.py:976] (1/7) Epoch 25, batch 3400, loss[loss=0.2135, simple_loss=0.2867, pruned_loss=0.07016, over 4823.00 frames. ], tot_loss[loss=0.1758, simple_loss=0.2483, pruned_loss=0.05171, over 955789.73 frames. ], batch size: 39, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:32:46,670 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140865.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:32:47,797 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.030e+02 1.564e+02 1.878e+02 2.236e+02 3.278e+02, threshold=3.756e+02, percent-clipped=0.0 +2023-03-27 06:32:49,696 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140870.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:33:39,160 INFO [finetune.py:976] (1/7) Epoch 25, batch 3450, loss[loss=0.1303, simple_loss=0.2031, pruned_loss=0.02876, over 4823.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2477, pruned_loss=0.05121, over 956279.59 frames. ], batch size: 33, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:33:39,279 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140915.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:33:40,471 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140917.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:33:45,361 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1885, 2.1540, 1.9275, 2.2625, 2.1666, 2.1535, 2.1008, 2.8881], + device='cuda:1'), covar=tensor([0.3574, 0.4844, 0.3510, 0.4133, 0.4467, 0.2400, 0.4622, 0.1529], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0263, 0.0236, 0.0277, 0.0259, 0.0229, 0.0256, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:34:01,691 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140947.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:34:09,592 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3081, 1.6077, 1.6756, 0.9063, 1.6156, 1.9049, 1.8872, 1.4996], + device='cuda:1'), covar=tensor([0.0860, 0.0612, 0.0549, 0.0501, 0.0493, 0.0577, 0.0375, 0.0625], + device='cuda:1'), in_proj_covar=tensor([0.0123, 0.0149, 0.0128, 0.0123, 0.0131, 0.0130, 0.0142, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.9712e-05, 1.0688e-04, 9.1227e-05, 8.6412e-05, 9.1595e-05, 9.2132e-05, + 1.0092e-04, 1.0589e-04], device='cuda:1') +2023-03-27 06:34:11,812 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=140963.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:34:12,942 INFO [finetune.py:976] (1/7) Epoch 25, batch 3500, loss[loss=0.151, simple_loss=0.2191, pruned_loss=0.0414, over 4927.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2447, pruned_loss=0.05048, over 957399.04 frames. ], batch size: 38, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:34:14,173 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.063e+02 1.468e+02 1.748e+02 2.204e+02 3.629e+02, threshold=3.496e+02, percent-clipped=0.0 +2023-03-27 06:34:20,900 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140978.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:34:41,924 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141008.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:34:46,048 INFO [finetune.py:976] (1/7) Epoch 25, batch 3550, loss[loss=0.1683, simple_loss=0.2397, pruned_loss=0.04842, over 4905.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2433, pruned_loss=0.05029, over 958582.83 frames. ], batch size: 35, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:35:04,133 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141043.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:35:04,784 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8593, 1.8336, 1.9781, 1.1935, 1.9483, 1.9706, 1.9782, 1.5793], + device='cuda:1'), covar=tensor([0.0630, 0.0640, 0.0656, 0.0895, 0.0749, 0.0656, 0.0582, 0.1175], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0136, 0.0140, 0.0119, 0.0126, 0.0137, 0.0139, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:35:19,345 INFO [finetune.py:976] (1/7) Epoch 25, batch 3600, loss[loss=0.2014, simple_loss=0.2696, pruned_loss=0.06658, over 4831.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2404, pruned_loss=0.04965, over 956610.53 frames. ], batch size: 33, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:35:20,525 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.001e+02 1.456e+02 1.796e+02 2.356e+02 3.995e+02, threshold=3.592e+02, percent-clipped=1.0 +2023-03-27 06:35:28,419 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141079.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:35:36,623 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=141091.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:35:49,608 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8231, 2.5571, 2.4067, 1.4005, 2.6107, 2.0018, 1.8228, 2.3872], + device='cuda:1'), covar=tensor([0.1085, 0.0720, 0.1533, 0.1872, 0.1375, 0.2120, 0.2188, 0.0930], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0189, 0.0198, 0.0180, 0.0208, 0.0209, 0.0222, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:35:57,495 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6451, 1.5576, 1.3266, 1.6524, 2.0851, 1.7230, 1.3725, 1.3787], + device='cuda:1'), covar=tensor([0.2071, 0.1829, 0.1918, 0.1603, 0.1476, 0.1187, 0.2294, 0.1777], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0211, 0.0214, 0.0197, 0.0245, 0.0191, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:35:58,692 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8665, 1.4007, 0.8039, 1.6045, 2.1322, 1.3117, 1.5471, 1.6362], + device='cuda:1'), covar=tensor([0.1393, 0.1958, 0.1813, 0.1172, 0.1856, 0.1822, 0.1396, 0.1858], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0093, 0.0109, 0.0092, 0.0119, 0.0093, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 06:36:00,469 INFO [finetune.py:976] (1/7) Epoch 25, batch 3650, loss[loss=0.159, simple_loss=0.2384, pruned_loss=0.03984, over 4798.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2429, pruned_loss=0.0506, over 956318.16 frames. ], batch size: 29, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:36:00,584 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1469, 1.6805, 2.6668, 1.6632, 2.2956, 2.3592, 1.5397, 2.5078], + device='cuda:1'), covar=tensor([0.1452, 0.2493, 0.1334, 0.2008, 0.0998, 0.1645, 0.2990, 0.1122], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0208, 0.0192, 0.0191, 0.0175, 0.0214, 0.0217, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:36:15,746 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1037, 1.9994, 2.1591, 1.4666, 2.0053, 2.2159, 2.1804, 1.6184], + device='cuda:1'), covar=tensor([0.0579, 0.0644, 0.0701, 0.0867, 0.0725, 0.0615, 0.0569, 0.1224], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0141, 0.0120, 0.0127, 0.0139, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:36:30,909 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141143.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:36:32,113 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141145.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:36:33,912 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5478, 3.4914, 3.2578, 1.6964, 3.6624, 2.7674, 1.0009, 2.5191], + device='cuda:1'), covar=tensor([0.3113, 0.2270, 0.1805, 0.3602, 0.1082, 0.1078, 0.4205, 0.1617], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0179, 0.0162, 0.0131, 0.0161, 0.0124, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 06:36:46,081 INFO [finetune.py:976] (1/7) Epoch 25, batch 3700, loss[loss=0.1704, simple_loss=0.2529, pruned_loss=0.04389, over 4843.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2458, pruned_loss=0.05127, over 954671.23 frames. ], batch size: 47, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:36:46,155 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141165.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:36:46,175 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141165.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:36:47,287 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.158e+02 1.708e+02 2.029e+02 2.382e+02 3.628e+02, threshold=4.058e+02, percent-clipped=1.0 +2023-03-27 06:37:03,879 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=141193.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:37:03,934 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9460, 1.9439, 2.0556, 1.3812, 1.9416, 2.0793, 2.0475, 1.5747], + device='cuda:1'), covar=tensor([0.0548, 0.0557, 0.0606, 0.0762, 0.0672, 0.0603, 0.0530, 0.1092], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0141, 0.0120, 0.0127, 0.0139, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:37:11,230 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141204.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:37:15,098 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3467, 3.7744, 3.9603, 4.2007, 4.1419, 3.8880, 4.4272, 1.2939], + device='cuda:1'), covar=tensor([0.0799, 0.0866, 0.0851, 0.0940, 0.1154, 0.1381, 0.0706, 0.5932], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0248, 0.0284, 0.0297, 0.0339, 0.0288, 0.0308, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:37:18,113 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=141213.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:37:19,742 INFO [finetune.py:976] (1/7) Epoch 25, batch 3750, loss[loss=0.1956, simple_loss=0.2758, pruned_loss=0.05775, over 4820.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2465, pruned_loss=0.05144, over 955072.54 frames. ], batch size: 33, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:37:45,906 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9946, 2.7070, 2.8142, 2.9296, 2.8346, 2.5844, 3.0258, 0.8541], + device='cuda:1'), covar=tensor([0.1065, 0.0928, 0.1046, 0.1035, 0.1569, 0.1725, 0.0990, 0.5277], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0248, 0.0283, 0.0296, 0.0339, 0.0288, 0.0308, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:37:52,662 INFO [finetune.py:976] (1/7) Epoch 25, batch 3800, loss[loss=0.193, simple_loss=0.269, pruned_loss=0.05852, over 4817.00 frames. ], tot_loss[loss=0.1759, simple_loss=0.2481, pruned_loss=0.05183, over 956578.01 frames. ], batch size: 40, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:37:54,343 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.018e+02 1.508e+02 1.827e+02 2.217e+02 6.513e+02, threshold=3.654e+02, percent-clipped=2.0 +2023-03-27 06:37:58,031 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141273.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:38:19,446 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141303.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:38:32,077 INFO [finetune.py:976] (1/7) Epoch 25, batch 3850, loss[loss=0.1843, simple_loss=0.2536, pruned_loss=0.0575, over 4853.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.246, pruned_loss=0.0508, over 955766.15 frames. ], batch size: 44, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:38:49,948 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141329.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:38:50,632 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.00 vs. limit=5.0 +2023-03-27 06:39:16,920 INFO [finetune.py:976] (1/7) Epoch 25, batch 3900, loss[loss=0.1479, simple_loss=0.212, pruned_loss=0.04192, over 4293.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.243, pruned_loss=0.05005, over 955130.85 frames. ], batch size: 19, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:39:18,104 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.088e+02 1.504e+02 1.773e+02 2.110e+02 6.012e+02, threshold=3.546e+02, percent-clipped=1.0 +2023-03-27 06:39:26,389 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141379.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:39:33,598 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141390.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:39:39,070 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8242, 4.0729, 3.8247, 1.9574, 4.1207, 3.1244, 1.0378, 2.9217], + device='cuda:1'), covar=tensor([0.2208, 0.1777, 0.1564, 0.3462, 0.1016, 0.0914, 0.4440, 0.1449], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0178, 0.0161, 0.0130, 0.0160, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 06:39:49,666 INFO [finetune.py:976] (1/7) Epoch 25, batch 3950, loss[loss=0.2068, simple_loss=0.265, pruned_loss=0.07432, over 4822.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2395, pruned_loss=0.04897, over 953583.47 frames. ], batch size: 30, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:39:51,020 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141417.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 06:39:58,916 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=141427.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:40:07,838 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141441.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:40:23,341 INFO [finetune.py:976] (1/7) Epoch 25, batch 4000, loss[loss=0.1908, simple_loss=0.2633, pruned_loss=0.05915, over 4780.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2409, pruned_loss=0.04992, over 953316.25 frames. ], batch size: 54, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:40:23,433 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141465.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:40:24,523 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.071e+02 1.514e+02 1.750e+02 2.113e+02 3.817e+02, threshold=3.500e+02, percent-clipped=1.0 +2023-03-27 06:40:33,172 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141478.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 06:40:46,342 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141499.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:40:48,236 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141502.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:40:51,772 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9209, 1.8810, 1.7372, 1.9546, 1.4835, 4.6342, 1.7824, 2.1523], + device='cuda:1'), covar=tensor([0.3118, 0.2354, 0.2022, 0.2261, 0.1591, 0.0122, 0.2494, 0.1207], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0123, 0.0112, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 06:40:55,307 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=141513.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:40:56,982 INFO [finetune.py:976] (1/7) Epoch 25, batch 4050, loss[loss=0.1446, simple_loss=0.2255, pruned_loss=0.0319, over 4753.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2427, pruned_loss=0.05017, over 953105.09 frames. ], batch size: 26, lr: 3.00e-03, grad_scale: 16.0 +2023-03-27 06:41:07,963 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7677, 1.5972, 1.9578, 3.4995, 2.3892, 2.4444, 1.0443, 2.8876], + device='cuda:1'), covar=tensor([0.1548, 0.1237, 0.1372, 0.0581, 0.0694, 0.1455, 0.1643, 0.0423], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0163, 0.0101, 0.0136, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 06:41:49,012 INFO [finetune.py:976] (1/7) Epoch 25, batch 4100, loss[loss=0.1841, simple_loss=0.2654, pruned_loss=0.05136, over 4895.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.246, pruned_loss=0.05139, over 953783.80 frames. ], batch size: 43, lr: 3.00e-03, grad_scale: 32.0 +2023-03-27 06:41:50,186 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.144e+02 1.605e+02 1.887e+02 2.173e+02 5.231e+02, threshold=3.774e+02, percent-clipped=3.0 +2023-03-27 06:41:54,377 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141573.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:42:14,756 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141603.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:42:22,402 INFO [finetune.py:976] (1/7) Epoch 25, batch 4150, loss[loss=0.1679, simple_loss=0.2481, pruned_loss=0.04386, over 4812.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2467, pruned_loss=0.05155, over 954625.14 frames. ], batch size: 40, lr: 3.00e-03, grad_scale: 32.0 +2023-03-27 06:42:26,085 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=141621.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:42:46,647 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=141651.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:42:52,746 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0 +2023-03-27 06:42:55,969 INFO [finetune.py:976] (1/7) Epoch 25, batch 4200, loss[loss=0.1939, simple_loss=0.2719, pruned_loss=0.05798, over 4749.00 frames. ], tot_loss[loss=0.1757, simple_loss=0.2477, pruned_loss=0.05184, over 953881.78 frames. ], batch size: 54, lr: 3.00e-03, grad_scale: 32.0 +2023-03-27 06:42:57,194 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.937e+01 1.553e+02 1.832e+02 2.223e+02 5.119e+02, threshold=3.664e+02, percent-clipped=2.0 +2023-03-27 06:43:09,606 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141685.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:43:29,319 INFO [finetune.py:976] (1/7) Epoch 25, batch 4250, loss[loss=0.1917, simple_loss=0.2538, pruned_loss=0.0648, over 4816.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2448, pruned_loss=0.05087, over 955029.66 frames. ], batch size: 39, lr: 3.00e-03, grad_scale: 32.0 +2023-03-27 06:43:38,445 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6517, 1.5167, 1.9481, 3.2979, 2.1984, 2.3795, 0.9441, 2.7145], + device='cuda:1'), covar=tensor([0.1664, 0.1284, 0.1275, 0.0496, 0.0777, 0.1272, 0.1752, 0.0445], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0163, 0.0101, 0.0136, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 06:43:59,174 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6237, 3.4948, 3.3710, 1.5141, 3.6286, 2.7538, 0.9726, 2.5512], + device='cuda:1'), covar=tensor([0.2451, 0.1713, 0.1509, 0.3273, 0.1104, 0.0903, 0.3774, 0.1473], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0178, 0.0160, 0.0130, 0.0160, 0.0123, 0.0147, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 06:44:21,245 INFO [finetune.py:976] (1/7) Epoch 25, batch 4300, loss[loss=0.1612, simple_loss=0.2376, pruned_loss=0.04243, over 4765.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2426, pruned_loss=0.05034, over 956524.97 frames. ], batch size: 27, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:44:22,428 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.814e+01 1.357e+02 1.630e+02 2.025e+02 3.929e+02, threshold=3.260e+02, percent-clipped=1.0 +2023-03-27 06:44:26,617 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141773.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 06:44:43,604 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141797.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:44:44,842 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141799.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:44:55,201 INFO [finetune.py:976] (1/7) Epoch 25, batch 4350, loss[loss=0.1356, simple_loss=0.2133, pruned_loss=0.02898, over 4772.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2385, pruned_loss=0.04858, over 956088.59 frames. ], batch size: 28, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:45:09,179 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-03-27 06:45:17,467 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=141847.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:45:28,845 INFO [finetune.py:976] (1/7) Epoch 25, batch 4400, loss[loss=0.2172, simple_loss=0.2923, pruned_loss=0.07105, over 4804.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2408, pruned_loss=0.05024, over 955044.35 frames. ], batch size: 45, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:45:30,033 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.038e+02 1.595e+02 1.814e+02 2.202e+02 4.275e+02, threshold=3.628e+02, percent-clipped=6.0 +2023-03-27 06:45:43,806 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3561, 3.3488, 3.1289, 1.5038, 3.4364, 2.5900, 0.9087, 2.3420], + device='cuda:1'), covar=tensor([0.2614, 0.2180, 0.1864, 0.3396, 0.1326, 0.1012, 0.4132, 0.1528], + device='cuda:1'), in_proj_covar=tensor([0.0149, 0.0176, 0.0159, 0.0129, 0.0159, 0.0122, 0.0146, 0.0123], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 06:45:48,455 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2289, 1.9931, 1.4304, 0.6546, 1.6591, 1.7761, 1.6741, 1.7988], + device='cuda:1'), covar=tensor([0.0765, 0.0756, 0.1493, 0.1930, 0.1191, 0.2343, 0.2097, 0.0824], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0192, 0.0200, 0.0181, 0.0210, 0.0212, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:46:01,881 INFO [finetune.py:976] (1/7) Epoch 25, batch 4450, loss[loss=0.1582, simple_loss=0.2319, pruned_loss=0.04225, over 4752.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2442, pruned_loss=0.0511, over 956653.08 frames. ], batch size: 28, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:46:02,592 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141916.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:46:10,451 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141928.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:46:46,899 INFO [finetune.py:976] (1/7) Epoch 25, batch 4500, loss[loss=0.1742, simple_loss=0.2494, pruned_loss=0.04951, over 4918.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2448, pruned_loss=0.05102, over 955041.32 frames. ], batch size: 42, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:46:52,641 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.869e+01 1.594e+02 1.826e+02 2.236e+02 4.959e+02, threshold=3.653e+02, percent-clipped=2.0 +2023-03-27 06:47:02,917 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141977.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:47:08,154 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141985.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:47:10,667 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141989.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:47:27,104 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1387, 2.0080, 2.1359, 1.3396, 1.9775, 2.0627, 2.1585, 1.6736], + device='cuda:1'), covar=tensor([0.0579, 0.0668, 0.0756, 0.1016, 0.0792, 0.0765, 0.0655, 0.1178], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0136, 0.0140, 0.0119, 0.0126, 0.0137, 0.0138, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:47:30,042 INFO [finetune.py:976] (1/7) Epoch 25, batch 4550, loss[loss=0.1516, simple_loss=0.2286, pruned_loss=0.03729, over 4742.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2461, pruned_loss=0.05088, over 955652.10 frames. ], batch size: 59, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:47:41,389 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=142033.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:48:03,343 INFO [finetune.py:976] (1/7) Epoch 25, batch 4600, loss[loss=0.1574, simple_loss=0.2375, pruned_loss=0.03864, over 4864.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2451, pruned_loss=0.05013, over 955413.34 frames. ], batch size: 34, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:48:04,586 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 7.716e+01 1.569e+02 1.799e+02 2.271e+02 3.318e+02, threshold=3.598e+02, percent-clipped=0.0 +2023-03-27 06:48:08,721 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=142073.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 06:48:15,506 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-27 06:48:23,761 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=142097.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:48:36,594 INFO [finetune.py:976] (1/7) Epoch 25, batch 4650, loss[loss=0.1768, simple_loss=0.2379, pruned_loss=0.05782, over 4253.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2431, pruned_loss=0.05003, over 955261.04 frames. ], batch size: 65, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:48:40,332 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=142121.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 06:48:57,213 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=142145.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:49:19,997 INFO [finetune.py:976] (1/7) Epoch 25, batch 4700, loss[loss=0.1601, simple_loss=0.2226, pruned_loss=0.04877, over 4870.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2405, pruned_loss=0.04937, over 955536.97 frames. ], batch size: 31, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:49:21,185 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.163e+01 1.384e+02 1.765e+02 2.088e+02 3.764e+02, threshold=3.531e+02, percent-clipped=1.0 +2023-03-27 06:50:00,898 INFO [finetune.py:976] (1/7) Epoch 25, batch 4750, loss[loss=0.1713, simple_loss=0.2387, pruned_loss=0.052, over 4933.00 frames. ], tot_loss[loss=0.169, simple_loss=0.2391, pruned_loss=0.0494, over 953348.65 frames. ], batch size: 33, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:50:34,332 INFO [finetune.py:976] (1/7) Epoch 25, batch 4800, loss[loss=0.1824, simple_loss=0.2337, pruned_loss=0.06559, over 4466.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2419, pruned_loss=0.05016, over 954335.25 frames. ], batch size: 19, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:50:35,544 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.947e+01 1.535e+02 1.762e+02 2.238e+02 3.446e+02, threshold=3.524e+02, percent-clipped=1.0 +2023-03-27 06:50:39,629 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=142272.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:50:47,477 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=142284.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:51:07,061 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0305, 1.9339, 1.6315, 1.6653, 1.8344, 1.7858, 1.8194, 2.4973], + device='cuda:1'), covar=tensor([0.3810, 0.3883, 0.3312, 0.3822, 0.3782, 0.2435, 0.3486, 0.1686], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0235, 0.0275, 0.0258, 0.0228, 0.0256, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:51:07,516 INFO [finetune.py:976] (1/7) Epoch 25, batch 4850, loss[loss=0.1993, simple_loss=0.2685, pruned_loss=0.06504, over 4866.00 frames. ], tot_loss[loss=0.1738, simple_loss=0.2456, pruned_loss=0.051, over 955255.05 frames. ], batch size: 34, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:51:11,728 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2558, 1.7872, 2.4128, 4.3192, 3.0219, 2.8625, 0.9900, 3.6262], + device='cuda:1'), covar=tensor([0.1707, 0.1516, 0.1587, 0.0475, 0.0695, 0.1350, 0.1936, 0.0401], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0135, 0.0165, 0.0102, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 06:51:39,151 INFO [finetune.py:976] (1/7) Epoch 25, batch 4900, loss[loss=0.2232, simple_loss=0.2813, pruned_loss=0.0825, over 4823.00 frames. ], tot_loss[loss=0.1744, simple_loss=0.2463, pruned_loss=0.05125, over 954487.47 frames. ], batch size: 49, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:51:40,866 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.096e+02 1.551e+02 1.812e+02 2.135e+02 6.918e+02, threshold=3.624e+02, percent-clipped=2.0 +2023-03-27 06:52:31,143 INFO [finetune.py:976] (1/7) Epoch 25, batch 4950, loss[loss=0.1731, simple_loss=0.2507, pruned_loss=0.04772, over 4888.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.2472, pruned_loss=0.05119, over 954398.06 frames. ], batch size: 43, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:53:03,768 INFO [finetune.py:976] (1/7) Epoch 25, batch 5000, loss[loss=0.1544, simple_loss=0.2371, pruned_loss=0.03591, over 4850.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2454, pruned_loss=0.05034, over 956013.21 frames. ], batch size: 44, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:53:04,976 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.190e+01 1.432e+02 1.813e+02 2.155e+02 3.992e+02, threshold=3.625e+02, percent-clipped=1.0 +2023-03-27 06:53:05,240 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-03-27 06:53:08,105 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2667, 2.2253, 2.3138, 1.5295, 2.2291, 2.3377, 2.3187, 1.8687], + device='cuda:1'), covar=tensor([0.0506, 0.0602, 0.0657, 0.0847, 0.0650, 0.0637, 0.0572, 0.1058], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0141, 0.0120, 0.0128, 0.0139, 0.0141, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:53:36,413 INFO [finetune.py:976] (1/7) Epoch 25, batch 5050, loss[loss=0.1595, simple_loss=0.2324, pruned_loss=0.04329, over 4896.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2434, pruned_loss=0.04989, over 955967.71 frames. ], batch size: 35, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:54:09,844 INFO [finetune.py:976] (1/7) Epoch 25, batch 5100, loss[loss=0.1547, simple_loss=0.2217, pruned_loss=0.04381, over 4913.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2397, pruned_loss=0.04858, over 955939.80 frames. ], batch size: 32, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:54:11,047 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.212e+01 1.519e+02 1.807e+02 2.247e+02 4.075e+02, threshold=3.613e+02, percent-clipped=2.0 +2023-03-27 06:54:14,209 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=142572.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:54:14,851 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=142573.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:54:27,733 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=142584.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:54:59,671 INFO [finetune.py:976] (1/7) Epoch 25, batch 5150, loss[loss=0.1675, simple_loss=0.242, pruned_loss=0.04646, over 4801.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.2398, pruned_loss=0.04884, over 956626.71 frames. ], batch size: 45, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:55:03,299 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=142620.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:55:10,596 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=142632.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:55:12,829 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=142634.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:55:33,009 INFO [finetune.py:976] (1/7) Epoch 25, batch 5200, loss[loss=0.1869, simple_loss=0.2528, pruned_loss=0.06051, over 4816.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2429, pruned_loss=0.04949, over 955658.70 frames. ], batch size: 38, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:55:34,191 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.084e+02 1.563e+02 1.762e+02 2.093e+02 3.679e+02, threshold=3.523e+02, percent-clipped=1.0 +2023-03-27 06:55:50,740 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4638, 1.3993, 1.9244, 1.7340, 1.5748, 3.3781, 1.3733, 1.5796], + device='cuda:1'), covar=tensor([0.0950, 0.1815, 0.1038, 0.0932, 0.1606, 0.0211, 0.1458, 0.1808], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0081, 0.0072, 0.0075, 0.0090, 0.0080, 0.0085, 0.0079], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 06:56:02,029 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4679, 2.2084, 2.7937, 1.8036, 2.2623, 2.6978, 2.0415, 2.6928], + device='cuda:1'), covar=tensor([0.1564, 0.2375, 0.1614, 0.2448, 0.1231, 0.1824, 0.2982, 0.1039], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0203, 0.0189, 0.0187, 0.0172, 0.0210, 0.0212, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:56:03,241 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6698, 1.4681, 2.1223, 3.3702, 2.2403, 2.4908, 1.0084, 2.8274], + device='cuda:1'), covar=tensor([0.1767, 0.1402, 0.1425, 0.0609, 0.0853, 0.1547, 0.1784, 0.0529], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0134, 0.0165, 0.0102, 0.0137, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 06:56:06,167 INFO [finetune.py:976] (1/7) Epoch 25, batch 5250, loss[loss=0.2132, simple_loss=0.2725, pruned_loss=0.07697, over 4185.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2457, pruned_loss=0.0501, over 956176.37 frames. ], batch size: 65, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:56:12,120 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=142724.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:56:14,566 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2827, 1.3214, 1.5748, 1.1085, 1.3017, 1.5091, 1.2958, 1.6363], + device='cuda:1'), covar=tensor([0.1217, 0.2391, 0.1271, 0.1478, 0.1021, 0.1238, 0.3216, 0.0813], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0204, 0.0189, 0.0188, 0.0172, 0.0211, 0.0213, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:56:33,395 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8334, 2.5357, 3.0517, 2.1151, 2.6777, 3.1489, 2.3270, 3.0384], + device='cuda:1'), covar=tensor([0.1283, 0.1874, 0.1594, 0.1960, 0.0962, 0.1343, 0.2433, 0.0758], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0204, 0.0190, 0.0188, 0.0172, 0.0211, 0.0213, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:56:39,096 INFO [finetune.py:976] (1/7) Epoch 25, batch 5300, loss[loss=0.1929, simple_loss=0.2656, pruned_loss=0.06007, over 4804.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2472, pruned_loss=0.05111, over 955779.10 frames. ], batch size: 40, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:56:40,273 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.031e+02 1.558e+02 1.826e+02 2.127e+02 3.045e+02, threshold=3.651e+02, percent-clipped=0.0 +2023-03-27 06:56:51,751 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=142785.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:57:06,678 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5321, 3.3780, 3.2385, 1.5967, 3.4647, 2.6121, 0.8179, 2.4125], + device='cuda:1'), covar=tensor([0.2262, 0.1688, 0.1513, 0.3096, 0.1127, 0.0964, 0.4032, 0.1261], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0179, 0.0161, 0.0130, 0.0161, 0.0123, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 06:57:19,473 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3480, 2.4299, 2.3466, 1.7305, 2.1995, 2.5901, 2.6167, 2.0524], + device='cuda:1'), covar=tensor([0.0515, 0.0584, 0.0700, 0.0844, 0.1251, 0.0630, 0.0492, 0.1019], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0141, 0.0120, 0.0129, 0.0139, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:57:19,998 INFO [finetune.py:976] (1/7) Epoch 25, batch 5350, loss[loss=0.1266, simple_loss=0.2052, pruned_loss=0.02401, over 4748.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2467, pruned_loss=0.0502, over 955019.27 frames. ], batch size: 54, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:57:50,302 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0 +2023-03-27 06:57:54,338 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7266, 2.4068, 2.9580, 1.9485, 2.5653, 2.9407, 2.1420, 2.8597], + device='cuda:1'), covar=tensor([0.1248, 0.1907, 0.1509, 0.2111, 0.1020, 0.1468, 0.2679, 0.0878], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0204, 0.0190, 0.0188, 0.0172, 0.0211, 0.0214, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:58:06,040 INFO [finetune.py:976] (1/7) Epoch 25, batch 5400, loss[loss=0.152, simple_loss=0.2251, pruned_loss=0.03949, over 4910.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2429, pruned_loss=0.04837, over 956445.33 frames. ], batch size: 37, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:58:07,255 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.131e+02 1.487e+02 1.682e+02 2.190e+02 4.832e+02, threshold=3.364e+02, percent-clipped=1.0 +2023-03-27 06:58:22,380 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2406, 2.1951, 1.7632, 2.0741, 2.1990, 1.9146, 2.4749, 2.2585], + device='cuda:1'), covar=tensor([0.1224, 0.1902, 0.2770, 0.2441, 0.2356, 0.1575, 0.2554, 0.1548], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0190, 0.0235, 0.0254, 0.0249, 0.0206, 0.0214, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 06:58:38,660 INFO [finetune.py:976] (1/7) Epoch 25, batch 5450, loss[loss=0.1708, simple_loss=0.2351, pruned_loss=0.05328, over 4911.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.2403, pruned_loss=0.04808, over 954833.99 frames. ], batch size: 32, lr: 2.99e-03, grad_scale: 32.0 +2023-03-27 06:58:47,628 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=142929.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 06:59:08,955 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2859, 2.8882, 2.7482, 1.2369, 2.9412, 2.2476, 0.7193, 1.9397], + device='cuda:1'), covar=tensor([0.2624, 0.2678, 0.2000, 0.3575, 0.1526, 0.1091, 0.4045, 0.1726], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0179, 0.0162, 0.0131, 0.0161, 0.0124, 0.0149, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 06:59:11,891 INFO [finetune.py:976] (1/7) Epoch 25, batch 5500, loss[loss=0.1619, simple_loss=0.2196, pruned_loss=0.05213, over 4809.00 frames. ], tot_loss[loss=0.167, simple_loss=0.2382, pruned_loss=0.04788, over 956804.50 frames. ], batch size: 25, lr: 2.99e-03, grad_scale: 16.0 +2023-03-27 06:59:13,716 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.198e+01 1.372e+02 1.708e+02 2.223e+02 4.314e+02, threshold=3.415e+02, percent-clipped=3.0 +2023-03-27 06:59:46,281 INFO [finetune.py:976] (1/7) Epoch 25, batch 5550, loss[loss=0.1677, simple_loss=0.249, pruned_loss=0.0432, over 4730.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2391, pruned_loss=0.04833, over 957043.64 frames. ], batch size: 54, lr: 2.99e-03, grad_scale: 16.0 +2023-03-27 07:00:29,924 INFO [finetune.py:976] (1/7) Epoch 25, batch 5600, loss[loss=0.1704, simple_loss=0.2421, pruned_loss=0.04941, over 4902.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2431, pruned_loss=0.04961, over 956129.78 frames. ], batch size: 35, lr: 2.99e-03, grad_scale: 16.0 +2023-03-27 07:00:31,669 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.665e+01 1.700e+02 1.937e+02 2.306e+02 4.675e+02, threshold=3.875e+02, percent-clipped=1.0 +2023-03-27 07:00:38,726 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=143080.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:01:00,341 INFO [finetune.py:976] (1/7) Epoch 25, batch 5650, loss[loss=0.1226, simple_loss=0.1954, pruned_loss=0.02496, over 4519.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2457, pruned_loss=0.05009, over 955024.32 frames. ], batch size: 20, lr: 2.99e-03, grad_scale: 16.0 +2023-03-27 07:01:05,198 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3822, 1.9964, 1.9896, 1.1365, 2.3511, 2.5462, 2.2407, 1.8978], + device='cuda:1'), covar=tensor([0.0904, 0.0872, 0.0531, 0.0617, 0.0455, 0.0655, 0.0403, 0.0668], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0127, 0.0121, 0.0129, 0.0128, 0.0141, 0.0147], + device='cuda:1'), out_proj_covar=tensor([8.8478e-05, 1.0546e-04, 9.0719e-05, 8.5230e-05, 9.0535e-05, 9.1216e-05, + 1.0024e-04, 1.0523e-04], device='cuda:1') +2023-03-27 07:01:29,839 INFO [finetune.py:976] (1/7) Epoch 25, batch 5700, loss[loss=0.1438, simple_loss=0.2118, pruned_loss=0.03789, over 4377.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.242, pruned_loss=0.04964, over 936485.49 frames. ], batch size: 19, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:01:31,567 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.213e+01 1.339e+02 1.671e+02 2.043e+02 4.216e+02, threshold=3.342e+02, percent-clipped=1.0 +2023-03-27 07:01:58,298 INFO [finetune.py:976] (1/7) Epoch 26, batch 0, loss[loss=0.1609, simple_loss=0.2378, pruned_loss=0.04205, over 4762.00 frames. ], tot_loss[loss=0.1609, simple_loss=0.2378, pruned_loss=0.04205, over 4762.00 frames. ], batch size: 28, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:01:58,298 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 07:02:01,135 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7832, 1.6821, 1.9796, 1.3431, 1.6580, 2.0034, 1.6032, 2.1157], + device='cuda:1'), covar=tensor([0.1215, 0.2202, 0.1387, 0.1830, 0.1039, 0.1370, 0.2862, 0.0877], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0205, 0.0190, 0.0190, 0.0173, 0.0212, 0.0215, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:02:09,054 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3866, 1.2923, 1.2550, 1.3788, 1.6274, 1.5494, 1.3554, 1.2093], + device='cuda:1'), covar=tensor([0.0397, 0.0359, 0.0664, 0.0356, 0.0276, 0.0452, 0.0433, 0.0450], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0107, 0.0146, 0.0112, 0.0102, 0.0116, 0.0104, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.8500e-05, 8.1959e-05, 1.1422e-04, 8.5531e-05, 7.8734e-05, 8.5958e-05, + 7.7183e-05, 8.6065e-05], device='cuda:1') +2023-03-27 07:02:12,071 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1472, 1.9221, 1.7712, 1.7175, 1.9028, 1.9363, 1.9058, 2.5665], + device='cuda:1'), covar=tensor([0.3706, 0.4526, 0.3408, 0.3853, 0.4228, 0.2470, 0.3950, 0.1850], + device='cuda:1'), in_proj_covar=tensor([0.0291, 0.0265, 0.0237, 0.0278, 0.0260, 0.0230, 0.0257, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:02:14,279 INFO [finetune.py:1010] (1/7) Epoch 26, validation: loss=0.1591, simple_loss=0.2269, pruned_loss=0.04565, over 2265189.00 frames. +2023-03-27 07:02:14,280 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 07:02:43,894 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=143229.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:03:00,610 INFO [finetune.py:976] (1/7) Epoch 26, batch 50, loss[loss=0.1495, simple_loss=0.2252, pruned_loss=0.03689, over 4805.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2427, pruned_loss=0.04867, over 215079.24 frames. ], batch size: 33, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:03:13,060 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4557, 1.5637, 1.4901, 0.8935, 1.6756, 1.8954, 1.8579, 1.4768], + device='cuda:1'), covar=tensor([0.0981, 0.0627, 0.0628, 0.0517, 0.0522, 0.0622, 0.0329, 0.0632], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0147, 0.0128, 0.0122, 0.0130, 0.0129, 0.0141, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.8943e-05, 1.0601e-04, 9.1068e-05, 8.5568e-05, 9.0961e-05, 9.1380e-05, + 1.0051e-04, 1.0563e-04], device='cuda:1') +2023-03-27 07:03:18,452 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.036e+02 1.460e+02 1.766e+02 2.058e+02 4.416e+02, threshold=3.532e+02, percent-clipped=3.0 +2023-03-27 07:03:19,188 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0037, 2.0321, 1.5666, 1.8304, 1.9668, 1.8417, 1.9224, 2.5489], + device='cuda:1'), covar=tensor([0.3801, 0.3869, 0.3195, 0.3538, 0.3417, 0.2378, 0.3348, 0.1754], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0264, 0.0236, 0.0277, 0.0259, 0.0229, 0.0256, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:03:24,476 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=143277.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:03:30,571 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4530, 2.3692, 1.9311, 2.5343, 2.4170, 2.1158, 2.8811, 2.5714], + device='cuda:1'), covar=tensor([0.1345, 0.2283, 0.2819, 0.2424, 0.2420, 0.1619, 0.2904, 0.1704], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0191, 0.0237, 0.0255, 0.0252, 0.0208, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:03:34,115 INFO [finetune.py:976] (1/7) Epoch 26, batch 100, loss[loss=0.1858, simple_loss=0.2505, pruned_loss=0.06053, over 4770.00 frames. ], tot_loss[loss=0.1658, simple_loss=0.2369, pruned_loss=0.04737, over 379699.17 frames. ], batch size: 27, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:03:34,190 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8154, 3.6582, 3.4842, 1.7356, 3.7787, 2.9593, 0.9708, 2.6664], + device='cuda:1'), covar=tensor([0.2515, 0.1667, 0.1646, 0.3401, 0.0933, 0.0913, 0.4303, 0.1520], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0178, 0.0161, 0.0130, 0.0160, 0.0123, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 07:04:07,508 INFO [finetune.py:976] (1/7) Epoch 26, batch 150, loss[loss=0.1445, simple_loss=0.2088, pruned_loss=0.0401, over 4832.00 frames. ], tot_loss[loss=0.1622, simple_loss=0.2321, pruned_loss=0.04614, over 508940.61 frames. ], batch size: 47, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:04:15,582 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.29 vs. limit=5.0 +2023-03-27 07:04:25,693 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.744e+01 1.335e+02 1.679e+02 2.114e+02 2.886e+02, threshold=3.358e+02, percent-clipped=0.0 +2023-03-27 07:04:33,622 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=143380.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:04:41,258 INFO [finetune.py:976] (1/7) Epoch 26, batch 200, loss[loss=0.1664, simple_loss=0.2332, pruned_loss=0.04975, over 4806.00 frames. ], tot_loss[loss=0.1631, simple_loss=0.2329, pruned_loss=0.04666, over 607642.54 frames. ], batch size: 25, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:04:46,841 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=143400.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 07:05:05,299 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=143428.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:05:05,375 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3136, 2.2871, 1.8133, 2.4259, 2.2803, 1.9641, 2.7094, 2.4223], + device='cuda:1'), covar=tensor([0.1402, 0.2382, 0.2904, 0.2679, 0.2568, 0.1771, 0.3259, 0.1799], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0191, 0.0237, 0.0256, 0.0252, 0.0208, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:05:22,139 INFO [finetune.py:976] (1/7) Epoch 26, batch 250, loss[loss=0.1769, simple_loss=0.2364, pruned_loss=0.05869, over 4703.00 frames. ], tot_loss[loss=0.1651, simple_loss=0.2358, pruned_loss=0.04722, over 685409.51 frames. ], batch size: 23, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:05:48,872 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=143461.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:05:51,340 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0840, 2.8140, 2.5088, 1.4403, 2.7501, 2.2639, 2.1827, 2.4358], + device='cuda:1'), covar=tensor([0.1187, 0.0743, 0.1806, 0.2083, 0.1594, 0.2107, 0.1960, 0.1161], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0193, 0.0202, 0.0183, 0.0212, 0.0213, 0.0226, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:05:53,071 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.741e+01 1.618e+02 1.961e+02 2.394e+02 5.476e+02, threshold=3.922e+02, percent-clipped=2.0 +2023-03-27 07:05:59,886 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=143473.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 07:06:12,193 INFO [finetune.py:976] (1/7) Epoch 26, batch 300, loss[loss=0.1944, simple_loss=0.2732, pruned_loss=0.05776, over 4850.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2404, pruned_loss=0.0482, over 743637.13 frames. ], batch size: 44, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:06:27,527 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3225, 1.8496, 2.7644, 1.7529, 2.3110, 2.5828, 1.7178, 2.5688], + device='cuda:1'), covar=tensor([0.1383, 0.2357, 0.1589, 0.2093, 0.0966, 0.1346, 0.2924, 0.0878], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0204, 0.0190, 0.0189, 0.0173, 0.0212, 0.0215, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:06:40,202 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=143534.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 07:06:44,912 INFO [finetune.py:976] (1/7) Epoch 26, batch 350, loss[loss=0.193, simple_loss=0.2754, pruned_loss=0.05532, over 4731.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2443, pruned_loss=0.04949, over 791165.52 frames. ], batch size: 59, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:07:03,062 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.042e+02 1.441e+02 1.724e+02 2.077e+02 3.544e+02, threshold=3.448e+02, percent-clipped=0.0 +2023-03-27 07:07:11,597 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0369, 4.8356, 4.6026, 2.9152, 4.9331, 3.7897, 1.1788, 3.5091], + device='cuda:1'), covar=tensor([0.1993, 0.1552, 0.1348, 0.2659, 0.0799, 0.0762, 0.4270, 0.1243], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0178, 0.0161, 0.0130, 0.0160, 0.0123, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 07:07:18,105 INFO [finetune.py:976] (1/7) Epoch 26, batch 400, loss[loss=0.1432, simple_loss=0.2175, pruned_loss=0.0345, over 4747.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2458, pruned_loss=0.0499, over 826157.63 frames. ], batch size: 27, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:07:54,070 INFO [finetune.py:976] (1/7) Epoch 26, batch 450, loss[loss=0.1384, simple_loss=0.219, pruned_loss=0.02889, over 4701.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2453, pruned_loss=0.04978, over 854679.97 frames. ], batch size: 23, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:08:22,190 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.245e+01 1.544e+02 1.809e+02 2.165e+02 3.752e+02, threshold=3.619e+02, percent-clipped=3.0 +2023-03-27 07:08:37,482 INFO [finetune.py:976] (1/7) Epoch 26, batch 500, loss[loss=0.1576, simple_loss=0.2444, pruned_loss=0.03542, over 4893.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2433, pruned_loss=0.04948, over 876053.07 frames. ], batch size: 35, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:09:00,534 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5030, 2.3647, 1.6275, 2.5342, 2.4176, 1.9936, 3.2201, 2.5058], + device='cuda:1'), covar=tensor([0.1314, 0.2250, 0.3437, 0.3121, 0.2746, 0.1817, 0.2441, 0.1845], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0191, 0.0236, 0.0255, 0.0250, 0.0207, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:09:11,113 INFO [finetune.py:976] (1/7) Epoch 26, batch 550, loss[loss=0.1972, simple_loss=0.2623, pruned_loss=0.06611, over 4811.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.2396, pruned_loss=0.04845, over 895578.61 frames. ], batch size: 45, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:09:20,263 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=143756.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:09:28,913 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.175e+01 1.443e+02 1.723e+02 1.984e+02 5.074e+02, threshold=3.446e+02, percent-clipped=2.0 +2023-03-27 07:09:44,563 INFO [finetune.py:976] (1/7) Epoch 26, batch 600, loss[loss=0.1852, simple_loss=0.2616, pruned_loss=0.05438, over 4817.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2415, pruned_loss=0.04953, over 908087.63 frames. ], batch size: 47, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:10:09,699 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=143829.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:10:17,582 INFO [finetune.py:976] (1/7) Epoch 26, batch 650, loss[loss=0.1648, simple_loss=0.2479, pruned_loss=0.04087, over 4822.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2433, pruned_loss=0.05011, over 919169.83 frames. ], batch size: 30, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:10:38,296 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6568, 1.4959, 2.1437, 2.0106, 1.8559, 4.1056, 1.5674, 1.6780], + device='cuda:1'), covar=tensor([0.0965, 0.1809, 0.1179, 0.0857, 0.1517, 0.0180, 0.1491, 0.1910], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0091, 0.0080, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:10:41,253 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.012e+02 1.600e+02 1.821e+02 2.293e+02 5.159e+02, threshold=3.642e+02, percent-clipped=4.0 +2023-03-27 07:11:07,602 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8244, 1.7797, 1.6911, 1.9452, 1.7382, 4.4550, 1.7157, 2.1056], + device='cuda:1'), covar=tensor([0.3209, 0.2281, 0.2026, 0.2229, 0.1361, 0.0138, 0.2297, 0.1143], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0121, 0.0124, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:11:12,407 INFO [finetune.py:976] (1/7) Epoch 26, batch 700, loss[loss=0.1606, simple_loss=0.2291, pruned_loss=0.04599, over 4922.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.2462, pruned_loss=0.05169, over 925705.40 frames. ], batch size: 33, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:11:22,134 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0483, 1.8188, 2.0334, 1.4206, 1.8555, 2.0478, 2.0656, 1.5245], + device='cuda:1'), covar=tensor([0.0537, 0.0708, 0.0716, 0.0854, 0.0973, 0.0592, 0.0546, 0.1196], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0138, 0.0142, 0.0120, 0.0129, 0.0139, 0.0141, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:11:52,958 INFO [finetune.py:976] (1/7) Epoch 26, batch 750, loss[loss=0.1988, simple_loss=0.2762, pruned_loss=0.06071, over 4828.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2463, pruned_loss=0.05136, over 932076.95 frames. ], batch size: 47, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:12:00,368 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1157, 1.7637, 2.1376, 2.1473, 1.8975, 1.8736, 2.1071, 2.0491], + device='cuda:1'), covar=tensor([0.4146, 0.4273, 0.3270, 0.4154, 0.5007, 0.4301, 0.5013, 0.2912], + device='cuda:1'), in_proj_covar=tensor([0.0264, 0.0246, 0.0266, 0.0293, 0.0293, 0.0270, 0.0299, 0.0249], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:12:09,862 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.567e+02 1.788e+02 2.169e+02 3.888e+02, threshold=3.576e+02, percent-clipped=1.0 +2023-03-27 07:12:26,435 INFO [finetune.py:976] (1/7) Epoch 26, batch 800, loss[loss=0.1965, simple_loss=0.2725, pruned_loss=0.06026, over 4893.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2466, pruned_loss=0.05085, over 937816.49 frames. ], batch size: 43, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:12:27,777 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9211, 1.8455, 1.7006, 2.0503, 2.4159, 2.0885, 1.9029, 1.6028], + device='cuda:1'), covar=tensor([0.1980, 0.1671, 0.1770, 0.1544, 0.1593, 0.1088, 0.2075, 0.1661], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0209, 0.0215, 0.0197, 0.0244, 0.0191, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:12:55,303 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9384, 1.9226, 1.7805, 2.0307, 1.5685, 4.6805, 1.7962, 2.4550], + device='cuda:1'), covar=tensor([0.3208, 0.2373, 0.2075, 0.2372, 0.1630, 0.0137, 0.2251, 0.1079], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0120, 0.0124, 0.0113, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:13:00,656 INFO [finetune.py:976] (1/7) Epoch 26, batch 850, loss[loss=0.1682, simple_loss=0.246, pruned_loss=0.04518, over 4851.00 frames. ], tot_loss[loss=0.171, simple_loss=0.243, pruned_loss=0.04956, over 939254.43 frames. ], batch size: 49, lr: 2.98e-03, grad_scale: 16.0 +2023-03-27 07:13:09,692 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=144056.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:13:16,916 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.067e+02 1.460e+02 1.746e+02 2.115e+02 7.519e+02, threshold=3.492e+02, percent-clipped=2.0 +2023-03-27 07:13:43,957 INFO [finetune.py:976] (1/7) Epoch 26, batch 900, loss[loss=0.209, simple_loss=0.269, pruned_loss=0.07446, over 4918.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2411, pruned_loss=0.04884, over 944962.53 frames. ], batch size: 36, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:13:45,338 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9011, 1.8516, 1.6235, 2.0526, 2.3499, 2.0537, 1.8870, 1.5700], + device='cuda:1'), covar=tensor([0.1863, 0.1547, 0.1644, 0.1369, 0.1507, 0.1018, 0.1873, 0.1659], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0209, 0.0214, 0.0197, 0.0243, 0.0190, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:13:51,326 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=144104.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 07:13:55,649 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-27 07:13:56,110 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5565, 1.4690, 1.4372, 1.4233, 0.8982, 2.3052, 0.7730, 1.2880], + device='cuda:1'), covar=tensor([0.3425, 0.2587, 0.2159, 0.2411, 0.1808, 0.0391, 0.2490, 0.1306], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0120, 0.0124, 0.0113, 0.0095, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:14:07,459 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=144129.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:14:16,860 INFO [finetune.py:976] (1/7) Epoch 26, batch 950, loss[loss=0.1683, simple_loss=0.2263, pruned_loss=0.05513, over 4745.00 frames. ], tot_loss[loss=0.1677, simple_loss=0.2388, pruned_loss=0.04836, over 948433.20 frames. ], batch size: 23, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:14:33,134 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.128e+02 1.468e+02 1.742e+02 2.065e+02 3.876e+02, threshold=3.485e+02, percent-clipped=2.0 +2023-03-27 07:14:39,231 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=144177.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 07:14:50,350 INFO [finetune.py:976] (1/7) Epoch 26, batch 1000, loss[loss=0.1525, simple_loss=0.2209, pruned_loss=0.04202, over 4699.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2418, pruned_loss=0.04977, over 950313.59 frames. ], batch size: 23, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:15:02,702 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8066, 1.1132, 1.7985, 1.8490, 1.6171, 1.5763, 1.7436, 1.7492], + device='cuda:1'), covar=tensor([0.3671, 0.3905, 0.3292, 0.3285, 0.4625, 0.3602, 0.4060, 0.2950], + device='cuda:1'), in_proj_covar=tensor([0.0264, 0.0246, 0.0266, 0.0293, 0.0293, 0.0269, 0.0300, 0.0250], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:15:18,927 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-27 07:15:22,313 INFO [finetune.py:976] (1/7) Epoch 26, batch 1050, loss[loss=0.1518, simple_loss=0.2238, pruned_loss=0.03997, over 4775.00 frames. ], tot_loss[loss=0.1728, simple_loss=0.2449, pruned_loss=0.0503, over 952348.66 frames. ], batch size: 26, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:15:40,002 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.069e+02 1.483e+02 1.785e+02 2.219e+02 5.161e+02, threshold=3.570e+02, percent-clipped=2.0 +2023-03-27 07:16:01,562 INFO [finetune.py:976] (1/7) Epoch 26, batch 1100, loss[loss=0.229, simple_loss=0.2948, pruned_loss=0.08156, over 4795.00 frames. ], tot_loss[loss=0.1748, simple_loss=0.2474, pruned_loss=0.05114, over 954093.40 frames. ], batch size: 51, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:16:45,103 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-03-27 07:16:45,525 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4626, 2.1462, 2.7156, 1.5864, 2.3391, 2.6976, 1.9357, 2.8028], + device='cuda:1'), covar=tensor([0.1399, 0.1939, 0.1539, 0.2264, 0.1041, 0.1539, 0.2729, 0.0814], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0205, 0.0191, 0.0190, 0.0174, 0.0212, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:16:56,472 INFO [finetune.py:976] (1/7) Epoch 26, batch 1150, loss[loss=0.1752, simple_loss=0.2709, pruned_loss=0.03978, over 4816.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.247, pruned_loss=0.05038, over 954040.61 frames. ], batch size: 39, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:17:13,862 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.072e+02 1.499e+02 1.760e+02 2.197e+02 4.327e+02, threshold=3.521e+02, percent-clipped=2.0 +2023-03-27 07:17:28,619 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9849, 1.8541, 2.0205, 1.1180, 1.9625, 2.0198, 1.9466, 1.6691], + device='cuda:1'), covar=tensor([0.0583, 0.0772, 0.0638, 0.0947, 0.0994, 0.0642, 0.0591, 0.1144], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0138, 0.0141, 0.0120, 0.0128, 0.0139, 0.0141, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:17:30,185 INFO [finetune.py:976] (1/7) Epoch 26, batch 1200, loss[loss=0.1341, simple_loss=0.2083, pruned_loss=0.02993, over 4714.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2456, pruned_loss=0.05039, over 953439.52 frames. ], batch size: 23, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:17:39,536 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-27 07:17:43,722 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=144411.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:18:03,374 INFO [finetune.py:976] (1/7) Epoch 26, batch 1250, loss[loss=0.1642, simple_loss=0.2348, pruned_loss=0.04684, over 4849.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2432, pruned_loss=0.05005, over 954862.25 frames. ], batch size: 47, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:18:11,745 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-03-27 07:18:21,714 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.894e+01 1.606e+02 1.805e+02 2.274e+02 3.881e+02, threshold=3.611e+02, percent-clipped=1.0 +2023-03-27 07:18:24,300 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=144472.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:18:37,272 INFO [finetune.py:976] (1/7) Epoch 26, batch 1300, loss[loss=0.1981, simple_loss=0.2538, pruned_loss=0.07123, over 4877.00 frames. ], tot_loss[loss=0.1697, simple_loss=0.2404, pruned_loss=0.04949, over 954167.32 frames. ], batch size: 31, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:19:00,384 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6863, 1.5004, 1.0788, 0.3174, 1.3037, 1.5267, 1.4592, 1.4011], + device='cuda:1'), covar=tensor([0.0988, 0.0811, 0.1344, 0.1851, 0.1280, 0.2195, 0.2268, 0.0896], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0192, 0.0201, 0.0182, 0.0211, 0.0211, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:19:21,353 INFO [finetune.py:976] (1/7) Epoch 26, batch 1350, loss[loss=0.1357, simple_loss=0.2057, pruned_loss=0.03288, over 4770.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2401, pruned_loss=0.04908, over 954688.19 frames. ], batch size: 28, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:19:39,472 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.034e+02 1.499e+02 1.803e+02 2.073e+02 4.281e+02, threshold=3.607e+02, percent-clipped=1.0 +2023-03-27 07:19:52,767 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8412, 1.8065, 1.5872, 1.9714, 2.4962, 2.1597, 1.8928, 1.5093], + device='cuda:1'), covar=tensor([0.2109, 0.1904, 0.1988, 0.1596, 0.1524, 0.1044, 0.2047, 0.1910], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0216, 0.0199, 0.0245, 0.0191, 0.0216, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:19:54,499 INFO [finetune.py:976] (1/7) Epoch 26, batch 1400, loss[loss=0.1699, simple_loss=0.2467, pruned_loss=0.04656, over 4849.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2433, pruned_loss=0.04946, over 956893.83 frames. ], batch size: 47, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:20:21,879 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-27 07:20:22,393 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4249, 2.3590, 1.8531, 2.5587, 2.4127, 1.9974, 2.7927, 2.4307], + device='cuda:1'), covar=tensor([0.1354, 0.2230, 0.3000, 0.2573, 0.2718, 0.1806, 0.3098, 0.1744], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0190, 0.0235, 0.0253, 0.0249, 0.0206, 0.0214, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:20:27,738 INFO [finetune.py:976] (1/7) Epoch 26, batch 1450, loss[loss=0.1846, simple_loss=0.2495, pruned_loss=0.05984, over 4761.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2453, pruned_loss=0.04982, over 956591.14 frames. ], batch size: 54, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:20:45,366 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9837, 1.4876, 2.1243, 2.0674, 1.8091, 1.8180, 1.9337, 1.9506], + device='cuda:1'), covar=tensor([0.3902, 0.4032, 0.3252, 0.3494, 0.4965, 0.3946, 0.4764, 0.2994], + device='cuda:1'), in_proj_covar=tensor([0.0264, 0.0247, 0.0267, 0.0294, 0.0294, 0.0271, 0.0301, 0.0250], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:20:45,823 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.092e+01 1.571e+02 1.855e+02 2.334e+02 4.645e+02, threshold=3.710e+02, percent-clipped=2.0 +2023-03-27 07:21:01,362 INFO [finetune.py:976] (1/7) Epoch 26, batch 1500, loss[loss=0.1677, simple_loss=0.2545, pruned_loss=0.04049, over 4806.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2451, pruned_loss=0.049, over 956937.20 frames. ], batch size: 45, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:21:50,337 INFO [finetune.py:976] (1/7) Epoch 26, batch 1550, loss[loss=0.1874, simple_loss=0.2587, pruned_loss=0.05801, over 4794.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2462, pruned_loss=0.04963, over 958406.08 frames. ], batch size: 29, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:21:58,276 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.13 vs. limit=5.0 +2023-03-27 07:22:18,427 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=144767.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:22:18,964 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.202e+02 1.547e+02 1.850e+02 2.044e+02 4.068e+02, threshold=3.700e+02, percent-clipped=1.0 +2023-03-27 07:22:30,226 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1199, 1.8414, 2.4251, 1.5131, 2.1256, 2.4330, 1.7566, 2.5096], + device='cuda:1'), covar=tensor([0.1451, 0.2083, 0.1542, 0.1994, 0.1026, 0.1291, 0.2748, 0.0950], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0204, 0.0189, 0.0188, 0.0172, 0.0211, 0.0213, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:22:35,464 INFO [finetune.py:976] (1/7) Epoch 26, batch 1600, loss[loss=0.1441, simple_loss=0.2116, pruned_loss=0.03829, over 4800.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.2442, pruned_loss=0.04922, over 958958.66 frames. ], batch size: 25, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:22:43,664 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 07:23:09,212 INFO [finetune.py:976] (1/7) Epoch 26, batch 1650, loss[loss=0.1887, simple_loss=0.2649, pruned_loss=0.05627, over 4760.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2422, pruned_loss=0.0491, over 957177.85 frames. ], batch size: 26, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:23:18,222 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.2879, 1.4845, 1.5263, 0.8703, 1.5771, 1.7532, 1.8085, 1.4226], + device='cuda:1'), covar=tensor([0.0872, 0.0595, 0.0510, 0.0498, 0.0409, 0.0652, 0.0294, 0.0625], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0147, 0.0128, 0.0122, 0.0130, 0.0129, 0.0142, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.9098e-05, 1.0608e-04, 9.0942e-05, 8.5846e-05, 9.0926e-05, 9.1489e-05, + 1.0098e-04, 1.0605e-04], device='cuda:1') +2023-03-27 07:23:26,320 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.503e+01 1.460e+02 1.738e+02 2.010e+02 3.428e+02, threshold=3.475e+02, percent-clipped=0.0 +2023-03-27 07:23:42,421 INFO [finetune.py:976] (1/7) Epoch 26, batch 1700, loss[loss=0.1505, simple_loss=0.2362, pruned_loss=0.03241, over 4790.00 frames. ], tot_loss[loss=0.1672, simple_loss=0.2389, pruned_loss=0.04776, over 957440.12 frames. ], batch size: 29, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:23:59,173 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7381, 1.6060, 1.5687, 1.7107, 1.1907, 3.6087, 1.4205, 1.8677], + device='cuda:1'), covar=tensor([0.3376, 0.2575, 0.2142, 0.2282, 0.1745, 0.0181, 0.2576, 0.1266], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0124, 0.0113, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:24:25,822 INFO [finetune.py:976] (1/7) Epoch 26, batch 1750, loss[loss=0.1291, simple_loss=0.208, pruned_loss=0.0251, over 4790.00 frames. ], tot_loss[loss=0.1675, simple_loss=0.2396, pruned_loss=0.04776, over 956244.65 frames. ], batch size: 25, lr: 2.97e-03, grad_scale: 16.0 +2023-03-27 07:24:42,911 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.967e+01 1.592e+02 1.823e+02 2.389e+02 4.337e+02, threshold=3.645e+02, percent-clipped=3.0 +2023-03-27 07:24:43,057 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5195, 1.7883, 1.4685, 1.4211, 2.1262, 2.0231, 1.7501, 1.7095], + device='cuda:1'), covar=tensor([0.0513, 0.0357, 0.0587, 0.0391, 0.0282, 0.0664, 0.0369, 0.0413], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0108, 0.0149, 0.0113, 0.0103, 0.0117, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.9816e-05, 8.2758e-05, 1.1597e-04, 8.6408e-05, 7.9600e-05, 8.6289e-05, + 7.7167e-05, 8.6889e-05], device='cuda:1') +2023-03-27 07:24:59,584 INFO [finetune.py:976] (1/7) Epoch 26, batch 1800, loss[loss=0.2195, simple_loss=0.2963, pruned_loss=0.07137, over 4815.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2433, pruned_loss=0.04877, over 955301.82 frames. ], batch size: 45, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:25:02,122 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=144996.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:25:06,997 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4104, 2.3935, 2.3917, 1.6282, 2.4405, 2.5671, 2.4896, 2.0398], + device='cuda:1'), covar=tensor([0.0557, 0.0564, 0.0664, 0.0847, 0.0560, 0.0578, 0.0617, 0.1000], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0136, 0.0140, 0.0119, 0.0127, 0.0138, 0.0139, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:25:23,070 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145027.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:25:33,497 INFO [finetune.py:976] (1/7) Epoch 26, batch 1850, loss[loss=0.2073, simple_loss=0.2698, pruned_loss=0.07244, over 4917.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.244, pruned_loss=0.04856, over 953747.41 frames. ], batch size: 38, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:25:43,148 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145057.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 07:25:48,484 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145066.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:25:49,087 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=145067.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:25:49,578 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.533e+02 1.829e+02 2.183e+02 4.392e+02, threshold=3.659e+02, percent-clipped=3.0 +2023-03-27 07:25:55,609 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-27 07:26:03,667 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145088.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:26:06,519 INFO [finetune.py:976] (1/7) Epoch 26, batch 1900, loss[loss=0.1414, simple_loss=0.2167, pruned_loss=0.033, over 4360.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2454, pruned_loss=0.04934, over 954400.41 frames. ], batch size: 19, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:26:21,295 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=145115.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:26:29,576 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145127.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:26:45,617 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7029, 1.6485, 1.5923, 1.7032, 1.1968, 3.5812, 1.3706, 1.8210], + device='cuda:1'), covar=tensor([0.3310, 0.2552, 0.2173, 0.2353, 0.1813, 0.0209, 0.2607, 0.1300], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0124, 0.0113, 0.0095, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:26:46,246 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2340, 2.2574, 1.9551, 2.3228, 2.1695, 2.1575, 2.1301, 2.9590], + device='cuda:1'), covar=tensor([0.3513, 0.4927, 0.3263, 0.4051, 0.4397, 0.2468, 0.4333, 0.1512], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0264, 0.0235, 0.0275, 0.0258, 0.0228, 0.0256, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:26:46,686 INFO [finetune.py:976] (1/7) Epoch 26, batch 1950, loss[loss=0.152, simple_loss=0.2241, pruned_loss=0.03994, over 4838.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.244, pruned_loss=0.04923, over 955757.51 frames. ], batch size: 47, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:26:57,617 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4544, 1.6341, 0.8887, 2.0967, 2.7136, 2.0696, 1.9728, 1.9509], + device='cuda:1'), covar=tensor([0.1227, 0.1883, 0.1827, 0.1146, 0.1558, 0.1620, 0.1351, 0.1979], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0120, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 07:27:06,392 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7794, 1.5528, 2.2296, 3.5509, 2.3886, 2.5490, 1.3577, 2.9877], + device='cuda:1'), covar=tensor([0.1527, 0.1370, 0.1245, 0.0463, 0.0741, 0.1250, 0.1590, 0.0423], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0164, 0.0101, 0.0135, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 07:27:15,912 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.125e+02 1.577e+02 1.831e+02 2.188e+02 4.363e+02, threshold=3.662e+02, percent-clipped=3.0 +2023-03-27 07:27:33,096 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6234, 1.9155, 1.6288, 1.5587, 2.1994, 2.2268, 1.8554, 1.8249], + device='cuda:1'), covar=tensor([0.0537, 0.0342, 0.0553, 0.0367, 0.0353, 0.0608, 0.0372, 0.0406], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0108, 0.0149, 0.0112, 0.0102, 0.0117, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.9592e-05, 8.2567e-05, 1.1592e-04, 8.6029e-05, 7.9356e-05, 8.6271e-05, + 7.7143e-05, 8.6779e-05], device='cuda:1') +2023-03-27 07:27:40,100 INFO [finetune.py:976] (1/7) Epoch 26, batch 2000, loss[loss=0.1645, simple_loss=0.2394, pruned_loss=0.04475, over 4758.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2419, pruned_loss=0.04885, over 956928.01 frames. ], batch size: 26, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:28:00,002 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3671, 2.3728, 2.0740, 2.5153, 2.9923, 2.4520, 2.3933, 1.8493], + device='cuda:1'), covar=tensor([0.2164, 0.1682, 0.1815, 0.1435, 0.1585, 0.1057, 0.1839, 0.1773], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0215, 0.0198, 0.0244, 0.0191, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:28:13,277 INFO [finetune.py:976] (1/7) Epoch 26, batch 2050, loss[loss=0.1216, simple_loss=0.2005, pruned_loss=0.02139, over 4763.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.2386, pruned_loss=0.04749, over 957012.44 frames. ], batch size: 28, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:28:19,939 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4258, 1.6558, 1.7376, 0.8933, 1.7429, 1.9413, 1.9424, 1.5456], + device='cuda:1'), covar=tensor([0.1009, 0.0666, 0.0565, 0.0638, 0.0536, 0.0748, 0.0346, 0.0747], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0128, 0.0123, 0.0130, 0.0129, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.8954e-05, 1.0618e-04, 9.1031e-05, 8.6206e-05, 9.0856e-05, 9.1632e-05, + 1.0083e-04, 1.0626e-04], device='cuda:1') +2023-03-27 07:28:30,375 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.187e+01 1.436e+02 1.792e+02 2.264e+02 4.038e+02, threshold=3.583e+02, percent-clipped=1.0 +2023-03-27 07:28:45,865 INFO [finetune.py:976] (1/7) Epoch 26, batch 2100, loss[loss=0.1464, simple_loss=0.2237, pruned_loss=0.0346, over 4802.00 frames. ], tot_loss[loss=0.1654, simple_loss=0.2372, pruned_loss=0.04682, over 955680.36 frames. ], batch size: 25, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:29:03,660 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7504, 2.6554, 2.2911, 1.2236, 2.4010, 2.1328, 1.9357, 2.3991], + device='cuda:1'), covar=tensor([0.0778, 0.0676, 0.1346, 0.1795, 0.1227, 0.1787, 0.2009, 0.0880], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0192, 0.0200, 0.0181, 0.0209, 0.0209, 0.0224, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:29:09,164 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1319, 1.7044, 1.8023, 0.8369, 1.9396, 2.1860, 2.0021, 1.7268], + device='cuda:1'), covar=tensor([0.1135, 0.0835, 0.0584, 0.0783, 0.0543, 0.0751, 0.0457, 0.0860], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0128, 0.0123, 0.0130, 0.0129, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.9137e-05, 1.0644e-04, 9.1238e-05, 8.6419e-05, 9.1066e-05, 9.1707e-05, + 1.0097e-04, 1.0656e-04], device='cuda:1') +2023-03-27 07:29:19,662 INFO [finetune.py:976] (1/7) Epoch 26, batch 2150, loss[loss=0.1798, simple_loss=0.2598, pruned_loss=0.04985, over 4733.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2415, pruned_loss=0.04869, over 955124.53 frames. ], batch size: 54, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:29:28,935 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145352.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 07:29:39,079 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0713, 1.9527, 1.6827, 1.8357, 1.8267, 1.8666, 1.8971, 2.6174], + device='cuda:1'), covar=tensor([0.3740, 0.4014, 0.3214, 0.3614, 0.3943, 0.2348, 0.3520, 0.1664], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0265, 0.0237, 0.0278, 0.0260, 0.0230, 0.0258, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:29:43,499 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5400, 3.9296, 4.1924, 4.4071, 4.2667, 3.9915, 4.6380, 1.3866], + device='cuda:1'), covar=tensor([0.0749, 0.0983, 0.0814, 0.0906, 0.1274, 0.1730, 0.0657, 0.5845], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0249, 0.0280, 0.0295, 0.0337, 0.0287, 0.0304, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:29:47,542 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.022e+02 1.494e+02 1.709e+02 2.298e+02 6.165e+02, threshold=3.419e+02, percent-clipped=3.0 +2023-03-27 07:29:49,488 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145371.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:29:56,781 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145383.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:30:02,627 INFO [finetune.py:976] (1/7) Epoch 26, batch 2200, loss[loss=0.1798, simple_loss=0.2525, pruned_loss=0.05351, over 4918.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2434, pruned_loss=0.04908, over 954971.52 frames. ], batch size: 42, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:30:03,787 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5853, 1.4184, 1.9414, 3.2000, 2.1277, 2.2824, 0.9113, 2.7527], + device='cuda:1'), covar=tensor([0.1736, 0.1401, 0.1252, 0.0546, 0.0823, 0.1339, 0.1848, 0.0426], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0162, 0.0101, 0.0134, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 07:30:22,233 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-27 07:30:23,255 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145422.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:30:29,362 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145432.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:30:36,241 INFO [finetune.py:976] (1/7) Epoch 26, batch 2250, loss[loss=0.1716, simple_loss=0.2401, pruned_loss=0.05148, over 4799.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2448, pruned_loss=0.04953, over 956514.17 frames. ], batch size: 51, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:30:37,627 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0705, 0.9912, 0.9665, 1.1920, 1.2461, 1.1870, 0.9954, 0.9633], + device='cuda:1'), covar=tensor([0.0437, 0.0360, 0.0692, 0.0346, 0.0319, 0.0448, 0.0399, 0.0502], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0109, 0.0149, 0.0113, 0.0103, 0.0118, 0.0104, 0.0115], + device='cuda:1'), out_proj_covar=tensor([8.0326e-05, 8.3024e-05, 1.1643e-04, 8.6571e-05, 7.9891e-05, 8.6772e-05, + 7.7549e-05, 8.7307e-05], device='cuda:1') +2023-03-27 07:30:44,585 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8037, 1.6147, 2.2767, 2.0448, 1.9663, 4.3876, 1.6848, 1.7825], + device='cuda:1'), covar=tensor([0.0942, 0.1841, 0.1223, 0.0900, 0.1449, 0.0176, 0.1469, 0.1794], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0072, 0.0076, 0.0090, 0.0080, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:30:53,949 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.189e+01 1.488e+02 1.760e+02 2.143e+02 3.776e+02, threshold=3.521e+02, percent-clipped=2.0 +2023-03-27 07:31:08,985 INFO [finetune.py:976] (1/7) Epoch 26, batch 2300, loss[loss=0.1529, simple_loss=0.2267, pruned_loss=0.03955, over 4792.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2449, pruned_loss=0.04903, over 956740.28 frames. ], batch size: 26, lr: 2.97e-03, grad_scale: 32.0 +2023-03-27 07:31:41,337 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6223, 3.7289, 3.5184, 1.8978, 3.7583, 2.7930, 0.9415, 2.7715], + device='cuda:1'), covar=tensor([0.2410, 0.1741, 0.1616, 0.3197, 0.1155, 0.1110, 0.4322, 0.1428], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0179, 0.0160, 0.0130, 0.0161, 0.0124, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 07:31:42,487 INFO [finetune.py:976] (1/7) Epoch 26, batch 2350, loss[loss=0.1182, simple_loss=0.192, pruned_loss=0.02218, over 4772.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2434, pruned_loss=0.04883, over 957383.14 frames. ], batch size: 28, lr: 2.96e-03, grad_scale: 32.0 +2023-03-27 07:31:53,135 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0801, 1.9227, 2.0127, 1.3236, 2.0148, 2.0872, 2.0515, 1.6681], + device='cuda:1'), covar=tensor([0.0472, 0.0608, 0.0691, 0.0942, 0.0814, 0.0605, 0.0602, 0.1068], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0137, 0.0141, 0.0119, 0.0128, 0.0138, 0.0139, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:32:06,623 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.008e+02 1.544e+02 1.840e+02 2.226e+02 4.643e+02, threshold=3.680e+02, percent-clipped=1.0 +2023-03-27 07:32:34,351 INFO [finetune.py:976] (1/7) Epoch 26, batch 2400, loss[loss=0.1726, simple_loss=0.2418, pruned_loss=0.05172, over 4824.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2409, pruned_loss=0.04843, over 955676.30 frames. ], batch size: 40, lr: 2.96e-03, grad_scale: 32.0 +2023-03-27 07:32:34,434 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.1448, 4.4676, 4.7215, 4.9451, 4.8513, 4.5688, 5.2161, 1.5752], + device='cuda:1'), covar=tensor([0.0649, 0.0798, 0.0735, 0.0848, 0.1132, 0.1552, 0.0547, 0.6026], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0249, 0.0280, 0.0296, 0.0337, 0.0287, 0.0304, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:32:35,672 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145594.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:33:17,100 INFO [finetune.py:976] (1/7) Epoch 26, batch 2450, loss[loss=0.1604, simple_loss=0.2338, pruned_loss=0.04348, over 4926.00 frames. ], tot_loss[loss=0.167, simple_loss=0.2387, pruned_loss=0.0477, over 957764.01 frames. ], batch size: 33, lr: 2.96e-03, grad_scale: 32.0 +2023-03-27 07:33:17,810 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145643.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:33:23,854 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=145652.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 07:33:26,159 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2506, 1.4723, 0.8752, 1.8963, 2.3475, 1.8279, 1.7418, 1.9014], + device='cuda:1'), covar=tensor([0.1209, 0.1950, 0.1847, 0.1109, 0.1866, 0.1750, 0.1370, 0.1850], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0119, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 07:33:26,197 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145655.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 07:33:34,916 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.521e+01 1.449e+02 1.824e+02 2.163e+02 4.630e+02, threshold=3.648e+02, percent-clipped=2.0 +2023-03-27 07:33:37,523 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-27 07:33:45,640 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=145683.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:33:51,040 INFO [finetune.py:976] (1/7) Epoch 26, batch 2500, loss[loss=0.1653, simple_loss=0.2371, pruned_loss=0.04672, over 4851.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.239, pruned_loss=0.04763, over 958265.16 frames. ], batch size: 44, lr: 2.96e-03, grad_scale: 32.0 +2023-03-27 07:33:55,943 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=145700.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:33:57,103 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4147, 2.1967, 1.8522, 0.8889, 2.1168, 1.7027, 1.4456, 1.9983], + device='cuda:1'), covar=tensor([0.0968, 0.1107, 0.1924, 0.2409, 0.1614, 0.2419, 0.2834, 0.1317], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0192, 0.0200, 0.0182, 0.0211, 0.0211, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:33:58,911 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145704.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:34:01,217 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2056, 2.1574, 1.9224, 2.2975, 2.1235, 2.1481, 2.0687, 2.7803], + device='cuda:1'), covar=tensor([0.3264, 0.4590, 0.3205, 0.3769, 0.4268, 0.2370, 0.4283, 0.1486], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0235, 0.0275, 0.0258, 0.0228, 0.0257, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:34:11,650 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=145722.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:34:12,402 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.15 vs. limit=5.0 +2023-03-27 07:34:15,144 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145727.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:34:17,567 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=145731.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:34:24,605 INFO [finetune.py:976] (1/7) Epoch 26, batch 2550, loss[loss=0.2133, simple_loss=0.2787, pruned_loss=0.07396, over 4816.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2425, pruned_loss=0.04839, over 956296.32 frames. ], batch size: 38, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:34:42,325 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.531e+01 1.551e+02 1.807e+02 2.106e+02 4.459e+02, threshold=3.615e+02, percent-clipped=2.0 +2023-03-27 07:34:43,554 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=145770.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:34:51,966 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7043, 1.5673, 1.4323, 1.8388, 2.0436, 1.8129, 1.3621, 1.4242], + device='cuda:1'), covar=tensor([0.2225, 0.2171, 0.2099, 0.1694, 0.1667, 0.1252, 0.2522, 0.2022], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0211, 0.0216, 0.0199, 0.0246, 0.0192, 0.0218, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:35:08,874 INFO [finetune.py:976] (1/7) Epoch 26, batch 2600, loss[loss=0.1368, simple_loss=0.2213, pruned_loss=0.02609, over 4795.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2439, pruned_loss=0.04881, over 954118.83 frames. ], batch size: 29, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:35:28,207 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145821.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:35:35,871 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0890, 1.9001, 1.8425, 0.8961, 2.0852, 2.2900, 2.0913, 1.8001], + device='cuda:1'), covar=tensor([0.0930, 0.0594, 0.0486, 0.0642, 0.0501, 0.0620, 0.0479, 0.0567], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0128, 0.0122, 0.0130, 0.0129, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.8925e-05, 1.0617e-04, 9.1291e-05, 8.6019e-05, 9.0989e-05, 9.1421e-05, + 1.0087e-04, 1.0643e-04], device='cuda:1') +2023-03-27 07:35:42,707 INFO [finetune.py:976] (1/7) Epoch 26, batch 2650, loss[loss=0.1464, simple_loss=0.227, pruned_loss=0.03285, over 4728.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2457, pruned_loss=0.04956, over 955680.05 frames. ], batch size: 23, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:35:55,358 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6491, 1.4293, 1.9291, 1.7312, 1.6479, 3.5431, 1.4060, 1.5676], + device='cuda:1'), covar=tensor([0.0943, 0.1851, 0.1037, 0.0969, 0.1618, 0.0232, 0.1536, 0.1811], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0091, 0.0081, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:36:00,019 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.986e+01 1.512e+02 1.783e+02 2.110e+02 4.476e+02, threshold=3.566e+02, percent-clipped=1.0 +2023-03-27 07:36:09,558 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145882.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 07:36:16,424 INFO [finetune.py:976] (1/7) Epoch 26, batch 2700, loss[loss=0.1615, simple_loss=0.2176, pruned_loss=0.05265, over 4595.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2454, pruned_loss=0.04933, over 955771.65 frames. ], batch size: 19, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:36:48,575 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4355, 1.4852, 1.9368, 1.7577, 1.6156, 3.3517, 1.3372, 1.5289], + device='cuda:1'), covar=tensor([0.1099, 0.2183, 0.1134, 0.1039, 0.1663, 0.0291, 0.1820, 0.2158], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0091, 0.0081, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 07:36:49,652 INFO [finetune.py:976] (1/7) Epoch 26, batch 2750, loss[loss=0.1796, simple_loss=0.2396, pruned_loss=0.05974, over 4821.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2431, pruned_loss=0.04898, over 957249.62 frames. ], batch size: 33, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:36:53,367 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1052, 1.7751, 2.1431, 2.1243, 1.8500, 1.8493, 2.0850, 1.9881], + device='cuda:1'), covar=tensor([0.4198, 0.4082, 0.3092, 0.3882, 0.4909, 0.4159, 0.4614, 0.2954], + device='cuda:1'), in_proj_covar=tensor([0.0266, 0.0247, 0.0266, 0.0294, 0.0294, 0.0271, 0.0300, 0.0251], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:36:55,118 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145950.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 07:36:56,605 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.45 vs. limit=2.0 +2023-03-27 07:37:00,652 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4909, 1.3806, 1.3088, 0.7256, 1.5527, 1.6235, 1.6735, 1.3236], + device='cuda:1'), covar=tensor([0.0815, 0.0600, 0.0574, 0.0550, 0.0461, 0.0556, 0.0306, 0.0627], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0128, 0.0122, 0.0130, 0.0129, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.9061e-05, 1.0631e-04, 9.1093e-05, 8.5941e-05, 9.1062e-05, 9.1470e-05, + 1.0093e-04, 1.0625e-04], device='cuda:1') +2023-03-27 07:37:07,589 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.011e+02 1.569e+02 1.804e+02 2.200e+02 3.850e+02, threshold=3.609e+02, percent-clipped=2.0 +2023-03-27 07:37:29,448 INFO [finetune.py:976] (1/7) Epoch 26, batch 2800, loss[loss=0.1856, simple_loss=0.2511, pruned_loss=0.06008, over 4910.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2405, pruned_loss=0.0483, over 957954.08 frames. ], batch size: 36, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:37:38,978 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145999.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:38:14,438 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=146027.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:38:18,650 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4784, 1.3794, 1.3681, 1.3359, 0.7749, 2.2941, 0.7575, 1.1992], + device='cuda:1'), covar=tensor([0.3524, 0.2736, 0.2250, 0.2552, 0.2055, 0.0372, 0.2794, 0.1373], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0124, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0005, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:38:22,680 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9474, 1.7278, 2.1946, 1.4472, 2.1035, 2.2528, 1.5684, 2.2938], + device='cuda:1'), covar=tensor([0.1329, 0.2089, 0.1478, 0.2051, 0.0841, 0.1293, 0.2970, 0.0848], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0207, 0.0192, 0.0189, 0.0174, 0.0213, 0.0217, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:38:24,986 INFO [finetune.py:976] (1/7) Epoch 26, batch 2850, loss[loss=0.16, simple_loss=0.2284, pruned_loss=0.04579, over 4764.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2392, pruned_loss=0.04831, over 957155.54 frames. ], batch size: 59, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:38:42,242 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.185e+01 1.523e+02 1.819e+02 2.110e+02 4.930e+02, threshold=3.638e+02, percent-clipped=2.0 +2023-03-27 07:38:46,429 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=146075.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:38:58,297 INFO [finetune.py:976] (1/7) Epoch 26, batch 2900, loss[loss=0.1966, simple_loss=0.2758, pruned_loss=0.05869, over 4922.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2436, pruned_loss=0.04978, over 957652.55 frames. ], batch size: 42, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:39:31,498 INFO [finetune.py:976] (1/7) Epoch 26, batch 2950, loss[loss=0.1884, simple_loss=0.2543, pruned_loss=0.06128, over 4821.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2462, pruned_loss=0.05039, over 956074.59 frames. ], batch size: 40, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:39:36,384 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=146149.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:39:49,295 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.099e+02 1.550e+02 1.859e+02 2.106e+02 3.478e+02, threshold=3.719e+02, percent-clipped=0.0 +2023-03-27 07:39:54,693 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=146177.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:40:04,822 INFO [finetune.py:976] (1/7) Epoch 26, batch 3000, loss[loss=0.1324, simple_loss=0.2136, pruned_loss=0.02562, over 4821.00 frames. ], tot_loss[loss=0.1739, simple_loss=0.2467, pruned_loss=0.0506, over 955197.36 frames. ], batch size: 30, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:40:04,822 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 07:40:10,104 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7676, 1.3448, 0.9761, 1.6209, 2.1321, 1.2846, 1.6460, 1.5746], + device='cuda:1'), covar=tensor([0.1260, 0.1661, 0.1655, 0.1037, 0.1740, 0.1811, 0.1170, 0.1799], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 07:40:12,771 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8091, 1.6616, 1.5293, 1.9038, 2.1546, 1.8845, 1.4650, 1.5358], + device='cuda:1'), covar=tensor([0.2205, 0.2093, 0.2072, 0.1712, 0.1622, 0.1183, 0.2428, 0.1894], + device='cuda:1'), in_proj_covar=tensor([0.0248, 0.0212, 0.0217, 0.0200, 0.0249, 0.0193, 0.0220, 0.0207], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:40:19,936 INFO [finetune.py:1010] (1/7) Epoch 26, validation: loss=0.1577, simple_loss=0.2252, pruned_loss=0.04507, over 2265189.00 frames. +2023-03-27 07:40:19,936 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 07:40:35,424 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=146210.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:40:56,712 INFO [finetune.py:976] (1/7) Epoch 26, batch 3050, loss[loss=0.1826, simple_loss=0.2409, pruned_loss=0.06219, over 4797.00 frames. ], tot_loss[loss=0.174, simple_loss=0.2475, pruned_loss=0.05024, over 956462.08 frames. ], batch size: 26, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:41:02,119 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=146250.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:41:14,892 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.646e+01 1.408e+02 1.795e+02 2.163e+02 4.679e+02, threshold=3.589e+02, percent-clipped=3.0 +2023-03-27 07:41:29,843 INFO [finetune.py:976] (1/7) Epoch 26, batch 3100, loss[loss=0.1709, simple_loss=0.2491, pruned_loss=0.04639, over 4716.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2446, pruned_loss=0.04936, over 956164.40 frames. ], batch size: 23, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:41:34,031 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=146298.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:41:34,698 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=146299.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:42:02,609 INFO [finetune.py:976] (1/7) Epoch 26, batch 3150, loss[loss=0.1614, simple_loss=0.2333, pruned_loss=0.04476, over 4213.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2424, pruned_loss=0.04857, over 956181.73 frames. ], batch size: 65, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:42:06,559 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=146347.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:42:06,618 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4841, 1.4086, 1.5539, 0.7630, 1.5407, 1.5767, 1.5145, 1.3463], + device='cuda:1'), covar=tensor([0.0645, 0.0862, 0.0714, 0.1032, 0.1003, 0.0711, 0.0655, 0.1406], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0137, 0.0141, 0.0120, 0.0128, 0.0138, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:42:21,141 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.044e+02 1.464e+02 1.851e+02 2.163e+02 3.423e+02, threshold=3.701e+02, percent-clipped=0.0 +2023-03-27 07:42:26,194 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-27 07:42:38,055 INFO [finetune.py:976] (1/7) Epoch 26, batch 3200, loss[loss=0.1347, simple_loss=0.2082, pruned_loss=0.03058, over 4707.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2407, pruned_loss=0.04853, over 958317.74 frames. ], batch size: 23, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:43:27,402 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1313, 2.0057, 1.4548, 0.6393, 1.6685, 1.7453, 1.6098, 1.8192], + device='cuda:1'), covar=tensor([0.0942, 0.0788, 0.1687, 0.2140, 0.1472, 0.2522, 0.2410, 0.0958], + device='cuda:1'), in_proj_covar=tensor([0.0169, 0.0191, 0.0199, 0.0181, 0.0208, 0.0209, 0.0223, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:43:35,343 INFO [finetune.py:976] (1/7) Epoch 26, batch 3250, loss[loss=0.2166, simple_loss=0.2912, pruned_loss=0.07094, over 4743.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2427, pruned_loss=0.04987, over 956292.29 frames. ], batch size: 59, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:43:53,729 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.041e+02 1.451e+02 1.808e+02 2.175e+02 4.535e+02, threshold=3.616e+02, percent-clipped=3.0 +2023-03-27 07:43:58,657 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=146477.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 07:44:08,659 INFO [finetune.py:976] (1/7) Epoch 26, batch 3300, loss[loss=0.1493, simple_loss=0.2301, pruned_loss=0.03427, over 4809.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2455, pruned_loss=0.05024, over 955781.40 frames. ], batch size: 25, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:44:17,123 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=146505.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:44:30,705 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=146525.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:44:41,544 INFO [finetune.py:976] (1/7) Epoch 26, batch 3350, loss[loss=0.1618, simple_loss=0.2283, pruned_loss=0.04761, over 4930.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2471, pruned_loss=0.05068, over 957362.69 frames. ], batch size: 42, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:44:48,757 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7941, 4.0607, 3.8697, 1.8435, 4.1328, 3.0595, 0.9576, 2.8830], + device='cuda:1'), covar=tensor([0.2284, 0.1828, 0.1313, 0.3346, 0.0834, 0.0995, 0.4400, 0.1391], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0180, 0.0161, 0.0131, 0.0162, 0.0125, 0.0150, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 07:44:51,698 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4573, 3.3194, 3.1239, 1.4693, 3.4108, 2.5228, 0.9137, 2.2995], + device='cuda:1'), covar=tensor([0.2679, 0.2011, 0.1664, 0.3481, 0.1249, 0.1101, 0.4316, 0.1466], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0180, 0.0161, 0.0131, 0.0162, 0.0125, 0.0150, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 07:45:00,357 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.852e+01 1.562e+02 1.841e+02 2.282e+02 4.006e+02, threshold=3.682e+02, percent-clipped=1.0 +2023-03-27 07:45:14,935 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=146591.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:45:15,430 INFO [finetune.py:976] (1/7) Epoch 26, batch 3400, loss[loss=0.2078, simple_loss=0.271, pruned_loss=0.07234, over 4877.00 frames. ], tot_loss[loss=0.1752, simple_loss=0.2488, pruned_loss=0.05079, over 959798.21 frames. ], batch size: 32, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:45:22,425 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.67 vs. limit=2.0 +2023-03-27 07:45:48,131 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=146626.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:45:58,654 INFO [finetune.py:976] (1/7) Epoch 26, batch 3450, loss[loss=0.1951, simple_loss=0.2587, pruned_loss=0.06572, over 4839.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.2468, pruned_loss=0.05022, over 954070.81 frames. ], batch size: 44, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:46:04,890 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=146652.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:46:17,061 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.040e+02 1.450e+02 1.708e+02 2.017e+02 4.995e+02, threshold=3.417e+02, percent-clipped=1.0 +2023-03-27 07:46:22,479 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7331, 1.2194, 0.8563, 1.4645, 2.1824, 1.3104, 1.4750, 1.4544], + device='cuda:1'), covar=tensor([0.1506, 0.2185, 0.2035, 0.1427, 0.1917, 0.2049, 0.1623, 0.2138], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0119, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 07:46:27,953 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1766, 1.9576, 1.4887, 0.6207, 1.6549, 1.8201, 1.6583, 1.8691], + device='cuda:1'), covar=tensor([0.0968, 0.0794, 0.1444, 0.1916, 0.1368, 0.2228, 0.2209, 0.0896], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0192, 0.0200, 0.0182, 0.0210, 0.0211, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:46:28,567 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=146687.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:46:32,388 INFO [finetune.py:976] (1/7) Epoch 26, batch 3500, loss[loss=0.1826, simple_loss=0.2579, pruned_loss=0.05367, over 4911.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2442, pruned_loss=0.04958, over 955650.02 frames. ], batch size: 37, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:47:05,273 INFO [finetune.py:976] (1/7) Epoch 26, batch 3550, loss[loss=0.1348, simple_loss=0.197, pruned_loss=0.0363, over 4240.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2398, pruned_loss=0.04785, over 955276.42 frames. ], batch size: 65, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:47:22,725 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.051e+02 1.430e+02 1.748e+02 2.315e+02 5.079e+02, threshold=3.497e+02, percent-clipped=6.0 +2023-03-27 07:47:32,136 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3555, 1.8674, 0.7835, 1.9939, 2.5410, 1.7564, 2.1039, 2.0231], + device='cuda:1'), covar=tensor([0.1290, 0.1779, 0.2068, 0.1111, 0.1787, 0.1885, 0.1261, 0.1904], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 07:47:38,113 INFO [finetune.py:976] (1/7) Epoch 26, batch 3600, loss[loss=0.1647, simple_loss=0.2441, pruned_loss=0.04266, over 4727.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2392, pruned_loss=0.04872, over 954529.77 frames. ], batch size: 23, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:47:47,320 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=146805.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:48:08,928 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-27 07:48:24,415 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6910, 1.4611, 1.8714, 1.1786, 1.7368, 1.8353, 1.3699, 1.9726], + device='cuda:1'), covar=tensor([0.1081, 0.2043, 0.1156, 0.1732, 0.0919, 0.1312, 0.2934, 0.0873], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0192, 0.0189, 0.0174, 0.0212, 0.0216, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:48:26,737 INFO [finetune.py:976] (1/7) Epoch 26, batch 3650, loss[loss=0.1465, simple_loss=0.2286, pruned_loss=0.03221, over 4757.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2413, pruned_loss=0.04952, over 954712.63 frames. ], batch size: 28, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:48:28,094 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=146844.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:48:39,244 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=146853.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:48:53,564 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.488e+01 1.511e+02 1.813e+02 2.229e+02 3.524e+02, threshold=3.627e+02, percent-clipped=1.0 +2023-03-27 07:49:12,953 INFO [finetune.py:976] (1/7) Epoch 26, batch 3700, loss[loss=0.1997, simple_loss=0.2767, pruned_loss=0.06135, over 4808.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2439, pruned_loss=0.04961, over 954754.69 frames. ], batch size: 40, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:49:21,434 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=146905.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:49:25,535 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0497, 1.6546, 2.2537, 1.4445, 2.0941, 2.1885, 1.5407, 2.2659], + device='cuda:1'), covar=tensor([0.1281, 0.2349, 0.1421, 0.2108, 0.0892, 0.1382, 0.3093, 0.0958], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0207, 0.0192, 0.0190, 0.0175, 0.0213, 0.0217, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:49:46,523 INFO [finetune.py:976] (1/7) Epoch 26, batch 3750, loss[loss=0.1639, simple_loss=0.224, pruned_loss=0.05196, over 4264.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2438, pruned_loss=0.04949, over 953510.23 frames. ], batch size: 18, lr: 2.96e-03, grad_scale: 16.0 +2023-03-27 07:49:49,608 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=146947.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:50:03,845 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.820e+01 1.503e+02 1.791e+02 2.461e+02 5.017e+02, threshold=3.581e+02, percent-clipped=5.0 +2023-03-27 07:50:12,647 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=146982.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 07:50:19,582 INFO [finetune.py:976] (1/7) Epoch 26, batch 3800, loss[loss=0.1702, simple_loss=0.2509, pruned_loss=0.04477, over 4774.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2452, pruned_loss=0.04952, over 955503.86 frames. ], batch size: 29, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:50:52,952 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5510, 1.4630, 1.9471, 2.9176, 1.9795, 2.2075, 0.9712, 2.4976], + device='cuda:1'), covar=tensor([0.1601, 0.1324, 0.1112, 0.0643, 0.0819, 0.1287, 0.1702, 0.0471], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0165, 0.0102, 0.0135, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 07:50:54,791 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4578, 3.8784, 4.0196, 4.2670, 4.2162, 3.9687, 4.5360, 1.3006], + device='cuda:1'), covar=tensor([0.0741, 0.0886, 0.0934, 0.0905, 0.1263, 0.1520, 0.0660, 0.5927], + device='cuda:1'), in_proj_covar=tensor([0.0349, 0.0248, 0.0281, 0.0294, 0.0338, 0.0287, 0.0304, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:50:55,309 INFO [finetune.py:976] (1/7) Epoch 26, batch 3850, loss[loss=0.1871, simple_loss=0.253, pruned_loss=0.06057, over 4827.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.244, pruned_loss=0.0492, over 958148.52 frames. ], batch size: 38, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:51:21,146 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.070e+02 1.517e+02 1.854e+02 2.266e+02 5.483e+02, threshold=3.707e+02, percent-clipped=2.0 +2023-03-27 07:51:37,019 INFO [finetune.py:976] (1/7) Epoch 26, batch 3900, loss[loss=0.1671, simple_loss=0.2357, pruned_loss=0.0492, over 4393.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2414, pruned_loss=0.04844, over 959182.69 frames. ], batch size: 65, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:52:02,067 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-03-27 07:52:09,602 INFO [finetune.py:976] (1/7) Epoch 26, batch 3950, loss[loss=0.1982, simple_loss=0.2354, pruned_loss=0.08054, over 4243.00 frames. ], tot_loss[loss=0.1666, simple_loss=0.2379, pruned_loss=0.04767, over 954672.10 frames. ], batch size: 18, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:52:27,907 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.014e+02 1.492e+02 1.682e+02 1.976e+02 2.814e+02, threshold=3.365e+02, percent-clipped=0.0 +2023-03-27 07:52:42,786 INFO [finetune.py:976] (1/7) Epoch 26, batch 4000, loss[loss=0.1699, simple_loss=0.251, pruned_loss=0.04439, over 4932.00 frames. ], tot_loss[loss=0.167, simple_loss=0.2378, pruned_loss=0.04804, over 956872.02 frames. ], batch size: 33, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:52:48,867 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147200.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:52:51,913 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6446, 3.4682, 3.2490, 1.4860, 3.5819, 2.6342, 0.8103, 2.3455], + device='cuda:1'), covar=tensor([0.2151, 0.2192, 0.1655, 0.3601, 0.1119, 0.1113, 0.4349, 0.1620], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0180, 0.0160, 0.0130, 0.0161, 0.0124, 0.0149, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 07:53:15,801 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147241.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:53:16,322 INFO [finetune.py:976] (1/7) Epoch 26, batch 4050, loss[loss=0.1584, simple_loss=0.2211, pruned_loss=0.04786, over 4711.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2418, pruned_loss=0.04902, over 956278.33 frames. ], batch size: 23, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:53:21,645 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147247.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:53:30,816 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2146, 3.6798, 3.9024, 4.0556, 3.9779, 3.7645, 4.2761, 1.5124], + device='cuda:1'), covar=tensor([0.0720, 0.0925, 0.0932, 0.0994, 0.1211, 0.1548, 0.0733, 0.5341], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0249, 0.0282, 0.0296, 0.0339, 0.0288, 0.0304, 0.0301], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:53:48,586 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.158e+02 1.660e+02 1.920e+02 2.375e+02 4.575e+02, threshold=3.840e+02, percent-clipped=6.0 +2023-03-27 07:53:58,889 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147282.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:54:08,318 INFO [finetune.py:976] (1/7) Epoch 26, batch 4100, loss[loss=0.1923, simple_loss=0.2744, pruned_loss=0.05508, over 4826.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2437, pruned_loss=0.04968, over 955249.92 frames. ], batch size: 51, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:54:13,960 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=147295.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:54:18,773 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147302.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:54:24,616 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147310.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:54:37,225 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=147330.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:54:44,910 INFO [finetune.py:976] (1/7) Epoch 26, batch 4150, loss[loss=0.1885, simple_loss=0.2673, pruned_loss=0.05489, over 4898.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2457, pruned_loss=0.05003, over 956788.51 frames. ], batch size: 46, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:55:03,459 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.008e+02 1.520e+02 1.873e+02 2.208e+02 5.004e+02, threshold=3.746e+02, percent-clipped=1.0 +2023-03-27 07:55:04,809 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147371.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:55:18,277 INFO [finetune.py:976] (1/7) Epoch 26, batch 4200, loss[loss=0.1525, simple_loss=0.2361, pruned_loss=0.03444, over 4799.00 frames. ], tot_loss[loss=0.1722, simple_loss=0.2456, pruned_loss=0.04945, over 956673.17 frames. ], batch size: 55, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:55:28,553 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147407.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:55:46,454 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147434.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:55:51,221 INFO [finetune.py:976] (1/7) Epoch 26, batch 4250, loss[loss=0.1855, simple_loss=0.2544, pruned_loss=0.05825, over 4936.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.244, pruned_loss=0.04926, over 958195.27 frames. ], batch size: 33, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:56:15,688 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9195, 1.8108, 1.7007, 1.7880, 1.5186, 3.8853, 1.6365, 2.0289], + device='cuda:1'), covar=tensor([0.3112, 0.2419, 0.2101, 0.2256, 0.1598, 0.0180, 0.2367, 0.1199], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0124, 0.0113, 0.0095, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 07:56:16,330 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147468.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:56:16,791 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.473e+02 1.815e+02 2.255e+02 8.587e+02, threshold=3.630e+02, percent-clipped=2.0 +2023-03-27 07:56:34,763 INFO [finetune.py:976] (1/7) Epoch 26, batch 4300, loss[loss=0.1391, simple_loss=0.2154, pruned_loss=0.03142, over 4777.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.2407, pruned_loss=0.04849, over 957243.34 frames. ], batch size: 26, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:56:36,733 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147495.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:56:40,224 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147500.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:57:08,536 INFO [finetune.py:976] (1/7) Epoch 26, batch 4350, loss[loss=0.1694, simple_loss=0.2263, pruned_loss=0.05625, over 4303.00 frames. ], tot_loss[loss=0.1649, simple_loss=0.2365, pruned_loss=0.04667, over 955585.43 frames. ], batch size: 65, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:57:12,262 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=147548.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:57:26,970 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.870e+01 1.462e+02 1.667e+02 1.917e+02 5.708e+02, threshold=3.333e+02, percent-clipped=2.0 +2023-03-27 07:57:30,005 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0511, 1.9010, 2.0611, 1.4937, 1.9759, 2.1100, 2.0863, 1.5861], + device='cuda:1'), covar=tensor([0.0599, 0.0786, 0.0709, 0.0894, 0.0834, 0.0721, 0.0702, 0.1270], + device='cuda:1'), in_proj_covar=tensor([0.0130, 0.0135, 0.0139, 0.0118, 0.0126, 0.0137, 0.0138, 0.0160], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:57:42,395 INFO [finetune.py:976] (1/7) Epoch 26, batch 4400, loss[loss=0.175, simple_loss=0.2539, pruned_loss=0.04804, over 4907.00 frames. ], tot_loss[loss=0.1661, simple_loss=0.2381, pruned_loss=0.04702, over 957409.81 frames. ], batch size: 37, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:57:45,502 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147597.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:57:58,113 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0813, 0.9708, 0.9590, 0.4145, 0.9997, 1.1503, 1.2147, 1.0054], + device='cuda:1'), covar=tensor([0.0817, 0.0563, 0.0585, 0.0488, 0.0463, 0.0600, 0.0394, 0.0652], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0128, 0.0122, 0.0130, 0.0129, 0.0141, 0.0148], + device='cuda:1'), out_proj_covar=tensor([8.8519e-05, 1.0640e-04, 9.0996e-05, 8.5679e-05, 9.0754e-05, 9.1143e-05, + 1.0028e-04, 1.0612e-04], device='cuda:1') +2023-03-27 07:58:16,297 INFO [finetune.py:976] (1/7) Epoch 26, batch 4450, loss[loss=0.2228, simple_loss=0.2878, pruned_loss=0.07892, over 4904.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2429, pruned_loss=0.04925, over 958282.87 frames. ], batch size: 35, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:58:27,315 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147659.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:58:34,815 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147666.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:58:35,486 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8905, 1.3920, 1.9301, 1.9519, 1.7495, 1.6924, 1.8969, 1.8518], + device='cuda:1'), covar=tensor([0.3657, 0.3784, 0.3121, 0.3423, 0.4496, 0.3464, 0.4111, 0.2782], + device='cuda:1'), in_proj_covar=tensor([0.0265, 0.0247, 0.0267, 0.0294, 0.0294, 0.0270, 0.0301, 0.0251], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:58:36,564 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.042e+02 1.542e+02 1.864e+02 2.192e+02 3.736e+02, threshold=3.727e+02, percent-clipped=4.0 +2023-03-27 07:59:06,170 INFO [finetune.py:976] (1/7) Epoch 26, batch 4500, loss[loss=0.171, simple_loss=0.2554, pruned_loss=0.04336, over 4896.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2447, pruned_loss=0.04949, over 959913.91 frames. ], batch size: 35, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 07:59:26,373 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3757, 2.2365, 1.8665, 2.2029, 2.2293, 1.9777, 2.5894, 2.3650], + device='cuda:1'), covar=tensor([0.1293, 0.1825, 0.2805, 0.2401, 0.2438, 0.1625, 0.2697, 0.1545], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0189, 0.0235, 0.0251, 0.0248, 0.0205, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 07:59:37,020 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147720.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 07:59:52,137 INFO [finetune.py:976] (1/7) Epoch 26, batch 4550, loss[loss=0.2029, simple_loss=0.2664, pruned_loss=0.06968, over 4860.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2461, pruned_loss=0.05019, over 959575.47 frames. ], batch size: 34, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 07:59:55,992 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-27 07:59:58,868 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9918, 1.7396, 2.2753, 1.5282, 2.0991, 2.2351, 1.6277, 2.3359], + device='cuda:1'), covar=tensor([0.1290, 0.2133, 0.1389, 0.1950, 0.0861, 0.1402, 0.2726, 0.0844], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0208, 0.0194, 0.0191, 0.0176, 0.0214, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:00:04,235 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2591, 2.1125, 2.7574, 4.4124, 2.9820, 2.9490, 1.2113, 3.7501], + device='cuda:1'), covar=tensor([0.1609, 0.1329, 0.1277, 0.0474, 0.0842, 0.1342, 0.1884, 0.0385], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0164, 0.0102, 0.0135, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 08:00:04,820 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147763.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:00:09,336 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.722e+01 1.488e+02 1.756e+02 2.293e+02 4.562e+02, threshold=3.512e+02, percent-clipped=2.0 +2023-03-27 08:00:09,557 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0 +2023-03-27 08:00:11,911 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-27 08:00:20,094 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.85 vs. limit=5.0 +2023-03-27 08:00:24,540 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147790.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:00:25,690 INFO [finetune.py:976] (1/7) Epoch 26, batch 4600, loss[loss=0.1415, simple_loss=0.2159, pruned_loss=0.03354, over 4822.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2442, pruned_loss=0.04899, over 958645.79 frames. ], batch size: 38, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:00:39,918 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147815.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:00:59,126 INFO [finetune.py:976] (1/7) Epoch 26, batch 4650, loss[loss=0.1307, simple_loss=0.2016, pruned_loss=0.02994, over 4825.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2415, pruned_loss=0.04861, over 958279.13 frames. ], batch size: 33, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:01:08,855 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147857.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:01:15,988 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 7.955e+01 1.458e+02 1.712e+02 2.175e+02 4.467e+02, threshold=3.424e+02, percent-clipped=3.0 +2023-03-27 08:01:16,551 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0476, 1.7441, 2.3619, 3.8836, 2.5938, 2.6544, 0.8404, 3.3017], + device='cuda:1'), covar=tensor([0.1596, 0.1460, 0.1317, 0.0436, 0.0772, 0.1754, 0.1955, 0.0363], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0164, 0.0102, 0.0135, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 08:01:21,846 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147876.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:01:25,237 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0170, 1.8248, 1.4381, 1.4171, 2.2985, 2.4418, 1.9964, 1.9280], + device='cuda:1'), covar=tensor([0.0417, 0.0517, 0.0774, 0.0511, 0.0290, 0.0644, 0.0434, 0.0477], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0105, 0.0145, 0.0110, 0.0100, 0.0114, 0.0102, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.7723e-05, 8.0618e-05, 1.1327e-04, 8.4262e-05, 7.7597e-05, 8.4282e-05, + 7.6040e-05, 8.4623e-05], device='cuda:1') +2023-03-27 08:01:39,628 INFO [finetune.py:976] (1/7) Epoch 26, batch 4700, loss[loss=0.1491, simple_loss=0.2103, pruned_loss=0.04392, over 4777.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.2388, pruned_loss=0.04768, over 957222.64 frames. ], batch size: 26, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:01:45,897 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1251, 2.0904, 1.9833, 2.2595, 1.9901, 4.7849, 1.8586, 2.1422], + device='cuda:1'), covar=tensor([0.3119, 0.2333, 0.1906, 0.2084, 0.1318, 0.0116, 0.2198, 0.1186], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0124, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 08:01:46,498 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147897.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:01:55,411 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147911.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:01:59,684 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147918.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:02:06,929 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-27 08:02:16,231 INFO [finetune.py:976] (1/7) Epoch 26, batch 4750, loss[loss=0.1694, simple_loss=0.248, pruned_loss=0.04541, over 4896.00 frames. ], tot_loss[loss=0.1655, simple_loss=0.2372, pruned_loss=0.04695, over 956213.18 frames. ], batch size: 32, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:02:18,627 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=147945.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:02:19,760 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8780, 1.8517, 1.7322, 2.1283, 2.2074, 2.1349, 1.7301, 1.5675], + device='cuda:1'), covar=tensor([0.2049, 0.1804, 0.1801, 0.1482, 0.1831, 0.1090, 0.2221, 0.1867], + device='cuda:1'), in_proj_covar=tensor([0.0248, 0.0213, 0.0216, 0.0200, 0.0248, 0.0192, 0.0220, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:02:30,700 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-27 08:02:32,379 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147966.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:02:34,089 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.372e+01 1.470e+02 1.654e+02 2.049e+02 2.990e+02, threshold=3.309e+02, percent-clipped=0.0 +2023-03-27 08:02:35,999 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147972.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:02:50,079 INFO [finetune.py:976] (1/7) Epoch 26, batch 4800, loss[loss=0.1412, simple_loss=0.2345, pruned_loss=0.02398, over 4811.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.2409, pruned_loss=0.04835, over 954487.24 frames. ], batch size: 45, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:03:06,222 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=148014.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:03:06,842 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148015.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:03:09,947 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9340, 1.1198, 1.9167, 1.9467, 1.7447, 1.6946, 1.8018, 1.8801], + device='cuda:1'), covar=tensor([0.3135, 0.3579, 0.3145, 0.3251, 0.4376, 0.3508, 0.3889, 0.2710], + device='cuda:1'), in_proj_covar=tensor([0.0266, 0.0247, 0.0267, 0.0295, 0.0295, 0.0271, 0.0301, 0.0251], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:03:24,536 INFO [finetune.py:976] (1/7) Epoch 26, batch 4850, loss[loss=0.2007, simple_loss=0.2905, pruned_loss=0.05542, over 4915.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2445, pruned_loss=0.04958, over 953687.75 frames. ], batch size: 38, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:03:35,620 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1999, 1.7660, 2.1771, 2.1922, 1.9126, 1.9093, 2.1216, 2.0211], + device='cuda:1'), covar=tensor([0.4416, 0.4155, 0.3208, 0.3975, 0.5128, 0.4282, 0.4771, 0.3073], + device='cuda:1'), in_proj_covar=tensor([0.0265, 0.0246, 0.0266, 0.0294, 0.0294, 0.0270, 0.0300, 0.0250], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:03:38,651 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1829, 1.9825, 2.1836, 1.4049, 2.0559, 2.2014, 2.1203, 1.7460], + device='cuda:1'), covar=tensor([0.0490, 0.0671, 0.0614, 0.0844, 0.0705, 0.0615, 0.0591, 0.1105], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0142, 0.0120, 0.0128, 0.0139, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:03:39,226 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148063.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:03:42,782 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.092e+02 1.530e+02 1.908e+02 2.333e+02 3.886e+02, threshold=3.817e+02, percent-clipped=4.0 +2023-03-27 08:03:45,326 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4240, 1.3315, 1.2709, 1.3717, 1.6358, 1.6032, 1.3683, 1.2604], + device='cuda:1'), covar=tensor([0.0363, 0.0323, 0.0634, 0.0313, 0.0251, 0.0416, 0.0373, 0.0428], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0147, 0.0111, 0.0101, 0.0115, 0.0103, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.8194e-05, 8.1079e-05, 1.1441e-04, 8.4886e-05, 7.7963e-05, 8.4826e-05, + 7.6663e-05, 8.5218e-05], device='cuda:1') +2023-03-27 08:04:04,119 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148090.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:04:05,252 INFO [finetune.py:976] (1/7) Epoch 26, batch 4900, loss[loss=0.2096, simple_loss=0.2762, pruned_loss=0.07148, over 4921.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2463, pruned_loss=0.05035, over 954272.82 frames. ], batch size: 33, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:04:30,816 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=148111.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:04:41,286 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6032, 1.4539, 2.1262, 3.2454, 2.1248, 2.4270, 0.7970, 2.7513], + device='cuda:1'), covar=tensor([0.1815, 0.1712, 0.1479, 0.0806, 0.0959, 0.1750, 0.2299, 0.0569], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0163, 0.0100, 0.0134, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 08:04:43,214 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-27 08:04:57,726 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=148138.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:05:00,563 INFO [finetune.py:976] (1/7) Epoch 26, batch 4950, loss[loss=0.1467, simple_loss=0.207, pruned_loss=0.04319, over 4119.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2471, pruned_loss=0.05053, over 953646.30 frames. ], batch size: 17, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:05:05,400 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0330, 1.8727, 1.6518, 1.8569, 2.2514, 2.2844, 1.8029, 1.7842], + device='cuda:1'), covar=tensor([0.0377, 0.0348, 0.0587, 0.0318, 0.0285, 0.0434, 0.0458, 0.0420], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0147, 0.0111, 0.0101, 0.0115, 0.0103, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.8143e-05, 8.0960e-05, 1.1435e-04, 8.4790e-05, 7.7911e-05, 8.4824e-05, + 7.6681e-05, 8.5168e-05], device='cuda:1') +2023-03-27 08:05:18,897 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.030e+01 1.572e+02 1.871e+02 2.257e+02 5.603e+02, threshold=3.742e+02, percent-clipped=1.0 +2023-03-27 08:05:20,235 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148171.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:05:33,993 INFO [finetune.py:976] (1/7) Epoch 26, batch 5000, loss[loss=0.139, simple_loss=0.217, pruned_loss=0.0305, over 4892.00 frames. ], tot_loss[loss=0.1724, simple_loss=0.2451, pruned_loss=0.04983, over 950014.96 frames. ], batch size: 43, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:05:48,850 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148213.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:06:07,383 INFO [finetune.py:976] (1/7) Epoch 26, batch 5050, loss[loss=0.1963, simple_loss=0.2687, pruned_loss=0.06195, over 4918.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2432, pruned_loss=0.04966, over 952649.35 frames. ], batch size: 37, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:06:22,868 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.59 vs. limit=5.0 +2023-03-27 08:06:25,053 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148267.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:06:26,158 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.004e+02 1.490e+02 1.796e+02 2.082e+02 4.496e+02, threshold=3.592e+02, percent-clipped=3.0 +2023-03-27 08:06:37,579 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148287.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:06:40,478 INFO [finetune.py:976] (1/7) Epoch 26, batch 5100, loss[loss=0.174, simple_loss=0.2346, pruned_loss=0.0567, over 3953.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2409, pruned_loss=0.04895, over 954327.46 frames. ], batch size: 17, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:07:02,511 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148315.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:07:18,063 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4002, 1.2945, 1.2940, 1.2952, 0.8543, 2.3087, 0.7184, 1.1410], + device='cuda:1'), covar=tensor([0.3365, 0.2559, 0.2186, 0.2496, 0.1814, 0.0333, 0.2848, 0.1316], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0124, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 08:07:22,780 INFO [finetune.py:976] (1/7) Epoch 26, batch 5150, loss[loss=0.1787, simple_loss=0.2559, pruned_loss=0.05077, over 4809.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2414, pruned_loss=0.04927, over 954078.69 frames. ], batch size: 38, lr: 2.95e-03, grad_scale: 32.0 +2023-03-27 08:07:27,033 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148348.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:07:34,761 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4770, 1.0785, 0.8123, 1.3480, 1.9232, 0.7477, 1.2800, 1.3360], + device='cuda:1'), covar=tensor([0.1519, 0.2040, 0.1635, 0.1191, 0.1985, 0.1935, 0.1469, 0.1997], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0093, 0.0109, 0.0091, 0.0119, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 08:07:36,970 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=148363.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:07:41,443 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.157e+02 1.549e+02 1.828e+02 2.233e+02 5.689e+02, threshold=3.657e+02, percent-clipped=4.0 +2023-03-27 08:07:55,902 INFO [finetune.py:976] (1/7) Epoch 26, batch 5200, loss[loss=0.1664, simple_loss=0.2329, pruned_loss=0.04993, over 4936.00 frames. ], tot_loss[loss=0.1731, simple_loss=0.2449, pruned_loss=0.05069, over 953090.16 frames. ], batch size: 33, lr: 2.95e-03, grad_scale: 16.0 +2023-03-27 08:07:58,521 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.67 vs. limit=5.0 +2023-03-27 08:08:07,369 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4425, 1.0898, 0.8154, 1.3330, 1.9451, 0.7462, 1.2033, 1.3243], + device='cuda:1'), covar=tensor([0.1606, 0.2070, 0.1667, 0.1225, 0.1928, 0.2019, 0.1534, 0.1991], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0093, 0.0110, 0.0091, 0.0120, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 08:08:21,975 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148429.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:08:29,695 INFO [finetune.py:976] (1/7) Epoch 26, batch 5250, loss[loss=0.1209, simple_loss=0.1902, pruned_loss=0.02577, over 4197.00 frames. ], tot_loss[loss=0.1734, simple_loss=0.2457, pruned_loss=0.05053, over 952684.53 frames. ], batch size: 18, lr: 2.94e-03, grad_scale: 16.0 +2023-03-27 08:08:44,056 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3549, 1.3000, 1.7189, 1.6346, 1.4640, 3.0991, 1.2464, 1.3993], + device='cuda:1'), covar=tensor([0.1063, 0.2003, 0.1344, 0.1027, 0.1794, 0.0299, 0.1747, 0.2116], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0091, 0.0080, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 08:08:48,646 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.102e+02 1.514e+02 1.742e+02 2.193e+02 4.299e+02, threshold=3.484e+02, percent-clipped=1.0 +2023-03-27 08:08:49,334 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148471.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:08:52,145 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3246, 2.3809, 1.8814, 2.5466, 2.2923, 1.9849, 2.6906, 2.4316], + device='cuda:1'), covar=tensor([0.1249, 0.2092, 0.2681, 0.2261, 0.2227, 0.1510, 0.3216, 0.1446], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0190, 0.0237, 0.0253, 0.0250, 0.0207, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:09:02,318 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148490.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:09:03,426 INFO [finetune.py:976] (1/7) Epoch 26, batch 5300, loss[loss=0.1638, simple_loss=0.2436, pruned_loss=0.04202, over 4885.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2467, pruned_loss=0.0504, over 955114.15 frames. ], batch size: 43, lr: 2.94e-03, grad_scale: 16.0 +2023-03-27 08:09:07,113 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148498.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:09:09,566 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0642, 1.7343, 2.2862, 1.5203, 2.0779, 2.2599, 1.6412, 2.3632], + device='cuda:1'), covar=tensor([0.1249, 0.2240, 0.1330, 0.2002, 0.0924, 0.1422, 0.2929, 0.0814], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0206, 0.0193, 0.0189, 0.0174, 0.0212, 0.0216, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:09:20,044 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148513.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:09:28,365 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=148519.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:09:35,151 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6369, 1.6034, 2.1822, 3.4058, 2.3320, 2.4992, 1.0999, 2.8833], + device='cuda:1'), covar=tensor([0.1698, 0.1356, 0.1318, 0.0628, 0.0803, 0.1502, 0.1804, 0.0415], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0132, 0.0163, 0.0101, 0.0135, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 08:09:39,314 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1990, 1.5974, 0.7085, 1.8953, 2.5601, 1.7447, 1.8918, 2.0239], + device='cuda:1'), covar=tensor([0.1532, 0.2154, 0.2354, 0.1370, 0.1933, 0.2040, 0.1516, 0.2067], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 08:09:54,001 INFO [finetune.py:976] (1/7) Epoch 26, batch 5350, loss[loss=0.1864, simple_loss=0.2516, pruned_loss=0.06064, over 4868.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2462, pruned_loss=0.04977, over 956616.14 frames. ], batch size: 34, lr: 2.94e-03, grad_scale: 16.0 +2023-03-27 08:10:04,819 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1700, 2.0597, 1.8247, 2.0436, 1.9778, 1.9844, 2.0037, 2.6587], + device='cuda:1'), covar=tensor([0.3280, 0.3817, 0.3100, 0.3709, 0.3883, 0.2335, 0.3624, 0.1697], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0264, 0.0236, 0.0275, 0.0259, 0.0229, 0.0257, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:10:12,422 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148559.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:10:13,592 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=148561.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:10:17,757 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148567.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:10:19,466 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.029e+02 1.412e+02 1.653e+02 1.941e+02 3.220e+02, threshold=3.306e+02, percent-clipped=0.0 +2023-03-27 08:10:34,700 INFO [finetune.py:976] (1/7) Epoch 26, batch 5400, loss[loss=0.1537, simple_loss=0.2121, pruned_loss=0.04768, over 4815.00 frames. ], tot_loss[loss=0.171, simple_loss=0.2434, pruned_loss=0.04928, over 954938.97 frames. ], batch size: 25, lr: 2.94e-03, grad_scale: 16.0 +2023-03-27 08:10:44,613 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1237, 1.7204, 2.3410, 1.5171, 2.1189, 2.2728, 1.6474, 2.3960], + device='cuda:1'), covar=tensor([0.1099, 0.2013, 0.1447, 0.1933, 0.0887, 0.1483, 0.2692, 0.0788], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0208, 0.0194, 0.0191, 0.0175, 0.0214, 0.0217, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:10:49,741 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=148615.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:11:07,906 INFO [finetune.py:976] (1/7) Epoch 26, batch 5450, loss[loss=0.1645, simple_loss=0.226, pruned_loss=0.05155, over 4779.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2398, pruned_loss=0.04843, over 954331.67 frames. ], batch size: 29, lr: 2.94e-03, grad_scale: 16.0 +2023-03-27 08:11:08,562 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148643.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:11:13,468 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148651.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:11:25,814 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.571e+02 1.855e+02 2.174e+02 4.125e+02, threshold=3.711e+02, percent-clipped=2.0 +2023-03-27 08:11:41,098 INFO [finetune.py:976] (1/7) Epoch 26, batch 5500, loss[loss=0.1835, simple_loss=0.2606, pruned_loss=0.05324, over 4816.00 frames. ], tot_loss[loss=0.1666, simple_loss=0.2378, pruned_loss=0.04772, over 955362.86 frames. ], batch size: 40, lr: 2.94e-03, grad_scale: 16.0 +2023-03-27 08:11:51,578 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.44 vs. limit=5.0 +2023-03-27 08:11:53,970 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148712.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:11:56,386 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148716.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:11:57,138 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-27 08:12:24,626 INFO [finetune.py:976] (1/7) Epoch 26, batch 5550, loss[loss=0.1761, simple_loss=0.2603, pruned_loss=0.04597, over 4847.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2391, pruned_loss=0.04825, over 954190.12 frames. ], batch size: 49, lr: 2.94e-03, grad_scale: 16.0 +2023-03-27 08:12:42,332 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.664e+01 1.578e+02 1.917e+02 2.288e+02 4.413e+02, threshold=3.834e+02, percent-clipped=2.0 +2023-03-27 08:12:47,520 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148777.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:12:52,459 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148785.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:12:56,500 INFO [finetune.py:976] (1/7) Epoch 26, batch 5600, loss[loss=0.115, simple_loss=0.1875, pruned_loss=0.02129, over 4339.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2426, pruned_loss=0.04871, over 955538.73 frames. ], batch size: 19, lr: 2.94e-03, grad_scale: 16.0 +2023-03-27 08:13:10,062 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8644, 1.8304, 1.7159, 1.8127, 1.5098, 4.5539, 1.8235, 2.2838], + device='cuda:1'), covar=tensor([0.3329, 0.2662, 0.2157, 0.2387, 0.1654, 0.0143, 0.2434, 0.1137], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0121, 0.0124, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 08:13:25,693 INFO [finetune.py:976] (1/7) Epoch 26, batch 5650, loss[loss=0.1719, simple_loss=0.2486, pruned_loss=0.04759, over 4810.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2457, pruned_loss=0.04926, over 953305.31 frames. ], batch size: 39, lr: 2.94e-03, grad_scale: 16.0 +2023-03-27 08:13:32,820 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148854.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:13:42,328 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.032e+02 1.500e+02 1.770e+02 2.150e+02 4.859e+02, threshold=3.539e+02, percent-clipped=1.0 +2023-03-27 08:13:51,287 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3003, 2.2080, 2.3288, 1.6469, 2.2898, 2.3564, 2.3767, 1.9115], + device='cuda:1'), covar=tensor([0.0528, 0.0570, 0.0645, 0.0888, 0.0868, 0.0680, 0.0661, 0.1092], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0136, 0.0141, 0.0119, 0.0128, 0.0138, 0.0139, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:13:52,147 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-27 08:13:55,304 INFO [finetune.py:976] (1/7) Epoch 26, batch 5700, loss[loss=0.1585, simple_loss=0.2063, pruned_loss=0.05533, over 4060.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2402, pruned_loss=0.04822, over 930963.00 frames. ], batch size: 17, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:14:24,188 INFO [finetune.py:976] (1/7) Epoch 27, batch 0, loss[loss=0.16, simple_loss=0.2404, pruned_loss=0.03977, over 4495.00 frames. ], tot_loss[loss=0.16, simple_loss=0.2404, pruned_loss=0.03977, over 4495.00 frames. ], batch size: 19, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:14:24,188 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 08:14:30,344 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8038, 1.2360, 0.9763, 1.7960, 2.2445, 1.2187, 1.5906, 1.6552], + device='cuda:1'), covar=tensor([0.1371, 0.1810, 0.1641, 0.1012, 0.1642, 0.1952, 0.1180, 0.1856], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 08:14:30,823 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5034, 1.3703, 1.3283, 1.4361, 1.7155, 1.6915, 1.4278, 1.3111], + device='cuda:1'), covar=tensor([0.0352, 0.0344, 0.0642, 0.0326, 0.0260, 0.0389, 0.0347, 0.0429], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0105, 0.0146, 0.0110, 0.0100, 0.0114, 0.0103, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.7685e-05, 8.0337e-05, 1.1349e-04, 8.4398e-05, 7.7821e-05, 8.4339e-05, + 7.6199e-05, 8.4954e-05], device='cuda:1') +2023-03-27 08:14:40,695 INFO [finetune.py:1010] (1/7) Epoch 27, validation: loss=0.1593, simple_loss=0.2269, pruned_loss=0.04586, over 2265189.00 frames. +2023-03-27 08:14:40,695 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 08:14:57,035 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148943.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:15:20,254 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 08:15:27,450 INFO [finetune.py:976] (1/7) Epoch 27, batch 50, loss[loss=0.1745, simple_loss=0.2506, pruned_loss=0.04918, over 4756.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2455, pruned_loss=0.049, over 215740.87 frames. ], batch size: 27, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:15:28,074 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.253e+01 1.427e+02 1.731e+02 2.058e+02 3.661e+02, threshold=3.462e+02, percent-clipped=4.0 +2023-03-27 08:15:35,636 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.5887, 3.9685, 4.2233, 4.4747, 4.3066, 4.0868, 4.6828, 1.4467], + device='cuda:1'), covar=tensor([0.0767, 0.0867, 0.0890, 0.0894, 0.1222, 0.1540, 0.0639, 0.5816], + device='cuda:1'), in_proj_covar=tensor([0.0354, 0.0250, 0.0285, 0.0298, 0.0338, 0.0289, 0.0307, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:15:44,091 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=148991.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:15:54,004 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=149007.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:16:03,651 INFO [finetune.py:976] (1/7) Epoch 27, batch 100, loss[loss=0.1607, simple_loss=0.2251, pruned_loss=0.04812, over 4755.00 frames. ], tot_loss[loss=0.1647, simple_loss=0.2379, pruned_loss=0.04578, over 381748.04 frames. ], batch size: 54, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:16:13,173 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0841, 1.9389, 2.0735, 1.3737, 2.0346, 2.0900, 2.0791, 1.6706], + device='cuda:1'), covar=tensor([0.0550, 0.0714, 0.0687, 0.0845, 0.0790, 0.0718, 0.0575, 0.1210], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0136, 0.0141, 0.0119, 0.0128, 0.0139, 0.0140, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:16:23,431 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8813, 4.1149, 3.8214, 1.9204, 4.2029, 3.1188, 0.8810, 3.0590], + device='cuda:1'), covar=tensor([0.2231, 0.1862, 0.1678, 0.3311, 0.1187, 0.1034, 0.4758, 0.1368], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0179, 0.0161, 0.0129, 0.0160, 0.0124, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 08:16:25,486 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.30 vs. limit=5.0 +2023-03-27 08:16:36,428 INFO [finetune.py:976] (1/7) Epoch 27, batch 150, loss[loss=0.1353, simple_loss=0.2054, pruned_loss=0.03259, over 4821.00 frames. ], tot_loss[loss=0.1614, simple_loss=0.2335, pruned_loss=0.04468, over 510750.14 frames. ], batch size: 40, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:16:37,488 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.570e+01 1.451e+02 1.770e+02 2.054e+02 3.397e+02, threshold=3.539e+02, percent-clipped=0.0 +2023-03-27 08:16:39,141 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=149072.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:16:42,923 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.25 vs. limit=5.0 +2023-03-27 08:16:47,475 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=149085.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:17:09,497 INFO [finetune.py:976] (1/7) Epoch 27, batch 200, loss[loss=0.1242, simple_loss=0.1918, pruned_loss=0.02828, over 4827.00 frames. ], tot_loss[loss=0.1589, simple_loss=0.2306, pruned_loss=0.04363, over 609362.67 frames. ], batch size: 30, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:17:19,403 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=149133.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:17:39,150 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=149154.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:17:52,876 INFO [finetune.py:976] (1/7) Epoch 27, batch 250, loss[loss=0.1622, simple_loss=0.2417, pruned_loss=0.04136, over 4748.00 frames. ], tot_loss[loss=0.1641, simple_loss=0.2356, pruned_loss=0.04635, over 685839.33 frames. ], batch size: 27, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:17:53,480 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.507e+01 1.547e+02 1.763e+02 2.073e+02 3.560e+02, threshold=3.526e+02, percent-clipped=1.0 +2023-03-27 08:18:03,844 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5965, 1.6373, 1.4171, 1.4897, 1.9901, 1.9290, 1.6208, 1.4244], + device='cuda:1'), covar=tensor([0.0395, 0.0379, 0.0694, 0.0409, 0.0243, 0.0465, 0.0429, 0.0516], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0107, 0.0148, 0.0112, 0.0102, 0.0116, 0.0105, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8960e-05, 8.1880e-05, 1.1554e-04, 8.5852e-05, 7.8998e-05, 8.5679e-05, + 7.8003e-05, 8.6201e-05], device='cuda:1') +2023-03-27 08:18:08,129 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-27 08:18:09,387 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.88 vs. limit=5.0 +2023-03-27 08:18:14,780 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=149202.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:18:23,975 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.11 vs. limit=2.0 +2023-03-27 08:18:25,585 INFO [finetune.py:976] (1/7) Epoch 27, batch 300, loss[loss=0.212, simple_loss=0.2795, pruned_loss=0.07221, over 4823.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2428, pruned_loss=0.0488, over 746634.57 frames. ], batch size: 40, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:18:28,683 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-27 08:18:55,219 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4823, 3.7791, 4.0534, 4.2728, 4.1995, 3.8710, 4.5200, 1.4160], + device='cuda:1'), covar=tensor([0.0718, 0.0906, 0.0896, 0.0920, 0.1197, 0.1693, 0.0645, 0.5714], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0248, 0.0284, 0.0296, 0.0336, 0.0288, 0.0305, 0.0299], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:18:58,793 INFO [finetune.py:976] (1/7) Epoch 27, batch 350, loss[loss=0.1627, simple_loss=0.2458, pruned_loss=0.03975, over 4825.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2458, pruned_loss=0.0496, over 794585.94 frames. ], batch size: 30, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:18:59,397 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.129e+02 1.601e+02 1.876e+02 2.140e+02 5.128e+02, threshold=3.753e+02, percent-clipped=1.0 +2023-03-27 08:19:14,647 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3246, 2.2308, 1.7414, 2.3803, 2.2803, 1.9864, 2.6114, 2.3135], + device='cuda:1'), covar=tensor([0.1394, 0.2113, 0.3089, 0.2426, 0.2559, 0.1726, 0.2804, 0.1798], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0192, 0.0240, 0.0256, 0.0252, 0.0209, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:19:24,859 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=149307.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:19:32,719 INFO [finetune.py:976] (1/7) Epoch 27, batch 400, loss[loss=0.1411, simple_loss=0.2262, pruned_loss=0.02801, over 4817.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2458, pruned_loss=0.04925, over 829217.99 frames. ], batch size: 38, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:19:34,665 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.2422, 1.2382, 1.2898, 0.7400, 1.2644, 1.4024, 1.5684, 1.2438], + device='cuda:1'), covar=tensor([0.0784, 0.0524, 0.0498, 0.0402, 0.0483, 0.0567, 0.0265, 0.0561], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0128, 0.0122, 0.0130, 0.0129, 0.0141, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.7996e-05, 1.0552e-04, 9.1176e-05, 8.5589e-05, 9.0718e-05, 9.1437e-05, + 1.0023e-04, 1.0660e-04], device='cuda:1') +2023-03-27 08:19:35,437 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.14 vs. limit=5.0 +2023-03-27 08:20:07,339 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=149355.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:20:07,378 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5137, 1.1343, 0.7453, 1.3963, 2.0383, 0.7510, 1.3103, 1.4402], + device='cuda:1'), covar=tensor([0.1579, 0.2099, 0.1758, 0.1212, 0.1906, 0.1974, 0.1490, 0.2003], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0092, 0.0120, 0.0094, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 08:20:16,468 INFO [finetune.py:976] (1/7) Epoch 27, batch 450, loss[loss=0.1553, simple_loss=0.2236, pruned_loss=0.04351, over 4699.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2445, pruned_loss=0.04892, over 856004.02 frames. ], batch size: 23, lr: 2.94e-03, grad_scale: 8.0 +2023-03-27 08:20:17,064 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.956e+01 1.496e+02 1.736e+02 2.126e+02 4.914e+02, threshold=3.471e+02, percent-clipped=1.0 +2023-03-27 08:20:17,773 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=149372.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:21:04,694 INFO [finetune.py:976] (1/7) Epoch 27, batch 500, loss[loss=0.1432, simple_loss=0.2239, pruned_loss=0.03128, over 4903.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2413, pruned_loss=0.04827, over 876267.83 frames. ], batch size: 36, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:21:04,756 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=149420.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:21:38,463 INFO [finetune.py:976] (1/7) Epoch 27, batch 550, loss[loss=0.1964, simple_loss=0.2654, pruned_loss=0.06371, over 4815.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2395, pruned_loss=0.04808, over 892308.89 frames. ], batch size: 45, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:21:39,060 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.918e+01 1.466e+02 1.717e+02 2.125e+02 3.295e+02, threshold=3.435e+02, percent-clipped=0.0 +2023-03-27 08:22:08,676 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=149514.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:22:12,133 INFO [finetune.py:976] (1/7) Epoch 27, batch 600, loss[loss=0.175, simple_loss=0.2483, pruned_loss=0.05085, over 4898.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2405, pruned_loss=0.04896, over 908410.46 frames. ], batch size: 32, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:22:40,002 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6437, 1.5725, 1.9452, 3.2424, 2.2459, 2.2661, 0.9401, 2.8033], + device='cuda:1'), covar=tensor([0.1569, 0.1300, 0.1259, 0.0637, 0.0772, 0.1391, 0.1791, 0.0436], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0131, 0.0163, 0.0100, 0.0135, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 08:22:45,157 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=149565.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:22:48,083 INFO [finetune.py:976] (1/7) Epoch 27, batch 650, loss[loss=0.211, simple_loss=0.2839, pruned_loss=0.06908, over 4848.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2436, pruned_loss=0.04982, over 918872.38 frames. ], batch size: 44, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:22:53,183 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.097e+02 1.592e+02 1.975e+02 2.434e+02 4.045e+02, threshold=3.949e+02, percent-clipped=4.0 +2023-03-27 08:22:55,787 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=149575.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:23:29,819 INFO [finetune.py:976] (1/7) Epoch 27, batch 700, loss[loss=0.2109, simple_loss=0.2974, pruned_loss=0.06215, over 4882.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2442, pruned_loss=0.0496, over 925552.81 frames. ], batch size: 43, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:23:33,644 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=149626.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 08:23:34,261 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1991, 1.3113, 1.3957, 0.6960, 1.3730, 1.5142, 1.5578, 1.2982], + device='cuda:1'), covar=tensor([0.0873, 0.0737, 0.0537, 0.0547, 0.0595, 0.0710, 0.0388, 0.0696], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0128, 0.0122, 0.0130, 0.0129, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.8205e-05, 1.0592e-04, 9.1076e-05, 8.6032e-05, 9.1016e-05, 9.1457e-05, + 1.0076e-04, 1.0694e-04], device='cuda:1') +2023-03-27 08:23:49,509 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5583, 1.5968, 1.2595, 1.5016, 1.8960, 1.8149, 1.5115, 1.3473], + device='cuda:1'), covar=tensor([0.0372, 0.0342, 0.0698, 0.0343, 0.0222, 0.0494, 0.0414, 0.0480], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0146, 0.0111, 0.0101, 0.0115, 0.0103, 0.0112], + device='cuda:1'), out_proj_covar=tensor([7.7944e-05, 8.0678e-05, 1.1404e-04, 8.4701e-05, 7.8103e-05, 8.4929e-05, + 7.6760e-05, 8.5247e-05], device='cuda:1') +2023-03-27 08:23:54,852 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-03-27 08:24:03,074 INFO [finetune.py:976] (1/7) Epoch 27, batch 750, loss[loss=0.191, simple_loss=0.2625, pruned_loss=0.05975, over 4848.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2456, pruned_loss=0.05008, over 930585.20 frames. ], batch size: 44, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:24:03,698 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.520e+02 1.783e+02 2.094e+02 3.998e+02, threshold=3.567e+02, percent-clipped=1.0 +2023-03-27 08:24:36,874 INFO [finetune.py:976] (1/7) Epoch 27, batch 800, loss[loss=0.1571, simple_loss=0.2261, pruned_loss=0.04406, over 4866.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2442, pruned_loss=0.04926, over 934146.50 frames. ], batch size: 31, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:24:47,264 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5339, 1.4103, 1.5532, 0.9425, 1.5163, 1.5866, 1.5322, 1.3536], + device='cuda:1'), covar=tensor([0.0465, 0.0651, 0.0556, 0.0813, 0.0929, 0.0532, 0.0543, 0.1076], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0137, 0.0141, 0.0119, 0.0129, 0.0139, 0.0141, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:24:53,181 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=149746.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 08:25:15,255 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8771, 1.8466, 1.6061, 2.0700, 2.4000, 2.0932, 1.8829, 1.4981], + device='cuda:1'), covar=tensor([0.2168, 0.1847, 0.1851, 0.1460, 0.1733, 0.1175, 0.2101, 0.1860], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0210, 0.0214, 0.0197, 0.0244, 0.0191, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:25:15,870 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2290, 2.2259, 1.9900, 2.3815, 2.6915, 2.3224, 2.4555, 1.7765], + device='cuda:1'), covar=tensor([0.2087, 0.1760, 0.1832, 0.1387, 0.1643, 0.1043, 0.1713, 0.1743], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0210, 0.0214, 0.0197, 0.0244, 0.0191, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:25:20,609 INFO [finetune.py:976] (1/7) Epoch 27, batch 850, loss[loss=0.1947, simple_loss=0.2542, pruned_loss=0.06757, over 4916.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.242, pruned_loss=0.04892, over 938134.71 frames. ], batch size: 37, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:25:21,208 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.021e+02 1.421e+02 1.714e+02 1.950e+02 4.580e+02, threshold=3.429e+02, percent-clipped=2.0 +2023-03-27 08:25:50,861 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5163, 1.3945, 1.2922, 1.5746, 1.4803, 1.5155, 0.9887, 1.3111], + device='cuda:1'), covar=tensor([0.1802, 0.1676, 0.1665, 0.1438, 0.1367, 0.1018, 0.2245, 0.1583], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0211, 0.0214, 0.0198, 0.0245, 0.0191, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:25:56,767 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=149807.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 08:26:09,033 INFO [finetune.py:976] (1/7) Epoch 27, batch 900, loss[loss=0.1431, simple_loss=0.2057, pruned_loss=0.04021, over 4732.00 frames. ], tot_loss[loss=0.1685, simple_loss=0.2399, pruned_loss=0.04853, over 942724.82 frames. ], batch size: 54, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:26:21,302 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.6901, 1.6526, 1.6517, 0.9684, 1.8361, 2.1069, 1.9353, 1.5164], + device='cuda:1'), covar=tensor([0.0993, 0.0809, 0.0587, 0.0659, 0.0500, 0.0568, 0.0389, 0.0805], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0129, 0.0123, 0.0131, 0.0130, 0.0143, 0.0151], + device='cuda:1'), out_proj_covar=tensor([8.8820e-05, 1.0642e-04, 9.2058e-05, 8.6621e-05, 9.1850e-05, 9.2317e-05, + 1.0165e-04, 1.0798e-04], device='cuda:1') +2023-03-27 08:26:42,230 INFO [finetune.py:976] (1/7) Epoch 27, batch 950, loss[loss=0.1993, simple_loss=0.2694, pruned_loss=0.06462, over 4812.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.2384, pruned_loss=0.0484, over 947599.70 frames. ], batch size: 45, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:26:42,299 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=149870.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:26:42,815 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.619e+01 1.503e+02 1.866e+02 2.296e+02 3.689e+02, threshold=3.732e+02, percent-clipped=3.0 +2023-03-27 08:27:15,533 INFO [finetune.py:976] (1/7) Epoch 27, batch 1000, loss[loss=0.2168, simple_loss=0.2866, pruned_loss=0.0735, over 4898.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2419, pruned_loss=0.04955, over 948144.93 frames. ], batch size: 43, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:27:16,176 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=149921.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 08:27:48,833 INFO [finetune.py:976] (1/7) Epoch 27, batch 1050, loss[loss=0.2587, simple_loss=0.3168, pruned_loss=0.1003, over 4809.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2448, pruned_loss=0.0499, over 951033.61 frames. ], batch size: 45, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:27:49,418 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.130e+02 1.561e+02 1.767e+02 2.240e+02 3.870e+02, threshold=3.534e+02, percent-clipped=1.0 +2023-03-27 08:28:33,249 INFO [finetune.py:976] (1/7) Epoch 27, batch 1100, loss[loss=0.1321, simple_loss=0.2137, pruned_loss=0.02526, over 4884.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2468, pruned_loss=0.05033, over 952446.41 frames. ], batch size: 43, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:29:06,475 INFO [finetune.py:976] (1/7) Epoch 27, batch 1150, loss[loss=0.181, simple_loss=0.2614, pruned_loss=0.05036, over 4918.00 frames. ], tot_loss[loss=0.1742, simple_loss=0.2471, pruned_loss=0.05067, over 953311.13 frames. ], batch size: 38, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:29:07,079 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.913e+01 1.470e+02 1.766e+02 2.217e+02 3.439e+02, threshold=3.531e+02, percent-clipped=0.0 +2023-03-27 08:29:13,024 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=150079.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:29:26,977 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=150102.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 08:29:39,274 INFO [finetune.py:976] (1/7) Epoch 27, batch 1200, loss[loss=0.209, simple_loss=0.2815, pruned_loss=0.06829, over 4728.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2455, pruned_loss=0.04992, over 955333.48 frames. ], batch size: 54, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:29:40,447 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9802, 1.8409, 2.3122, 3.2810, 2.4740, 2.4970, 1.2974, 2.8388], + device='cuda:1'), covar=tensor([0.1365, 0.1190, 0.1152, 0.0609, 0.0660, 0.1286, 0.1586, 0.0512], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0131, 0.0162, 0.0100, 0.0135, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 08:29:51,697 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9318, 4.2121, 3.9147, 1.9651, 4.3892, 3.2172, 1.1867, 3.1081], + device='cuda:1'), covar=tensor([0.2113, 0.1864, 0.1652, 0.3347, 0.0940, 0.1000, 0.4092, 0.1410], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0178, 0.0159, 0.0129, 0.0160, 0.0123, 0.0148, 0.0124], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 08:29:52,953 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=150140.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:30:14,458 INFO [finetune.py:976] (1/7) Epoch 27, batch 1250, loss[loss=0.1216, simple_loss=0.1902, pruned_loss=0.02655, over 4814.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2426, pruned_loss=0.04932, over 953944.14 frames. ], batch size: 51, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:30:15,063 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=150170.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:30:15,534 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.012e+02 1.554e+02 1.886e+02 2.235e+02 6.588e+02, threshold=3.772e+02, percent-clipped=2.0 +2023-03-27 08:30:56,374 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=150218.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:30:57,557 INFO [finetune.py:976] (1/7) Epoch 27, batch 1300, loss[loss=0.1853, simple_loss=0.2486, pruned_loss=0.06103, over 4219.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2405, pruned_loss=0.04839, over 955401.16 frames. ], batch size: 65, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:31:02,850 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=150221.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:31:42,208 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=150269.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:31:42,763 INFO [finetune.py:976] (1/7) Epoch 27, batch 1350, loss[loss=0.1772, simple_loss=0.2535, pruned_loss=0.05045, over 4900.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2393, pruned_loss=0.04809, over 955150.64 frames. ], batch size: 35, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:31:43,345 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.009e+02 1.453e+02 1.768e+02 2.125e+02 3.830e+02, threshold=3.537e+02, percent-clipped=1.0 +2023-03-27 08:31:45,772 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=150274.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:32:16,593 INFO [finetune.py:976] (1/7) Epoch 27, batch 1400, loss[loss=0.1464, simple_loss=0.2295, pruned_loss=0.03168, over 4902.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2432, pruned_loss=0.04925, over 957110.24 frames. ], batch size: 35, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:32:28,265 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=150335.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:32:32,527 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-27 08:32:49,840 INFO [finetune.py:976] (1/7) Epoch 27, batch 1450, loss[loss=0.211, simple_loss=0.2868, pruned_loss=0.06758, over 4905.00 frames. ], tot_loss[loss=0.171, simple_loss=0.2438, pruned_loss=0.04912, over 954144.48 frames. ], batch size: 36, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:32:50,436 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.045e+02 1.587e+02 1.925e+02 2.309e+02 4.827e+02, threshold=3.851e+02, percent-clipped=3.0 +2023-03-27 08:32:55,336 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-27 08:33:09,885 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8382, 1.7530, 1.9336, 1.2038, 1.8090, 1.8718, 1.9114, 1.5476], + device='cuda:1'), covar=tensor([0.0583, 0.0743, 0.0683, 0.0898, 0.0865, 0.0695, 0.0657, 0.1215], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0142, 0.0120, 0.0129, 0.0139, 0.0141, 0.0161], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:33:11,836 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=150402.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 08:33:29,002 INFO [finetune.py:976] (1/7) Epoch 27, batch 1500, loss[loss=0.1766, simple_loss=0.2588, pruned_loss=0.04718, over 4834.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2457, pruned_loss=0.05004, over 952980.60 frames. ], batch size: 47, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:33:42,492 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=150435.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:33:53,597 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=150450.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 08:34:02,118 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-03-27 08:34:05,547 INFO [finetune.py:976] (1/7) Epoch 27, batch 1550, loss[loss=0.1734, simple_loss=0.2424, pruned_loss=0.0522, over 4813.00 frames. ], tot_loss[loss=0.1735, simple_loss=0.2462, pruned_loss=0.05041, over 952621.19 frames. ], batch size: 39, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:34:06,129 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.290e+01 1.580e+02 1.863e+02 2.206e+02 4.598e+02, threshold=3.727e+02, percent-clipped=2.0 +2023-03-27 08:34:37,008 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3078, 1.4767, 1.7852, 1.6687, 1.6181, 3.2787, 1.3228, 1.5672], + device='cuda:1'), covar=tensor([0.1052, 0.1727, 0.1242, 0.0953, 0.1505, 0.0248, 0.1478, 0.1698], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0091, 0.0080, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 08:34:38,724 INFO [finetune.py:976] (1/7) Epoch 27, batch 1600, loss[loss=0.1456, simple_loss=0.2177, pruned_loss=0.03674, over 4874.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.244, pruned_loss=0.04905, over 955919.24 frames. ], batch size: 34, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:34:50,054 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=150537.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:35:02,707 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.41 vs. limit=5.0 +2023-03-27 08:35:11,518 INFO [finetune.py:976] (1/7) Epoch 27, batch 1650, loss[loss=0.1391, simple_loss=0.2164, pruned_loss=0.03089, over 4809.00 frames. ], tot_loss[loss=0.1693, simple_loss=0.2418, pruned_loss=0.04842, over 955125.88 frames. ], batch size: 25, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:35:12,131 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.130e+02 1.535e+02 1.741e+02 2.182e+02 5.670e+02, threshold=3.482e+02, percent-clipped=1.0 +2023-03-27 08:35:37,671 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=150598.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:35:54,937 INFO [finetune.py:976] (1/7) Epoch 27, batch 1700, loss[loss=0.1731, simple_loss=0.2395, pruned_loss=0.05335, over 4848.00 frames. ], tot_loss[loss=0.1663, simple_loss=0.2384, pruned_loss=0.04709, over 956422.11 frames. ], batch size: 49, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:36:01,032 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=150630.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:36:04,095 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6666, 1.5003, 2.0971, 3.4713, 2.2959, 2.4872, 0.9859, 2.9358], + device='cuda:1'), covar=tensor([0.1659, 0.1351, 0.1296, 0.0566, 0.0780, 0.1200, 0.1798, 0.0422], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0114, 0.0131, 0.0162, 0.0100, 0.0135, 0.0123, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 08:36:41,944 INFO [finetune.py:976] (1/7) Epoch 27, batch 1750, loss[loss=0.1666, simple_loss=0.244, pruned_loss=0.04465, over 4898.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2415, pruned_loss=0.0487, over 957948.18 frames. ], batch size: 35, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:36:42,539 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.512e+01 1.530e+02 1.821e+02 2.198e+02 3.521e+02, threshold=3.642e+02, percent-clipped=1.0 +2023-03-27 08:36:43,450 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-27 08:37:15,428 INFO [finetune.py:976] (1/7) Epoch 27, batch 1800, loss[loss=0.1845, simple_loss=0.2606, pruned_loss=0.05421, over 4813.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2446, pruned_loss=0.04994, over 958140.38 frames. ], batch size: 40, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:37:27,749 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=150732.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:37:33,302 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=150735.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:37:38,135 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8544, 1.8065, 1.8583, 1.0871, 1.9675, 1.8814, 1.9319, 1.5772], + device='cuda:1'), covar=tensor([0.0566, 0.0690, 0.0674, 0.0927, 0.0712, 0.0701, 0.0626, 0.1157], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0138, 0.0142, 0.0121, 0.0130, 0.0139, 0.0142, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:37:57,190 INFO [finetune.py:976] (1/7) Epoch 27, batch 1850, loss[loss=0.1944, simple_loss=0.2633, pruned_loss=0.0628, over 4922.00 frames. ], tot_loss[loss=0.1747, simple_loss=0.2467, pruned_loss=0.05129, over 957966.02 frames. ], batch size: 38, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:37:57,786 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.063e+02 1.537e+02 1.800e+02 2.248e+02 4.542e+02, threshold=3.600e+02, percent-clipped=6.0 +2023-03-27 08:38:05,038 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=150783.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:38:11,143 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=150793.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 08:38:30,227 INFO [finetune.py:976] (1/7) Epoch 27, batch 1900, loss[loss=0.1765, simple_loss=0.2534, pruned_loss=0.04981, over 4776.00 frames. ], tot_loss[loss=0.1743, simple_loss=0.2465, pruned_loss=0.05099, over 954092.64 frames. ], batch size: 51, lr: 2.93e-03, grad_scale: 8.0 +2023-03-27 08:39:14,069 INFO [finetune.py:976] (1/7) Epoch 27, batch 1950, loss[loss=0.1554, simple_loss=0.2274, pruned_loss=0.04167, over 4750.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2446, pruned_loss=0.04972, over 952994.50 frames. ], batch size: 27, lr: 2.92e-03, grad_scale: 8.0 +2023-03-27 08:39:14,655 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.033e+02 1.460e+02 1.651e+02 1.933e+02 3.642e+02, threshold=3.302e+02, percent-clipped=1.0 +2023-03-27 08:39:28,770 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=150893.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:39:33,108 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=150900.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:39:47,871 INFO [finetune.py:976] (1/7) Epoch 27, batch 2000, loss[loss=0.1819, simple_loss=0.2516, pruned_loss=0.05608, over 4855.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2428, pruned_loss=0.04923, over 954292.59 frames. ], batch size: 49, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:39:54,517 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=150930.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:40:15,357 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=150961.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:40:21,606 INFO [finetune.py:976] (1/7) Epoch 27, batch 2050, loss[loss=0.1322, simple_loss=0.2012, pruned_loss=0.03157, over 4759.00 frames. ], tot_loss[loss=0.1685, simple_loss=0.2398, pruned_loss=0.04859, over 954544.89 frames. ], batch size: 28, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:40:22,190 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.304e+01 1.432e+02 1.658e+02 2.071e+02 3.830e+02, threshold=3.317e+02, percent-clipped=1.0 +2023-03-27 08:40:27,052 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=150978.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:40:56,341 INFO [finetune.py:976] (1/7) Epoch 27, batch 2100, loss[loss=0.1704, simple_loss=0.2462, pruned_loss=0.04732, over 4933.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.24, pruned_loss=0.04881, over 954855.44 frames. ], batch size: 38, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:41:47,142 INFO [finetune.py:976] (1/7) Epoch 27, batch 2150, loss[loss=0.1888, simple_loss=0.2663, pruned_loss=0.05563, over 4905.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2434, pruned_loss=0.04995, over 955660.72 frames. ], batch size: 36, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:41:48,292 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.039e+02 1.525e+02 1.813e+02 2.166e+02 3.448e+02, threshold=3.626e+02, percent-clipped=1.0 +2023-03-27 08:42:02,497 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151088.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 08:42:23,614 INFO [finetune.py:976] (1/7) Epoch 27, batch 2200, loss[loss=0.2075, simple_loss=0.2678, pruned_loss=0.07356, over 4828.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2443, pruned_loss=0.05021, over 953369.09 frames. ], batch size: 30, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:43:04,288 INFO [finetune.py:976] (1/7) Epoch 27, batch 2250, loss[loss=0.1627, simple_loss=0.2392, pruned_loss=0.04314, over 4827.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2449, pruned_loss=0.05007, over 951916.85 frames. ], batch size: 47, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:43:04,898 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.120e+01 1.457e+02 1.754e+02 2.221e+02 3.820e+02, threshold=3.509e+02, percent-clipped=1.0 +2023-03-27 08:43:14,780 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151184.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:43:20,786 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151193.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:43:32,989 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6519, 1.4220, 1.9939, 3.1581, 2.1424, 2.3963, 1.1840, 2.6572], + device='cuda:1'), covar=tensor([0.1646, 0.1394, 0.1246, 0.0639, 0.0815, 0.1367, 0.1576, 0.0501], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0163, 0.0101, 0.0135, 0.0125, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 08:43:37,565 INFO [finetune.py:976] (1/7) Epoch 27, batch 2300, loss[loss=0.1923, simple_loss=0.262, pruned_loss=0.06126, over 4912.00 frames. ], tot_loss[loss=0.1722, simple_loss=0.2446, pruned_loss=0.04992, over 951826.69 frames. ], batch size: 38, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:43:51,733 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=151241.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:43:54,684 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151245.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:44:06,983 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151256.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:44:18,928 INFO [finetune.py:976] (1/7) Epoch 27, batch 2350, loss[loss=0.1801, simple_loss=0.2498, pruned_loss=0.05516, over 4756.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2432, pruned_loss=0.04964, over 953473.47 frames. ], batch size: 54, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:44:19,967 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 7.208e+01 1.503e+02 1.827e+02 2.189e+02 3.264e+02, threshold=3.653e+02, percent-clipped=0.0 +2023-03-27 08:44:28,769 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151282.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:44:39,441 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151298.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:44:47,302 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8713, 1.8522, 1.8751, 1.1697, 1.9410, 1.9309, 1.8762, 1.6569], + device='cuda:1'), covar=tensor([0.0640, 0.0736, 0.0730, 0.0949, 0.0701, 0.0699, 0.0659, 0.1118], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0138, 0.0143, 0.0121, 0.0129, 0.0139, 0.0142, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:44:52,608 INFO [finetune.py:976] (1/7) Epoch 27, batch 2400, loss[loss=0.1297, simple_loss=0.1917, pruned_loss=0.03384, over 4488.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.241, pruned_loss=0.04913, over 954729.78 frames. ], batch size: 19, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:45:09,244 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151343.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:45:12,322 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3252, 2.2272, 2.0134, 2.3473, 2.3590, 2.1237, 2.6047, 2.4223], + device='cuda:1'), covar=tensor([0.1421, 0.1948, 0.2889, 0.2178, 0.2428, 0.1715, 0.2279, 0.1724], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0190, 0.0236, 0.0254, 0.0250, 0.0207, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:45:19,399 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151359.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:45:26,012 INFO [finetune.py:976] (1/7) Epoch 27, batch 2450, loss[loss=0.1604, simple_loss=0.2306, pruned_loss=0.04516, over 4816.00 frames. ], tot_loss[loss=0.1677, simple_loss=0.239, pruned_loss=0.04821, over 955631.84 frames. ], batch size: 51, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:45:26,601 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.046e+02 1.415e+02 1.689e+02 1.968e+02 4.441e+02, threshold=3.378e+02, percent-clipped=1.0 +2023-03-27 08:45:31,571 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.71 vs. limit=2.0 +2023-03-27 08:45:38,464 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151388.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:45:39,186 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-27 08:45:58,928 INFO [finetune.py:976] (1/7) Epoch 27, batch 2500, loss[loss=0.2003, simple_loss=0.268, pruned_loss=0.06627, over 4904.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2414, pruned_loss=0.04962, over 953137.38 frames. ], batch size: 32, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:46:12,814 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=151436.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:46:48,695 INFO [finetune.py:976] (1/7) Epoch 27, batch 2550, loss[loss=0.1345, simple_loss=0.2075, pruned_loss=0.03074, over 4770.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2431, pruned_loss=0.04967, over 953176.08 frames. ], batch size: 26, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:46:49,279 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.209e+01 1.472e+02 1.881e+02 2.470e+02 3.912e+02, threshold=3.762e+02, percent-clipped=2.0 +2023-03-27 08:47:01,542 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3034, 2.2800, 1.8143, 2.3288, 2.2808, 1.9396, 2.5882, 2.3932], + device='cuda:1'), covar=tensor([0.1319, 0.1941, 0.2897, 0.2468, 0.2495, 0.1727, 0.2724, 0.1637], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0187, 0.0234, 0.0251, 0.0247, 0.0205, 0.0213, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:47:17,821 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-27 08:47:24,842 INFO [finetune.py:976] (1/7) Epoch 27, batch 2600, loss[loss=0.2153, simple_loss=0.2813, pruned_loss=0.07461, over 4824.00 frames. ], tot_loss[loss=0.1733, simple_loss=0.2452, pruned_loss=0.05072, over 953734.61 frames. ], batch size: 39, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:47:42,186 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151540.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:48:03,452 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151556.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:48:11,485 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8365, 1.6676, 1.4141, 1.1956, 1.6365, 1.6706, 1.6046, 2.2094], + device='cuda:1'), covar=tensor([0.3765, 0.3655, 0.3308, 0.3618, 0.3556, 0.2430, 0.3251, 0.1917], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0263, 0.0236, 0.0274, 0.0259, 0.0229, 0.0258, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:48:16,178 INFO [finetune.py:976] (1/7) Epoch 27, batch 2650, loss[loss=0.1808, simple_loss=0.2529, pruned_loss=0.05437, over 4889.00 frames. ], tot_loss[loss=0.175, simple_loss=0.2473, pruned_loss=0.05134, over 953816.01 frames. ], batch size: 43, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:48:16,788 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.199e+02 1.649e+02 1.887e+02 2.270e+02 4.456e+02, threshold=3.774e+02, percent-clipped=3.0 +2023-03-27 08:48:39,918 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=151604.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:48:49,946 INFO [finetune.py:976] (1/7) Epoch 27, batch 2700, loss[loss=0.1586, simple_loss=0.219, pruned_loss=0.04913, over 4232.00 frames. ], tot_loss[loss=0.1749, simple_loss=0.2474, pruned_loss=0.05117, over 954424.89 frames. ], batch size: 18, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:49:01,305 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151638.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:49:12,818 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151654.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:49:29,775 INFO [finetune.py:976] (1/7) Epoch 27, batch 2750, loss[loss=0.1512, simple_loss=0.2222, pruned_loss=0.04006, over 4860.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2449, pruned_loss=0.05021, over 956467.31 frames. ], batch size: 49, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:49:30,372 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.940e+01 1.418e+02 1.693e+02 2.178e+02 3.976e+02, threshold=3.385e+02, percent-clipped=1.0 +2023-03-27 08:49:44,710 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151688.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:49:44,802 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.52 vs. limit=5.0 +2023-03-27 08:49:53,361 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2996, 1.3553, 1.5956, 1.5562, 1.5442, 2.9439, 1.2073, 1.4929], + device='cuda:1'), covar=tensor([0.1103, 0.1861, 0.1143, 0.0970, 0.1660, 0.0312, 0.1682, 0.1850], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0092, 0.0080, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 08:50:06,331 INFO [finetune.py:976] (1/7) Epoch 27, batch 2800, loss[loss=0.2224, simple_loss=0.2825, pruned_loss=0.08116, over 4312.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.242, pruned_loss=0.04942, over 955828.01 frames. ], batch size: 65, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:50:16,620 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6344, 1.5400, 1.4713, 1.5416, 1.8475, 1.8758, 1.6533, 1.3872], + device='cuda:1'), covar=tensor([0.0327, 0.0313, 0.0636, 0.0340, 0.0254, 0.0367, 0.0296, 0.0426], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0106, 0.0146, 0.0111, 0.0101, 0.0115, 0.0103, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.7334e-05, 8.0800e-05, 1.1364e-04, 8.4584e-05, 7.8030e-05, 8.4735e-05, + 7.6039e-05, 8.5548e-05], device='cuda:1') +2023-03-27 08:50:17,992 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-27 08:50:24,987 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151749.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:50:28,611 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-27 08:50:39,488 INFO [finetune.py:976] (1/7) Epoch 27, batch 2850, loss[loss=0.151, simple_loss=0.2274, pruned_loss=0.03732, over 4893.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2405, pruned_loss=0.04887, over 952102.53 frames. ], batch size: 35, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:50:40,098 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.092e+02 1.485e+02 1.795e+02 2.169e+02 3.375e+02, threshold=3.589e+02, percent-clipped=0.0 +2023-03-27 08:50:51,024 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7379, 1.6117, 1.9801, 1.2426, 1.6829, 1.9161, 1.5601, 2.1392], + device='cuda:1'), covar=tensor([0.1217, 0.2166, 0.1238, 0.1666, 0.0938, 0.1164, 0.2878, 0.0796], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0207, 0.0193, 0.0190, 0.0175, 0.0213, 0.0218, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:50:59,489 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151801.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:51:12,914 INFO [finetune.py:976] (1/7) Epoch 27, batch 2900, loss[loss=0.1804, simple_loss=0.2566, pruned_loss=0.05213, over 4813.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2442, pruned_loss=0.05026, over 954020.03 frames. ], batch size: 33, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:51:25,535 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151840.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:51:53,899 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151862.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:52:04,134 INFO [finetune.py:976] (1/7) Epoch 27, batch 2950, loss[loss=0.1483, simple_loss=0.2365, pruned_loss=0.03007, over 4806.00 frames. ], tot_loss[loss=0.1741, simple_loss=0.2466, pruned_loss=0.0508, over 954158.42 frames. ], batch size: 45, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:52:04,749 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.132e+02 1.531e+02 1.876e+02 2.281e+02 4.815e+02, threshold=3.752e+02, percent-clipped=2.0 +2023-03-27 08:52:15,638 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=151888.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:52:37,406 INFO [finetune.py:976] (1/7) Epoch 27, batch 3000, loss[loss=0.1394, simple_loss=0.2098, pruned_loss=0.03448, over 4746.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.246, pruned_loss=0.05021, over 954091.60 frames. ], batch size: 26, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:52:37,407 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 08:52:43,653 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9034, 1.7543, 1.6552, 2.0031, 2.1222, 2.0063, 1.4903, 1.6531], + device='cuda:1'), covar=tensor([0.2360, 0.2028, 0.2079, 0.1777, 0.1649, 0.1206, 0.2432, 0.1960], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0211, 0.0215, 0.0200, 0.0247, 0.0191, 0.0219, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:52:44,167 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6295, 1.5329, 2.0169, 2.9447, 2.0099, 2.3925, 1.0507, 2.5735], + device='cuda:1'), covar=tensor([0.1624, 0.1313, 0.1159, 0.0563, 0.0873, 0.1152, 0.1746, 0.0472], + device='cuda:1'), in_proj_covar=tensor([0.0098, 0.0115, 0.0132, 0.0162, 0.0100, 0.0134, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 08:52:50,758 INFO [finetune.py:1010] (1/7) Epoch 27, validation: loss=0.1572, simple_loss=0.2248, pruned_loss=0.04486, over 2265189.00 frames. +2023-03-27 08:52:50,759 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 08:52:55,100 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3274, 3.7605, 3.9970, 4.1880, 4.0803, 3.8371, 4.3998, 1.4040], + device='cuda:1'), covar=tensor([0.0720, 0.0864, 0.0821, 0.0851, 0.1166, 0.1666, 0.0780, 0.6111], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0249, 0.0282, 0.0297, 0.0335, 0.0288, 0.0307, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:52:55,725 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5059, 1.4372, 1.9934, 1.7899, 1.5976, 3.1464, 1.3191, 1.5608], + device='cuda:1'), covar=tensor([0.0934, 0.1721, 0.1184, 0.0869, 0.1464, 0.0252, 0.1413, 0.1651], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0091, 0.0080, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 08:53:01,834 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151938.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:53:13,708 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-27 08:53:14,199 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151954.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:53:32,161 INFO [finetune.py:976] (1/7) Epoch 27, batch 3050, loss[loss=0.2045, simple_loss=0.269, pruned_loss=0.07003, over 4922.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2462, pruned_loss=0.05008, over 951927.52 frames. ], batch size: 33, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:53:32,746 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.022e+02 1.546e+02 1.837e+02 2.199e+02 4.500e+02, threshold=3.674e+02, percent-clipped=2.0 +2023-03-27 08:53:44,356 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=151986.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:53:46,190 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4957, 1.5270, 2.0906, 1.8039, 1.7417, 3.2138, 1.4462, 1.7041], + device='cuda:1'), covar=tensor([0.1031, 0.1688, 0.1454, 0.0922, 0.1444, 0.0282, 0.1385, 0.1599], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0083, 0.0073, 0.0076, 0.0091, 0.0080, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 08:53:55,881 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=152002.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:54:06,689 INFO [finetune.py:976] (1/7) Epoch 27, batch 3100, loss[loss=0.1317, simple_loss=0.2027, pruned_loss=0.03035, over 4724.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2446, pruned_loss=0.04942, over 951625.69 frames. ], batch size: 23, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:54:09,308 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-03-27 08:54:23,087 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4957, 1.5379, 1.9946, 1.8186, 1.6231, 3.6174, 1.3980, 1.6313], + device='cuda:1'), covar=tensor([0.0997, 0.1841, 0.1098, 0.0902, 0.1589, 0.0261, 0.1538, 0.1811], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0083, 0.0074, 0.0077, 0.0092, 0.0080, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 08:54:23,670 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152044.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:54:41,698 INFO [finetune.py:976] (1/7) Epoch 27, batch 3150, loss[loss=0.1841, simple_loss=0.2525, pruned_loss=0.05789, over 4918.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2417, pruned_loss=0.0483, over 953302.11 frames. ], batch size: 43, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:54:42,282 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.945e+01 1.491e+02 1.827e+02 2.202e+02 3.039e+02, threshold=3.654e+02, percent-clipped=0.0 +2023-03-27 08:55:04,606 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-03-27 08:55:14,603 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3255, 2.3701, 2.3144, 1.5962, 2.3405, 2.4413, 2.4180, 2.0178], + device='cuda:1'), covar=tensor([0.0620, 0.0595, 0.0755, 0.0891, 0.0627, 0.0770, 0.0711, 0.1042], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0142, 0.0120, 0.0129, 0.0138, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:55:20,717 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1455, 2.1293, 1.9358, 2.2720, 2.4940, 2.2204, 2.3786, 1.6887], + device='cuda:1'), covar=tensor([0.2030, 0.1756, 0.1683, 0.1473, 0.1782, 0.1088, 0.1794, 0.1748], + device='cuda:1'), in_proj_covar=tensor([0.0250, 0.0214, 0.0218, 0.0203, 0.0250, 0.0194, 0.0222, 0.0209], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:55:21,810 INFO [finetune.py:976] (1/7) Epoch 27, batch 3200, loss[loss=0.1724, simple_loss=0.2445, pruned_loss=0.05018, over 4889.00 frames. ], tot_loss[loss=0.1667, simple_loss=0.2388, pruned_loss=0.04732, over 953126.50 frames. ], batch size: 32, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:55:46,785 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152157.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:55:54,655 INFO [finetune.py:976] (1/7) Epoch 27, batch 3250, loss[loss=0.1573, simple_loss=0.2236, pruned_loss=0.04552, over 4769.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2397, pruned_loss=0.04806, over 953486.84 frames. ], batch size: 28, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:55:55,263 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.795e+01 1.453e+02 1.756e+02 2.073e+02 3.538e+02, threshold=3.512e+02, percent-clipped=0.0 +2023-03-27 08:56:13,672 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1530, 2.2030, 2.1068, 1.5044, 2.1508, 2.2820, 2.2759, 1.8129], + device='cuda:1'), covar=tensor([0.0577, 0.0643, 0.0855, 0.0990, 0.0752, 0.0730, 0.0632, 0.1222], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0137, 0.0142, 0.0120, 0.0129, 0.0138, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:56:28,344 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.96 vs. limit=5.0 +2023-03-27 08:56:29,452 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 08:56:32,279 INFO [finetune.py:976] (1/7) Epoch 27, batch 3300, loss[loss=0.1774, simple_loss=0.261, pruned_loss=0.04694, over 4821.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2436, pruned_loss=0.04937, over 954910.90 frames. ], batch size: 40, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:56:35,413 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152225.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:57:13,840 INFO [finetune.py:976] (1/7) Epoch 27, batch 3350, loss[loss=0.1762, simple_loss=0.2501, pruned_loss=0.0512, over 4766.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2439, pruned_loss=0.04928, over 953464.13 frames. ], batch size: 28, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:57:14,395 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.581e+01 1.606e+02 1.884e+02 2.337e+02 3.345e+02, threshold=3.768e+02, percent-clipped=0.0 +2023-03-27 08:57:20,866 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152273.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:57:33,569 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152286.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:57:46,619 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4371, 1.6576, 1.3827, 1.5487, 1.9857, 1.9874, 1.7912, 1.7097], + device='cuda:1'), covar=tensor([0.0578, 0.0365, 0.0577, 0.0334, 0.0288, 0.0536, 0.0326, 0.0357], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0112, 0.0101, 0.0116, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8333e-05, 8.1390e-05, 1.1500e-04, 8.5254e-05, 7.8538e-05, 8.5332e-05, + 7.6825e-05, 8.6400e-05], device='cuda:1') +2023-03-27 08:58:01,023 INFO [finetune.py:976] (1/7) Epoch 27, batch 3400, loss[loss=0.176, simple_loss=0.2495, pruned_loss=0.05123, over 4889.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2435, pruned_loss=0.049, over 950956.98 frames. ], batch size: 32, lr: 2.92e-03, grad_scale: 16.0 +2023-03-27 08:58:09,651 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152334.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:58:15,623 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9026, 1.4597, 1.9541, 1.9752, 1.7342, 1.7080, 1.8562, 1.8630], + device='cuda:1'), covar=tensor([0.4632, 0.4317, 0.3562, 0.3839, 0.5154, 0.4144, 0.4867, 0.3243], + device='cuda:1'), in_proj_covar=tensor([0.0266, 0.0248, 0.0268, 0.0297, 0.0296, 0.0272, 0.0302, 0.0253], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 08:58:16,673 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=152344.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:58:36,172 INFO [finetune.py:976] (1/7) Epoch 27, batch 3450, loss[loss=0.1576, simple_loss=0.2332, pruned_loss=0.04106, over 4819.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2421, pruned_loss=0.04813, over 948805.24 frames. ], batch size: 39, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 08:58:36,740 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.916e+01 1.467e+02 1.787e+02 2.252e+02 4.149e+02, threshold=3.573e+02, percent-clipped=3.0 +2023-03-27 08:58:46,199 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.00 vs. limit=5.0 +2023-03-27 08:58:58,043 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-27 08:58:58,954 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=152392.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:58:59,612 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2285, 1.7951, 1.8737, 0.8621, 2.1755, 2.4382, 2.1244, 1.7424], + device='cuda:1'), covar=tensor([0.0956, 0.0833, 0.0632, 0.0704, 0.0664, 0.0673, 0.0490, 0.0802], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0130, 0.0123, 0.0132, 0.0131, 0.0142, 0.0151], + device='cuda:1'), out_proj_covar=tensor([8.9144e-05, 1.0625e-04, 9.2644e-05, 8.6586e-05, 9.2706e-05, 9.2695e-05, + 1.0120e-04, 1.0785e-04], device='cuda:1') +2023-03-27 08:59:18,827 INFO [finetune.py:976] (1/7) Epoch 27, batch 3500, loss[loss=0.2191, simple_loss=0.2778, pruned_loss=0.08016, over 4892.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2404, pruned_loss=0.04766, over 951434.75 frames. ], batch size: 32, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 08:59:34,549 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152445.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:59:43,766 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=152457.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 08:59:52,114 INFO [finetune.py:976] (1/7) Epoch 27, batch 3550, loss[loss=0.1374, simple_loss=0.2134, pruned_loss=0.03068, over 4773.00 frames. ], tot_loss[loss=0.165, simple_loss=0.2371, pruned_loss=0.04641, over 950896.95 frames. ], batch size: 27, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 08:59:52,706 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.528e+01 1.381e+02 1.664e+02 2.040e+02 3.997e+02, threshold=3.328e+02, percent-clipped=1.0 +2023-03-27 09:00:03,390 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3275, 2.2956, 2.2982, 1.5564, 2.3420, 2.4961, 2.3984, 2.0136], + device='cuda:1'), covar=tensor([0.0530, 0.0602, 0.0741, 0.0981, 0.0657, 0.0650, 0.0620, 0.1082], + device='cuda:1'), in_proj_covar=tensor([0.0134, 0.0139, 0.0143, 0.0121, 0.0130, 0.0140, 0.0142, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:00:25,120 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=152505.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:00:26,312 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152506.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:00:30,982 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0 +2023-03-27 09:00:36,269 INFO [finetune.py:976] (1/7) Epoch 27, batch 3600, loss[loss=0.1644, simple_loss=0.2505, pruned_loss=0.03916, over 4915.00 frames. ], tot_loss[loss=0.1621, simple_loss=0.2337, pruned_loss=0.04519, over 952425.04 frames. ], batch size: 37, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 09:00:43,664 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9028, 2.0688, 1.7402, 1.8576, 2.4552, 2.5206, 2.1996, 2.0347], + device='cuda:1'), covar=tensor([0.0433, 0.0367, 0.0637, 0.0361, 0.0251, 0.0536, 0.0347, 0.0377], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0112, 0.0101, 0.0116, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8326e-05, 8.1328e-05, 1.1535e-04, 8.5385e-05, 7.8548e-05, 8.5260e-05, + 7.6838e-05, 8.6385e-05], device='cuda:1') +2023-03-27 09:01:10,221 INFO [finetune.py:976] (1/7) Epoch 27, batch 3650, loss[loss=0.1615, simple_loss=0.2442, pruned_loss=0.03935, over 4741.00 frames. ], tot_loss[loss=0.1637, simple_loss=0.2356, pruned_loss=0.04588, over 952930.29 frames. ], batch size: 27, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 09:01:10,827 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.982e+01 1.594e+02 1.907e+02 2.265e+02 4.160e+02, threshold=3.814e+02, percent-clipped=3.0 +2023-03-27 09:01:17,556 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152581.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:01:18,211 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152582.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:01:22,486 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152589.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:01:46,436 INFO [finetune.py:976] (1/7) Epoch 27, batch 3700, loss[loss=0.1328, simple_loss=0.2122, pruned_loss=0.02667, over 4782.00 frames. ], tot_loss[loss=0.1656, simple_loss=0.2384, pruned_loss=0.04641, over 954452.44 frames. ], batch size: 26, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 09:01:48,429 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.4008, 1.3852, 1.3854, 0.8090, 1.4503, 1.6418, 1.7270, 1.3330], + device='cuda:1'), covar=tensor([0.0893, 0.0604, 0.0536, 0.0533, 0.0452, 0.0600, 0.0266, 0.0676], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0147, 0.0130, 0.0123, 0.0132, 0.0131, 0.0142, 0.0150], + device='cuda:1'), out_proj_covar=tensor([8.8925e-05, 1.0566e-04, 9.2731e-05, 8.6069e-05, 9.2187e-05, 9.2409e-05, + 1.0086e-04, 1.0769e-04], device='cuda:1') +2023-03-27 09:01:52,534 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152629.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:01:59,329 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1963, 1.8460, 2.6877, 4.1693, 2.8103, 2.8018, 1.1637, 3.5298], + device='cuda:1'), covar=tensor([0.1686, 0.1447, 0.1324, 0.0524, 0.0753, 0.1779, 0.1866, 0.0336], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0133, 0.0164, 0.0101, 0.0135, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 09:02:01,182 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152643.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:02:05,484 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152650.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:02:22,199 INFO [finetune.py:976] (1/7) Epoch 27, batch 3750, loss[loss=0.168, simple_loss=0.2289, pruned_loss=0.05355, over 4722.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2409, pruned_loss=0.04697, over 955501.50 frames. ], batch size: 23, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 09:02:22,799 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.041e+02 1.491e+02 1.751e+02 2.166e+02 4.226e+02, threshold=3.502e+02, percent-clipped=3.0 +2023-03-27 09:02:58,170 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2457, 2.0935, 1.7935, 2.2664, 2.6382, 2.2596, 2.2722, 1.6712], + device='cuda:1'), covar=tensor([0.2275, 0.1963, 0.1968, 0.1622, 0.1714, 0.1199, 0.1904, 0.1898], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0211, 0.0215, 0.0200, 0.0246, 0.0192, 0.0219, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:03:07,114 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-03-27 09:03:12,606 INFO [finetune.py:976] (1/7) Epoch 27, batch 3800, loss[loss=0.2035, simple_loss=0.2601, pruned_loss=0.07339, over 4242.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2419, pruned_loss=0.04719, over 955251.81 frames. ], batch size: 66, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 09:03:16,881 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152726.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:03:35,092 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6518, 1.5621, 1.4757, 1.6103, 1.0390, 3.6502, 1.3711, 1.7721], + device='cuda:1'), covar=tensor([0.3278, 0.2506, 0.2179, 0.2440, 0.1926, 0.0191, 0.2549, 0.1252], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0124, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 09:03:45,579 INFO [finetune.py:976] (1/7) Epoch 27, batch 3850, loss[loss=0.164, simple_loss=0.2353, pruned_loss=0.04633, over 4816.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2414, pruned_loss=0.04705, over 957318.42 frames. ], batch size: 41, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 09:03:46,653 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.206e+01 1.335e+02 1.631e+02 2.144e+02 3.589e+02, threshold=3.262e+02, percent-clipped=1.0 +2023-03-27 09:04:00,014 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152787.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:04:10,895 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6726, 1.6478, 2.0229, 1.3536, 1.7865, 1.9760, 1.4746, 2.2074], + device='cuda:1'), covar=tensor([0.1383, 0.1908, 0.1352, 0.1878, 0.0991, 0.1355, 0.2816, 0.0733], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0207, 0.0194, 0.0190, 0.0175, 0.0214, 0.0218, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:04:12,777 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152801.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:04:28,201 INFO [finetune.py:976] (1/7) Epoch 27, batch 3900, loss[loss=0.1262, simple_loss=0.2053, pruned_loss=0.0235, over 4899.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2401, pruned_loss=0.04737, over 955406.66 frames. ], batch size: 35, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 09:04:41,786 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7607, 0.7478, 1.7606, 1.7251, 1.6188, 1.5692, 1.6115, 1.7401], + device='cuda:1'), covar=tensor([0.3631, 0.3655, 0.3126, 0.3329, 0.4148, 0.3549, 0.3783, 0.2854], + device='cuda:1'), in_proj_covar=tensor([0.0266, 0.0248, 0.0268, 0.0297, 0.0295, 0.0272, 0.0301, 0.0253], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:04:47,818 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.82 vs. limit=5.0 +2023-03-27 09:05:01,432 INFO [finetune.py:976] (1/7) Epoch 27, batch 3950, loss[loss=0.1411, simple_loss=0.2085, pruned_loss=0.03682, over 4837.00 frames. ], tot_loss[loss=0.165, simple_loss=0.2371, pruned_loss=0.04645, over 953419.33 frames. ], batch size: 25, lr: 2.91e-03, grad_scale: 16.0 +2023-03-27 09:05:02,041 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.044e+02 1.440e+02 1.686e+02 2.039e+02 3.105e+02, threshold=3.372e+02, percent-clipped=0.0 +2023-03-27 09:05:09,749 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=152881.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:05:43,211 INFO [finetune.py:976] (1/7) Epoch 27, batch 4000, loss[loss=0.1498, simple_loss=0.2301, pruned_loss=0.03473, over 4755.00 frames. ], tot_loss[loss=0.1648, simple_loss=0.2366, pruned_loss=0.04653, over 952845.92 frames. ], batch size: 54, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:05:49,728 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=152929.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:05:49,761 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=152929.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:05:52,627 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7724, 2.5387, 2.3424, 1.1934, 2.3648, 2.0781, 2.0158, 2.4180], + device='cuda:1'), covar=tensor([0.0812, 0.0806, 0.1534, 0.2108, 0.1387, 0.2215, 0.1980, 0.1079], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0192, 0.0202, 0.0183, 0.0210, 0.0212, 0.0226, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:05:53,948 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-27 09:05:56,131 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152938.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:06:00,329 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152945.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:06:16,490 INFO [finetune.py:976] (1/7) Epoch 27, batch 4050, loss[loss=0.153, simple_loss=0.2295, pruned_loss=0.03826, over 4865.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2407, pruned_loss=0.04868, over 951720.15 frames. ], batch size: 31, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:06:17,094 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.013e+02 1.469e+02 1.767e+02 2.180e+02 3.425e+02, threshold=3.534e+02, percent-clipped=1.0 +2023-03-27 09:06:20,835 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=152977.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:06:49,253 INFO [finetune.py:976] (1/7) Epoch 27, batch 4100, loss[loss=0.1571, simple_loss=0.2375, pruned_loss=0.03832, over 4915.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2436, pruned_loss=0.04932, over 952168.50 frames. ], batch size: 36, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:07:05,554 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7075, 1.0253, 1.7603, 1.7252, 1.5374, 1.4742, 1.6414, 1.7058], + device='cuda:1'), covar=tensor([0.3371, 0.3654, 0.2904, 0.3175, 0.4029, 0.3658, 0.3739, 0.2641], + device='cuda:1'), in_proj_covar=tensor([0.0267, 0.0248, 0.0269, 0.0298, 0.0296, 0.0273, 0.0302, 0.0253], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:07:15,797 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.73 vs. limit=2.0 +2023-03-27 09:07:22,842 INFO [finetune.py:976] (1/7) Epoch 27, batch 4150, loss[loss=0.2061, simple_loss=0.2707, pruned_loss=0.07076, over 4841.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2447, pruned_loss=0.04919, over 952290.15 frames. ], batch size: 49, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:07:23,441 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.058e+02 1.644e+02 1.926e+02 2.373e+02 3.999e+02, threshold=3.851e+02, percent-clipped=3.0 +2023-03-27 09:07:31,208 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=153082.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:07:32,456 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2914, 1.9342, 2.4791, 1.6784, 2.1946, 2.5087, 1.6879, 2.5886], + device='cuda:1'), covar=tensor([0.1259, 0.1970, 0.1473, 0.1889, 0.0971, 0.1392, 0.2762, 0.0746], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0207, 0.0193, 0.0189, 0.0175, 0.0214, 0.0217, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:07:51,123 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153101.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:08:07,059 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153117.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:08:13,346 INFO [finetune.py:976] (1/7) Epoch 27, batch 4200, loss[loss=0.1728, simple_loss=0.2411, pruned_loss=0.05222, over 4840.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2454, pruned_loss=0.04931, over 952997.65 frames. ], batch size: 49, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:08:36,711 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=153149.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:08:38,012 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-27 09:08:49,826 INFO [finetune.py:976] (1/7) Epoch 27, batch 4250, loss[loss=0.1739, simple_loss=0.2508, pruned_loss=0.04855, over 4904.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2428, pruned_loss=0.04842, over 954271.69 frames. ], batch size: 36, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:08:50,415 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.682e+01 1.570e+02 1.909e+02 2.227e+02 3.978e+02, threshold=3.818e+02, percent-clipped=1.0 +2023-03-27 09:08:54,787 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153178.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:09:33,254 INFO [finetune.py:976] (1/7) Epoch 27, batch 4300, loss[loss=0.1493, simple_loss=0.2196, pruned_loss=0.03948, over 4895.00 frames. ], tot_loss[loss=0.1675, simple_loss=0.2396, pruned_loss=0.04764, over 954908.40 frames. ], batch size: 32, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:09:42,854 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6302, 1.6030, 1.3558, 1.5182, 1.8778, 1.8385, 1.6176, 1.3700], + device='cuda:1'), covar=tensor([0.0362, 0.0300, 0.0699, 0.0316, 0.0228, 0.0451, 0.0350, 0.0447], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0107, 0.0149, 0.0113, 0.0102, 0.0116, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8541e-05, 8.1590e-05, 1.1568e-04, 8.6001e-05, 7.8876e-05, 8.5681e-05, + 7.7433e-05, 8.6510e-05], device='cuda:1') +2023-03-27 09:09:44,697 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153238.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:09:49,327 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153245.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:09:49,356 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.2140, 1.3728, 1.3000, 0.7582, 1.2840, 1.5253, 1.5847, 1.2791], + device='cuda:1'), covar=tensor([0.0746, 0.0424, 0.0529, 0.0451, 0.0510, 0.0458, 0.0257, 0.0542], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0147, 0.0130, 0.0123, 0.0132, 0.0130, 0.0142, 0.0150], + device='cuda:1'), out_proj_covar=tensor([8.9025e-05, 1.0571e-04, 9.2473e-05, 8.6261e-05, 9.2266e-05, 9.2429e-05, + 1.0131e-04, 1.0739e-04], device='cuda:1') +2023-03-27 09:10:00,393 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.67 vs. limit=2.0 +2023-03-27 09:10:06,729 INFO [finetune.py:976] (1/7) Epoch 27, batch 4350, loss[loss=0.1561, simple_loss=0.2336, pruned_loss=0.03927, over 4828.00 frames. ], tot_loss[loss=0.1659, simple_loss=0.2377, pruned_loss=0.04708, over 955351.52 frames. ], batch size: 33, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:10:07,328 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.429e+01 1.453e+02 1.753e+02 2.112e+02 4.699e+02, threshold=3.507e+02, percent-clipped=1.0 +2023-03-27 09:10:09,381 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-27 09:10:11,662 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153278.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:10:16,461 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=153286.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:10:18,803 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4549, 1.4444, 1.8222, 1.7291, 1.6808, 3.2722, 1.3928, 1.6068], + device='cuda:1'), covar=tensor([0.0984, 0.1900, 0.1102, 0.0990, 0.1573, 0.0245, 0.1569, 0.1835], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0077, 0.0091, 0.0080, 0.0086, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 09:10:21,194 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=153293.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:10:41,653 INFO [finetune.py:976] (1/7) Epoch 27, batch 4400, loss[loss=0.1674, simple_loss=0.2447, pruned_loss=0.04504, over 4913.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2396, pruned_loss=0.04808, over 955671.01 frames. ], batch size: 37, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:11:01,179 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153339.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:11:23,046 INFO [finetune.py:976] (1/7) Epoch 27, batch 4450, loss[loss=0.1467, simple_loss=0.2309, pruned_loss=0.03126, over 4905.00 frames. ], tot_loss[loss=0.1693, simple_loss=0.2422, pruned_loss=0.04817, over 956113.12 frames. ], batch size: 37, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:11:23,634 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.005e+02 1.494e+02 1.793e+02 2.132e+02 3.020e+02, threshold=3.586e+02, percent-clipped=0.0 +2023-03-27 09:11:30,906 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153382.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:11:56,789 INFO [finetune.py:976] (1/7) Epoch 27, batch 4500, loss[loss=0.207, simple_loss=0.2739, pruned_loss=0.07002, over 4795.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2428, pruned_loss=0.04838, over 955966.30 frames. ], batch size: 45, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:11:57,470 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153421.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:12:03,376 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=153430.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:12:11,380 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1404, 2.0706, 2.2629, 1.3483, 2.1041, 2.2224, 2.1319, 1.8212], + device='cuda:1'), covar=tensor([0.0596, 0.0640, 0.0612, 0.0931, 0.0733, 0.0708, 0.0636, 0.1080], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0138, 0.0142, 0.0121, 0.0130, 0.0140, 0.0141, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:12:29,935 INFO [finetune.py:976] (1/7) Epoch 27, batch 4550, loss[loss=0.1861, simple_loss=0.2686, pruned_loss=0.05176, over 4756.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2435, pruned_loss=0.04814, over 953548.41 frames. ], batch size: 54, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:12:30,510 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.518e+01 1.590e+02 1.867e+02 2.229e+02 3.919e+02, threshold=3.734e+02, percent-clipped=1.0 +2023-03-27 09:12:31,761 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=153473.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:12:37,684 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153482.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:12:54,365 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2786, 2.2853, 1.7574, 2.2102, 2.1371, 1.9048, 2.5331, 2.2771], + device='cuda:1'), covar=tensor([0.1205, 0.1769, 0.2720, 0.2381, 0.2362, 0.1551, 0.2659, 0.1566], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0191, 0.0237, 0.0254, 0.0250, 0.0207, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:13:14,341 INFO [finetune.py:976] (1/7) Epoch 27, batch 4600, loss[loss=0.1262, simple_loss=0.1914, pruned_loss=0.0305, over 4338.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.244, pruned_loss=0.04851, over 953962.24 frames. ], batch size: 18, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:13:33,453 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0951, 1.9004, 2.3411, 1.5197, 2.1168, 2.3732, 1.7903, 2.4778], + device='cuda:1'), covar=tensor([0.1470, 0.2011, 0.1401, 0.1846, 0.1116, 0.1457, 0.2905, 0.0917], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0209, 0.0196, 0.0191, 0.0175, 0.0216, 0.0218, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:13:38,998 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3781, 2.3050, 1.8342, 2.2666, 2.2303, 1.9690, 2.6098, 2.3709], + device='cuda:1'), covar=tensor([0.1369, 0.2045, 0.3020, 0.2499, 0.2568, 0.1780, 0.2758, 0.1821], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0191, 0.0237, 0.0254, 0.0250, 0.0207, 0.0216, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:13:41,975 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5989, 1.1132, 0.7554, 1.4618, 2.1102, 1.0470, 1.4071, 1.5179], + device='cuda:1'), covar=tensor([0.1578, 0.2238, 0.2049, 0.1182, 0.1887, 0.1969, 0.1541, 0.1895], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0093, 0.0109, 0.0091, 0.0119, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 09:13:56,974 INFO [finetune.py:976] (1/7) Epoch 27, batch 4650, loss[loss=0.1449, simple_loss=0.2132, pruned_loss=0.03829, over 4779.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2424, pruned_loss=0.04847, over 956163.15 frames. ], batch size: 29, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:13:57,586 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.034e+02 1.461e+02 1.737e+02 2.165e+02 3.643e+02, threshold=3.475e+02, percent-clipped=0.0 +2023-03-27 09:14:31,817 INFO [finetune.py:976] (1/7) Epoch 27, batch 4700, loss[loss=0.1593, simple_loss=0.2265, pruned_loss=0.0461, over 4778.00 frames. ], tot_loss[loss=0.1673, simple_loss=0.2395, pruned_loss=0.04753, over 958073.66 frames. ], batch size: 26, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:14:47,932 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=153634.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:14:53,327 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1945, 2.0326, 2.2266, 1.7217, 2.0153, 2.3051, 2.2936, 1.6922], + device='cuda:1'), covar=tensor([0.0470, 0.0593, 0.0597, 0.0849, 0.1064, 0.0512, 0.0462, 0.1140], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0137, 0.0140, 0.0119, 0.0128, 0.0138, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:15:12,022 INFO [finetune.py:976] (1/7) Epoch 27, batch 4750, loss[loss=0.1957, simple_loss=0.2608, pruned_loss=0.0653, over 4825.00 frames. ], tot_loss[loss=0.1662, simple_loss=0.2378, pruned_loss=0.04729, over 958703.19 frames. ], batch size: 39, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:15:13,083 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.023e+02 1.473e+02 1.795e+02 2.173e+02 4.465e+02, threshold=3.590e+02, percent-clipped=3.0 +2023-03-27 09:15:28,190 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9743, 2.0197, 1.6032, 1.8616, 1.9537, 1.8809, 1.9378, 2.5609], + device='cuda:1'), covar=tensor([0.3798, 0.3754, 0.3568, 0.3764, 0.3883, 0.2506, 0.3665, 0.1795], + device='cuda:1'), in_proj_covar=tensor([0.0291, 0.0266, 0.0239, 0.0278, 0.0263, 0.0231, 0.0261, 0.0240], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:15:45,874 INFO [finetune.py:976] (1/7) Epoch 27, batch 4800, loss[loss=0.2092, simple_loss=0.2913, pruned_loss=0.06359, over 4800.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2401, pruned_loss=0.04799, over 958720.87 frames. ], batch size: 45, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:15:59,364 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5839, 3.3902, 3.2248, 1.5301, 3.5520, 2.6303, 0.7467, 2.3063], + device='cuda:1'), covar=tensor([0.2440, 0.2546, 0.1813, 0.3887, 0.1175, 0.1170, 0.4720, 0.1820], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0179, 0.0159, 0.0130, 0.0161, 0.0123, 0.0149, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 09:16:08,222 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0661, 1.9985, 1.6091, 0.7691, 1.7338, 1.6755, 1.5394, 1.8475], + device='cuda:1'), covar=tensor([0.0783, 0.0648, 0.1147, 0.1644, 0.1044, 0.2058, 0.2072, 0.0704], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0191, 0.0201, 0.0182, 0.0210, 0.0211, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:16:25,134 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4612, 1.4920, 1.9548, 3.0056, 2.0166, 2.2814, 1.1530, 2.5143], + device='cuda:1'), covar=tensor([0.1864, 0.1407, 0.1256, 0.0589, 0.0853, 0.1220, 0.1606, 0.0581], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0116, 0.0132, 0.0163, 0.0100, 0.0135, 0.0124, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 09:16:28,411 INFO [finetune.py:976] (1/7) Epoch 27, batch 4850, loss[loss=0.2014, simple_loss=0.2765, pruned_loss=0.06316, over 4739.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2444, pruned_loss=0.04959, over 958630.53 frames. ], batch size: 59, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:16:28,977 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.100e+02 1.544e+02 1.777e+02 2.223e+02 4.381e+02, threshold=3.554e+02, percent-clipped=2.0 +2023-03-27 09:16:30,767 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153773.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:16:33,639 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=153777.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:17:00,914 INFO [finetune.py:976] (1/7) Epoch 27, batch 4900, loss[loss=0.1774, simple_loss=0.2365, pruned_loss=0.0591, over 4673.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2457, pruned_loss=0.04977, over 958001.16 frames. ], batch size: 23, lr: 2.91e-03, grad_scale: 32.0 +2023-03-27 09:17:01,618 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=153821.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:17:34,602 INFO [finetune.py:976] (1/7) Epoch 27, batch 4950, loss[loss=0.1461, simple_loss=0.232, pruned_loss=0.03011, over 4850.00 frames. ], tot_loss[loss=0.1736, simple_loss=0.247, pruned_loss=0.05016, over 955998.25 frames. ], batch size: 31, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:17:35,192 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.496e+02 1.750e+02 2.158e+02 3.393e+02, threshold=3.501e+02, percent-clipped=0.0 +2023-03-27 09:18:10,031 INFO [finetune.py:976] (1/7) Epoch 27, batch 5000, loss[loss=0.135, simple_loss=0.2031, pruned_loss=0.03343, over 4782.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2459, pruned_loss=0.05021, over 955522.73 frames. ], batch size: 26, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:18:12,426 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153923.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:18:28,725 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153934.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:19:01,715 INFO [finetune.py:976] (1/7) Epoch 27, batch 5050, loss[loss=0.1151, simple_loss=0.1752, pruned_loss=0.0275, over 4080.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2425, pruned_loss=0.04954, over 953872.36 frames. ], batch size: 17, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:19:02,311 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.210e+01 1.435e+02 1.808e+02 2.168e+02 4.775e+02, threshold=3.617e+02, percent-clipped=1.0 +2023-03-27 09:19:10,570 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=153982.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:19:11,850 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153984.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:19:11,870 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.1258, 2.8992, 2.5962, 3.4725, 3.0940, 2.8489, 3.4999, 3.0961], + device='cuda:1'), covar=tensor([0.1135, 0.1882, 0.2591, 0.1963, 0.1954, 0.1333, 0.2015, 0.1557], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0192, 0.0239, 0.0256, 0.0252, 0.0209, 0.0218, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:19:26,151 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.09 vs. limit=5.0 +2023-03-27 09:19:27,256 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4967, 1.4130, 1.3999, 1.3959, 0.8687, 2.3105, 0.8372, 1.2864], + device='cuda:1'), covar=tensor([0.3351, 0.2702, 0.2272, 0.2618, 0.1983, 0.0375, 0.2826, 0.1410], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0121, 0.0125, 0.0114, 0.0096, 0.0095, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 09:19:28,625 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-27 09:19:33,106 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7179, 1.5102, 1.9304, 1.2482, 1.6766, 1.8444, 1.4345, 2.0000], + device='cuda:1'), covar=tensor([0.1141, 0.1965, 0.1270, 0.1630, 0.0926, 0.1206, 0.2736, 0.0741], + device='cuda:1'), in_proj_covar=tensor([0.0194, 0.0209, 0.0195, 0.0191, 0.0176, 0.0215, 0.0219, 0.0200], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:19:36,570 INFO [finetune.py:976] (1/7) Epoch 27, batch 5100, loss[loss=0.1715, simple_loss=0.2393, pruned_loss=0.05187, over 4828.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2396, pruned_loss=0.04845, over 955311.83 frames. ], batch size: 30, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:20:06,452 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154049.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 09:20:19,662 INFO [finetune.py:976] (1/7) Epoch 27, batch 5150, loss[loss=0.101, simple_loss=0.1711, pruned_loss=0.01547, over 4693.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2388, pruned_loss=0.04805, over 954225.49 frames. ], batch size: 23, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:20:20,254 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.883e+01 1.500e+02 1.789e+02 2.109e+02 4.792e+02, threshold=3.578e+02, percent-clipped=3.0 +2023-03-27 09:20:22,334 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-27 09:20:23,487 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154076.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:20:24,062 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6154, 1.1245, 0.8928, 1.5697, 1.9838, 1.2838, 1.3949, 1.5291], + device='cuda:1'), covar=tensor([0.1422, 0.2075, 0.1740, 0.1065, 0.1926, 0.1905, 0.1449, 0.1718], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0093, 0.0109, 0.0091, 0.0119, 0.0092, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 09:20:24,075 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154077.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:20:28,828 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154084.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:20:46,961 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154110.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:20:53,421 INFO [finetune.py:976] (1/7) Epoch 27, batch 5200, loss[loss=0.1262, simple_loss=0.2074, pruned_loss=0.0225, over 4775.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2425, pruned_loss=0.0488, over 953850.53 frames. ], batch size: 26, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:20:56,440 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=154125.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:21:04,315 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154137.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 09:21:12,336 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154145.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:21:34,732 INFO [finetune.py:976] (1/7) Epoch 27, batch 5250, loss[loss=0.1806, simple_loss=0.2572, pruned_loss=0.052, over 4819.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2447, pruned_loss=0.04938, over 952438.57 frames. ], batch size: 38, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:21:35,331 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.122e+02 1.556e+02 1.889e+02 2.346e+02 3.556e+02, threshold=3.778e+02, percent-clipped=0.0 +2023-03-27 09:21:38,505 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2310, 2.0284, 1.7342, 2.1794, 2.6496, 2.2215, 2.2052, 1.6459], + device='cuda:1'), covar=tensor([0.2125, 0.1998, 0.1966, 0.1635, 0.1783, 0.1157, 0.1958, 0.1904], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0211, 0.0215, 0.0200, 0.0246, 0.0191, 0.0218, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:21:58,196 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9650, 4.7278, 4.4607, 2.6341, 4.7863, 3.7830, 0.8891, 3.4884], + device='cuda:1'), covar=tensor([0.2328, 0.1781, 0.1424, 0.3140, 0.0778, 0.0793, 0.4819, 0.1261], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0179, 0.0159, 0.0130, 0.0161, 0.0123, 0.0149, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 09:22:08,469 INFO [finetune.py:976] (1/7) Epoch 27, batch 5300, loss[loss=0.1564, simple_loss=0.2416, pruned_loss=0.03563, over 4910.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2456, pruned_loss=0.04933, over 952983.50 frames. ], batch size: 42, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:22:09,177 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154221.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:22:41,917 INFO [finetune.py:976] (1/7) Epoch 27, batch 5350, loss[loss=0.2202, simple_loss=0.2832, pruned_loss=0.07856, over 4908.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.2466, pruned_loss=0.04989, over 952530.28 frames. ], batch size: 38, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:22:42,509 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.999e+01 1.513e+02 1.798e+02 2.139e+02 3.270e+02, threshold=3.596e+02, percent-clipped=0.0 +2023-03-27 09:22:47,880 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154279.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:22:49,748 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154282.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 09:22:52,176 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154286.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:22:55,614 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154291.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:23:15,347 INFO [finetune.py:976] (1/7) Epoch 27, batch 5400, loss[loss=0.131, simple_loss=0.2145, pruned_loss=0.02375, over 4826.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2448, pruned_loss=0.05009, over 955005.54 frames. ], batch size: 30, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:23:19,157 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-27 09:23:33,012 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154337.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:23:42,777 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154347.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:23:44,020 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-03-27 09:23:46,782 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154352.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:24:08,868 INFO [finetune.py:976] (1/7) Epoch 27, batch 5450, loss[loss=0.1741, simple_loss=0.2451, pruned_loss=0.05157, over 4756.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2418, pruned_loss=0.04906, over 955627.95 frames. ], batch size: 54, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:24:09,460 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.752e+01 1.439e+02 1.730e+02 2.063e+02 4.741e+02, threshold=3.460e+02, percent-clipped=1.0 +2023-03-27 09:24:26,429 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154398.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 09:24:32,222 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154405.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:24:42,541 INFO [finetune.py:976] (1/7) Epoch 27, batch 5500, loss[loss=0.1462, simple_loss=0.2084, pruned_loss=0.04206, over 4892.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.238, pruned_loss=0.04778, over 955667.17 frames. ], batch size: 32, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:24:49,886 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154432.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 09:24:54,769 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154440.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:25:14,103 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.65 vs. limit=2.0 +2023-03-27 09:25:27,010 INFO [finetune.py:976] (1/7) Epoch 27, batch 5550, loss[loss=0.156, simple_loss=0.2293, pruned_loss=0.04136, over 4794.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.24, pruned_loss=0.04887, over 953662.45 frames. ], batch size: 29, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:25:27,599 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.282e+01 1.501e+02 1.802e+02 2.038e+02 5.335e+02, threshold=3.603e+02, percent-clipped=2.0 +2023-03-27 09:25:29,563 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154474.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:25:57,466 INFO [finetune.py:976] (1/7) Epoch 27, batch 5600, loss[loss=0.1938, simple_loss=0.2696, pruned_loss=0.05895, over 4897.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2435, pruned_loss=0.04931, over 955198.79 frames. ], batch size: 43, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:26:07,303 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154535.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:26:23,016 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-27 09:26:30,304 INFO [finetune.py:976] (1/7) Epoch 27, batch 5650, loss[loss=0.1922, simple_loss=0.26, pruned_loss=0.06219, over 4870.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2459, pruned_loss=0.04934, over 952721.67 frames. ], batch size: 34, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:26:30,863 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.301e+01 1.398e+02 1.740e+02 2.119e+02 4.723e+02, threshold=3.480e+02, percent-clipped=2.0 +2023-03-27 09:26:39,108 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154577.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:26:40,301 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154579.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:27:03,305 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.66 vs. limit=2.0 +2023-03-27 09:27:07,855 INFO [finetune.py:976] (1/7) Epoch 27, batch 5700, loss[loss=0.1506, simple_loss=0.2148, pruned_loss=0.04324, over 3963.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2421, pruned_loss=0.04903, over 932041.05 frames. ], batch size: 17, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:27:12,038 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=154627.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:27:12,683 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154628.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:27:20,863 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154642.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:27:34,184 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154647.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:27:34,737 INFO [finetune.py:976] (1/7) Epoch 28, batch 0, loss[loss=0.1717, simple_loss=0.2359, pruned_loss=0.05373, over 4906.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2359, pruned_loss=0.05373, over 4906.00 frames. ], batch size: 36, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:27:34,737 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 09:27:44,452 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3400, 1.3841, 1.3368, 0.8203, 1.3017, 1.5262, 1.5634, 1.2860], + device='cuda:1'), covar=tensor([0.0779, 0.0492, 0.0594, 0.0448, 0.0581, 0.0562, 0.0292, 0.0554], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0129, 0.0121, 0.0131, 0.0130, 0.0142, 0.0150], + device='cuda:1'), out_proj_covar=tensor([8.8337e-05, 1.0504e-04, 9.1824e-05, 8.5222e-05, 9.1862e-05, 9.1862e-05, + 1.0083e-04, 1.0708e-04], device='cuda:1') +2023-03-27 09:27:54,285 INFO [finetune.py:1010] (1/7) Epoch 28, validation: loss=0.1583, simple_loss=0.2265, pruned_loss=0.04511, over 2265189.00 frames. +2023-03-27 09:27:54,286 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 09:27:55,417 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9246, 1.3971, 1.0899, 1.7656, 2.3050, 1.5807, 1.7244, 1.6651], + device='cuda:1'), covar=tensor([0.1392, 0.1947, 0.1688, 0.1090, 0.1713, 0.1752, 0.1313, 0.1977], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0093, 0.0109, 0.0091, 0.0119, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 09:27:59,639 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4615, 1.3259, 1.8520, 2.6829, 1.7510, 2.1937, 0.9252, 2.3433], + device='cuda:1'), covar=tensor([0.1737, 0.1728, 0.1284, 0.0832, 0.0981, 0.1579, 0.1835, 0.0627], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0163, 0.0100, 0.0135, 0.0124, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 09:28:08,714 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.011e+02 1.488e+02 1.773e+02 2.221e+02 3.199e+02, threshold=3.546e+02, percent-clipped=0.0 +2023-03-27 09:28:19,538 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1190, 2.0470, 2.1944, 1.4230, 2.0874, 2.1941, 2.2004, 1.7249], + device='cuda:1'), covar=tensor([0.0546, 0.0551, 0.0598, 0.0866, 0.0645, 0.0632, 0.0506, 0.1067], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0138, 0.0141, 0.0120, 0.0129, 0.0140, 0.0140, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:28:21,212 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154689.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:28:24,050 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154693.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:28:27,013 INFO [finetune.py:976] (1/7) Epoch 28, batch 50, loss[loss=0.1272, simple_loss=0.1986, pruned_loss=0.02788, over 4783.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2459, pruned_loss=0.0501, over 216869.09 frames. ], batch size: 26, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:28:32,658 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154705.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 09:28:35,792 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1909, 2.2238, 1.9048, 2.3493, 2.1209, 2.1999, 2.1270, 2.9213], + device='cuda:1'), covar=tensor([0.3407, 0.4959, 0.3157, 0.3899, 0.3987, 0.2291, 0.4247, 0.1522], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0263, 0.0236, 0.0274, 0.0260, 0.0228, 0.0258, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:28:41,924 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154720.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:28:52,101 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154732.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:28:57,940 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154740.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:29:03,133 INFO [finetune.py:976] (1/7) Epoch 28, batch 100, loss[loss=0.1978, simple_loss=0.2593, pruned_loss=0.06814, over 4939.00 frames. ], tot_loss[loss=0.1664, simple_loss=0.2383, pruned_loss=0.04725, over 381612.19 frames. ], batch size: 33, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:29:10,674 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.4483, 4.7681, 4.9993, 5.0832, 4.9972, 4.7539, 5.5413, 1.9720], + device='cuda:1'), covar=tensor([0.0716, 0.1259, 0.1138, 0.1231, 0.1598, 0.1920, 0.0849, 0.7461], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0247, 0.0283, 0.0296, 0.0335, 0.0287, 0.0305, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:29:11,904 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=154753.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 09:29:13,662 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5782, 2.7230, 2.6050, 1.8861, 2.6252, 2.8534, 2.8525, 2.2039], + device='cuda:1'), covar=tensor([0.0546, 0.0505, 0.0605, 0.0770, 0.0674, 0.0576, 0.0489, 0.0997], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0138, 0.0141, 0.0120, 0.0128, 0.0139, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:29:26,904 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.078e+02 1.413e+02 1.713e+02 2.093e+02 4.180e+02, threshold=3.426e+02, percent-clipped=2.0 +2023-03-27 09:29:32,448 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=154780.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:29:33,106 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154781.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:29:37,834 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=154788.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:29:44,900 INFO [finetune.py:976] (1/7) Epoch 28, batch 150, loss[loss=0.1646, simple_loss=0.2349, pruned_loss=0.04714, over 4859.00 frames. ], tot_loss[loss=0.1643, simple_loss=0.2354, pruned_loss=0.04659, over 508547.80 frames. ], batch size: 31, lr: 2.90e-03, grad_scale: 32.0 +2023-03-27 09:29:50,543 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0 +2023-03-27 09:29:56,985 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9461, 1.8468, 1.5899, 2.0038, 2.4674, 2.1072, 1.9369, 1.6022], + device='cuda:1'), covar=tensor([0.2022, 0.1817, 0.1803, 0.1509, 0.1594, 0.1051, 0.2006, 0.1789], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0212, 0.0216, 0.0200, 0.0246, 0.0192, 0.0218, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:29:57,697 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-03-27 09:29:59,450 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.10 vs. limit=2.0 +2023-03-27 09:30:06,029 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154830.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:30:18,388 INFO [finetune.py:976] (1/7) Epoch 28, batch 200, loss[loss=0.1427, simple_loss=0.2111, pruned_loss=0.03712, over 4821.00 frames. ], tot_loss[loss=0.1615, simple_loss=0.2323, pruned_loss=0.04534, over 608524.93 frames. ], batch size: 30, lr: 2.89e-03, grad_scale: 32.0 +2023-03-27 09:30:40,611 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.190e+01 1.565e+02 1.831e+02 2.234e+02 3.641e+02, threshold=3.662e+02, percent-clipped=1.0 +2023-03-27 09:30:48,051 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154877.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 09:31:02,678 INFO [finetune.py:976] (1/7) Epoch 28, batch 250, loss[loss=0.1843, simple_loss=0.2642, pruned_loss=0.05218, over 4819.00 frames. ], tot_loss[loss=0.1635, simple_loss=0.2352, pruned_loss=0.04586, over 685890.93 frames. ], batch size: 33, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:31:14,727 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.69 vs. limit=5.0 +2023-03-27 09:31:20,576 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=154925.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:31:31,417 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154942.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:31:34,494 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154947.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:31:35,507 INFO [finetune.py:976] (1/7) Epoch 28, batch 300, loss[loss=0.2106, simple_loss=0.2845, pruned_loss=0.06836, over 4265.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2416, pruned_loss=0.04788, over 745556.90 frames. ], batch size: 65, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:31:42,227 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3529, 1.2354, 1.1966, 1.2585, 1.5414, 1.4307, 1.3338, 1.1557], + device='cuda:1'), covar=tensor([0.0363, 0.0316, 0.0638, 0.0320, 0.0242, 0.0478, 0.0325, 0.0424], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0147, 0.0111, 0.0101, 0.0115, 0.0103, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.7876e-05, 8.1050e-05, 1.1455e-04, 8.4770e-05, 7.8252e-05, 8.4632e-05, + 7.6211e-05, 8.5487e-05], device='cuda:1') +2023-03-27 09:31:51,475 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.026e+02 1.523e+02 1.869e+02 2.212e+02 3.864e+02, threshold=3.739e+02, percent-clipped=1.0 +2023-03-27 09:31:59,667 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154976.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:32:07,902 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154984.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:32:11,536 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=154990.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:32:13,397 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154993.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 09:32:15,017 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=154995.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:32:16,795 INFO [finetune.py:976] (1/7) Epoch 28, batch 350, loss[loss=0.1618, simple_loss=0.2368, pruned_loss=0.04337, over 4846.00 frames. ], tot_loss[loss=0.1693, simple_loss=0.2432, pruned_loss=0.04769, over 793983.86 frames. ], batch size: 44, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:32:28,697 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0741, 2.8814, 2.5180, 3.4517, 3.0113, 2.7639, 3.4836, 3.0484], + device='cuda:1'), covar=tensor([0.1069, 0.1792, 0.2662, 0.1873, 0.2326, 0.1574, 0.1912, 0.1505], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0190, 0.0236, 0.0254, 0.0250, 0.0207, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:32:31,990 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.7915, 3.3091, 3.5116, 3.6797, 3.5750, 3.3373, 3.8890, 1.1576], + device='cuda:1'), covar=tensor([0.0970, 0.0999, 0.0927, 0.1048, 0.1451, 0.1636, 0.0888, 0.5980], + device='cuda:1'), in_proj_covar=tensor([0.0354, 0.0248, 0.0285, 0.0297, 0.0336, 0.0288, 0.0306, 0.0304], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:32:43,048 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155037.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:32:45,383 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=155041.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:32:49,880 INFO [finetune.py:976] (1/7) Epoch 28, batch 400, loss[loss=0.2304, simple_loss=0.302, pruned_loss=0.07937, over 4831.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2446, pruned_loss=0.04822, over 830059.05 frames. ], batch size: 47, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:33:13,472 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155069.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:33:15,062 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.906e+01 1.559e+02 1.879e+02 2.352e+02 4.263e+02, threshold=3.758e+02, percent-clipped=3.0 +2023-03-27 09:33:18,275 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155076.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:33:19,629 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.60 vs. limit=5.0 +2023-03-27 09:33:31,495 INFO [finetune.py:976] (1/7) Epoch 28, batch 450, loss[loss=0.1825, simple_loss=0.2468, pruned_loss=0.0591, over 4785.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2429, pruned_loss=0.04751, over 858103.71 frames. ], batch size: 25, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:33:54,283 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155130.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:33:54,330 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155130.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:34:03,589 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.68 vs. limit=5.0 +2023-03-27 09:34:05,148 INFO [finetune.py:976] (1/7) Epoch 28, batch 500, loss[loss=0.1664, simple_loss=0.2366, pruned_loss=0.04813, over 4800.00 frames. ], tot_loss[loss=0.1664, simple_loss=0.2397, pruned_loss=0.04654, over 879216.54 frames. ], batch size: 51, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:34:28,553 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.687e+01 1.475e+02 1.683e+02 2.204e+02 4.497e+02, threshold=3.366e+02, percent-clipped=1.0 +2023-03-27 09:34:33,445 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=155178.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:34:49,250 INFO [finetune.py:976] (1/7) Epoch 28, batch 550, loss[loss=0.1373, simple_loss=0.2076, pruned_loss=0.03353, over 4908.00 frames. ], tot_loss[loss=0.1639, simple_loss=0.2366, pruned_loss=0.04559, over 896487.82 frames. ], batch size: 36, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:35:01,234 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.3121, 3.0235, 2.8230, 1.5392, 2.9912, 2.3967, 2.3279, 2.7445], + device='cuda:1'), covar=tensor([0.0940, 0.0876, 0.1620, 0.2091, 0.1623, 0.2309, 0.2101, 0.1131], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0192, 0.0202, 0.0182, 0.0210, 0.0210, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:35:11,073 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0602, 2.0545, 1.6270, 1.8561, 2.0175, 1.7310, 2.2228, 2.0657], + device='cuda:1'), covar=tensor([0.1399, 0.1845, 0.3023, 0.2539, 0.2530, 0.1745, 0.3052, 0.1663], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0189, 0.0235, 0.0253, 0.0249, 0.0206, 0.0214, 0.0201], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:35:14,006 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3372, 1.1892, 1.6034, 1.1876, 1.3685, 1.4625, 1.1725, 1.6207], + device='cuda:1'), covar=tensor([0.1180, 0.2494, 0.1161, 0.1313, 0.0946, 0.1191, 0.3434, 0.0905], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0208, 0.0194, 0.0189, 0.0174, 0.0213, 0.0219, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:35:23,081 INFO [finetune.py:976] (1/7) Epoch 28, batch 600, loss[loss=0.1924, simple_loss=0.2689, pruned_loss=0.05797, over 4912.00 frames. ], tot_loss[loss=0.1659, simple_loss=0.2385, pruned_loss=0.04667, over 910522.11 frames. ], batch size: 37, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:35:38,977 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.640e+01 1.454e+02 1.702e+02 1.998e+02 4.828e+02, threshold=3.403e+02, percent-clipped=2.0 +2023-03-27 09:35:53,266 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155284.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:36:05,097 INFO [finetune.py:976] (1/7) Epoch 28, batch 650, loss[loss=0.2212, simple_loss=0.2903, pruned_loss=0.07606, over 4906.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.2411, pruned_loss=0.04767, over 920603.78 frames. ], batch size: 35, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:36:27,995 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8025, 1.6844, 1.4554, 1.3081, 1.6167, 1.6014, 1.6135, 2.1968], + device='cuda:1'), covar=tensor([0.3895, 0.3626, 0.3128, 0.3525, 0.3784, 0.2213, 0.3376, 0.1777], + device='cuda:1'), in_proj_covar=tensor([0.0291, 0.0265, 0.0238, 0.0277, 0.0262, 0.0230, 0.0260, 0.0240], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:36:29,094 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=155332.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:36:29,104 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155332.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:36:38,732 INFO [finetune.py:976] (1/7) Epoch 28, batch 700, loss[loss=0.2031, simple_loss=0.2651, pruned_loss=0.07052, over 4832.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.244, pruned_loss=0.04908, over 926606.43 frames. ], batch size: 47, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:36:54,658 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.124e+02 1.563e+02 1.812e+02 2.261e+02 4.160e+02, threshold=3.625e+02, percent-clipped=3.0 +2023-03-27 09:36:57,796 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155376.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:37:05,077 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3497, 3.7826, 3.9993, 4.2120, 4.1130, 3.8391, 4.4281, 1.3107], + device='cuda:1'), covar=tensor([0.0842, 0.0939, 0.0882, 0.0944, 0.1338, 0.1733, 0.0762, 0.6000], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0247, 0.0284, 0.0297, 0.0335, 0.0287, 0.0305, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:37:19,492 INFO [finetune.py:976] (1/7) Epoch 28, batch 750, loss[loss=0.2028, simple_loss=0.2907, pruned_loss=0.05748, over 4926.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2456, pruned_loss=0.0499, over 933894.26 frames. ], batch size: 41, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:37:40,160 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=155424.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:37:41,309 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155425.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:37:49,651 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5124, 1.3996, 1.3979, 1.3746, 0.8829, 2.2586, 0.7184, 1.1942], + device='cuda:1'), covar=tensor([0.3302, 0.2723, 0.2187, 0.2555, 0.1927, 0.0375, 0.2784, 0.1385], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0124, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 09:37:56,711 INFO [finetune.py:976] (1/7) Epoch 28, batch 800, loss[loss=0.1274, simple_loss=0.2073, pruned_loss=0.02372, over 4730.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2444, pruned_loss=0.04857, over 940452.95 frames. ], batch size: 54, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:38:01,223 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.62 vs. limit=5.0 +2023-03-27 09:38:02,348 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155457.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:38:11,933 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.111e+02 1.495e+02 1.724e+02 1.967e+02 3.002e+02, threshold=3.447e+02, percent-clipped=0.0 +2023-03-27 09:38:39,884 INFO [finetune.py:976] (1/7) Epoch 28, batch 850, loss[loss=0.1666, simple_loss=0.2289, pruned_loss=0.05212, over 4825.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2424, pruned_loss=0.04771, over 943547.23 frames. ], batch size: 40, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:38:52,661 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155518.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:39:12,068 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6731, 2.5104, 2.3267, 1.1409, 2.4224, 2.0524, 1.7937, 2.3754], + device='cuda:1'), covar=tensor([0.0912, 0.0851, 0.1537, 0.2047, 0.1329, 0.2302, 0.2264, 0.0916], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0191, 0.0201, 0.0181, 0.0210, 0.0210, 0.0224, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:39:13,741 INFO [finetune.py:976] (1/7) Epoch 28, batch 900, loss[loss=0.1431, simple_loss=0.209, pruned_loss=0.03856, over 4837.00 frames. ], tot_loss[loss=0.1667, simple_loss=0.2394, pruned_loss=0.04697, over 947759.74 frames. ], batch size: 30, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:39:28,255 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.783e+01 1.413e+02 1.787e+02 2.288e+02 4.282e+02, threshold=3.575e+02, percent-clipped=3.0 +2023-03-27 09:39:34,789 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6381, 1.5667, 1.5079, 1.5965, 1.2042, 3.2752, 1.3022, 1.7326], + device='cuda:1'), covar=tensor([0.3213, 0.2407, 0.2085, 0.2307, 0.1685, 0.0242, 0.2644, 0.1206], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0123, 0.0112, 0.0095, 0.0093, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 09:39:54,473 INFO [finetune.py:976] (1/7) Epoch 28, batch 950, loss[loss=0.1629, simple_loss=0.2291, pruned_loss=0.04841, over 4923.00 frames. ], tot_loss[loss=0.1662, simple_loss=0.2382, pruned_loss=0.04713, over 949987.65 frames. ], batch size: 43, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:40:02,635 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155605.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:40:11,183 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5607, 1.4251, 1.3970, 1.4948, 0.9832, 2.9460, 1.0671, 1.4817], + device='cuda:1'), covar=tensor([0.3475, 0.2646, 0.2285, 0.2439, 0.1901, 0.0299, 0.2522, 0.1340], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0123, 0.0112, 0.0095, 0.0093, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 09:40:20,552 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155632.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:40:31,714 INFO [finetune.py:976] (1/7) Epoch 28, batch 1000, loss[loss=0.284, simple_loss=0.3306, pruned_loss=0.1186, over 4222.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2407, pruned_loss=0.04807, over 951740.66 frames. ], batch size: 65, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:40:37,330 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155657.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:40:42,785 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155666.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:40:45,714 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.619e+01 1.519e+02 1.814e+02 2.190e+02 3.109e+02, threshold=3.628e+02, percent-clipped=0.0 +2023-03-27 09:40:49,976 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0717, 0.9468, 0.9146, 0.3808, 0.9492, 1.0923, 1.1323, 0.9165], + device='cuda:1'), covar=tensor([0.0854, 0.0697, 0.0595, 0.0614, 0.0589, 0.0737, 0.0490, 0.0748], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0147, 0.0130, 0.0123, 0.0132, 0.0130, 0.0143, 0.0150], + device='cuda:1'), out_proj_covar=tensor([8.8758e-05, 1.0555e-04, 9.2581e-05, 8.6096e-05, 9.2301e-05, 9.2359e-05, + 1.0139e-04, 1.0743e-04], device='cuda:1') +2023-03-27 09:40:54,172 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=155680.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:41:13,977 INFO [finetune.py:976] (1/7) Epoch 28, batch 1050, loss[loss=0.1589, simple_loss=0.2247, pruned_loss=0.04653, over 4719.00 frames. ], tot_loss[loss=0.1701, simple_loss=0.2433, pruned_loss=0.04844, over 952021.75 frames. ], batch size: 23, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:41:26,782 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155718.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:41:31,003 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155725.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:41:46,713 INFO [finetune.py:976] (1/7) Epoch 28, batch 1100, loss[loss=0.1591, simple_loss=0.2406, pruned_loss=0.03882, over 4865.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2446, pruned_loss=0.04883, over 952466.71 frames. ], batch size: 35, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:41:49,119 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.0979, 3.6145, 3.7588, 3.9647, 3.8407, 3.5436, 4.1722, 1.3476], + device='cuda:1'), covar=tensor([0.0805, 0.0875, 0.1044, 0.0952, 0.1372, 0.1773, 0.0807, 0.6090], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0246, 0.0283, 0.0297, 0.0335, 0.0287, 0.0305, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:42:01,622 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.104e+02 1.629e+02 1.915e+02 2.256e+02 9.973e+02, threshold=3.830e+02, percent-clipped=2.0 +2023-03-27 09:42:02,900 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=155773.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:42:02,965 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3920, 1.2963, 1.2068, 1.3829, 1.6136, 1.5308, 1.3830, 1.2426], + device='cuda:1'), covar=tensor([0.0331, 0.0333, 0.0670, 0.0292, 0.0218, 0.0498, 0.0334, 0.0437], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0111, 0.0101, 0.0116, 0.0103, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.8543e-05, 8.1141e-05, 1.1533e-04, 8.4781e-05, 7.8351e-05, 8.5137e-05, + 7.6640e-05, 8.6105e-05], device='cuda:1') +2023-03-27 09:42:10,635 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155784.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:42:21,324 INFO [finetune.py:976] (1/7) Epoch 28, batch 1150, loss[loss=0.1436, simple_loss=0.2299, pruned_loss=0.02865, over 4838.00 frames. ], tot_loss[loss=0.172, simple_loss=0.2461, pruned_loss=0.04895, over 955288.32 frames. ], batch size: 30, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:42:39,751 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155813.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:43:01,208 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155845.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:43:02,855 INFO [finetune.py:976] (1/7) Epoch 28, batch 1200, loss[loss=0.1498, simple_loss=0.2329, pruned_loss=0.03337, over 4747.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2443, pruned_loss=0.04869, over 955901.95 frames. ], batch size: 27, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:43:18,196 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.094e+02 1.526e+02 1.746e+02 2.167e+02 3.236e+02, threshold=3.492e+02, percent-clipped=0.0 +2023-03-27 09:43:45,612 INFO [finetune.py:976] (1/7) Epoch 28, batch 1250, loss[loss=0.1455, simple_loss=0.2312, pruned_loss=0.02988, over 4826.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.2423, pruned_loss=0.04829, over 956880.57 frames. ], batch size: 33, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:44:01,993 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.11 vs. limit=2.0 +2023-03-27 09:44:15,242 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.85 vs. limit=5.0 +2023-03-27 09:44:19,205 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5701, 1.4782, 1.3687, 1.7139, 1.5411, 1.7114, 1.0541, 1.4181], + device='cuda:1'), covar=tensor([0.2240, 0.1957, 0.1891, 0.1576, 0.1752, 0.1236, 0.2677, 0.1854], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0212, 0.0215, 0.0199, 0.0246, 0.0190, 0.0217, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:44:22,108 INFO [finetune.py:976] (1/7) Epoch 28, batch 1300, loss[loss=0.1342, simple_loss=0.2062, pruned_loss=0.03107, over 4903.00 frames. ], tot_loss[loss=0.1672, simple_loss=0.2394, pruned_loss=0.04755, over 958091.33 frames. ], batch size: 43, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:44:32,015 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155961.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:44:38,006 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.818e+01 1.476e+02 1.729e+02 2.197e+02 4.050e+02, threshold=3.458e+02, percent-clipped=1.0 +2023-03-27 09:44:55,322 INFO [finetune.py:976] (1/7) Epoch 28, batch 1350, loss[loss=0.198, simple_loss=0.2657, pruned_loss=0.06514, over 4825.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2402, pruned_loss=0.04824, over 957384.81 frames. ], batch size: 33, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:45:10,151 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156013.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:45:30,428 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7141, 1.5895, 2.2506, 3.5946, 2.4232, 2.5024, 1.0663, 2.9779], + device='cuda:1'), covar=tensor([0.1812, 0.1400, 0.1355, 0.0652, 0.0785, 0.1404, 0.1923, 0.0551], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0117, 0.0134, 0.0167, 0.0101, 0.0137, 0.0126, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 09:45:31,627 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.3303, 3.0132, 3.1160, 3.2401, 3.1079, 2.9991, 3.3900, 0.9331], + device='cuda:1'), covar=tensor([0.1154, 0.1053, 0.1142, 0.1262, 0.1711, 0.1802, 0.1084, 0.6120], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0284, 0.0296, 0.0334, 0.0287, 0.0304, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:45:32,752 INFO [finetune.py:976] (1/7) Epoch 28, batch 1400, loss[loss=0.1802, simple_loss=0.2552, pruned_loss=0.05257, over 4828.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2437, pruned_loss=0.04954, over 956224.62 frames. ], batch size: 38, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:45:48,224 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156070.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:45:48,708 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.069e+02 1.532e+02 1.806e+02 2.221e+02 4.474e+02, threshold=3.612e+02, percent-clipped=3.0 +2023-03-27 09:46:06,150 INFO [finetune.py:976] (1/7) Epoch 28, batch 1450, loss[loss=0.1639, simple_loss=0.2383, pruned_loss=0.04476, over 4106.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.246, pruned_loss=0.04993, over 955698.06 frames. ], batch size: 65, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:46:15,764 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7660, 1.3107, 0.6992, 1.8346, 2.1709, 1.5132, 1.6695, 1.7499], + device='cuda:1'), covar=tensor([0.1400, 0.1950, 0.1904, 0.1070, 0.1865, 0.1881, 0.1286, 0.1703], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0109, 0.0092, 0.0120, 0.0093, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 09:46:23,135 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156113.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:46:32,730 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3307, 2.3212, 1.8051, 2.3212, 2.2620, 1.9557, 2.5379, 2.3594], + device='cuda:1'), covar=tensor([0.1374, 0.1881, 0.2934, 0.2350, 0.2459, 0.1708, 0.2958, 0.1666], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0189, 0.0235, 0.0253, 0.0249, 0.0206, 0.0214, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:46:33,991 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-27 09:46:35,163 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156131.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:46:40,535 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156140.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 09:46:45,776 INFO [finetune.py:976] (1/7) Epoch 28, batch 1500, loss[loss=0.1741, simple_loss=0.2523, pruned_loss=0.04795, over 4871.00 frames. ], tot_loss[loss=0.1737, simple_loss=0.2471, pruned_loss=0.05019, over 956535.81 frames. ], batch size: 35, lr: 2.89e-03, grad_scale: 64.0 +2023-03-27 09:46:48,917 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-27 09:46:54,175 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=156161.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:47:02,080 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.065e+02 1.585e+02 1.899e+02 2.331e+02 3.577e+02, threshold=3.799e+02, percent-clipped=0.0 +2023-03-27 09:47:04,696 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-03-27 09:47:09,485 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3937, 1.2991, 1.2236, 1.2916, 1.6297, 1.5278, 1.3812, 1.2481], + device='cuda:1'), covar=tensor([0.0325, 0.0310, 0.0644, 0.0308, 0.0224, 0.0450, 0.0302, 0.0391], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0111, 0.0101, 0.0115, 0.0103, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.8344e-05, 8.0912e-05, 1.1521e-04, 8.4887e-05, 7.7947e-05, 8.4940e-05, + 7.6758e-05, 8.6043e-05], device='cuda:1') +2023-03-27 09:47:18,928 INFO [finetune.py:976] (1/7) Epoch 28, batch 1550, loss[loss=0.1517, simple_loss=0.2179, pruned_loss=0.04272, over 4755.00 frames. ], tot_loss[loss=0.1727, simple_loss=0.2457, pruned_loss=0.04981, over 954339.55 frames. ], batch size: 23, lr: 2.89e-03, grad_scale: 32.0 +2023-03-27 09:47:23,439 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-27 09:47:30,563 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156216.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:47:59,288 INFO [finetune.py:976] (1/7) Epoch 28, batch 1600, loss[loss=0.1676, simple_loss=0.2282, pruned_loss=0.05349, over 4786.00 frames. ], tot_loss[loss=0.1697, simple_loss=0.2421, pruned_loss=0.0486, over 954746.27 frames. ], batch size: 26, lr: 2.89e-03, grad_scale: 32.0 +2023-03-27 09:48:08,067 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8167, 1.7606, 1.9291, 1.2036, 1.8630, 1.9221, 1.9298, 1.5950], + device='cuda:1'), covar=tensor([0.0542, 0.0675, 0.0610, 0.0866, 0.0818, 0.0635, 0.0564, 0.1096], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0139, 0.0141, 0.0119, 0.0128, 0.0139, 0.0140, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:48:08,681 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156261.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:48:15,839 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.043e+02 1.435e+02 1.767e+02 2.111e+02 3.704e+02, threshold=3.535e+02, percent-clipped=0.0 +2023-03-27 09:48:19,966 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156277.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:48:27,858 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156290.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:48:32,610 INFO [finetune.py:976] (1/7) Epoch 28, batch 1650, loss[loss=0.1679, simple_loss=0.2361, pruned_loss=0.04991, over 4828.00 frames. ], tot_loss[loss=0.1667, simple_loss=0.2387, pruned_loss=0.04731, over 955178.64 frames. ], batch size: 39, lr: 2.89e-03, grad_scale: 32.0 +2023-03-27 09:48:40,815 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=156309.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:48:41,474 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0482, 1.8587, 2.5120, 1.7358, 2.2062, 2.4864, 1.8084, 2.6370], + device='cuda:1'), covar=tensor([0.1298, 0.2083, 0.1212, 0.1854, 0.0959, 0.1233, 0.2731, 0.0686], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0207, 0.0192, 0.0189, 0.0174, 0.0212, 0.0218, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:48:42,698 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1288, 2.0212, 1.7893, 2.0438, 1.9746, 2.0037, 1.9589, 2.7380], + device='cuda:1'), covar=tensor([0.3761, 0.4383, 0.3177, 0.3820, 0.4020, 0.2363, 0.3779, 0.1582], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0264, 0.0237, 0.0276, 0.0261, 0.0230, 0.0259, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:48:43,242 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156313.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:49:18,367 INFO [finetune.py:976] (1/7) Epoch 28, batch 1700, loss[loss=0.1886, simple_loss=0.2564, pruned_loss=0.06046, over 4859.00 frames. ], tot_loss[loss=0.1658, simple_loss=0.2373, pruned_loss=0.04718, over 956086.74 frames. ], batch size: 44, lr: 2.88e-03, grad_scale: 32.0 +2023-03-27 09:49:20,333 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156351.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:49:26,960 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=156361.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:49:34,437 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.260e+01 1.525e+02 1.796e+02 2.204e+02 4.546e+02, threshold=3.593e+02, percent-clipped=3.0 +2023-03-27 09:49:51,288 INFO [finetune.py:976] (1/7) Epoch 28, batch 1750, loss[loss=0.1838, simple_loss=0.2531, pruned_loss=0.05725, over 4888.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.2407, pruned_loss=0.04843, over 955938.58 frames. ], batch size: 35, lr: 2.88e-03, grad_scale: 32.0 +2023-03-27 09:50:09,513 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156426.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:50:18,801 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9383, 4.3379, 4.5582, 4.7705, 4.6795, 4.4294, 5.0360, 1.7240], + device='cuda:1'), covar=tensor([0.0787, 0.0870, 0.0850, 0.0980, 0.1160, 0.1625, 0.0544, 0.5908], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0246, 0.0284, 0.0295, 0.0334, 0.0288, 0.0305, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:50:19,435 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156440.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 09:50:24,162 INFO [finetune.py:976] (1/7) Epoch 28, batch 1800, loss[loss=0.2309, simple_loss=0.2895, pruned_loss=0.08622, over 4799.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.243, pruned_loss=0.04886, over 955335.46 frames. ], batch size: 51, lr: 2.88e-03, grad_scale: 32.0 +2023-03-27 09:50:27,907 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1362, 0.9266, 0.9005, 0.5040, 0.8051, 1.0438, 1.0802, 0.9097], + device='cuda:1'), covar=tensor([0.0764, 0.0480, 0.0525, 0.0491, 0.0592, 0.0569, 0.0332, 0.0548], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0130, 0.0122, 0.0131, 0.0130, 0.0142, 0.0149], + device='cuda:1'), out_proj_covar=tensor([8.8238e-05, 1.0516e-04, 9.2254e-05, 8.5614e-05, 9.1453e-05, 9.1925e-05, + 1.0076e-04, 1.0666e-04], device='cuda:1') +2023-03-27 09:50:33,745 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5164, 1.5082, 1.8741, 1.8029, 1.5533, 3.4113, 1.3905, 1.6208], + device='cuda:1'), covar=tensor([0.1041, 0.1868, 0.1167, 0.0967, 0.1684, 0.0225, 0.1597, 0.1922], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0091, 0.0080, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 09:50:39,986 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.070e+02 1.522e+02 1.851e+02 2.291e+02 4.651e+02, threshold=3.702e+02, percent-clipped=5.0 +2023-03-27 09:50:43,352 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-03-27 09:50:51,502 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=156488.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:50:57,541 INFO [finetune.py:976] (1/7) Epoch 28, batch 1850, loss[loss=0.146, simple_loss=0.2368, pruned_loss=0.02754, over 4811.00 frames. ], tot_loss[loss=0.1715, simple_loss=0.2442, pruned_loss=0.04937, over 954497.85 frames. ], batch size: 40, lr: 2.88e-03, grad_scale: 32.0 +2023-03-27 09:51:40,567 INFO [finetune.py:976] (1/7) Epoch 28, batch 1900, loss[loss=0.1817, simple_loss=0.252, pruned_loss=0.05569, over 4893.00 frames. ], tot_loss[loss=0.1732, simple_loss=0.246, pruned_loss=0.05016, over 954801.89 frames. ], batch size: 36, lr: 2.88e-03, grad_scale: 32.0 +2023-03-27 09:51:56,029 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.831e+01 1.540e+02 1.871e+02 2.242e+02 4.934e+02, threshold=3.741e+02, percent-clipped=1.0 +2023-03-27 09:51:56,115 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156572.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:52:09,723 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.93 vs. limit=5.0 +2023-03-27 09:52:10,212 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9101, 1.8148, 1.7526, 1.8542, 1.6709, 4.4766, 1.7080, 2.1642], + device='cuda:1'), covar=tensor([0.3102, 0.2424, 0.2035, 0.2266, 0.1435, 0.0137, 0.2292, 0.1121], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0121, 0.0124, 0.0114, 0.0096, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 09:52:12,053 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1543, 1.9976, 1.7297, 1.9764, 1.8254, 1.8831, 1.9273, 2.6216], + device='cuda:1'), covar=tensor([0.3443, 0.3905, 0.2937, 0.3323, 0.3869, 0.2214, 0.3448, 0.1582], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0264, 0.0237, 0.0275, 0.0261, 0.0229, 0.0258, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:52:13,683 INFO [finetune.py:976] (1/7) Epoch 28, batch 1950, loss[loss=0.1585, simple_loss=0.2249, pruned_loss=0.04601, over 4754.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2443, pruned_loss=0.04962, over 953357.01 frames. ], batch size: 27, lr: 2.88e-03, grad_scale: 32.0 +2023-03-27 09:52:19,468 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4487, 1.3798, 1.5377, 0.8061, 1.5161, 1.5319, 1.4826, 1.3144], + device='cuda:1'), covar=tensor([0.0643, 0.0847, 0.0731, 0.0942, 0.0923, 0.0715, 0.0661, 0.1309], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0138, 0.0140, 0.0118, 0.0128, 0.0138, 0.0139, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:52:39,057 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6388, 1.5148, 1.3633, 1.7238, 1.6313, 1.6685, 1.0522, 1.4038], + device='cuda:1'), covar=tensor([0.2079, 0.1940, 0.1875, 0.1502, 0.1622, 0.1180, 0.2432, 0.1841], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0211, 0.0215, 0.0198, 0.0245, 0.0190, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:52:46,376 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156646.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:52:47,509 INFO [finetune.py:976] (1/7) Epoch 28, batch 2000, loss[loss=0.1778, simple_loss=0.2389, pruned_loss=0.05829, over 4902.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2428, pruned_loss=0.04969, over 955257.47 frames. ], batch size: 36, lr: 2.88e-03, grad_scale: 32.0 +2023-03-27 09:53:04,320 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.919e+01 1.529e+02 1.760e+02 2.169e+02 4.761e+02, threshold=3.520e+02, percent-clipped=1.0 +2023-03-27 09:53:23,850 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1563, 1.9418, 1.4852, 1.5075, 2.4330, 2.5355, 2.2171, 1.9865], + device='cuda:1'), covar=tensor([0.0355, 0.0471, 0.0920, 0.0481, 0.0301, 0.0622, 0.0316, 0.0501], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0111, 0.0102, 0.0116, 0.0104, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.8397e-05, 8.1109e-05, 1.1550e-04, 8.5079e-05, 7.8497e-05, 8.5353e-05, + 7.6857e-05, 8.6004e-05], device='cuda:1') +2023-03-27 09:53:29,982 INFO [finetune.py:976] (1/7) Epoch 28, batch 2050, loss[loss=0.1219, simple_loss=0.1963, pruned_loss=0.02373, over 4781.00 frames. ], tot_loss[loss=0.1666, simple_loss=0.2381, pruned_loss=0.04757, over 956126.71 frames. ], batch size: 28, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:53:44,496 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6001, 0.7743, 1.7004, 1.6376, 1.5094, 1.4394, 1.5607, 1.6576], + device='cuda:1'), covar=tensor([0.3103, 0.3336, 0.2638, 0.2939, 0.3940, 0.3178, 0.3373, 0.2359], + device='cuda:1'), in_proj_covar=tensor([0.0268, 0.0249, 0.0269, 0.0298, 0.0297, 0.0274, 0.0303, 0.0253], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:53:47,440 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156726.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 09:54:08,957 INFO [finetune.py:976] (1/7) Epoch 28, batch 2100, loss[loss=0.1271, simple_loss=0.1967, pruned_loss=0.02872, over 4761.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.2382, pruned_loss=0.04768, over 957255.79 frames. ], batch size: 27, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:54:09,684 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156749.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:54:20,564 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.56 vs. limit=5.0 +2023-03-27 09:54:37,294 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4095, 1.4860, 1.2267, 1.5122, 1.7145, 1.6669, 1.4834, 1.3031], + device='cuda:1'), covar=tensor([0.0407, 0.0373, 0.0660, 0.0307, 0.0249, 0.0484, 0.0384, 0.0480], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0112, 0.0101, 0.0116, 0.0104, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.8444e-05, 8.1122e-05, 1.1557e-04, 8.5166e-05, 7.8461e-05, 8.5241e-05, + 7.6821e-05, 8.5899e-05], device='cuda:1') +2023-03-27 09:54:37,771 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.070e+02 1.521e+02 1.844e+02 2.179e+02 3.224e+02, threshold=3.687e+02, percent-clipped=0.0 +2023-03-27 09:54:38,476 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=156774.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:54:54,962 INFO [finetune.py:976] (1/7) Epoch 28, batch 2150, loss[loss=0.1809, simple_loss=0.2505, pruned_loss=0.05572, over 4854.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2418, pruned_loss=0.04851, over 956512.77 frames. ], batch size: 31, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:55:03,496 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156810.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:55:16,189 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-27 09:55:17,861 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7979, 1.7051, 1.6401, 1.7496, 1.3490, 4.3319, 1.6588, 2.0060], + device='cuda:1'), covar=tensor([0.3148, 0.2495, 0.2081, 0.2254, 0.1675, 0.0180, 0.2441, 0.1190], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0124, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 09:55:27,789 INFO [finetune.py:976] (1/7) Epoch 28, batch 2200, loss[loss=0.1681, simple_loss=0.2428, pruned_loss=0.04666, over 4753.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2428, pruned_loss=0.04861, over 955416.93 frames. ], batch size: 28, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:55:43,603 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9429, 2.5828, 2.9122, 2.8814, 2.6441, 2.6572, 2.8271, 2.7187], + device='cuda:1'), covar=tensor([0.3025, 0.3406, 0.2700, 0.3049, 0.4207, 0.3186, 0.3909, 0.2508], + device='cuda:1'), in_proj_covar=tensor([0.0269, 0.0250, 0.0270, 0.0299, 0.0298, 0.0275, 0.0304, 0.0254], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:55:44,139 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156872.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:55:44,662 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.040e+02 1.543e+02 1.740e+02 2.127e+02 4.555e+02, threshold=3.480e+02, percent-clipped=1.0 +2023-03-27 09:55:52,561 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156885.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:55:57,154 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2974, 2.2886, 1.7049, 2.2907, 2.2680, 1.9849, 2.6016, 2.3270], + device='cuda:1'), covar=tensor([0.1240, 0.1849, 0.2924, 0.2308, 0.2365, 0.1560, 0.2513, 0.1493], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0190, 0.0237, 0.0255, 0.0251, 0.0208, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:56:01,289 INFO [finetune.py:976] (1/7) Epoch 28, batch 2250, loss[loss=0.213, simple_loss=0.2929, pruned_loss=0.06652, over 4842.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2437, pruned_loss=0.04889, over 955116.78 frames. ], batch size: 49, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:56:16,206 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=156920.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:56:31,635 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3620, 3.7561, 4.0029, 4.2018, 4.1309, 3.8474, 4.4482, 1.4520], + device='cuda:1'), covar=tensor([0.0765, 0.0930, 0.0864, 0.0912, 0.1132, 0.1672, 0.0599, 0.5790], + device='cuda:1'), in_proj_covar=tensor([0.0350, 0.0244, 0.0282, 0.0293, 0.0333, 0.0285, 0.0304, 0.0302], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:56:32,874 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156946.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:56:32,891 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156946.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:56:33,999 INFO [finetune.py:976] (1/7) Epoch 28, batch 2300, loss[loss=0.161, simple_loss=0.2344, pruned_loss=0.04381, over 4758.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2433, pruned_loss=0.04836, over 952995.52 frames. ], batch size: 28, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:56:35,395 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-27 09:57:00,119 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.437e+02 1.655e+02 2.039e+02 3.893e+02, threshold=3.311e+02, percent-clipped=1.0 +2023-03-27 09:57:10,748 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-03-27 09:57:17,059 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=156994.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:57:20,392 INFO [finetune.py:976] (1/7) Epoch 28, batch 2350, loss[loss=0.1478, simple_loss=0.2156, pruned_loss=0.03998, over 4034.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2413, pruned_loss=0.04749, over 954376.13 frames. ], batch size: 17, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:57:20,571 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.63 vs. limit=2.0 +2023-03-27 09:57:28,241 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157010.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:57:45,930 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.6283, 4.9187, 5.1950, 5.4570, 5.3156, 5.0271, 5.7198, 2.2567], + device='cuda:1'), covar=tensor([0.0633, 0.0765, 0.0745, 0.0664, 0.1165, 0.1549, 0.0478, 0.5059], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0245, 0.0284, 0.0294, 0.0335, 0.0287, 0.0305, 0.0304], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:57:52,968 INFO [finetune.py:976] (1/7) Epoch 28, batch 2400, loss[loss=0.2016, simple_loss=0.2777, pruned_loss=0.06272, over 4758.00 frames. ], tot_loss[loss=0.1672, simple_loss=0.2396, pruned_loss=0.04737, over 953038.07 frames. ], batch size: 26, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:57:56,633 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8610, 1.7771, 1.5828, 1.8948, 2.4452, 2.0240, 1.8267, 1.5906], + device='cuda:1'), covar=tensor([0.2203, 0.1989, 0.1939, 0.1639, 0.1574, 0.1186, 0.2170, 0.1744], + device='cuda:1'), in_proj_covar=tensor([0.0244, 0.0210, 0.0214, 0.0198, 0.0244, 0.0189, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:58:08,932 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157071.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:58:12,822 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.019e+02 1.490e+02 1.799e+02 2.218e+02 3.254e+02, threshold=3.597e+02, percent-clipped=0.0 +2023-03-27 09:58:16,664 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-27 09:58:28,589 INFO [finetune.py:976] (1/7) Epoch 28, batch 2450, loss[loss=0.1421, simple_loss=0.221, pruned_loss=0.03166, over 4786.00 frames. ], tot_loss[loss=0.1656, simple_loss=0.2372, pruned_loss=0.04695, over 951792.96 frames. ], batch size: 29, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:58:33,536 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157105.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:58:50,988 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.3144, 2.9485, 3.0577, 3.2544, 3.0627, 2.9038, 3.3499, 0.9896], + device='cuda:1'), covar=tensor([0.1236, 0.1118, 0.1208, 0.1222, 0.1829, 0.1947, 0.1153, 0.6262], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0247, 0.0285, 0.0295, 0.0337, 0.0288, 0.0307, 0.0305], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 09:58:55,869 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0783, 1.4032, 0.9070, 2.0176, 2.3114, 1.9373, 1.6607, 2.0367], + device='cuda:1'), covar=tensor([0.1275, 0.1836, 0.1851, 0.1015, 0.1837, 0.1727, 0.1245, 0.1622], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0093, 0.0109, 0.0092, 0.0120, 0.0092, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 09:58:57,708 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157142.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 09:59:01,794 INFO [finetune.py:976] (1/7) Epoch 28, batch 2500, loss[loss=0.1394, simple_loss=0.2136, pruned_loss=0.03256, over 4829.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2398, pruned_loss=0.04817, over 954525.40 frames. ], batch size: 25, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:59:09,563 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1526, 1.2513, 1.3501, 0.7174, 1.2894, 1.5993, 1.5848, 1.2971], + device='cuda:1'), covar=tensor([0.0989, 0.0853, 0.0565, 0.0565, 0.0680, 0.0645, 0.0410, 0.0754], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0130, 0.0123, 0.0132, 0.0130, 0.0143, 0.0151], + device='cuda:1'), out_proj_covar=tensor([8.8684e-05, 1.0615e-04, 9.1997e-05, 8.5980e-05, 9.2301e-05, 9.2087e-05, + 1.0123e-04, 1.0749e-04], device='cuda:1') +2023-03-27 09:59:23,963 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.120e+02 1.586e+02 1.871e+02 2.257e+02 5.817e+02, threshold=3.742e+02, percent-clipped=3.0 +2023-03-27 09:59:27,293 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-03-27 09:59:51,556 INFO [finetune.py:976] (1/7) Epoch 28, batch 2550, loss[loss=0.2008, simple_loss=0.2703, pruned_loss=0.06568, over 4929.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2424, pruned_loss=0.04883, over 953788.71 frames. ], batch size: 33, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 09:59:55,380 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157203.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:00:20,724 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157241.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:00:24,903 INFO [finetune.py:976] (1/7) Epoch 28, batch 2600, loss[loss=0.1787, simple_loss=0.2588, pruned_loss=0.0493, over 4808.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2436, pruned_loss=0.04901, over 952786.15 frames. ], batch size: 51, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:00:37,826 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9562, 3.7893, 3.7927, 2.0276, 4.0097, 3.1673, 1.0334, 2.8088], + device='cuda:1'), covar=tensor([0.2804, 0.1989, 0.1442, 0.2988, 0.0981, 0.0886, 0.3973, 0.1498], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0180, 0.0160, 0.0130, 0.0163, 0.0123, 0.0149, 0.0126], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 10:00:41,899 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.034e+02 1.561e+02 1.846e+02 2.212e+02 4.271e+02, threshold=3.692e+02, percent-clipped=1.0 +2023-03-27 10:00:57,757 INFO [finetune.py:976] (1/7) Epoch 28, batch 2650, loss[loss=0.1683, simple_loss=0.2381, pruned_loss=0.04921, over 4865.00 frames. ], tot_loss[loss=0.1719, simple_loss=0.2447, pruned_loss=0.0495, over 951914.96 frames. ], batch size: 34, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:01:30,662 INFO [finetune.py:976] (1/7) Epoch 28, batch 2700, loss[loss=0.1381, simple_loss=0.2094, pruned_loss=0.03337, over 4924.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2431, pruned_loss=0.04832, over 951832.99 frames. ], batch size: 33, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:01:35,645 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157356.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:01:42,597 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157366.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:01:47,169 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.080e+02 1.455e+02 1.749e+02 2.146e+02 4.370e+02, threshold=3.498e+02, percent-clipped=1.0 +2023-03-27 10:01:53,265 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157382.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:02:12,809 INFO [finetune.py:976] (1/7) Epoch 28, batch 2750, loss[loss=0.1574, simple_loss=0.2288, pruned_loss=0.04299, over 4863.00 frames. ], tot_loss[loss=0.1693, simple_loss=0.2416, pruned_loss=0.04845, over 952905.41 frames. ], batch size: 44, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:02:20,319 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157404.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 10:02:20,947 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157405.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:02:21,574 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8141, 1.7268, 1.5060, 1.9230, 2.2204, 1.9359, 1.4386, 1.5041], + device='cuda:1'), covar=tensor([0.1934, 0.1854, 0.1865, 0.1457, 0.1566, 0.1182, 0.2457, 0.1799], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0212, 0.0216, 0.0200, 0.0246, 0.0191, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:02:22,851 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=5.62 vs. limit=5.0 +2023-03-27 10:02:28,676 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157417.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:02:33,283 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3071, 2.1130, 1.7960, 2.1021, 2.2002, 1.9303, 2.4234, 2.2687], + device='cuda:1'), covar=tensor([0.1240, 0.2128, 0.2902, 0.2460, 0.2545, 0.1681, 0.3256, 0.1590], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0191, 0.0238, 0.0255, 0.0251, 0.0208, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:02:37,514 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0596, 4.7076, 4.4175, 2.4126, 4.7131, 3.6603, 0.9050, 3.3659], + device='cuda:1'), covar=tensor([0.2152, 0.1507, 0.1495, 0.3018, 0.0769, 0.0815, 0.4535, 0.1475], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0180, 0.0160, 0.0130, 0.0164, 0.0124, 0.0149, 0.0126], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 10:02:46,860 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157443.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:02:50,279 INFO [finetune.py:976] (1/7) Epoch 28, batch 2800, loss[loss=0.176, simple_loss=0.2475, pruned_loss=0.05225, over 4908.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2395, pruned_loss=0.04803, over 954123.20 frames. ], batch size: 36, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:02:53,300 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=157453.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:02:53,353 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8050, 1.7568, 1.6404, 1.8091, 1.5993, 4.5704, 1.7383, 2.2458], + device='cuda:1'), covar=tensor([0.3162, 0.2444, 0.2123, 0.2286, 0.1499, 0.0100, 0.2366, 0.1094], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0124, 0.0114, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 10:02:57,023 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0 +2023-03-27 10:03:01,071 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157465.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 10:03:06,319 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.288e+01 1.520e+02 1.688e+02 2.071e+02 7.416e+02, threshold=3.376e+02, percent-clipped=4.0 +2023-03-27 10:03:18,778 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4796, 1.5351, 1.2690, 1.5068, 1.8435, 1.7153, 1.5757, 1.3848], + device='cuda:1'), covar=tensor([0.0419, 0.0339, 0.0682, 0.0338, 0.0240, 0.0506, 0.0286, 0.0440], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0106, 0.0148, 0.0111, 0.0102, 0.0116, 0.0104, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.8687e-05, 8.0964e-05, 1.1521e-04, 8.5012e-05, 7.8758e-05, 8.5260e-05, + 7.7045e-05, 8.5938e-05], device='cuda:1') +2023-03-27 10:03:23,412 INFO [finetune.py:976] (1/7) Epoch 28, batch 2850, loss[loss=0.2038, simple_loss=0.2607, pruned_loss=0.07345, over 4913.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.2381, pruned_loss=0.04775, over 955188.00 frames. ], batch size: 36, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:03:23,481 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157498.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:03:28,862 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157506.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 10:03:52,498 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157541.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:03:57,078 INFO [finetune.py:976] (1/7) Epoch 28, batch 2900, loss[loss=0.1786, simple_loss=0.255, pruned_loss=0.05111, over 4219.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2405, pruned_loss=0.04804, over 955150.92 frames. ], batch size: 65, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:04:09,785 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157567.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 10:04:10,365 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157568.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:04:13,242 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.581e+01 1.484e+02 1.768e+02 2.098e+02 4.175e+02, threshold=3.535e+02, percent-clipped=1.0 +2023-03-27 10:04:18,533 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8715, 3.9333, 3.7052, 1.7886, 4.0690, 3.1277, 0.8029, 2.7880], + device='cuda:1'), covar=tensor([0.2305, 0.1943, 0.1477, 0.3244, 0.0900, 0.0865, 0.4293, 0.1378], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0182, 0.0161, 0.0131, 0.0165, 0.0125, 0.0150, 0.0127], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 10:04:24,444 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=157589.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:04:32,418 INFO [finetune.py:976] (1/7) Epoch 28, batch 2950, loss[loss=0.1551, simple_loss=0.2057, pruned_loss=0.05223, over 4022.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2441, pruned_loss=0.04911, over 955610.20 frames. ], batch size: 17, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:04:40,962 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.73 vs. limit=2.0 +2023-03-27 10:05:06,827 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157629.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:05:23,713 INFO [finetune.py:976] (1/7) Epoch 28, batch 3000, loss[loss=0.1599, simple_loss=0.2316, pruned_loss=0.04409, over 4825.00 frames. ], tot_loss[loss=0.173, simple_loss=0.2459, pruned_loss=0.04999, over 956625.13 frames. ], batch size: 30, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:05:23,714 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 10:05:26,830 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2547, 2.0823, 1.8175, 1.9681, 2.2046, 1.9503, 2.3114, 2.2436], + device='cuda:1'), covar=tensor([0.1308, 0.2021, 0.2915, 0.2344, 0.2499, 0.1678, 0.2723, 0.1725], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0190, 0.0237, 0.0254, 0.0250, 0.0207, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:05:29,613 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.1333, 1.4022, 1.4787, 0.6352, 1.4417, 1.6518, 1.7264, 1.4331], + device='cuda:1'), covar=tensor([0.1058, 0.0679, 0.0622, 0.0635, 0.0542, 0.0719, 0.0406, 0.0708], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0149, 0.0130, 0.0123, 0.0132, 0.0131, 0.0143, 0.0152], + device='cuda:1'), out_proj_covar=tensor([8.8922e-05, 1.0667e-04, 9.2651e-05, 8.6451e-05, 9.2698e-05, 9.2844e-05, + 1.0146e-04, 1.0827e-04], device='cuda:1') +2023-03-27 10:05:32,076 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8939, 3.4879, 3.5892, 3.7813, 3.6217, 3.4635, 3.9486, 1.2915], + device='cuda:1'), covar=tensor([0.0891, 0.0866, 0.0870, 0.0946, 0.1388, 0.1704, 0.0760, 0.5803], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0285, 0.0294, 0.0334, 0.0285, 0.0305, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:05:34,508 INFO [finetune.py:1010] (1/7) Epoch 28, validation: loss=0.1567, simple_loss=0.2243, pruned_loss=0.04455, over 2265189.00 frames. +2023-03-27 10:05:34,508 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 10:05:46,488 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157666.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:05:50,618 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.041e+02 1.524e+02 1.841e+02 2.248e+02 4.082e+02, threshold=3.682e+02, percent-clipped=3.0 +2023-03-27 10:06:07,193 INFO [finetune.py:976] (1/7) Epoch 28, batch 3050, loss[loss=0.1629, simple_loss=0.2308, pruned_loss=0.04749, over 4723.00 frames. ], tot_loss[loss=0.1726, simple_loss=0.2458, pruned_loss=0.04966, over 956400.23 frames. ], batch size: 59, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:06:16,638 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157712.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:06:17,820 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=157714.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:06:33,274 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157738.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:06:38,050 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0137, 1.7776, 2.1777, 1.5760, 2.1168, 2.2991, 1.5989, 2.3751], + device='cuda:1'), covar=tensor([0.1137, 0.1991, 0.1230, 0.1529, 0.0864, 0.1048, 0.2741, 0.0727], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0205, 0.0192, 0.0188, 0.0173, 0.0211, 0.0217, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:06:40,233 INFO [finetune.py:976] (1/7) Epoch 28, batch 3100, loss[loss=0.1468, simple_loss=0.2163, pruned_loss=0.03866, over 4846.00 frames. ], tot_loss[loss=0.171, simple_loss=0.2441, pruned_loss=0.04897, over 955799.21 frames. ], batch size: 49, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:06:48,809 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157760.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 10:06:57,050 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.055e+02 1.389e+02 1.744e+02 2.105e+02 3.209e+02, threshold=3.488e+02, percent-clipped=0.0 +2023-03-27 10:07:19,510 INFO [finetune.py:976] (1/7) Epoch 28, batch 3150, loss[loss=0.18, simple_loss=0.2532, pruned_loss=0.0534, over 4816.00 frames. ], tot_loss[loss=0.169, simple_loss=0.2413, pruned_loss=0.04833, over 955115.29 frames. ], batch size: 45, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:07:19,600 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157798.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:08:03,993 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=157846.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:08:05,667 INFO [finetune.py:976] (1/7) Epoch 28, batch 3200, loss[loss=0.1702, simple_loss=0.2221, pruned_loss=0.05919, over 4878.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.2389, pruned_loss=0.04738, over 956535.12 frames. ], batch size: 31, lr: 2.88e-03, grad_scale: 16.0 +2023-03-27 10:08:09,358 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157853.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:08:15,274 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157862.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 10:08:22,848 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.817e+01 1.555e+02 1.831e+02 2.254e+02 7.078e+02, threshold=3.662e+02, percent-clipped=7.0 +2023-03-27 10:08:38,478 INFO [finetune.py:976] (1/7) Epoch 28, batch 3250, loss[loss=0.1729, simple_loss=0.2434, pruned_loss=0.05118, over 4803.00 frames. ], tot_loss[loss=0.167, simple_loss=0.2392, pruned_loss=0.04744, over 953900.54 frames. ], batch size: 45, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:08:49,843 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157914.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:08:56,861 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157924.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:09:11,705 INFO [finetune.py:976] (1/7) Epoch 28, batch 3300, loss[loss=0.1837, simple_loss=0.2714, pruned_loss=0.04795, over 4853.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.244, pruned_loss=0.04928, over 953978.30 frames. ], batch size: 44, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:09:29,137 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.081e+02 1.521e+02 1.758e+02 2.140e+02 4.599e+02, threshold=3.515e+02, percent-clipped=1.0 +2023-03-27 10:09:41,792 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4293, 3.2565, 3.1664, 1.4143, 3.4635, 2.5655, 1.0144, 2.2959], + device='cuda:1'), covar=tensor([0.2620, 0.2275, 0.1501, 0.3142, 0.1129, 0.1000, 0.3637, 0.1460], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0181, 0.0161, 0.0131, 0.0165, 0.0125, 0.0151, 0.0127], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 10:09:44,716 INFO [finetune.py:976] (1/7) Epoch 28, batch 3350, loss[loss=0.1432, simple_loss=0.2203, pruned_loss=0.03306, over 4825.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.2455, pruned_loss=0.04972, over 955912.73 frames. ], batch size: 33, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:09:58,176 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158012.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:10:29,907 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158038.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:10:38,995 INFO [finetune.py:976] (1/7) Epoch 28, batch 3400, loss[loss=0.143, simple_loss=0.2178, pruned_loss=0.03406, over 4757.00 frames. ], tot_loss[loss=0.1721, simple_loss=0.2453, pruned_loss=0.04943, over 956316.58 frames. ], batch size: 27, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:10:46,886 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=158060.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:10:46,937 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158060.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 10:10:55,546 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.103e+02 1.550e+02 1.897e+02 2.350e+02 3.360e+02, threshold=3.793e+02, percent-clipped=0.0 +2023-03-27 10:11:04,972 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=158086.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:11:12,581 INFO [finetune.py:976] (1/7) Epoch 28, batch 3450, loss[loss=0.1425, simple_loss=0.2125, pruned_loss=0.03623, over 4710.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2453, pruned_loss=0.04914, over 957066.61 frames. ], batch size: 23, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:11:18,072 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3428, 1.4446, 1.2639, 1.4765, 1.6914, 1.6440, 1.4706, 1.2889], + device='cuda:1'), covar=tensor([0.0412, 0.0287, 0.0598, 0.0265, 0.0215, 0.0427, 0.0322, 0.0394], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0105, 0.0147, 0.0111, 0.0101, 0.0115, 0.0103, 0.0113], + device='cuda:1'), out_proj_covar=tensor([7.8117e-05, 8.0599e-05, 1.1453e-04, 8.4551e-05, 7.8315e-05, 8.4887e-05, + 7.6747e-05, 8.5679e-05], device='cuda:1') +2023-03-27 10:11:18,606 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=158108.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 10:11:28,458 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158122.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:11:43,772 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-03-27 10:11:44,265 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5705, 1.5422, 1.9296, 2.9934, 1.9750, 2.2896, 1.0203, 2.5344], + device='cuda:1'), covar=tensor([0.1757, 0.1274, 0.1215, 0.0549, 0.0815, 0.1184, 0.1798, 0.0519], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0133, 0.0164, 0.0100, 0.0135, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 10:11:45,461 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8374, 4.2954, 3.9335, 2.0378, 4.3627, 3.4418, 0.9680, 3.1032], + device='cuda:1'), covar=tensor([0.2403, 0.2350, 0.1677, 0.3547, 0.0853, 0.0857, 0.4789, 0.1562], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0181, 0.0161, 0.0131, 0.0164, 0.0124, 0.0150, 0.0127], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 10:11:46,008 INFO [finetune.py:976] (1/7) Epoch 28, batch 3500, loss[loss=0.1362, simple_loss=0.2076, pruned_loss=0.03238, over 4784.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2431, pruned_loss=0.04877, over 955408.24 frames. ], batch size: 23, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:11:52,155 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0204, 1.5281, 0.9344, 2.0181, 2.2831, 2.0677, 1.8297, 1.9800], + device='cuda:1'), covar=tensor([0.1213, 0.1749, 0.2023, 0.0951, 0.1787, 0.1646, 0.1159, 0.1632], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0110, 0.0093, 0.0120, 0.0092, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 10:11:54,573 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158162.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 10:12:02,509 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.003e+02 1.432e+02 1.800e+02 2.074e+02 3.411e+02, threshold=3.600e+02, percent-clipped=0.0 +2023-03-27 10:12:08,977 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4443, 2.3103, 2.4435, 1.6421, 2.3316, 2.4741, 2.5927, 2.0655], + device='cuda:1'), covar=tensor([0.0497, 0.0589, 0.0587, 0.0878, 0.0935, 0.0635, 0.0517, 0.0985], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0139, 0.0141, 0.0120, 0.0130, 0.0141, 0.0141, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:12:09,607 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158183.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:12:19,503 INFO [finetune.py:976] (1/7) Epoch 28, batch 3550, loss[loss=0.1462, simple_loss=0.2244, pruned_loss=0.03399, over 4912.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2412, pruned_loss=0.04855, over 955759.99 frames. ], batch size: 43, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:12:24,063 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.5633, 1.6151, 1.6006, 0.8416, 1.7674, 1.9515, 1.9183, 1.5022], + device='cuda:1'), covar=tensor([0.0947, 0.0601, 0.0550, 0.0627, 0.0433, 0.0577, 0.0303, 0.0670], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0131, 0.0123, 0.0133, 0.0131, 0.0143, 0.0151], + device='cuda:1'), out_proj_covar=tensor([8.9188e-05, 1.0632e-04, 9.2847e-05, 8.6320e-05, 9.3033e-05, 9.2798e-05, + 1.0162e-04, 1.0810e-04], device='cuda:1') +2023-03-27 10:12:28,596 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=158209.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:12:29,188 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=158210.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 10:12:37,004 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1950, 1.9702, 2.4588, 1.6671, 2.2109, 2.5243, 1.7797, 2.5016], + device='cuda:1'), covar=tensor([0.1254, 0.1867, 0.1382, 0.1775, 0.0916, 0.1138, 0.2704, 0.0850], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0194, 0.0189, 0.0174, 0.0213, 0.0219, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:12:38,690 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158224.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:12:41,523 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2001, 1.4411, 1.6899, 1.5523, 1.6706, 2.8746, 1.2757, 1.5768], + device='cuda:1'), covar=tensor([0.1091, 0.1981, 0.1074, 0.0883, 0.1467, 0.0372, 0.1726, 0.1913], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0091, 0.0080, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 10:13:02,523 INFO [finetune.py:976] (1/7) Epoch 28, batch 3600, loss[loss=0.1815, simple_loss=0.2535, pruned_loss=0.05471, over 4848.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2396, pruned_loss=0.04851, over 954193.06 frames. ], batch size: 49, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:13:18,162 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=158272.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:13:18,717 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.065e+02 1.425e+02 1.679e+02 2.016e+02 3.584e+02, threshold=3.358e+02, percent-clipped=0.0 +2023-03-27 10:13:36,304 INFO [finetune.py:976] (1/7) Epoch 28, batch 3650, loss[loss=0.1306, simple_loss=0.2058, pruned_loss=0.02767, over 4762.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.2409, pruned_loss=0.049, over 951114.58 frames. ], batch size: 26, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:14:00,839 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0969, 1.2105, 0.5050, 1.9481, 2.2820, 1.6214, 1.7333, 1.6925], + device='cuda:1'), covar=tensor([0.1506, 0.2318, 0.2224, 0.1231, 0.2020, 0.1924, 0.1436, 0.2087], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0095, 0.0110, 0.0093, 0.0121, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 10:14:10,065 INFO [finetune.py:976] (1/7) Epoch 28, batch 3700, loss[loss=0.1786, simple_loss=0.2626, pruned_loss=0.04728, over 4764.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2436, pruned_loss=0.04945, over 950899.77 frames. ], batch size: 28, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:14:26,136 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.573e+02 1.954e+02 2.338e+02 5.991e+02, threshold=3.909e+02, percent-clipped=5.0 +2023-03-27 10:14:33,449 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158384.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:14:39,373 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4796, 1.0878, 0.7498, 1.3820, 1.9265, 0.7183, 1.3756, 1.3375], + device='cuda:1'), covar=tensor([0.1542, 0.2063, 0.1694, 0.1210, 0.1945, 0.2031, 0.1394, 0.2004], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0095, 0.0110, 0.0093, 0.0121, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 10:14:43,253 INFO [finetune.py:976] (1/7) Epoch 28, batch 3750, loss[loss=0.1832, simple_loss=0.2645, pruned_loss=0.051, over 4898.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.2443, pruned_loss=0.04919, over 953815.62 frames. ], batch size: 36, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:15:19,912 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158445.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:15:22,040 INFO [finetune.py:976] (1/7) Epoch 28, batch 3800, loss[loss=0.1604, simple_loss=0.2367, pruned_loss=0.04199, over 4693.00 frames. ], tot_loss[loss=0.1718, simple_loss=0.2455, pruned_loss=0.04902, over 954808.22 frames. ], batch size: 59, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:15:51,767 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.090e+02 1.580e+02 1.909e+02 2.258e+02 3.504e+02, threshold=3.818e+02, percent-clipped=0.0 +2023-03-27 10:15:55,428 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=158478.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:16:08,887 INFO [finetune.py:976] (1/7) Epoch 28, batch 3850, loss[loss=0.1591, simple_loss=0.2429, pruned_loss=0.03763, over 4755.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2438, pruned_loss=0.04834, over 952472.83 frames. ], batch size: 27, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:16:15,407 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2312, 1.6361, 0.7772, 2.1001, 2.4839, 1.8967, 2.0757, 2.0126], + device='cuda:1'), covar=tensor([0.1238, 0.1683, 0.2020, 0.1018, 0.1754, 0.1795, 0.1102, 0.1794], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0109, 0.0093, 0.0120, 0.0093, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 10:16:15,507 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.07 vs. limit=5.0 +2023-03-27 10:16:16,593 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158509.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:16:42,035 INFO [finetune.py:976] (1/7) Epoch 28, batch 3900, loss[loss=0.1492, simple_loss=0.2159, pruned_loss=0.04128, over 4784.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2404, pruned_loss=0.04726, over 954457.43 frames. ], batch size: 29, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:16:48,969 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=158557.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:16:58,921 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.091e+02 1.423e+02 1.682e+02 2.011e+02 3.434e+02, threshold=3.365e+02, percent-clipped=0.0 +2023-03-27 10:17:15,470 INFO [finetune.py:976] (1/7) Epoch 28, batch 3950, loss[loss=0.1702, simple_loss=0.2285, pruned_loss=0.05594, over 4797.00 frames. ], tot_loss[loss=0.1653, simple_loss=0.2375, pruned_loss=0.04656, over 954878.44 frames. ], batch size: 29, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:17:22,872 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1941, 2.0008, 1.7859, 1.9478, 1.9207, 1.9488, 1.9494, 2.7097], + device='cuda:1'), covar=tensor([0.3478, 0.4587, 0.3095, 0.3578, 0.4113, 0.2260, 0.3627, 0.1539], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0265, 0.0238, 0.0276, 0.0261, 0.0231, 0.0260, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:17:26,908 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1422, 1.4110, 0.8444, 2.0779, 2.4987, 1.9331, 1.8338, 1.9699], + device='cuda:1'), covar=tensor([0.1318, 0.1958, 0.1937, 0.1092, 0.1759, 0.1707, 0.1238, 0.1862], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0093, 0.0109, 0.0092, 0.0120, 0.0092, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 10:17:31,075 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.9346, 4.3379, 4.4935, 4.7530, 4.7102, 4.4078, 5.0841, 1.4786], + device='cuda:1'), covar=tensor([0.0883, 0.0875, 0.0848, 0.0991, 0.1255, 0.1681, 0.0602, 0.5984], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0246, 0.0286, 0.0296, 0.0338, 0.0286, 0.0307, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:17:48,877 INFO [finetune.py:976] (1/7) Epoch 28, batch 4000, loss[loss=0.1566, simple_loss=0.2373, pruned_loss=0.03799, over 4754.00 frames. ], tot_loss[loss=0.1645, simple_loss=0.2362, pruned_loss=0.04641, over 951884.47 frames. ], batch size: 23, lr: 2.87e-03, grad_scale: 16.0 +2023-03-27 10:18:15,605 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.131e+01 1.444e+02 1.852e+02 2.094e+02 7.470e+02, threshold=3.703e+02, percent-clipped=2.0 +2023-03-27 10:18:27,619 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.18 vs. limit=5.0 +2023-03-27 10:18:32,166 INFO [finetune.py:976] (1/7) Epoch 28, batch 4050, loss[loss=0.22, simple_loss=0.2934, pruned_loss=0.07329, over 4935.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.2399, pruned_loss=0.04718, over 953257.17 frames. ], batch size: 33, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:18:59,957 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=158740.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:19:05,203 INFO [finetune.py:976] (1/7) Epoch 28, batch 4100, loss[loss=0.1625, simple_loss=0.2375, pruned_loss=0.0437, over 4050.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2415, pruned_loss=0.04719, over 952517.07 frames. ], batch size: 65, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:19:16,952 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6486, 3.6521, 3.4653, 1.7617, 3.7167, 2.8056, 1.0124, 2.5434], + device='cuda:1'), covar=tensor([0.2719, 0.1669, 0.1509, 0.3123, 0.1044, 0.1043, 0.3947, 0.1470], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0181, 0.0160, 0.0130, 0.0163, 0.0124, 0.0149, 0.0126], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 10:19:22,718 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.067e+02 1.534e+02 1.832e+02 2.269e+02 3.411e+02, threshold=3.665e+02, percent-clipped=0.0 +2023-03-27 10:19:25,300 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158777.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:19:25,445 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-27 10:19:25,893 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158778.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:19:29,387 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158783.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:19:31,240 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3442, 1.4087, 1.4275, 0.7774, 1.4799, 1.6360, 1.7018, 1.3350], + device='cuda:1'), covar=tensor([0.0863, 0.0530, 0.0456, 0.0462, 0.0426, 0.0609, 0.0251, 0.0631], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0130, 0.0122, 0.0131, 0.0129, 0.0142, 0.0151], + device='cuda:1'), out_proj_covar=tensor([8.8194e-05, 1.0519e-04, 9.2014e-05, 8.5494e-05, 9.1881e-05, 9.1588e-05, + 1.0105e-04, 1.0768e-04], device='cuda:1') +2023-03-27 10:19:38,839 INFO [finetune.py:976] (1/7) Epoch 28, batch 4150, loss[loss=0.2157, simple_loss=0.2861, pruned_loss=0.0726, over 4717.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2432, pruned_loss=0.04817, over 950073.48 frames. ], batch size: 59, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:19:58,146 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=158826.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:19:59,983 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.8463, 3.3611, 3.5432, 3.7145, 3.6198, 3.3699, 3.9086, 1.2030], + device='cuda:1'), covar=tensor([0.0977, 0.0909, 0.0964, 0.1141, 0.1333, 0.1668, 0.0902, 0.5896], + device='cuda:1'), in_proj_covar=tensor([0.0352, 0.0246, 0.0286, 0.0296, 0.0338, 0.0286, 0.0306, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:20:03,628 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158835.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 10:20:05,934 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158838.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:20:09,552 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158844.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:20:11,899 INFO [finetune.py:976] (1/7) Epoch 28, batch 4200, loss[loss=0.1592, simple_loss=0.2349, pruned_loss=0.04176, over 4809.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2425, pruned_loss=0.04731, over 951725.41 frames. ], batch size: 40, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:20:18,378 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7636, 3.6798, 3.5872, 1.7236, 3.8624, 2.8953, 0.8227, 2.6906], + device='cuda:1'), covar=tensor([0.2114, 0.2278, 0.1466, 0.3105, 0.1002, 0.1014, 0.4253, 0.1367], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0181, 0.0160, 0.0130, 0.0163, 0.0124, 0.0150, 0.0126], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 10:20:35,368 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.763e+01 1.457e+02 1.627e+02 2.050e+02 3.601e+02, threshold=3.253e+02, percent-clipped=0.0 +2023-03-27 10:20:53,831 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8684, 1.8389, 1.6981, 1.9435, 1.5918, 4.4104, 1.6724, 1.9904], + device='cuda:1'), covar=tensor([0.3041, 0.2380, 0.1997, 0.2306, 0.1516, 0.0153, 0.2321, 0.1197], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0124, 0.0114, 0.0095, 0.0093, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 10:20:55,698 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2317, 2.0984, 1.8477, 2.1212, 1.9884, 2.0066, 2.0337, 2.7809], + device='cuda:1'), covar=tensor([0.3628, 0.4343, 0.3319, 0.3883, 0.4450, 0.2414, 0.3819, 0.1651], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0237, 0.0274, 0.0260, 0.0230, 0.0258, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:20:58,734 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158896.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 10:21:04,570 INFO [finetune.py:976] (1/7) Epoch 28, batch 4250, loss[loss=0.2122, simple_loss=0.2701, pruned_loss=0.07718, over 4230.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.2402, pruned_loss=0.04698, over 952167.81 frames. ], batch size: 65, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:21:14,157 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2917, 1.9041, 2.2956, 2.2935, 2.0039, 2.0163, 2.2510, 2.1038], + device='cuda:1'), covar=tensor([0.3923, 0.3930, 0.3138, 0.3761, 0.5144, 0.4104, 0.4495, 0.3021], + device='cuda:1'), in_proj_covar=tensor([0.0269, 0.0249, 0.0269, 0.0298, 0.0297, 0.0274, 0.0304, 0.0253], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:21:38,818 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.69 vs. limit=5.0 +2023-03-27 10:21:46,471 INFO [finetune.py:976] (1/7) Epoch 28, batch 4300, loss[loss=0.1415, simple_loss=0.2066, pruned_loss=0.03815, over 4857.00 frames. ], tot_loss[loss=0.1657, simple_loss=0.2381, pruned_loss=0.04668, over 953819.34 frames. ], batch size: 47, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:22:03,616 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.096e+02 1.473e+02 1.789e+02 2.045e+02 3.501e+02, threshold=3.577e+02, percent-clipped=2.0 +2023-03-27 10:22:20,211 INFO [finetune.py:976] (1/7) Epoch 28, batch 4350, loss[loss=0.1609, simple_loss=0.2408, pruned_loss=0.04049, over 4897.00 frames. ], tot_loss[loss=0.1646, simple_loss=0.2365, pruned_loss=0.04634, over 953012.58 frames. ], batch size: 43, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:22:48,304 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159040.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:22:53,108 INFO [finetune.py:976] (1/7) Epoch 28, batch 4400, loss[loss=0.2092, simple_loss=0.2941, pruned_loss=0.06213, over 4807.00 frames. ], tot_loss[loss=0.1643, simple_loss=0.2365, pruned_loss=0.0461, over 952577.62 frames. ], batch size: 41, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:23:06,910 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159069.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:23:09,712 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.014e+02 1.529e+02 1.726e+02 2.218e+02 4.795e+02, threshold=3.452e+02, percent-clipped=2.0 +2023-03-27 10:23:24,707 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=159088.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:23:27,819 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9808, 1.3951, 1.8631, 1.9179, 1.7463, 1.7428, 1.8260, 1.8598], + device='cuda:1'), covar=tensor([0.4780, 0.4841, 0.4274, 0.4550, 0.6108, 0.5004, 0.5837, 0.4007], + device='cuda:1'), in_proj_covar=tensor([0.0268, 0.0248, 0.0268, 0.0297, 0.0296, 0.0273, 0.0302, 0.0253], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:23:35,365 INFO [finetune.py:976] (1/7) Epoch 28, batch 4450, loss[loss=0.2078, simple_loss=0.2713, pruned_loss=0.0721, over 4859.00 frames. ], tot_loss[loss=0.1669, simple_loss=0.2399, pruned_loss=0.047, over 952838.26 frames. ], batch size: 31, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:24:00,776 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159130.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:24:02,992 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159133.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:24:04,915 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-27 10:24:07,592 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159139.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:24:12,912 INFO [finetune.py:976] (1/7) Epoch 28, batch 4500, loss[loss=0.1607, simple_loss=0.2401, pruned_loss=0.04059, over 4923.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2405, pruned_loss=0.04719, over 950717.77 frames. ], batch size: 33, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:24:20,200 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159159.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:24:28,962 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.131e+02 1.553e+02 1.848e+02 2.152e+02 3.966e+02, threshold=3.696e+02, percent-clipped=2.0 +2023-03-27 10:24:41,949 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159191.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 10:24:44,833 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6420, 1.1143, 0.9194, 1.5487, 1.9999, 1.2050, 1.4741, 1.5821], + device='cuda:1'), covar=tensor([0.1349, 0.1947, 0.1613, 0.1134, 0.1916, 0.1861, 0.1278, 0.1659], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0109, 0.0093, 0.0121, 0.0093, 0.0099, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 10:24:46,582 INFO [finetune.py:976] (1/7) Epoch 28, batch 4550, loss[loss=0.1841, simple_loss=0.2466, pruned_loss=0.06081, over 4814.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2425, pruned_loss=0.04789, over 953173.24 frames. ], batch size: 38, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:25:00,548 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159220.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:25:10,424 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2453, 2.1684, 2.3811, 1.6884, 2.1876, 2.3299, 2.4064, 1.8618], + device='cuda:1'), covar=tensor([0.0569, 0.0623, 0.0617, 0.0835, 0.0682, 0.0679, 0.0584, 0.1086], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0139, 0.0141, 0.0120, 0.0129, 0.0140, 0.0141, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:25:20,091 INFO [finetune.py:976] (1/7) Epoch 28, batch 4600, loss[loss=0.1936, simple_loss=0.2763, pruned_loss=0.05548, over 4839.00 frames. ], tot_loss[loss=0.1692, simple_loss=0.2425, pruned_loss=0.04796, over 953072.35 frames. ], batch size: 49, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:25:35,683 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.005e+02 1.454e+02 1.668e+02 2.062e+02 3.965e+02, threshold=3.336e+02, percent-clipped=3.0 +2023-03-27 10:25:45,538 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9363, 1.8873, 2.4397, 2.1386, 2.0724, 3.7771, 1.7824, 2.1308], + device='cuda:1'), covar=tensor([0.0832, 0.1453, 0.0820, 0.0760, 0.1236, 0.0213, 0.1210, 0.1350], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0083, 0.0074, 0.0077, 0.0092, 0.0081, 0.0086, 0.0081], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 10:25:54,165 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159290.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:26:05,013 INFO [finetune.py:976] (1/7) Epoch 28, batch 4650, loss[loss=0.1907, simple_loss=0.261, pruned_loss=0.06017, over 4823.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2405, pruned_loss=0.0476, over 953806.56 frames. ], batch size: 40, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:26:24,438 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4439, 1.4795, 2.1186, 1.6388, 1.6320, 3.5522, 1.4646, 1.6781], + device='cuda:1'), covar=tensor([0.1007, 0.1871, 0.0967, 0.1008, 0.1633, 0.0257, 0.1462, 0.1717], + device='cuda:1'), in_proj_covar=tensor([0.0076, 0.0083, 0.0074, 0.0077, 0.0092, 0.0081, 0.0086, 0.0081], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 10:26:25,675 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3455, 1.3815, 1.5509, 1.0875, 1.3310, 1.4849, 1.3591, 1.6739], + device='cuda:1'), covar=tensor([0.1138, 0.2150, 0.1131, 0.1385, 0.0909, 0.1184, 0.2877, 0.0823], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0193, 0.0189, 0.0174, 0.0214, 0.0218, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:26:55,598 INFO [finetune.py:976] (1/7) Epoch 28, batch 4700, loss[loss=0.146, simple_loss=0.2246, pruned_loss=0.03374, over 4775.00 frames. ], tot_loss[loss=0.1663, simple_loss=0.2388, pruned_loss=0.04692, over 956261.45 frames. ], batch size: 28, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:26:58,020 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159351.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:27:11,660 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.356e+01 1.407e+02 1.592e+02 1.978e+02 3.098e+02, threshold=3.183e+02, percent-clipped=0.0 +2023-03-27 10:27:28,646 INFO [finetune.py:976] (1/7) Epoch 28, batch 4750, loss[loss=0.2137, simple_loss=0.2782, pruned_loss=0.07454, over 4919.00 frames. ], tot_loss[loss=0.1666, simple_loss=0.2383, pruned_loss=0.04742, over 955964.43 frames. ], batch size: 38, lr: 2.87e-03, grad_scale: 32.0 +2023-03-27 10:27:37,573 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9567, 1.4597, 2.0874, 2.0485, 1.8568, 1.8246, 1.9738, 1.9616], + device='cuda:1'), covar=tensor([0.3982, 0.4110, 0.3372, 0.3704, 0.4908, 0.3920, 0.4570, 0.3052], + device='cuda:1'), in_proj_covar=tensor([0.0269, 0.0248, 0.0270, 0.0298, 0.0298, 0.0275, 0.0303, 0.0253], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:27:41,276 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-27 10:27:46,456 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159425.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:27:48,336 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7035, 1.8104, 1.6164, 1.5486, 2.3141, 2.3070, 2.0349, 1.9226], + device='cuda:1'), covar=tensor([0.0477, 0.0392, 0.0610, 0.0379, 0.0255, 0.0561, 0.0358, 0.0411], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0107, 0.0149, 0.0112, 0.0102, 0.0117, 0.0105, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.9069e-05, 8.1488e-05, 1.1551e-04, 8.5300e-05, 7.8801e-05, 8.5898e-05, + 7.7663e-05, 8.6910e-05], device='cuda:1') +2023-03-27 10:27:51,860 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159433.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:27:55,927 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159439.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:28:02,248 INFO [finetune.py:976] (1/7) Epoch 28, batch 4800, loss[loss=0.1634, simple_loss=0.2342, pruned_loss=0.04634, over 4814.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.24, pruned_loss=0.04761, over 954159.73 frames. ], batch size: 25, lr: 2.86e-03, grad_scale: 32.0 +2023-03-27 10:28:11,745 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5787, 1.1930, 0.8594, 1.5181, 1.9912, 1.0810, 1.4167, 1.5932], + device='cuda:1'), covar=tensor([0.1503, 0.2087, 0.1789, 0.1182, 0.2042, 0.1991, 0.1369, 0.1792], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0108, 0.0093, 0.0120, 0.0092, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 10:28:18,791 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.120e+02 1.565e+02 1.779e+02 2.195e+02 3.956e+02, threshold=3.558e+02, percent-clipped=1.0 +2023-03-27 10:28:23,704 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=159481.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:28:27,810 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=159487.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:28:30,761 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159491.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 10:28:36,824 INFO [finetune.py:976] (1/7) Epoch 28, batch 4850, loss[loss=0.1854, simple_loss=0.2621, pruned_loss=0.05435, over 4804.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.2416, pruned_loss=0.04784, over 954295.94 frames. ], batch size: 41, lr: 2.86e-03, grad_scale: 32.0 +2023-03-27 10:28:39,355 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-27 10:28:57,883 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159515.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:29:09,687 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2646, 2.2609, 1.9690, 2.3763, 2.1746, 2.1422, 2.2249, 3.0067], + device='cuda:1'), covar=tensor([0.3698, 0.4825, 0.3228, 0.4021, 0.4615, 0.2380, 0.4483, 0.1607], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0264, 0.0237, 0.0274, 0.0261, 0.0231, 0.0259, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:29:17,410 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=159539.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 10:29:23,228 INFO [finetune.py:976] (1/7) Epoch 28, batch 4900, loss[loss=0.144, simple_loss=0.2278, pruned_loss=0.03007, over 4929.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2431, pruned_loss=0.0484, over 955109.65 frames. ], batch size: 38, lr: 2.86e-03, grad_scale: 32.0 +2023-03-27 10:29:36,215 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3454, 1.3414, 1.3732, 0.9526, 1.4234, 1.6130, 1.7189, 1.2473], + device='cuda:1'), covar=tensor([0.0973, 0.0751, 0.0630, 0.0503, 0.0521, 0.0667, 0.0335, 0.0841], + device='cuda:1'), in_proj_covar=tensor([0.0122, 0.0148, 0.0130, 0.0123, 0.0132, 0.0130, 0.0143, 0.0152], + device='cuda:1'), out_proj_covar=tensor([8.8820e-05, 1.0615e-04, 9.2589e-05, 8.5907e-05, 9.2606e-05, 9.2088e-05, + 1.0157e-04, 1.0855e-04], device='cuda:1') +2023-03-27 10:29:40,327 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.030e+02 1.553e+02 1.825e+02 2.271e+02 5.584e+02, threshold=3.651e+02, percent-clipped=3.0 +2023-03-27 10:29:56,963 INFO [finetune.py:976] (1/7) Epoch 28, batch 4950, loss[loss=0.1701, simple_loss=0.244, pruned_loss=0.04812, over 4723.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2453, pruned_loss=0.04908, over 954476.89 frames. ], batch size: 54, lr: 2.86e-03, grad_scale: 32.0 +2023-03-27 10:30:18,784 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159631.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:30:28,810 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159646.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:30:29,933 INFO [finetune.py:976] (1/7) Epoch 28, batch 5000, loss[loss=0.1782, simple_loss=0.2484, pruned_loss=0.05399, over 4801.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2441, pruned_loss=0.0487, over 953941.90 frames. ], batch size: 25, lr: 2.86e-03, grad_scale: 32.0 +2023-03-27 10:30:43,385 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2984, 2.3130, 2.0170, 2.3968, 2.2130, 2.1720, 2.2304, 3.0549], + device='cuda:1'), covar=tensor([0.3669, 0.4220, 0.3354, 0.3842, 0.4177, 0.2386, 0.3977, 0.1612], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0264, 0.0238, 0.0275, 0.0261, 0.0231, 0.0259, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:30:47,440 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.023e+02 1.490e+02 1.734e+02 1.957e+02 3.436e+02, threshold=3.469e+02, percent-clipped=0.0 +2023-03-27 10:30:59,471 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159692.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:31:05,617 INFO [finetune.py:976] (1/7) Epoch 28, batch 5050, loss[loss=0.1433, simple_loss=0.2086, pruned_loss=0.03896, over 4708.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2405, pruned_loss=0.04784, over 953620.87 frames. ], batch size: 23, lr: 2.86e-03, grad_scale: 32.0 +2023-03-27 10:31:24,319 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0303, 0.9984, 0.9511, 1.1410, 1.2202, 1.1544, 1.0227, 0.9428], + device='cuda:1'), covar=tensor([0.0447, 0.0332, 0.0716, 0.0327, 0.0317, 0.0514, 0.0447, 0.0491], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0111, 0.0101, 0.0116, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8613e-05, 8.0847e-05, 1.1486e-04, 8.4828e-05, 7.8273e-05, 8.5298e-05, + 7.7071e-05, 8.6439e-05], device='cuda:1') +2023-03-27 10:31:28,015 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-27 10:31:32,253 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159725.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:31:42,972 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-27 10:31:55,160 INFO [finetune.py:976] (1/7) Epoch 28, batch 5100, loss[loss=0.183, simple_loss=0.2525, pruned_loss=0.05672, over 4927.00 frames. ], tot_loss[loss=0.1647, simple_loss=0.2368, pruned_loss=0.04627, over 956103.97 frames. ], batch size: 33, lr: 2.86e-03, grad_scale: 32.0 +2023-03-27 10:32:21,761 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.008e+02 1.553e+02 1.866e+02 2.180e+02 3.771e+02, threshold=3.731e+02, percent-clipped=1.0 +2023-03-27 10:32:21,834 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=159773.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:32:37,623 INFO [finetune.py:976] (1/7) Epoch 28, batch 5150, loss[loss=0.161, simple_loss=0.2376, pruned_loss=0.04223, over 4766.00 frames. ], tot_loss[loss=0.1653, simple_loss=0.2373, pruned_loss=0.04669, over 956603.51 frames. ], batch size: 26, lr: 2.86e-03, grad_scale: 32.0 +2023-03-27 10:32:49,247 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159815.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:32:54,384 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6534, 1.5300, 2.0895, 3.3399, 2.2076, 2.2659, 0.8957, 2.8247], + device='cuda:1'), covar=tensor([0.1619, 0.1321, 0.1277, 0.0572, 0.0760, 0.1827, 0.1831, 0.0435], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0165, 0.0100, 0.0136, 0.0125, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 10:33:11,650 INFO [finetune.py:976] (1/7) Epoch 28, batch 5200, loss[loss=0.183, simple_loss=0.2515, pruned_loss=0.05725, over 4864.00 frames. ], tot_loss[loss=0.1678, simple_loss=0.2405, pruned_loss=0.04756, over 957598.60 frames. ], batch size: 31, lr: 2.86e-03, grad_scale: 32.0 +2023-03-27 10:33:16,448 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159855.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:33:21,708 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=159863.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:33:28,751 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.072e+02 1.499e+02 1.871e+02 2.174e+02 4.313e+02, threshold=3.743e+02, percent-clipped=1.0 +2023-03-27 10:33:44,908 INFO [finetune.py:976] (1/7) Epoch 28, batch 5250, loss[loss=0.1621, simple_loss=0.2414, pruned_loss=0.04138, over 4816.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2425, pruned_loss=0.0483, over 953445.42 frames. ], batch size: 39, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:33:53,298 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159910.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:33:59,356 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159916.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:34:14,944 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4372, 1.4500, 0.8259, 2.2011, 2.7030, 1.7993, 1.9466, 2.1069], + device='cuda:1'), covar=tensor([0.1371, 0.2105, 0.1884, 0.1151, 0.1597, 0.1817, 0.1338, 0.1964], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0093, 0.0108, 0.0093, 0.0120, 0.0092, 0.0098, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 10:34:26,204 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159946.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:34:27,320 INFO [finetune.py:976] (1/7) Epoch 28, batch 5300, loss[loss=0.2081, simple_loss=0.2687, pruned_loss=0.07372, over 4888.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2453, pruned_loss=0.04965, over 952547.31 frames. ], batch size: 32, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:34:50,614 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159971.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:34:52,303 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.066e+02 1.498e+02 1.765e+02 2.107e+02 3.764e+02, threshold=3.530e+02, percent-clipped=1.0 +2023-03-27 10:34:55,873 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4708, 2.6398, 2.5632, 1.9963, 2.5823, 2.8099, 2.9862, 2.3413], + device='cuda:1'), covar=tensor([0.0688, 0.0651, 0.0770, 0.0783, 0.0651, 0.0801, 0.0573, 0.1079], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0139, 0.0141, 0.0120, 0.0130, 0.0140, 0.0141, 0.0165], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:35:01,757 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159987.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:35:06,011 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=159994.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:35:08,412 INFO [finetune.py:976] (1/7) Epoch 28, batch 5350, loss[loss=0.1567, simple_loss=0.2307, pruned_loss=0.04132, over 4815.00 frames. ], tot_loss[loss=0.1725, simple_loss=0.246, pruned_loss=0.04953, over 955048.12 frames. ], batch size: 33, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:35:43,140 INFO [finetune.py:976] (1/7) Epoch 28, batch 5400, loss[loss=0.1298, simple_loss=0.2019, pruned_loss=0.02888, over 4803.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.2437, pruned_loss=0.04945, over 956386.14 frames. ], batch size: 29, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:36:00,195 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.280e+01 1.494e+02 1.877e+02 2.215e+02 3.734e+02, threshold=3.755e+02, percent-clipped=1.0 +2023-03-27 10:36:16,670 INFO [finetune.py:976] (1/7) Epoch 28, batch 5450, loss[loss=0.1786, simple_loss=0.2404, pruned_loss=0.05835, over 4770.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2408, pruned_loss=0.0487, over 954941.53 frames. ], batch size: 27, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:37:05,405 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.2314, 5.0742, 4.7460, 2.5723, 5.1953, 3.8746, 1.0302, 3.5872], + device='cuda:1'), covar=tensor([0.1871, 0.1601, 0.1268, 0.3136, 0.0616, 0.0777, 0.4416, 0.1360], + device='cuda:1'), in_proj_covar=tensor([0.0151, 0.0179, 0.0159, 0.0129, 0.0163, 0.0124, 0.0148, 0.0126], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 10:37:07,792 INFO [finetune.py:976] (1/7) Epoch 28, batch 5500, loss[loss=0.1392, simple_loss=0.2149, pruned_loss=0.03177, over 4830.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2388, pruned_loss=0.04845, over 954746.65 frames. ], batch size: 39, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:37:38,238 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.035e+02 1.497e+02 1.808e+02 2.088e+02 3.346e+02, threshold=3.616e+02, percent-clipped=0.0 +2023-03-27 10:37:55,667 INFO [finetune.py:976] (1/7) Epoch 28, batch 5550, loss[loss=0.1863, simple_loss=0.2619, pruned_loss=0.05541, over 4811.00 frames. ], tot_loss[loss=0.1672, simple_loss=0.2385, pruned_loss=0.04794, over 953120.10 frames. ], batch size: 39, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:38:03,851 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=160211.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:38:13,471 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3939, 1.8868, 2.7821, 1.6573, 2.3068, 2.6394, 1.7148, 2.6376], + device='cuda:1'), covar=tensor([0.1349, 0.2284, 0.1394, 0.2101, 0.0950, 0.1306, 0.2798, 0.0921], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0203, 0.0191, 0.0187, 0.0172, 0.0211, 0.0215, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:38:19,314 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2842, 1.4618, 1.2438, 1.4820, 1.6101, 1.6207, 1.4957, 1.4187], + device='cuda:1'), covar=tensor([0.0450, 0.0271, 0.0577, 0.0277, 0.0242, 0.0415, 0.0353, 0.0342], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0111, 0.0101, 0.0116, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8397e-05, 8.0873e-05, 1.1478e-04, 8.4523e-05, 7.8005e-05, 8.5452e-05, + 7.6918e-05, 8.6377e-05], device='cuda:1') +2023-03-27 10:38:27,206 INFO [finetune.py:976] (1/7) Epoch 28, batch 5600, loss[loss=0.1922, simple_loss=0.2689, pruned_loss=0.05777, over 4833.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2421, pruned_loss=0.04917, over 952064.29 frames. ], batch size: 33, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:38:37,609 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=160266.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:38:42,223 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.097e+02 1.448e+02 1.891e+02 2.366e+02 4.690e+02, threshold=3.782e+02, percent-clipped=2.0 +2023-03-27 10:38:49,853 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=160287.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:38:52,189 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6176, 1.5020, 1.4879, 1.5081, 1.1407, 3.5815, 1.3279, 1.8238], + device='cuda:1'), covar=tensor([0.3499, 0.2700, 0.2244, 0.2653, 0.1866, 0.0186, 0.2946, 0.1337], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0117, 0.0121, 0.0124, 0.0114, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 10:38:56,601 INFO [finetune.py:976] (1/7) Epoch 28, batch 5650, loss[loss=0.2137, simple_loss=0.2786, pruned_loss=0.07446, over 4908.00 frames. ], tot_loss[loss=0.1713, simple_loss=0.2443, pruned_loss=0.04917, over 952851.23 frames. ], batch size: 37, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:39:01,985 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=160307.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 10:39:04,333 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.3338, 1.1637, 1.1941, 0.6670, 1.1570, 1.4295, 1.4197, 1.1933], + device='cuda:1'), covar=tensor([0.0915, 0.0706, 0.0574, 0.0501, 0.0645, 0.0532, 0.0377, 0.0647], + device='cuda:1'), in_proj_covar=tensor([0.0120, 0.0147, 0.0130, 0.0122, 0.0130, 0.0129, 0.0142, 0.0150], + device='cuda:1'), out_proj_covar=tensor([8.7481e-05, 1.0549e-04, 9.2170e-05, 8.5213e-05, 9.1351e-05, 9.1622e-05, + 1.0084e-04, 1.0734e-04], device='cuda:1') +2023-03-27 10:39:24,239 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-03-27 10:39:24,562 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=160335.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:39:32,241 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6690, 2.5743, 2.5618, 1.9391, 2.4785, 2.7673, 2.8789, 2.2789], + device='cuda:1'), covar=tensor([0.0472, 0.0560, 0.0625, 0.0774, 0.0826, 0.0566, 0.0460, 0.0941], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0139, 0.0141, 0.0120, 0.0129, 0.0141, 0.0141, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:39:35,787 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2504, 2.2524, 2.0550, 2.4163, 2.6819, 2.4363, 2.1415, 1.9611], + device='cuda:1'), covar=tensor([0.1924, 0.1642, 0.1640, 0.1450, 0.1448, 0.0966, 0.1866, 0.1707], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0210, 0.0214, 0.0198, 0.0244, 0.0190, 0.0215, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:39:36,255 INFO [finetune.py:976] (1/7) Epoch 28, batch 5700, loss[loss=0.1419, simple_loss=0.2029, pruned_loss=0.04046, over 4177.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.241, pruned_loss=0.04853, over 934273.22 frames. ], batch size: 17, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:39:36,942 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0490, 2.7991, 2.5661, 3.3587, 2.9697, 2.7600, 3.4064, 3.0313], + device='cuda:1'), covar=tensor([0.1243, 0.1959, 0.2794, 0.1902, 0.2377, 0.1450, 0.2197, 0.1722], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0189, 0.0235, 0.0252, 0.0249, 0.0207, 0.0215, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:39:48,025 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=160368.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 10:39:53,235 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.932e+01 1.353e+02 1.695e+02 2.046e+02 5.031e+02, threshold=3.390e+02, percent-clipped=3.0 +2023-03-27 10:40:10,866 INFO [finetune.py:976] (1/7) Epoch 29, batch 0, loss[loss=0.161, simple_loss=0.2427, pruned_loss=0.03964, over 4783.00 frames. ], tot_loss[loss=0.161, simple_loss=0.2427, pruned_loss=0.03964, over 4783.00 frames. ], batch size: 29, lr: 2.86e-03, grad_scale: 16.0 +2023-03-27 10:40:10,866 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 10:40:21,879 INFO [finetune.py:1010] (1/7) Epoch 29, validation: loss=0.1588, simple_loss=0.2262, pruned_loss=0.04569, over 2265189.00 frames. +2023-03-27 10:40:21,879 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 10:40:31,722 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-03-27 10:40:57,964 INFO [finetune.py:976] (1/7) Epoch 29, batch 50, loss[loss=0.1921, simple_loss=0.2666, pruned_loss=0.0588, over 4817.00 frames. ], tot_loss[loss=0.1745, simple_loss=0.2492, pruned_loss=0.04993, over 218756.77 frames. ], batch size: 38, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:41:38,860 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.219e+02 1.505e+02 1.852e+02 2.145e+02 8.823e+02, threshold=3.704e+02, percent-clipped=1.0 +2023-03-27 10:41:39,934 INFO [finetune.py:976] (1/7) Epoch 29, batch 100, loss[loss=0.1472, simple_loss=0.2298, pruned_loss=0.03225, over 4769.00 frames. ], tot_loss[loss=0.1658, simple_loss=0.2386, pruned_loss=0.04647, over 383185.07 frames. ], batch size: 26, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:42:14,753 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=160511.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:42:15,883 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6349, 2.5872, 1.9808, 2.7748, 2.5943, 2.2496, 3.1006, 2.6749], + device='cuda:1'), covar=tensor([0.1186, 0.2180, 0.2927, 0.2311, 0.2372, 0.1594, 0.2758, 0.1579], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0191, 0.0237, 0.0253, 0.0251, 0.0208, 0.0216, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:42:34,271 INFO [finetune.py:976] (1/7) Epoch 29, batch 150, loss[loss=0.1683, simple_loss=0.237, pruned_loss=0.04983, over 4936.00 frames. ], tot_loss[loss=0.1623, simple_loss=0.2339, pruned_loss=0.0454, over 511603.85 frames. ], batch size: 33, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:42:55,943 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=160559.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:42:56,609 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0872, 1.7951, 2.2964, 1.5106, 2.0439, 2.4252, 1.6565, 2.4201], + device='cuda:1'), covar=tensor([0.1332, 0.2043, 0.1456, 0.2007, 0.1107, 0.1315, 0.2924, 0.0882], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0205, 0.0193, 0.0189, 0.0174, 0.0213, 0.0218, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:43:00,754 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=160566.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:43:06,534 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.027e+02 1.446e+02 1.774e+02 2.135e+02 3.412e+02, threshold=3.547e+02, percent-clipped=0.0 +2023-03-27 10:43:07,129 INFO [finetune.py:976] (1/7) Epoch 29, batch 200, loss[loss=0.1771, simple_loss=0.2531, pruned_loss=0.05053, over 4811.00 frames. ], tot_loss[loss=0.1631, simple_loss=0.2335, pruned_loss=0.04635, over 608620.21 frames. ], batch size: 41, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:43:30,739 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4920, 1.5121, 1.3547, 1.5182, 1.8725, 1.7708, 1.6101, 1.3957], + device='cuda:1'), covar=tensor([0.0363, 0.0292, 0.0622, 0.0329, 0.0201, 0.0489, 0.0275, 0.0415], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0111, 0.0101, 0.0116, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8404e-05, 8.0991e-05, 1.1512e-04, 8.4740e-05, 7.8258e-05, 8.5371e-05, + 7.7249e-05, 8.6834e-05], device='cuda:1') +2023-03-27 10:43:32,501 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=160614.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:43:39,908 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9374, 1.4082, 2.0248, 2.0189, 1.8161, 1.8015, 1.9017, 1.9675], + device='cuda:1'), covar=tensor([0.4012, 0.3916, 0.3298, 0.3468, 0.5080, 0.3767, 0.4452, 0.3037], + device='cuda:1'), in_proj_covar=tensor([0.0268, 0.0248, 0.0269, 0.0298, 0.0298, 0.0274, 0.0303, 0.0253], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:43:40,981 INFO [finetune.py:976] (1/7) Epoch 29, batch 250, loss[loss=0.2305, simple_loss=0.2915, pruned_loss=0.08476, over 4893.00 frames. ], tot_loss[loss=0.1656, simple_loss=0.2369, pruned_loss=0.04709, over 687445.01 frames. ], batch size: 35, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:44:05,536 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=160663.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 10:44:12,915 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.082e+02 1.455e+02 1.868e+02 2.134e+02 3.558e+02, threshold=3.736e+02, percent-clipped=1.0 +2023-03-27 10:44:13,967 INFO [finetune.py:976] (1/7) Epoch 29, batch 300, loss[loss=0.1612, simple_loss=0.2258, pruned_loss=0.04831, over 4931.00 frames. ], tot_loss[loss=0.169, simple_loss=0.242, pruned_loss=0.04803, over 749352.66 frames. ], batch size: 38, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:44:16,380 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0258, 2.0205, 2.1516, 1.5740, 2.0434, 2.2185, 2.2368, 1.6746], + device='cuda:1'), covar=tensor([0.0554, 0.0629, 0.0613, 0.0756, 0.0719, 0.0621, 0.0541, 0.1136], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0138, 0.0140, 0.0119, 0.0128, 0.0140, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:44:42,899 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3446, 1.2247, 1.0607, 1.1627, 1.5959, 1.5417, 1.3663, 1.1771], + device='cuda:1'), covar=tensor([0.0381, 0.0351, 0.0865, 0.0418, 0.0273, 0.0523, 0.0352, 0.0459], + device='cuda:1'), in_proj_covar=tensor([0.0101, 0.0106, 0.0148, 0.0111, 0.0101, 0.0116, 0.0104, 0.0114], + device='cuda:1'), out_proj_covar=tensor([7.8271e-05, 8.0882e-05, 1.1512e-04, 8.4626e-05, 7.8226e-05, 8.5415e-05, + 7.7299e-05, 8.6607e-05], device='cuda:1') +2023-03-27 10:44:57,148 INFO [finetune.py:976] (1/7) Epoch 29, batch 350, loss[loss=0.1837, simple_loss=0.2511, pruned_loss=0.05812, over 4933.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2437, pruned_loss=0.04886, over 794724.31 frames. ], batch size: 29, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:45:37,469 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.928e+01 1.581e+02 1.864e+02 2.210e+02 3.251e+02, threshold=3.728e+02, percent-clipped=0.0 +2023-03-27 10:45:38,107 INFO [finetune.py:976] (1/7) Epoch 29, batch 400, loss[loss=0.183, simple_loss=0.2708, pruned_loss=0.04765, over 4814.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.244, pruned_loss=0.0486, over 827317.23 frames. ], batch size: 33, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:46:04,906 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0562, 0.9142, 0.9458, 0.4563, 0.9660, 1.1353, 1.1445, 0.8766], + device='cuda:1'), covar=tensor([0.0853, 0.0637, 0.0677, 0.0540, 0.0668, 0.0632, 0.0515, 0.0778], + device='cuda:1'), in_proj_covar=tensor([0.0120, 0.0148, 0.0130, 0.0122, 0.0131, 0.0130, 0.0142, 0.0151], + device='cuda:1'), out_proj_covar=tensor([8.7865e-05, 1.0578e-04, 9.2259e-05, 8.5376e-05, 9.1669e-05, 9.1751e-05, + 1.0126e-04, 1.0756e-04], device='cuda:1') +2023-03-27 10:46:11,857 INFO [finetune.py:976] (1/7) Epoch 29, batch 450, loss[loss=0.1394, simple_loss=0.2138, pruned_loss=0.03249, over 4815.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2431, pruned_loss=0.04807, over 857254.02 frames. ], batch size: 39, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:46:55,070 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.046e+02 1.473e+02 1.674e+02 2.082e+02 4.783e+02, threshold=3.348e+02, percent-clipped=2.0 +2023-03-27 10:46:55,687 INFO [finetune.py:976] (1/7) Epoch 29, batch 500, loss[loss=0.1812, simple_loss=0.259, pruned_loss=0.05172, over 4818.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2411, pruned_loss=0.04761, over 880446.55 frames. ], batch size: 39, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:47:15,113 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=160901.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:47:36,924 INFO [finetune.py:976] (1/7) Epoch 29, batch 550, loss[loss=0.1416, simple_loss=0.2163, pruned_loss=0.03342, over 4901.00 frames. ], tot_loss[loss=0.1648, simple_loss=0.2373, pruned_loss=0.0462, over 897227.54 frames. ], batch size: 35, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:47:59,839 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7045, 1.5036, 1.9345, 3.0322, 2.1457, 2.3238, 0.9653, 2.5988], + device='cuda:1'), covar=tensor([0.1592, 0.1370, 0.1238, 0.0614, 0.0753, 0.1311, 0.1752, 0.0467], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0115, 0.0132, 0.0163, 0.0100, 0.0134, 0.0124, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 10:48:08,211 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=160962.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:48:08,796 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=160963.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 10:48:10,035 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=160965.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:48:15,382 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.000e+01 1.387e+02 1.769e+02 2.082e+02 3.377e+02, threshold=3.538e+02, percent-clipped=1.0 +2023-03-27 10:48:16,011 INFO [finetune.py:976] (1/7) Epoch 29, batch 600, loss[loss=0.1664, simple_loss=0.2389, pruned_loss=0.04695, over 4934.00 frames. ], tot_loss[loss=0.1646, simple_loss=0.2367, pruned_loss=0.0462, over 910885.30 frames. ], batch size: 38, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:48:18,027 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.65 vs. limit=5.0 +2023-03-27 10:48:20,850 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3151, 2.1431, 1.8503, 0.8485, 1.9661, 1.7968, 1.6862, 2.0141], + device='cuda:1'), covar=tensor([0.0917, 0.0766, 0.1474, 0.1897, 0.1329, 0.2395, 0.2272, 0.0880], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0189, 0.0201, 0.0180, 0.0208, 0.0208, 0.0222, 0.0195], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:48:41,019 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=161011.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 10:48:49,430 INFO [finetune.py:976] (1/7) Epoch 29, batch 650, loss[loss=0.183, simple_loss=0.2549, pruned_loss=0.05549, over 4752.00 frames. ], tot_loss[loss=0.1658, simple_loss=0.2384, pruned_loss=0.04665, over 922067.31 frames. ], batch size: 27, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:48:50,190 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161026.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:49:22,503 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.055e+02 1.587e+02 1.904e+02 2.267e+02 4.706e+02, threshold=3.807e+02, percent-clipped=2.0 +2023-03-27 10:49:23,105 INFO [finetune.py:976] (1/7) Epoch 29, batch 700, loss[loss=0.2116, simple_loss=0.2844, pruned_loss=0.06943, over 4811.00 frames. ], tot_loss[loss=0.1673, simple_loss=0.2404, pruned_loss=0.04708, over 928164.50 frames. ], batch size: 33, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:50:03,474 INFO [finetune.py:976] (1/7) Epoch 29, batch 750, loss[loss=0.1798, simple_loss=0.2569, pruned_loss=0.05135, over 4856.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2426, pruned_loss=0.04784, over 934178.05 frames. ], batch size: 31, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:50:46,352 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.613e+01 1.645e+02 1.965e+02 2.346e+02 5.118e+02, threshold=3.931e+02, percent-clipped=1.0 +2023-03-27 10:50:46,988 INFO [finetune.py:976] (1/7) Epoch 29, batch 800, loss[loss=0.1522, simple_loss=0.2264, pruned_loss=0.03894, over 4762.00 frames. ], tot_loss[loss=0.1685, simple_loss=0.2422, pruned_loss=0.04736, over 940477.12 frames. ], batch size: 26, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:51:06,563 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4905, 1.3406, 1.3740, 1.4042, 0.9285, 2.1993, 0.7209, 1.1734], + device='cuda:1'), covar=tensor([0.3001, 0.2464, 0.2075, 0.2374, 0.1774, 0.0420, 0.2798, 0.1313], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0117, 0.0121, 0.0124, 0.0114, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 10:51:15,283 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161216.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:51:16,483 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0817, 2.0477, 2.0951, 1.6597, 2.0394, 2.3149, 2.3097, 1.6977], + device='cuda:1'), covar=tensor([0.0643, 0.0696, 0.0794, 0.0889, 0.0826, 0.0691, 0.0570, 0.1292], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0139, 0.0141, 0.0119, 0.0129, 0.0141, 0.0141, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:51:20,578 INFO [finetune.py:976] (1/7) Epoch 29, batch 850, loss[loss=0.1577, simple_loss=0.225, pruned_loss=0.04525, over 4820.00 frames. ], tot_loss[loss=0.1673, simple_loss=0.2405, pruned_loss=0.04703, over 943659.14 frames. ], batch size: 39, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:51:24,327 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161231.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:51:27,343 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.4435, 3.8127, 4.0555, 4.2476, 4.1951, 3.9540, 4.5013, 1.4599], + device='cuda:1'), covar=tensor([0.0722, 0.0982, 0.0976, 0.0994, 0.1235, 0.1779, 0.0709, 0.5851], + device='cuda:1'), in_proj_covar=tensor([0.0356, 0.0250, 0.0289, 0.0301, 0.0343, 0.0290, 0.0309, 0.0308], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:51:48,478 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161257.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:52:02,089 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-03-27 10:52:04,267 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.030e+02 1.396e+02 1.676e+02 2.040e+02 3.102e+02, threshold=3.353e+02, percent-clipped=0.0 +2023-03-27 10:52:04,925 INFO [finetune.py:976] (1/7) Epoch 29, batch 900, loss[loss=0.148, simple_loss=0.2159, pruned_loss=0.04002, over 4897.00 frames. ], tot_loss[loss=0.1649, simple_loss=0.2377, pruned_loss=0.0461, over 946031.53 frames. ], batch size: 35, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:52:06,253 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161277.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:52:15,354 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161292.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:52:35,802 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161321.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:52:38,179 INFO [finetune.py:976] (1/7) Epoch 29, batch 950, loss[loss=0.182, simple_loss=0.2604, pruned_loss=0.05186, over 4811.00 frames. ], tot_loss[loss=0.1646, simple_loss=0.2367, pruned_loss=0.04627, over 948840.19 frames. ], batch size: 45, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:52:56,960 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1075, 1.5867, 0.9497, 2.0903, 2.2856, 1.9956, 1.6874, 2.0732], + device='cuda:1'), covar=tensor([0.1160, 0.1626, 0.1866, 0.0889, 0.1724, 0.1731, 0.1175, 0.1497], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0109, 0.0093, 0.0120, 0.0093, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 10:53:00,091 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-27 10:53:28,289 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.815e+01 1.511e+02 1.765e+02 2.170e+02 3.612e+02, threshold=3.529e+02, percent-clipped=1.0 +2023-03-27 10:53:28,920 INFO [finetune.py:976] (1/7) Epoch 29, batch 1000, loss[loss=0.1532, simple_loss=0.2394, pruned_loss=0.03355, over 4942.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.239, pruned_loss=0.04729, over 948966.78 frames. ], batch size: 33, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:53:48,057 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1687, 1.7540, 1.8186, 0.8615, 2.0818, 2.1846, 1.9872, 1.7006], + device='cuda:1'), covar=tensor([0.0854, 0.0711, 0.0524, 0.0691, 0.0492, 0.0728, 0.0493, 0.0808], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0148, 0.0131, 0.0123, 0.0132, 0.0130, 0.0143, 0.0152], + device='cuda:1'), out_proj_covar=tensor([8.8578e-05, 1.0634e-04, 9.3134e-05, 8.6019e-05, 9.2506e-05, 9.2271e-05, + 1.0194e-04, 1.0874e-04], device='cuda:1') +2023-03-27 10:53:49,430 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161401.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:54:06,939 INFO [finetune.py:976] (1/7) Epoch 29, batch 1050, loss[loss=0.1433, simple_loss=0.2265, pruned_loss=0.02999, over 4928.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2412, pruned_loss=0.04747, over 950602.00 frames. ], batch size: 42, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:54:13,083 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4695, 1.2430, 1.2756, 1.2295, 1.6272, 1.5567, 1.4061, 1.2280], + device='cuda:1'), covar=tensor([0.0330, 0.0326, 0.0659, 0.0330, 0.0233, 0.0473, 0.0329, 0.0415], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0107, 0.0149, 0.0113, 0.0102, 0.0118, 0.0105, 0.0115], + device='cuda:1'), out_proj_covar=tensor([7.9509e-05, 8.1921e-05, 1.1587e-04, 8.5688e-05, 7.9166e-05, 8.6512e-05, + 7.8162e-05, 8.7466e-05], device='cuda:1') +2023-03-27 10:54:28,543 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161460.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:54:30,825 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161462.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:54:39,307 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.696e+01 1.443e+02 1.825e+02 2.108e+02 3.139e+02, threshold=3.650e+02, percent-clipped=0.0 +2023-03-27 10:54:39,917 INFO [finetune.py:976] (1/7) Epoch 29, batch 1100, loss[loss=0.1786, simple_loss=0.2349, pruned_loss=0.06118, over 4426.00 frames. ], tot_loss[loss=0.17, simple_loss=0.243, pruned_loss=0.04852, over 951477.43 frames. ], batch size: 19, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:55:11,598 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161521.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:55:14,357 INFO [finetune.py:976] (1/7) Epoch 29, batch 1150, loss[loss=0.1796, simple_loss=0.2547, pruned_loss=0.05227, over 4778.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2442, pruned_loss=0.04899, over 953716.59 frames. ], batch size: 51, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:55:15,168 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.60 vs. limit=2.0 +2023-03-27 10:55:33,350 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-27 10:55:39,774 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161557.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:55:50,718 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161572.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:55:51,853 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.769e+01 1.420e+02 1.779e+02 2.194e+02 3.095e+02, threshold=3.558e+02, percent-clipped=0.0 +2023-03-27 10:55:52,952 INFO [finetune.py:976] (1/7) Epoch 29, batch 1200, loss[loss=0.1544, simple_loss=0.2127, pruned_loss=0.048, over 4917.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2435, pruned_loss=0.04897, over 954815.27 frames. ], batch size: 46, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:56:03,033 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161587.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:56:21,393 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-03-27 10:56:21,955 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-03-27 10:56:22,440 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=161605.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:56:33,593 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161621.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:56:36,406 INFO [finetune.py:976] (1/7) Epoch 29, batch 1250, loss[loss=0.1381, simple_loss=0.2105, pruned_loss=0.03282, over 4890.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2406, pruned_loss=0.04807, over 953299.33 frames. ], batch size: 32, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:56:40,127 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-03-27 10:56:51,958 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5956, 0.7463, 1.7529, 1.5901, 1.4969, 1.3982, 1.5579, 1.6976], + device='cuda:1'), covar=tensor([0.2907, 0.3256, 0.2616, 0.2850, 0.3866, 0.3080, 0.3314, 0.2356], + device='cuda:1'), in_proj_covar=tensor([0.0271, 0.0251, 0.0271, 0.0300, 0.0300, 0.0277, 0.0307, 0.0255], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:57:07,717 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=161669.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:57:07,782 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0752, 2.7736, 2.5988, 1.3958, 2.8095, 2.2138, 2.0851, 2.6021], + device='cuda:1'), covar=tensor([0.1444, 0.0708, 0.1729, 0.2000, 0.1537, 0.2117, 0.2059, 0.1082], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0188, 0.0200, 0.0179, 0.0208, 0.0209, 0.0222, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:57:07,793 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1946, 1.9755, 2.0612, 0.9545, 2.3146, 2.4945, 2.1671, 1.8611], + device='cuda:1'), covar=tensor([0.1022, 0.0783, 0.0455, 0.0697, 0.0615, 0.0667, 0.0694, 0.0874], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0148, 0.0131, 0.0122, 0.0132, 0.0130, 0.0143, 0.0152], + device='cuda:1'), out_proj_covar=tensor([8.8020e-05, 1.0589e-04, 9.3013e-05, 8.5697e-05, 9.2468e-05, 9.2013e-05, + 1.0156e-04, 1.0853e-04], device='cuda:1') +2023-03-27 10:57:10,129 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6123, 1.5897, 1.4462, 1.5198, 1.9634, 1.9397, 1.6839, 1.4588], + device='cuda:1'), covar=tensor([0.0376, 0.0339, 0.0661, 0.0349, 0.0237, 0.0480, 0.0326, 0.0465], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0107, 0.0149, 0.0112, 0.0102, 0.0118, 0.0105, 0.0115], + device='cuda:1'), out_proj_covar=tensor([7.9511e-05, 8.1784e-05, 1.1567e-04, 8.5501e-05, 7.9208e-05, 8.6742e-05, + 7.8247e-05, 8.7668e-05], device='cuda:1') +2023-03-27 10:57:11,726 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.453e+02 1.676e+02 2.058e+02 3.499e+02, threshold=3.351e+02, percent-clipped=0.0 +2023-03-27 10:57:12,869 INFO [finetune.py:976] (1/7) Epoch 29, batch 1300, loss[loss=0.1982, simple_loss=0.2663, pruned_loss=0.06502, over 4931.00 frames. ], tot_loss[loss=0.1657, simple_loss=0.2374, pruned_loss=0.04697, over 953652.37 frames. ], batch size: 38, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:57:49,625 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 10:57:50,741 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.81 vs. limit=5.0 +2023-03-27 10:57:54,737 INFO [finetune.py:976] (1/7) Epoch 29, batch 1350, loss[loss=0.1765, simple_loss=0.2609, pruned_loss=0.04608, over 4860.00 frames. ], tot_loss[loss=0.1664, simple_loss=0.2381, pruned_loss=0.04737, over 952272.61 frames. ], batch size: 44, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:58:24,061 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.54 vs. limit=2.0 +2023-03-27 10:58:29,755 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161757.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:58:49,497 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.001e+02 1.561e+02 1.829e+02 2.338e+02 6.236e+02, threshold=3.658e+02, percent-clipped=4.0 +2023-03-27 10:58:50,570 INFO [finetune.py:976] (1/7) Epoch 29, batch 1400, loss[loss=0.2016, simple_loss=0.2545, pruned_loss=0.07436, over 4766.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2407, pruned_loss=0.04804, over 952792.41 frames. ], batch size: 27, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:58:54,683 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9903, 1.8680, 2.1607, 1.4692, 1.9124, 2.1687, 2.0273, 1.5906], + device='cuda:1'), covar=tensor([0.0604, 0.0680, 0.0611, 0.0841, 0.0732, 0.0660, 0.0656, 0.1225], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0138, 0.0141, 0.0119, 0.0129, 0.0140, 0.0141, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:59:02,216 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5727, 1.5235, 1.4572, 1.5913, 1.3046, 3.6414, 1.4206, 1.7931], + device='cuda:1'), covar=tensor([0.3322, 0.2530, 0.2202, 0.2403, 0.1630, 0.0208, 0.2598, 0.1302], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0116, 0.0121, 0.0124, 0.0113, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 10:59:03,462 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3294, 2.2224, 1.7804, 2.3626, 2.1465, 1.8880, 2.5307, 2.3011], + device='cuda:1'), covar=tensor([0.1338, 0.1861, 0.2864, 0.2292, 0.2647, 0.1739, 0.2673, 0.1641], + device='cuda:1'), in_proj_covar=tensor([0.0188, 0.0189, 0.0235, 0.0251, 0.0248, 0.0207, 0.0214, 0.0202], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:59:17,540 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161815.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:59:18,106 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161816.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:59:19,383 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1637, 2.0454, 1.7741, 2.2431, 2.6992, 2.2830, 2.0699, 1.6136], + device='cuda:1'), covar=tensor([0.2128, 0.1779, 0.1773, 0.1545, 0.1602, 0.1074, 0.1979, 0.1877], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0212, 0.0216, 0.0199, 0.0246, 0.0191, 0.0217, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:59:23,992 INFO [finetune.py:976] (1/7) Epoch 29, batch 1450, loss[loss=0.2147, simple_loss=0.2851, pruned_loss=0.07216, over 4830.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2411, pruned_loss=0.04776, over 951564.93 frames. ], batch size: 49, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:59:41,425 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8937, 1.5571, 2.1216, 1.3892, 1.8902, 2.0347, 1.5209, 2.2195], + device='cuda:1'), covar=tensor([0.1270, 0.2298, 0.1336, 0.1905, 0.0955, 0.1465, 0.2927, 0.0893], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0206, 0.0194, 0.0191, 0.0175, 0.0214, 0.0219, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 10:59:55,388 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161872.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 10:59:56,503 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.032e+02 1.509e+02 1.810e+02 2.171e+02 4.074e+02, threshold=3.620e+02, percent-clipped=1.0 +2023-03-27 10:59:57,118 INFO [finetune.py:976] (1/7) Epoch 29, batch 1500, loss[loss=0.2019, simple_loss=0.2681, pruned_loss=0.06786, over 4801.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.242, pruned_loss=0.04788, over 953223.73 frames. ], batch size: 25, lr: 2.85e-03, grad_scale: 16.0 +2023-03-27 10:59:58,308 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161876.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:00:05,892 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161887.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:00:16,519 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-27 11:00:27,810 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=161920.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:00:32,887 INFO [finetune.py:976] (1/7) Epoch 29, batch 1550, loss[loss=0.1873, simple_loss=0.2584, pruned_loss=0.05811, over 4890.00 frames. ], tot_loss[loss=0.1682, simple_loss=0.242, pruned_loss=0.04718, over 953795.16 frames. ], batch size: 35, lr: 2.85e-03, grad_scale: 32.0 +2023-03-27 11:00:44,437 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=161935.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:01:09,394 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1263, 1.8858, 2.4915, 1.6391, 2.0912, 2.4250, 1.7710, 2.5189], + device='cuda:1'), covar=tensor([0.1320, 0.2001, 0.1518, 0.2206, 0.1045, 0.1493, 0.2662, 0.0899], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0206, 0.0194, 0.0191, 0.0175, 0.0214, 0.0219, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:01:16,694 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.413e+02 1.667e+02 1.984e+02 3.261e+02, threshold=3.334e+02, percent-clipped=0.0 +2023-03-27 11:01:17,325 INFO [finetune.py:976] (1/7) Epoch 29, batch 1600, loss[loss=0.1956, simple_loss=0.2556, pruned_loss=0.06781, over 4868.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2402, pruned_loss=0.04733, over 953762.35 frames. ], batch size: 34, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:01:59,455 INFO [finetune.py:976] (1/7) Epoch 29, batch 1650, loss[loss=0.1671, simple_loss=0.2281, pruned_loss=0.05311, over 4930.00 frames. ], tot_loss[loss=0.1654, simple_loss=0.2378, pruned_loss=0.04656, over 953876.51 frames. ], batch size: 33, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:02:22,371 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=162057.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:02:34,949 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.499e+01 1.468e+02 1.758e+02 2.171e+02 4.899e+02, threshold=3.516e+02, percent-clipped=3.0 +2023-03-27 11:02:35,583 INFO [finetune.py:976] (1/7) Epoch 29, batch 1700, loss[loss=0.1811, simple_loss=0.2448, pruned_loss=0.0587, over 4921.00 frames. ], tot_loss[loss=0.1626, simple_loss=0.2348, pruned_loss=0.04519, over 956330.53 frames. ], batch size: 37, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:02:36,902 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=162077.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:03:01,317 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=162105.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:03:08,006 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=162116.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:03:12,362 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1555, 2.0538, 2.2465, 1.6187, 2.0490, 2.2947, 2.2963, 1.7161], + device='cuda:1'), covar=tensor([0.0677, 0.0731, 0.0744, 0.0883, 0.0818, 0.0788, 0.0628, 0.1236], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0137, 0.0140, 0.0118, 0.0128, 0.0140, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:03:16,125 INFO [finetune.py:976] (1/7) Epoch 29, batch 1750, loss[loss=0.1727, simple_loss=0.2404, pruned_loss=0.05247, over 4829.00 frames. ], tot_loss[loss=0.1661, simple_loss=0.2386, pruned_loss=0.04681, over 956938.64 frames. ], batch size: 25, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:03:24,609 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=162138.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:03:59,952 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=162164.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:04:04,278 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=162171.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:04:10,047 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.681e+01 1.626e+02 1.856e+02 2.284e+02 4.131e+02, threshold=3.712e+02, percent-clipped=2.0 +2023-03-27 11:04:10,647 INFO [finetune.py:976] (1/7) Epoch 29, batch 1800, loss[loss=0.2011, simple_loss=0.264, pruned_loss=0.06911, over 4935.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2409, pruned_loss=0.04758, over 957720.77 frames. ], batch size: 33, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:04:44,452 INFO [finetune.py:976] (1/7) Epoch 29, batch 1850, loss[loss=0.1398, simple_loss=0.2083, pruned_loss=0.0356, over 4725.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2424, pruned_loss=0.04863, over 954159.70 frames. ], batch size: 23, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:04:48,176 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=162231.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:05:17,384 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.145e+02 1.604e+02 1.802e+02 2.267e+02 3.875e+02, threshold=3.605e+02, percent-clipped=1.0 +2023-03-27 11:05:17,994 INFO [finetune.py:976] (1/7) Epoch 29, batch 1900, loss[loss=0.1784, simple_loss=0.2506, pruned_loss=0.05315, over 4866.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2432, pruned_loss=0.04878, over 953322.97 frames. ], batch size: 34, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:05:20,544 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7972, 1.9308, 1.6988, 1.7145, 2.3496, 2.4267, 2.0862, 1.9628], + device='cuda:1'), covar=tensor([0.0490, 0.0389, 0.0594, 0.0347, 0.0311, 0.0515, 0.0337, 0.0434], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0108, 0.0150, 0.0113, 0.0103, 0.0119, 0.0106, 0.0117], + device='cuda:1'), out_proj_covar=tensor([8.0084e-05, 8.2346e-05, 1.1699e-04, 8.6112e-05, 7.9963e-05, 8.7505e-05, + 7.8961e-05, 8.8445e-05], device='cuda:1') +2023-03-27 11:05:28,433 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=162292.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:05:51,672 INFO [finetune.py:976] (1/7) Epoch 29, batch 1950, loss[loss=0.2009, simple_loss=0.2697, pruned_loss=0.06604, over 4804.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2424, pruned_loss=0.04837, over 955219.78 frames. ], batch size: 45, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:06:36,624 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.724e+01 1.430e+02 1.855e+02 2.130e+02 3.474e+02, threshold=3.709e+02, percent-clipped=0.0 +2023-03-27 11:06:37,251 INFO [finetune.py:976] (1/7) Epoch 29, batch 2000, loss[loss=0.1453, simple_loss=0.2191, pruned_loss=0.03579, over 4789.00 frames. ], tot_loss[loss=0.167, simple_loss=0.2395, pruned_loss=0.04723, over 953675.63 frames. ], batch size: 25, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:06:38,693 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.04 vs. limit=2.0 +2023-03-27 11:07:14,705 INFO [finetune.py:976] (1/7) Epoch 29, batch 2050, loss[loss=0.1218, simple_loss=0.1994, pruned_loss=0.0221, over 4713.00 frames. ], tot_loss[loss=0.165, simple_loss=0.237, pruned_loss=0.04653, over 954112.41 frames. ], batch size: 23, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:07:19,654 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=162433.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:07:30,191 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0701, 2.0414, 1.6118, 1.8393, 2.0340, 1.7652, 2.2974, 2.0438], + device='cuda:1'), covar=tensor([0.1284, 0.1809, 0.2756, 0.2409, 0.2402, 0.1604, 0.2503, 0.1636], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0191, 0.0237, 0.0254, 0.0251, 0.0209, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:07:45,860 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=162471.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:07:47,599 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.116e+01 1.601e+02 1.964e+02 2.330e+02 4.874e+02, threshold=3.928e+02, percent-clipped=3.0 +2023-03-27 11:07:48,226 INFO [finetune.py:976] (1/7) Epoch 29, batch 2100, loss[loss=0.1525, simple_loss=0.2253, pruned_loss=0.03983, over 4902.00 frames. ], tot_loss[loss=0.1659, simple_loss=0.2378, pruned_loss=0.047, over 953615.12 frames. ], batch size: 32, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:08:22,585 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=162511.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:08:28,422 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=162519.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:08:34,946 INFO [finetune.py:976] (1/7) Epoch 29, batch 2150, loss[loss=0.1429, simple_loss=0.212, pruned_loss=0.03691, over 4821.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2405, pruned_loss=0.04783, over 954383.83 frames. ], batch size: 25, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:08:55,400 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9719, 1.2954, 1.9489, 1.9235, 1.7592, 1.7413, 1.8853, 1.8445], + device='cuda:1'), covar=tensor([0.3889, 0.4100, 0.3370, 0.3764, 0.4993, 0.4042, 0.4359, 0.3051], + device='cuda:1'), in_proj_covar=tensor([0.0269, 0.0250, 0.0270, 0.0299, 0.0297, 0.0276, 0.0303, 0.0253], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:08:59,799 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.69 vs. limit=5.0 +2023-03-27 11:09:14,983 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=162572.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 11:09:18,163 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.053e+02 1.475e+02 1.779e+02 2.116e+02 3.329e+02, threshold=3.558e+02, percent-clipped=0.0 +2023-03-27 11:09:18,789 INFO [finetune.py:976] (1/7) Epoch 29, batch 2200, loss[loss=0.1935, simple_loss=0.2677, pruned_loss=0.05969, over 4812.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.242, pruned_loss=0.04758, over 956482.12 frames. ], batch size: 51, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:09:22,365 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7355, 2.8296, 2.6417, 1.8360, 2.7022, 2.9985, 3.0310, 2.4084], + device='cuda:1'), covar=tensor([0.0588, 0.0551, 0.0715, 0.0896, 0.0575, 0.0674, 0.0559, 0.1048], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0137, 0.0139, 0.0118, 0.0128, 0.0139, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:09:27,171 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=162587.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:10:00,795 INFO [finetune.py:976] (1/7) Epoch 29, batch 2250, loss[loss=0.1448, simple_loss=0.2425, pruned_loss=0.02356, over 4753.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2441, pruned_loss=0.04821, over 957098.45 frames. ], batch size: 28, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:10:06,545 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7319, 3.9306, 3.7254, 2.1163, 4.1199, 3.0706, 0.8608, 2.7884], + device='cuda:1'), covar=tensor([0.2243, 0.1791, 0.1495, 0.2821, 0.1004, 0.0849, 0.4418, 0.1382], + device='cuda:1'), in_proj_covar=tensor([0.0150, 0.0180, 0.0159, 0.0128, 0.0162, 0.0123, 0.0148, 0.0125], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 11:10:19,417 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1483, 1.6932, 2.4512, 1.5070, 2.0224, 2.3614, 1.6731, 2.3979], + device='cuda:1'), covar=tensor([0.1310, 0.2181, 0.1412, 0.2211, 0.1008, 0.1354, 0.2819, 0.0912], + device='cuda:1'), in_proj_covar=tensor([0.0192, 0.0206, 0.0193, 0.0190, 0.0175, 0.0213, 0.0219, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:10:33,516 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.222e+01 1.558e+02 1.861e+02 2.054e+02 5.864e+02, threshold=3.723e+02, percent-clipped=1.0 +2023-03-27 11:10:34,114 INFO [finetune.py:976] (1/7) Epoch 29, batch 2300, loss[loss=0.175, simple_loss=0.2517, pruned_loss=0.04914, over 4817.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2457, pruned_loss=0.04858, over 957126.05 frames. ], batch size: 38, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:10:46,490 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1772, 2.1559, 1.7120, 1.9954, 2.0668, 2.0101, 2.0121, 2.7395], + device='cuda:1'), covar=tensor([0.3757, 0.3809, 0.3274, 0.3723, 0.4003, 0.2417, 0.3475, 0.1771], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0264, 0.0239, 0.0275, 0.0261, 0.0232, 0.0260, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:11:09,593 INFO [finetune.py:976] (1/7) Epoch 29, batch 2350, loss[loss=0.1783, simple_loss=0.2464, pruned_loss=0.05507, over 4926.00 frames. ], tot_loss[loss=0.1699, simple_loss=0.2434, pruned_loss=0.04817, over 958156.12 frames. ], batch size: 33, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:11:20,050 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=162733.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:11:29,724 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.69 vs. limit=2.0 +2023-03-27 11:11:43,571 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8391, 1.6600, 1.9838, 1.2074, 1.7239, 1.9831, 1.6157, 2.1191], + device='cuda:1'), covar=tensor([0.1209, 0.1992, 0.1394, 0.1927, 0.0937, 0.1309, 0.2626, 0.0850], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0206, 0.0193, 0.0189, 0.0174, 0.0212, 0.0218, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:11:50,496 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.467e+01 1.478e+02 1.866e+02 2.325e+02 4.091e+02, threshold=3.732e+02, percent-clipped=2.0 +2023-03-27 11:11:51,107 INFO [finetune.py:976] (1/7) Epoch 29, batch 2400, loss[loss=0.1434, simple_loss=0.2086, pruned_loss=0.0391, over 4761.00 frames. ], tot_loss[loss=0.1675, simple_loss=0.2403, pruned_loss=0.04736, over 958568.08 frames. ], batch size: 28, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:11:56,644 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=162781.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:12:16,314 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.61 vs. limit=5.0 +2023-03-27 11:12:33,621 INFO [finetune.py:976] (1/7) Epoch 29, batch 2450, loss[loss=0.153, simple_loss=0.2261, pruned_loss=0.03995, over 4837.00 frames. ], tot_loss[loss=0.1657, simple_loss=0.2377, pruned_loss=0.04687, over 955542.32 frames. ], batch size: 30, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:12:35,949 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6668, 1.4942, 2.1793, 3.1144, 2.0542, 2.3211, 0.9294, 2.6333], + device='cuda:1'), covar=tensor([0.1600, 0.1344, 0.1138, 0.0613, 0.0831, 0.1590, 0.1779, 0.0492], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0114, 0.0132, 0.0163, 0.0100, 0.0135, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 11:12:35,973 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=162827.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:13:02,268 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=162867.0, num_to_drop=1, layers_to_drop={3} +2023-03-27 11:13:09,318 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.738e+01 1.481e+02 1.739e+02 2.016e+02 4.521e+02, threshold=3.479e+02, percent-clipped=1.0 +2023-03-27 11:13:09,948 INFO [finetune.py:976] (1/7) Epoch 29, batch 2500, loss[loss=0.1804, simple_loss=0.2537, pruned_loss=0.05356, over 4858.00 frames. ], tot_loss[loss=0.1658, simple_loss=0.2381, pruned_loss=0.04676, over 954807.32 frames. ], batch size: 31, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:13:23,119 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=162887.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:13:28,126 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=162888.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:13:34,641 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.17 vs. limit=5.0 +2023-03-27 11:13:43,018 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-03-27 11:13:52,415 INFO [finetune.py:976] (1/7) Epoch 29, batch 2550, loss[loss=0.1949, simple_loss=0.2675, pruned_loss=0.06115, over 4897.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2415, pruned_loss=0.04766, over 955134.78 frames. ], batch size: 35, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:14:01,024 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5632, 1.4225, 1.3572, 1.4217, 1.7649, 1.7715, 1.5164, 1.3347], + device='cuda:1'), covar=tensor([0.0330, 0.0372, 0.0614, 0.0384, 0.0221, 0.0460, 0.0400, 0.0467], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0107, 0.0149, 0.0113, 0.0103, 0.0118, 0.0105, 0.0116], + device='cuda:1'), out_proj_covar=tensor([7.9631e-05, 8.1608e-05, 1.1610e-04, 8.5718e-05, 7.9392e-05, 8.6870e-05, + 7.8210e-05, 8.7732e-05], device='cuda:1') +2023-03-27 11:14:01,594 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=162935.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:14:01,913 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-27 11:14:09,141 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8027, 2.5442, 2.1486, 1.0807, 2.3479, 2.1831, 1.9150, 2.4180], + device='cuda:1'), covar=tensor([0.0919, 0.0754, 0.1704, 0.2161, 0.1372, 0.2110, 0.2360, 0.1000], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0187, 0.0201, 0.0180, 0.0208, 0.0209, 0.0222, 0.0194], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:14:36,960 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.986e+01 1.516e+02 1.791e+02 2.248e+02 3.208e+02, threshold=3.582e+02, percent-clipped=0.0 +2023-03-27 11:14:37,596 INFO [finetune.py:976] (1/7) Epoch 29, batch 2600, loss[loss=0.1885, simple_loss=0.2586, pruned_loss=0.05921, over 4808.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2441, pruned_loss=0.04872, over 954620.66 frames. ], batch size: 40, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:15:09,250 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4491, 1.3216, 1.3348, 1.3069, 0.7901, 2.2798, 0.7563, 1.1733], + device='cuda:1'), covar=tensor([0.3388, 0.2575, 0.2275, 0.2529, 0.1955, 0.0351, 0.2762, 0.1399], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0123, 0.0112, 0.0095, 0.0093, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 11:15:17,741 INFO [finetune.py:976] (1/7) Epoch 29, batch 2650, loss[loss=0.1578, simple_loss=0.246, pruned_loss=0.03483, over 4907.00 frames. ], tot_loss[loss=0.1717, simple_loss=0.2459, pruned_loss=0.04874, over 956393.01 frames. ], batch size: 37, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:15:29,773 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2467, 2.2430, 1.8814, 2.4834, 2.2367, 2.0531, 2.5604, 2.3176], + device='cuda:1'), covar=tensor([0.1210, 0.1988, 0.2493, 0.2046, 0.2062, 0.1414, 0.2569, 0.1390], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0190, 0.0236, 0.0253, 0.0250, 0.0208, 0.0216, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:15:51,094 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.228e+01 1.484e+02 1.662e+02 2.061e+02 3.704e+02, threshold=3.324e+02, percent-clipped=1.0 +2023-03-27 11:15:51,717 INFO [finetune.py:976] (1/7) Epoch 29, batch 2700, loss[loss=0.1771, simple_loss=0.2394, pruned_loss=0.05736, over 4816.00 frames. ], tot_loss[loss=0.1705, simple_loss=0.2445, pruned_loss=0.04824, over 952652.81 frames. ], batch size: 33, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:16:02,251 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=163090.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:16:25,140 INFO [finetune.py:976] (1/7) Epoch 29, batch 2750, loss[loss=0.1761, simple_loss=0.2363, pruned_loss=0.0579, over 4813.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2411, pruned_loss=0.04768, over 950992.03 frames. ], batch size: 41, lr: 2.84e-03, grad_scale: 32.0 +2023-03-27 11:16:52,306 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=163151.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:17:03,845 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=163167.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 11:17:10,463 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.013e+02 1.483e+02 1.714e+02 1.995e+02 3.279e+02, threshold=3.429e+02, percent-clipped=0.0 +2023-03-27 11:17:10,479 INFO [finetune.py:976] (1/7) Epoch 29, batch 2800, loss[loss=0.1297, simple_loss=0.2095, pruned_loss=0.02492, over 4934.00 frames. ], tot_loss[loss=0.1659, simple_loss=0.2381, pruned_loss=0.04685, over 952976.02 frames. ], batch size: 38, lr: 2.84e-03, grad_scale: 16.0 +2023-03-27 11:17:15,475 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=163183.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:17:37,945 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=163215.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:17:44,448 INFO [finetune.py:976] (1/7) Epoch 29, batch 2850, loss[loss=0.1427, simple_loss=0.2197, pruned_loss=0.03287, over 4766.00 frames. ], tot_loss[loss=0.1642, simple_loss=0.236, pruned_loss=0.04614, over 953127.32 frames. ], batch size: 27, lr: 2.84e-03, grad_scale: 16.0 +2023-03-27 11:17:56,658 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.3030, 5.1599, 4.9904, 3.4590, 5.3448, 4.1568, 1.8298, 3.9883], + device='cuda:1'), covar=tensor([0.2284, 0.2085, 0.1436, 0.2657, 0.0826, 0.0718, 0.4165, 0.1182], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0181, 0.0160, 0.0130, 0.0163, 0.0124, 0.0150, 0.0126], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 11:17:59,031 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=163240.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:17:59,638 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7643, 1.6844, 1.4976, 1.8262, 2.1663, 1.8371, 1.7248, 1.5115], + device='cuda:1'), covar=tensor([0.1893, 0.1832, 0.1781, 0.1505, 0.1558, 0.1234, 0.2152, 0.1683], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0213, 0.0217, 0.0199, 0.0246, 0.0191, 0.0218, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:18:31,673 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.921e+01 1.528e+02 1.890e+02 2.266e+02 3.566e+02, threshold=3.779e+02, percent-clipped=1.0 +2023-03-27 11:18:31,689 INFO [finetune.py:976] (1/7) Epoch 29, batch 2900, loss[loss=0.1979, simple_loss=0.2899, pruned_loss=0.05293, over 4905.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2409, pruned_loss=0.04797, over 952617.94 frames. ], batch size: 43, lr: 2.84e-03, grad_scale: 16.0 +2023-03-27 11:18:33,624 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2692, 2.2464, 2.8367, 2.5344, 2.5217, 4.6776, 2.3900, 2.4214], + device='cuda:1'), covar=tensor([0.0980, 0.1710, 0.1002, 0.0902, 0.1408, 0.0225, 0.1391, 0.1817], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0083, 0.0074, 0.0077, 0.0092, 0.0081, 0.0086, 0.0081], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 11:18:38,923 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0705, 0.9684, 1.0130, 0.4934, 0.9283, 1.1842, 1.1941, 1.0147], + device='cuda:1'), covar=tensor([0.0803, 0.0536, 0.0558, 0.0467, 0.0559, 0.0600, 0.0381, 0.0680], + device='cuda:1'), in_proj_covar=tensor([0.0120, 0.0147, 0.0131, 0.0122, 0.0131, 0.0130, 0.0142, 0.0152], + device='cuda:1'), out_proj_covar=tensor([8.7871e-05, 1.0573e-04, 9.3068e-05, 8.5318e-05, 9.1872e-05, 9.1989e-05, + 1.0058e-04, 1.0841e-04], device='cuda:1') +2023-03-27 11:18:52,057 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=163301.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:19:08,249 INFO [finetune.py:976] (1/7) Epoch 29, batch 2950, loss[loss=0.1686, simple_loss=0.2426, pruned_loss=0.04733, over 4840.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.244, pruned_loss=0.04895, over 954255.48 frames. ], batch size: 47, lr: 2.84e-03, grad_scale: 16.0 +2023-03-27 11:19:20,687 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.21 vs. limit=5.0 +2023-03-27 11:19:21,370 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.25 vs. limit=5.0 +2023-03-27 11:19:49,295 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.216e+02 1.530e+02 1.927e+02 2.309e+02 4.929e+02, threshold=3.855e+02, percent-clipped=3.0 +2023-03-27 11:19:49,311 INFO [finetune.py:976] (1/7) Epoch 29, batch 3000, loss[loss=0.2082, simple_loss=0.2793, pruned_loss=0.06855, over 4840.00 frames. ], tot_loss[loss=0.1709, simple_loss=0.2441, pruned_loss=0.04888, over 953428.40 frames. ], batch size: 47, lr: 2.84e-03, grad_scale: 16.0 +2023-03-27 11:19:49,311 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 11:20:05,061 INFO [finetune.py:1010] (1/7) Epoch 29, validation: loss=0.158, simple_loss=0.2251, pruned_loss=0.04545, over 2265189.00 frames. +2023-03-27 11:20:05,061 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 11:20:15,057 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-03-27 11:20:43,032 INFO [finetune.py:976] (1/7) Epoch 29, batch 3050, loss[loss=0.1738, simple_loss=0.2436, pruned_loss=0.05193, over 4868.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.2435, pruned_loss=0.04777, over 954004.79 frames. ], batch size: 34, lr: 2.84e-03, grad_scale: 16.0 +2023-03-27 11:20:57,920 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=163446.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:21:16,333 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.820e+01 1.385e+02 1.622e+02 1.937e+02 3.745e+02, threshold=3.244e+02, percent-clipped=0.0 +2023-03-27 11:21:16,348 INFO [finetune.py:976] (1/7) Epoch 29, batch 3100, loss[loss=0.1736, simple_loss=0.2309, pruned_loss=0.05818, over 4892.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.242, pruned_loss=0.04737, over 954023.02 frames. ], batch size: 32, lr: 2.84e-03, grad_scale: 16.0 +2023-03-27 11:21:17,689 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9705, 1.9204, 1.7426, 2.1343, 2.3455, 2.1326, 1.8167, 1.6412], + device='cuda:1'), covar=tensor([0.2182, 0.1979, 0.1961, 0.1635, 0.1615, 0.1078, 0.2126, 0.2002], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0212, 0.0216, 0.0198, 0.0245, 0.0191, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:21:22,347 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=163483.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:21:51,332 INFO [finetune.py:976] (1/7) Epoch 29, batch 3150, loss[loss=0.1687, simple_loss=0.2305, pruned_loss=0.05344, over 4814.00 frames. ], tot_loss[loss=0.1676, simple_loss=0.2403, pruned_loss=0.04741, over 956977.40 frames. ], batch size: 38, lr: 2.84e-03, grad_scale: 16.0 +2023-03-27 11:21:51,444 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3390, 2.1209, 1.7043, 0.7759, 2.0330, 1.9184, 1.5959, 2.0744], + device='cuda:1'), covar=tensor([0.0840, 0.0921, 0.1634, 0.2098, 0.1110, 0.2115, 0.2445, 0.0915], + device='cuda:1'), in_proj_covar=tensor([0.0170, 0.0187, 0.0200, 0.0180, 0.0207, 0.0208, 0.0222, 0.0193], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:21:55,027 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=163531.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:22:33,519 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.004e+02 1.497e+02 1.704e+02 2.227e+02 4.213e+02, threshold=3.407e+02, percent-clipped=5.0 +2023-03-27 11:22:33,535 INFO [finetune.py:976] (1/7) Epoch 29, batch 3200, loss[loss=0.2128, simple_loss=0.2736, pruned_loss=0.07603, over 4244.00 frames. ], tot_loss[loss=0.166, simple_loss=0.2377, pruned_loss=0.04714, over 956214.02 frames. ], batch size: 65, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:22:49,203 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=163596.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:22:56,135 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2808, 2.1281, 1.7353, 2.1318, 2.2024, 1.9286, 2.4501, 2.2121], + device='cuda:1'), covar=tensor([0.1220, 0.1977, 0.2901, 0.2459, 0.2568, 0.1728, 0.2798, 0.1698], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0190, 0.0236, 0.0253, 0.0249, 0.0208, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:23:05,131 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2482, 2.1413, 1.6592, 2.2066, 2.1257, 1.8636, 2.4619, 2.2098], + device='cuda:1'), covar=tensor([0.1204, 0.1820, 0.2872, 0.2234, 0.2403, 0.1708, 0.2692, 0.1634], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0189, 0.0235, 0.0252, 0.0249, 0.0207, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:23:07,473 INFO [finetune.py:976] (1/7) Epoch 29, batch 3250, loss[loss=0.1806, simple_loss=0.2655, pruned_loss=0.04783, over 4843.00 frames. ], tot_loss[loss=0.1663, simple_loss=0.2385, pruned_loss=0.04708, over 955557.08 frames. ], batch size: 49, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:23:30,851 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8071, 1.7708, 1.8178, 1.1595, 1.8294, 1.9103, 1.8825, 1.5417], + device='cuda:1'), covar=tensor([0.0646, 0.0675, 0.0713, 0.0895, 0.0783, 0.0735, 0.0629, 0.1282], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0138, 0.0140, 0.0119, 0.0128, 0.0139, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:23:52,619 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.061e+02 1.629e+02 1.981e+02 2.506e+02 4.570e+02, threshold=3.962e+02, percent-clipped=6.0 +2023-03-27 11:23:52,635 INFO [finetune.py:976] (1/7) Epoch 29, batch 3300, loss[loss=0.193, simple_loss=0.2675, pruned_loss=0.05923, over 4936.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2435, pruned_loss=0.04898, over 955341.24 frames. ], batch size: 38, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:24:21,187 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1102, 1.3019, 0.7155, 1.8395, 2.3236, 1.6573, 1.6288, 1.8994], + device='cuda:1'), covar=tensor([0.1396, 0.2141, 0.2159, 0.1153, 0.1829, 0.2067, 0.1351, 0.1863], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0093, 0.0108, 0.0093, 0.0120, 0.0091, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 11:24:29,884 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5102, 2.3927, 2.5721, 1.4330, 2.9892, 3.1124, 2.6818, 2.2087], + device='cuda:1'), covar=tensor([0.0932, 0.0841, 0.0431, 0.0676, 0.0443, 0.0609, 0.0483, 0.0704], + device='cuda:1'), in_proj_covar=tensor([0.0120, 0.0147, 0.0130, 0.0121, 0.0130, 0.0129, 0.0141, 0.0151], + device='cuda:1'), out_proj_covar=tensor([8.7347e-05, 1.0525e-04, 9.2507e-05, 8.4922e-05, 9.1214e-05, 9.1387e-05, + 1.0021e-04, 1.0733e-04], device='cuda:1') +2023-03-27 11:24:37,006 INFO [finetune.py:976] (1/7) Epoch 29, batch 3350, loss[loss=0.144, simple_loss=0.2214, pruned_loss=0.03325, over 4221.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.2445, pruned_loss=0.04909, over 954102.14 frames. ], batch size: 18, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:24:45,558 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-03-27 11:24:55,339 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=163746.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:25:10,096 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.01 vs. limit=5.0 +2023-03-27 11:25:21,355 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.117e+02 1.486e+02 1.801e+02 2.174e+02 4.171e+02, threshold=3.603e+02, percent-clipped=1.0 +2023-03-27 11:25:21,370 INFO [finetune.py:976] (1/7) Epoch 29, batch 3400, loss[loss=0.1449, simple_loss=0.2243, pruned_loss=0.03273, over 4821.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2441, pruned_loss=0.04857, over 952647.06 frames. ], batch size: 25, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:25:37,709 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=163794.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:25:58,974 INFO [finetune.py:976] (1/7) Epoch 29, batch 3450, loss[loss=0.1329, simple_loss=0.2168, pruned_loss=0.02447, over 4760.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2438, pruned_loss=0.04814, over 954202.85 frames. ], batch size: 26, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:26:04,511 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.4041, 3.4402, 3.1938, 1.5205, 3.5514, 2.6003, 0.7493, 2.3282], + device='cuda:1'), covar=tensor([0.2652, 0.2241, 0.1790, 0.3623, 0.1218, 0.1128, 0.4662, 0.1661], + device='cuda:1'), in_proj_covar=tensor([0.0152, 0.0181, 0.0161, 0.0130, 0.0163, 0.0124, 0.0150, 0.0126], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 11:26:05,208 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3245, 2.3664, 1.9794, 2.3720, 2.2321, 2.2429, 2.1793, 3.0626], + device='cuda:1'), covar=tensor([0.3534, 0.4355, 0.3324, 0.3920, 0.4423, 0.2558, 0.4143, 0.1523], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0265, 0.0239, 0.0275, 0.0261, 0.0232, 0.0260, 0.0240], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:26:41,250 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.103e+01 1.539e+02 1.877e+02 2.200e+02 3.171e+02, threshold=3.754e+02, percent-clipped=0.0 +2023-03-27 11:26:41,266 INFO [finetune.py:976] (1/7) Epoch 29, batch 3500, loss[loss=0.1616, simple_loss=0.2397, pruned_loss=0.04174, over 4920.00 frames. ], tot_loss[loss=0.1685, simple_loss=0.2415, pruned_loss=0.04773, over 954191.44 frames. ], batch size: 38, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:26:55,488 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=163896.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:26:56,711 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1012, 3.5643, 3.7526, 3.9161, 3.8491, 3.5673, 4.1617, 1.2230], + device='cuda:1'), covar=tensor([0.0924, 0.0958, 0.1074, 0.1082, 0.1423, 0.1960, 0.0876, 0.6214], + device='cuda:1'), in_proj_covar=tensor([0.0351, 0.0245, 0.0285, 0.0294, 0.0337, 0.0284, 0.0304, 0.0300], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:27:17,137 INFO [finetune.py:976] (1/7) Epoch 29, batch 3550, loss[loss=0.1245, simple_loss=0.1989, pruned_loss=0.02512, over 4758.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.2393, pruned_loss=0.04713, over 954853.58 frames. ], batch size: 27, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:27:38,585 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=163944.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:27:52,972 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3411, 2.3796, 2.0087, 2.3885, 2.2938, 2.2192, 2.2686, 3.1330], + device='cuda:1'), covar=tensor([0.3835, 0.4751, 0.3394, 0.4162, 0.4428, 0.2655, 0.4294, 0.1690], + device='cuda:1'), in_proj_covar=tensor([0.0291, 0.0266, 0.0240, 0.0276, 0.0262, 0.0233, 0.0261, 0.0240], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:27:59,307 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.875e+01 1.446e+02 1.734e+02 2.162e+02 4.667e+02, threshold=3.468e+02, percent-clipped=1.0 +2023-03-27 11:27:59,322 INFO [finetune.py:976] (1/7) Epoch 29, batch 3600, loss[loss=0.1532, simple_loss=0.2265, pruned_loss=0.03997, over 4887.00 frames. ], tot_loss[loss=0.165, simple_loss=0.2371, pruned_loss=0.04647, over 953529.67 frames. ], batch size: 32, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:28:36,274 INFO [finetune.py:976] (1/7) Epoch 29, batch 3650, loss[loss=0.175, simple_loss=0.2577, pruned_loss=0.04621, over 4830.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2405, pruned_loss=0.04785, over 953939.20 frames. ], batch size: 47, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:28:44,009 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.9983, 0.9361, 0.9338, 1.0336, 1.2132, 1.1424, 1.0262, 0.9119], + device='cuda:1'), covar=tensor([0.0439, 0.0373, 0.0728, 0.0382, 0.0325, 0.0537, 0.0401, 0.0508], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0108, 0.0150, 0.0113, 0.0104, 0.0119, 0.0105, 0.0116], + device='cuda:1'), out_proj_covar=tensor([7.9967e-05, 8.2250e-05, 1.1628e-04, 8.5788e-05, 8.0070e-05, 8.7553e-05, + 7.8388e-05, 8.7996e-05], device='cuda:1') +2023-03-27 11:28:44,597 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4366, 1.7125, 1.4499, 1.4822, 2.0115, 2.0525, 1.7175, 1.6779], + device='cuda:1'), covar=tensor([0.0644, 0.0440, 0.0679, 0.0400, 0.0367, 0.0604, 0.0453, 0.0484], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0108, 0.0150, 0.0113, 0.0104, 0.0119, 0.0105, 0.0116], + device='cuda:1'), out_proj_covar=tensor([7.9951e-05, 8.2237e-05, 1.1625e-04, 8.5771e-05, 8.0054e-05, 8.7528e-05, + 7.8376e-05, 8.7984e-05], device='cuda:1') +2023-03-27 11:29:19,111 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.932e+01 1.541e+02 1.828e+02 2.331e+02 7.133e+02, threshold=3.656e+02, percent-clipped=4.0 +2023-03-27 11:29:19,127 INFO [finetune.py:976] (1/7) Epoch 29, batch 3700, loss[loss=0.1871, simple_loss=0.2737, pruned_loss=0.05026, over 4821.00 frames. ], tot_loss[loss=0.17, simple_loss=0.2432, pruned_loss=0.04843, over 956132.31 frames. ], batch size: 39, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:29:19,852 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7020, 1.6948, 1.8806, 1.2849, 1.5638, 1.9686, 1.6266, 2.0714], + device='cuda:1'), covar=tensor([0.1025, 0.1832, 0.1121, 0.1396, 0.0872, 0.0954, 0.2540, 0.0684], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0206, 0.0192, 0.0188, 0.0173, 0.0212, 0.0217, 0.0196], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:29:28,316 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-03-27 11:29:36,831 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=164090.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:30:00,904 INFO [finetune.py:976] (1/7) Epoch 29, batch 3750, loss[loss=0.179, simple_loss=0.2578, pruned_loss=0.05008, over 4887.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2438, pruned_loss=0.04843, over 957179.83 frames. ], batch size: 43, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:30:17,223 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=164151.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:30:20,021 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=164155.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:30:37,149 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.292e+01 1.655e+02 1.832e+02 2.179e+02 3.362e+02, threshold=3.664e+02, percent-clipped=0.0 +2023-03-27 11:30:37,165 INFO [finetune.py:976] (1/7) Epoch 29, batch 3800, loss[loss=0.1215, simple_loss=0.1917, pruned_loss=0.02569, over 4725.00 frames. ], tot_loss[loss=0.1707, simple_loss=0.2443, pruned_loss=0.0485, over 957188.59 frames. ], batch size: 23, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:30:57,290 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.3457, 1.4402, 1.2747, 1.3498, 1.6719, 1.5932, 1.4046, 1.2574], + device='cuda:1'), covar=tensor([0.0398, 0.0286, 0.0640, 0.0302, 0.0237, 0.0476, 0.0343, 0.0445], + device='cuda:1'), in_proj_covar=tensor([0.0102, 0.0107, 0.0148, 0.0112, 0.0103, 0.0118, 0.0104, 0.0115], + device='cuda:1'), out_proj_covar=tensor([7.9027e-05, 8.1594e-05, 1.1520e-04, 8.5009e-05, 7.9385e-05, 8.6631e-05, + 7.7524e-05, 8.7181e-05], device='cuda:1') +2023-03-27 11:31:03,881 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=164216.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:31:09,782 INFO [finetune.py:976] (1/7) Epoch 29, batch 3850, loss[loss=0.1597, simple_loss=0.224, pruned_loss=0.0477, over 4685.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2424, pruned_loss=0.04771, over 954917.89 frames. ], batch size: 23, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:31:29,005 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9481, 1.8248, 1.7354, 1.9156, 1.4768, 4.4045, 1.5857, 2.0054], + device='cuda:1'), covar=tensor([0.3128, 0.2322, 0.1986, 0.2216, 0.1528, 0.0130, 0.2500, 0.1118], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0124, 0.0113, 0.0095, 0.0093, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 11:31:45,644 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.748e+01 1.394e+02 1.747e+02 2.221e+02 3.425e+02, threshold=3.494e+02, percent-clipped=0.0 +2023-03-27 11:31:45,659 INFO [finetune.py:976] (1/7) Epoch 29, batch 3900, loss[loss=0.13, simple_loss=0.2029, pruned_loss=0.02855, over 4759.00 frames. ], tot_loss[loss=0.1663, simple_loss=0.2393, pruned_loss=0.04664, over 955431.08 frames. ], batch size: 28, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:32:27,501 INFO [finetune.py:976] (1/7) Epoch 29, batch 3950, loss[loss=0.1442, simple_loss=0.2211, pruned_loss=0.0337, over 4773.00 frames. ], tot_loss[loss=0.1642, simple_loss=0.2361, pruned_loss=0.0461, over 954163.83 frames. ], batch size: 26, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:32:31,665 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.7083, 2.4454, 2.0664, 2.8753, 2.5894, 2.3192, 3.0931, 2.6400], + device='cuda:1'), covar=tensor([0.1289, 0.2335, 0.2994, 0.2479, 0.2636, 0.1770, 0.2799, 0.1881], + device='cuda:1'), in_proj_covar=tensor([0.0189, 0.0190, 0.0236, 0.0253, 0.0250, 0.0209, 0.0215, 0.0203], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:32:58,638 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=164360.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:33:07,942 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.65 vs. limit=2.0 +2023-03-27 11:33:10,186 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9428, 1.8754, 1.7645, 1.9109, 1.5795, 3.6239, 1.7684, 2.2086], + device='cuda:1'), covar=tensor([0.3428, 0.2566, 0.2055, 0.2420, 0.1504, 0.0301, 0.2170, 0.0920], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0124, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 11:33:11,862 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.819e+01 1.491e+02 1.748e+02 1.970e+02 3.605e+02, threshold=3.496e+02, percent-clipped=1.0 +2023-03-27 11:33:11,878 INFO [finetune.py:976] (1/7) Epoch 29, batch 4000, loss[loss=0.1487, simple_loss=0.2284, pruned_loss=0.03451, over 4198.00 frames. ], tot_loss[loss=0.163, simple_loss=0.235, pruned_loss=0.04551, over 953643.52 frames. ], batch size: 65, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:33:25,596 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8765, 1.3124, 1.9391, 1.9845, 1.7792, 1.7310, 1.9061, 1.9044], + device='cuda:1'), covar=tensor([0.3871, 0.3793, 0.2967, 0.3352, 0.4341, 0.3595, 0.3983, 0.2821], + device='cuda:1'), in_proj_covar=tensor([0.0269, 0.0250, 0.0270, 0.0300, 0.0299, 0.0276, 0.0304, 0.0254], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:33:34,783 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.61 vs. limit=2.0 +2023-03-27 11:33:42,370 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4757, 1.3925, 1.3399, 1.4279, 1.0444, 2.8495, 1.0846, 1.4342], + device='cuda:1'), covar=tensor([0.3317, 0.2566, 0.2246, 0.2436, 0.1838, 0.0281, 0.2962, 0.1316], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0116, 0.0120, 0.0124, 0.0113, 0.0095, 0.0094, 0.0094], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 11:33:42,974 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=164421.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:33:45,304 INFO [finetune.py:976] (1/7) Epoch 29, batch 4050, loss[loss=0.1531, simple_loss=0.2321, pruned_loss=0.03709, over 4791.00 frames. ], tot_loss[loss=0.167, simple_loss=0.2394, pruned_loss=0.0473, over 952246.18 frames. ], batch size: 29, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:34:06,765 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164446.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:34:27,956 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4790, 1.3072, 1.2020, 1.4385, 1.6136, 1.5355, 1.0290, 1.2349], + device='cuda:1'), covar=tensor([0.2389, 0.2178, 0.2257, 0.1883, 0.1693, 0.1337, 0.2708, 0.2057], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0212, 0.0217, 0.0199, 0.0246, 0.0190, 0.0218, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:34:29,030 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.788e+01 1.600e+02 1.814e+02 2.182e+02 3.811e+02, threshold=3.628e+02, percent-clipped=1.0 +2023-03-27 11:34:29,046 INFO [finetune.py:976] (1/7) Epoch 29, batch 4100, loss[loss=0.1827, simple_loss=0.2669, pruned_loss=0.04924, over 4820.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.242, pruned_loss=0.04778, over 952294.55 frames. ], batch size: 40, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:35:04,639 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164511.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:35:13,895 INFO [finetune.py:976] (1/7) Epoch 29, batch 4150, loss[loss=0.1785, simple_loss=0.2517, pruned_loss=0.0526, over 4909.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2438, pruned_loss=0.04826, over 953478.16 frames. ], batch size: 36, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:35:42,374 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=164568.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:35:46,474 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.025e+02 1.582e+02 1.835e+02 2.360e+02 4.097e+02, threshold=3.670e+02, percent-clipped=1.0 +2023-03-27 11:35:46,490 INFO [finetune.py:976] (1/7) Epoch 29, batch 4200, loss[loss=0.1665, simple_loss=0.2394, pruned_loss=0.04678, over 4743.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.2436, pruned_loss=0.04772, over 955909.90 frames. ], batch size: 54, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:35:48,882 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=164578.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:35:50,745 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9745, 1.3900, 2.0399, 1.9978, 1.7815, 1.7563, 1.9269, 1.9386], + device='cuda:1'), covar=tensor([0.3898, 0.3872, 0.3175, 0.3679, 0.4893, 0.3963, 0.4545, 0.2921], + device='cuda:1'), in_proj_covar=tensor([0.0269, 0.0250, 0.0270, 0.0300, 0.0300, 0.0277, 0.0305, 0.0254], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:36:13,888 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7563, 1.6721, 1.4555, 1.8386, 2.2423, 1.9203, 1.6752, 1.4200], + device='cuda:1'), covar=tensor([0.2120, 0.1848, 0.1955, 0.1677, 0.1499, 0.1162, 0.2186, 0.1982], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0212, 0.0217, 0.0199, 0.0247, 0.0191, 0.0218, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:36:20,292 INFO [finetune.py:976] (1/7) Epoch 29, batch 4250, loss[loss=0.1763, simple_loss=0.2481, pruned_loss=0.05224, over 4910.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2414, pruned_loss=0.04715, over 955753.98 frames. ], batch size: 37, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:36:21,019 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.3778, 2.2295, 2.3799, 1.6147, 2.3572, 2.5590, 2.5023, 1.9554], + device='cuda:1'), covar=tensor([0.0599, 0.0654, 0.0673, 0.0820, 0.0683, 0.0621, 0.0592, 0.1172], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0138, 0.0141, 0.0118, 0.0129, 0.0139, 0.0140, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:36:23,249 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=164629.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:36:29,280 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=164639.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 11:36:41,352 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6850, 1.5695, 1.5028, 1.6213, 1.1494, 3.4332, 1.2683, 1.5924], + device='cuda:1'), covar=tensor([0.3167, 0.2419, 0.2147, 0.2313, 0.1764, 0.0198, 0.2606, 0.1258], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0121, 0.0124, 0.0114, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 11:36:43,173 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=164658.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:36:53,782 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.101e+02 1.450e+02 1.663e+02 2.112e+02 3.483e+02, threshold=3.326e+02, percent-clipped=0.0 +2023-03-27 11:36:53,798 INFO [finetune.py:976] (1/7) Epoch 29, batch 4300, loss[loss=0.1363, simple_loss=0.2196, pruned_loss=0.0265, over 4906.00 frames. ], tot_loss[loss=0.1668, simple_loss=0.2393, pruned_loss=0.04712, over 956184.23 frames. ], batch size: 36, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:37:12,582 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.53 vs. limit=2.0 +2023-03-27 11:37:32,858 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0418, 1.9039, 1.6375, 1.6745, 1.8597, 1.8073, 1.8570, 2.4987], + device='cuda:1'), covar=tensor([0.3330, 0.3551, 0.3090, 0.3353, 0.3446, 0.2377, 0.3070, 0.1660], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0264, 0.0239, 0.0275, 0.0262, 0.0233, 0.0260, 0.0240], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:37:36,001 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-27 11:37:39,419 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164716.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:37:41,306 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=164719.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:37:44,825 INFO [finetune.py:976] (1/7) Epoch 29, batch 4350, loss[loss=0.1664, simple_loss=0.2362, pruned_loss=0.04827, over 4916.00 frames. ], tot_loss[loss=0.164, simple_loss=0.2361, pruned_loss=0.04593, over 954946.20 frames. ], batch size: 37, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:37:58,678 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=164746.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:37:58,697 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=164746.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:38:20,795 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.196e+01 1.395e+02 1.786e+02 2.060e+02 3.738e+02, threshold=3.572e+02, percent-clipped=2.0 +2023-03-27 11:38:20,811 INFO [finetune.py:976] (1/7) Epoch 29, batch 4400, loss[loss=0.1245, simple_loss=0.1986, pruned_loss=0.02515, over 4774.00 frames. ], tot_loss[loss=0.1641, simple_loss=0.2362, pruned_loss=0.04595, over 955642.59 frames. ], batch size: 26, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:38:33,467 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=164794.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:38:43,374 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=164807.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:38:46,307 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=164811.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:38:54,767 INFO [finetune.py:976] (1/7) Epoch 29, batch 4450, loss[loss=0.2438, simple_loss=0.3119, pruned_loss=0.08784, over 4904.00 frames. ], tot_loss[loss=0.1677, simple_loss=0.2411, pruned_loss=0.04714, over 957147.60 frames. ], batch size: 36, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:39:02,790 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=2.02 vs. limit=2.0 +2023-03-27 11:39:12,257 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.76 vs. limit=5.0 +2023-03-27 11:39:23,087 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=164859.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:39:37,138 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.009e+02 1.552e+02 1.889e+02 2.217e+02 4.905e+02, threshold=3.778e+02, percent-clipped=1.0 +2023-03-27 11:39:37,154 INFO [finetune.py:976] (1/7) Epoch 29, batch 4500, loss[loss=0.2071, simple_loss=0.2632, pruned_loss=0.0755, over 4902.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2433, pruned_loss=0.04792, over 958259.08 frames. ], batch size: 46, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:39:57,354 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8670, 1.8973, 1.5937, 2.0787, 2.4036, 2.0359, 1.7621, 1.5130], + device='cuda:1'), covar=tensor([0.1992, 0.1713, 0.1789, 0.1412, 0.1547, 0.1133, 0.2126, 0.1862], + device='cuda:1'), in_proj_covar=tensor([0.0245, 0.0211, 0.0216, 0.0198, 0.0245, 0.0190, 0.0217, 0.0205], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:40:13,939 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-27 11:40:22,144 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164924.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:40:22,679 INFO [finetune.py:976] (1/7) Epoch 29, batch 4550, loss[loss=0.1667, simple_loss=0.2343, pruned_loss=0.04958, over 4930.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2427, pruned_loss=0.04723, over 957206.99 frames. ], batch size: 33, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:40:28,154 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164934.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 11:40:55,993 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.117e+02 1.494e+02 1.764e+02 2.048e+02 3.220e+02, threshold=3.528e+02, percent-clipped=0.0 +2023-03-27 11:40:56,009 INFO [finetune.py:976] (1/7) Epoch 29, batch 4600, loss[loss=0.1709, simple_loss=0.2346, pruned_loss=0.05364, over 4822.00 frames. ], tot_loss[loss=0.1683, simple_loss=0.2425, pruned_loss=0.04711, over 957185.13 frames. ], batch size: 40, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:40:56,733 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8310, 1.8036, 1.5639, 1.9372, 2.2549, 1.9666, 1.6059, 1.4918], + device='cuda:1'), covar=tensor([0.2077, 0.1793, 0.1890, 0.1519, 0.1597, 0.1127, 0.2338, 0.1834], + device='cuda:1'), in_proj_covar=tensor([0.0246, 0.0211, 0.0216, 0.0198, 0.0245, 0.0190, 0.0217, 0.0206], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:41:14,790 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9444, 1.8491, 1.5227, 1.6698, 1.7600, 1.7195, 1.7764, 2.3999], + device='cuda:1'), covar=tensor([0.3998, 0.3689, 0.3505, 0.3482, 0.4097, 0.2579, 0.3634, 0.1924], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0265, 0.0240, 0.0275, 0.0262, 0.0233, 0.0260, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:41:22,216 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=165014.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:41:23,394 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165016.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:41:29,265 INFO [finetune.py:976] (1/7) Epoch 29, batch 4650, loss[loss=0.1869, simple_loss=0.2483, pruned_loss=0.06278, over 4893.00 frames. ], tot_loss[loss=0.168, simple_loss=0.241, pruned_loss=0.04752, over 956828.94 frames. ], batch size: 32, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:41:38,982 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9800, 2.5426, 2.4845, 1.4527, 2.6916, 2.1573, 2.0295, 2.4916], + device='cuda:1'), covar=tensor([0.1219, 0.0895, 0.1773, 0.2113, 0.1634, 0.2118, 0.2172, 0.1222], + device='cuda:1'), in_proj_covar=tensor([0.0173, 0.0191, 0.0205, 0.0182, 0.0211, 0.0212, 0.0224, 0.0199], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:41:49,650 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-03-27 11:41:53,745 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=165064.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:42:01,328 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.062e+02 1.478e+02 1.798e+02 2.114e+02 1.110e+03, threshold=3.597e+02, percent-clipped=3.0 +2023-03-27 11:42:01,344 INFO [finetune.py:976] (1/7) Epoch 29, batch 4700, loss[loss=0.1367, simple_loss=0.2178, pruned_loss=0.02787, over 4896.00 frames. ], tot_loss[loss=0.1659, simple_loss=0.2383, pruned_loss=0.04674, over 957002.60 frames. ], batch size: 35, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:42:27,131 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=165102.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:42:43,742 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165123.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:42:44,888 INFO [finetune.py:976] (1/7) Epoch 29, batch 4750, loss[loss=0.1499, simple_loss=0.2197, pruned_loss=0.04005, over 4928.00 frames. ], tot_loss[loss=0.1647, simple_loss=0.2366, pruned_loss=0.04635, over 957935.11 frames. ], batch size: 33, lr: 2.83e-03, grad_scale: 16.0 +2023-03-27 11:43:11,116 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-27 11:43:21,524 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.139e+02 1.579e+02 1.786e+02 2.031e+02 3.721e+02, threshold=3.572e+02, percent-clipped=1.0 +2023-03-27 11:43:21,540 INFO [finetune.py:976] (1/7) Epoch 29, batch 4800, loss[loss=0.1618, simple_loss=0.2285, pruned_loss=0.04749, over 4756.00 frames. ], tot_loss[loss=0.1656, simple_loss=0.238, pruned_loss=0.04662, over 956399.28 frames. ], batch size: 27, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:43:26,590 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-03-27 11:43:27,582 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165184.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:43:53,907 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165224.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:43:54,433 INFO [finetune.py:976] (1/7) Epoch 29, batch 4850, loss[loss=0.2231, simple_loss=0.2909, pruned_loss=0.07771, over 4815.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2411, pruned_loss=0.04739, over 955671.82 frames. ], batch size: 38, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:44:00,506 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165234.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 11:44:22,090 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.2887, 1.4761, 1.5163, 0.8230, 1.5450, 1.8114, 1.7201, 1.4507], + device='cuda:1'), covar=tensor([0.0991, 0.0813, 0.0623, 0.0592, 0.0568, 0.0668, 0.0533, 0.0749], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0131, 0.0121, 0.0131, 0.0130, 0.0141, 0.0151], + device='cuda:1'), out_proj_covar=tensor([8.8130e-05, 1.0505e-04, 9.2971e-05, 8.4948e-05, 9.1720e-05, 9.2080e-05, + 1.0047e-04, 1.0795e-04], device='cuda:1') +2023-03-27 11:44:24,480 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=165272.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:44:26,733 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.028e+02 1.742e+02 1.982e+02 2.317e+02 4.079e+02, threshold=3.965e+02, percent-clipped=3.0 +2023-03-27 11:44:26,748 INFO [finetune.py:976] (1/7) Epoch 29, batch 4900, loss[loss=0.1892, simple_loss=0.283, pruned_loss=0.04765, over 4901.00 frames. ], tot_loss[loss=0.1677, simple_loss=0.2413, pruned_loss=0.04712, over 954839.21 frames. ], batch size: 43, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:44:34,164 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5518, 1.3930, 1.3871, 1.4597, 1.3710, 3.5379, 1.3675, 1.8977], + device='cuda:1'), covar=tensor([0.4037, 0.3166, 0.2548, 0.3010, 0.1729, 0.0279, 0.2675, 0.1160], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0121, 0.0125, 0.0114, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 11:44:41,265 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=165282.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:44:46,928 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 11:45:03,726 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165314.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:45:16,147 INFO [finetune.py:976] (1/7) Epoch 29, batch 4950, loss[loss=0.1328, simple_loss=0.2186, pruned_loss=0.02354, over 4773.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2431, pruned_loss=0.04731, over 953580.10 frames. ], batch size: 29, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:45:20,047 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-03-27 11:45:48,520 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=165362.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:45:56,863 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.088e+01 1.449e+02 1.743e+02 2.087e+02 3.629e+02, threshold=3.486e+02, percent-clipped=0.0 +2023-03-27 11:45:56,879 INFO [finetune.py:976] (1/7) Epoch 29, batch 5000, loss[loss=0.1577, simple_loss=0.2382, pruned_loss=0.03861, over 4913.00 frames. ], tot_loss[loss=0.1669, simple_loss=0.2405, pruned_loss=0.04661, over 952503.54 frames. ], batch size: 37, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:45:59,524 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-27 11:46:01,621 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8398, 1.2802, 0.7527, 1.6967, 2.2034, 1.4223, 1.5941, 1.6718], + device='cuda:1'), covar=tensor([0.1387, 0.2056, 0.1905, 0.1154, 0.1842, 0.1919, 0.1354, 0.1876], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0093, 0.0108, 0.0092, 0.0119, 0.0091, 0.0097, 0.0088], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 11:46:15,401 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165402.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:46:26,288 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0 +2023-03-27 11:46:30,164 INFO [finetune.py:976] (1/7) Epoch 29, batch 5050, loss[loss=0.1903, simple_loss=0.2523, pruned_loss=0.06413, over 4733.00 frames. ], tot_loss[loss=0.165, simple_loss=0.238, pruned_loss=0.04605, over 953657.99 frames. ], batch size: 23, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:46:47,815 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=165450.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:46:47,840 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.3212, 2.9714, 3.0662, 3.2481, 3.1387, 2.8815, 3.3579, 1.0121], + device='cuda:1'), covar=tensor([0.1133, 0.1076, 0.1304, 0.1104, 0.1564, 0.2176, 0.1142, 0.6016], + device='cuda:1'), in_proj_covar=tensor([0.0357, 0.0249, 0.0290, 0.0299, 0.0340, 0.0289, 0.0309, 0.0305], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:46:57,527 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1700, 2.0417, 1.7640, 2.2190, 2.6360, 2.2137, 1.9198, 1.6649], + device='cuda:1'), covar=tensor([0.1963, 0.1789, 0.1791, 0.1493, 0.1764, 0.1070, 0.2248, 0.1778], + device='cuda:1'), in_proj_covar=tensor([0.0247, 0.0212, 0.0217, 0.0199, 0.0246, 0.0191, 0.0219, 0.0207], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:47:02,934 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1139, 1.7785, 2.3612, 1.6111, 2.1364, 2.2788, 1.7231, 2.4207], + device='cuda:1'), covar=tensor([0.1184, 0.2087, 0.1491, 0.2003, 0.0909, 0.1314, 0.2716, 0.0759], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0194, 0.0189, 0.0174, 0.0212, 0.0218, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:47:03,404 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.069e+02 1.397e+02 1.700e+02 2.095e+02 3.650e+02, threshold=3.400e+02, percent-clipped=1.0 +2023-03-27 11:47:03,420 INFO [finetune.py:976] (1/7) Epoch 29, batch 5100, loss[loss=0.1694, simple_loss=0.2381, pruned_loss=0.05034, over 4917.00 frames. ], tot_loss[loss=0.1633, simple_loss=0.2355, pruned_loss=0.04554, over 954966.66 frames. ], batch size: 43, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:47:06,336 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=165479.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:47:46,453 INFO [finetune.py:976] (1/7) Epoch 29, batch 5150, loss[loss=0.1596, simple_loss=0.2337, pruned_loss=0.04272, over 4816.00 frames. ], tot_loss[loss=0.1648, simple_loss=0.2368, pruned_loss=0.04643, over 955445.08 frames. ], batch size: 39, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:47:54,886 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1793, 2.1852, 2.2631, 1.7423, 2.1649, 2.4085, 2.4050, 1.9827], + device='cuda:1'), covar=tensor([0.0615, 0.0663, 0.0709, 0.0842, 0.1169, 0.0655, 0.0515, 0.0990], + device='cuda:1'), in_proj_covar=tensor([0.0133, 0.0140, 0.0142, 0.0119, 0.0130, 0.0141, 0.0141, 0.0164], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:48:04,787 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165551.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:48:20,244 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.818e+01 1.606e+02 1.856e+02 2.249e+02 3.990e+02, threshold=3.713e+02, percent-clipped=3.0 +2023-03-27 11:48:20,260 INFO [finetune.py:976] (1/7) Epoch 29, batch 5200, loss[loss=0.19, simple_loss=0.2615, pruned_loss=0.05926, over 4828.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2407, pruned_loss=0.04752, over 952917.85 frames. ], batch size: 30, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:48:45,763 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165612.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:48:53,937 INFO [finetune.py:976] (1/7) Epoch 29, batch 5250, loss[loss=0.1978, simple_loss=0.2669, pruned_loss=0.06435, over 4902.00 frames. ], tot_loss[loss=0.1693, simple_loss=0.2429, pruned_loss=0.04783, over 955156.26 frames. ], batch size: 43, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:49:00,465 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.2697, 3.6740, 3.8934, 4.1018, 4.0552, 3.7529, 4.3628, 1.2781], + device='cuda:1'), covar=tensor([0.0858, 0.0927, 0.0918, 0.1112, 0.1175, 0.1697, 0.0672, 0.6205], + device='cuda:1'), in_proj_covar=tensor([0.0357, 0.0249, 0.0290, 0.0299, 0.0340, 0.0289, 0.0309, 0.0305], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:49:26,774 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.319e+01 1.511e+02 1.772e+02 2.250e+02 3.554e+02, threshold=3.544e+02, percent-clipped=0.0 +2023-03-27 11:49:26,790 INFO [finetune.py:976] (1/7) Epoch 29, batch 5300, loss[loss=0.2004, simple_loss=0.2703, pruned_loss=0.06523, over 4892.00 frames. ], tot_loss[loss=0.1712, simple_loss=0.2447, pruned_loss=0.04888, over 955206.95 frames. ], batch size: 35, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:49:52,110 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0032, 1.2827, 1.9462, 1.9101, 1.7538, 1.7572, 1.8324, 1.9397], + device='cuda:1'), covar=tensor([0.4325, 0.4409, 0.4276, 0.4326, 0.6085, 0.4906, 0.5096, 0.3874], + device='cuda:1'), in_proj_covar=tensor([0.0270, 0.0250, 0.0269, 0.0301, 0.0301, 0.0278, 0.0306, 0.0256], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:50:10,068 INFO [finetune.py:976] (1/7) Epoch 29, batch 5350, loss[loss=0.1571, simple_loss=0.2346, pruned_loss=0.03978, over 4832.00 frames. ], tot_loss[loss=0.1714, simple_loss=0.245, pruned_loss=0.04892, over 954818.78 frames. ], batch size: 30, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:50:33,771 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165750.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:50:59,727 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.046e+02 1.485e+02 1.788e+02 2.126e+02 3.231e+02, threshold=3.576e+02, percent-clipped=0.0 +2023-03-27 11:50:59,742 INFO [finetune.py:976] (1/7) Epoch 29, batch 5400, loss[loss=0.2129, simple_loss=0.2692, pruned_loss=0.07834, over 4868.00 frames. ], tot_loss[loss=0.1702, simple_loss=0.2429, pruned_loss=0.04877, over 954471.87 frames. ], batch size: 34, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:51:02,237 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165779.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:51:17,904 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165803.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:51:24,167 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165811.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:51:33,003 INFO [finetune.py:976] (1/7) Epoch 29, batch 5450, loss[loss=0.1481, simple_loss=0.2243, pruned_loss=0.03591, over 4829.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2397, pruned_loss=0.04752, over 954416.88 frames. ], batch size: 33, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:51:34,293 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=165827.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:51:45,078 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9267, 1.7149, 1.5461, 1.2025, 1.6522, 1.6586, 1.6571, 2.2319], + device='cuda:1'), covar=tensor([0.3474, 0.3372, 0.3094, 0.3377, 0.3716, 0.2166, 0.3128, 0.1746], + device='cuda:1'), in_proj_covar=tensor([0.0289, 0.0263, 0.0238, 0.0273, 0.0260, 0.0231, 0.0258, 0.0238], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:51:58,116 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2114, 1.9919, 1.8197, 1.8580, 1.9212, 1.9346, 1.9214, 2.5873], + device='cuda:1'), covar=tensor([0.3801, 0.4304, 0.3208, 0.3673, 0.4211, 0.2609, 0.4126, 0.1746], + device='cuda:1'), in_proj_covar=tensor([0.0288, 0.0263, 0.0237, 0.0273, 0.0260, 0.0231, 0.0257, 0.0237], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:51:59,789 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165864.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:52:06,351 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.128e+01 1.456e+02 1.721e+02 2.001e+02 5.868e+02, threshold=3.442e+02, percent-clipped=2.0 +2023-03-27 11:52:06,367 INFO [finetune.py:976] (1/7) Epoch 29, batch 5500, loss[loss=0.1731, simple_loss=0.2386, pruned_loss=0.05378, over 4148.00 frames. ], tot_loss[loss=0.1651, simple_loss=0.237, pruned_loss=0.04659, over 954792.28 frames. ], batch size: 65, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:52:27,330 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=165907.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:52:30,844 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.55 vs. limit=5.0 +2023-03-27 11:52:32,588 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0836, 1.4833, 0.7183, 2.0236, 2.4556, 1.7285, 1.8144, 2.0282], + device='cuda:1'), covar=tensor([0.1535, 0.2111, 0.2177, 0.1185, 0.1821, 0.1817, 0.1380, 0.1906], + device='cuda:1'), in_proj_covar=tensor([0.0089, 0.0094, 0.0109, 0.0093, 0.0120, 0.0092, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 11:52:34,334 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.7591, 1.5479, 1.9022, 1.2917, 1.7836, 1.8230, 1.4307, 2.0296], + device='cuda:1'), covar=tensor([0.1023, 0.1953, 0.1345, 0.1606, 0.0775, 0.1221, 0.2787, 0.0672], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0193, 0.0189, 0.0174, 0.0213, 0.0219, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:52:40,105 INFO [finetune.py:976] (1/7) Epoch 29, batch 5550, loss[loss=0.1706, simple_loss=0.2431, pruned_loss=0.04904, over 4768.00 frames. ], tot_loss[loss=0.1656, simple_loss=0.2379, pruned_loss=0.04662, over 954158.15 frames. ], batch size: 28, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:53:16,527 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8215, 1.7666, 1.6346, 2.0856, 2.1913, 2.0857, 1.6335, 1.5566], + device='cuda:1'), covar=tensor([0.2119, 0.1855, 0.1806, 0.1439, 0.1627, 0.1043, 0.2253, 0.1917], + device='cuda:1'), in_proj_covar=tensor([0.0250, 0.0214, 0.0218, 0.0201, 0.0249, 0.0193, 0.0220, 0.0209], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:53:22,712 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.003e+02 1.570e+02 1.926e+02 2.203e+02 4.633e+02, threshold=3.853e+02, percent-clipped=1.0 +2023-03-27 11:53:22,727 INFO [finetune.py:976] (1/7) Epoch 29, batch 5600, loss[loss=0.1742, simple_loss=0.2483, pruned_loss=0.05008, over 4904.00 frames. ], tot_loss[loss=0.1673, simple_loss=0.2403, pruned_loss=0.04718, over 952294.02 frames. ], batch size: 37, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:53:24,061 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-03-27 11:53:35,810 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-03-27 11:53:50,125 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1831, 1.8663, 2.4423, 4.0790, 2.8588, 2.6535, 0.7458, 3.5754], + device='cuda:1'), covar=tensor([0.1593, 0.1245, 0.1292, 0.0480, 0.0701, 0.1561, 0.2022, 0.0347], + device='cuda:1'), in_proj_covar=tensor([0.0099, 0.0114, 0.0131, 0.0163, 0.0099, 0.0134, 0.0123, 0.0100], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 11:53:54,009 INFO [finetune.py:976] (1/7) Epoch 29, batch 5650, loss[loss=0.1419, simple_loss=0.2235, pruned_loss=0.03015, over 4871.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2429, pruned_loss=0.04792, over 951275.64 frames. ], batch size: 34, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:54:11,163 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166054.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 11:54:12,918 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.3403, 3.8259, 4.0270, 4.1953, 4.1212, 3.8235, 4.4157, 1.8906], + device='cuda:1'), covar=tensor([0.0945, 0.0997, 0.1036, 0.1357, 0.1309, 0.1769, 0.0832, 0.5284], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0248, 0.0287, 0.0297, 0.0338, 0.0288, 0.0305, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:54:23,590 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.094e+01 1.435e+02 1.719e+02 2.053e+02 3.744e+02, threshold=3.437e+02, percent-clipped=0.0 +2023-03-27 11:54:23,606 INFO [finetune.py:976] (1/7) Epoch 29, batch 5700, loss[loss=0.1109, simple_loss=0.1737, pruned_loss=0.02403, over 4429.00 frames. ], tot_loss[loss=0.1669, simple_loss=0.2395, pruned_loss=0.04718, over 935217.38 frames. ], batch size: 19, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:54:32,085 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.5484, 2.4106, 2.2415, 2.6085, 2.3086, 2.3551, 2.3195, 3.0050], + device='cuda:1'), covar=tensor([0.3043, 0.3343, 0.2726, 0.2638, 0.3083, 0.2143, 0.3046, 0.1476], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0264, 0.0239, 0.0274, 0.0261, 0.0232, 0.0259, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:54:50,333 INFO [finetune.py:976] (1/7) Epoch 30, batch 0, loss[loss=0.2251, simple_loss=0.281, pruned_loss=0.08455, over 4752.00 frames. ], tot_loss[loss=0.2251, simple_loss=0.281, pruned_loss=0.08455, over 4752.00 frames. ], batch size: 54, lr: 2.82e-03, grad_scale: 32.0 +2023-03-27 11:54:50,333 INFO [finetune.py:1001] (1/7) Computing validation loss +2023-03-27 11:54:52,266 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9391, 1.1723, 2.0452, 1.9832, 1.8293, 1.7481, 1.8299, 2.0233], + device='cuda:1'), covar=tensor([0.4184, 0.4214, 0.3449, 0.3991, 0.5236, 0.3766, 0.4657, 0.2954], + device='cuda:1'), in_proj_covar=tensor([0.0270, 0.0250, 0.0269, 0.0301, 0.0300, 0.0277, 0.0306, 0.0255], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:54:57,656 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1289, 1.8406, 1.7353, 1.7496, 1.8136, 1.8343, 1.8490, 2.5226], + device='cuda:1'), covar=tensor([0.3871, 0.4584, 0.3452, 0.3663, 0.4514, 0.2570, 0.3922, 0.1717], + device='cuda:1'), in_proj_covar=tensor([0.0290, 0.0264, 0.0239, 0.0274, 0.0261, 0.0232, 0.0259, 0.0239], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:55:06,737 INFO [finetune.py:1010] (1/7) Epoch 30, validation: loss=0.1598, simple_loss=0.2264, pruned_loss=0.04658, over 2265189.00 frames. +2023-03-27 11:55:06,738 INFO [finetune.py:1011] (1/7) Maximum memory allocated so far is 6364MB +2023-03-27 11:55:11,872 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166106.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:55:21,137 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166115.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 11:55:37,391 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-03-27 11:55:43,743 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166147.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:55:52,370 INFO [finetune.py:976] (1/7) Epoch 30, batch 50, loss[loss=0.1805, simple_loss=0.253, pruned_loss=0.05405, over 4852.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2424, pruned_loss=0.04785, over 215061.64 frames. ], batch size: 49, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 11:56:02,559 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166159.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:56:16,047 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.060e+01 1.403e+02 1.686e+02 1.983e+02 3.736e+02, threshold=3.372e+02, percent-clipped=1.0 +2023-03-27 11:56:19,229 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-27 11:56:34,673 INFO [finetune.py:976] (1/7) Epoch 30, batch 100, loss[loss=0.2174, simple_loss=0.2778, pruned_loss=0.07855, over 4872.00 frames. ], tot_loss[loss=0.1636, simple_loss=0.2367, pruned_loss=0.04527, over 380526.62 frames. ], batch size: 34, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 11:56:38,187 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166207.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:56:39,330 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166208.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:56:49,688 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-27 11:56:56,085 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.87 vs. limit=5.0 +2023-03-27 11:57:07,203 INFO [finetune.py:976] (1/7) Epoch 30, batch 150, loss[loss=0.1332, simple_loss=0.2116, pruned_loss=0.02739, over 4768.00 frames. ], tot_loss[loss=0.1625, simple_loss=0.2339, pruned_loss=0.04549, over 507341.48 frames. ], batch size: 28, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 11:57:08,471 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=166255.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:57:12,006 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.50 vs. limit=2.0 +2023-03-27 11:57:21,826 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.042e+02 1.442e+02 1.804e+02 2.095e+02 4.016e+02, threshold=3.609e+02, percent-clipped=1.0 +2023-03-27 11:57:23,180 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0233, 1.6984, 2.1902, 1.4971, 1.9513, 2.1184, 1.6406, 2.2310], + device='cuda:1'), covar=tensor([0.1051, 0.1879, 0.1291, 0.1731, 0.0866, 0.1333, 0.2728, 0.0753], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0206, 0.0194, 0.0189, 0.0174, 0.0213, 0.0219, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:57:39,804 INFO [finetune.py:976] (1/7) Epoch 30, batch 200, loss[loss=0.1477, simple_loss=0.2238, pruned_loss=0.03579, over 4791.00 frames. ], tot_loss[loss=0.1633, simple_loss=0.2344, pruned_loss=0.04607, over 605556.50 frames. ], batch size: 51, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 11:57:42,331 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166307.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:57:48,715 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.24 vs. limit=5.0 +2023-03-27 11:58:14,802 INFO [finetune.py:976] (1/7) Epoch 30, batch 250, loss[loss=0.1617, simple_loss=0.2303, pruned_loss=0.04658, over 4773.00 frames. ], tot_loss[loss=0.1662, simple_loss=0.2379, pruned_loss=0.04724, over 682906.74 frames. ], batch size: 27, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 11:58:17,956 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5037, 1.4115, 1.3784, 1.4190, 0.9780, 3.0021, 1.0970, 1.4280], + device='cuda:1'), covar=tensor([0.3410, 0.2652, 0.2294, 0.2627, 0.1912, 0.0264, 0.2830, 0.1393], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0117, 0.0121, 0.0125, 0.0114, 0.0096, 0.0094, 0.0095], + device='cuda:1'), out_proj_covar=tensor([0.0006, 0.0006, 0.0005, 0.0006, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:1') +2023-03-27 11:58:26,043 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166368.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:58:26,771 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-03-27 11:58:30,118 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.080e+02 1.508e+02 1.830e+02 2.341e+02 4.348e+02, threshold=3.661e+02, percent-clipped=2.0 +2023-03-27 11:58:39,829 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.0843, 1.0433, 1.0140, 0.4436, 0.8955, 1.1549, 1.1823, 1.0098], + device='cuda:1'), covar=tensor([0.0806, 0.0482, 0.0539, 0.0525, 0.0554, 0.0521, 0.0355, 0.0612], + device='cuda:1'), in_proj_covar=tensor([0.0121, 0.0147, 0.0131, 0.0121, 0.0131, 0.0130, 0.0141, 0.0151], + device='cuda:1'), out_proj_covar=tensor([8.7934e-05, 1.0476e-04, 9.2951e-05, 8.4790e-05, 9.1881e-05, 9.2105e-05, + 1.0031e-04, 1.0788e-04], device='cuda:1') +2023-03-27 11:58:46,435 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8426, 1.6918, 1.4677, 1.9660, 2.3344, 1.9785, 1.6531, 1.4764], + device='cuda:1'), covar=tensor([0.2193, 0.2029, 0.2047, 0.1664, 0.1641, 0.1230, 0.2312, 0.1948], + device='cuda:1'), in_proj_covar=tensor([0.0251, 0.0216, 0.0220, 0.0202, 0.0250, 0.0194, 0.0221, 0.0210], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 11:58:48,138 INFO [finetune.py:976] (1/7) Epoch 30, batch 300, loss[loss=0.1365, simple_loss=0.215, pruned_loss=0.02906, over 4914.00 frames. ], tot_loss[loss=0.1696, simple_loss=0.2429, pruned_loss=0.0481, over 745300.87 frames. ], batch size: 36, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 11:58:50,087 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166406.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:58:52,503 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166410.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 11:59:21,479 INFO [finetune.py:976] (1/7) Epoch 30, batch 350, loss[loss=0.2093, simple_loss=0.2812, pruned_loss=0.06864, over 4920.00 frames. ], tot_loss[loss=0.1729, simple_loss=0.2463, pruned_loss=0.04974, over 793092.26 frames. ], batch size: 42, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 11:59:22,660 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=166454.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:59:25,723 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166459.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:59:37,684 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.476e+02 1.813e+02 2.161e+02 3.890e+02, threshold=3.626e+02, percent-clipped=2.0 +2023-03-27 11:59:41,436 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166481.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:59:55,115 INFO [finetune.py:976] (1/7) Epoch 30, batch 400, loss[loss=0.1771, simple_loss=0.2529, pruned_loss=0.05064, over 4862.00 frames. ], tot_loss[loss=0.1723, simple_loss=0.2464, pruned_loss=0.04907, over 827938.57 frames. ], batch size: 34, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 11:59:55,183 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166503.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:59:58,012 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=166507.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:59:59,266 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166509.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 11:59:59,343 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-27 12:00:20,212 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166531.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:00:35,348 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166542.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:00:45,991 INFO [finetune.py:976] (1/7) Epoch 30, batch 450, loss[loss=0.1816, simple_loss=0.2666, pruned_loss=0.04828, over 4887.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2445, pruned_loss=0.04855, over 855634.02 frames. ], batch size: 35, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:00:56,911 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166570.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:01:00,792 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.283e+01 1.415e+02 1.694e+02 2.083e+02 4.695e+02, threshold=3.388e+02, percent-clipped=2.0 +2023-03-27 12:01:03,265 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2267, 2.0587, 1.7229, 1.9211, 2.1502, 1.9085, 2.3485, 2.1699], + device='cuda:1'), covar=tensor([0.1283, 0.1962, 0.3091, 0.2354, 0.2636, 0.1756, 0.2588, 0.1679], + device='cuda:1'), in_proj_covar=tensor([0.0191, 0.0191, 0.0238, 0.0254, 0.0251, 0.0210, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:01:22,336 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166592.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:01:32,864 INFO [finetune.py:976] (1/7) Epoch 30, batch 500, loss[loss=0.1614, simple_loss=0.2342, pruned_loss=0.04425, over 4873.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2409, pruned_loss=0.04757, over 876884.78 frames. ], batch size: 34, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:01:46,181 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([4.1660, 3.6592, 3.8226, 4.0394, 3.9576, 3.7075, 4.2582, 1.2874], + device='cuda:1'), covar=tensor([0.0845, 0.0919, 0.1041, 0.0992, 0.1245, 0.1676, 0.0798, 0.5997], + device='cuda:1'), in_proj_covar=tensor([0.0355, 0.0249, 0.0288, 0.0298, 0.0339, 0.0288, 0.0306, 0.0304], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:02:06,082 INFO [finetune.py:976] (1/7) Epoch 30, batch 550, loss[loss=0.1589, simple_loss=0.2333, pruned_loss=0.04223, over 4730.00 frames. ], tot_loss[loss=0.167, simple_loss=0.2388, pruned_loss=0.04764, over 893967.94 frames. ], batch size: 23, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:02:12,185 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166663.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:02:20,414 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 7.871e+01 1.519e+02 1.786e+02 2.255e+02 3.834e+02, threshold=3.573e+02, percent-clipped=3.0 +2023-03-27 12:02:39,362 INFO [finetune.py:976] (1/7) Epoch 30, batch 600, loss[loss=0.1444, simple_loss=0.2158, pruned_loss=0.03648, over 4797.00 frames. ], tot_loss[loss=0.1667, simple_loss=0.2385, pruned_loss=0.04749, over 908817.32 frames. ], batch size: 26, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:02:43,710 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166710.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 12:02:45,057 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-03-27 12:03:00,925 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166735.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:03:12,732 INFO [finetune.py:976] (1/7) Epoch 30, batch 650, loss[loss=0.1614, simple_loss=0.2398, pruned_loss=0.04148, over 4874.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2406, pruned_loss=0.04772, over 918747.52 frames. ], batch size: 34, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:03:15,788 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=166758.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 12:03:19,468 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166764.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:03:21,892 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5814, 0.7761, 1.6609, 1.6272, 1.4949, 1.4072, 1.5639, 1.6393], + device='cuda:1'), covar=tensor([0.3576, 0.3482, 0.2965, 0.3122, 0.4013, 0.3404, 0.3521, 0.2659], + device='cuda:1'), in_proj_covar=tensor([0.0271, 0.0251, 0.0270, 0.0301, 0.0301, 0.0279, 0.0306, 0.0255], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:03:26,540 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.144e+02 1.552e+02 1.821e+02 2.218e+02 3.611e+02, threshold=3.642e+02, percent-clipped=1.0 +2023-03-27 12:03:41,558 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166796.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:03:45,812 INFO [finetune.py:976] (1/7) Epoch 30, batch 700, loss[loss=0.1755, simple_loss=0.247, pruned_loss=0.05205, over 4931.00 frames. ], tot_loss[loss=0.1706, simple_loss=0.2435, pruned_loss=0.04881, over 927105.57 frames. ], batch size: 41, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:03:45,923 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166803.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:04:00,347 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166825.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:04:08,140 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166837.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:04:18,069 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=166851.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:04:19,223 INFO [finetune.py:976] (1/7) Epoch 30, batch 750, loss[loss=0.1708, simple_loss=0.2524, pruned_loss=0.0446, over 4844.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2434, pruned_loss=0.0486, over 932683.37 frames. ], batch size: 47, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:04:27,147 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166865.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:04:27,837 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.9958, 2.4059, 2.3656, 1.1683, 2.6045, 2.0478, 1.7947, 2.2972], + device='cuda:1'), covar=tensor([0.1051, 0.1166, 0.2311, 0.2640, 0.1821, 0.2473, 0.2592, 0.1490], + device='cuda:1'), in_proj_covar=tensor([0.0171, 0.0189, 0.0202, 0.0181, 0.0209, 0.0210, 0.0224, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:04:33,234 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.605e+01 1.409e+02 1.752e+02 2.054e+02 3.389e+02, threshold=3.504e+02, percent-clipped=0.0 +2023-03-27 12:04:41,696 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166887.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:04:52,671 INFO [finetune.py:976] (1/7) Epoch 30, batch 800, loss[loss=0.1864, simple_loss=0.2551, pruned_loss=0.05888, over 4820.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2422, pruned_loss=0.04776, over 938156.03 frames. ], batch size: 33, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:05:09,488 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166929.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:05:32,102 INFO [finetune.py:976] (1/7) Epoch 30, batch 850, loss[loss=0.1411, simple_loss=0.2194, pruned_loss=0.03142, over 4928.00 frames. ], tot_loss[loss=0.1684, simple_loss=0.2413, pruned_loss=0.04772, over 942751.69 frames. ], batch size: 33, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:05:42,348 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166963.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:05:53,845 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.050e+02 1.507e+02 1.769e+02 2.105e+02 4.355e+02, threshold=3.539e+02, percent-clipped=2.0 +2023-03-27 12:06:03,571 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0 +2023-03-27 12:06:07,526 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166990.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:06:11,173 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-27 12:06:16,010 INFO [finetune.py:976] (1/7) Epoch 30, batch 900, loss[loss=0.1735, simple_loss=0.2533, pruned_loss=0.0469, over 4914.00 frames. ], tot_loss[loss=0.1667, simple_loss=0.2392, pruned_loss=0.04708, over 946625.58 frames. ], batch size: 43, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:06:23,185 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=167011.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:06:52,329 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-27 12:06:59,310 INFO [finetune.py:976] (1/7) Epoch 30, batch 950, loss[loss=0.1379, simple_loss=0.2131, pruned_loss=0.03135, over 4784.00 frames. ], tot_loss[loss=0.1664, simple_loss=0.2386, pruned_loss=0.04709, over 948436.60 frames. ], batch size: 28, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:07:13,662 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.122e+02 1.522e+02 1.848e+02 2.259e+02 3.601e+02, threshold=3.696e+02, percent-clipped=1.0 +2023-03-27 12:07:18,092 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.05 vs. limit=5.0 +2023-03-27 12:07:24,435 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167091.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:07:31,384 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167100.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:07:33,102 INFO [finetune.py:976] (1/7) Epoch 30, batch 1000, loss[loss=0.2324, simple_loss=0.2952, pruned_loss=0.08479, over 4750.00 frames. ], tot_loss[loss=0.1681, simple_loss=0.2405, pruned_loss=0.04785, over 948675.45 frames. ], batch size: 54, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:07:33,823 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167104.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:07:44,460 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167120.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:07:52,609 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.15 vs. limit=5.0 +2023-03-27 12:07:55,374 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167137.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:08:01,846 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-27 12:08:06,884 INFO [finetune.py:976] (1/7) Epoch 30, batch 1050, loss[loss=0.1951, simple_loss=0.2602, pruned_loss=0.06496, over 4818.00 frames. ], tot_loss[loss=0.1694, simple_loss=0.2425, pruned_loss=0.04817, over 949305.89 frames. ], batch size: 30, lr: 2.81e-03, grad_scale: 64.0 +2023-03-27 12:08:09,416 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8467, 1.6430, 2.2382, 3.2069, 2.1521, 2.4224, 1.3334, 2.6702], + device='cuda:1'), covar=tensor([0.1435, 0.1190, 0.1084, 0.0503, 0.0755, 0.1984, 0.1513, 0.0461], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0115, 0.0133, 0.0164, 0.0100, 0.0135, 0.0125, 0.0101], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 12:08:11,863 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167161.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:08:14,776 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167165.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:08:14,804 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167165.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:08:21,250 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.094e+02 1.490e+02 1.818e+02 2.230e+02 3.401e+02, threshold=3.636e+02, percent-clipped=0.0 +2023-03-27 12:08:27,349 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=167185.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:08:28,585 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167187.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:08:39,896 INFO [finetune.py:976] (1/7) Epoch 30, batch 1100, loss[loss=0.1905, simple_loss=0.2595, pruned_loss=0.0608, over 4911.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.244, pruned_loss=0.04875, over 950894.36 frames. ], batch size: 36, lr: 2.81e-03, grad_scale: 64.0 +2023-03-27 12:08:45,818 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.1503, 1.2556, 1.3940, 1.3398, 1.3353, 2.4066, 1.1194, 1.3170], + device='cuda:1'), covar=tensor([0.1094, 0.2076, 0.1316, 0.1038, 0.1877, 0.0452, 0.1820, 0.2139], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0083, 0.0073, 0.0076, 0.0092, 0.0081, 0.0086, 0.0081], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 12:08:46,972 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=167213.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:09:01,418 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=167235.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:09:13,765 INFO [finetune.py:976] (1/7) Epoch 30, batch 1150, loss[loss=0.1665, simple_loss=0.2468, pruned_loss=0.04311, over 4830.00 frames. ], tot_loss[loss=0.1703, simple_loss=0.2442, pruned_loss=0.04818, over 952439.29 frames. ], batch size: 47, lr: 2.81e-03, grad_scale: 64.0 +2023-03-27 12:09:28,392 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.072e+02 1.511e+02 1.772e+02 2.131e+02 4.281e+02, threshold=3.544e+02, percent-clipped=3.0 +2023-03-27 12:09:34,971 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167285.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:09:47,289 INFO [finetune.py:976] (1/7) Epoch 30, batch 1200, loss[loss=0.1343, simple_loss=0.2252, pruned_loss=0.0217, over 4792.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2432, pruned_loss=0.04825, over 951882.36 frames. ], batch size: 29, lr: 2.81e-03, grad_scale: 64.0 +2023-03-27 12:10:20,449 INFO [finetune.py:976] (1/7) Epoch 30, batch 1250, loss[loss=0.1908, simple_loss=0.2446, pruned_loss=0.06855, over 4932.00 frames. ], tot_loss[loss=0.1675, simple_loss=0.24, pruned_loss=0.04752, over 953023.49 frames. ], batch size: 33, lr: 2.81e-03, grad_scale: 64.0 +2023-03-27 12:10:36,986 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.068e+02 1.462e+02 1.711e+02 2.091e+02 4.240e+02, threshold=3.422e+02, percent-clipped=1.0 +2023-03-27 12:10:47,778 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=3.75 vs. limit=5.0 +2023-03-27 12:10:56,488 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167391.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:11:06,190 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5762, 1.1567, 0.7817, 1.4083, 1.9901, 0.8289, 1.3665, 1.4458], + device='cuda:1'), covar=tensor([0.1437, 0.2064, 0.1687, 0.1210, 0.1836, 0.1885, 0.1457, 0.1939], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0109, 0.0093, 0.0120, 0.0092, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 12:11:09,175 INFO [finetune.py:976] (1/7) Epoch 30, batch 1300, loss[loss=0.1879, simple_loss=0.251, pruned_loss=0.06238, over 4796.00 frames. ], tot_loss[loss=0.166, simple_loss=0.2381, pruned_loss=0.04697, over 954733.31 frames. ], batch size: 45, lr: 2.81e-03, grad_scale: 64.0 +2023-03-27 12:11:24,565 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167420.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:11:32,403 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167432.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:11:38,998 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=167439.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:11:47,729 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([0.9847, 0.9460, 0.9234, 1.0600, 1.1584, 1.0865, 0.9860, 0.9365], + device='cuda:1'), covar=tensor([0.0408, 0.0322, 0.0736, 0.0328, 0.0284, 0.0530, 0.0355, 0.0451], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0107, 0.0149, 0.0112, 0.0103, 0.0118, 0.0104, 0.0116], + device='cuda:1'), out_proj_covar=tensor([7.9990e-05, 8.1833e-05, 1.1603e-04, 8.4729e-05, 7.9501e-05, 8.7181e-05, + 7.7443e-05, 8.7955e-05], device='cuda:1') +2023-03-27 12:11:55,875 INFO [finetune.py:976] (1/7) Epoch 30, batch 1350, loss[loss=0.231, simple_loss=0.2918, pruned_loss=0.08508, over 4832.00 frames. ], tot_loss[loss=0.1648, simple_loss=0.237, pruned_loss=0.04629, over 954624.50 frames. ], batch size: 47, lr: 2.81e-03, grad_scale: 64.0 +2023-03-27 12:11:58,247 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167456.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:12:00,697 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167460.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:12:07,067 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=167468.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:12:11,228 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.213e+01 1.429e+02 1.665e+02 1.960e+02 3.889e+02, threshold=3.329e+02, percent-clipped=2.0 +2023-03-27 12:12:23,210 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167493.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:12:29,688 INFO [finetune.py:976] (1/7) Epoch 30, batch 1400, loss[loss=0.191, simple_loss=0.2668, pruned_loss=0.05757, over 4807.00 frames. ], tot_loss[loss=0.1673, simple_loss=0.2402, pruned_loss=0.04715, over 954480.20 frames. ], batch size: 45, lr: 2.81e-03, grad_scale: 64.0 +2023-03-27 12:13:02,946 INFO [finetune.py:976] (1/7) Epoch 30, batch 1450, loss[loss=0.1554, simple_loss=0.2282, pruned_loss=0.04133, over 4783.00 frames. ], tot_loss[loss=0.1671, simple_loss=0.2407, pruned_loss=0.04674, over 956595.86 frames. ], batch size: 51, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:13:12,954 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([3.0617, 2.6496, 2.5892, 1.2897, 2.6976, 2.1831, 2.1330, 2.5142], + device='cuda:1'), covar=tensor([0.0911, 0.0825, 0.1766, 0.2192, 0.1355, 0.2252, 0.2056, 0.1188], + device='cuda:1'), in_proj_covar=tensor([0.0172, 0.0190, 0.0203, 0.0182, 0.0210, 0.0212, 0.0225, 0.0197], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:13:19,227 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.032e+02 1.535e+02 1.844e+02 2.174e+02 4.375e+02, threshold=3.689e+02, percent-clipped=4.0 +2023-03-27 12:13:25,325 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167585.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:13:36,787 INFO [finetune.py:976] (1/7) Epoch 30, batch 1500, loss[loss=0.1263, simple_loss=0.1944, pruned_loss=0.02911, over 4783.00 frames. ], tot_loss[loss=0.1686, simple_loss=0.2426, pruned_loss=0.04733, over 957165.74 frames. ], batch size: 25, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:13:52,729 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-03-27 12:13:57,372 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=167633.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:14:10,430 INFO [finetune.py:976] (1/7) Epoch 30, batch 1550, loss[loss=0.1598, simple_loss=0.2302, pruned_loss=0.04463, over 4816.00 frames. ], tot_loss[loss=0.1674, simple_loss=0.2418, pruned_loss=0.04649, over 957666.36 frames. ], batch size: 40, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:14:26,674 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.931e+01 1.416e+02 1.751e+02 2.105e+02 4.024e+02, threshold=3.503e+02, percent-clipped=1.0 +2023-03-27 12:14:44,013 INFO [finetune.py:976] (1/7) Epoch 30, batch 1600, loss[loss=0.1378, simple_loss=0.2059, pruned_loss=0.03489, over 4770.00 frames. ], tot_loss[loss=0.166, simple_loss=0.2394, pruned_loss=0.04628, over 956689.76 frames. ], batch size: 28, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:14:54,741 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167719.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:15:17,590 INFO [finetune.py:976] (1/7) Epoch 30, batch 1650, loss[loss=0.132, simple_loss=0.2001, pruned_loss=0.03197, over 4740.00 frames. ], tot_loss[loss=0.1641, simple_loss=0.2372, pruned_loss=0.0455, over 959022.61 frames. ], batch size: 54, lr: 2.81e-03, grad_scale: 32.0 +2023-03-27 12:15:19,515 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167756.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:15:21,892 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167760.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:15:26,699 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167767.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:15:32,527 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.722e+01 1.428e+02 1.633e+02 1.926e+02 4.440e+02, threshold=3.266e+02, percent-clipped=1.0 +2023-03-27 12:15:36,067 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167780.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:15:39,493 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0593, 1.8553, 1.9740, 1.3838, 1.9658, 2.0685, 2.0533, 1.5849], + device='cuda:1'), covar=tensor([0.0536, 0.0714, 0.0702, 0.0876, 0.0774, 0.0642, 0.0559, 0.1241], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0138, 0.0141, 0.0119, 0.0129, 0.0139, 0.0139, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:15:41,289 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167788.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:15:55,582 INFO [finetune.py:976] (1/7) Epoch 30, batch 1700, loss[loss=0.1386, simple_loss=0.1987, pruned_loss=0.0393, over 4090.00 frames. ], tot_loss[loss=0.1617, simple_loss=0.2345, pruned_loss=0.04444, over 958387.71 frames. ], batch size: 17, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:15:56,725 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=167804.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:16:03,659 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=167808.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:16:25,152 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167828.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:16:42,118 INFO [finetune.py:976] (1/7) Epoch 30, batch 1750, loss[loss=0.1859, simple_loss=0.2607, pruned_loss=0.05552, over 4892.00 frames. ], tot_loss[loss=0.1636, simple_loss=0.2369, pruned_loss=0.0452, over 957528.72 frames. ], batch size: 35, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:16:47,661 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9015, 1.7678, 1.8590, 1.2383, 1.8305, 1.9228, 1.8983, 1.5625], + device='cuda:1'), covar=tensor([0.0561, 0.0657, 0.0658, 0.0820, 0.0810, 0.0642, 0.0558, 0.1175], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0137, 0.0141, 0.0118, 0.0128, 0.0138, 0.0139, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:16:54,726 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.8999, 1.5707, 2.3294, 3.6370, 2.4546, 2.6039, 1.3862, 3.0367], + device='cuda:1'), covar=tensor([0.1614, 0.1475, 0.1323, 0.0582, 0.0784, 0.1497, 0.1727, 0.0440], + device='cuda:1'), in_proj_covar=tensor([0.0100, 0.0116, 0.0133, 0.0164, 0.0100, 0.0136, 0.0125, 0.0102], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0003], + device='cuda:1') +2023-03-27 12:16:56,580 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5400, 1.5881, 1.2886, 1.5412, 1.8856, 1.8266, 1.5842, 1.3494], + device='cuda:1'), covar=tensor([0.0363, 0.0318, 0.0707, 0.0340, 0.0215, 0.0436, 0.0330, 0.0454], + device='cuda:1'), in_proj_covar=tensor([0.0104, 0.0107, 0.0150, 0.0112, 0.0103, 0.0119, 0.0105, 0.0116], + device='cuda:1'), out_proj_covar=tensor([7.9997e-05, 8.2062e-05, 1.1646e-04, 8.5244e-05, 7.9962e-05, 8.7409e-05, + 7.8003e-05, 8.8186e-05], device='cuda:1') +2023-03-27 12:16:57,034 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.117e+02 1.486e+02 1.829e+02 2.140e+02 4.770e+02, threshold=3.658e+02, percent-clipped=2.0 +2023-03-27 12:17:20,224 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1605, 1.4161, 0.9007, 1.9913, 2.3456, 1.8153, 1.7804, 1.8504], + device='cuda:1'), covar=tensor([0.1356, 0.2068, 0.1971, 0.1175, 0.1873, 0.1880, 0.1401, 0.1982], + device='cuda:1'), in_proj_covar=tensor([0.0090, 0.0094, 0.0109, 0.0094, 0.0120, 0.0093, 0.0098, 0.0089], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 12:17:25,520 INFO [finetune.py:976] (1/7) Epoch 30, batch 1800, loss[loss=0.1326, simple_loss=0.2158, pruned_loss=0.0247, over 4780.00 frames. ], tot_loss[loss=0.1667, simple_loss=0.2408, pruned_loss=0.04624, over 958123.47 frames. ], batch size: 29, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:17:58,716 INFO [finetune.py:976] (1/7) Epoch 30, batch 1850, loss[loss=0.195, simple_loss=0.2696, pruned_loss=0.06022, over 4819.00 frames. ], tot_loss[loss=0.1687, simple_loss=0.243, pruned_loss=0.04715, over 957772.62 frames. ], batch size: 39, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:18:13,649 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.002e+02 1.458e+02 1.787e+02 2.144e+02 3.700e+02, threshold=3.573e+02, percent-clipped=1.0 +2023-03-27 12:18:33,562 INFO [finetune.py:976] (1/7) Epoch 30, batch 1900, loss[loss=0.1874, simple_loss=0.2704, pruned_loss=0.05217, over 4820.00 frames. ], tot_loss[loss=0.1688, simple_loss=0.2437, pruned_loss=0.04693, over 957182.35 frames. ], batch size: 33, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:18:45,500 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.8529, 4.7340, 4.4267, 2.3638, 4.8007, 3.6928, 0.9849, 3.3555], + device='cuda:1'), covar=tensor([0.2186, 0.1678, 0.1426, 0.2916, 0.0769, 0.0761, 0.4183, 0.1248], + device='cuda:1'), in_proj_covar=tensor([0.0153, 0.0181, 0.0161, 0.0131, 0.0164, 0.0125, 0.0149, 0.0127], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:1') +2023-03-27 12:18:55,007 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=168035.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:18:55,592 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4401, 1.1005, 0.7304, 1.3209, 1.9030, 0.8068, 1.3172, 1.3942], + device='cuda:1'), covar=tensor([0.1571, 0.2138, 0.1780, 0.1285, 0.2009, 0.1897, 0.1511, 0.1931], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0095, 0.0110, 0.0094, 0.0121, 0.0093, 0.0099, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 12:18:56,840 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.6776, 1.1707, 0.9383, 1.5357, 2.1268, 1.2283, 1.5943, 1.5063], + device='cuda:1'), covar=tensor([0.1547, 0.2212, 0.1795, 0.1340, 0.2018, 0.1871, 0.1436, 0.2171], + device='cuda:1'), in_proj_covar=tensor([0.0091, 0.0095, 0.0110, 0.0094, 0.0121, 0.0093, 0.0099, 0.0090], + device='cuda:1'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:1') +2023-03-27 12:19:07,329 INFO [finetune.py:976] (1/7) Epoch 30, batch 1950, loss[loss=0.1588, simple_loss=0.2265, pruned_loss=0.04558, over 4889.00 frames. ], tot_loss[loss=0.1679, simple_loss=0.2425, pruned_loss=0.04667, over 955561.39 frames. ], batch size: 43, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:19:21,544 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=168075.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:19:22,071 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.065e+02 1.458e+02 1.758e+02 2.118e+02 3.555e+02, threshold=3.516e+02, percent-clipped=0.0 +2023-03-27 12:19:30,464 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=168088.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:19:35,803 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=168096.0, num_to_drop=1, layers_to_drop={2} +2023-03-27 12:19:40,898 INFO [finetune.py:976] (1/7) Epoch 30, batch 2000, loss[loss=0.1509, simple_loss=0.2183, pruned_loss=0.04174, over 4392.00 frames. ], tot_loss[loss=0.1657, simple_loss=0.2397, pruned_loss=0.0459, over 956128.49 frames. ], batch size: 19, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:19:54,027 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=168123.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:20:01,892 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=168136.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:20:14,572 INFO [finetune.py:976] (1/7) Epoch 30, batch 2050, loss[loss=0.1518, simple_loss=0.2194, pruned_loss=0.04211, over 4324.00 frames. ], tot_loss[loss=0.1634, simple_loss=0.2365, pruned_loss=0.04519, over 954487.78 frames. ], batch size: 65, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:20:24,871 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2146, 1.8945, 2.5348, 1.6896, 2.2419, 2.4201, 1.6577, 2.5636], + device='cuda:1'), covar=tensor([0.1423, 0.2100, 0.1459, 0.2034, 0.0995, 0.1546, 0.3005, 0.0855], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0207, 0.0193, 0.0189, 0.0174, 0.0212, 0.0219, 0.0198], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:20:28,418 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.9259, 1.8302, 1.5534, 1.9231, 2.5052, 2.0178, 1.7947, 1.5071], + device='cuda:1'), covar=tensor([0.1959, 0.1808, 0.1816, 0.1574, 0.1466, 0.1129, 0.2146, 0.1783], + device='cuda:1'), in_proj_covar=tensor([0.0251, 0.0215, 0.0219, 0.0202, 0.0249, 0.0195, 0.0222, 0.0209], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:20:29,509 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.016e+02 1.484e+02 1.729e+02 2.115e+02 4.273e+02, threshold=3.459e+02, percent-clipped=3.0 +2023-03-27 12:20:29,645 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1092, 1.8496, 2.0679, 1.3843, 2.0802, 2.0532, 2.0779, 1.6351], + device='cuda:1'), covar=tensor([0.0539, 0.0677, 0.0545, 0.0811, 0.0726, 0.0607, 0.0556, 0.1244], + device='cuda:1'), in_proj_covar=tensor([0.0131, 0.0137, 0.0141, 0.0119, 0.0128, 0.0138, 0.0139, 0.0162], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:20:47,506 INFO [finetune.py:976] (1/7) Epoch 30, batch 2100, loss[loss=0.1324, simple_loss=0.2083, pruned_loss=0.02829, over 4762.00 frames. ], tot_loss[loss=0.1635, simple_loss=0.2363, pruned_loss=0.04534, over 954365.61 frames. ], batch size: 26, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:21:19,160 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=168233.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:21:31,921 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4769, 1.5629, 1.2573, 1.4881, 1.8763, 1.8249, 1.5375, 1.3873], + device='cuda:1'), covar=tensor([0.0411, 0.0333, 0.0766, 0.0328, 0.0236, 0.0485, 0.0392, 0.0450], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0107, 0.0149, 0.0111, 0.0102, 0.0118, 0.0105, 0.0115], + device='cuda:1'), out_proj_covar=tensor([7.9703e-05, 8.1330e-05, 1.1559e-04, 8.4698e-05, 7.9205e-05, 8.6514e-05, + 7.7565e-05, 8.7434e-05], device='cuda:1') +2023-03-27 12:21:41,940 INFO [finetune.py:976] (1/7) Epoch 30, batch 2150, loss[loss=0.1666, simple_loss=0.2323, pruned_loss=0.0504, over 4839.00 frames. ], tot_loss[loss=0.1656, simple_loss=0.2394, pruned_loss=0.04594, over 955286.37 frames. ], batch size: 30, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:22:00,990 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.149e+02 1.551e+02 1.892e+02 2.224e+02 4.404e+02, threshold=3.784e+02, percent-clipped=3.0 +2023-03-27 12:22:12,543 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=168294.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:22:18,788 INFO [finetune.py:976] (1/7) Epoch 30, batch 2200, loss[loss=0.1899, simple_loss=0.2574, pruned_loss=0.0612, over 4814.00 frames. ], tot_loss[loss=0.168, simple_loss=0.2421, pruned_loss=0.04697, over 953017.61 frames. ], batch size: 38, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:22:29,007 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.0510, 0.9759, 0.9187, 1.0866, 1.1950, 1.1745, 1.0049, 0.9583], + device='cuda:1'), covar=tensor([0.0460, 0.0352, 0.0741, 0.0366, 0.0321, 0.0449, 0.0381, 0.0462], + device='cuda:1'), in_proj_covar=tensor([0.0103, 0.0106, 0.0148, 0.0111, 0.0102, 0.0117, 0.0104, 0.0115], + device='cuda:1'), out_proj_covar=tensor([7.9634e-05, 8.1154e-05, 1.1539e-04, 8.4615e-05, 7.8958e-05, 8.6296e-05, + 7.7407e-05, 8.7262e-05], device='cuda:1') +2023-03-27 12:23:02,560 INFO [finetune.py:976] (1/7) Epoch 30, batch 2250, loss[loss=0.1615, simple_loss=0.2437, pruned_loss=0.03965, over 4876.00 frames. ], tot_loss[loss=0.1689, simple_loss=0.2431, pruned_loss=0.0474, over 953295.40 frames. ], batch size: 35, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:23:17,396 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=168375.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:23:17,906 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.029e+01 1.513e+02 1.826e+02 2.132e+02 3.584e+02, threshold=3.652e+02, percent-clipped=0.0 +2023-03-27 12:23:28,073 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=168391.0, num_to_drop=1, layers_to_drop={0} +2023-03-27 12:23:36,282 INFO [finetune.py:976] (1/7) Epoch 30, batch 2300, loss[loss=0.1604, simple_loss=0.2309, pruned_loss=0.04491, over 4838.00 frames. ], tot_loss[loss=0.1691, simple_loss=0.2436, pruned_loss=0.04726, over 954247.51 frames. ], batch size: 47, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:23:41,163 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.6368, 2.4089, 2.5465, 1.9020, 2.5008, 2.6113, 2.7585, 2.1922], + device='cuda:1'), covar=tensor([0.0531, 0.0632, 0.0579, 0.0769, 0.0827, 0.0615, 0.0437, 0.0969], + device='cuda:1'), in_proj_covar=tensor([0.0132, 0.0138, 0.0142, 0.0120, 0.0129, 0.0140, 0.0140, 0.0163], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:23:49,959 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=168423.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:23:49,982 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=168423.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:24:09,555 INFO [finetune.py:976] (1/7) Epoch 30, batch 2350, loss[loss=0.1303, simple_loss=0.2058, pruned_loss=0.02743, over 4831.00 frames. ], tot_loss[loss=0.1666, simple_loss=0.2403, pruned_loss=0.04643, over 952182.59 frames. ], batch size: 30, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:24:21,868 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=168471.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:24:24,770 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.258e+01 1.458e+02 1.699e+02 2.116e+02 4.301e+02, threshold=3.398e+02, percent-clipped=1.0 +2023-03-27 12:24:42,042 INFO [finetune.py:976] (1/7) Epoch 30, batch 2400, loss[loss=0.1271, simple_loss=0.1978, pruned_loss=0.02816, over 4730.00 frames. ], tot_loss[loss=0.1658, simple_loss=0.239, pruned_loss=0.04626, over 953433.56 frames. ], batch size: 23, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:25:15,074 INFO [finetune.py:976] (1/7) Epoch 30, batch 2450, loss[loss=0.2068, simple_loss=0.2731, pruned_loss=0.07028, over 4910.00 frames. ], tot_loss[loss=0.164, simple_loss=0.2364, pruned_loss=0.0458, over 953854.65 frames. ], batch size: 37, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:25:30,928 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.020e+02 1.487e+02 1.838e+02 2.245e+02 3.083e+02, threshold=3.676e+02, percent-clipped=0.0 +2023-03-27 12:25:39,363 INFO [zipformer.py:1188] (1/7) warmup_begin=1333.3, warmup_end=2000.0, batch_count=168589.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:25:48,868 INFO [finetune.py:976] (1/7) Epoch 30, batch 2500, loss[loss=0.1714, simple_loss=0.2419, pruned_loss=0.05043, over 4936.00 frames. ], tot_loss[loss=0.1663, simple_loss=0.2388, pruned_loss=0.04686, over 953220.79 frames. ], batch size: 33, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:25:50,180 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2422, 2.1784, 1.6756, 2.2493, 2.2346, 1.8953, 2.5353, 2.2532], + device='cuda:1'), covar=tensor([0.1221, 0.1691, 0.2829, 0.2364, 0.2272, 0.1593, 0.2704, 0.1551], + device='cuda:1'), in_proj_covar=tensor([0.0190, 0.0191, 0.0238, 0.0252, 0.0249, 0.0209, 0.0216, 0.0204], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:25:53,262 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-03-27 12:26:27,843 INFO [finetune.py:976] (1/7) Epoch 30, batch 2550, loss[loss=0.1899, simple_loss=0.2655, pruned_loss=0.05713, over 4885.00 frames. ], tot_loss[loss=0.1698, simple_loss=0.2431, pruned_loss=0.04824, over 954194.39 frames. ], batch size: 32, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:26:55,337 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.157e+02 1.538e+02 1.821e+02 2.163e+02 4.307e+02, threshold=3.641e+02, percent-clipped=2.0 +2023-03-27 12:27:09,676 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=168691.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:27:17,863 INFO [finetune.py:976] (1/7) Epoch 30, batch 2600, loss[loss=0.1744, simple_loss=0.2618, pruned_loss=0.04345, over 4791.00 frames. ], tot_loss[loss=0.1708, simple_loss=0.2442, pruned_loss=0.04873, over 953857.15 frames. ], batch size: 45, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:27:26,313 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([5.1814, 4.5930, 4.7255, 5.0610, 4.8421, 4.6080, 5.3661, 1.7653], + device='cuda:1'), covar=tensor([0.0882, 0.0868, 0.0840, 0.1034, 0.1454, 0.1966, 0.0592, 0.6285], + device='cuda:1'), in_proj_covar=tensor([0.0353, 0.0247, 0.0287, 0.0298, 0.0337, 0.0287, 0.0307, 0.0303], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:27:44,431 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=168739.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:28:01,792 INFO [finetune.py:976] (1/7) Epoch 30, batch 2650, loss[loss=0.1636, simple_loss=0.2173, pruned_loss=0.05499, over 4118.00 frames. ], tot_loss[loss=0.1711, simple_loss=0.2445, pruned_loss=0.0489, over 952288.73 frames. ], batch size: 17, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:28:17,251 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-03-27 12:28:21,696 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 1.069e+02 1.510e+02 1.724e+02 1.979e+02 3.263e+02, threshold=3.448e+02, percent-clipped=0.0 +2023-03-27 12:28:43,723 INFO [finetune.py:976] (1/7) Epoch 30, batch 2700, loss[loss=0.1379, simple_loss=0.2068, pruned_loss=0.0345, over 4817.00 frames. ], tot_loss[loss=0.1695, simple_loss=0.243, pruned_loss=0.04801, over 952852.91 frames. ], batch size: 39, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:29:09,817 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.1938, 1.7790, 2.1811, 2.1662, 1.9143, 1.9202, 2.1182, 2.0435], + device='cuda:1'), covar=tensor([0.4736, 0.4185, 0.3265, 0.4177, 0.4952, 0.4605, 0.4714, 0.3212], + device='cuda:1'), in_proj_covar=tensor([0.0271, 0.0252, 0.0271, 0.0301, 0.0302, 0.0281, 0.0308, 0.0257], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:29:12,416 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-03-27 12:29:17,014 INFO [finetune.py:976] (1/7) Epoch 30, batch 2750, loss[loss=0.1529, simple_loss=0.2266, pruned_loss=0.0396, over 4757.00 frames. ], tot_loss[loss=0.1672, simple_loss=0.24, pruned_loss=0.04719, over 953925.64 frames. ], batch size: 27, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:29:32,259 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.461e+01 1.436e+02 1.670e+02 1.989e+02 2.987e+02, threshold=3.340e+02, percent-clipped=0.0 +2023-03-27 12:29:41,551 INFO [zipformer.py:1188] (1/7) warmup_begin=2000.0, warmup_end=2666.7, batch_count=168889.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:29:47,128 INFO [scaling.py:679] (1/7) Whitening: num_groups=1, num_channels=384, metric=4.36 vs. limit=5.0 +2023-03-27 12:29:50,098 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-03-27 12:29:50,503 INFO [finetune.py:976] (1/7) Epoch 30, batch 2800, loss[loss=0.1206, simple_loss=0.1789, pruned_loss=0.03118, over 3880.00 frames. ], tot_loss[loss=0.1641, simple_loss=0.2362, pruned_loss=0.04601, over 951924.56 frames. ], batch size: 16, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:30:05,459 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.4178, 1.3990, 1.8081, 1.6104, 1.5959, 3.1627, 1.2964, 1.5380], + device='cuda:1'), covar=tensor([0.1035, 0.1879, 0.1177, 0.1022, 0.1591, 0.0294, 0.1609, 0.1879], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0082, 0.0073, 0.0076, 0.0091, 0.0081, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 12:30:13,011 INFO [zipformer.py:1188] (1/7) warmup_begin=666.7, warmup_end=1333.3, batch_count=168937.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:30:20,510 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.2172, 2.0378, 1.9056, 2.1682, 2.0310, 2.0237, 2.0024, 2.7704], + device='cuda:1'), covar=tensor([0.3609, 0.4488, 0.3297, 0.4070, 0.4335, 0.2788, 0.4175, 0.1746], + device='cuda:1'), in_proj_covar=tensor([0.0291, 0.0265, 0.0240, 0.0276, 0.0264, 0.0234, 0.0261, 0.0241], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:30:23,986 INFO [finetune.py:976] (1/7) Epoch 30, batch 2850, loss[loss=0.1839, simple_loss=0.2514, pruned_loss=0.05824, over 4821.00 frames. ], tot_loss[loss=0.1637, simple_loss=0.235, pruned_loss=0.04617, over 953572.92 frames. ], batch size: 30, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:30:33,496 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=168967.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:30:34,709 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=168969.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 12:30:38,878 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 8.606e+01 1.456e+02 1.728e+02 2.104e+02 5.266e+02, threshold=3.457e+02, percent-clipped=2.0 +2023-03-27 12:30:38,995 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.5367, 1.5661, 2.2218, 1.7947, 1.7684, 4.1522, 1.4631, 1.6369], + device='cuda:1'), covar=tensor([0.1012, 0.1862, 0.1266, 0.1035, 0.1648, 0.0201, 0.1572, 0.1910], + device='cuda:1'), in_proj_covar=tensor([0.0074, 0.0082, 0.0073, 0.0076, 0.0091, 0.0081, 0.0085, 0.0080], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 12:30:43,294 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-27 12:30:46,787 INFO [scaling.py:679] (1/7) Whitening: num_groups=8, num_channels=192, metric=1.64 vs. limit=2.0 +2023-03-27 12:30:57,835 INFO [finetune.py:976] (1/7) Epoch 30, batch 2900, loss[loss=0.1623, simple_loss=0.2477, pruned_loss=0.03844, over 4820.00 frames. ], tot_loss[loss=0.1675, simple_loss=0.2391, pruned_loss=0.04792, over 951648.20 frames. ], batch size: 39, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:30:59,829 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([2.0267, 1.9421, 1.6984, 1.9874, 1.9703, 1.7659, 2.1764, 2.0701], + device='cuda:1'), covar=tensor([0.1148, 0.1792, 0.2558, 0.2139, 0.2390, 0.1491, 0.2960, 0.1447], + device='cuda:1'), in_proj_covar=tensor([0.0193, 0.0194, 0.0242, 0.0256, 0.0253, 0.0212, 0.0219, 0.0207], + device='cuda:1'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:1') +2023-03-27 12:31:14,731 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=169028.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:31:15,299 INFO [zipformer.py:2441] (1/7) attn_weights_entropy = tensor([1.2398, 1.2891, 1.5443, 1.4183, 1.4339, 2.8585, 1.2129, 1.4035], + device='cuda:1'), covar=tensor([0.1113, 0.1907, 0.1295, 0.1047, 0.1709, 0.0294, 0.1617, 0.1902], + device='cuda:1'), in_proj_covar=tensor([0.0075, 0.0082, 0.0073, 0.0076, 0.0092, 0.0081, 0.0085, 0.0081], + device='cuda:1'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0005, 0.0005], + device='cuda:1') +2023-03-27 12:31:15,927 INFO [zipformer.py:1188] (1/7) warmup_begin=3333.3, warmup_end=4000.0, batch_count=169030.0, num_to_drop=1, layers_to_drop={1} +2023-03-27 12:31:31,770 INFO [finetune.py:976] (1/7) Epoch 30, batch 2950, loss[loss=0.1846, simple_loss=0.257, pruned_loss=0.05608, over 4750.00 frames. ], tot_loss[loss=0.1704, simple_loss=0.2427, pruned_loss=0.04905, over 951802.00 frames. ], batch size: 54, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:31:49,078 INFO [optim.py:369] (1/7) Clipping_scale=2.0, grad-norm quartiles 9.406e+01 1.615e+02 1.887e+02 2.255e+02 4.054e+02, threshold=3.773e+02, percent-clipped=1.0 +2023-03-27 12:31:53,308 INFO [zipformer.py:1188] (1/7) warmup_begin=2666.7, warmup_end=3333.3, batch_count=169082.0, num_to_drop=0, layers_to_drop=set() +2023-03-27 12:32:19,262 INFO [finetune.py:976] (1/7) Epoch 30, batch 3000, loss[loss=0.2712, simple_loss=0.3217, pruned_loss=0.1104, over 4112.00 frames. ], tot_loss[loss=0.1716, simple_loss=0.2444, pruned_loss=0.04942, over 951200.15 frames. ], batch size: 65, lr: 2.80e-03, grad_scale: 32.0 +2023-03-27 12:32:19,262 INFO [finetune.py:1001] (1/7) Computing validation loss