diff --git "a/log/log-train-2023-03-31-18-51-54-0" "b/log/log-train-2023-03-31-18-51-54-0" new file mode 100644--- /dev/null +++ "b/log/log-train-2023-03-31-18-51-54-0" @@ -0,0 +1,25561 @@ +2023-03-31 18:51:54,767 INFO [train.py:975] (0/4) Training started +2023-03-31 18:51:54,771 INFO [train.py:985] (0/4) Device: cuda:0 +2023-03-31 18:51:54,825 INFO [train.py:994] (0/4) {'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': '1.23.3', 'k2-build-type': 'Debug', 'k2-with-cuda': True, 'k2-git-sha1': '1c9950559223ec24d187f56bc424c3b43904bed3', 'k2-git-date': 'Thu Jan 26 22:00:26 2023', 'lhotse-version': '1.13.0.dev+git.ca98c73.dirty', 'torch-version': '2.0.0+cu117', 'torch-cuda-available': True, 'torch-cuda-version': '11.7', 'python-version': '3.8', 'icefall-git-branch': 'surt', 'icefall-git-sha1': '51e6a8a-dirty', 'icefall-git-date': 'Fri Mar 17 11:23:13 2023', 'icefall-path': '/exp/draj/mini_scale_2022/icefall', 'k2-path': '/exp/draj/mini_scale_2022/k2/k2/python/k2/__init__.py', 'lhotse-path': '/exp/draj/mini_scale_2022/lhotse/lhotse/__init__.py', 'hostname': 'r2n03', 'IP address': '10.1.2.3'}, 'world_size': 4, 'master_port': 54321, 'tensorboard': True, 'num_epochs': 30, 'start_epoch': 1, 'start_batch': 0, 'exp_dir': PosixPath('pruned_transducer_stateless7_streaming/exp/v2'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'base_lr': 0.05, 'lr_batches': 5000, 'lr_epochs': 3.5, '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': 10, 'average_period': 200, 'use_fp16': True, 'num_encoder_layers': '2,2,2,2,2', 'feedforward_dims': '768,768,768,768,768', 'nhead': '8,8,8,8,8', 'encoder_dims': '256,256,256,256,256', 'attention_dims': '192,192,192,192,192', 'encoder_unmasked_dims': '192,192,192,192,192', 'zipformer_downsampling_factors': '1,2,4,8,2', 'cnn_module_kernels': '31,31,31,31,31', 'decoder_dim': 512, 'joiner_dim': 512, 'short_chunk_size': 50, 'num_left_chunks': 4, 'decode_chunk_len': 32, 'full_libri': True, 'manifest_dir': PosixPath('data/manifests'), 'max_duration': 800, '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} +2023-03-31 18:51:54,826 INFO [train.py:996] (0/4) About to create model +2023-03-31 18:51:55,663 INFO [zipformer.py:405] (0/4) 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. +2023-03-31 18:51:55,687 INFO [train.py:1000] (0/4) Number of model parameters: 20697573 +2023-03-31 18:52:03,347 INFO [train.py:1019] (0/4) Using DDP +2023-03-31 18:52:03,649 INFO [asr_datamodule.py:429] (0/4) About to get the shuffled train-clean-100, train-clean-360 and train-other-500 cuts, combined with their reverberated versions +2023-03-31 18:52:03,689 INFO [asr_datamodule.py:224] (0/4) Enable MUSAN +2023-03-31 18:52:03,689 INFO [asr_datamodule.py:225] (0/4) About to get Musan cuts +2023-03-31 18:52:05,928 INFO [asr_datamodule.py:249] (0/4) Enable SpecAugment +2023-03-31 18:52:05,928 INFO [asr_datamodule.py:250] (0/4) Time warp factor: 80 +2023-03-31 18:52:05,928 INFO [asr_datamodule.py:260] (0/4) Num frame mask: 10 +2023-03-31 18:52:05,928 INFO [asr_datamodule.py:273] (0/4) About to create train dataset +2023-03-31 18:52:05,928 INFO [asr_datamodule.py:300] (0/4) Using DynamicBucketingSampler. +2023-03-31 18:52:08,259 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-03-31 18:52:09,266 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-03-31 18:52:09,538 INFO [asr_datamodule.py:315] (0/4) About to create train dataloader +2023-03-31 18:52:09,538 INFO [asr_datamodule.py:440] (0/4) About to get dev-clean cuts +2023-03-31 18:52:09,540 INFO [asr_datamodule.py:447] (0/4) About to get dev-other cuts +2023-03-31 18:52:09,540 INFO [asr_datamodule.py:346] (0/4) About to create dev dataset +2023-03-31 18:52:09,987 INFO [asr_datamodule.py:363] (0/4) About to create dev dataloader +2023-03-31 18:52:23,164 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-03-31 18:52:24,169 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-03-31 18:52:36,261 INFO [train.py:903] (0/4) Epoch 1, batch 0, loss[loss=7.146, simple_loss=6.465, pruned_loss=6.793, over 19712.00 frames. ], tot_loss[loss=7.146, simple_loss=6.465, pruned_loss=6.793, over 19712.00 frames. ], batch size: 46, lr: 2.50e-02, grad_scale: 2.0 +2023-03-31 18:52:36,262 INFO [train.py:928] (0/4) Computing validation loss +2023-03-31 18:52:49,139 INFO [train.py:937] (0/4) Epoch 1, validation: loss=6.888, simple_loss=6.229, pruned_loss=6.575, over 944034.00 frames. +2023-03-31 18:52:49,140 INFO [train.py:938] (0/4) Maximum memory allocated so far is 11725MB +2023-03-31 18:53:03,044 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-03-31 18:53:58,990 INFO [train.py:903] (0/4) Epoch 1, batch 50, loss[loss=1.375, simple_loss=1.22, pruned_loss=1.381, over 17522.00 frames. ], tot_loss[loss=2.154, simple_loss=1.946, pruned_loss=2.001, over 868928.52 frames. ], batch size: 101, lr: 2.75e-02, grad_scale: 0.125 +2023-03-31 18:54:00,549 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 18:54:26,751 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.4230, 4.4169, 4.4217, 4.4202, 4.4120, 4.4227, 4.4136, 4.4169], + device='cuda:0'), covar=tensor([0.0023, 0.0099, 0.0043, 0.0073, 0.0042, 0.0048, 0.0045, 0.0039], + device='cuda:0'), in_proj_covar=tensor([0.0014, 0.0014, 0.0014, 0.0015, 0.0014, 0.0014, 0.0015, 0.0014], + device='cuda:0'), out_proj_covar=tensor([9.4309e-06, 9.5826e-06, 9.4325e-06, 9.2044e-06, 9.7229e-06, 9.3247e-06, + 9.7607e-06, 9.4217e-06], device='cuda:0') +2023-03-31 18:54:36,661 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 18:54:41,866 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-03-31 18:54:51,554 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=17.36 vs. limit=2.0 +2023-03-31 18:55:11,351 INFO [train.py:903] (0/4) Epoch 1, batch 100, loss[loss=1.066, simple_loss=0.9158, pruned_loss=1.191, over 19727.00 frames. ], tot_loss[loss=1.62, simple_loss=1.442, pruned_loss=1.606, over 1527941.54 frames. ], batch size: 51, lr: 3.00e-02, grad_scale: 0.25 +2023-03-31 18:55:11,538 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=101.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 18:55:17,834 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 9.479e+01 1.678e+02 3.237e+02 1.260e+03 8.630e+04, threshold=6.475e+02, percent-clipped=0.0 +2023-03-31 18:55:25,890 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-03-31 18:55:46,080 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=21.98 vs. limit=2.0 +2023-03-31 18:56:16,543 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=7.37 vs. limit=2.0 +2023-03-31 18:56:20,069 INFO [train.py:903] (0/4) Epoch 1, batch 150, loss[loss=1.071, simple_loss=0.911, pruned_loss=1.156, over 19727.00 frames. ], tot_loss[loss=1.389, simple_loss=1.22, pruned_loss=1.424, over 2038834.94 frames. ], batch size: 63, lr: 3.25e-02, grad_scale: 0.25 +2023-03-31 18:56:21,697 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=3.96 vs. limit=2.0 +2023-03-31 18:56:46,162 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=59.42 vs. limit=5.0 +2023-03-31 18:57:32,417 INFO [train.py:903] (0/4) Epoch 1, batch 200, loss[loss=0.8904, simple_loss=0.7512, pruned_loss=0.9287, over 19058.00 frames. ], tot_loss[loss=1.242, simple_loss=1.081, pruned_loss=1.281, over 2447046.74 frames. ], batch size: 42, lr: 3.50e-02, grad_scale: 0.5 +2023-03-31 18:57:32,453 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-03-31 18:57:39,439 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 7.067e+01 1.186e+02 1.653e+02 2.090e+02 5.158e+02, threshold=3.307e+02, percent-clipped=0.0 +2023-03-31 18:57:50,595 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=10.94 vs. limit=5.0 +2023-03-31 18:58:03,185 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=12.78 vs. limit=5.0 +2023-03-31 18:58:15,856 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=9.66 vs. limit=5.0 +2023-03-31 18:58:43,202 INFO [train.py:903] (0/4) Epoch 1, batch 250, loss[loss=0.9894, simple_loss=0.8377, pruned_loss=0.9593, over 18157.00 frames. ], tot_loss[loss=1.154, simple_loss=0.9975, pruned_loss=1.181, over 2756766.97 frames. ], batch size: 83, lr: 3.75e-02, grad_scale: 0.5 +2023-03-31 18:58:45,588 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.08 vs. limit=2.0 +2023-03-31 18:59:35,280 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9899, 3.9901, 3.9886, 3.9893, 3.9897, 3.9906, 3.9848, 3.9905], + device='cuda:0'), covar=tensor([0.0035, 0.0045, 0.0049, 0.0054, 0.0051, 0.0049, 0.0057, 0.0049], + device='cuda:0'), in_proj_covar=tensor([0.0013, 0.0013, 0.0013, 0.0014, 0.0013, 0.0013, 0.0014, 0.0013], + device='cuda:0'), out_proj_covar=tensor([9.0738e-06, 9.0535e-06, 9.0144e-06, 8.7269e-06, 8.9411e-06, 8.9467e-06, + 8.8238e-06, 8.7904e-06], device='cuda:0') +2023-03-31 18:59:51,886 INFO [train.py:903] (0/4) Epoch 1, batch 300, loss[loss=0.8041, simple_loss=0.6691, pruned_loss=0.7872, over 19816.00 frames. ], tot_loss[loss=1.095, simple_loss=0.9391, pruned_loss=1.108, over 2991484.96 frames. ], batch size: 49, lr: 4.00e-02, grad_scale: 1.0 +2023-03-31 18:59:56,685 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 8.717e+01 1.166e+02 1.521e+02 1.991e+02 3.277e+02, threshold=3.043e+02, percent-clipped=0.0 +2023-03-31 18:59:58,372 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=306.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:00:09,458 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=314.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:00:58,531 INFO [train.py:903] (0/4) Epoch 1, batch 350, loss[loss=0.9322, simple_loss=0.7719, pruned_loss=0.884, over 19676.00 frames. ], tot_loss[loss=1.057, simple_loss=0.9003, pruned_loss=1.054, over 3165631.78 frames. ], batch size: 53, lr: 4.25e-02, grad_scale: 1.0 +2023-03-31 19:01:05,504 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-03-31 19:02:08,281 INFO [train.py:903] (0/4) Epoch 1, batch 400, loss[loss=0.8909, simple_loss=0.7342, pruned_loss=0.8213, over 19745.00 frames. ], tot_loss[loss=1.023, simple_loss=0.8651, pruned_loss=1.004, over 3318503.55 frames. ], batch size: 51, lr: 4.50e-02, grad_scale: 2.0 +2023-03-31 19:02:13,389 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 9.539e+01 1.278e+02 1.546e+02 1.978e+02 5.474e+02, threshold=3.091e+02, percent-clipped=7.0 +2023-03-31 19:02:13,648 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=405.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:02:33,547 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=421.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:03:05,037 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=445.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:03:07,167 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=5.48 vs. limit=5.0 +2023-03-31 19:03:12,886 INFO [train.py:903] (0/4) Epoch 1, batch 450, loss[loss=0.9719, simple_loss=0.8015, pruned_loss=0.8608, over 18138.00 frames. ], tot_loss[loss=0.9995, simple_loss=0.8406, pruned_loss=0.9633, over 3437746.07 frames. ], batch size: 83, lr: 4.75e-02, grad_scale: 2.0 +2023-03-31 19:03:26,470 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.8292, 3.4111, 4.8035, 3.3214, 5.9386, 3.2284, 5.9385, 2.9695], + device='cuda:0'), covar=tensor([0.0049, 0.1632, 0.0288, 0.1983, 0.0055, 0.1800, 0.0054, 0.1742], + device='cuda:0'), in_proj_covar=tensor([0.0016, 0.0018, 0.0016, 0.0019, 0.0015, 0.0018, 0.0016, 0.0018], + device='cuda:0'), out_proj_covar=tensor([1.0884e-05, 1.2206e-05, 1.1148e-05, 1.1953e-05, 1.0952e-05, 1.1976e-05, + 1.0788e-05, 1.3087e-05], device='cuda:0') +2023-03-31 19:03:46,825 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6804, 4.6723, 4.6801, 4.6671, 4.5651, 4.6717, 4.6791, 4.6772], + device='cuda:0'), covar=tensor([0.0088, 0.0064, 0.0084, 0.0071, 0.0091, 0.0083, 0.0096, 0.0082], + device='cuda:0'), in_proj_covar=tensor([0.0014, 0.0013, 0.0014, 0.0013, 0.0014, 0.0013, 0.0014, 0.0014], + device='cuda:0'), out_proj_covar=tensor([9.1936e-06, 8.7371e-06, 9.0572e-06, 8.9951e-06, 9.5862e-06, 9.0783e-06, + 8.8703e-06, 9.3240e-06], device='cuda:0') +2023-03-31 19:03:49,660 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-03-31 19:03:51,779 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-03-31 19:04:19,402 INFO [train.py:903] (0/4) Epoch 1, batch 500, loss[loss=1.004, simple_loss=0.8314, pruned_loss=0.8508, over 18132.00 frames. ], tot_loss[loss=0.981, simple_loss=0.8219, pruned_loss=0.9235, over 3518521.20 frames. ], batch size: 83, lr: 4.99e-02, grad_scale: 2.0 +2023-03-31 19:04:22,543 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=4.04 vs. limit=2.0 +2023-03-31 19:04:25,197 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 9.539e+01 1.386e+02 1.860e+02 2.529e+02 4.736e+02, threshold=3.719e+02, percent-clipped=12.0 +2023-03-31 19:05:06,540 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6726, 1.3132, 1.6530, 1.2703, 2.1970, 1.7914, 1.1416, 1.4102], + device='cuda:0'), covar=tensor([0.6733, 0.8468, 0.6514, 0.9552, 0.5075, 0.6402, 0.6547, 0.7141], + device='cuda:0'), in_proj_covar=tensor([0.0041, 0.0043, 0.0040, 0.0044, 0.0035, 0.0046, 0.0041, 0.0040], + device='cuda:0'), out_proj_covar=tensor([2.6551e-05, 2.9653e-05, 2.5329e-05, 2.3989e-05, 2.1307e-05, 2.8475e-05, + 2.4338e-05, 2.2266e-05], device='cuda:0') +2023-03-31 19:05:27,834 INFO [train.py:903] (0/4) Epoch 1, batch 550, loss[loss=0.901, simple_loss=0.7521, pruned_loss=0.7258, over 19530.00 frames. ], tot_loss[loss=0.9627, simple_loss=0.805, pruned_loss=0.8816, over 3589697.23 frames. ], batch size: 54, lr: 4.98e-02, grad_scale: 2.0 +2023-03-31 19:05:40,607 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=560.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:06:03,851 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=580.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:06:12,412 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=586.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:06:32,552 INFO [train.py:903] (0/4) Epoch 1, batch 600, loss[loss=0.8931, simple_loss=0.7527, pruned_loss=0.6844, over 19674.00 frames. ], tot_loss[loss=0.9393, simple_loss=0.7856, pruned_loss=0.8344, over 3657401.10 frames. ], batch size: 60, lr: 4.98e-02, grad_scale: 2.0 +2023-03-31 19:06:36,898 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 1.512e+02 2.910e+02 4.086e+02 6.136e+02 1.097e+03, threshold=8.173e+02, percent-clipped=60.0 +2023-03-31 19:06:40,860 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=608.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:07:11,480 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-03-31 19:07:13,974 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.92 vs. limit=2.0 +2023-03-31 19:07:38,076 INFO [train.py:903] (0/4) Epoch 1, batch 650, loss[loss=0.6846, simple_loss=0.581, pruned_loss=0.5043, over 19778.00 frames. ], tot_loss[loss=0.9158, simple_loss=0.7673, pruned_loss=0.7881, over 3686324.51 frames. ], batch size: 47, lr: 4.98e-02, grad_scale: 2.0 +2023-03-31 19:07:47,314 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=658.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:08:11,546 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=677.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:08:41,839 INFO [train.py:903] (0/4) Epoch 1, batch 700, loss[loss=0.7609, simple_loss=0.6523, pruned_loss=0.5364, over 19533.00 frames. ], tot_loss[loss=0.886, simple_loss=0.7446, pruned_loss=0.7384, over 3721873.14 frames. ], batch size: 54, lr: 4.98e-02, grad_scale: 2.0 +2023-03-31 19:08:42,981 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=702.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:08:46,604 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.000e+02 5.173e+02 6.580e+02 8.914e+02 3.039e+03, threshold=1.316e+03, percent-clipped=29.0 +2023-03-31 19:09:04,075 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=2.99 vs. limit=2.0 +2023-03-31 19:09:43,554 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=749.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:09:45,317 INFO [train.py:903] (0/4) Epoch 1, batch 750, loss[loss=0.7706, simple_loss=0.6647, pruned_loss=0.5266, over 19657.00 frames. ], tot_loss[loss=0.8585, simple_loss=0.724, pruned_loss=0.6927, over 3741912.36 frames. ], batch size: 60, lr: 4.97e-02, grad_scale: 2.0 +2023-03-31 19:10:09,419 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1803, 0.8552, 1.0174, 1.0734, 1.2735, 1.7623, 0.9972, 1.2472], + device='cuda:0'), covar=tensor([0.9400, 1.4627, 1.5910, 1.2003, 0.5534, 1.1186, 1.2665, 0.9890], + device='cuda:0'), in_proj_covar=tensor([0.0081, 0.0099, 0.0107, 0.0091, 0.0061, 0.0112, 0.0090, 0.0087], + device='cuda:0'), out_proj_covar=tensor([5.3175e-05, 6.5926e-05, 6.9401e-05, 5.2876e-05, 3.5719e-05, 7.4215e-05, + 5.5713e-05, 5.1770e-05], device='cuda:0') +2023-03-31 19:10:14,495 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=773.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:10:48,969 INFO [train.py:903] (0/4) Epoch 1, batch 800, loss[loss=0.7628, simple_loss=0.6544, pruned_loss=0.52, over 19668.00 frames. ], tot_loss[loss=0.8324, simple_loss=0.7045, pruned_loss=0.6515, over 3768970.45 frames. ], batch size: 55, lr: 4.97e-02, grad_scale: 4.0 +2023-03-31 19:10:53,085 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.110e+02 5.827e+02 7.991e+02 1.030e+03 2.888e+03, threshold=1.598e+03, percent-clipped=14.0 +2023-03-31 19:11:01,626 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-03-31 19:11:08,320 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=816.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:11:39,603 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=841.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:11:40,561 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-31 19:11:52,174 INFO [train.py:903] (0/4) Epoch 1, batch 850, loss[loss=0.6531, simple_loss=0.564, pruned_loss=0.4328, over 19771.00 frames. ], tot_loss[loss=0.8074, simple_loss=0.6858, pruned_loss=0.614, over 3773175.14 frames. ], batch size: 47, lr: 4.96e-02, grad_scale: 4.0 +2023-03-31 19:12:10,469 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=864.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:12:11,228 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=865.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:12:15,097 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0 +2023-03-31 19:12:43,328 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-03-31 19:12:43,565 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5622, 2.0142, 1.3027, 2.9773, 3.7648, 2.0498, 3.2785, 3.4262], + device='cuda:0'), covar=tensor([0.3922, 0.7327, 1.3081, 0.3319, 0.3507, 1.0428, 0.3158, 0.2801], + device='cuda:0'), in_proj_covar=tensor([0.0063, 0.0071, 0.0096, 0.0058, 0.0072, 0.0090, 0.0074, 0.0059], + device='cuda:0'), out_proj_covar=tensor([3.9235e-05, 5.0024e-05, 6.6715e-05, 3.8274e-05, 4.3811e-05, 6.2202e-05, + 4.4756e-05, 4.0852e-05], device='cuda:0') +2023-03-31 19:12:46,097 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5870, 1.2061, 1.7887, 1.3091, 1.8403, 1.7956, 1.6903, 1.6813], + device='cuda:0'), covar=tensor([0.3519, 0.5668, 0.2932, 0.4128, 0.3150, 0.2355, 0.3173, 0.3253], + device='cuda:0'), in_proj_covar=tensor([0.0060, 0.0064, 0.0057, 0.0063, 0.0061, 0.0052, 0.0061, 0.0054], + device='cuda:0'), out_proj_covar=tensor([3.8451e-05, 4.6669e-05, 3.4982e-05, 4.4182e-05, 3.7468e-05, 3.0544e-05, + 3.8210e-05, 3.4099e-05], device='cuda:0') +2023-03-31 19:12:54,615 INFO [train.py:903] (0/4) Epoch 1, batch 900, loss[loss=0.7565, simple_loss=0.6521, pruned_loss=0.497, over 19308.00 frames. ], tot_loss[loss=0.7855, simple_loss=0.6699, pruned_loss=0.5804, over 3784655.08 frames. ], batch size: 66, lr: 4.96e-02, grad_scale: 4.0 +2023-03-31 19:12:59,599 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.676e+02 6.072e+02 7.456e+02 9.579e+02 1.181e+04, threshold=1.491e+03, percent-clipped=3.0 +2023-03-31 19:13:21,564 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=924.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:13:30,344 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=930.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:13:53,706 INFO [train.py:903] (0/4) Epoch 1, batch 950, loss[loss=0.6747, simple_loss=0.5881, pruned_loss=0.4289, over 19495.00 frames. ], tot_loss[loss=0.7674, simple_loss=0.6569, pruned_loss=0.5519, over 3792702.56 frames. ], batch size: 49, lr: 4.96e-02, grad_scale: 4.0 +2023-03-31 19:13:53,741 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-03-31 19:13:55,996 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=952.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:14:00,790 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.97 vs. limit=5.0 +2023-03-31 19:14:51,788 INFO [train.py:903] (0/4) Epoch 1, batch 1000, loss[loss=0.673, simple_loss=0.593, pruned_loss=0.415, over 17676.00 frames. ], tot_loss[loss=0.7484, simple_loss=0.643, pruned_loss=0.5251, over 3798892.75 frames. ], batch size: 102, lr: 4.95e-02, grad_scale: 4.0 +2023-03-31 19:14:56,987 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.794e+02 5.980e+02 7.509e+02 1.052e+03 2.029e+03, threshold=1.502e+03, percent-clipped=4.0 +2023-03-31 19:15:14,919 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2826, 1.4962, 1.3002, 2.1021, 1.6819, 2.3933, 2.5087, 1.4233], + device='cuda:0'), covar=tensor([0.4102, 0.9753, 1.0359, 0.4361, 1.4447, 0.4055, 0.3919, 0.8111], + device='cuda:0'), in_proj_covar=tensor([0.0040, 0.0057, 0.0071, 0.0044, 0.0081, 0.0044, 0.0046, 0.0051], + device='cuda:0'), out_proj_covar=tensor([2.5678e-05, 4.0111e-05, 5.0702e-05, 2.8625e-05, 6.0590e-05, 2.5738e-05, + 2.7663e-05, 3.5199e-05], device='cuda:0') +2023-03-31 19:15:25,830 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1029.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:15:38,832 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1039.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:15:41,741 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-03-31 19:15:45,014 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1045.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:15:52,666 INFO [train.py:903] (0/4) Epoch 1, batch 1050, loss[loss=0.5841, simple_loss=0.516, pruned_loss=0.3555, over 19366.00 frames. ], tot_loss[loss=0.7296, simple_loss=0.6293, pruned_loss=0.5002, over 3791228.60 frames. ], batch size: 47, lr: 4.95e-02, grad_scale: 4.0 +2023-03-31 19:15:56,716 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1054.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:16:12,357 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1067.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:16:20,733 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-03-31 19:16:53,093 INFO [train.py:903] (0/4) Epoch 1, batch 1100, loss[loss=0.5984, simple_loss=0.5341, pruned_loss=0.355, over 19766.00 frames. ], tot_loss[loss=0.7141, simple_loss=0.6181, pruned_loss=0.479, over 3787483.63 frames. ], batch size: 51, lr: 4.94e-02, grad_scale: 4.0 +2023-03-31 19:16:57,405 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.036e+02 7.117e+02 8.563e+02 1.068e+03 2.368e+03, threshold=1.713e+03, percent-clipped=4.0 +2023-03-31 19:17:14,920 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1120.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:17:23,436 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.01 vs. limit=2.0 +2023-03-31 19:17:44,289 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1145.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:17:51,776 INFO [train.py:903] (0/4) Epoch 1, batch 1150, loss[loss=0.6175, simple_loss=0.544, pruned_loss=0.3721, over 19708.00 frames. ], tot_loss[loss=0.6988, simple_loss=0.6071, pruned_loss=0.4593, over 3789169.86 frames. ], batch size: 51, lr: 4.94e-02, grad_scale: 4.0 +2023-03-31 19:18:13,009 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1171.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:18:47,616 INFO [train.py:903] (0/4) Epoch 1, batch 1200, loss[loss=0.5968, simple_loss=0.5419, pruned_loss=0.3398, over 19526.00 frames. ], tot_loss[loss=0.6888, simple_loss=0.5999, pruned_loss=0.4448, over 3797569.53 frames. ], batch size: 56, lr: 4.93e-02, grad_scale: 8.0 +2023-03-31 19:18:52,223 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.229e+02 7.433e+02 9.314e+02 1.239e+03 3.000e+03, threshold=1.863e+03, percent-clipped=16.0 +2023-03-31 19:18:56,108 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1209.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:19:16,396 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-03-31 19:19:42,377 INFO [train.py:903] (0/4) Epoch 1, batch 1250, loss[loss=0.6163, simple_loss=0.5506, pruned_loss=0.3589, over 19661.00 frames. ], tot_loss[loss=0.6732, simple_loss=0.5894, pruned_loss=0.4262, over 3811134.84 frames. ], batch size: 53, lr: 4.92e-02, grad_scale: 8.0 +2023-03-31 19:19:55,590 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3130, 1.5912, 1.2793, 2.6692, 3.0474, 1.5798, 2.7197, 2.9856], + device='cuda:0'), covar=tensor([0.1155, 0.5102, 0.9183, 0.2436, 0.1545, 0.8546, 0.1752, 0.1691], + device='cuda:0'), in_proj_covar=tensor([0.0077, 0.0099, 0.0129, 0.0083, 0.0101, 0.0153, 0.0094, 0.0079], + device='cuda:0'), out_proj_covar=tensor([4.4735e-05, 6.8349e-05, 8.9177e-05, 5.7078e-05, 6.1210e-05, 1.0065e-04, + 5.8989e-05, 5.3781e-05], device='cuda:0') +2023-03-31 19:20:32,301 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1295.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:20:39,090 INFO [train.py:903] (0/4) Epoch 1, batch 1300, loss[loss=0.6739, simple_loss=0.585, pruned_loss=0.4072, over 19538.00 frames. ], tot_loss[loss=0.6612, simple_loss=0.5807, pruned_loss=0.4119, over 3818189.75 frames. ], batch size: 56, lr: 4.92e-02, grad_scale: 8.0 +2023-03-31 19:20:39,463 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1301.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:20:43,725 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.522e+02 7.699e+02 1.048e+03 1.379e+03 4.741e+03, threshold=2.097e+03, percent-clipped=13.0 +2023-03-31 19:20:59,996 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1320.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:21:03,670 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1323.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:21:04,874 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1324.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:21:06,781 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1326.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:21:33,403 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1348.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:21:35,927 INFO [train.py:903] (0/4) Epoch 1, batch 1350, loss[loss=0.6695, simple_loss=0.5923, pruned_loss=0.3916, over 19400.00 frames. ], tot_loss[loss=0.6537, simple_loss=0.576, pruned_loss=0.4013, over 3814240.90 frames. ], batch size: 70, lr: 4.91e-02, grad_scale: 8.0 +2023-03-31 19:22:31,179 INFO [train.py:903] (0/4) Epoch 1, batch 1400, loss[loss=0.6235, simple_loss=0.5669, pruned_loss=0.3493, over 19668.00 frames. ], tot_loss[loss=0.6409, simple_loss=0.5678, pruned_loss=0.3869, over 3829234.77 frames. ], batch size: 58, lr: 4.91e-02, grad_scale: 8.0 +2023-03-31 19:22:35,192 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.777e+02 7.620e+02 9.515e+02 1.230e+03 4.278e+03, threshold=1.903e+03, percent-clipped=3.0 +2023-03-31 19:23:24,293 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-03-31 19:23:25,213 INFO [train.py:903] (0/4) Epoch 1, batch 1450, loss[loss=0.6027, simple_loss=0.5475, pruned_loss=0.3371, over 18242.00 frames. ], tot_loss[loss=0.6327, simple_loss=0.562, pruned_loss=0.3775, over 3828520.83 frames. ], batch size: 83, lr: 4.90e-02, grad_scale: 8.0 +2023-03-31 19:23:26,441 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1452.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:23:34,536 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.21 vs. limit=2.0 +2023-03-31 19:23:56,780 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.25 vs. limit=2.0 +2023-03-31 19:24:18,018 INFO [train.py:903] (0/4) Epoch 1, batch 1500, loss[loss=0.4681, simple_loss=0.4393, pruned_loss=0.2497, over 19382.00 frames. ], tot_loss[loss=0.6247, simple_loss=0.5569, pruned_loss=0.3682, over 3822547.94 frames. ], batch size: 47, lr: 4.89e-02, grad_scale: 8.0 +2023-03-31 19:24:23,055 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.214e+02 9.104e+02 1.060e+03 1.370e+03 5.981e+03, threshold=2.119e+03, percent-clipped=12.0 +2023-03-31 19:24:35,369 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1515.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:25:14,430 INFO [train.py:903] (0/4) Epoch 1, batch 1550, loss[loss=0.6301, simple_loss=0.5559, pruned_loss=0.3631, over 13548.00 frames. ], tot_loss[loss=0.6153, simple_loss=0.5512, pruned_loss=0.3582, over 3829774.35 frames. ], batch size: 136, lr: 4.89e-02, grad_scale: 8.0 +2023-03-31 19:25:27,687 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-31 19:25:43,546 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1580.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:25:47,315 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1584.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:26:05,877 INFO [train.py:903] (0/4) Epoch 1, batch 1600, loss[loss=0.5265, simple_loss=0.484, pruned_loss=0.2879, over 19796.00 frames. ], tot_loss[loss=0.6061, simple_loss=0.5451, pruned_loss=0.349, over 3835439.24 frames. ], batch size: 47, lr: 4.88e-02, grad_scale: 8.0 +2023-03-31 19:26:10,815 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.811e+02 9.198e+02 1.152e+03 1.497e+03 2.578e+03, threshold=2.303e+03, percent-clipped=3.0 +2023-03-31 19:26:11,200 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1605.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:26:25,444 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-03-31 19:26:35,458 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1629.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:26:36,331 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1630.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:26:57,914 INFO [train.py:903] (0/4) Epoch 1, batch 1650, loss[loss=0.4915, simple_loss=0.462, pruned_loss=0.2612, over 19604.00 frames. ], tot_loss[loss=0.5993, simple_loss=0.5409, pruned_loss=0.3417, over 3830103.64 frames. ], batch size: 50, lr: 4.87e-02, grad_scale: 8.0 +2023-03-31 19:27:52,350 INFO [train.py:903] (0/4) Epoch 1, batch 1700, loss[loss=0.561, simple_loss=0.5255, pruned_loss=0.2992, over 19520.00 frames. ], tot_loss[loss=0.5891, simple_loss=0.5341, pruned_loss=0.3326, over 3829326.22 frames. ], batch size: 56, lr: 4.86e-02, grad_scale: 8.0 +2023-03-31 19:27:56,185 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.719e+02 9.402e+02 1.223e+03 1.535e+03 2.582e+03, threshold=2.447e+03, percent-clipped=3.0 +2023-03-31 19:28:26,602 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-03-31 19:28:46,835 INFO [train.py:903] (0/4) Epoch 1, batch 1750, loss[loss=0.5212, simple_loss=0.4845, pruned_loss=0.2801, over 19727.00 frames. ], tot_loss[loss=0.5819, simple_loss=0.5299, pruned_loss=0.3257, over 3823502.02 frames. ], batch size: 51, lr: 4.86e-02, grad_scale: 8.0 +2023-03-31 19:29:37,345 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1796.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:29:43,454 INFO [train.py:903] (0/4) Epoch 1, batch 1800, loss[loss=0.5464, simple_loss=0.5096, pruned_loss=0.2924, over 19779.00 frames. ], tot_loss[loss=0.5735, simple_loss=0.5238, pruned_loss=0.3187, over 3818986.45 frames. ], batch size: 54, lr: 4.85e-02, grad_scale: 8.0 +2023-03-31 19:29:47,613 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.927e+02 9.266e+02 1.209e+03 1.539e+03 2.564e+03, threshold=2.418e+03, percent-clipped=2.0 +2023-03-31 19:30:36,189 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-03-31 19:30:40,571 INFO [train.py:903] (0/4) Epoch 1, batch 1850, loss[loss=0.5994, simple_loss=0.552, pruned_loss=0.3247, over 19735.00 frames. ], tot_loss[loss=0.5677, simple_loss=0.5207, pruned_loss=0.3131, over 3810574.60 frames. ], batch size: 63, lr: 4.84e-02, grad_scale: 8.0 +2023-03-31 19:30:45,827 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1856.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:31:06,943 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1875.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:31:11,911 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-03-31 19:31:20,275 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=1886.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:31:22,125 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=1888.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:31:36,186 INFO [train.py:903] (0/4) Epoch 1, batch 1900, loss[loss=0.4577, simple_loss=0.442, pruned_loss=0.2363, over 19751.00 frames. ], tot_loss[loss=0.5635, simple_loss=0.5186, pruned_loss=0.3088, over 3810490.24 frames. ], batch size: 46, lr: 4.83e-02, grad_scale: 8.0 +2023-03-31 19:31:40,294 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.070e+02 9.078e+02 1.104e+03 1.499e+03 2.754e+03, threshold=2.207e+03, percent-clipped=2.0 +2023-03-31 19:31:47,404 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=1911.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:31:47,424 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=1911.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:31:52,273 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-03-31 19:31:56,268 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-03-31 19:32:06,353 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1928.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:32:19,691 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-03-31 19:32:31,510 INFO [train.py:903] (0/4) Epoch 1, batch 1950, loss[loss=0.5672, simple_loss=0.5261, pruned_loss=0.3044, over 18824.00 frames. ], tot_loss[loss=0.5604, simple_loss=0.5174, pruned_loss=0.3053, over 3818182.08 frames. ], batch size: 74, lr: 4.83e-02, grad_scale: 8.0 +2023-03-31 19:32:57,186 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=1973.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:33:27,214 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-2000.pt +2023-03-31 19:33:29,175 INFO [train.py:903] (0/4) Epoch 1, batch 2000, loss[loss=0.597, simple_loss=0.5495, pruned_loss=0.3222, over 19640.00 frames. ], tot_loss[loss=0.5567, simple_loss=0.5154, pruned_loss=0.3018, over 3829921.80 frames. ], batch size: 60, lr: 4.82e-02, grad_scale: 8.0 +2023-03-31 19:33:33,541 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.917e+02 1.007e+03 1.260e+03 1.703e+03 3.255e+03, threshold=2.521e+03, percent-clipped=11.0 +2023-03-31 19:34:18,968 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2043.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:34:25,230 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-03-31 19:34:27,591 INFO [train.py:903] (0/4) Epoch 1, batch 2050, loss[loss=0.5722, simple_loss=0.5363, pruned_loss=0.304, over 18240.00 frames. ], tot_loss[loss=0.5458, simple_loss=0.5096, pruned_loss=0.2932, over 3843484.33 frames. ], batch size: 83, lr: 4.81e-02, grad_scale: 16.0 +2023-03-31 19:34:43,694 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-03-31 19:34:43,730 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-03-31 19:35:06,976 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-03-31 19:35:12,669 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2088.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:35:27,372 INFO [train.py:903] (0/4) Epoch 1, batch 2100, loss[loss=0.561, simple_loss=0.5319, pruned_loss=0.295, over 19670.00 frames. ], tot_loss[loss=0.5382, simple_loss=0.5056, pruned_loss=0.2871, over 3841917.07 frames. ], batch size: 60, lr: 4.80e-02, grad_scale: 16.0 +2023-03-31 19:35:31,663 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.951e+02 9.211e+02 1.091e+03 1.524e+03 2.851e+03, threshold=2.182e+03, percent-clipped=6.0 +2023-03-31 19:35:56,703 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-03-31 19:36:17,081 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-03-31 19:36:24,974 INFO [train.py:903] (0/4) Epoch 1, batch 2150, loss[loss=0.4437, simple_loss=0.4417, pruned_loss=0.2229, over 19748.00 frames. ], tot_loss[loss=0.5305, simple_loss=0.5014, pruned_loss=0.2811, over 3836517.68 frames. ], batch size: 46, lr: 4.79e-02, grad_scale: 16.0 +2023-03-31 19:36:38,452 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-03-31 19:36:45,564 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2167.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:37:13,898 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2192.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:37:24,838 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2200.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:37:25,812 INFO [train.py:903] (0/4) Epoch 1, batch 2200, loss[loss=0.429, simple_loss=0.4263, pruned_loss=0.2159, over 19762.00 frames. ], tot_loss[loss=0.5242, simple_loss=0.498, pruned_loss=0.2762, over 3828303.93 frames. ], batch size: 45, lr: 4.78e-02, grad_scale: 16.0 +2023-03-31 19:37:31,687 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.347e+02 9.332e+02 1.145e+03 1.435e+03 3.303e+03, threshold=2.290e+03, percent-clipped=7.0 +2023-03-31 19:37:49,456 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2219.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:38:03,593 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2232.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:38:27,470 INFO [train.py:903] (0/4) Epoch 1, batch 2250, loss[loss=0.4595, simple_loss=0.4455, pruned_loss=0.2367, over 19747.00 frames. ], tot_loss[loss=0.5199, simple_loss=0.4961, pruned_loss=0.2727, over 3842504.62 frames. ], batch size: 46, lr: 4.77e-02, grad_scale: 16.0 +2023-03-31 19:39:24,157 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2299.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:39:25,909 INFO [train.py:903] (0/4) Epoch 1, batch 2300, loss[loss=0.3842, simple_loss=0.3978, pruned_loss=0.1853, over 19706.00 frames. ], tot_loss[loss=0.5143, simple_loss=0.4927, pruned_loss=0.2686, over 3830064.83 frames. ], batch size: 45, lr: 4.77e-02, grad_scale: 8.0 +2023-03-31 19:39:31,324 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.883e+02 9.458e+02 1.205e+03 1.557e+03 3.326e+03, threshold=2.410e+03, percent-clipped=10.0 +2023-03-31 19:39:39,116 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-03-31 19:39:41,661 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2315.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:39:52,808 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6341, 1.2358, 1.3032, 1.4180, 1.5748, 2.0309, 1.6730, 1.7201], + device='cuda:0'), covar=tensor([0.1632, 0.2489, 0.3255, 0.2157, 0.4948, 0.1066, 0.2747, 0.1710], + device='cuda:0'), in_proj_covar=tensor([0.0149, 0.0165, 0.0209, 0.0161, 0.0256, 0.0159, 0.0188, 0.0169], + device='cuda:0'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0002, 0.0001, 0.0001, 0.0001], + device='cuda:0') +2023-03-31 19:39:53,803 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8115, 3.3905, 4.0015, 3.9181, 1.6348, 3.8286, 3.4032, 4.0125], + device='cuda:0'), covar=tensor([0.0232, 0.0484, 0.0243, 0.0148, 0.2734, 0.0215, 0.0411, 0.0271], + device='cuda:0'), in_proj_covar=tensor([0.0087, 0.0117, 0.0108, 0.0082, 0.0210, 0.0082, 0.0115, 0.0105], + device='cuda:0'), out_proj_covar=tensor([5.3152e-05, 8.0129e-05, 6.1727e-05, 4.8572e-05, 1.2797e-04, 4.8900e-05, + 7.2126e-05, 6.2226e-05], device='cuda:0') +2023-03-31 19:39:53,909 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2324.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:40:04,946 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2334.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:40:17,232 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2344.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:40:20,632 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2347.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:40:24,828 INFO [train.py:903] (0/4) Epoch 1, batch 2350, loss[loss=0.4817, simple_loss=0.4831, pruned_loss=0.2402, over 19668.00 frames. ], tot_loss[loss=0.5085, simple_loss=0.4891, pruned_loss=0.2645, over 3826569.52 frames. ], batch size: 55, lr: 4.76e-02, grad_scale: 8.0 +2023-03-31 19:40:33,247 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2358.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:40:48,353 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2369.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:40:55,229 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.08 vs. limit=2.0 +2023-03-31 19:41:07,244 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-03-31 19:41:23,365 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-03-31 19:41:26,485 INFO [train.py:903] (0/4) Epoch 1, batch 2400, loss[loss=0.3963, simple_loss=0.4059, pruned_loss=0.1933, over 19712.00 frames. ], tot_loss[loss=0.5031, simple_loss=0.4858, pruned_loss=0.2606, over 3811040.67 frames. ], batch size: 46, lr: 4.75e-02, grad_scale: 8.0 +2023-03-31 19:41:33,163 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.460e+02 9.458e+02 1.226e+03 1.613e+03 2.603e+03, threshold=2.451e+03, percent-clipped=4.0 +2023-03-31 19:42:26,157 INFO [train.py:903] (0/4) Epoch 1, batch 2450, loss[loss=0.4956, simple_loss=0.4921, pruned_loss=0.2495, over 19692.00 frames. ], tot_loss[loss=0.4989, simple_loss=0.4839, pruned_loss=0.2573, over 3808027.52 frames. ], batch size: 59, lr: 4.74e-02, grad_scale: 8.0 +2023-03-31 19:42:30,970 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6255, 1.7265, 1.7256, 0.9989, 2.2592, 2.9343, 2.6946, 2.6101], + device='cuda:0'), covar=tensor([0.2409, 0.2267, 0.1689, 0.3475, 0.0945, 0.0192, 0.0331, 0.0429], + device='cuda:0'), in_proj_covar=tensor([0.0192, 0.0162, 0.0139, 0.0197, 0.0135, 0.0078, 0.0101, 0.0086], + device='cuda:0'), out_proj_covar=tensor([1.2660e-04, 1.0630e-04, 8.9364e-05, 1.2763e-04, 9.4105e-05, 4.3811e-05, + 5.8731e-05, 5.0579e-05], device='cuda:0') +2023-03-31 19:43:20,049 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0522, 1.1387, 1.5920, 1.3550, 2.0389, 2.7786, 2.3732, 1.6335], + device='cuda:0'), covar=tensor([0.3209, 0.2079, 0.2486, 0.3159, 0.1235, 0.0370, 0.0804, 0.2015], + device='cuda:0'), in_proj_covar=tensor([0.0136, 0.0108, 0.0126, 0.0150, 0.0139, 0.0066, 0.0115, 0.0140], + device='cuda:0'), out_proj_covar=tensor([9.2116e-05, 7.1223e-05, 8.7536e-05, 1.0373e-04, 9.4163e-05, 4.0542e-05, + 7.7590e-05, 9.2573e-05], device='cuda:0') +2023-03-31 19:43:24,722 INFO [train.py:903] (0/4) Epoch 1, batch 2500, loss[loss=0.5268, simple_loss=0.4844, pruned_loss=0.2846, over 19405.00 frames. ], tot_loss[loss=0.495, simple_loss=0.4814, pruned_loss=0.2546, over 3804397.54 frames. ], batch size: 48, lr: 4.73e-02, grad_scale: 8.0 +2023-03-31 19:43:30,995 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.049e+02 1.082e+03 1.390e+03 1.742e+03 4.873e+03, threshold=2.779e+03, percent-clipped=5.0 +2023-03-31 19:43:41,913 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6601, 3.2049, 1.9137, 2.8510, 1.4318, 3.3909, 2.9559, 3.0899], + device='cuda:0'), covar=tensor([0.0665, 0.1270, 0.2628, 0.0679, 0.2884, 0.0480, 0.0743, 0.0819], + device='cuda:0'), in_proj_covar=tensor([0.0167, 0.0200, 0.0208, 0.0139, 0.0220, 0.0129, 0.0133, 0.0128], + device='cuda:0'), out_proj_covar=tensor([1.3689e-04, 1.5780e-04, 1.3921e-04, 1.0297e-04, 1.5396e-04, 9.3544e-05, + 9.5972e-05, 9.3732e-05], device='cuda:0') +2023-03-31 19:44:22,068 INFO [train.py:903] (0/4) Epoch 1, batch 2550, loss[loss=0.5195, simple_loss=0.5046, pruned_loss=0.2672, over 19550.00 frames. ], tot_loss[loss=0.4949, simple_loss=0.4817, pruned_loss=0.2543, over 3799435.64 frames. ], batch size: 56, lr: 4.72e-02, grad_scale: 8.0 +2023-03-31 19:44:47,124 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2571.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:45:09,013 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2590.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:45:14,138 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-03-31 19:45:15,592 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2596.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:45:21,665 INFO [train.py:903] (0/4) Epoch 1, batch 2600, loss[loss=0.4601, simple_loss=0.4629, pruned_loss=0.2287, over 19540.00 frames. ], tot_loss[loss=0.4888, simple_loss=0.4784, pruned_loss=0.2497, over 3812652.54 frames. ], batch size: 56, lr: 4.71e-02, grad_scale: 8.0 +2023-03-31 19:45:24,457 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=2603.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:45:28,255 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.353e+02 9.154e+02 1.259e+03 1.710e+03 2.682e+03, threshold=2.519e+03, percent-clipped=0.0 +2023-03-31 19:45:39,460 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2615.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:45:45,370 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-03-31 19:45:49,965 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.37 vs. limit=5.0 +2023-03-31 19:45:55,203 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=2628.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:46:22,931 INFO [train.py:903] (0/4) Epoch 1, batch 2650, loss[loss=0.4675, simple_loss=0.477, pruned_loss=0.2289, over 19677.00 frames. ], tot_loss[loss=0.4836, simple_loss=0.4754, pruned_loss=0.246, over 3811747.55 frames. ], batch size: 60, lr: 4.70e-02, grad_scale: 8.0 +2023-03-31 19:46:39,346 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-03-31 19:47:23,190 INFO [train.py:903] (0/4) Epoch 1, batch 2700, loss[loss=0.3989, simple_loss=0.4141, pruned_loss=0.1919, over 19767.00 frames. ], tot_loss[loss=0.48, simple_loss=0.4734, pruned_loss=0.2433, over 3826453.55 frames. ], batch size: 48, lr: 4.69e-02, grad_scale: 8.0 +2023-03-31 19:47:24,580 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=2702.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:47:25,677 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2703.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:47:29,292 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-31 19:47:29,729 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.007e+02 8.490e+02 1.133e+03 1.436e+03 3.154e+03, threshold=2.267e+03, percent-clipped=3.0 +2023-03-31 19:47:49,723 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.91 vs. limit=2.0 +2023-03-31 19:48:24,813 INFO [train.py:903] (0/4) Epoch 1, batch 2750, loss[loss=0.4561, simple_loss=0.4627, pruned_loss=0.2248, over 19536.00 frames. ], tot_loss[loss=0.4762, simple_loss=0.4712, pruned_loss=0.2406, over 3817492.91 frames. ], batch size: 56, lr: 4.68e-02, grad_scale: 8.0 +2023-03-31 19:49:25,712 INFO [train.py:903] (0/4) Epoch 1, batch 2800, loss[loss=0.4176, simple_loss=0.4175, pruned_loss=0.2089, over 19077.00 frames. ], tot_loss[loss=0.4739, simple_loss=0.4701, pruned_loss=0.2389, over 3817851.16 frames. ], batch size: 42, lr: 4.67e-02, grad_scale: 8.0 +2023-03-31 19:49:31,054 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.211e+02 1.002e+03 1.265e+03 1.511e+03 4.462e+03, threshold=2.529e+03, percent-clipped=7.0 +2023-03-31 19:49:44,664 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9834, 1.4751, 1.4038, 1.6058, 1.8916, 2.0391, 2.1340, 1.8861], + device='cuda:0'), covar=tensor([0.1329, 0.2093, 0.2483, 0.2545, 0.3126, 0.2307, 0.3219, 0.1657], + device='cuda:0'), in_proj_covar=tensor([0.0197, 0.0244, 0.0255, 0.0251, 0.0341, 0.0242, 0.0306, 0.0219], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 19:49:45,688 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=2817.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:50:06,590 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7005, 1.3421, 1.7658, 0.9726, 2.8271, 2.5730, 2.5508, 2.3098], + device='cuda:0'), covar=tensor([0.1926, 0.2493, 0.1559, 0.3074, 0.0489, 0.0230, 0.0312, 0.0396], + device='cuda:0'), in_proj_covar=tensor([0.0232, 0.0195, 0.0177, 0.0240, 0.0168, 0.0095, 0.0118, 0.0101], + device='cuda:0'), out_proj_covar=tensor([1.5737e-04, 1.3078e-04, 1.1887e-04, 1.5954e-04, 1.2695e-04, 5.8135e-05, + 7.5061e-05, 6.6786e-05], device='cuda:0') +2023-03-31 19:50:15,441 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2842.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:50:26,221 INFO [train.py:903] (0/4) Epoch 1, batch 2850, loss[loss=0.445, simple_loss=0.4562, pruned_loss=0.2169, over 19367.00 frames. ], tot_loss[loss=0.4721, simple_loss=0.4688, pruned_loss=0.2377, over 3810427.28 frames. ], batch size: 70, lr: 4.66e-02, grad_scale: 8.0 +2023-03-31 19:51:16,291 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=2.02 vs. limit=2.0 +2023-03-31 19:51:22,260 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-03-31 19:51:25,259 INFO [train.py:903] (0/4) Epoch 1, batch 2900, loss[loss=0.506, simple_loss=0.4885, pruned_loss=0.2617, over 18913.00 frames. ], tot_loss[loss=0.4699, simple_loss=0.4671, pruned_loss=0.2363, over 3825749.22 frames. ], batch size: 74, lr: 4.65e-02, grad_scale: 8.0 +2023-03-31 19:51:30,481 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.237e+02 1.045e+03 1.349e+03 1.754e+03 3.463e+03, threshold=2.699e+03, percent-clipped=4.0 +2023-03-31 19:52:25,011 INFO [train.py:903] (0/4) Epoch 1, batch 2950, loss[loss=0.4398, simple_loss=0.4507, pruned_loss=0.2144, over 19651.00 frames. ], tot_loss[loss=0.4651, simple_loss=0.464, pruned_loss=0.2331, over 3818577.77 frames. ], batch size: 55, lr: 4.64e-02, grad_scale: 8.0 +2023-03-31 19:52:30,817 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=2955.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:53:26,138 INFO [train.py:903] (0/4) Epoch 1, batch 3000, loss[loss=0.4467, simple_loss=0.4632, pruned_loss=0.2151, over 18729.00 frames. ], tot_loss[loss=0.4631, simple_loss=0.4633, pruned_loss=0.2315, over 3825880.41 frames. ], batch size: 74, lr: 4.63e-02, grad_scale: 8.0 +2023-03-31 19:53:26,139 INFO [train.py:928] (0/4) Computing validation loss +2023-03-31 19:53:38,708 INFO [train.py:937] (0/4) Epoch 1, validation: loss=0.3995, simple_loss=0.4801, pruned_loss=0.1594, over 944034.00 frames. +2023-03-31 19:53:38,709 INFO [train.py:938] (0/4) Maximum memory allocated so far is 15735MB +2023-03-31 19:53:43,194 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-03-31 19:53:45,670 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.188e+02 9.060e+02 1.151e+03 1.550e+03 2.691e+03, threshold=2.303e+03, percent-clipped=0.0 +2023-03-31 19:54:23,238 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3037.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:54:35,488 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3047.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:54:40,047 INFO [train.py:903] (0/4) Epoch 1, batch 3050, loss[loss=0.4571, simple_loss=0.4725, pruned_loss=0.2208, over 19530.00 frames. ], tot_loss[loss=0.4611, simple_loss=0.462, pruned_loss=0.2301, over 3831683.75 frames. ], batch size: 56, lr: 4.62e-02, grad_scale: 8.0 +2023-03-31 19:54:50,176 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3353, 0.9244, 1.0992, 1.2790, 1.1405, 1.7529, 1.4913, 1.4117], + device='cuda:0'), covar=tensor([0.1308, 0.2754, 0.2399, 0.1703, 0.2965, 0.1098, 0.1807, 0.1537], + device='cuda:0'), in_proj_covar=tensor([0.0115, 0.0156, 0.0167, 0.0139, 0.0199, 0.0128, 0.0141, 0.0129], + device='cuda:0'), out_proj_covar=tensor([8.4845e-05, 1.1301e-04, 1.1499e-04, 1.0105e-04, 1.4184e-04, 9.1249e-05, + 9.8527e-05, 9.1916e-05], device='cuda:0') +2023-03-31 19:54:52,332 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0 +2023-03-31 19:55:07,342 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3073.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 19:55:36,288 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3098.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:55:39,212 INFO [train.py:903] (0/4) Epoch 1, batch 3100, loss[loss=0.4429, simple_loss=0.4367, pruned_loss=0.2245, over 19691.00 frames. ], tot_loss[loss=0.4623, simple_loss=0.4622, pruned_loss=0.2312, over 3827373.90 frames. ], batch size: 53, lr: 4.61e-02, grad_scale: 8.0 +2023-03-31 19:55:45,833 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.197e+02 1.021e+03 1.362e+03 1.815e+03 5.785e+03, threshold=2.723e+03, percent-clipped=14.0 +2023-03-31 19:56:17,182 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6042, 1.9688, 2.3053, 1.7458, 1.8235, 0.9392, 0.9093, 1.4751], + device='cuda:0'), covar=tensor([0.1669, 0.0710, 0.0383, 0.1112, 0.1382, 0.1474, 0.2307, 0.1462], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0128, 0.0125, 0.0156, 0.0125, 0.0173, 0.0196, 0.0177], + device='cuda:0'), out_proj_covar=tensor([1.5921e-04, 9.9970e-05, 9.8031e-05, 1.2227e-04, 1.0202e-04, 1.3412e-04, + 1.4646e-04, 1.3607e-04], device='cuda:0') +2023-03-31 19:56:21,061 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8101, 1.1784, 1.6889, 1.1726, 2.5839, 3.3961, 2.9821, 3.0194], + device='cuda:0'), covar=tensor([0.2483, 0.3235, 0.2193, 0.3496, 0.0876, 0.0163, 0.0336, 0.0313], + device='cuda:0'), in_proj_covar=tensor([0.0247, 0.0212, 0.0199, 0.0256, 0.0179, 0.0101, 0.0127, 0.0103], + device='cuda:0'), out_proj_covar=tensor([1.7110e-04, 1.4508e-04, 1.3720e-04, 1.7537e-04, 1.4149e-04, 6.5771e-05, + 8.7744e-05, 7.2562e-05], device='cuda:0') +2023-03-31 19:56:41,856 INFO [train.py:903] (0/4) Epoch 1, batch 3150, loss[loss=0.4726, simple_loss=0.4749, pruned_loss=0.2352, over 19639.00 frames. ], tot_loss[loss=0.4586, simple_loss=0.4599, pruned_loss=0.2287, over 3829200.19 frames. ], batch size: 60, lr: 4.60e-02, grad_scale: 8.0 +2023-03-31 19:56:54,520 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3162.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:57:08,989 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-03-31 19:57:24,072 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3186.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:57:30,846 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.67 vs. limit=5.0 +2023-03-31 19:57:40,829 INFO [train.py:903] (0/4) Epoch 1, batch 3200, loss[loss=0.3935, simple_loss=0.4167, pruned_loss=0.1851, over 19500.00 frames. ], tot_loss[loss=0.4558, simple_loss=0.4585, pruned_loss=0.2265, over 3835957.49 frames. ], batch size: 49, lr: 4.59e-02, grad_scale: 8.0 +2023-03-31 19:57:46,448 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.226e+02 9.158e+02 1.127e+03 1.418e+03 2.574e+03, threshold=2.253e+03, percent-clipped=0.0 +2023-03-31 19:58:41,775 INFO [train.py:903] (0/4) Epoch 1, batch 3250, loss[loss=0.4397, simple_loss=0.4546, pruned_loss=0.2124, over 18766.00 frames. ], tot_loss[loss=0.4536, simple_loss=0.4571, pruned_loss=0.225, over 3830299.68 frames. ], batch size: 74, lr: 4.58e-02, grad_scale: 8.0 +2023-03-31 19:59:40,386 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3299.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:59:42,435 INFO [train.py:903] (0/4) Epoch 1, batch 3300, loss[loss=0.4811, simple_loss=0.487, pruned_loss=0.2376, over 19329.00 frames. ], tot_loss[loss=0.4496, simple_loss=0.4548, pruned_loss=0.2223, over 3810273.52 frames. ], batch size: 66, lr: 4.57e-02, grad_scale: 8.0 +2023-03-31 19:59:42,803 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3301.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 19:59:48,777 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.005e+02 9.991e+02 1.183e+03 1.562e+03 4.237e+03, threshold=2.366e+03, percent-clipped=7.0 +2023-03-31 19:59:48,826 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-03-31 19:59:49,156 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3306.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 19:59:54,697 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2965, 1.0628, 1.0108, 1.2678, 1.2097, 1.3980, 1.2427, 1.3417], + device='cuda:0'), covar=tensor([0.1132, 0.1895, 0.1920, 0.1551, 0.2102, 0.1668, 0.2166, 0.1278], + device='cuda:0'), in_proj_covar=tensor([0.0204, 0.0254, 0.0256, 0.0271, 0.0361, 0.0245, 0.0319, 0.0221], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 20:00:43,740 INFO [train.py:903] (0/4) Epoch 1, batch 3350, loss[loss=0.3835, simple_loss=0.4044, pruned_loss=0.1814, over 18987.00 frames. ], tot_loss[loss=0.448, simple_loss=0.4538, pruned_loss=0.2211, over 3809269.82 frames. ], batch size: 42, lr: 4.56e-02, grad_scale: 8.0 +2023-03-31 20:00:49,625 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0428, 4.2892, 5.6605, 5.3730, 1.8778, 4.9086, 4.5654, 5.1230], + device='cuda:0'), covar=tensor([0.0169, 0.0393, 0.0222, 0.0134, 0.3079, 0.0324, 0.0410, 0.0431], + device='cuda:0'), in_proj_covar=tensor([0.0131, 0.0176, 0.0199, 0.0132, 0.0307, 0.0116, 0.0167, 0.0182], + device='cuda:0'), out_proj_covar=tensor([9.3834e-05, 1.2303e-04, 1.2869e-04, 8.2271e-05, 1.7637e-04, 7.6271e-05, + 1.1318e-04, 1.1617e-04], device='cuda:0') +2023-03-31 20:01:22,256 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3381.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:01:46,107 INFO [train.py:903] (0/4) Epoch 1, batch 3400, loss[loss=0.4229, simple_loss=0.4161, pruned_loss=0.2149, over 19773.00 frames. ], tot_loss[loss=0.4471, simple_loss=0.4534, pruned_loss=0.2204, over 3816417.29 frames. ], batch size: 45, lr: 4.55e-02, grad_scale: 8.0 +2023-03-31 20:01:46,920 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0 +2023-03-31 20:01:48,080 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.36 vs. limit=5.0 +2023-03-31 20:01:52,903 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.525e+02 9.967e+02 1.253e+03 1.611e+03 4.007e+03, threshold=2.507e+03, percent-clipped=3.0 +2023-03-31 20:02:02,054 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3414.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:02:06,791 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3418.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:02:38,902 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3443.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:02:48,480 INFO [train.py:903] (0/4) Epoch 1, batch 3450, loss[loss=0.4488, simple_loss=0.4712, pruned_loss=0.2132, over 19658.00 frames. ], tot_loss[loss=0.4484, simple_loss=0.4545, pruned_loss=0.2211, over 3812897.10 frames. ], batch size: 58, lr: 4.54e-02, grad_scale: 8.0 +2023-03-31 20:02:50,748 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-03-31 20:03:43,982 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3496.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:03:50,182 INFO [train.py:903] (0/4) Epoch 1, batch 3500, loss[loss=0.4489, simple_loss=0.4651, pruned_loss=0.2164, over 19662.00 frames. ], tot_loss[loss=0.4509, simple_loss=0.456, pruned_loss=0.2229, over 3803575.46 frames. ], batch size: 60, lr: 4.53e-02, grad_scale: 8.0 +2023-03-31 20:03:56,677 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.496e+02 9.703e+02 1.213e+03 1.703e+03 9.610e+03, threshold=2.427e+03, percent-clipped=9.0 +2023-03-31 20:04:48,394 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-31 20:04:52,060 INFO [train.py:903] (0/4) Epoch 1, batch 3550, loss[loss=0.5704, simple_loss=0.5257, pruned_loss=0.3076, over 13170.00 frames. ], tot_loss[loss=0.4486, simple_loss=0.4547, pruned_loss=0.2213, over 3793975.90 frames. ], batch size: 137, lr: 4.51e-02, grad_scale: 8.0 +2023-03-31 20:05:00,240 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3557.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 20:05:07,862 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3564.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:05:16,035 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.46 vs. limit=5.0 +2023-03-31 20:05:30,977 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3582.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:05:53,929 INFO [train.py:903] (0/4) Epoch 1, batch 3600, loss[loss=0.4809, simple_loss=0.4817, pruned_loss=0.24, over 19364.00 frames. ], tot_loss[loss=0.4588, simple_loss=0.4609, pruned_loss=0.2284, over 3806675.61 frames. ], batch size: 70, lr: 4.50e-02, grad_scale: 8.0 +2023-03-31 20:06:00,961 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.467e+02 9.459e+02 1.417e+03 1.964e+03 2.103e+04, threshold=2.834e+03, percent-clipped=17.0 +2023-03-31 20:06:39,631 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7945, 1.8325, 1.6856, 1.1837, 1.0480, 1.5303, 0.3630, 1.0178], + device='cuda:0'), covar=tensor([0.0904, 0.0401, 0.0555, 0.1044, 0.1163, 0.1163, 0.2374, 0.1593], + device='cuda:0'), in_proj_covar=tensor([0.0104, 0.0100, 0.0115, 0.0136, 0.0145, 0.0141, 0.0168, 0.0158], + device='cuda:0'), out_proj_covar=tensor([8.7750e-05, 7.6651e-05, 8.8901e-05, 1.0320e-04, 1.1117e-04, 1.0574e-04, + 1.2572e-04, 1.2421e-04], device='cuda:0') +2023-03-31 20:06:55,026 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3650.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:06:55,794 INFO [train.py:903] (0/4) Epoch 1, batch 3650, loss[loss=0.5615, simple_loss=0.5245, pruned_loss=0.2993, over 19669.00 frames. ], tot_loss[loss=0.4577, simple_loss=0.4608, pruned_loss=0.2272, over 3807279.50 frames. ], batch size: 59, lr: 4.49e-02, grad_scale: 8.0 +2023-03-31 20:07:18,704 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3670.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:07:40,586 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-03-31 20:07:49,495 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3694.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 20:07:50,816 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3695.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:07:56,860 INFO [train.py:903] (0/4) Epoch 1, batch 3700, loss[loss=0.4792, simple_loss=0.4771, pruned_loss=0.2407, over 19313.00 frames. ], tot_loss[loss=0.4609, simple_loss=0.4626, pruned_loss=0.2296, over 3815459.11 frames. ], batch size: 66, lr: 4.48e-02, grad_scale: 8.0 +2023-03-31 20:08:05,842 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.255e+02 1.022e+03 1.666e+03 2.666e+03 1.441e+04, threshold=3.331e+03, percent-clipped=22.0 +2023-03-31 20:09:01,162 INFO [train.py:903] (0/4) Epoch 1, batch 3750, loss[loss=0.4905, simple_loss=0.5051, pruned_loss=0.2379, over 19046.00 frames. ], tot_loss[loss=0.4577, simple_loss=0.4609, pruned_loss=0.2273, over 3817826.15 frames. ], batch size: 69, lr: 4.47e-02, grad_scale: 8.0 +2023-03-31 20:09:02,741 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=3752.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:09:19,346 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=3765.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:09:34,317 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=3777.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:09:53,230 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8234, 1.9141, 1.3393, 2.1012, 1.6666, 0.8033, 1.0295, 1.3723], + device='cuda:0'), covar=tensor([0.1035, 0.0841, 0.1189, 0.0847, 0.1220, 0.1387, 0.2485, 0.1627], + device='cuda:0'), in_proj_covar=tensor([0.0104, 0.0091, 0.0127, 0.0137, 0.0142, 0.0081, 0.0135, 0.0122], + device='cuda:0'), out_proj_covar=tensor([7.4978e-05, 6.7821e-05, 8.5290e-05, 9.6979e-05, 9.6875e-05, 5.1451e-05, + 1.0229e-04, 8.7621e-05], device='cuda:0') +2023-03-31 20:09:57,039 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-31 20:10:05,361 INFO [train.py:903] (0/4) Epoch 1, batch 3800, loss[loss=0.4145, simple_loss=0.4436, pruned_loss=0.1927, over 19651.00 frames. ], tot_loss[loss=0.4532, simple_loss=0.4583, pruned_loss=0.2241, over 3829235.14 frames. ], batch size: 58, lr: 4.46e-02, grad_scale: 8.0 +2023-03-31 20:10:12,575 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.844e+02 1.035e+03 1.394e+03 1.973e+03 4.112e+03, threshold=2.788e+03, percent-clipped=1.0 +2023-03-31 20:10:15,317 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2520, 1.0461, 1.1187, 1.3175, 1.2756, 1.3372, 1.2687, 1.2914], + device='cuda:0'), covar=tensor([0.1123, 0.1880, 0.1760, 0.1544, 0.2111, 0.1820, 0.2026, 0.1329], + device='cuda:0'), in_proj_covar=tensor([0.0218, 0.0274, 0.0276, 0.0294, 0.0378, 0.0266, 0.0330, 0.0244], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 20:10:41,863 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-03-31 20:11:08,747 INFO [train.py:903] (0/4) Epoch 1, batch 3850, loss[loss=0.4337, simple_loss=0.4544, pruned_loss=0.2065, over 19534.00 frames. ], tot_loss[loss=0.4476, simple_loss=0.4544, pruned_loss=0.2204, over 3843992.31 frames. ], batch size: 54, lr: 4.45e-02, grad_scale: 8.0 +2023-03-31 20:11:15,011 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9493, 1.6075, 1.2469, 1.3864, 1.0445, 1.2175, 0.6264, 1.0840], + device='cuda:0'), covar=tensor([0.0786, 0.0641, 0.0610, 0.0882, 0.1408, 0.1246, 0.2475, 0.1556], + device='cuda:0'), in_proj_covar=tensor([0.0110, 0.0113, 0.0122, 0.0146, 0.0160, 0.0153, 0.0180, 0.0173], + device='cuda:0'), out_proj_covar=tensor([9.5432e-05, 8.7071e-05, 9.6328e-05, 1.1030e-04, 1.2398e-04, 1.1514e-04, + 1.3441e-04, 1.3575e-04], device='cuda:0') +2023-03-31 20:12:13,123 INFO [train.py:903] (0/4) Epoch 1, batch 3900, loss[loss=0.4476, simple_loss=0.4553, pruned_loss=0.22, over 19510.00 frames. ], tot_loss[loss=0.445, simple_loss=0.4525, pruned_loss=0.2187, over 3845526.11 frames. ], batch size: 64, lr: 4.44e-02, grad_scale: 8.0 +2023-03-31 20:12:22,004 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.917e+02 1.152e+03 1.441e+03 1.935e+03 3.736e+03, threshold=2.883e+03, percent-clipped=2.0 +2023-03-31 20:12:23,355 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=3908.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:13:07,839 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=3944.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:13:18,194 INFO [train.py:903] (0/4) Epoch 1, batch 3950, loss[loss=0.3641, simple_loss=0.3992, pruned_loss=0.1645, over 19629.00 frames. ], tot_loss[loss=0.4471, simple_loss=0.4535, pruned_loss=0.2204, over 3845411.23 frames. ], batch size: 50, lr: 4.43e-02, grad_scale: 8.0 +2023-03-31 20:13:23,976 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-03-31 20:14:21,835 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-4000.pt +2023-03-31 20:14:23,994 INFO [train.py:903] (0/4) Epoch 1, batch 4000, loss[loss=0.457, simple_loss=0.4578, pruned_loss=0.2281, over 13288.00 frames. ], tot_loss[loss=0.4425, simple_loss=0.4505, pruned_loss=0.2173, over 3830160.22 frames. ], batch size: 136, lr: 4.42e-02, grad_scale: 8.0 +2023-03-31 20:14:30,937 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.942e+02 1.053e+03 1.358e+03 1.948e+03 9.883e+03, threshold=2.717e+03, percent-clipped=12.0 +2023-03-31 20:14:49,301 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=4021.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 20:14:51,511 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4023.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:15:11,863 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=4038.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:15:12,778 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-03-31 20:15:22,641 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=4046.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 20:15:28,266 INFO [train.py:903] (0/4) Epoch 1, batch 4050, loss[loss=0.3981, simple_loss=0.4169, pruned_loss=0.1896, over 19535.00 frames. ], tot_loss[loss=0.4397, simple_loss=0.4481, pruned_loss=0.2156, over 3811991.79 frames. ], batch size: 54, lr: 4.41e-02, grad_scale: 8.0 +2023-03-31 20:16:32,898 INFO [train.py:903] (0/4) Epoch 1, batch 4100, loss[loss=0.4153, simple_loss=0.4401, pruned_loss=0.1952, over 19724.00 frames. ], tot_loss[loss=0.4398, simple_loss=0.4485, pruned_loss=0.2156, over 3806729.48 frames. ], batch size: 63, lr: 4.40e-02, grad_scale: 8.0 +2023-03-31 20:16:41,803 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.649e+02 1.198e+03 1.458e+03 1.833e+03 3.490e+03, threshold=2.915e+03, percent-clipped=3.0 +2023-03-31 20:17:08,499 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-03-31 20:17:38,861 INFO [train.py:903] (0/4) Epoch 1, batch 4150, loss[loss=0.3858, simple_loss=0.4108, pruned_loss=0.1804, over 19871.00 frames. ], tot_loss[loss=0.4384, simple_loss=0.4476, pruned_loss=0.2146, over 3793583.29 frames. ], batch size: 52, lr: 4.39e-02, grad_scale: 8.0 +2023-03-31 20:17:41,670 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4153.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:17:49,786 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4159.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:17:58,237 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6162, 1.5301, 1.6696, 2.5717, 3.3161, 1.3906, 2.1397, 3.3681], + device='cuda:0'), covar=tensor([0.0351, 0.2607, 0.2908, 0.1589, 0.0353, 0.2841, 0.1245, 0.0366], + device='cuda:0'), in_proj_covar=tensor([0.0165, 0.0263, 0.0254, 0.0264, 0.0190, 0.0321, 0.0226, 0.0199], + device='cuda:0'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:18:06,904 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0891, 1.9513, 1.5685, 2.2938, 1.7963, 0.5663, 1.1694, 1.5231], + device='cuda:0'), covar=tensor([0.1112, 0.0988, 0.1041, 0.0960, 0.1420, 0.1938, 0.2671, 0.1530], + device='cuda:0'), in_proj_covar=tensor([0.0091, 0.0078, 0.0107, 0.0114, 0.0119, 0.0080, 0.0122, 0.0104], + device='cuda:0'), out_proj_covar=tensor([6.3573e-05, 5.3425e-05, 6.8721e-05, 7.5055e-05, 7.5545e-05, 4.9392e-05, + 9.0291e-05, 7.0564e-05], device='cuda:0') +2023-03-31 20:18:14,049 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.24 vs. limit=5.0 +2023-03-31 20:18:44,276 INFO [train.py:903] (0/4) Epoch 1, batch 4200, loss[loss=0.4719, simple_loss=0.4829, pruned_loss=0.2305, over 19494.00 frames. ], tot_loss[loss=0.4342, simple_loss=0.4449, pruned_loss=0.2117, over 3796359.61 frames. ], batch size: 64, lr: 4.38e-02, grad_scale: 8.0 +2023-03-31 20:18:46,652 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-03-31 20:18:51,465 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.163e+02 8.919e+02 1.098e+03 1.489e+03 3.268e+03, threshold=2.196e+03, percent-clipped=3.0 +2023-03-31 20:19:43,641 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.66 vs. limit=2.0 +2023-03-31 20:19:47,735 INFO [train.py:903] (0/4) Epoch 1, batch 4250, loss[loss=0.482, simple_loss=0.4817, pruned_loss=0.2412, over 19778.00 frames. ], tot_loss[loss=0.4349, simple_loss=0.4453, pruned_loss=0.2123, over 3785945.18 frames. ], batch size: 56, lr: 4.36e-02, grad_scale: 8.0 +2023-03-31 20:20:02,077 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-03-31 20:20:15,225 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-03-31 20:20:25,100 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=4279.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 20:20:36,567 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=4288.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:20:52,968 INFO [train.py:903] (0/4) Epoch 1, batch 4300, loss[loss=0.3873, simple_loss=0.4076, pruned_loss=0.1835, over 19628.00 frames. ], tot_loss[loss=0.4335, simple_loss=0.4439, pruned_loss=0.2115, over 3795109.33 frames. ], batch size: 50, lr: 4.35e-02, grad_scale: 8.0 +2023-03-31 20:20:57,802 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=4304.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 20:21:00,179 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=4306.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:21:02,056 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.021e+02 1.171e+03 1.478e+03 2.100e+03 3.660e+03, threshold=2.957e+03, percent-clipped=20.0 +2023-03-31 20:21:44,163 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6484, 3.3520, 1.9167, 3.0791, 1.2756, 3.2982, 3.0312, 2.9783], + device='cuda:0'), covar=tensor([0.0633, 0.1190, 0.2275, 0.0624, 0.3180, 0.0753, 0.0558, 0.0833], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0231, 0.0256, 0.0205, 0.0273, 0.0207, 0.0155, 0.0171], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0001], + device='cuda:0') +2023-03-31 20:21:46,488 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-03-31 20:21:59,415 INFO [train.py:903] (0/4) Epoch 1, batch 4350, loss[loss=0.4244, simple_loss=0.443, pruned_loss=0.2028, over 19535.00 frames. ], tot_loss[loss=0.4308, simple_loss=0.4429, pruned_loss=0.2094, over 3793305.98 frames. ], batch size: 56, lr: 4.34e-02, grad_scale: 8.0 +2023-03-31 20:22:20,197 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7679, 1.1824, 1.4409, 0.7814, 2.6004, 2.8400, 2.7358, 2.6387], + device='cuda:0'), covar=tensor([0.1636, 0.2831, 0.2278, 0.2904, 0.0490, 0.0183, 0.0263, 0.0254], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0269, 0.0279, 0.0304, 0.0208, 0.0127, 0.0158, 0.0121], + device='cuda:0'), out_proj_covar=tensor([2.2508e-04, 2.0494e-04, 2.1278e-04, 2.3036e-04, 1.8519e-04, 9.1116e-05, + 1.2465e-04, 1.0075e-04], device='cuda:0') +2023-03-31 20:23:03,132 INFO [train.py:903] (0/4) Epoch 1, batch 4400, loss[loss=0.4095, simple_loss=0.4339, pruned_loss=0.1925, over 19576.00 frames. ], tot_loss[loss=0.4284, simple_loss=0.4414, pruned_loss=0.2077, over 3799435.99 frames. ], batch size: 61, lr: 4.33e-02, grad_scale: 8.0 +2023-03-31 20:23:05,845 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4403.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:23:11,485 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.454e+02 9.391e+02 1.114e+03 1.514e+03 3.216e+03, threshold=2.228e+03, percent-clipped=1.0 +2023-03-31 20:23:14,571 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=4409.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 20:23:30,201 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-03-31 20:23:39,401 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-03-31 20:23:46,823 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=4434.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:24:06,851 INFO [train.py:903] (0/4) Epoch 1, batch 4450, loss[loss=0.5332, simple_loss=0.5091, pruned_loss=0.2787, over 13491.00 frames. ], tot_loss[loss=0.4282, simple_loss=0.4413, pruned_loss=0.2076, over 3798897.60 frames. ], batch size: 136, lr: 4.32e-02, grad_scale: 8.0 +2023-03-31 20:24:10,359 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5166, 1.2301, 1.2535, 1.6075, 1.3156, 1.4140, 1.2663, 1.5917], + device='cuda:0'), covar=tensor([0.1231, 0.2377, 0.1877, 0.1303, 0.2049, 0.1053, 0.1837, 0.1008], + device='cuda:0'), in_proj_covar=tensor([0.0185, 0.0276, 0.0246, 0.0208, 0.0276, 0.0201, 0.0226, 0.0184], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001], + device='cuda:0') +2023-03-31 20:24:29,565 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-31 20:25:09,799 INFO [train.py:903] (0/4) Epoch 1, batch 4500, loss[loss=0.3622, simple_loss=0.3919, pruned_loss=0.1662, over 19593.00 frames. ], tot_loss[loss=0.4274, simple_loss=0.4413, pruned_loss=0.2067, over 3819646.46 frames. ], batch size: 52, lr: 4.31e-02, grad_scale: 8.0 +2023-03-31 20:25:12,517 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=4503.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:25:18,130 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.500e+02 1.049e+03 1.357e+03 1.620e+03 3.962e+03, threshold=2.713e+03, percent-clipped=8.0 +2023-03-31 20:26:14,040 INFO [train.py:903] (0/4) Epoch 1, batch 4550, loss[loss=0.4148, simple_loss=0.4336, pruned_loss=0.198, over 18699.00 frames. ], tot_loss[loss=0.4238, simple_loss=0.4387, pruned_loss=0.2045, over 3823733.93 frames. ], batch size: 74, lr: 4.30e-02, grad_scale: 8.0 +2023-03-31 20:26:24,214 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-03-31 20:26:47,306 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-03-31 20:27:04,912 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4705, 0.9841, 1.3617, 1.7375, 2.1825, 1.3682, 1.8597, 2.0524], + device='cuda:0'), covar=tensor([0.0375, 0.2571, 0.2437, 0.1309, 0.0440, 0.1790, 0.0755, 0.0656], + device='cuda:0'), in_proj_covar=tensor([0.0173, 0.0267, 0.0256, 0.0258, 0.0194, 0.0315, 0.0225, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:27:16,034 INFO [train.py:903] (0/4) Epoch 1, batch 4600, loss[loss=0.4428, simple_loss=0.4605, pruned_loss=0.2125, over 18792.00 frames. ], tot_loss[loss=0.4244, simple_loss=0.4392, pruned_loss=0.2048, over 3810891.59 frames. ], batch size: 75, lr: 4.29e-02, grad_scale: 4.0 +2023-03-31 20:27:24,059 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.210e+02 9.691e+02 1.279e+03 1.723e+03 8.130e+03, threshold=2.557e+03, percent-clipped=7.0 +2023-03-31 20:27:36,627 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4618.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:27:54,576 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6032, 2.6742, 2.1692, 2.9033, 1.9958, 2.5664, 2.1711, 1.9642], + device='cuda:0'), covar=tensor([0.1073, 0.0855, 0.0894, 0.0909, 0.1394, 0.0771, 0.2022, 0.1241], + device='cuda:0'), in_proj_covar=tensor([0.0124, 0.0108, 0.0138, 0.0154, 0.0162, 0.0090, 0.0163, 0.0133], + device='cuda:0'), out_proj_covar=tensor([8.7656e-05, 7.4996e-05, 9.1657e-05, 1.0148e-04, 1.0391e-04, 5.5804e-05, + 1.1812e-04, 9.0435e-05], device='cuda:0') +2023-03-31 20:28:17,144 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=4650.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:28:18,171 INFO [train.py:903] (0/4) Epoch 1, batch 4650, loss[loss=0.4609, simple_loss=0.4698, pruned_loss=0.2259, over 19611.00 frames. ], tot_loss[loss=0.4224, simple_loss=0.4376, pruned_loss=0.2036, over 3813322.71 frames. ], batch size: 61, lr: 4.28e-02, grad_scale: 4.0 +2023-03-31 20:28:27,851 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=4659.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:28:33,642 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5965, 4.2585, 2.2465, 3.6460, 1.5327, 3.9913, 3.7255, 3.9716], + device='cuda:0'), covar=tensor([0.0534, 0.1048, 0.2426, 0.0659, 0.3131, 0.0748, 0.0538, 0.0567], + device='cuda:0'), in_proj_covar=tensor([0.0229, 0.0226, 0.0268, 0.0214, 0.0281, 0.0218, 0.0164, 0.0171], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001], + device='cuda:0') +2023-03-31 20:28:34,610 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-03-31 20:28:44,788 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-03-31 20:28:58,639 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=4684.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:29:18,946 INFO [train.py:903] (0/4) Epoch 1, batch 4700, loss[loss=0.4106, simple_loss=0.4279, pruned_loss=0.1967, over 19512.00 frames. ], tot_loss[loss=0.4202, simple_loss=0.4361, pruned_loss=0.2022, over 3811753.40 frames. ], batch size: 56, lr: 4.27e-02, grad_scale: 4.0 +2023-03-31 20:29:27,997 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.618e+02 9.658e+02 1.202e+03 1.526e+03 2.859e+03, threshold=2.405e+03, percent-clipped=1.0 +2023-03-31 20:29:39,223 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-03-31 20:30:21,813 INFO [train.py:903] (0/4) Epoch 1, batch 4750, loss[loss=0.4233, simple_loss=0.4349, pruned_loss=0.2058, over 19708.00 frames. ], tot_loss[loss=0.4171, simple_loss=0.4335, pruned_loss=0.2004, over 3808871.83 frames. ], batch size: 63, lr: 4.26e-02, grad_scale: 4.0 +2023-03-31 20:30:39,317 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=4765.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:31:23,846 INFO [train.py:903] (0/4) Epoch 1, batch 4800, loss[loss=0.4119, simple_loss=0.4259, pruned_loss=0.199, over 17561.00 frames. ], tot_loss[loss=0.4188, simple_loss=0.4351, pruned_loss=0.2012, over 3813173.89 frames. ], batch size: 101, lr: 4.25e-02, grad_scale: 8.0 +2023-03-31 20:31:32,985 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 7.504e+02 1.038e+03 1.224e+03 1.522e+03 3.175e+03, threshold=2.447e+03, percent-clipped=5.0 +2023-03-31 20:32:13,379 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.1207, 3.7950, 1.9968, 3.4665, 1.4572, 3.5450, 3.3186, 3.6026], + device='cuda:0'), covar=tensor([0.0634, 0.1266, 0.2606, 0.0705, 0.3658, 0.1028, 0.0684, 0.0671], + device='cuda:0'), in_proj_covar=tensor([0.0231, 0.0236, 0.0264, 0.0214, 0.0288, 0.0220, 0.0167, 0.0172], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001], + device='cuda:0') +2023-03-31 20:32:24,211 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-03-31 20:32:25,500 INFO [train.py:903] (0/4) Epoch 1, batch 4850, loss[loss=0.3844, simple_loss=0.3937, pruned_loss=0.1876, over 19394.00 frames. ], tot_loss[loss=0.4169, simple_loss=0.4334, pruned_loss=0.2002, over 3826733.91 frames. ], batch size: 48, lr: 4.24e-02, grad_scale: 8.0 +2023-03-31 20:32:46,763 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-03-31 20:32:54,326 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=4874.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:33:04,558 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4470, 1.7365, 1.5310, 1.4395, 1.4845, 1.0019, 1.0925, 1.4791], + device='cuda:0'), covar=tensor([0.1087, 0.0445, 0.0739, 0.0932, 0.0873, 0.1373, 0.1353, 0.0908], + device='cuda:0'), in_proj_covar=tensor([0.0247, 0.0149, 0.0176, 0.0212, 0.0152, 0.0237, 0.0244, 0.0230], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:33:06,426 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-03-31 20:33:12,764 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-03-31 20:33:12,785 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-03-31 20:33:23,799 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-03-31 20:33:25,317 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=4899.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:33:27,241 INFO [train.py:903] (0/4) Epoch 1, batch 4900, loss[loss=0.3553, simple_loss=0.3906, pruned_loss=0.16, over 19665.00 frames. ], tot_loss[loss=0.4137, simple_loss=0.431, pruned_loss=0.1982, over 3821467.67 frames. ], batch size: 53, lr: 4.23e-02, grad_scale: 8.0 +2023-03-31 20:33:36,363 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8082, 1.7053, 1.3147, 1.3626, 1.4245, 1.2762, 0.2733, 1.0140], + device='cuda:0'), covar=tensor([0.0753, 0.0561, 0.0436, 0.0715, 0.0916, 0.0835, 0.1712, 0.1364], + device='cuda:0'), in_proj_covar=tensor([0.0141, 0.0145, 0.0144, 0.0182, 0.0206, 0.0188, 0.0211, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:33:37,035 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.319e+02 9.845e+02 1.167e+03 1.485e+03 2.856e+03, threshold=2.333e+03, percent-clipped=2.0 +2023-03-31 20:33:44,813 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-03-31 20:34:29,508 INFO [train.py:903] (0/4) Epoch 1, batch 4950, loss[loss=0.5644, simple_loss=0.5311, pruned_loss=0.2989, over 19768.00 frames. ], tot_loss[loss=0.4134, simple_loss=0.431, pruned_loss=0.1979, over 3815176.53 frames. ], batch size: 56, lr: 4.21e-02, grad_scale: 8.0 +2023-03-31 20:34:38,643 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4203, 1.3842, 1.1946, 1.3709, 1.2744, 1.4526, 1.2536, 1.4455], + device='cuda:0'), covar=tensor([0.0815, 0.1358, 0.1327, 0.0970, 0.1575, 0.0709, 0.1199, 0.0674], + device='cuda:0'), in_proj_covar=tensor([0.0205, 0.0307, 0.0262, 0.0224, 0.0293, 0.0219, 0.0247, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:34:40,549 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-03-31 20:35:05,330 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-31 20:35:05,872 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-03-31 20:35:31,811 INFO [train.py:903] (0/4) Epoch 1, batch 5000, loss[loss=0.3839, simple_loss=0.3992, pruned_loss=0.1843, over 19741.00 frames. ], tot_loss[loss=0.4119, simple_loss=0.4299, pruned_loss=0.1969, over 3813177.26 frames. ], batch size: 45, lr: 4.20e-02, grad_scale: 8.0 +2023-03-31 20:35:35,556 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-03-31 20:35:40,125 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.692e+02 8.720e+02 1.063e+03 1.451e+03 3.452e+03, threshold=2.125e+03, percent-clipped=4.0 +2023-03-31 20:35:46,918 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-03-31 20:35:56,135 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=5021.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:36:18,949 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3566, 3.8898, 2.4864, 3.5852, 1.2670, 3.8963, 3.4202, 3.4909], + device='cuda:0'), covar=tensor([0.0770, 0.1444, 0.2394, 0.0800, 0.3853, 0.0841, 0.0747, 0.0692], + device='cuda:0'), in_proj_covar=tensor([0.0239, 0.0242, 0.0270, 0.0223, 0.0288, 0.0223, 0.0163, 0.0179], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001], + device='cuda:0') +2023-03-31 20:36:27,130 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=5046.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:36:33,279 INFO [train.py:903] (0/4) Epoch 1, batch 5050, loss[loss=0.4203, simple_loss=0.4476, pruned_loss=0.1965, over 19664.00 frames. ], tot_loss[loss=0.4114, simple_loss=0.4301, pruned_loss=0.1963, over 3826132.29 frames. ], batch size: 55, lr: 4.19e-02, grad_scale: 8.0 +2023-03-31 20:36:41,705 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7763, 2.1753, 2.4870, 2.1339, 1.8227, 2.1265, 0.3707, 1.7279], + device='cuda:0'), covar=tensor([0.1138, 0.0663, 0.0320, 0.0710, 0.1043, 0.0880, 0.2108, 0.1881], + device='cuda:0'), in_proj_covar=tensor([0.0138, 0.0144, 0.0147, 0.0181, 0.0209, 0.0191, 0.0213, 0.0209], + device='cuda:0'), out_proj_covar=tensor([0.0001, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:37:02,502 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-03-31 20:37:34,522 INFO [train.py:903] (0/4) Epoch 1, batch 5100, loss[loss=0.4921, simple_loss=0.4828, pruned_loss=0.2507, over 19658.00 frames. ], tot_loss[loss=0.4101, simple_loss=0.4291, pruned_loss=0.1956, over 3826830.44 frames. ], batch size: 60, lr: 4.18e-02, grad_scale: 8.0 +2023-03-31 20:37:39,766 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-03-31 20:37:43,116 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.551e+02 1.065e+03 1.254e+03 1.490e+03 3.647e+03, threshold=2.509e+03, percent-clipped=6.0 +2023-03-31 20:37:43,189 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-03-31 20:37:47,668 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-03-31 20:38:26,587 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2238, 3.3012, 3.6972, 3.4799, 1.1726, 3.2820, 2.9811, 3.1371], + device='cuda:0'), covar=tensor([0.0328, 0.0497, 0.0418, 0.0285, 0.3247, 0.0249, 0.0421, 0.1002], + device='cuda:0'), in_proj_covar=tensor([0.0174, 0.0224, 0.0273, 0.0191, 0.0366, 0.0137, 0.0201, 0.0284], + device='cuda:0'), out_proj_covar=tensor([1.1615e-04, 1.4757e-04, 1.8180e-04, 1.1621e-04, 2.0351e-04, 9.0705e-05, + 1.2792e-04, 1.6713e-04], device='cuda:0') +2023-03-31 20:38:33,532 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8662, 1.3250, 1.5369, 1.0470, 2.7291, 3.1549, 3.1207, 3.4918], + device='cuda:0'), covar=tensor([0.1733, 0.2778, 0.2477, 0.2731, 0.0537, 0.0198, 0.0232, 0.0178], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0277, 0.0294, 0.0304, 0.0205, 0.0124, 0.0171, 0.0126], + device='cuda:0'), out_proj_covar=tensor([2.3973e-04, 2.2267e-04, 2.3410e-04, 2.4378e-04, 1.8999e-04, 9.8038e-05, + 1.4029e-04, 1.1110e-04], device='cuda:0') +2023-03-31 20:38:36,973 INFO [train.py:903] (0/4) Epoch 1, batch 5150, loss[loss=0.3708, simple_loss=0.3904, pruned_loss=0.1756, over 15154.00 frames. ], tot_loss[loss=0.4103, simple_loss=0.429, pruned_loss=0.1957, over 3827175.25 frames. ], batch size: 33, lr: 4.17e-02, grad_scale: 8.0 +2023-03-31 20:38:46,431 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-03-31 20:39:08,042 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-03-31 20:39:20,325 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-03-31 20:39:33,876 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-31 20:39:37,961 INFO [train.py:903] (0/4) Epoch 1, batch 5200, loss[loss=0.3755, simple_loss=0.3922, pruned_loss=0.1794, over 19720.00 frames. ], tot_loss[loss=0.4104, simple_loss=0.4289, pruned_loss=0.196, over 3832722.06 frames. ], batch size: 46, lr: 4.16e-02, grad_scale: 8.0 +2023-03-31 20:39:45,882 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.640e+02 1.028e+03 1.252e+03 1.630e+03 4.880e+03, threshold=2.504e+03, percent-clipped=1.0 +2023-03-31 20:39:50,382 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-03-31 20:40:32,320 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-03-31 20:40:39,295 INFO [train.py:903] (0/4) Epoch 1, batch 5250, loss[loss=0.4409, simple_loss=0.4598, pruned_loss=0.211, over 18801.00 frames. ], tot_loss[loss=0.4099, simple_loss=0.4288, pruned_loss=0.1955, over 3828000.66 frames. ], batch size: 74, lr: 4.15e-02, grad_scale: 8.0 +2023-03-31 20:41:24,115 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9224, 4.3564, 5.5941, 5.2400, 1.5825, 5.0440, 4.6878, 4.7335], + device='cuda:0'), covar=tensor([0.0237, 0.0470, 0.0395, 0.0254, 0.3654, 0.0199, 0.0340, 0.0995], + device='cuda:0'), in_proj_covar=tensor([0.0182, 0.0228, 0.0281, 0.0197, 0.0368, 0.0142, 0.0206, 0.0293], + device='cuda:0'), out_proj_covar=tensor([1.2050e-04, 1.5085e-04, 1.8654e-04, 1.1817e-04, 2.0418e-04, 9.2051e-05, + 1.3018e-04, 1.7155e-04], device='cuda:0') +2023-03-31 20:41:39,595 INFO [train.py:903] (0/4) Epoch 1, batch 5300, loss[loss=0.4195, simple_loss=0.4456, pruned_loss=0.1967, over 19677.00 frames. ], tot_loss[loss=0.409, simple_loss=0.4281, pruned_loss=0.1949, over 3832457.56 frames. ], batch size: 60, lr: 4.14e-02, grad_scale: 8.0 +2023-03-31 20:41:48,685 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.962e+02 9.394e+02 1.191e+03 1.647e+03 4.206e+03, threshold=2.383e+03, percent-clipped=5.0 +2023-03-31 20:41:53,347 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-03-31 20:42:06,837 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9401, 1.0561, 1.5918, 1.2522, 2.2098, 2.2189, 2.3324, 1.4211], + device='cuda:0'), covar=tensor([0.1759, 0.2019, 0.1410, 0.1716, 0.0830, 0.0770, 0.0846, 0.1518], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0252, 0.0245, 0.0273, 0.0281, 0.0214, 0.0290, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:42:41,824 INFO [train.py:903] (0/4) Epoch 1, batch 5350, loss[loss=0.3744, simple_loss=0.4162, pruned_loss=0.1663, over 19675.00 frames. ], tot_loss[loss=0.4077, simple_loss=0.4275, pruned_loss=0.194, over 3836743.67 frames. ], batch size: 55, lr: 4.13e-02, grad_scale: 8.0 +2023-03-31 20:43:07,200 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9308, 1.4029, 1.6641, 1.3428, 2.6490, 3.1896, 2.9271, 3.1747], + device='cuda:0'), covar=tensor([0.1500, 0.2553, 0.2096, 0.2338, 0.0526, 0.0141, 0.0265, 0.0193], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0276, 0.0292, 0.0299, 0.0199, 0.0118, 0.0173, 0.0125], + device='cuda:0'), out_proj_covar=tensor([2.3952e-04, 2.2431e-04, 2.3378e-04, 2.4216e-04, 1.8555e-04, 9.3408e-05, + 1.4158e-04, 1.1072e-04], device='cuda:0') +2023-03-31 20:43:13,707 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-03-31 20:43:43,593 INFO [train.py:903] (0/4) Epoch 1, batch 5400, loss[loss=0.3997, simple_loss=0.4262, pruned_loss=0.1866, over 19763.00 frames. ], tot_loss[loss=0.4093, simple_loss=0.4288, pruned_loss=0.1949, over 3836785.10 frames. ], batch size: 54, lr: 4.12e-02, grad_scale: 8.0 +2023-03-31 20:43:51,063 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.073e+02 9.364e+02 1.084e+03 1.611e+03 4.795e+03, threshold=2.168e+03, percent-clipped=7.0 +2023-03-31 20:44:38,333 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-03-31 20:44:44,645 INFO [train.py:903] (0/4) Epoch 1, batch 5450, loss[loss=0.3671, simple_loss=0.4071, pruned_loss=0.1635, over 19660.00 frames. ], tot_loss[loss=0.4078, simple_loss=0.4275, pruned_loss=0.194, over 3822688.65 frames. ], batch size: 55, lr: 4.11e-02, grad_scale: 8.0 +2023-03-31 20:45:33,606 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9011, 0.9721, 1.4539, 1.0347, 2.0206, 1.8432, 1.8709, 1.0512], + device='cuda:0'), covar=tensor([0.1886, 0.2000, 0.1360, 0.1877, 0.0710, 0.0857, 0.0768, 0.1479], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0262, 0.0250, 0.0280, 0.0289, 0.0223, 0.0301, 0.0282], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:45:46,527 INFO [train.py:903] (0/4) Epoch 1, batch 5500, loss[loss=0.3573, simple_loss=0.3944, pruned_loss=0.1601, over 19743.00 frames. ], tot_loss[loss=0.4074, simple_loss=0.4272, pruned_loss=0.1938, over 3824435.73 frames. ], batch size: 51, lr: 4.10e-02, grad_scale: 8.0 +2023-03-31 20:45:54,009 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.842e+02 9.504e+02 1.107e+03 1.412e+03 4.004e+03, threshold=2.214e+03, percent-clipped=7.0 +2023-03-31 20:46:08,674 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-03-31 20:46:46,625 INFO [train.py:903] (0/4) Epoch 1, batch 5550, loss[loss=0.3367, simple_loss=0.3739, pruned_loss=0.1497, over 18181.00 frames. ], tot_loss[loss=0.4062, simple_loss=0.4261, pruned_loss=0.1931, over 3827132.10 frames. ], batch size: 40, lr: 4.09e-02, grad_scale: 8.0 +2023-03-31 20:46:53,467 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-03-31 20:47:42,835 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-03-31 20:47:47,482 INFO [train.py:903] (0/4) Epoch 1, batch 5600, loss[loss=0.441, simple_loss=0.4545, pruned_loss=0.2138, over 18213.00 frames. ], tot_loss[loss=0.405, simple_loss=0.4253, pruned_loss=0.1923, over 3829139.35 frames. ], batch size: 83, lr: 4.08e-02, grad_scale: 8.0 +2023-03-31 20:47:56,555 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.862e+02 1.009e+03 1.185e+03 1.400e+03 2.216e+03, threshold=2.370e+03, percent-clipped=2.0 +2023-03-31 20:48:12,710 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4273, 2.0506, 2.3324, 2.1189, 3.0387, 3.6877, 3.3637, 3.6761], + device='cuda:0'), covar=tensor([0.1036, 0.1871, 0.1558, 0.1613, 0.0462, 0.0121, 0.0166, 0.0164], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0274, 0.0295, 0.0300, 0.0193, 0.0118, 0.0166, 0.0122], + device='cuda:0'), out_proj_covar=tensor([2.3614e-04, 2.2589e-04, 2.3904e-04, 2.4745e-04, 1.8182e-04, 9.6208e-05, + 1.3747e-04, 1.0952e-04], device='cuda:0') +2023-03-31 20:48:48,825 INFO [train.py:903] (0/4) Epoch 1, batch 5650, loss[loss=0.3308, simple_loss=0.3679, pruned_loss=0.1468, over 19377.00 frames. ], tot_loss[loss=0.4056, simple_loss=0.4255, pruned_loss=0.1929, over 3814040.42 frames. ], batch size: 47, lr: 4.07e-02, grad_scale: 8.0 +2023-03-31 20:49:08,755 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=5668.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 20:49:32,789 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-03-31 20:49:49,723 INFO [train.py:903] (0/4) Epoch 1, batch 5700, loss[loss=0.3851, simple_loss=0.4223, pruned_loss=0.1739, over 19505.00 frames. ], tot_loss[loss=0.4077, simple_loss=0.4271, pruned_loss=0.1941, over 3822422.22 frames. ], batch size: 64, lr: 4.06e-02, grad_scale: 8.0 +2023-03-31 20:49:57,492 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.193e+02 1.084e+03 1.385e+03 1.754e+03 4.325e+03, threshold=2.770e+03, percent-clipped=14.0 +2023-03-31 20:49:57,788 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1855, 3.1952, 3.6306, 3.4206, 1.2986, 3.0414, 2.8334, 3.0134], + device='cuda:0'), covar=tensor([0.0329, 0.0519, 0.0428, 0.0329, 0.3130, 0.0365, 0.0489, 0.0998], + device='cuda:0'), in_proj_covar=tensor([0.0184, 0.0223, 0.0271, 0.0194, 0.0367, 0.0142, 0.0205, 0.0287], + device='cuda:0'), out_proj_covar=tensor([1.2092e-04, 1.4406e-04, 1.7789e-04, 1.1421e-04, 2.0079e-04, 9.2285e-05, + 1.2764e-04, 1.6519e-04], device='cuda:0') +2023-03-31 20:50:51,579 INFO [train.py:903] (0/4) Epoch 1, batch 5750, loss[loss=0.4694, simple_loss=0.4633, pruned_loss=0.2377, over 13594.00 frames. ], tot_loss[loss=0.4046, simple_loss=0.4253, pruned_loss=0.192, over 3834142.32 frames. ], batch size: 135, lr: 4.05e-02, grad_scale: 8.0 +2023-03-31 20:50:51,602 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-03-31 20:50:59,677 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-03-31 20:51:06,045 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-03-31 20:51:18,541 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=5773.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:51:52,609 INFO [train.py:903] (0/4) Epoch 1, batch 5800, loss[loss=0.3816, simple_loss=0.428, pruned_loss=0.1676, over 19294.00 frames. ], tot_loss[loss=0.4021, simple_loss=0.4236, pruned_loss=0.1903, over 3843984.68 frames. ], batch size: 66, lr: 4.04e-02, grad_scale: 8.0 +2023-03-31 20:52:02,181 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.310e+02 8.969e+02 1.169e+03 1.352e+03 2.735e+03, threshold=2.337e+03, percent-clipped=0.0 +2023-03-31 20:52:16,012 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-03-31 20:52:53,453 INFO [train.py:903] (0/4) Epoch 1, batch 5850, loss[loss=0.4171, simple_loss=0.4466, pruned_loss=0.1937, over 19580.00 frames. ], tot_loss[loss=0.404, simple_loss=0.4251, pruned_loss=0.1915, over 3849203.83 frames. ], batch size: 61, lr: 4.03e-02, grad_scale: 8.0 +2023-03-31 20:53:23,809 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=5876.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:53:55,089 INFO [train.py:903] (0/4) Epoch 1, batch 5900, loss[loss=0.3919, simple_loss=0.4291, pruned_loss=0.1773, over 19617.00 frames. ], tot_loss[loss=0.4025, simple_loss=0.4243, pruned_loss=0.1903, over 3851201.67 frames. ], batch size: 57, lr: 4.02e-02, grad_scale: 8.0 +2023-03-31 20:53:58,589 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-03-31 20:54:03,114 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.633e+02 8.668e+02 1.127e+03 1.397e+03 3.736e+03, threshold=2.255e+03, percent-clipped=4.0 +2023-03-31 20:54:21,185 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-03-31 20:54:32,234 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4081, 2.4159, 1.8346, 1.6319, 1.8821, 0.8795, 0.9966, 1.6698], + device='cuda:0'), covar=tensor([0.1217, 0.0349, 0.0813, 0.0682, 0.0786, 0.1458, 0.1375, 0.0814], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0148, 0.0199, 0.0221, 0.0156, 0.0254, 0.0250, 0.0231], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:54:50,229 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0 +2023-03-31 20:54:55,344 INFO [train.py:903] (0/4) Epoch 1, batch 5950, loss[loss=0.3951, simple_loss=0.4225, pruned_loss=0.1839, over 19681.00 frames. ], tot_loss[loss=0.4027, simple_loss=0.4245, pruned_loss=0.1905, over 3846178.79 frames. ], batch size: 53, lr: 4.01e-02, grad_scale: 8.0 +2023-03-31 20:55:55,469 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-6000.pt +2023-03-31 20:55:57,492 INFO [train.py:903] (0/4) Epoch 1, batch 6000, loss[loss=0.4388, simple_loss=0.4688, pruned_loss=0.2044, over 19664.00 frames. ], tot_loss[loss=0.4042, simple_loss=0.4257, pruned_loss=0.1914, over 3843480.00 frames. ], batch size: 58, lr: 4.00e-02, grad_scale: 8.0 +2023-03-31 20:55:57,493 INFO [train.py:928] (0/4) Computing validation loss +2023-03-31 20:56:10,587 INFO [train.py:937] (0/4) Epoch 1, validation: loss=0.2784, simple_loss=0.3626, pruned_loss=0.09714, over 944034.00 frames. +2023-03-31 20:56:10,588 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18153MB +2023-03-31 20:56:19,577 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.849e+02 9.012e+02 1.240e+03 1.620e+03 2.952e+03, threshold=2.480e+03, percent-clipped=5.0 +2023-03-31 20:56:19,902 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6008.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:56:24,067 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6012.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 20:57:10,741 INFO [train.py:903] (0/4) Epoch 1, batch 6050, loss[loss=0.5052, simple_loss=0.4994, pruned_loss=0.2555, over 19546.00 frames. ], tot_loss[loss=0.4069, simple_loss=0.4274, pruned_loss=0.1932, over 3830296.51 frames. ], batch size: 61, lr: 3.99e-02, grad_scale: 8.0 +2023-03-31 20:57:24,316 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6061.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:57:59,147 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0033, 1.1159, 1.7429, 1.2978, 2.3970, 2.2370, 2.4764, 1.2654], + device='cuda:0'), covar=tensor([0.1252, 0.1489, 0.0996, 0.1218, 0.0598, 0.0600, 0.0649, 0.1239], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0284, 0.0265, 0.0288, 0.0308, 0.0245, 0.0330, 0.0300], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 20:58:12,892 INFO [train.py:903] (0/4) Epoch 1, batch 6100, loss[loss=0.3818, simple_loss=0.4163, pruned_loss=0.1737, over 19766.00 frames. ], tot_loss[loss=0.4046, simple_loss=0.4259, pruned_loss=0.1917, over 3834135.57 frames. ], batch size: 54, lr: 3.98e-02, grad_scale: 8.0 +2023-03-31 20:58:20,961 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.479e+02 9.509e+02 1.169e+03 1.489e+03 2.977e+03, threshold=2.338e+03, percent-clipped=4.0 +2023-03-31 20:58:32,474 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6117.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:58:44,702 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6127.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 20:59:13,360 INFO [train.py:903] (0/4) Epoch 1, batch 6150, loss[loss=0.4191, simple_loss=0.4473, pruned_loss=0.1954, over 19666.00 frames. ], tot_loss[loss=0.401, simple_loss=0.4232, pruned_loss=0.1893, over 3839302.54 frames. ], batch size: 60, lr: 3.97e-02, grad_scale: 8.0 +2023-03-31 20:59:15,665 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6153.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 20:59:28,307 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.63 vs. limit=2.0 +2023-03-31 20:59:42,671 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-03-31 20:59:52,046 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-31 20:59:58,491 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6188.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:00:13,786 INFO [train.py:903] (0/4) Epoch 1, batch 6200, loss[loss=0.3562, simple_loss=0.3973, pruned_loss=0.1576, over 19596.00 frames. ], tot_loss[loss=0.4021, simple_loss=0.4236, pruned_loss=0.1903, over 3822706.56 frames. ], batch size: 52, lr: 3.96e-02, grad_scale: 8.0 +2023-03-31 21:00:20,830 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8445, 1.3835, 1.2043, 1.7998, 1.2967, 1.8217, 1.7556, 1.6728], + device='cuda:0'), covar=tensor([0.0863, 0.1713, 0.1908, 0.1387, 0.2143, 0.1357, 0.1898, 0.1134], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0305, 0.0301, 0.0322, 0.0391, 0.0280, 0.0343, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 21:00:22,726 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.993e+02 9.585e+02 1.181e+03 1.511e+03 2.920e+03, threshold=2.362e+03, percent-clipped=2.0 +2023-03-31 21:00:38,370 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6220.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:00:39,666 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6221.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:00:41,250 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-03-31 21:00:49,011 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6229.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:00:52,526 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6232.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:01:16,025 INFO [train.py:903] (0/4) Epoch 1, batch 6250, loss[loss=0.4073, simple_loss=0.4277, pruned_loss=0.1935, over 19571.00 frames. ], tot_loss[loss=0.4008, simple_loss=0.4228, pruned_loss=0.1894, over 3820827.23 frames. ], batch size: 61, lr: 3.95e-02, grad_scale: 8.0 +2023-03-31 21:01:46,722 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-03-31 21:01:56,437 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6284.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:02:18,554 INFO [train.py:903] (0/4) Epoch 1, batch 6300, loss[loss=0.3701, simple_loss=0.399, pruned_loss=0.1706, over 19468.00 frames. ], tot_loss[loss=0.3976, simple_loss=0.4211, pruned_loss=0.1871, over 3813678.02 frames. ], batch size: 49, lr: 3.94e-02, grad_scale: 8.0 +2023-03-31 21:02:26,556 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.984e+02 8.812e+02 1.125e+03 1.363e+03 2.149e+03, threshold=2.249e+03, percent-clipped=0.0 +2023-03-31 21:02:45,504 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0 +2023-03-31 21:02:58,273 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6335.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:03:06,727 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8188, 1.2901, 1.4316, 1.0233, 2.3941, 3.2153, 2.8436, 3.3749], + device='cuda:0'), covar=tensor([0.1899, 0.4121, 0.3749, 0.3031, 0.0653, 0.0220, 0.0454, 0.0270], + device='cuda:0'), in_proj_covar=tensor([0.0314, 0.0287, 0.0313, 0.0303, 0.0198, 0.0114, 0.0177, 0.0121], + device='cuda:0'), out_proj_covar=tensor([2.5964e-04, 2.4466e-04, 2.6142e-04, 2.5841e-04, 1.9106e-04, 9.6840e-05, + 1.4820e-04, 1.1165e-04], device='cuda:0') +2023-03-31 21:03:17,629 INFO [train.py:903] (0/4) Epoch 1, batch 6350, loss[loss=0.3822, simple_loss=0.4196, pruned_loss=0.1724, over 19485.00 frames. ], tot_loss[loss=0.3992, simple_loss=0.4219, pruned_loss=0.1883, over 3831716.59 frames. ], batch size: 64, lr: 3.93e-02, grad_scale: 8.0 +2023-03-31 21:03:18,874 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6352.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:03:50,810 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6866, 1.3438, 1.2215, 1.8167, 1.5442, 1.7998, 1.9163, 1.6664], + device='cuda:0'), covar=tensor([0.0902, 0.1495, 0.1626, 0.1220, 0.1790, 0.1212, 0.1395, 0.0922], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0302, 0.0294, 0.0314, 0.0380, 0.0279, 0.0337, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 21:03:58,092 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6383.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 21:04:11,220 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4715, 1.0275, 1.1570, 0.4076, 2.3096, 2.1184, 1.8671, 2.0358], + device='cuda:0'), covar=tensor([0.1307, 0.2543, 0.2507, 0.2445, 0.0372, 0.0197, 0.0338, 0.0248], + device='cuda:0'), in_proj_covar=tensor([0.0313, 0.0285, 0.0316, 0.0303, 0.0200, 0.0114, 0.0179, 0.0122], + device='cuda:0'), out_proj_covar=tensor([2.5990e-04, 2.4399e-04, 2.6353e-04, 2.5817e-04, 1.9322e-04, 9.6715e-05, + 1.5059e-04, 1.1409e-04], device='cuda:0') +2023-03-31 21:04:12,771 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.34 vs. limit=5.0 +2023-03-31 21:04:18,887 INFO [train.py:903] (0/4) Epoch 1, batch 6400, loss[loss=0.4028, simple_loss=0.437, pruned_loss=0.1843, over 19618.00 frames. ], tot_loss[loss=0.3982, simple_loss=0.4217, pruned_loss=0.1873, over 3824453.68 frames. ], batch size: 57, lr: 3.92e-02, grad_scale: 8.0 +2023-03-31 21:04:24,485 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6405.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:04:27,872 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.880e+02 9.359e+02 1.206e+03 1.547e+03 5.333e+03, threshold=2.412e+03, percent-clipped=7.0 +2023-03-31 21:04:28,271 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6408.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 21:05:09,509 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-03-31 21:05:19,076 INFO [train.py:903] (0/4) Epoch 1, batch 6450, loss[loss=0.4841, simple_loss=0.4883, pruned_loss=0.24, over 17944.00 frames. ], tot_loss[loss=0.399, simple_loss=0.4222, pruned_loss=0.1879, over 3813154.84 frames. ], batch size: 83, lr: 3.91e-02, grad_scale: 8.0 +2023-03-31 21:05:39,895 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6467.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:05:55,618 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6480.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:06:01,942 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6485.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:06:05,061 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-03-31 21:06:05,478 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6488.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:06:15,201 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6497.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:06:21,198 INFO [train.py:903] (0/4) Epoch 1, batch 6500, loss[loss=0.4885, simple_loss=0.4846, pruned_loss=0.2462, over 18763.00 frames. ], tot_loss[loss=0.3985, simple_loss=0.4222, pruned_loss=0.1874, over 3817228.47 frames. ], batch size: 74, lr: 3.90e-02, grad_scale: 8.0 +2023-03-31 21:06:25,573 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-03-31 21:06:28,780 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.341e+02 9.704e+02 1.201e+03 1.443e+03 2.205e+03, threshold=2.402e+03, percent-clipped=0.0 +2023-03-31 21:06:35,097 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6513.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:06:43,916 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6520.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:06:58,399 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6532.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:07:15,431 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-03-31 21:07:21,792 INFO [train.py:903] (0/4) Epoch 1, batch 6550, loss[loss=0.3841, simple_loss=0.4182, pruned_loss=0.1749, over 19673.00 frames. ], tot_loss[loss=0.3982, simple_loss=0.4219, pruned_loss=0.1872, over 3812635.48 frames. ], batch size: 58, lr: 3.89e-02, grad_scale: 8.0 +2023-03-31 21:07:39,359 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6565.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:07:45,286 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1785, 1.2627, 1.0073, 0.9246, 1.0326, 0.7693, 0.4130, 1.0647], + device='cuda:0'), covar=tensor([0.0676, 0.0541, 0.1110, 0.0741, 0.0821, 0.1407, 0.1377, 0.0816], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0156, 0.0227, 0.0236, 0.0163, 0.0265, 0.0253, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:07:48,404 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6573.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:08:10,418 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6591.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:08:22,325 INFO [train.py:903] (0/4) Epoch 1, batch 6600, loss[loss=0.428, simple_loss=0.4511, pruned_loss=0.2024, over 19662.00 frames. ], tot_loss[loss=0.3946, simple_loss=0.4193, pruned_loss=0.1849, over 3820048.89 frames. ], batch size: 58, lr: 3.89e-02, grad_scale: 16.0 +2023-03-31 21:08:31,047 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.089e+02 8.736e+02 1.082e+03 1.231e+03 3.386e+03, threshold=2.164e+03, percent-clipped=2.0 +2023-03-31 21:08:36,154 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6612.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:08:40,729 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6616.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:08:56,077 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6628.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:09:05,509 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9993, 1.0847, 0.9038, 0.8808, 0.8220, 1.0241, 0.0619, 0.4902], + device='cuda:0'), covar=tensor([0.0461, 0.0510, 0.0330, 0.0417, 0.0914, 0.0461, 0.1190, 0.0942], + device='cuda:0'), in_proj_covar=tensor([0.0177, 0.0178, 0.0179, 0.0223, 0.0252, 0.0227, 0.0234, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:09:19,944 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6647.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:09:20,015 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6647.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:09:24,251 INFO [train.py:903] (0/4) Epoch 1, batch 6650, loss[loss=0.4575, simple_loss=0.4507, pruned_loss=0.2322, over 12854.00 frames. ], tot_loss[loss=0.3953, simple_loss=0.42, pruned_loss=0.1852, over 3817105.37 frames. ], batch size: 136, lr: 3.88e-02, grad_scale: 4.0 +2023-03-31 21:09:42,128 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3226, 1.0028, 1.0486, 1.3616, 1.0081, 1.2976, 1.2609, 1.2422], + device='cuda:0'), covar=tensor([0.0792, 0.1508, 0.1593, 0.0993, 0.1509, 0.1151, 0.1297, 0.0959], + device='cuda:0'), in_proj_covar=tensor([0.0244, 0.0303, 0.0296, 0.0330, 0.0379, 0.0272, 0.0341, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 21:09:56,611 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9964, 1.6552, 1.5094, 1.7933, 1.5526, 1.7824, 1.8959, 1.8407], + device='cuda:0'), covar=tensor([0.0793, 0.1451, 0.1533, 0.1431, 0.2168, 0.1309, 0.1798, 0.0926], + device='cuda:0'), in_proj_covar=tensor([0.0247, 0.0305, 0.0299, 0.0334, 0.0382, 0.0274, 0.0345, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 21:09:59,883 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6680.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:10:09,261 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6688.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:10:11,575 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8688, 1.4262, 1.5888, 1.2061, 2.7443, 3.3211, 3.1851, 3.5750], + device='cuda:0'), covar=tensor([0.1708, 0.2912, 0.2758, 0.2632, 0.0538, 0.0145, 0.0214, 0.0154], + device='cuda:0'), in_proj_covar=tensor([0.0313, 0.0294, 0.0323, 0.0309, 0.0205, 0.0111, 0.0180, 0.0123], + device='cuda:0'), out_proj_covar=tensor([2.6290e-04, 2.5506e-04, 2.7242e-04, 2.6599e-04, 2.0068e-04, 9.8375e-05, + 1.5637e-04, 1.1721e-04], device='cuda:0') +2023-03-31 21:10:23,711 INFO [train.py:903] (0/4) Epoch 1, batch 6700, loss[loss=0.4138, simple_loss=0.4396, pruned_loss=0.194, over 19509.00 frames. ], tot_loss[loss=0.3955, simple_loss=0.4196, pruned_loss=0.1857, over 3828002.39 frames. ], batch size: 64, lr: 3.87e-02, grad_scale: 4.0 +2023-03-31 21:10:35,447 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.729e+02 8.354e+02 1.101e+03 1.693e+03 1.016e+04, threshold=2.202e+03, percent-clipped=16.0 +2023-03-31 21:10:51,612 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6723.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:11:14,953 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6743.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:11:19,487 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9753, 3.6917, 2.2900, 3.2950, 1.4769, 3.4693, 3.2486, 3.3720], + device='cuda:0'), covar=tensor([0.0725, 0.1021, 0.2015, 0.0865, 0.3081, 0.0867, 0.0689, 0.0884], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0244, 0.0282, 0.0242, 0.0304, 0.0237, 0.0185, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 21:11:19,670 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5159, 1.5340, 1.1411, 1.5667, 1.4253, 1.4624, 1.3508, 1.6604], + device='cuda:0'), covar=tensor([0.0830, 0.1634, 0.1382, 0.1038, 0.1542, 0.0751, 0.1225, 0.0619], + device='cuda:0'), in_proj_covar=tensor([0.0229, 0.0365, 0.0271, 0.0244, 0.0310, 0.0238, 0.0268, 0.0214], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:11:20,754 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6748.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:11:23,799 INFO [train.py:903] (0/4) Epoch 1, batch 6750, loss[loss=0.3761, simple_loss=0.4097, pruned_loss=0.1713, over 19373.00 frames. ], tot_loss[loss=0.3926, simple_loss=0.4182, pruned_loss=0.1835, over 3837255.19 frames. ], batch size: 48, lr: 3.86e-02, grad_scale: 4.0 +2023-03-31 21:11:51,614 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6776.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:12:18,952 INFO [train.py:903] (0/4) Epoch 1, batch 6800, loss[loss=0.4861, simple_loss=0.4739, pruned_loss=0.2491, over 13014.00 frames. ], tot_loss[loss=0.3927, simple_loss=0.4184, pruned_loss=0.1835, over 3824598.94 frames. ], batch size: 135, lr: 3.85e-02, grad_scale: 8.0 +2023-03-31 21:12:19,292 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6801.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:12:28,853 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.487e+02 9.125e+02 1.072e+03 1.412e+03 3.162e+03, threshold=2.143e+03, percent-clipped=4.0 +2023-03-31 21:12:43,729 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6824.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:12:48,499 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-1.pt +2023-03-31 21:13:03,356 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-03-31 21:13:04,444 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-03-31 21:13:06,742 INFO [train.py:903] (0/4) Epoch 2, batch 0, loss[loss=0.4562, simple_loss=0.4637, pruned_loss=0.2244, over 19043.00 frames. ], tot_loss[loss=0.4562, simple_loss=0.4637, pruned_loss=0.2244, over 19043.00 frames. ], batch size: 69, lr: 3.77e-02, grad_scale: 8.0 +2023-03-31 21:13:06,743 INFO [train.py:928] (0/4) Computing validation loss +2023-03-31 21:13:18,419 INFO [train.py:937] (0/4) Epoch 2, validation: loss=0.2802, simple_loss=0.3637, pruned_loss=0.09835, over 944034.00 frames. +2023-03-31 21:13:18,420 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18153MB +2023-03-31 21:13:18,572 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6829.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:13:28,805 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-03-31 21:14:06,576 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6868.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:14:20,898 INFO [train.py:903] (0/4) Epoch 2, batch 50, loss[loss=0.3667, simple_loss=0.3922, pruned_loss=0.1706, over 19843.00 frames. ], tot_loss[loss=0.3882, simple_loss=0.4138, pruned_loss=0.1813, over 867290.57 frames. ], batch size: 52, lr: 3.76e-02, grad_scale: 8.0 +2023-03-31 21:14:37,427 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6893.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:14:41,557 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6896.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 21:14:49,439 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6903.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:14:54,401 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-03-31 21:14:57,930 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.488e+02 9.076e+02 1.150e+03 1.515e+03 2.802e+03, threshold=2.301e+03, percent-clipped=3.0 +2023-03-31 21:15:19,035 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=6927.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:15:20,147 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6928.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:15:20,723 INFO [train.py:903] (0/4) Epoch 2, batch 100, loss[loss=0.5355, simple_loss=0.5121, pruned_loss=0.2794, over 17444.00 frames. ], tot_loss[loss=0.3987, simple_loss=0.4214, pruned_loss=0.1879, over 1515834.46 frames. ], batch size: 101, lr: 3.75e-02, grad_scale: 8.0 +2023-03-31 21:15:29,299 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6936.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:15:32,189 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-03-31 21:15:33,689 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6939.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:15:39,443 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=6944.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:15:39,518 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6944.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:15:43,891 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.54 vs. limit=5.0 +2023-03-31 21:16:01,540 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6961.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:16:10,909 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=6969.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:16:23,040 INFO [train.py:903] (0/4) Epoch 2, batch 150, loss[loss=0.3371, simple_loss=0.3823, pruned_loss=0.146, over 19775.00 frames. ], tot_loss[loss=0.3923, simple_loss=0.4172, pruned_loss=0.1837, over 2027462.14 frames. ], batch size: 54, lr: 3.74e-02, grad_scale: 4.0 +2023-03-31 21:16:38,559 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=6991.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:16:49,224 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=6999.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:17:03,027 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.128e+02 7.870e+02 9.855e+02 1.288e+03 4.108e+03, threshold=1.971e+03, percent-clipped=4.0 +2023-03-31 21:17:19,658 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7024.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:17:24,032 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-03-31 21:17:25,200 INFO [train.py:903] (0/4) Epoch 2, batch 200, loss[loss=0.4042, simple_loss=0.4352, pruned_loss=0.1866, over 19307.00 frames. ], tot_loss[loss=0.3856, simple_loss=0.4128, pruned_loss=0.1792, over 2437375.74 frames. ], batch size: 66, lr: 3.73e-02, grad_scale: 4.0 +2023-03-31 21:17:54,551 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.1090, 3.7815, 4.5865, 4.4044, 1.6319, 4.0600, 3.6830, 4.0447], + device='cuda:0'), covar=tensor([0.0249, 0.0485, 0.0365, 0.0186, 0.2798, 0.0189, 0.0356, 0.0761], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0257, 0.0316, 0.0220, 0.0398, 0.0155, 0.0238, 0.0342], + device='cuda:0'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 21:18:11,606 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7066.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 21:18:29,133 INFO [train.py:903] (0/4) Epoch 2, batch 250, loss[loss=0.3867, simple_loss=0.4201, pruned_loss=0.1766, over 19528.00 frames. ], tot_loss[loss=0.3831, simple_loss=0.4111, pruned_loss=0.1776, over 2753309.82 frames. ], batch size: 54, lr: 3.72e-02, grad_scale: 4.0 +2023-03-31 21:19:03,847 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7106.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:19:09,455 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.258e+02 7.747e+02 9.740e+02 1.157e+03 2.695e+03, threshold=1.948e+03, percent-clipped=1.0 +2023-03-31 21:19:33,822 INFO [train.py:903] (0/4) Epoch 2, batch 300, loss[loss=0.2842, simple_loss=0.3301, pruned_loss=0.1192, over 19737.00 frames. ], tot_loss[loss=0.3833, simple_loss=0.412, pruned_loss=0.1773, over 2984945.74 frames. ], batch size: 46, lr: 3.72e-02, grad_scale: 4.0 +2023-03-31 21:20:35,435 INFO [train.py:903] (0/4) Epoch 2, batch 350, loss[loss=0.3827, simple_loss=0.4197, pruned_loss=0.1729, over 19657.00 frames. ], tot_loss[loss=0.3853, simple_loss=0.4129, pruned_loss=0.1789, over 3175006.36 frames. ], batch size: 60, lr: 3.71e-02, grad_scale: 4.0 +2023-03-31 21:20:37,925 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7181.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:20:39,829 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-03-31 21:20:54,814 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7195.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:21:01,700 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7200.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:21:15,795 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.882e+02 9.781e+02 1.245e+03 1.512e+03 3.081e+03, threshold=2.489e+03, percent-clipped=8.0 +2023-03-31 21:21:26,506 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7220.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:21:32,335 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7225.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:21:37,138 INFO [train.py:903] (0/4) Epoch 2, batch 400, loss[loss=0.3932, simple_loss=0.4007, pruned_loss=0.1928, over 19763.00 frames. ], tot_loss[loss=0.3868, simple_loss=0.4135, pruned_loss=0.1801, over 3316961.96 frames. ], batch size: 45, lr: 3.70e-02, grad_scale: 8.0 +2023-03-31 21:21:51,080 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7240.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 21:22:30,090 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7271.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:22:40,124 INFO [train.py:903] (0/4) Epoch 2, batch 450, loss[loss=0.41, simple_loss=0.4378, pruned_loss=0.1911, over 19686.00 frames. ], tot_loss[loss=0.3854, simple_loss=0.4129, pruned_loss=0.179, over 3444386.82 frames. ], batch size: 59, lr: 3.69e-02, grad_scale: 8.0 +2023-03-31 21:22:58,508 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7293.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:23:14,127 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-03-31 21:23:15,292 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-03-31 21:23:19,468 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.877e+02 8.763e+02 1.192e+03 1.491e+03 2.950e+03, threshold=2.384e+03, percent-clipped=4.0 +2023-03-31 21:23:43,129 INFO [train.py:903] (0/4) Epoch 2, batch 500, loss[loss=0.3243, simple_loss=0.3543, pruned_loss=0.1472, over 19779.00 frames. ], tot_loss[loss=0.3847, simple_loss=0.4124, pruned_loss=0.1784, over 3535667.02 frames. ], batch size: 48, lr: 3.68e-02, grad_scale: 8.0 +2023-03-31 21:24:14,671 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7355.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 21:24:23,762 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7362.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:24:37,496 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2387, 3.3390, 3.6636, 3.6151, 1.3006, 3.1628, 3.0620, 3.1328], + device='cuda:0'), covar=tensor([0.0330, 0.0582, 0.0533, 0.0318, 0.3082, 0.0269, 0.0404, 0.1009], + device='cuda:0'), in_proj_covar=tensor([0.0216, 0.0255, 0.0323, 0.0218, 0.0396, 0.0155, 0.0231, 0.0332], + device='cuda:0'), out_proj_covar=tensor([1.3372e-04, 1.5776e-04, 2.0237e-04, 1.2543e-04, 2.1244e-04, 9.9618e-05, + 1.3496e-04, 1.8547e-04], device='cuda:0') +2023-03-31 21:24:45,279 INFO [train.py:903] (0/4) Epoch 2, batch 550, loss[loss=0.3556, simple_loss=0.4002, pruned_loss=0.1555, over 19658.00 frames. ], tot_loss[loss=0.3858, simple_loss=0.4135, pruned_loss=0.179, over 3606109.16 frames. ], batch size: 55, lr: 3.67e-02, grad_scale: 8.0 +2023-03-31 21:24:53,909 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7386.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:24:55,088 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7387.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:25:24,040 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7410.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 21:25:26,117 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.968e+02 9.206e+02 1.127e+03 1.377e+03 2.659e+03, threshold=2.254e+03, percent-clipped=2.0 +2023-03-31 21:25:43,338 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7425.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:25:47,659 INFO [train.py:903] (0/4) Epoch 2, batch 600, loss[loss=0.3646, simple_loss=0.3793, pruned_loss=0.1749, over 19743.00 frames. ], tot_loss[loss=0.3852, simple_loss=0.4129, pruned_loss=0.1788, over 3668463.98 frames. ], batch size: 45, lr: 3.66e-02, grad_scale: 8.0 +2023-03-31 21:26:13,260 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0541, 3.9347, 4.5480, 4.4863, 1.4713, 4.0333, 3.7156, 3.9842], + device='cuda:0'), covar=tensor([0.0265, 0.0447, 0.0388, 0.0194, 0.3081, 0.0219, 0.0358, 0.0906], + device='cuda:0'), in_proj_covar=tensor([0.0220, 0.0256, 0.0325, 0.0217, 0.0398, 0.0157, 0.0233, 0.0336], + device='cuda:0'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 21:26:26,674 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8169, 1.3272, 1.3040, 1.8682, 1.3948, 1.8009, 2.0529, 2.0713], + device='cuda:0'), covar=tensor([0.1132, 0.1776, 0.1823, 0.1693, 0.2209, 0.1376, 0.1787, 0.0907], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0307, 0.0303, 0.0328, 0.0376, 0.0277, 0.0340, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 21:26:29,908 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-03-31 21:26:50,215 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-03-31 21:26:50,577 INFO [train.py:903] (0/4) Epoch 2, batch 650, loss[loss=0.3356, simple_loss=0.3785, pruned_loss=0.1463, over 19734.00 frames. ], tot_loss[loss=0.3837, simple_loss=0.4119, pruned_loss=0.1777, over 3699557.05 frames. ], batch size: 51, lr: 3.66e-02, grad_scale: 8.0 +2023-03-31 21:27:07,972 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.9658, 5.4437, 2.9988, 4.6588, 1.6755, 5.4665, 5.2931, 5.4825], + device='cuda:0'), covar=tensor([0.0492, 0.0977, 0.2084, 0.0716, 0.3533, 0.0768, 0.0509, 0.0479], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0268, 0.0305, 0.0249, 0.0320, 0.0256, 0.0202, 0.0222], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:27:30,827 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.799e+02 8.519e+02 1.041e+03 1.431e+03 3.840e+03, threshold=2.082e+03, percent-clipped=3.0 +2023-03-31 21:27:49,374 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7525.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:27:49,584 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7525.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 21:27:53,956 INFO [train.py:903] (0/4) Epoch 2, batch 700, loss[loss=0.3813, simple_loss=0.4264, pruned_loss=0.1681, over 19661.00 frames. ], tot_loss[loss=0.3815, simple_loss=0.4099, pruned_loss=0.1766, over 3731587.76 frames. ], batch size: 58, lr: 3.65e-02, grad_scale: 8.0 +2023-03-31 21:28:56,975 INFO [train.py:903] (0/4) Epoch 2, batch 750, loss[loss=0.4171, simple_loss=0.4345, pruned_loss=0.1999, over 18455.00 frames. ], tot_loss[loss=0.3844, simple_loss=0.4121, pruned_loss=0.1783, over 3753916.13 frames. ], batch size: 84, lr: 3.64e-02, grad_scale: 8.0 +2023-03-31 21:29:24,821 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6233, 2.8547, 3.0089, 2.9428, 1.1499, 2.7175, 2.4952, 2.5493], + device='cuda:0'), covar=tensor([0.0521, 0.0648, 0.0628, 0.0406, 0.3135, 0.0322, 0.0502, 0.1341], + device='cuda:0'), in_proj_covar=tensor([0.0226, 0.0261, 0.0330, 0.0224, 0.0398, 0.0162, 0.0234, 0.0344], + device='cuda:0'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 21:29:35,963 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.266e+02 8.676e+02 1.032e+03 1.220e+03 3.020e+03, threshold=2.064e+03, percent-clipped=5.0 +2023-03-31 21:29:36,435 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7611.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 21:29:55,584 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6872, 1.5532, 1.1953, 1.7705, 1.5343, 1.5378, 1.5204, 1.9568], + device='cuda:0'), covar=tensor([0.0950, 0.2151, 0.1741, 0.1208, 0.1792, 0.0765, 0.1352, 0.0601], + device='cuda:0'), in_proj_covar=tensor([0.0248, 0.0370, 0.0290, 0.0253, 0.0327, 0.0259, 0.0285, 0.0215], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:29:58,614 INFO [train.py:903] (0/4) Epoch 2, batch 800, loss[loss=0.3877, simple_loss=0.4156, pruned_loss=0.1799, over 19691.00 frames. ], tot_loss[loss=0.3829, simple_loss=0.4113, pruned_loss=0.1772, over 3773181.13 frames. ], batch size: 53, lr: 3.63e-02, grad_scale: 8.0 +2023-03-31 21:30:01,363 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1400, 1.6394, 1.2706, 2.0925, 1.3805, 2.2003, 2.1142, 1.7614], + device='cuda:0'), covar=tensor([0.0923, 0.1383, 0.1861, 0.1210, 0.2087, 0.0990, 0.1671, 0.0935], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0303, 0.0311, 0.0329, 0.0378, 0.0274, 0.0340, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 21:30:07,109 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7636.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 21:30:09,131 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7637.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:30:11,708 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9943, 1.1799, 1.7178, 1.2112, 2.2492, 2.3533, 2.4399, 1.1114], + device='cuda:0'), covar=tensor([0.1397, 0.1755, 0.1207, 0.1416, 0.0794, 0.0739, 0.0870, 0.1596], + device='cuda:0'), in_proj_covar=tensor([0.0314, 0.0321, 0.0304, 0.0314, 0.0352, 0.0286, 0.0408, 0.0334], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 21:30:12,812 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7640.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:30:15,236 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7642.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:30:15,378 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7642.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:30:16,157 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-03-31 21:30:46,859 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7667.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:31:01,250 INFO [train.py:903] (0/4) Epoch 2, batch 850, loss[loss=0.3752, simple_loss=0.4087, pruned_loss=0.1709, over 19843.00 frames. ], tot_loss[loss=0.3842, simple_loss=0.4128, pruned_loss=0.1778, over 3797338.71 frames. ], batch size: 52, lr: 3.62e-02, grad_scale: 8.0 +2023-03-31 21:31:41,593 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.436e+02 9.016e+02 1.057e+03 1.450e+03 5.160e+03, threshold=2.114e+03, percent-clipped=6.0 +2023-03-31 21:31:56,564 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-03-31 21:32:02,492 INFO [train.py:903] (0/4) Epoch 2, batch 900, loss[loss=0.3395, simple_loss=0.3657, pruned_loss=0.1567, over 18131.00 frames. ], tot_loss[loss=0.3848, simple_loss=0.4128, pruned_loss=0.1784, over 3794843.97 frames. ], batch size: 40, lr: 3.61e-02, grad_scale: 4.0 +2023-03-31 21:32:04,688 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-03-31 21:32:26,842 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2082, 1.7613, 1.4606, 1.2026, 1.5722, 0.8850, 0.5274, 1.6857], + device='cuda:0'), covar=tensor([0.0785, 0.0414, 0.0962, 0.0720, 0.0544, 0.1395, 0.1190, 0.0512], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0174, 0.0246, 0.0238, 0.0178, 0.0277, 0.0258, 0.0246], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:32:32,755 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7752.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:32:43,835 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7761.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:32:44,208 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.96 vs. limit=2.0 +2023-03-31 21:32:53,062 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7769.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:32:58,707 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6798, 1.4046, 1.1650, 1.5630, 1.1980, 1.3837, 1.3791, 1.5981], + device='cuda:0'), covar=tensor([0.0855, 0.1737, 0.1429, 0.1101, 0.1595, 0.0776, 0.1194, 0.0718], + device='cuda:0'), in_proj_covar=tensor([0.0245, 0.0362, 0.0280, 0.0252, 0.0317, 0.0249, 0.0281, 0.0214], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:33:06,311 INFO [train.py:903] (0/4) Epoch 2, batch 950, loss[loss=0.4446, simple_loss=0.4608, pruned_loss=0.2142, over 19297.00 frames. ], tot_loss[loss=0.384, simple_loss=0.4123, pruned_loss=0.1779, over 3815827.83 frames. ], batch size: 66, lr: 3.61e-02, grad_scale: 4.0 +2023-03-31 21:33:09,105 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7781.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 21:33:10,900 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-03-31 21:33:39,978 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7806.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 21:33:46,539 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.660e+02 8.698e+02 1.089e+03 1.494e+03 2.916e+03, threshold=2.178e+03, percent-clipped=6.0 +2023-03-31 21:33:58,941 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=7820.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:34:09,630 INFO [train.py:903] (0/4) Epoch 2, batch 1000, loss[loss=0.3629, simple_loss=0.4067, pruned_loss=0.1595, over 19664.00 frames. ], tot_loss[loss=0.3822, simple_loss=0.4114, pruned_loss=0.1765, over 3824385.25 frames. ], batch size: 58, lr: 3.60e-02, grad_scale: 4.0 +2023-03-31 21:35:04,945 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-03-31 21:35:11,833 INFO [train.py:903] (0/4) Epoch 2, batch 1050, loss[loss=0.3442, simple_loss=0.371, pruned_loss=0.1587, over 19765.00 frames. ], tot_loss[loss=0.3804, simple_loss=0.41, pruned_loss=0.1754, over 3802499.69 frames. ], batch size: 45, lr: 3.59e-02, grad_scale: 4.0 +2023-03-31 21:35:18,830 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=7884.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:35:33,736 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=7896.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:35:46,810 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-03-31 21:35:53,415 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.580e+02 8.878e+02 9.952e+02 1.228e+03 3.126e+03, threshold=1.990e+03, percent-clipped=5.0 +2023-03-31 21:36:05,089 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=7921.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:36:14,154 INFO [train.py:903] (0/4) Epoch 2, batch 1100, loss[loss=0.438, simple_loss=0.447, pruned_loss=0.2144, over 17495.00 frames. ], tot_loss[loss=0.3815, simple_loss=0.4106, pruned_loss=0.1762, over 3801683.46 frames. ], batch size: 101, lr: 3.58e-02, grad_scale: 4.0 +2023-03-31 21:36:31,415 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7545, 2.9408, 3.0644, 3.0692, 1.1108, 2.7590, 2.6037, 2.7097], + device='cuda:0'), covar=tensor([0.0483, 0.0566, 0.0635, 0.0391, 0.3120, 0.0320, 0.0461, 0.1253], + device='cuda:0'), in_proj_covar=tensor([0.0228, 0.0251, 0.0331, 0.0228, 0.0391, 0.0157, 0.0228, 0.0353], + device='cuda:0'), out_proj_covar=tensor([1.3955e-04, 1.5496e-04, 2.0549e-04, 1.2859e-04, 2.0777e-04, 9.9032e-05, + 1.3016e-04, 1.9373e-04], device='cuda:0') +2023-03-31 21:36:40,181 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2634, 1.1042, 1.3561, 0.3240, 2.6102, 2.2788, 1.7213, 2.2712], + device='cuda:0'), covar=tensor([0.1433, 0.2671, 0.2434, 0.2567, 0.0302, 0.0212, 0.0468, 0.0245], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0282, 0.0324, 0.0297, 0.0196, 0.0111, 0.0186, 0.0115], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-03-31 21:37:16,838 INFO [train.py:903] (0/4) Epoch 2, batch 1150, loss[loss=0.492, simple_loss=0.4811, pruned_loss=0.2515, over 19587.00 frames. ], tot_loss[loss=0.3824, simple_loss=0.4112, pruned_loss=0.1768, over 3800276.00 frames. ], batch size: 61, lr: 3.57e-02, grad_scale: 4.0 +2023-03-31 21:37:26,671 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=7986.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:37:43,808 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-8000.pt +2023-03-31 21:37:54,467 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8008.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:37:58,738 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.862e+02 8.681e+02 1.035e+03 1.273e+03 2.854e+03, threshold=2.070e+03, percent-clipped=5.0 +2023-03-31 21:38:21,506 INFO [train.py:903] (0/4) Epoch 2, batch 1200, loss[loss=0.4473, simple_loss=0.4448, pruned_loss=0.2249, over 13376.00 frames. ], tot_loss[loss=0.3814, simple_loss=0.4105, pruned_loss=0.1762, over 3791795.22 frames. ], batch size: 136, lr: 3.56e-02, grad_scale: 8.0 +2023-03-31 21:38:26,392 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8033.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:38:52,432 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-03-31 21:39:22,668 INFO [train.py:903] (0/4) Epoch 2, batch 1250, loss[loss=0.372, simple_loss=0.4097, pruned_loss=0.1671, over 19619.00 frames. ], tot_loss[loss=0.3825, simple_loss=0.4112, pruned_loss=0.177, over 3796588.26 frames. ], batch size: 57, lr: 3.56e-02, grad_scale: 8.0 +2023-03-31 21:39:50,757 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8101.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:39:57,180 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=8105.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:40:05,139 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.516e+02 8.392e+02 1.041e+03 1.254e+03 3.427e+03, threshold=2.083e+03, percent-clipped=3.0 +2023-03-31 21:40:25,457 INFO [train.py:903] (0/4) Epoch 2, batch 1300, loss[loss=0.3869, simple_loss=0.4219, pruned_loss=0.1759, over 19317.00 frames. ], tot_loss[loss=0.3803, simple_loss=0.4096, pruned_loss=0.1755, over 3795935.85 frames. ], batch size: 66, lr: 3.55e-02, grad_scale: 8.0 +2023-03-31 21:40:39,774 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8140.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:41:09,939 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=8164.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:41:11,405 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8165.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:41:28,092 INFO [train.py:903] (0/4) Epoch 2, batch 1350, loss[loss=0.3858, simple_loss=0.422, pruned_loss=0.1748, over 19298.00 frames. ], tot_loss[loss=0.38, simple_loss=0.409, pruned_loss=0.1755, over 3809187.03 frames. ], batch size: 66, lr: 3.54e-02, grad_scale: 8.0 +2023-03-31 21:41:31,735 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8759, 1.3831, 1.6613, 1.4212, 2.7452, 3.0915, 3.1457, 3.3129], + device='cuda:0'), covar=tensor([0.1323, 0.2483, 0.2362, 0.2031, 0.0463, 0.0154, 0.0227, 0.0135], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0280, 0.0321, 0.0290, 0.0199, 0.0110, 0.0183, 0.0111], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-03-31 21:41:42,272 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-03-31 21:42:08,819 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.809e+02 9.088e+02 1.109e+03 1.527e+03 2.312e+03, threshold=2.218e+03, percent-clipped=6.0 +2023-03-31 21:42:19,621 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8220.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:42:30,883 INFO [train.py:903] (0/4) Epoch 2, batch 1400, loss[loss=0.334, simple_loss=0.3814, pruned_loss=0.1433, over 19724.00 frames. ], tot_loss[loss=0.3764, simple_loss=0.4067, pruned_loss=0.1731, over 3829762.47 frames. ], batch size: 51, lr: 3.53e-02, grad_scale: 8.0 +2023-03-31 21:43:32,681 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-03-31 21:43:33,840 INFO [train.py:903] (0/4) Epoch 2, batch 1450, loss[loss=0.3765, simple_loss=0.4028, pruned_loss=0.1751, over 19470.00 frames. ], tot_loss[loss=0.3749, simple_loss=0.4058, pruned_loss=0.1721, over 3833242.76 frames. ], batch size: 49, lr: 3.53e-02, grad_scale: 8.0 +2023-03-31 21:43:34,146 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8279.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:43:38,616 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=8283.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:44:15,421 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.216e+02 8.724e+02 1.078e+03 1.353e+03 2.729e+03, threshold=2.156e+03, percent-clipped=3.0 +2023-03-31 21:44:30,498 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6645, 1.8203, 1.2636, 1.2062, 1.2048, 1.4419, 0.0978, 0.7758], + device='cuda:0'), covar=tensor([0.0499, 0.0465, 0.0326, 0.0434, 0.0940, 0.0552, 0.1068, 0.0857], + device='cuda:0'), in_proj_covar=tensor([0.0198, 0.0199, 0.0190, 0.0226, 0.0266, 0.0236, 0.0240, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:44:35,646 INFO [train.py:903] (0/4) Epoch 2, batch 1500, loss[loss=0.3766, simple_loss=0.3989, pruned_loss=0.1771, over 19576.00 frames. ], tot_loss[loss=0.3766, simple_loss=0.4065, pruned_loss=0.1733, over 3820863.57 frames. ], batch size: 52, lr: 3.52e-02, grad_scale: 8.0 +2023-03-31 21:45:10,884 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8357.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:45:38,213 INFO [train.py:903] (0/4) Epoch 2, batch 1550, loss[loss=0.3911, simple_loss=0.4228, pruned_loss=0.1797, over 19588.00 frames. ], tot_loss[loss=0.3758, simple_loss=0.4059, pruned_loss=0.1728, over 3830874.48 frames. ], batch size: 52, lr: 3.51e-02, grad_scale: 8.0 +2023-03-31 21:45:42,320 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8382.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:46:19,242 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.712e+02 9.525e+02 1.175e+03 1.582e+03 3.285e+03, threshold=2.351e+03, percent-clipped=5.0 +2023-03-31 21:46:41,077 INFO [train.py:903] (0/4) Epoch 2, batch 1600, loss[loss=0.3627, simple_loss=0.394, pruned_loss=0.1657, over 19841.00 frames. ], tot_loss[loss=0.3747, simple_loss=0.4054, pruned_loss=0.172, over 3808778.47 frames. ], batch size: 52, lr: 3.50e-02, grad_scale: 8.0 +2023-03-31 21:47:02,053 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-03-31 21:47:39,723 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8476.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:47:42,841 INFO [train.py:903] (0/4) Epoch 2, batch 1650, loss[loss=0.332, simple_loss=0.3704, pruned_loss=0.1468, over 19576.00 frames. ], tot_loss[loss=0.3723, simple_loss=0.4035, pruned_loss=0.1705, over 3812333.28 frames. ], batch size: 52, lr: 3.49e-02, grad_scale: 8.0 +2023-03-31 21:47:56,679 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7657, 4.3566, 2.5378, 3.9938, 1.5313, 4.1533, 3.8967, 4.1299], + device='cuda:0'), covar=tensor([0.0470, 0.0922, 0.1881, 0.0618, 0.3306, 0.0789, 0.0683, 0.0610], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0268, 0.0301, 0.0242, 0.0320, 0.0258, 0.0201, 0.0223], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 21:48:10,065 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8501.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:48:23,186 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.971e+02 8.791e+02 1.048e+03 1.403e+03 4.696e+03, threshold=2.096e+03, percent-clipped=2.0 +2023-03-31 21:48:44,117 INFO [train.py:903] (0/4) Epoch 2, batch 1700, loss[loss=0.3832, simple_loss=0.4168, pruned_loss=0.1749, over 19690.00 frames. ], tot_loss[loss=0.3745, simple_loss=0.4048, pruned_loss=0.1721, over 3819875.70 frames. ], batch size: 59, lr: 3.49e-02, grad_scale: 8.0 +2023-03-31 21:48:51,035 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8535.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:49:17,272 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9430, 2.0422, 1.5988, 1.6429, 1.5684, 1.6323, 0.7244, 1.3635], + device='cuda:0'), covar=tensor([0.0390, 0.0383, 0.0253, 0.0334, 0.0537, 0.0447, 0.0790, 0.0651], + device='cuda:0'), in_proj_covar=tensor([0.0196, 0.0199, 0.0190, 0.0231, 0.0270, 0.0239, 0.0242, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:49:23,079 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=8560.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:49:23,924 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-03-31 21:49:31,719 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3334, 0.9762, 1.2231, 1.1482, 2.0540, 1.0978, 1.8212, 1.9503], + device='cuda:0'), covar=tensor([0.0652, 0.2940, 0.2707, 0.1877, 0.0738, 0.1984, 0.1039, 0.0863], + device='cuda:0'), in_proj_covar=tensor([0.0216, 0.0293, 0.0281, 0.0272, 0.0240, 0.0314, 0.0256, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:49:46,288 INFO [train.py:903] (0/4) Epoch 2, batch 1750, loss[loss=0.3709, simple_loss=0.4124, pruned_loss=0.1646, over 19525.00 frames. ], tot_loss[loss=0.3759, simple_loss=0.4059, pruned_loss=0.173, over 3812070.02 frames. ], batch size: 54, lr: 3.48e-02, grad_scale: 8.0 +2023-03-31 21:50:27,276 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.711e+02 8.589e+02 1.062e+03 1.367e+03 2.706e+03, threshold=2.124e+03, percent-clipped=6.0 +2023-03-31 21:50:46,999 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=8627.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:50:48,943 INFO [train.py:903] (0/4) Epoch 2, batch 1800, loss[loss=0.3974, simple_loss=0.4168, pruned_loss=0.189, over 19663.00 frames. ], tot_loss[loss=0.3726, simple_loss=0.4035, pruned_loss=0.1709, over 3822914.00 frames. ], batch size: 53, lr: 3.47e-02, grad_scale: 8.0 +2023-03-31 21:51:48,862 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-03-31 21:51:52,355 INFO [train.py:903] (0/4) Epoch 2, batch 1850, loss[loss=0.3081, simple_loss=0.3531, pruned_loss=0.1316, over 19351.00 frames. ], tot_loss[loss=0.3726, simple_loss=0.4039, pruned_loss=0.1707, over 3820789.65 frames. ], batch size: 47, lr: 3.46e-02, grad_scale: 8.0 +2023-03-31 21:52:26,719 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-03-31 21:52:32,412 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.683e+02 8.545e+02 1.022e+03 1.402e+03 2.945e+03, threshold=2.044e+03, percent-clipped=4.0 +2023-03-31 21:52:53,437 INFO [train.py:903] (0/4) Epoch 2, batch 1900, loss[loss=0.5349, simple_loss=0.4922, pruned_loss=0.2889, over 13397.00 frames. ], tot_loss[loss=0.3717, simple_loss=0.403, pruned_loss=0.1702, over 3817412.42 frames. ], batch size: 138, lr: 3.46e-02, grad_scale: 8.0 +2023-03-31 21:53:10,022 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=8742.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:53:13,048 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-03-31 21:53:19,062 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-03-31 21:53:44,522 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-03-31 21:53:51,870 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.8445, 5.4256, 2.9988, 4.6512, 1.7866, 5.4857, 5.0549, 5.3056], + device='cuda:0'), covar=tensor([0.0410, 0.0865, 0.1672, 0.0589, 0.3167, 0.0612, 0.0536, 0.0576], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0267, 0.0300, 0.0243, 0.0320, 0.0260, 0.0206, 0.0221], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 21:53:55,409 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=8778.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:53:57,531 INFO [train.py:903] (0/4) Epoch 2, batch 1950, loss[loss=0.3861, simple_loss=0.4163, pruned_loss=0.1779, over 19746.00 frames. ], tot_loss[loss=0.3698, simple_loss=0.402, pruned_loss=0.1688, over 3827961.52 frames. ], batch size: 63, lr: 3.45e-02, grad_scale: 8.0 +2023-03-31 21:54:38,567 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.007e+02 8.441e+02 1.013e+03 1.280e+03 2.038e+03, threshold=2.026e+03, percent-clipped=0.0 +2023-03-31 21:55:00,983 INFO [train.py:903] (0/4) Epoch 2, batch 2000, loss[loss=0.3455, simple_loss=0.3718, pruned_loss=0.1596, over 18612.00 frames. ], tot_loss[loss=0.3697, simple_loss=0.4022, pruned_loss=0.1686, over 3826922.78 frames. ], batch size: 41, lr: 3.44e-02, grad_scale: 8.0 +2023-03-31 21:56:00,559 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-03-31 21:56:01,573 INFO [train.py:903] (0/4) Epoch 2, batch 2050, loss[loss=0.3429, simple_loss=0.3732, pruned_loss=0.1562, over 19125.00 frames. ], tot_loss[loss=0.3715, simple_loss=0.4033, pruned_loss=0.1698, over 3827805.17 frames. ], batch size: 42, lr: 3.43e-02, grad_scale: 8.0 +2023-03-31 21:56:19,561 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-03-31 21:56:20,748 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-03-31 21:56:41,655 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-03-31 21:56:44,118 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.978e+02 1.029e+03 1.194e+03 1.427e+03 4.040e+03, threshold=2.389e+03, percent-clipped=7.0 +2023-03-31 21:56:51,785 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-03-31 21:57:05,505 INFO [train.py:903] (0/4) Epoch 2, batch 2100, loss[loss=0.3194, simple_loss=0.3698, pruned_loss=0.1345, over 19741.00 frames. ], tot_loss[loss=0.3722, simple_loss=0.4041, pruned_loss=0.1701, over 3828279.55 frames. ], batch size: 51, lr: 3.43e-02, grad_scale: 8.0 +2023-03-31 21:57:25,717 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=8945.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:57:36,778 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-03-31 21:57:59,444 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-03-31 21:58:07,609 INFO [train.py:903] (0/4) Epoch 2, batch 2150, loss[loss=0.322, simple_loss=0.3653, pruned_loss=0.1393, over 19649.00 frames. ], tot_loss[loss=0.3683, simple_loss=0.4014, pruned_loss=0.1676, over 3828995.49 frames. ], batch size: 53, lr: 3.42e-02, grad_scale: 8.0 +2023-03-31 21:58:30,522 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.57 vs. limit=2.0 +2023-03-31 21:58:32,326 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=8998.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:58:40,102 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9004.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:58:49,979 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.755e+02 7.392e+02 9.171e+02 1.181e+03 2.165e+03, threshold=1.834e+03, percent-clipped=0.0 +2023-03-31 21:59:04,434 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9023.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:59:11,992 INFO [train.py:903] (0/4) Epoch 2, batch 2200, loss[loss=0.333, simple_loss=0.3716, pruned_loss=0.1472, over 19770.00 frames. ], tot_loss[loss=0.3668, simple_loss=0.4, pruned_loss=0.1668, over 3824046.26 frames. ], batch size: 48, lr: 3.41e-02, grad_scale: 8.0 +2023-03-31 21:59:14,465 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9031.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 21:59:14,585 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4665, 2.1338, 1.6080, 1.6203, 1.6030, 1.4751, 0.3069, 1.0612], + device='cuda:0'), covar=tensor([0.0477, 0.0404, 0.0335, 0.0484, 0.0863, 0.0642, 0.1132, 0.0796], + device='cuda:0'), in_proj_covar=tensor([0.0204, 0.0208, 0.0201, 0.0245, 0.0286, 0.0253, 0.0256, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 22:00:11,531 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-03-31 22:00:13,986 INFO [train.py:903] (0/4) Epoch 2, batch 2250, loss[loss=0.4021, simple_loss=0.4317, pruned_loss=0.1863, over 18681.00 frames. ], tot_loss[loss=0.3683, simple_loss=0.4017, pruned_loss=0.1674, over 3813707.48 frames. ], batch size: 74, lr: 3.41e-02, grad_scale: 8.0 +2023-03-31 22:00:56,131 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.993e+02 8.727e+02 9.943e+02 1.293e+03 2.077e+03, threshold=1.989e+03, percent-clipped=4.0 +2023-03-31 22:01:08,135 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9122.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:01:17,887 INFO [train.py:903] (0/4) Epoch 2, batch 2300, loss[loss=0.3751, simple_loss=0.4078, pruned_loss=0.1711, over 19673.00 frames. ], tot_loss[loss=0.3665, simple_loss=0.4007, pruned_loss=0.1662, over 3829350.59 frames. ], batch size: 59, lr: 3.40e-02, grad_scale: 8.0 +2023-03-31 22:01:29,610 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6821, 1.8317, 2.2615, 1.9485, 3.0236, 3.4419, 3.4705, 3.5562], + device='cuda:0'), covar=tensor([0.1586, 0.2012, 0.2008, 0.1635, 0.0517, 0.0143, 0.0170, 0.0151], + device='cuda:0'), in_proj_covar=tensor([0.0292, 0.0280, 0.0323, 0.0288, 0.0197, 0.0107, 0.0188, 0.0114], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-03-31 22:01:30,529 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-03-31 22:02:11,734 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8288, 1.1758, 1.3524, 1.5425, 2.4692, 1.1540, 1.7629, 2.4146], + device='cuda:0'), covar=tensor([0.0576, 0.2980, 0.2610, 0.1657, 0.0576, 0.2199, 0.1126, 0.0685], + device='cuda:0'), in_proj_covar=tensor([0.0220, 0.0302, 0.0280, 0.0269, 0.0250, 0.0318, 0.0257, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 22:02:19,087 INFO [train.py:903] (0/4) Epoch 2, batch 2350, loss[loss=0.4073, simple_loss=0.4275, pruned_loss=0.1935, over 18757.00 frames. ], tot_loss[loss=0.368, simple_loss=0.4014, pruned_loss=0.1673, over 3818127.56 frames. ], batch size: 74, lr: 3.39e-02, grad_scale: 8.0 +2023-03-31 22:03:00,735 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.988e+02 9.117e+02 1.090e+03 1.432e+03 2.529e+03, threshold=2.180e+03, percent-clipped=5.0 +2023-03-31 22:03:00,800 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-03-31 22:03:18,246 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-03-31 22:03:22,888 INFO [train.py:903] (0/4) Epoch 2, batch 2400, loss[loss=0.4009, simple_loss=0.4336, pruned_loss=0.1841, over 19504.00 frames. ], tot_loss[loss=0.3706, simple_loss=0.4035, pruned_loss=0.1689, over 3813640.02 frames. ], batch size: 64, lr: 3.38e-02, grad_scale: 8.0 +2023-03-31 22:03:32,266 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9237.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:04:24,306 INFO [train.py:903] (0/4) Epoch 2, batch 2450, loss[loss=0.4605, simple_loss=0.4584, pruned_loss=0.2314, over 13617.00 frames. ], tot_loss[loss=0.3701, simple_loss=0.4031, pruned_loss=0.1685, over 3808790.52 frames. ], batch size: 135, lr: 3.38e-02, grad_scale: 8.0 +2023-03-31 22:04:38,145 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9289.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:04:57,186 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-03-31 22:05:06,391 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.471e+02 8.791e+02 1.096e+03 1.484e+03 3.289e+03, threshold=2.192e+03, percent-clipped=7.0 +2023-03-31 22:05:27,995 INFO [train.py:903] (0/4) Epoch 2, batch 2500, loss[loss=0.3447, simple_loss=0.3936, pruned_loss=0.1479, over 19537.00 frames. ], tot_loss[loss=0.3691, simple_loss=0.4028, pruned_loss=0.1677, over 3818691.58 frames. ], batch size: 56, lr: 3.37e-02, grad_scale: 8.0 +2023-03-31 22:05:52,423 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9348.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:06:25,085 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9375.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:06:29,686 INFO [train.py:903] (0/4) Epoch 2, batch 2550, loss[loss=0.4652, simple_loss=0.4568, pruned_loss=0.2368, over 13373.00 frames. ], tot_loss[loss=0.3689, simple_loss=0.4019, pruned_loss=0.1679, over 3813956.48 frames. ], batch size: 136, lr: 3.36e-02, grad_scale: 8.0 +2023-03-31 22:07:01,775 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9404.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:07:11,556 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.591e+02 8.134e+02 9.492e+02 1.275e+03 2.544e+03, threshold=1.898e+03, percent-clipped=3.0 +2023-03-31 22:07:12,954 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.8418, 5.1677, 2.3863, 4.6086, 1.4547, 5.2615, 5.0664, 5.1846], + device='cuda:0'), covar=tensor([0.0509, 0.1151, 0.2403, 0.0628, 0.3767, 0.0760, 0.0534, 0.0623], + device='cuda:0'), in_proj_covar=tensor([0.0294, 0.0280, 0.0314, 0.0259, 0.0328, 0.0267, 0.0214, 0.0233], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 22:07:25,225 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-03-31 22:07:33,274 INFO [train.py:903] (0/4) Epoch 2, batch 2600, loss[loss=0.3651, simple_loss=0.4067, pruned_loss=0.1617, over 19770.00 frames. ], tot_loss[loss=0.3664, simple_loss=0.4004, pruned_loss=0.1661, over 3832392.05 frames. ], batch size: 56, lr: 3.36e-02, grad_scale: 8.0 +2023-03-31 22:07:36,924 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9432.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:08:14,519 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9463.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:08:34,691 INFO [train.py:903] (0/4) Epoch 2, batch 2650, loss[loss=0.3679, simple_loss=0.4088, pruned_loss=0.1635, over 19250.00 frames. ], tot_loss[loss=0.3673, simple_loss=0.4011, pruned_loss=0.1668, over 3823923.92 frames. ], batch size: 66, lr: 3.35e-02, grad_scale: 8.0 +2023-03-31 22:08:49,014 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9490.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:08:52,630 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6156, 1.2220, 1.5941, 1.2154, 2.8166, 3.3828, 3.4791, 3.5928], + device='cuda:0'), covar=tensor([0.1504, 0.2748, 0.2634, 0.2242, 0.0407, 0.0145, 0.0170, 0.0153], + device='cuda:0'), in_proj_covar=tensor([0.0292, 0.0283, 0.0321, 0.0292, 0.0194, 0.0108, 0.0183, 0.0115], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-03-31 22:08:52,710 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9493.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:08:53,487 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-03-31 22:09:09,376 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.83 vs. limit=5.0 +2023-03-31 22:09:16,684 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.258e+02 8.887e+02 1.023e+03 1.383e+03 3.476e+03, threshold=2.047e+03, percent-clipped=7.0 +2023-03-31 22:09:24,004 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9518.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:09:36,963 INFO [train.py:903] (0/4) Epoch 2, batch 2700, loss[loss=0.4111, simple_loss=0.44, pruned_loss=0.1911, over 19690.00 frames. ], tot_loss[loss=0.3652, simple_loss=0.3995, pruned_loss=0.1655, over 3841283.22 frames. ], batch size: 60, lr: 3.34e-02, grad_scale: 8.0 +2023-03-31 22:10:39,553 INFO [train.py:903] (0/4) Epoch 2, batch 2750, loss[loss=0.345, simple_loss=0.3847, pruned_loss=0.1526, over 19753.00 frames. ], tot_loss[loss=0.366, simple_loss=0.4, pruned_loss=0.166, over 3839921.12 frames. ], batch size: 54, lr: 3.34e-02, grad_scale: 8.0 +2023-03-31 22:11:08,785 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9602.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:11:20,990 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.046e+02 9.047e+02 1.065e+03 1.297e+03 2.590e+03, threshold=2.130e+03, percent-clipped=3.0 +2023-03-31 22:11:43,322 INFO [train.py:903] (0/4) Epoch 2, batch 2800, loss[loss=0.4125, simple_loss=0.4196, pruned_loss=0.2027, over 19665.00 frames. ], tot_loss[loss=0.364, simple_loss=0.3984, pruned_loss=0.1648, over 3835929.53 frames. ], batch size: 53, lr: 3.33e-02, grad_scale: 8.0 +2023-03-31 22:11:56,240 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0419, 1.8546, 1.8612, 2.8072, 2.0373, 2.6834, 2.1398, 1.8532], + device='cuda:0'), covar=tensor([0.1087, 0.0946, 0.0636, 0.0529, 0.1082, 0.0326, 0.1191, 0.0869], + device='cuda:0'), in_proj_covar=tensor([0.0323, 0.0296, 0.0315, 0.0407, 0.0385, 0.0214, 0.0414, 0.0325], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:12:20,914 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9660.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:12:41,292 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9676.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:12:44,609 INFO [train.py:903] (0/4) Epoch 2, batch 2850, loss[loss=0.3159, simple_loss=0.3532, pruned_loss=0.1393, over 19730.00 frames. ], tot_loss[loss=0.3628, simple_loss=0.3972, pruned_loss=0.1641, over 3840014.17 frames. ], batch size: 46, lr: 3.32e-02, grad_scale: 8.0 +2023-03-31 22:12:51,698 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9685.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:13:26,587 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 6.408e+02 8.898e+02 1.125e+03 1.384e+03 2.599e+03, threshold=2.251e+03, percent-clipped=6.0 +2023-03-31 22:13:35,219 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9719.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:13:45,574 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-03-31 22:13:46,841 INFO [train.py:903] (0/4) Epoch 2, batch 2900, loss[loss=0.3726, simple_loss=0.4078, pruned_loss=0.1687, over 19778.00 frames. ], tot_loss[loss=0.3632, simple_loss=0.398, pruned_loss=0.1642, over 3828873.78 frames. ], batch size: 56, lr: 3.31e-02, grad_scale: 16.0 +2023-03-31 22:13:58,075 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.34 vs. limit=5.0 +2023-03-31 22:14:05,839 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9744.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:14:09,036 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=9746.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:14:27,836 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=9761.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:14:39,277 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=9771.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:14:45,585 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9776.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:14:49,525 INFO [train.py:903] (0/4) Epoch 2, batch 2950, loss[loss=0.414, simple_loss=0.433, pruned_loss=0.1975, over 19365.00 frames. ], tot_loss[loss=0.3652, simple_loss=0.3996, pruned_loss=0.1654, over 3827686.46 frames. ], batch size: 70, lr: 3.31e-02, grad_scale: 8.0 +2023-03-31 22:15:31,703 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.533e+02 8.475e+02 1.131e+03 1.413e+03 3.215e+03, threshold=2.262e+03, percent-clipped=4.0 +2023-03-31 22:15:53,069 INFO [train.py:903] (0/4) Epoch 2, batch 3000, loss[loss=0.4689, simple_loss=0.4646, pruned_loss=0.2366, over 13128.00 frames. ], tot_loss[loss=0.3656, simple_loss=0.3999, pruned_loss=0.1656, over 3835819.41 frames. ], batch size: 137, lr: 3.30e-02, grad_scale: 4.0 +2023-03-31 22:15:53,070 INFO [train.py:928] (0/4) Computing validation loss +2023-03-31 22:16:04,424 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7867, 1.2247, 1.4847, 1.6075, 2.5024, 1.2682, 1.8054, 2.4733], + device='cuda:0'), covar=tensor([0.0580, 0.3315, 0.3190, 0.2045, 0.0596, 0.2661, 0.1274, 0.0745], + device='cuda:0'), in_proj_covar=tensor([0.0218, 0.0293, 0.0284, 0.0271, 0.0242, 0.0309, 0.0252, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 22:16:06,238 INFO [train.py:937] (0/4) Epoch 2, validation: loss=0.2513, simple_loss=0.3423, pruned_loss=0.08019, over 944034.00 frames. +2023-03-31 22:16:06,240 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18153MB +2023-03-31 22:16:12,106 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-03-31 22:16:53,686 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0664, 1.1118, 1.9409, 1.3096, 2.6445, 2.4787, 3.0451, 1.3158], + device='cuda:0'), covar=tensor([0.1378, 0.2011, 0.1099, 0.1310, 0.0833, 0.0783, 0.0999, 0.1787], + device='cuda:0'), in_proj_covar=tensor([0.0349, 0.0372, 0.0340, 0.0345, 0.0396, 0.0318, 0.0465, 0.0355], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:17:08,087 INFO [train.py:903] (0/4) Epoch 2, batch 3050, loss[loss=0.3934, simple_loss=0.4155, pruned_loss=0.1857, over 13285.00 frames. ], tot_loss[loss=0.3646, simple_loss=0.399, pruned_loss=0.1651, over 3832343.06 frames. ], batch size: 137, lr: 3.29e-02, grad_scale: 4.0 +2023-03-31 22:17:22,786 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=9891.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:17:50,735 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.558e+02 8.806e+02 1.158e+03 1.402e+03 3.282e+03, threshold=2.315e+03, percent-clipped=2.0 +2023-03-31 22:18:10,032 INFO [train.py:903] (0/4) Epoch 2, batch 3100, loss[loss=0.4447, simple_loss=0.4463, pruned_loss=0.2215, over 13484.00 frames. ], tot_loss[loss=0.3654, simple_loss=0.3993, pruned_loss=0.1657, over 3814990.23 frames. ], batch size: 135, lr: 3.29e-02, grad_scale: 4.0 +2023-03-31 22:18:29,665 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=9946.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:19:11,238 INFO [train.py:903] (0/4) Epoch 2, batch 3150, loss[loss=0.3929, simple_loss=0.418, pruned_loss=0.1839, over 19483.00 frames. ], tot_loss[loss=0.3645, simple_loss=0.3992, pruned_loss=0.1649, over 3823152.07 frames. ], batch size: 64, lr: 3.28e-02, grad_scale: 4.0 +2023-03-31 22:19:39,329 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-10000.pt +2023-03-31 22:19:42,630 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-03-31 22:19:48,737 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7861, 3.8843, 4.3066, 4.1500, 1.3647, 3.7845, 3.4299, 3.8068], + device='cuda:0'), covar=tensor([0.0362, 0.0493, 0.0412, 0.0288, 0.3150, 0.0212, 0.0363, 0.0832], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0278, 0.0367, 0.0265, 0.0413, 0.0178, 0.0252, 0.0371], + device='cuda:0'), out_proj_covar=tensor([0.0001, 0.0002, 0.0002, 0.0001, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 22:19:50,333 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.05 vs. limit=2.0 +2023-03-31 22:19:56,512 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.945e+02 7.345e+02 9.404e+02 1.247e+03 3.615e+03, threshold=1.881e+03, percent-clipped=3.0 +2023-03-31 22:20:03,763 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10020.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:20:15,208 INFO [train.py:903] (0/4) Epoch 2, batch 3200, loss[loss=0.3473, simple_loss=0.3856, pruned_loss=0.1545, over 19579.00 frames. ], tot_loss[loss=0.3638, simple_loss=0.3989, pruned_loss=0.1644, over 3821589.12 frames. ], batch size: 52, lr: 3.27e-02, grad_scale: 8.0 +2023-03-31 22:20:50,849 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10057.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:20:55,736 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10061.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:21:19,931 INFO [train.py:903] (0/4) Epoch 2, batch 3250, loss[loss=0.4166, simple_loss=0.4348, pruned_loss=0.1992, over 19664.00 frames. ], tot_loss[loss=0.3634, simple_loss=0.3989, pruned_loss=0.1639, over 3826267.97 frames. ], batch size: 53, lr: 3.27e-02, grad_scale: 8.0 +2023-03-31 22:21:21,258 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10080.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 22:21:41,233 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3190, 1.4532, 1.4850, 1.9350, 2.9991, 1.2752, 1.9238, 2.8162], + device='cuda:0'), covar=tensor([0.0339, 0.2628, 0.2782, 0.1461, 0.0451, 0.2252, 0.1306, 0.0541], + device='cuda:0'), in_proj_covar=tensor([0.0218, 0.0299, 0.0287, 0.0272, 0.0246, 0.0315, 0.0253, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:21:50,583 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10105.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:22:02,472 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.757e+02 8.486e+02 1.037e+03 1.291e+03 3.604e+03, threshold=2.074e+03, percent-clipped=6.0 +2023-03-31 22:22:20,719 INFO [train.py:903] (0/4) Epoch 2, batch 3300, loss[loss=0.3509, simple_loss=0.3913, pruned_loss=0.1552, over 19728.00 frames. ], tot_loss[loss=0.3633, simple_loss=0.3987, pruned_loss=0.1639, over 3839872.92 frames. ], batch size: 51, lr: 3.26e-02, grad_scale: 8.0 +2023-03-31 22:22:25,200 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-03-31 22:22:27,887 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10135.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:22:42,679 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10147.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:23:14,925 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10172.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:23:22,721 INFO [train.py:903] (0/4) Epoch 2, batch 3350, loss[loss=0.3901, simple_loss=0.4221, pruned_loss=0.179, over 19518.00 frames. ], tot_loss[loss=0.3628, simple_loss=0.3983, pruned_loss=0.1636, over 3839546.30 frames. ], batch size: 54, lr: 3.26e-02, grad_scale: 8.0 +2023-03-31 22:23:51,296 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9639, 1.2507, 1.5379, 1.9932, 1.5275, 1.9626, 2.1140, 1.9016], + device='cuda:0'), covar=tensor([0.0822, 0.1620, 0.1455, 0.1170, 0.1655, 0.1001, 0.1265, 0.0756], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0309, 0.0304, 0.0323, 0.0363, 0.0269, 0.0327, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 22:24:07,629 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.516e+02 8.361e+02 9.909e+02 1.198e+03 2.844e+03, threshold=1.982e+03, percent-clipped=3.0 +2023-03-31 22:24:14,899 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6936, 1.8912, 1.7108, 2.5064, 4.3641, 1.4395, 2.1960, 3.9048], + device='cuda:0'), covar=tensor([0.0281, 0.2536, 0.2522, 0.1578, 0.0315, 0.2274, 0.1127, 0.0466], + device='cuda:0'), in_proj_covar=tensor([0.0217, 0.0298, 0.0286, 0.0274, 0.0247, 0.0316, 0.0254, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:24:14,941 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10220.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:24:25,935 INFO [train.py:903] (0/4) Epoch 2, batch 3400, loss[loss=0.3441, simple_loss=0.397, pruned_loss=0.1456, over 19337.00 frames. ], tot_loss[loss=0.3617, simple_loss=0.3979, pruned_loss=0.1627, over 3832583.61 frames. ], batch size: 66, lr: 3.25e-02, grad_scale: 8.0 +2023-03-31 22:25:29,366 INFO [train.py:903] (0/4) Epoch 2, batch 3450, loss[loss=0.3218, simple_loss=0.3718, pruned_loss=0.1359, over 19673.00 frames. ], tot_loss[loss=0.3639, simple_loss=0.3996, pruned_loss=0.1642, over 3834807.66 frames. ], batch size: 53, lr: 3.24e-02, grad_scale: 4.0 +2023-03-31 22:25:32,668 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-03-31 22:26:13,346 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.771e+02 9.409e+02 1.166e+03 1.453e+03 2.796e+03, threshold=2.333e+03, percent-clipped=9.0 +2023-03-31 22:26:17,036 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10317.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:26:31,778 INFO [train.py:903] (0/4) Epoch 2, batch 3500, loss[loss=0.3795, simple_loss=0.3963, pruned_loss=0.1814, over 19356.00 frames. ], tot_loss[loss=0.3644, simple_loss=0.3997, pruned_loss=0.1645, over 3821311.26 frames. ], batch size: 47, lr: 3.24e-02, grad_scale: 4.0 +2023-03-31 22:26:47,007 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10342.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:27:33,760 INFO [train.py:903] (0/4) Epoch 2, batch 3550, loss[loss=0.3438, simple_loss=0.393, pruned_loss=0.1473, over 19787.00 frames. ], tot_loss[loss=0.3605, simple_loss=0.3968, pruned_loss=0.1621, over 3815299.35 frames. ], batch size: 56, lr: 3.23e-02, grad_scale: 4.0 +2023-03-31 22:27:49,477 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10391.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:27:50,564 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3588, 1.0142, 1.4067, 0.3459, 2.6952, 2.3015, 2.0539, 2.4480], + device='cuda:0'), covar=tensor([0.1224, 0.2681, 0.2640, 0.2284, 0.0281, 0.0169, 0.0377, 0.0186], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0274, 0.0316, 0.0280, 0.0191, 0.0101, 0.0184, 0.0110], + device='cuda:0'), out_proj_covar=tensor([2.6526e-04, 2.7567e-04, 2.9945e-04, 2.7519e-04, 2.0554e-04, 9.9056e-05, + 1.7069e-04, 1.1471e-04], device='cuda:0') +2023-03-31 22:28:04,073 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10401.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:28:20,155 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.354e+02 7.776e+02 1.013e+03 1.369e+03 3.978e+03, threshold=2.027e+03, percent-clipped=2.0 +2023-03-31 22:28:21,557 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10416.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:28:30,580 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10424.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 22:28:35,007 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10427.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:28:37,126 INFO [train.py:903] (0/4) Epoch 2, batch 3600, loss[loss=0.3096, simple_loss=0.3663, pruned_loss=0.1265, over 19601.00 frames. ], tot_loss[loss=0.3585, simple_loss=0.3958, pruned_loss=0.1606, over 3822131.71 frames. ], batch size: 57, lr: 3.22e-02, grad_scale: 8.0 +2023-03-31 22:29:32,507 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10472.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:29:37,366 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10476.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:29:41,348 INFO [train.py:903] (0/4) Epoch 2, batch 3650, loss[loss=0.4036, simple_loss=0.4334, pruned_loss=0.1869, over 19753.00 frames. ], tot_loss[loss=0.3558, simple_loss=0.3935, pruned_loss=0.1591, over 3835726.81 frames. ], batch size: 63, lr: 3.22e-02, grad_scale: 8.0 +2023-03-31 22:29:48,911 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-31 22:30:09,087 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10501.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:30:26,009 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.209e+02 8.454e+02 1.072e+03 1.396e+03 2.688e+03, threshold=2.143e+03, percent-clipped=6.0 +2023-03-31 22:30:27,454 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10516.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:30:45,121 INFO [train.py:903] (0/4) Epoch 2, batch 3700, loss[loss=0.3153, simple_loss=0.3689, pruned_loss=0.1308, over 19417.00 frames. ], tot_loss[loss=0.3574, simple_loss=0.3946, pruned_loss=0.1602, over 3841434.90 frames. ], batch size: 48, lr: 3.21e-02, grad_scale: 8.0 +2023-03-31 22:30:51,539 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0174, 0.9905, 2.0680, 1.2399, 2.7113, 2.6590, 3.0486, 1.2613], + device='cuda:0'), covar=tensor([0.1417, 0.2074, 0.1080, 0.1349, 0.0717, 0.0739, 0.0955, 0.1848], + device='cuda:0'), in_proj_covar=tensor([0.0351, 0.0373, 0.0345, 0.0346, 0.0397, 0.0327, 0.0474, 0.0359], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:30:57,266 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10539.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 22:31:47,575 INFO [train.py:903] (0/4) Epoch 2, batch 3750, loss[loss=0.4319, simple_loss=0.4514, pruned_loss=0.2062, over 19705.00 frames. ], tot_loss[loss=0.3556, simple_loss=0.3935, pruned_loss=0.1589, over 3845011.67 frames. ], batch size: 59, lr: 3.20e-02, grad_scale: 8.0 +2023-03-31 22:32:33,531 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.274e+02 8.986e+02 1.057e+03 1.326e+03 2.585e+03, threshold=2.114e+03, percent-clipped=3.0 +2023-03-31 22:32:50,179 INFO [train.py:903] (0/4) Epoch 2, batch 3800, loss[loss=0.5113, simple_loss=0.4989, pruned_loss=0.2619, over 19502.00 frames. ], tot_loss[loss=0.3564, simple_loss=0.3936, pruned_loss=0.1596, over 3843504.08 frames. ], batch size: 64, lr: 3.20e-02, grad_scale: 8.0 +2023-03-31 22:33:19,196 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.68 vs. limit=2.0 +2023-03-31 22:33:23,948 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-03-31 22:33:51,757 INFO [train.py:903] (0/4) Epoch 2, batch 3850, loss[loss=0.3575, simple_loss=0.4012, pruned_loss=0.1569, over 19656.00 frames. ], tot_loss[loss=0.3561, simple_loss=0.3932, pruned_loss=0.1595, over 3830941.12 frames. ], batch size: 58, lr: 3.19e-02, grad_scale: 8.0 +2023-03-31 22:34:37,544 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.106e+02 8.604e+02 1.077e+03 1.442e+03 2.985e+03, threshold=2.155e+03, percent-clipped=6.0 +2023-03-31 22:34:56,856 INFO [train.py:903] (0/4) Epoch 2, batch 3900, loss[loss=0.3087, simple_loss=0.3677, pruned_loss=0.1249, over 19590.00 frames. ], tot_loss[loss=0.3567, simple_loss=0.3938, pruned_loss=0.1598, over 3824548.44 frames. ], batch size: 52, lr: 3.19e-02, grad_scale: 8.0 +2023-03-31 22:35:36,172 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10762.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:35:43,253 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-03-31 22:35:49,448 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10771.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:35:50,801 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10772.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:35:58,736 INFO [train.py:903] (0/4) Epoch 2, batch 3950, loss[loss=0.3444, simple_loss=0.3965, pruned_loss=0.1461, over 19784.00 frames. ], tot_loss[loss=0.3566, simple_loss=0.3938, pruned_loss=0.1597, over 3818501.29 frames. ], batch size: 56, lr: 3.18e-02, grad_scale: 8.0 +2023-03-31 22:36:04,539 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-03-31 22:36:15,281 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10792.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:36:18,964 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=10795.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 22:36:21,098 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10797.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:36:44,827 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.316e+02 7.972e+02 1.008e+03 1.220e+03 2.629e+03, threshold=2.016e+03, percent-clipped=1.0 +2023-03-31 22:36:46,226 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=10816.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:36:51,361 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=10820.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 22:37:01,757 INFO [train.py:903] (0/4) Epoch 2, batch 4000, loss[loss=0.3128, simple_loss=0.3544, pruned_loss=0.1356, over 19741.00 frames. ], tot_loss[loss=0.3547, simple_loss=0.3923, pruned_loss=0.1586, over 3824394.88 frames. ], batch size: 51, lr: 3.17e-02, grad_scale: 8.0 +2023-03-31 22:37:49,240 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-03-31 22:38:04,673 INFO [train.py:903] (0/4) Epoch 2, batch 4050, loss[loss=0.3351, simple_loss=0.3884, pruned_loss=0.1409, over 19563.00 frames. ], tot_loss[loss=0.3572, simple_loss=0.3943, pruned_loss=0.16, over 3818591.19 frames. ], batch size: 61, lr: 3.17e-02, grad_scale: 8.0 +2023-03-31 22:38:15,742 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10886.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:38:35,397 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=10903.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 22:38:49,942 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.194e+02 9.186e+02 1.101e+03 1.338e+03 4.215e+03, threshold=2.201e+03, percent-clipped=7.0 +2023-03-31 22:39:10,606 INFO [train.py:903] (0/4) Epoch 2, batch 4100, loss[loss=0.3089, simple_loss=0.3458, pruned_loss=0.136, over 19744.00 frames. ], tot_loss[loss=0.3548, simple_loss=0.393, pruned_loss=0.1584, over 3818074.13 frames. ], batch size: 47, lr: 3.16e-02, grad_scale: 8.0 +2023-03-31 22:39:13,344 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=10931.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:39:45,312 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-03-31 22:39:54,833 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6934, 1.4247, 1.4349, 2.0925, 1.8395, 1.6253, 1.5324, 1.9039], + device='cuda:0'), covar=tensor([0.0871, 0.1724, 0.1234, 0.0825, 0.1139, 0.0615, 0.1007, 0.0581], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0374, 0.0280, 0.0254, 0.0310, 0.0266, 0.0276, 0.0227], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:40:04,230 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5287, 1.2616, 1.4012, 2.1353, 3.1007, 1.8425, 2.2264, 2.8849], + device='cuda:0'), covar=tensor([0.0519, 0.3082, 0.2699, 0.1431, 0.0522, 0.1997, 0.1496, 0.0672], + device='cuda:0'), in_proj_covar=tensor([0.0223, 0.0305, 0.0286, 0.0280, 0.0259, 0.0321, 0.0262, 0.0269], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:40:13,106 INFO [train.py:903] (0/4) Epoch 2, batch 4150, loss[loss=0.3726, simple_loss=0.4132, pruned_loss=0.166, over 17288.00 frames. ], tot_loss[loss=0.3556, simple_loss=0.3939, pruned_loss=0.1587, over 3808129.85 frames. ], batch size: 101, lr: 3.16e-02, grad_scale: 8.0 +2023-03-31 22:40:59,342 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.653e+02 7.731e+02 1.007e+03 1.258e+03 2.097e+03, threshold=2.015e+03, percent-clipped=0.0 +2023-03-31 22:41:15,490 INFO [train.py:903] (0/4) Epoch 2, batch 4200, loss[loss=0.4689, simple_loss=0.461, pruned_loss=0.2384, over 12750.00 frames. ], tot_loss[loss=0.3572, simple_loss=0.3947, pruned_loss=0.1599, over 3800510.25 frames. ], batch size: 137, lr: 3.15e-02, grad_scale: 8.0 +2023-03-31 22:41:18,940 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-03-31 22:41:27,259 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6802, 1.2859, 1.2288, 1.8293, 1.5767, 1.5048, 1.4352, 1.6280], + device='cuda:0'), covar=tensor([0.0844, 0.1718, 0.1422, 0.0893, 0.1130, 0.0621, 0.1016, 0.0639], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0383, 0.0287, 0.0255, 0.0314, 0.0264, 0.0281, 0.0229], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:42:10,710 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8387, 1.2844, 1.2514, 1.8828, 1.5133, 1.9197, 2.1259, 1.9991], + device='cuda:0'), covar=tensor([0.0941, 0.1454, 0.1794, 0.1316, 0.1742, 0.1034, 0.1271, 0.0837], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0305, 0.0294, 0.0322, 0.0354, 0.0262, 0.0320, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 22:42:18,010 INFO [train.py:903] (0/4) Epoch 2, batch 4250, loss[loss=0.4325, simple_loss=0.4407, pruned_loss=0.2121, over 19541.00 frames. ], tot_loss[loss=0.3565, simple_loss=0.394, pruned_loss=0.1595, over 3795109.10 frames. ], batch size: 56, lr: 3.14e-02, grad_scale: 8.0 +2023-03-31 22:42:35,128 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-03-31 22:42:45,402 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-03-31 22:42:52,660 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11106.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:43:03,987 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.611e+02 8.485e+02 1.107e+03 1.406e+03 3.284e+03, threshold=2.214e+03, percent-clipped=7.0 +2023-03-31 22:43:21,882 INFO [train.py:903] (0/4) Epoch 2, batch 4300, loss[loss=0.4145, simple_loss=0.4359, pruned_loss=0.1965, over 18268.00 frames. ], tot_loss[loss=0.3533, simple_loss=0.3916, pruned_loss=0.1575, over 3813621.99 frames. ], batch size: 83, lr: 3.14e-02, grad_scale: 8.0 +2023-03-31 22:43:30,381 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11136.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:43:37,578 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11142.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:43:45,978 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.95 vs. limit=5.0 +2023-03-31 22:43:50,382 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.36 vs. limit=5.0 +2023-03-31 22:43:53,033 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-03-31 22:44:08,988 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11167.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:44:16,377 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-03-31 22:44:23,251 INFO [train.py:903] (0/4) Epoch 2, batch 4350, loss[loss=0.4015, simple_loss=0.4231, pruned_loss=0.19, over 13087.00 frames. ], tot_loss[loss=0.3515, simple_loss=0.3908, pruned_loss=0.1561, over 3824043.90 frames. ], batch size: 136, lr: 3.13e-02, grad_scale: 8.0 +2023-03-31 22:44:32,827 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11187.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:44:39,297 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11192.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:45:04,729 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11212.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:45:09,081 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.029e+02 7.841e+02 9.559e+02 1.160e+03 2.939e+03, threshold=1.912e+03, percent-clipped=2.0 +2023-03-31 22:45:15,635 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11221.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:45:24,371 INFO [train.py:903] (0/4) Epoch 2, batch 4400, loss[loss=0.335, simple_loss=0.3652, pruned_loss=0.1524, over 19732.00 frames. ], tot_loss[loss=0.3541, simple_loss=0.3927, pruned_loss=0.1578, over 3815346.00 frames. ], batch size: 46, lr: 3.13e-02, grad_scale: 8.0 +2023-03-31 22:45:48,410 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11247.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 22:45:50,515 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-03-31 22:45:54,201 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11251.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:46:00,721 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-03-31 22:46:20,829 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3996, 1.1349, 1.3105, 2.0173, 1.6066, 1.6850, 2.7828, 1.7099], + device='cuda:0'), covar=tensor([0.0993, 0.2296, 0.2072, 0.1819, 0.2112, 0.1806, 0.1475, 0.1393], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0303, 0.0295, 0.0329, 0.0353, 0.0265, 0.0318, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 22:46:27,487 INFO [train.py:903] (0/4) Epoch 2, batch 4450, loss[loss=0.3942, simple_loss=0.4188, pruned_loss=0.1848, over 19764.00 frames. ], tot_loss[loss=0.3562, simple_loss=0.3945, pruned_loss=0.159, over 3829713.03 frames. ], batch size: 54, lr: 3.12e-02, grad_scale: 8.0 +2023-03-31 22:47:14,189 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.163e+02 8.820e+02 1.081e+03 1.372e+03 2.333e+03, threshold=2.162e+03, percent-clipped=5.0 +2023-03-31 22:47:31,602 INFO [train.py:903] (0/4) Epoch 2, batch 4500, loss[loss=0.3639, simple_loss=0.405, pruned_loss=0.1614, over 19517.00 frames. ], tot_loss[loss=0.3528, simple_loss=0.3917, pruned_loss=0.1569, over 3829811.04 frames. ], batch size: 54, lr: 3.12e-02, grad_scale: 8.0 +2023-03-31 22:47:35,926 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.16 vs. limit=2.0 +2023-03-31 22:48:12,940 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11362.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 22:48:14,066 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0136, 1.3867, 1.2528, 1.9328, 1.4896, 1.8500, 1.9387, 1.8331], + device='cuda:0'), covar=tensor([0.0694, 0.1355, 0.1506, 0.1121, 0.1485, 0.1119, 0.1280, 0.0739], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0306, 0.0294, 0.0326, 0.0349, 0.0267, 0.0320, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0003], + device='cuda:0') +2023-03-31 22:48:34,554 INFO [train.py:903] (0/4) Epoch 2, batch 4550, loss[loss=0.3699, simple_loss=0.4069, pruned_loss=0.1665, over 19784.00 frames. ], tot_loss[loss=0.3533, simple_loss=0.3919, pruned_loss=0.1573, over 3831977.74 frames. ], batch size: 56, lr: 3.11e-02, grad_scale: 8.0 +2023-03-31 22:48:45,431 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-03-31 22:49:08,241 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-03-31 22:49:21,599 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.316e+02 7.788e+02 1.003e+03 1.220e+03 2.356e+03, threshold=2.005e+03, percent-clipped=1.0 +2023-03-31 22:49:37,109 INFO [train.py:903] (0/4) Epoch 2, batch 4600, loss[loss=0.3891, simple_loss=0.4163, pruned_loss=0.1809, over 17265.00 frames. ], tot_loss[loss=0.3522, simple_loss=0.3914, pruned_loss=0.1565, over 3841315.25 frames. ], batch size: 101, lr: 3.10e-02, grad_scale: 8.0 +2023-03-31 22:50:06,616 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4153, 1.7419, 1.7933, 1.9104, 3.0457, 4.5655, 4.7153, 5.0952], + device='cuda:0'), covar=tensor([0.1025, 0.2180, 0.2471, 0.1630, 0.0373, 0.0091, 0.0095, 0.0053], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0273, 0.0321, 0.0275, 0.0187, 0.0103, 0.0183, 0.0106], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-03-31 22:50:19,685 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1767, 2.0778, 1.8875, 3.0512, 2.2120, 3.2429, 2.4081, 1.7014], + device='cuda:0'), covar=tensor([0.0914, 0.0723, 0.0470, 0.0495, 0.0873, 0.0190, 0.0842, 0.0793], + device='cuda:0'), in_proj_covar=tensor([0.0366, 0.0327, 0.0346, 0.0452, 0.0412, 0.0244, 0.0454, 0.0357], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0001, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:50:37,512 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11477.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:50:39,645 INFO [train.py:903] (0/4) Epoch 2, batch 4650, loss[loss=0.305, simple_loss=0.3559, pruned_loss=0.1271, over 19656.00 frames. ], tot_loss[loss=0.3538, simple_loss=0.3925, pruned_loss=0.1576, over 3827373.55 frames. ], batch size: 53, lr: 3.10e-02, grad_scale: 8.0 +2023-03-31 22:50:59,945 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-03-31 22:51:09,648 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11502.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:51:10,451 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-03-31 22:51:15,594 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11507.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:51:26,520 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.039e+02 8.212e+02 1.126e+03 1.409e+03 2.689e+03, threshold=2.252e+03, percent-clipped=6.0 +2023-03-31 22:51:42,949 INFO [train.py:903] (0/4) Epoch 2, batch 4700, loss[loss=0.3552, simple_loss=0.4026, pruned_loss=0.1539, over 18214.00 frames. ], tot_loss[loss=0.3503, simple_loss=0.3905, pruned_loss=0.1551, over 3831038.33 frames. ], batch size: 84, lr: 3.09e-02, grad_scale: 8.0 +2023-03-31 22:51:47,850 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11532.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:51:52,264 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11536.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:52:04,173 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8689, 1.3493, 1.4200, 2.1371, 1.3319, 1.7646, 1.7284, 1.8385], + device='cuda:0'), covar=tensor([0.0761, 0.1890, 0.1319, 0.0813, 0.1390, 0.0591, 0.0838, 0.0612], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0379, 0.0295, 0.0258, 0.0324, 0.0264, 0.0278, 0.0227], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 22:52:05,035 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-03-31 22:52:08,641 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11550.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:52:18,079 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7306, 1.2987, 1.3305, 1.8349, 1.4740, 1.9996, 2.0428, 1.6372], + device='cuda:0'), covar=tensor([0.0964, 0.1448, 0.1425, 0.1142, 0.1282, 0.0871, 0.1076, 0.0813], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0312, 0.0295, 0.0326, 0.0351, 0.0269, 0.0319, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0004, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-03-31 22:52:46,145 INFO [train.py:903] (0/4) Epoch 2, batch 4750, loss[loss=0.4276, simple_loss=0.4415, pruned_loss=0.2069, over 19539.00 frames. ], tot_loss[loss=0.3498, simple_loss=0.3905, pruned_loss=0.1545, over 3836630.12 frames. ], batch size: 54, lr: 3.09e-02, grad_scale: 8.0 +2023-03-31 22:52:58,627 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.02 vs. limit=2.0 +2023-03-31 22:53:32,495 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.603e+02 7.433e+02 1.012e+03 1.332e+03 3.283e+03, threshold=2.025e+03, percent-clipped=2.0 +2023-03-31 22:53:35,185 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11618.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 22:53:47,420 INFO [train.py:903] (0/4) Epoch 2, batch 4800, loss[loss=0.2923, simple_loss=0.3399, pruned_loss=0.1224, over 19773.00 frames. ], tot_loss[loss=0.3477, simple_loss=0.3885, pruned_loss=0.1535, over 3830714.43 frames. ], batch size: 47, lr: 3.08e-02, grad_scale: 8.0 +2023-03-31 22:54:04,594 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11643.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 22:54:15,067 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=11651.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:54:19,421 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11653.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:54:49,929 INFO [train.py:903] (0/4) Epoch 2, batch 4850, loss[loss=0.2865, simple_loss=0.3412, pruned_loss=0.1159, over 19482.00 frames. ], tot_loss[loss=0.3475, simple_loss=0.3882, pruned_loss=0.1534, over 3824511.49 frames. ], batch size: 49, lr: 3.08e-02, grad_scale: 8.0 +2023-03-31 22:55:14,985 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-03-31 22:55:34,576 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-03-31 22:55:36,942 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.407e+02 7.585e+02 1.008e+03 1.288e+03 2.592e+03, threshold=2.016e+03, percent-clipped=4.0 +2023-03-31 22:55:38,521 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11717.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:55:40,632 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-03-31 22:55:41,791 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-03-31 22:55:51,154 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-03-31 22:55:53,218 INFO [train.py:903] (0/4) Epoch 2, batch 4900, loss[loss=0.3169, simple_loss=0.3593, pruned_loss=0.1373, over 19754.00 frames. ], tot_loss[loss=0.3472, simple_loss=0.3873, pruned_loss=0.1535, over 3820881.96 frames. ], batch size: 51, lr: 3.07e-02, grad_scale: 8.0 +2023-03-31 22:56:13,254 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-03-31 22:56:55,831 INFO [train.py:903] (0/4) Epoch 2, batch 4950, loss[loss=0.3505, simple_loss=0.3858, pruned_loss=0.1576, over 19607.00 frames. ], tot_loss[loss=0.3475, simple_loss=0.3875, pruned_loss=0.1537, over 3827756.53 frames. ], batch size: 50, lr: 3.06e-02, grad_scale: 8.0 +2023-03-31 22:57:12,053 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-03-31 22:57:37,547 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-03-31 22:57:41,841 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.086e+02 9.257e+02 1.136e+03 1.412e+03 3.441e+03, threshold=2.272e+03, percent-clipped=4.0 +2023-03-31 22:57:57,804 INFO [train.py:903] (0/4) Epoch 2, batch 5000, loss[loss=0.3891, simple_loss=0.4073, pruned_loss=0.1854, over 19668.00 frames. ], tot_loss[loss=0.3476, simple_loss=0.3875, pruned_loss=0.1538, over 3824023.44 frames. ], batch size: 55, lr: 3.06e-02, grad_scale: 8.0 +2023-03-31 22:58:04,608 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11834.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:58:07,424 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-03-31 22:58:16,638 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-03-31 22:58:42,460 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11864.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:59:00,548 INFO [train.py:903] (0/4) Epoch 2, batch 5050, loss[loss=0.2931, simple_loss=0.341, pruned_loss=0.1226, over 19728.00 frames. ], tot_loss[loss=0.3481, simple_loss=0.3877, pruned_loss=0.1542, over 3822364.59 frames. ], batch size: 46, lr: 3.05e-02, grad_scale: 8.0 +2023-03-31 22:59:19,011 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11894.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:59:32,446 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.68 vs. limit=5.0 +2023-03-31 22:59:36,628 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=11907.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 22:59:38,652 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-03-31 22:59:48,010 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.479e+02 8.836e+02 1.119e+03 1.460e+03 3.605e+03, threshold=2.237e+03, percent-clipped=7.0 +2023-03-31 23:00:03,339 INFO [train.py:903] (0/4) Epoch 2, batch 5100, loss[loss=0.3023, simple_loss=0.3522, pruned_loss=0.1262, over 19481.00 frames. ], tot_loss[loss=0.3473, simple_loss=0.3874, pruned_loss=0.1536, over 3823468.92 frames. ], batch size: 49, lr: 3.05e-02, grad_scale: 8.0 +2023-03-31 23:00:08,016 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=11932.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:00:09,170 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7408, 1.0719, 1.3720, 1.0454, 2.7369, 3.2887, 3.2567, 3.6471], + device='cuda:0'), covar=tensor([0.1374, 0.2851, 0.2893, 0.2256, 0.0445, 0.0141, 0.0201, 0.0101], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0276, 0.0317, 0.0278, 0.0193, 0.0104, 0.0190, 0.0108], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-03-31 23:00:16,215 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11938.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:00:19,216 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-03-31 23:00:22,846 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-03-31 23:00:26,271 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-03-31 23:00:50,293 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=11966.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:01:01,941 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9871, 1.9820, 1.4122, 1.5185, 1.3276, 1.4451, 0.1523, 0.9526], + device='cuda:0'), covar=tensor([0.0358, 0.0299, 0.0201, 0.0250, 0.0634, 0.0421, 0.0639, 0.0499], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0231, 0.0229, 0.0244, 0.0308, 0.0264, 0.0257, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 23:01:07,384 INFO [train.py:903] (0/4) Epoch 2, batch 5150, loss[loss=0.3209, simple_loss=0.3739, pruned_loss=0.1339, over 19779.00 frames. ], tot_loss[loss=0.3479, simple_loss=0.3878, pruned_loss=0.154, over 3817631.63 frames. ], batch size: 54, lr: 3.04e-02, grad_scale: 8.0 +2023-03-31 23:01:20,955 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-03-31 23:01:28,128 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-03-31 23:01:29,876 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=11997.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:01:33,200 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-12000.pt +2023-03-31 23:01:36,604 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12002.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:01:44,865 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12009.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:01:54,416 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.835e+02 8.254e+02 9.989e+02 1.287e+03 2.673e+03, threshold=1.998e+03, percent-clipped=4.0 +2023-03-31 23:01:57,693 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-03-31 23:02:10,373 INFO [train.py:903] (0/4) Epoch 2, batch 5200, loss[loss=0.3688, simple_loss=0.4009, pruned_loss=0.1684, over 17467.00 frames. ], tot_loss[loss=0.3477, simple_loss=0.3875, pruned_loss=0.1539, over 3816083.55 frames. ], batch size: 101, lr: 3.04e-02, grad_scale: 8.0 +2023-03-31 23:02:25,397 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-03-31 23:02:52,111 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12061.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:03:01,733 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3058, 2.1804, 1.9501, 3.5828, 2.5437, 4.4702, 3.4862, 1.9807], + device='cuda:0'), covar=tensor([0.1133, 0.0828, 0.0518, 0.0471, 0.0938, 0.0124, 0.0759, 0.0756], + device='cuda:0'), in_proj_covar=tensor([0.0369, 0.0336, 0.0355, 0.0466, 0.0421, 0.0261, 0.0455, 0.0360], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:03:11,163 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-03-31 23:03:13,773 INFO [train.py:903] (0/4) Epoch 2, batch 5250, loss[loss=0.3042, simple_loss=0.3498, pruned_loss=0.1294, over 19297.00 frames. ], tot_loss[loss=0.3478, simple_loss=0.3877, pruned_loss=0.1539, over 3817098.03 frames. ], batch size: 44, lr: 3.03e-02, grad_scale: 8.0 +2023-03-31 23:03:27,392 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9821, 1.9229, 1.4737, 1.2316, 1.4277, 1.5006, 0.1494, 0.8157], + device='cuda:0'), covar=tensor([0.0286, 0.0289, 0.0200, 0.0293, 0.0536, 0.0380, 0.0635, 0.0530], + device='cuda:0'), in_proj_covar=tensor([0.0215, 0.0231, 0.0223, 0.0246, 0.0309, 0.0262, 0.0261, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 23:03:32,701 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12094.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:03:55,673 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12112.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:03:59,697 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.115e+02 8.666e+02 1.057e+03 1.421e+03 4.195e+03, threshold=2.115e+03, percent-clipped=5.0 +2023-03-31 23:04:14,783 INFO [train.py:903] (0/4) Epoch 2, batch 5300, loss[loss=0.3272, simple_loss=0.3837, pruned_loss=0.1354, over 19778.00 frames. ], tot_loss[loss=0.3503, simple_loss=0.3898, pruned_loss=0.1554, over 3810112.88 frames. ], batch size: 56, lr: 3.03e-02, grad_scale: 8.0 +2023-03-31 23:04:16,280 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0435, 1.9373, 2.1953, 2.6624, 4.6642, 1.5329, 2.2598, 4.4562], + device='cuda:0'), covar=tensor([0.0168, 0.2035, 0.1958, 0.1204, 0.0237, 0.1899, 0.1013, 0.0319], + device='cuda:0'), in_proj_covar=tensor([0.0227, 0.0297, 0.0286, 0.0272, 0.0257, 0.0309, 0.0260, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:04:24,809 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.80 vs. limit=5.0 +2023-03-31 23:04:29,181 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5486, 3.7713, 4.0197, 3.9771, 1.4085, 3.4970, 3.3707, 3.5422], + device='cuda:0'), covar=tensor([0.0438, 0.0477, 0.0473, 0.0294, 0.3068, 0.0250, 0.0388, 0.0971], + device='cuda:0'), in_proj_covar=tensor([0.0307, 0.0285, 0.0399, 0.0294, 0.0428, 0.0196, 0.0273, 0.0396], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 23:04:35,573 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-03-31 23:05:13,665 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12176.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:05:15,907 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12178.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:05:16,930 INFO [train.py:903] (0/4) Epoch 2, batch 5350, loss[loss=0.3118, simple_loss=0.3637, pruned_loss=0.13, over 19851.00 frames. ], tot_loss[loss=0.3498, simple_loss=0.3892, pruned_loss=0.1552, over 3821393.49 frames. ], batch size: 52, lr: 3.02e-02, grad_scale: 8.0 +2023-03-31 23:05:53,517 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-03-31 23:05:53,639 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12208.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:06:03,423 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.241e+02 9.122e+02 1.169e+03 1.506e+03 3.477e+03, threshold=2.338e+03, percent-clipped=13.0 +2023-03-31 23:06:20,776 INFO [train.py:903] (0/4) Epoch 2, batch 5400, loss[loss=0.3335, simple_loss=0.3701, pruned_loss=0.1484, over 19730.00 frames. ], tot_loss[loss=0.3501, simple_loss=0.3892, pruned_loss=0.1555, over 3824388.23 frames. ], batch size: 51, lr: 3.02e-02, grad_scale: 8.0 +2023-03-31 23:07:06,371 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12265.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:07:23,450 INFO [train.py:903] (0/4) Epoch 2, batch 5450, loss[loss=0.3694, simple_loss=0.4073, pruned_loss=0.1658, over 18644.00 frames. ], tot_loss[loss=0.3483, simple_loss=0.3886, pruned_loss=0.154, over 3814794.39 frames. ], batch size: 74, lr: 3.01e-02, grad_scale: 8.0 +2023-03-31 23:07:27,018 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12282.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:07:36,430 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12290.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:07:39,788 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12293.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:08:01,525 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12310.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:08:02,767 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0865, 2.8653, 2.0283, 2.6420, 1.4544, 2.6879, 2.5294, 2.6338], + device='cuda:0'), covar=tensor([0.0846, 0.0954, 0.1628, 0.0779, 0.2391, 0.0979, 0.0730, 0.0945], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0284, 0.0311, 0.0262, 0.0328, 0.0278, 0.0222, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 23:08:08,443 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.315e+02 7.539e+02 9.464e+02 1.137e+03 1.898e+03, threshold=1.893e+03, percent-clipped=0.0 +2023-03-31 23:08:17,817 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12323.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:08:23,992 INFO [train.py:903] (0/4) Epoch 2, batch 5500, loss[loss=0.3768, simple_loss=0.406, pruned_loss=0.1738, over 19751.00 frames. ], tot_loss[loss=0.349, simple_loss=0.3891, pruned_loss=0.1545, over 3810369.28 frames. ], batch size: 54, lr: 3.01e-02, grad_scale: 8.0 +2023-03-31 23:08:47,584 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12346.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:08:49,878 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-03-31 23:09:14,231 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12368.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:09:28,513 INFO [train.py:903] (0/4) Epoch 2, batch 5550, loss[loss=0.3137, simple_loss=0.3663, pruned_loss=0.1305, over 19660.00 frames. ], tot_loss[loss=0.3481, simple_loss=0.3886, pruned_loss=0.1538, over 3799075.78 frames. ], batch size: 53, lr: 3.00e-02, grad_scale: 8.0 +2023-03-31 23:09:36,187 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-03-31 23:09:46,100 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12393.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:09:50,647 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12397.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:10:14,306 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.827e+02 8.577e+02 1.018e+03 1.198e+03 2.956e+03, threshold=2.037e+03, percent-clipped=3.0 +2023-03-31 23:10:26,765 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-03-31 23:10:27,083 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12425.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:10:29,728 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.18 vs. limit=5.0 +2023-03-31 23:10:31,326 INFO [train.py:903] (0/4) Epoch 2, batch 5600, loss[loss=0.2953, simple_loss=0.3472, pruned_loss=0.1217, over 19750.00 frames. ], tot_loss[loss=0.3468, simple_loss=0.3879, pruned_loss=0.1529, over 3801541.26 frames. ], batch size: 51, lr: 3.00e-02, grad_scale: 8.0 +2023-03-31 23:10:35,282 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12432.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:10:41,856 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12438.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:11:06,549 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12457.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:11:11,966 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12461.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:11:16,317 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12464.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:11:33,647 INFO [train.py:903] (0/4) Epoch 2, batch 5650, loss[loss=0.3338, simple_loss=0.3794, pruned_loss=0.1441, over 18181.00 frames. ], tot_loss[loss=0.3457, simple_loss=0.3866, pruned_loss=0.1524, over 3797451.41 frames. ], batch size: 83, lr: 2.99e-02, grad_scale: 8.0 +2023-03-31 23:12:19,592 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.393e+02 8.152e+02 1.040e+03 1.340e+03 2.595e+03, threshold=2.080e+03, percent-clipped=4.0 +2023-03-31 23:12:21,928 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-03-31 23:12:35,556 INFO [train.py:903] (0/4) Epoch 2, batch 5700, loss[loss=0.3426, simple_loss=0.3892, pruned_loss=0.148, over 19526.00 frames. ], tot_loss[loss=0.3492, simple_loss=0.3887, pruned_loss=0.1548, over 3808766.02 frames. ], batch size: 56, lr: 2.98e-02, grad_scale: 8.0 +2023-03-31 23:13:02,373 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12549.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:13:06,807 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12553.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:13:32,333 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12574.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:13:37,666 INFO [train.py:903] (0/4) Epoch 2, batch 5750, loss[loss=0.3577, simple_loss=0.3915, pruned_loss=0.162, over 19585.00 frames. ], tot_loss[loss=0.3471, simple_loss=0.3873, pruned_loss=0.1534, over 3812538.41 frames. ], batch size: 52, lr: 2.98e-02, grad_scale: 8.0 +2023-03-31 23:13:38,090 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12579.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:13:42,210 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-03-31 23:13:50,267 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-03-31 23:13:56,018 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9161, 1.0061, 1.3256, 1.5642, 2.6184, 1.4157, 1.8528, 2.5918], + device='cuda:0'), covar=tensor([0.0383, 0.2450, 0.2307, 0.1408, 0.0440, 0.1815, 0.0942, 0.0526], + device='cuda:0'), in_proj_covar=tensor([0.0227, 0.0302, 0.0285, 0.0273, 0.0260, 0.0318, 0.0258, 0.0271], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-03-31 23:13:56,869 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-03-31 23:14:10,383 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12604.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:14:23,848 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.880e+02 8.083e+02 9.516e+02 1.322e+03 3.330e+03, threshold=1.903e+03, percent-clipped=5.0 +2023-03-31 23:14:40,962 INFO [train.py:903] (0/4) Epoch 2, batch 5800, loss[loss=0.4148, simple_loss=0.4435, pruned_loss=0.193, over 19635.00 frames. ], tot_loss[loss=0.3458, simple_loss=0.3863, pruned_loss=0.1526, over 3837665.42 frames. ], batch size: 57, lr: 2.97e-02, grad_scale: 8.0 +2023-03-31 23:14:58,391 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6510, 1.4721, 1.2955, 1.8303, 1.6442, 1.6092, 1.2923, 1.7533], + device='cuda:0'), covar=tensor([0.0829, 0.1667, 0.1318, 0.0925, 0.1181, 0.0580, 0.1127, 0.0613], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0376, 0.0281, 0.0252, 0.0319, 0.0260, 0.0277, 0.0229], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:15:10,924 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12653.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:15:42,712 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12678.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:15:43,511 INFO [train.py:903] (0/4) Epoch 2, batch 5850, loss[loss=0.3034, simple_loss=0.344, pruned_loss=0.1315, over 19732.00 frames. ], tot_loss[loss=0.3451, simple_loss=0.3861, pruned_loss=0.1521, over 3829495.26 frames. ], batch size: 45, lr: 2.97e-02, grad_scale: 8.0 +2023-03-31 23:15:46,286 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12681.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:16:18,388 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12706.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:16:30,350 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.479e+02 7.980e+02 9.827e+02 1.217e+03 2.781e+03, threshold=1.965e+03, percent-clipped=6.0 +2023-03-31 23:16:31,907 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12717.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:16:46,106 INFO [train.py:903] (0/4) Epoch 2, batch 5900, loss[loss=0.3675, simple_loss=0.4078, pruned_loss=0.1636, over 19700.00 frames. ], tot_loss[loss=0.3438, simple_loss=0.3847, pruned_loss=0.1515, over 3809982.40 frames. ], batch size: 59, lr: 2.96e-02, grad_scale: 8.0 +2023-03-31 23:16:49,438 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-03-31 23:17:03,709 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12742.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:17:11,474 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-03-31 23:17:49,032 INFO [train.py:903] (0/4) Epoch 2, batch 5950, loss[loss=0.3404, simple_loss=0.3923, pruned_loss=0.1442, over 19283.00 frames. ], tot_loss[loss=0.346, simple_loss=0.3864, pruned_loss=0.1528, over 3812110.12 frames. ], batch size: 66, lr: 2.96e-02, grad_scale: 8.0 +2023-03-31 23:17:51,524 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=12781.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:18:25,354 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=12808.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:18:26,806 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=12809.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:18:34,471 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.856e+02 9.090e+02 1.139e+03 1.452e+03 3.383e+03, threshold=2.279e+03, percent-clipped=8.0 +2023-03-31 23:18:51,783 INFO [train.py:903] (0/4) Epoch 2, batch 6000, loss[loss=0.337, simple_loss=0.388, pruned_loss=0.143, over 19753.00 frames. ], tot_loss[loss=0.3457, simple_loss=0.386, pruned_loss=0.1527, over 3818779.62 frames. ], batch size: 63, lr: 2.95e-02, grad_scale: 8.0 +2023-03-31 23:18:51,784 INFO [train.py:928] (0/4) Computing validation loss +2023-03-31 23:19:06,008 INFO [train.py:937] (0/4) Epoch 2, validation: loss=0.246, simple_loss=0.337, pruned_loss=0.07745, over 944034.00 frames. +2023-03-31 23:19:06,009 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18153MB +2023-03-31 23:19:13,259 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=12834.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:20:08,251 INFO [train.py:903] (0/4) Epoch 2, batch 6050, loss[loss=0.4387, simple_loss=0.4483, pruned_loss=0.2145, over 13499.00 frames. ], tot_loss[loss=0.3465, simple_loss=0.3865, pruned_loss=0.1533, over 3814646.94 frames. ], batch size: 136, lr: 2.95e-02, grad_scale: 4.0 +2023-03-31 23:20:56,514 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.060e+02 7.840e+02 9.937e+02 1.323e+03 8.220e+03, threshold=1.987e+03, percent-clipped=9.0 +2023-03-31 23:21:03,468 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=12923.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:21:03,953 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-03-31 23:21:10,259 INFO [train.py:903] (0/4) Epoch 2, batch 6100, loss[loss=0.3465, simple_loss=0.3886, pruned_loss=0.1522, over 17456.00 frames. ], tot_loss[loss=0.3454, simple_loss=0.3863, pruned_loss=0.1523, over 3806695.36 frames. ], batch size: 101, lr: 2.94e-02, grad_scale: 4.0 +2023-03-31 23:22:05,400 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-03-31 23:22:11,863 INFO [train.py:903] (0/4) Epoch 2, batch 6150, loss[loss=0.3229, simple_loss=0.3623, pruned_loss=0.1417, over 19595.00 frames. ], tot_loss[loss=0.3468, simple_loss=0.3875, pruned_loss=0.153, over 3818908.06 frames. ], batch size: 52, lr: 2.94e-02, grad_scale: 4.0 +2023-03-31 23:22:13,426 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3839, 1.2462, 2.3647, 1.8483, 3.3814, 3.4641, 3.7729, 1.6472], + device='cuda:0'), covar=tensor([0.1271, 0.2001, 0.1096, 0.1072, 0.0738, 0.0695, 0.0924, 0.1893], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0397, 0.0366, 0.0364, 0.0443, 0.0353, 0.0512, 0.0385], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:22:43,816 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-03-31 23:23:00,059 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2575, 2.2971, 1.4542, 1.5088, 1.4150, 1.5503, 0.2640, 0.9276], + device='cuda:0'), covar=tensor([0.0251, 0.0210, 0.0192, 0.0249, 0.0630, 0.0344, 0.0569, 0.0490], + device='cuda:0'), in_proj_covar=tensor([0.0217, 0.0227, 0.0225, 0.0245, 0.0314, 0.0259, 0.0251, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-03-31 23:23:00,761 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.553e+02 7.749e+02 1.029e+03 1.287e+03 3.235e+03, threshold=2.059e+03, percent-clipped=7.0 +2023-03-31 23:23:04,771 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3352, 1.2759, 1.9064, 1.5927, 2.6251, 2.6863, 2.7815, 1.1765], + device='cuda:0'), covar=tensor([0.1376, 0.2063, 0.1219, 0.1232, 0.0922, 0.0826, 0.1185, 0.2074], + device='cuda:0'), in_proj_covar=tensor([0.0365, 0.0397, 0.0366, 0.0365, 0.0439, 0.0350, 0.0514, 0.0386], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:23:13,346 INFO [train.py:903] (0/4) Epoch 2, batch 6200, loss[loss=0.3272, simple_loss=0.3819, pruned_loss=0.1362, over 19663.00 frames. ], tot_loss[loss=0.3475, simple_loss=0.3877, pruned_loss=0.1537, over 3812515.69 frames. ], batch size: 55, lr: 2.93e-02, grad_scale: 4.0 +2023-03-31 23:24:15,386 INFO [train.py:903] (0/4) Epoch 2, batch 6250, loss[loss=0.3118, simple_loss=0.3694, pruned_loss=0.1271, over 19689.00 frames. ], tot_loss[loss=0.3496, simple_loss=0.3896, pruned_loss=0.1547, over 3802567.88 frames. ], batch size: 59, lr: 2.93e-02, grad_scale: 4.0 +2023-03-31 23:24:47,223 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-03-31 23:25:04,356 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.358e+02 8.552e+02 1.023e+03 1.333e+03 3.705e+03, threshold=2.046e+03, percent-clipped=2.0 +2023-03-31 23:25:04,789 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9896, 2.0673, 1.6366, 2.7745, 2.0912, 3.0287, 2.3318, 1.8145], + device='cuda:0'), covar=tensor([0.0778, 0.0623, 0.0461, 0.0393, 0.0689, 0.0171, 0.0715, 0.0590], + device='cuda:0'), in_proj_covar=tensor([0.0388, 0.0367, 0.0380, 0.0492, 0.0445, 0.0277, 0.0474, 0.0375], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:25:13,024 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=13125.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:25:17,708 INFO [train.py:903] (0/4) Epoch 2, batch 6300, loss[loss=0.3719, simple_loss=0.4072, pruned_loss=0.1683, over 19765.00 frames. ], tot_loss[loss=0.3491, simple_loss=0.3891, pruned_loss=0.1546, over 3823542.79 frames. ], batch size: 56, lr: 2.92e-02, grad_scale: 4.0 +2023-03-31 23:25:59,096 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6766, 2.3898, 2.1995, 3.6809, 2.4750, 4.4123, 3.5125, 2.3151], + device='cuda:0'), covar=tensor([0.0838, 0.0684, 0.0401, 0.0430, 0.0863, 0.0116, 0.0554, 0.0560], + device='cuda:0'), in_proj_covar=tensor([0.0381, 0.0365, 0.0373, 0.0484, 0.0441, 0.0276, 0.0467, 0.0371], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:26:19,874 INFO [train.py:903] (0/4) Epoch 2, batch 6350, loss[loss=0.2986, simple_loss=0.3608, pruned_loss=0.1182, over 19749.00 frames. ], tot_loss[loss=0.3491, simple_loss=0.3893, pruned_loss=0.1544, over 3826012.09 frames. ], batch size: 54, lr: 2.92e-02, grad_scale: 4.0 +2023-03-31 23:26:20,284 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=13179.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:26:49,643 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=13204.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:26:51,304 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.90 vs. limit=5.0 +2023-03-31 23:27:06,688 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.673e+02 8.539e+02 1.071e+03 1.407e+03 4.202e+03, threshold=2.141e+03, percent-clipped=6.0 +2023-03-31 23:27:08,226 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5649, 1.4380, 1.3788, 1.8058, 1.5320, 1.3950, 1.3005, 1.6887], + device='cuda:0'), covar=tensor([0.0961, 0.1809, 0.1486, 0.0960, 0.1367, 0.0908, 0.1443, 0.0790], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0382, 0.0291, 0.0261, 0.0317, 0.0264, 0.0282, 0.0234], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:27:14,123 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5294, 1.5619, 2.0451, 2.6107, 2.0767, 2.1026, 1.7239, 2.5192], + device='cuda:0'), covar=tensor([0.0774, 0.2323, 0.1326, 0.0919, 0.1485, 0.0542, 0.1355, 0.0604], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0382, 0.0291, 0.0261, 0.0319, 0.0264, 0.0283, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:27:19,358 INFO [train.py:903] (0/4) Epoch 2, batch 6400, loss[loss=0.3798, simple_loss=0.4091, pruned_loss=0.1752, over 19566.00 frames. ], tot_loss[loss=0.3476, simple_loss=0.3876, pruned_loss=0.1538, over 3835053.16 frames. ], batch size: 52, lr: 2.92e-02, grad_scale: 8.0 +2023-03-31 23:27:33,772 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=13240.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:27:52,433 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-03-31 23:28:12,858 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.55 vs. limit=2.0 +2023-03-31 23:28:22,168 INFO [train.py:903] (0/4) Epoch 2, batch 6450, loss[loss=0.3555, simple_loss=0.3959, pruned_loss=0.1576, over 19700.00 frames. ], tot_loss[loss=0.3491, simple_loss=0.3888, pruned_loss=0.1547, over 3835366.51 frames. ], batch size: 59, lr: 2.91e-02, grad_scale: 8.0 +2023-03-31 23:29:07,670 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0269, 3.9286, 2.3439, 2.9968, 3.3232, 1.5819, 1.3372, 1.8004], + device='cuda:0'), covar=tensor([0.1191, 0.0243, 0.0887, 0.0496, 0.0414, 0.1212, 0.1074, 0.0794], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0216, 0.0307, 0.0256, 0.0208, 0.0310, 0.0274, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-03-31 23:29:09,603 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-03-31 23:29:10,633 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.859e+02 7.874e+02 9.907e+02 1.203e+03 2.411e+03, threshold=1.981e+03, percent-clipped=4.0 +2023-03-31 23:29:24,102 INFO [train.py:903] (0/4) Epoch 2, batch 6500, loss[loss=0.2663, simple_loss=0.3314, pruned_loss=0.1006, over 19574.00 frames. ], tot_loss[loss=0.3482, simple_loss=0.3884, pruned_loss=0.154, over 3826642.73 frames. ], batch size: 52, lr: 2.91e-02, grad_scale: 8.0 +2023-03-31 23:29:29,952 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8401, 4.7326, 5.6667, 5.6112, 1.9779, 5.1178, 4.6325, 5.0743], + device='cuda:0'), covar=tensor([0.0432, 0.0416, 0.0366, 0.0211, 0.3006, 0.0153, 0.0307, 0.0762], + device='cuda:0'), in_proj_covar=tensor([0.0321, 0.0302, 0.0406, 0.0307, 0.0445, 0.0204, 0.0274, 0.0408], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 23:29:30,898 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-03-31 23:30:26,693 INFO [train.py:903] (0/4) Epoch 2, batch 6550, loss[loss=0.348, simple_loss=0.3921, pruned_loss=0.1519, over 18018.00 frames. ], tot_loss[loss=0.3454, simple_loss=0.3867, pruned_loss=0.1521, over 3835955.45 frames. ], batch size: 83, lr: 2.90e-02, grad_scale: 8.0 +2023-03-31 23:31:14,493 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.467e+02 7.924e+02 9.507e+02 1.218e+03 2.525e+03, threshold=1.901e+03, percent-clipped=3.0 +2023-03-31 23:31:27,085 INFO [train.py:903] (0/4) Epoch 2, batch 6600, loss[loss=0.371, simple_loss=0.4075, pruned_loss=0.1673, over 18152.00 frames. ], tot_loss[loss=0.3457, simple_loss=0.3866, pruned_loss=0.1524, over 3824927.12 frames. ], batch size: 83, lr: 2.90e-02, grad_scale: 8.0 +2023-03-31 23:32:29,087 INFO [train.py:903] (0/4) Epoch 2, batch 6650, loss[loss=0.4237, simple_loss=0.4398, pruned_loss=0.2038, over 19673.00 frames. ], tot_loss[loss=0.3467, simple_loss=0.3871, pruned_loss=0.1532, over 3829570.31 frames. ], batch size: 58, lr: 2.89e-02, grad_scale: 8.0 +2023-03-31 23:32:50,439 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=13496.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:33:08,242 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-03-31 23:33:17,438 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.558e+02 9.096e+02 1.181e+03 1.471e+03 3.411e+03, threshold=2.361e+03, percent-clipped=10.0 +2023-03-31 23:33:21,111 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=13521.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:33:29,657 INFO [train.py:903] (0/4) Epoch 2, batch 6700, loss[loss=0.3724, simple_loss=0.4135, pruned_loss=0.1657, over 19528.00 frames. ], tot_loss[loss=0.3469, simple_loss=0.3877, pruned_loss=0.1531, over 3836526.49 frames. ], batch size: 56, lr: 2.89e-02, grad_scale: 8.0 +2023-03-31 23:34:27,765 INFO [train.py:903] (0/4) Epoch 2, batch 6750, loss[loss=0.3826, simple_loss=0.4202, pruned_loss=0.1724, over 19254.00 frames. ], tot_loss[loss=0.3468, simple_loss=0.3876, pruned_loss=0.153, over 3835548.81 frames. ], batch size: 66, lr: 2.88e-02, grad_scale: 8.0 +2023-03-31 23:34:38,864 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0 +2023-03-31 23:34:51,595 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8614, 4.8414, 5.7286, 5.5934, 2.0302, 5.3092, 4.6756, 5.1404], + device='cuda:0'), covar=tensor([0.0366, 0.0424, 0.0324, 0.0201, 0.2718, 0.0136, 0.0254, 0.0708], + device='cuda:0'), in_proj_covar=tensor([0.0325, 0.0306, 0.0410, 0.0307, 0.0444, 0.0210, 0.0272, 0.0408], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 23:35:12,679 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.052e+02 8.174e+02 9.996e+02 1.293e+03 2.664e+03, threshold=1.999e+03, percent-clipped=1.0 +2023-03-31 23:35:25,228 INFO [train.py:903] (0/4) Epoch 2, batch 6800, loss[loss=0.3522, simple_loss=0.3921, pruned_loss=0.1562, over 19736.00 frames. ], tot_loss[loss=0.3482, simple_loss=0.3882, pruned_loss=0.1541, over 3822076.75 frames. ], batch size: 63, lr: 2.88e-02, grad_scale: 8.0 +2023-03-31 23:35:54,699 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-2.pt +2023-03-31 23:36:09,998 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-03-31 23:36:11,043 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-03-31 23:36:13,727 INFO [train.py:903] (0/4) Epoch 3, batch 0, loss[loss=0.3343, simple_loss=0.3632, pruned_loss=0.1527, over 19725.00 frames. ], tot_loss[loss=0.3343, simple_loss=0.3632, pruned_loss=0.1527, over 19725.00 frames. ], batch size: 46, lr: 2.73e-02, grad_scale: 8.0 +2023-03-31 23:36:13,727 INFO [train.py:928] (0/4) Computing validation loss +2023-03-31 23:36:24,488 INFO [train.py:937] (0/4) Epoch 3, validation: loss=0.241, simple_loss=0.3346, pruned_loss=0.07374, over 944034.00 frames. +2023-03-31 23:36:24,489 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18153MB +2023-03-31 23:36:37,425 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-03-31 23:36:45,585 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1523, 0.9135, 1.0139, 1.5048, 1.1118, 1.2184, 1.2620, 1.1744], + device='cuda:0'), covar=tensor([0.1106, 0.1800, 0.1578, 0.0859, 0.1274, 0.1199, 0.1259, 0.1120], + device='cuda:0'), in_proj_covar=tensor([0.0269, 0.0308, 0.0293, 0.0325, 0.0338, 0.0270, 0.0309, 0.0269], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-03-31 23:37:25,524 INFO [train.py:903] (0/4) Epoch 3, batch 50, loss[loss=0.3896, simple_loss=0.4165, pruned_loss=0.1813, over 17457.00 frames. ], tot_loss[loss=0.3434, simple_loss=0.3848, pruned_loss=0.151, over 865249.12 frames. ], batch size: 101, lr: 2.73e-02, grad_scale: 8.0 +2023-03-31 23:37:38,309 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.119e+02 7.787e+02 9.326e+02 1.115e+03 3.182e+03, threshold=1.865e+03, percent-clipped=5.0 +2023-03-31 23:37:58,962 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-03-31 23:38:23,958 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=13755.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:38:25,858 INFO [train.py:903] (0/4) Epoch 3, batch 100, loss[loss=0.4201, simple_loss=0.4386, pruned_loss=0.2008, over 19237.00 frames. ], tot_loss[loss=0.3415, simple_loss=0.3849, pruned_loss=0.1491, over 1536049.28 frames. ], batch size: 69, lr: 2.72e-02, grad_scale: 8.0 +2023-03-31 23:38:35,102 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-03-31 23:39:27,419 INFO [train.py:903] (0/4) Epoch 3, batch 150, loss[loss=0.3332, simple_loss=0.3853, pruned_loss=0.1406, over 19465.00 frames. ], tot_loss[loss=0.3407, simple_loss=0.3841, pruned_loss=0.1486, over 2055816.99 frames. ], batch size: 64, lr: 2.72e-02, grad_scale: 8.0 +2023-03-31 23:39:30,164 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4818, 0.9886, 1.2917, 1.3026, 2.2360, 1.1883, 1.7991, 2.1205], + device='cuda:0'), covar=tensor([0.0412, 0.2271, 0.1959, 0.1314, 0.0514, 0.1390, 0.0724, 0.0557], + device='cuda:0'), in_proj_covar=tensor([0.0231, 0.0303, 0.0287, 0.0275, 0.0273, 0.0318, 0.0265, 0.0271], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:39:40,067 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.782e+02 7.521e+02 1.009e+03 1.351e+03 3.530e+03, threshold=2.018e+03, percent-clipped=10.0 +2023-03-31 23:40:28,875 INFO [train.py:903] (0/4) Epoch 3, batch 200, loss[loss=0.4297, simple_loss=0.4499, pruned_loss=0.2048, over 19308.00 frames. ], tot_loss[loss=0.3418, simple_loss=0.3844, pruned_loss=0.1496, over 2441726.99 frames. ], batch size: 66, lr: 2.71e-02, grad_scale: 8.0 +2023-03-31 23:40:28,917 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-03-31 23:41:28,996 INFO [train.py:903] (0/4) Epoch 3, batch 250, loss[loss=0.2713, simple_loss=0.3269, pruned_loss=0.1079, over 19768.00 frames. ], tot_loss[loss=0.3416, simple_loss=0.3838, pruned_loss=0.1497, over 2754507.22 frames. ], batch size: 46, lr: 2.71e-02, grad_scale: 8.0 +2023-03-31 23:41:44,255 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.886e+02 8.729e+02 1.056e+03 1.304e+03 3.760e+03, threshold=2.113e+03, percent-clipped=6.0 +2023-03-31 23:41:49,454 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-03-31 23:42:33,072 INFO [train.py:903] (0/4) Epoch 3, batch 300, loss[loss=0.2925, simple_loss=0.3554, pruned_loss=0.1147, over 19512.00 frames. ], tot_loss[loss=0.3386, simple_loss=0.3808, pruned_loss=0.1481, over 2988693.02 frames. ], batch size: 54, lr: 2.71e-02, grad_scale: 8.0 +2023-03-31 23:43:25,964 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-14000.pt +2023-03-31 23:43:34,502 INFO [train.py:903] (0/4) Epoch 3, batch 350, loss[loss=0.369, simple_loss=0.3933, pruned_loss=0.1724, over 19587.00 frames. ], tot_loss[loss=0.3386, simple_loss=0.3814, pruned_loss=0.1479, over 3176080.10 frames. ], batch size: 52, lr: 2.70e-02, grad_scale: 8.0 +2023-03-31 23:43:40,055 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-03-31 23:43:46,963 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.582e+02 7.628e+02 9.853e+02 1.217e+03 3.369e+03, threshold=1.971e+03, percent-clipped=3.0 +2023-03-31 23:44:34,933 INFO [train.py:903] (0/4) Epoch 3, batch 400, loss[loss=0.4001, simple_loss=0.4169, pruned_loss=0.1916, over 13128.00 frames. ], tot_loss[loss=0.339, simple_loss=0.3813, pruned_loss=0.1483, over 3313546.68 frames. ], batch size: 138, lr: 2.70e-02, grad_scale: 8.0 +2023-03-31 23:45:05,470 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7859, 1.5383, 1.5337, 1.7143, 1.7599, 1.7488, 1.7138, 2.0766], + device='cuda:0'), covar=tensor([0.0846, 0.1714, 0.1293, 0.0916, 0.1170, 0.0534, 0.0883, 0.0559], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0381, 0.0288, 0.0260, 0.0319, 0.0272, 0.0273, 0.0231], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:45:17,145 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14090.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:45:27,292 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14099.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:45:27,630 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8702, 1.7761, 1.7142, 2.6007, 1.7508, 2.3097, 2.1655, 1.6575], + device='cuda:0'), covar=tensor([0.0819, 0.0664, 0.0438, 0.0337, 0.0717, 0.0257, 0.0766, 0.0686], + device='cuda:0'), in_proj_covar=tensor([0.0409, 0.0379, 0.0388, 0.0510, 0.0454, 0.0285, 0.0484, 0.0391], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-03-31 23:45:36,306 INFO [train.py:903] (0/4) Epoch 3, batch 450, loss[loss=0.3443, simple_loss=0.3927, pruned_loss=0.1479, over 19662.00 frames. ], tot_loss[loss=0.3378, simple_loss=0.3806, pruned_loss=0.1475, over 3427201.02 frames. ], batch size: 55, lr: 2.69e-02, grad_scale: 8.0 +2023-03-31 23:45:52,383 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.440e+02 8.213e+02 1.015e+03 1.206e+03 3.609e+03, threshold=2.029e+03, percent-clipped=6.0 +2023-03-31 23:45:55,949 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14121.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:46:10,162 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-03-31 23:46:11,132 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-03-31 23:46:14,944 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14138.0, num_to_drop=1, layers_to_drop={1} +2023-03-31 23:46:38,896 INFO [train.py:903] (0/4) Epoch 3, batch 500, loss[loss=0.2997, simple_loss=0.3525, pruned_loss=0.1235, over 19374.00 frames. ], tot_loss[loss=0.339, simple_loss=0.3812, pruned_loss=0.1484, over 3516381.84 frames. ], batch size: 47, lr: 2.69e-02, grad_scale: 8.0 +2023-03-31 23:47:38,213 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6804, 1.3312, 1.7899, 1.4928, 2.8076, 4.2540, 4.3199, 4.7014], + device='cuda:0'), covar=tensor([0.1289, 0.2558, 0.2481, 0.1871, 0.0472, 0.0109, 0.0138, 0.0074], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0277, 0.0324, 0.0274, 0.0192, 0.0103, 0.0195, 0.0109], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-03-31 23:47:38,988 INFO [train.py:903] (0/4) Epoch 3, batch 550, loss[loss=0.3121, simple_loss=0.3721, pruned_loss=0.1261, over 19707.00 frames. ], tot_loss[loss=0.3386, simple_loss=0.3813, pruned_loss=0.148, over 3576647.13 frames. ], batch size: 63, lr: 2.68e-02, grad_scale: 8.0 +2023-03-31 23:47:47,470 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14214.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:47:51,334 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.716e+02 8.063e+02 9.949e+02 1.307e+03 2.222e+03, threshold=1.990e+03, percent-clipped=3.0 +2023-03-31 23:47:55,490 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.20 vs. limit=5.0 +2023-03-31 23:48:38,813 INFO [train.py:903] (0/4) Epoch 3, batch 600, loss[loss=0.3774, simple_loss=0.4076, pruned_loss=0.1736, over 19776.00 frames. ], tot_loss[loss=0.3378, simple_loss=0.3807, pruned_loss=0.1475, over 3637875.06 frames. ], batch size: 56, lr: 2.68e-02, grad_scale: 8.0 +2023-03-31 23:49:16,706 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-03-31 23:49:39,186 INFO [train.py:903] (0/4) Epoch 3, batch 650, loss[loss=0.3193, simple_loss=0.366, pruned_loss=0.1363, over 19536.00 frames. ], tot_loss[loss=0.338, simple_loss=0.3812, pruned_loss=0.1474, over 3671102.82 frames. ], batch size: 54, lr: 2.68e-02, grad_scale: 8.0 +2023-03-31 23:49:54,624 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.103e+02 8.257e+02 1.098e+03 1.322e+03 3.191e+03, threshold=2.197e+03, percent-clipped=10.0 +2023-03-31 23:50:41,496 INFO [train.py:903] (0/4) Epoch 3, batch 700, loss[loss=0.3372, simple_loss=0.3739, pruned_loss=0.1502, over 19866.00 frames. ], tot_loss[loss=0.3377, simple_loss=0.3808, pruned_loss=0.1473, over 3717364.12 frames. ], batch size: 52, lr: 2.67e-02, grad_scale: 8.0 +2023-03-31 23:51:43,785 INFO [train.py:903] (0/4) Epoch 3, batch 750, loss[loss=0.3306, simple_loss=0.395, pruned_loss=0.1331, over 19605.00 frames. ], tot_loss[loss=0.3356, simple_loss=0.3799, pruned_loss=0.1456, over 3756478.95 frames. ], batch size: 57, lr: 2.67e-02, grad_scale: 8.0 +2023-03-31 23:51:56,444 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.299e+02 7.798e+02 9.551e+02 1.191e+03 2.807e+03, threshold=1.910e+03, percent-clipped=6.0 +2023-03-31 23:52:11,634 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14431.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:52:14,916 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14434.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:52:44,747 INFO [train.py:903] (0/4) Epoch 3, batch 800, loss[loss=0.3001, simple_loss=0.3416, pruned_loss=0.1294, over 19430.00 frames. ], tot_loss[loss=0.3341, simple_loss=0.3782, pruned_loss=0.145, over 3770625.82 frames. ], batch size: 48, lr: 2.66e-02, grad_scale: 8.0 +2023-03-31 23:52:53,861 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14465.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:52:54,899 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-03-31 23:52:59,795 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14470.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:53:15,236 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14482.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 23:53:31,015 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14495.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:53:42,648 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14505.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:53:44,740 INFO [train.py:903] (0/4) Epoch 3, batch 850, loss[loss=0.3574, simple_loss=0.4076, pruned_loss=0.1536, over 19518.00 frames. ], tot_loss[loss=0.3355, simple_loss=0.3792, pruned_loss=0.1459, over 3793879.79 frames. ], batch size: 64, lr: 2.66e-02, grad_scale: 8.0 +2023-03-31 23:53:58,373 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.292e+02 8.633e+02 1.105e+03 1.534e+03 3.114e+03, threshold=2.210e+03, percent-clipped=11.0 +2023-03-31 23:54:32,134 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-03-31 23:54:35,855 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14549.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:54:45,563 INFO [train.py:903] (0/4) Epoch 3, batch 900, loss[loss=0.2984, simple_loss=0.358, pruned_loss=0.1194, over 19616.00 frames. ], tot_loss[loss=0.3361, simple_loss=0.3799, pruned_loss=0.1461, over 3800957.25 frames. ], batch size: 57, lr: 2.66e-02, grad_scale: 8.0 +2023-03-31 23:55:15,225 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14580.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:55:23,135 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5691, 2.9020, 2.9751, 2.9973, 1.1656, 2.7074, 2.4640, 2.5951], + device='cuda:0'), covar=tensor([0.0721, 0.0690, 0.0601, 0.0479, 0.2838, 0.0360, 0.0506, 0.1134], + device='cuda:0'), in_proj_covar=tensor([0.0348, 0.0313, 0.0440, 0.0328, 0.0469, 0.0223, 0.0292, 0.0434], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-03-31 23:55:33,870 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14597.0, num_to_drop=1, layers_to_drop={0} +2023-03-31 23:55:39,424 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7624, 1.3813, 1.2933, 1.8296, 1.4096, 1.9020, 1.8268, 1.8667], + device='cuda:0'), covar=tensor([0.0748, 0.1170, 0.1355, 0.1016, 0.1169, 0.0863, 0.1031, 0.0692], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0299, 0.0287, 0.0324, 0.0322, 0.0267, 0.0301, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-03-31 23:55:46,733 INFO [train.py:903] (0/4) Epoch 3, batch 950, loss[loss=0.3533, simple_loss=0.4022, pruned_loss=0.1522, over 19624.00 frames. ], tot_loss[loss=0.336, simple_loss=0.3798, pruned_loss=0.1461, over 3813776.51 frames. ], batch size: 57, lr: 2.65e-02, grad_scale: 4.0 +2023-03-31 23:55:46,747 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-03-31 23:56:00,986 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.858e+02 7.381e+02 9.246e+02 1.263e+03 4.500e+03, threshold=1.849e+03, percent-clipped=5.0 +2023-03-31 23:56:21,391 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.60 vs. limit=2.0 +2023-03-31 23:56:46,895 INFO [train.py:903] (0/4) Epoch 3, batch 1000, loss[loss=0.3005, simple_loss=0.3535, pruned_loss=0.1237, over 19616.00 frames. ], tot_loss[loss=0.3371, simple_loss=0.381, pruned_loss=0.1465, over 3830613.78 frames. ], batch size: 50, lr: 2.65e-02, grad_scale: 4.0 +2023-03-31 23:57:38,693 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-03-31 23:57:47,678 INFO [train.py:903] (0/4) Epoch 3, batch 1050, loss[loss=0.3566, simple_loss=0.3884, pruned_loss=0.1624, over 19738.00 frames. ], tot_loss[loss=0.3359, simple_loss=0.3806, pruned_loss=0.1456, over 3837529.43 frames. ], batch size: 45, lr: 2.64e-02, grad_scale: 4.0 +2023-03-31 23:58:01,055 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.125e+02 7.305e+02 8.951e+02 1.118e+03 2.421e+03, threshold=1.790e+03, percent-clipped=2.0 +2023-03-31 23:58:15,792 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.50 vs. limit=5.0 +2023-03-31 23:58:17,620 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-03-31 23:58:48,366 INFO [train.py:903] (0/4) Epoch 3, batch 1100, loss[loss=0.371, simple_loss=0.4042, pruned_loss=0.1689, over 19583.00 frames. ], tot_loss[loss=0.3353, simple_loss=0.38, pruned_loss=0.1453, over 3835569.75 frames. ], batch size: 61, lr: 2.64e-02, grad_scale: 4.0 +2023-03-31 23:58:49,821 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=14758.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:59:10,908 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14775.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:59:46,865 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14805.0, num_to_drop=0, layers_to_drop=set() +2023-03-31 23:59:49,456 INFO [train.py:903] (0/4) Epoch 3, batch 1150, loss[loss=0.3683, simple_loss=0.3997, pruned_loss=0.1684, over 19715.00 frames. ], tot_loss[loss=0.3342, simple_loss=0.3788, pruned_loss=0.1448, over 3836916.93 frames. ], batch size: 63, lr: 2.64e-02, grad_scale: 4.0 +2023-04-01 00:00:03,884 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.960e+02 7.465e+02 1.021e+03 1.238e+03 3.548e+03, threshold=2.043e+03, percent-clipped=7.0 +2023-04-01 00:00:16,780 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14830.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:00:23,653 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14836.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:00:40,916 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=14849.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:00:45,731 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=14853.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 00:00:49,934 INFO [train.py:903] (0/4) Epoch 3, batch 1200, loss[loss=0.3489, simple_loss=0.4013, pruned_loss=0.1482, over 19558.00 frames. ], tot_loss[loss=0.3348, simple_loss=0.3795, pruned_loss=0.1451, over 3831137.21 frames. ], batch size: 61, lr: 2.63e-02, grad_scale: 8.0 +2023-04-01 00:00:55,819 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14861.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:01:15,101 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=14878.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 00:01:17,851 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 00:01:21,324 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2205, 1.2668, 2.0708, 1.6304, 3.0376, 3.1552, 3.7332, 1.5210], + device='cuda:0'), covar=tensor([0.1419, 0.2215, 0.1235, 0.1253, 0.0889, 0.0823, 0.0959, 0.2033], + device='cuda:0'), in_proj_covar=tensor([0.0404, 0.0435, 0.0402, 0.0391, 0.0473, 0.0386, 0.0553, 0.0400], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 00:01:31,250 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14890.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:01:51,485 INFO [train.py:903] (0/4) Epoch 3, batch 1250, loss[loss=0.3512, simple_loss=0.3799, pruned_loss=0.1613, over 19609.00 frames. ], tot_loss[loss=0.3334, simple_loss=0.3789, pruned_loss=0.1439, over 3831980.39 frames. ], batch size: 50, lr: 2.63e-02, grad_scale: 8.0 +2023-04-01 00:02:00,410 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2302, 1.2601, 1.9591, 1.4892, 3.1144, 2.7560, 3.1302, 0.9951], + device='cuda:0'), covar=tensor([0.1166, 0.1880, 0.1011, 0.1021, 0.0600, 0.0712, 0.0801, 0.1894], + device='cuda:0'), in_proj_covar=tensor([0.0400, 0.0434, 0.0402, 0.0387, 0.0471, 0.0383, 0.0554, 0.0401], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 00:02:05,938 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.325e+02 7.701e+02 1.002e+03 1.250e+03 2.941e+03, threshold=2.004e+03, percent-clipped=3.0 +2023-04-01 00:02:53,120 INFO [train.py:903] (0/4) Epoch 3, batch 1300, loss[loss=0.2944, simple_loss=0.3611, pruned_loss=0.1138, over 19807.00 frames. ], tot_loss[loss=0.3347, simple_loss=0.38, pruned_loss=0.1447, over 3815231.49 frames. ], batch size: 56, lr: 2.62e-02, grad_scale: 8.0 +2023-04-01 00:03:02,029 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=14964.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:03:54,923 INFO [train.py:903] (0/4) Epoch 3, batch 1350, loss[loss=0.3516, simple_loss=0.3911, pruned_loss=0.156, over 18753.00 frames. ], tot_loss[loss=0.3381, simple_loss=0.3825, pruned_loss=0.1469, over 3799285.29 frames. ], batch size: 74, lr: 2.62e-02, grad_scale: 8.0 +2023-04-01 00:03:57,804 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.94 vs. limit=5.0 +2023-04-01 00:04:10,657 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.789e+02 8.351e+02 9.883e+02 1.225e+03 3.360e+03, threshold=1.977e+03, percent-clipped=2.0 +2023-04-01 00:04:57,758 INFO [train.py:903] (0/4) Epoch 3, batch 1400, loss[loss=0.3787, simple_loss=0.4123, pruned_loss=0.1725, over 17116.00 frames. ], tot_loss[loss=0.3362, simple_loss=0.3808, pruned_loss=0.1458, over 3799081.01 frames. ], batch size: 101, lr: 2.62e-02, grad_scale: 8.0 +2023-04-01 00:05:08,569 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-01 00:05:53,731 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=15102.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:05:59,088 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 00:06:00,251 INFO [train.py:903] (0/4) Epoch 3, batch 1450, loss[loss=0.349, simple_loss=0.3893, pruned_loss=0.1544, over 19446.00 frames. ], tot_loss[loss=0.3365, simple_loss=0.381, pruned_loss=0.146, over 3799700.06 frames. ], batch size: 64, lr: 2.61e-02, grad_scale: 8.0 +2023-04-01 00:06:13,786 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.668e+02 7.923e+02 9.351e+02 1.150e+03 2.880e+03, threshold=1.870e+03, percent-clipped=3.0 +2023-04-01 00:06:49,290 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=15146.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:07:01,533 INFO [train.py:903] (0/4) Epoch 3, batch 1500, loss[loss=0.2737, simple_loss=0.3285, pruned_loss=0.1095, over 19491.00 frames. ], tot_loss[loss=0.3339, simple_loss=0.3788, pruned_loss=0.1445, over 3808481.93 frames. ], batch size: 49, lr: 2.61e-02, grad_scale: 8.0 +2023-04-01 00:07:20,407 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=15171.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:08:03,674 INFO [train.py:903] (0/4) Epoch 3, batch 1550, loss[loss=0.3174, simple_loss=0.3739, pruned_loss=0.1304, over 19130.00 frames. ], tot_loss[loss=0.3323, simple_loss=0.3772, pruned_loss=0.1437, over 3812875.38 frames. ], batch size: 69, lr: 2.60e-02, grad_scale: 8.0 +2023-04-01 00:08:06,003 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2913, 0.9669, 1.5305, 1.0508, 2.2659, 2.9050, 2.8843, 3.1457], + device='cuda:0'), covar=tensor([0.1785, 0.3973, 0.3604, 0.2478, 0.0613, 0.0235, 0.0309, 0.0164], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0282, 0.0329, 0.0278, 0.0191, 0.0103, 0.0198, 0.0117], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 00:08:12,797 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=15213.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:08:18,397 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15217.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:08:20,277 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.312e+02 7.825e+02 9.432e+02 1.149e+03 3.008e+03, threshold=1.886e+03, percent-clipped=3.0 +2023-04-01 00:08:21,877 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=15220.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:08:52,221 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=15245.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:09:08,812 INFO [train.py:903] (0/4) Epoch 3, batch 1600, loss[loss=0.2928, simple_loss=0.346, pruned_loss=0.1198, over 19385.00 frames. ], tot_loss[loss=0.3323, simple_loss=0.3774, pruned_loss=0.1436, over 3819798.78 frames. ], batch size: 47, lr: 2.60e-02, grad_scale: 8.0 +2023-04-01 00:09:32,830 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 00:09:38,988 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3301, 2.8661, 1.8794, 2.4365, 2.1443, 2.0399, 0.5930, 2.2975], + device='cuda:0'), covar=tensor([0.0285, 0.0207, 0.0271, 0.0246, 0.0456, 0.0438, 0.0554, 0.0440], + device='cuda:0'), in_proj_covar=tensor([0.0242, 0.0237, 0.0242, 0.0257, 0.0319, 0.0266, 0.0254, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 00:10:10,173 INFO [train.py:903] (0/4) Epoch 3, batch 1650, loss[loss=0.3535, simple_loss=0.3763, pruned_loss=0.1654, over 19779.00 frames. ], tot_loss[loss=0.3304, simple_loss=0.3761, pruned_loss=0.1424, over 3827579.81 frames. ], batch size: 48, lr: 2.60e-02, grad_scale: 8.0 +2023-04-01 00:10:24,984 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.775e+02 8.096e+02 9.310e+02 1.117e+03 2.889e+03, threshold=1.862e+03, percent-clipped=6.0 +2023-04-01 00:10:35,606 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5207, 1.0868, 1.0532, 1.7393, 1.2196, 1.7084, 1.7597, 1.4276], + device='cuda:0'), covar=tensor([0.0762, 0.1308, 0.1413, 0.0978, 0.1195, 0.0878, 0.1029, 0.0875], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0294, 0.0285, 0.0318, 0.0324, 0.0262, 0.0302, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 00:11:11,890 INFO [train.py:903] (0/4) Epoch 3, batch 1700, loss[loss=0.2716, simple_loss=0.3193, pruned_loss=0.112, over 19330.00 frames. ], tot_loss[loss=0.3304, simple_loss=0.3757, pruned_loss=0.1426, over 3827694.70 frames. ], batch size: 44, lr: 2.59e-02, grad_scale: 8.0 +2023-04-01 00:11:30,482 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7277, 1.7103, 1.6166, 2.2860, 1.5730, 1.9454, 1.9628, 1.5531], + device='cuda:0'), covar=tensor([0.0811, 0.0629, 0.0461, 0.0326, 0.0671, 0.0283, 0.0757, 0.0681], + device='cuda:0'), in_proj_covar=tensor([0.0432, 0.0405, 0.0412, 0.0548, 0.0482, 0.0316, 0.0505, 0.0409], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 00:11:50,333 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 00:12:13,187 INFO [train.py:903] (0/4) Epoch 3, batch 1750, loss[loss=0.2628, simple_loss=0.3169, pruned_loss=0.1043, over 19763.00 frames. ], tot_loss[loss=0.331, simple_loss=0.3758, pruned_loss=0.143, over 3817518.84 frames. ], batch size: 46, lr: 2.59e-02, grad_scale: 8.0 +2023-04-01 00:12:30,181 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.268e+02 8.359e+02 1.065e+03 1.276e+03 4.198e+03, threshold=2.129e+03, percent-clipped=6.0 +2023-04-01 00:13:17,302 INFO [train.py:903] (0/4) Epoch 3, batch 1800, loss[loss=0.3056, simple_loss=0.3663, pruned_loss=0.1224, over 18683.00 frames. ], tot_loss[loss=0.3295, simple_loss=0.3751, pruned_loss=0.142, over 3822164.55 frames. ], batch size: 74, lr: 2.58e-02, grad_scale: 8.0 +2023-04-01 00:13:37,278 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=15473.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:14:07,872 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=15498.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:14:14,391 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 00:14:17,663 INFO [train.py:903] (0/4) Epoch 3, batch 1850, loss[loss=0.2696, simple_loss=0.3197, pruned_loss=0.1097, over 19744.00 frames. ], tot_loss[loss=0.3282, simple_loss=0.3743, pruned_loss=0.141, over 3825430.47 frames. ], batch size: 46, lr: 2.58e-02, grad_scale: 8.0 +2023-04-01 00:14:32,125 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.736e+02 7.448e+02 9.407e+02 1.169e+03 2.273e+03, threshold=1.881e+03, percent-clipped=1.0 +2023-04-01 00:14:50,377 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 00:15:18,052 INFO [train.py:903] (0/4) Epoch 3, batch 1900, loss[loss=0.4264, simple_loss=0.4337, pruned_loss=0.2095, over 19673.00 frames. ], tot_loss[loss=0.3297, simple_loss=0.3753, pruned_loss=0.142, over 3810316.16 frames. ], batch size: 53, lr: 2.58e-02, grad_scale: 8.0 +2023-04-01 00:15:18,210 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=15557.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:15:36,172 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 00:15:41,730 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 00:16:04,583 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 00:16:19,662 INFO [train.py:903] (0/4) Epoch 3, batch 1950, loss[loss=0.4068, simple_loss=0.4305, pruned_loss=0.1916, over 13528.00 frames. ], tot_loss[loss=0.3316, simple_loss=0.3766, pruned_loss=0.1433, over 3801971.69 frames. ], batch size: 136, lr: 2.57e-02, grad_scale: 8.0 +2023-04-01 00:16:31,838 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-04-01 00:16:36,885 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.560e+02 7.789e+02 9.617e+02 1.296e+03 2.448e+03, threshold=1.923e+03, percent-clipped=3.0 +2023-04-01 00:17:22,855 INFO [train.py:903] (0/4) Epoch 3, batch 2000, loss[loss=0.2821, simple_loss=0.3302, pruned_loss=0.117, over 19759.00 frames. ], tot_loss[loss=0.3317, simple_loss=0.3769, pruned_loss=0.1432, over 3806604.49 frames. ], batch size: 48, lr: 2.57e-02, grad_scale: 8.0 +2023-04-01 00:17:41,425 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=15672.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:18:20,112 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 00:18:23,586 INFO [train.py:903] (0/4) Epoch 3, batch 2050, loss[loss=0.3066, simple_loss=0.3702, pruned_loss=0.1215, over 19781.00 frames. ], tot_loss[loss=0.3312, simple_loss=0.3766, pruned_loss=0.1429, over 3812689.17 frames. ], batch size: 63, lr: 2.57e-02, grad_scale: 8.0 +2023-04-01 00:18:38,231 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.734e+02 7.410e+02 9.269e+02 1.172e+03 2.915e+03, threshold=1.854e+03, percent-clipped=8.0 +2023-04-01 00:18:38,299 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 00:18:39,611 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 00:18:58,658 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 00:19:24,999 INFO [train.py:903] (0/4) Epoch 3, batch 2100, loss[loss=0.3852, simple_loss=0.4282, pruned_loss=0.1711, over 19488.00 frames. ], tot_loss[loss=0.3314, simple_loss=0.3772, pruned_loss=0.1428, over 3812064.95 frames. ], batch size: 64, lr: 2.56e-02, grad_scale: 8.0 +2023-04-01 00:19:52,227 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 00:20:09,577 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.2397, 4.8653, 6.0474, 5.8231, 1.9670, 5.5368, 4.9068, 5.4227], + device='cuda:0'), covar=tensor([0.0461, 0.0503, 0.0294, 0.0239, 0.3159, 0.0166, 0.0310, 0.0725], + device='cuda:0'), in_proj_covar=tensor([0.0363, 0.0336, 0.0458, 0.0345, 0.0476, 0.0236, 0.0306, 0.0439], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 00:20:13,816 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 00:20:25,854 INFO [train.py:903] (0/4) Epoch 3, batch 2150, loss[loss=0.3456, simple_loss=0.3928, pruned_loss=0.1492, over 19476.00 frames. ], tot_loss[loss=0.3305, simple_loss=0.3763, pruned_loss=0.1424, over 3817305.10 frames. ], batch size: 49, lr: 2.56e-02, grad_scale: 8.0 +2023-04-01 00:20:42,351 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.336e+02 7.356e+02 9.022e+02 1.284e+03 2.686e+03, threshold=1.804e+03, percent-clipped=4.0 +2023-04-01 00:21:28,851 INFO [train.py:903] (0/4) Epoch 3, batch 2200, loss[loss=0.2741, simple_loss=0.3449, pruned_loss=0.1016, over 19669.00 frames. ], tot_loss[loss=0.331, simple_loss=0.377, pruned_loss=0.1426, over 3818165.50 frames. ], batch size: 53, lr: 2.55e-02, grad_scale: 8.0 +2023-04-01 00:22:23,669 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=15901.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:22:30,390 INFO [train.py:903] (0/4) Epoch 3, batch 2250, loss[loss=0.3916, simple_loss=0.4093, pruned_loss=0.1869, over 13319.00 frames. ], tot_loss[loss=0.3302, simple_loss=0.3762, pruned_loss=0.142, over 3790084.53 frames. ], batch size: 136, lr: 2.55e-02, grad_scale: 8.0 +2023-04-01 00:22:44,786 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.843e+02 7.012e+02 9.226e+02 1.146e+03 2.721e+03, threshold=1.845e+03, percent-clipped=4.0 +2023-04-01 00:22:55,658 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=15928.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:23:08,212 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-01 00:23:26,925 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=15953.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:23:31,838 INFO [train.py:903] (0/4) Epoch 3, batch 2300, loss[loss=0.3393, simple_loss=0.3839, pruned_loss=0.1473, over 19707.00 frames. ], tot_loss[loss=0.3281, simple_loss=0.375, pruned_loss=0.1406, over 3811776.66 frames. ], batch size: 59, lr: 2.55e-02, grad_scale: 8.0 +2023-04-01 00:23:44,534 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 00:23:57,229 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-04-01 00:24:24,635 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-16000.pt +2023-04-01 00:24:33,426 INFO [train.py:903] (0/4) Epoch 3, batch 2350, loss[loss=0.4111, simple_loss=0.4177, pruned_loss=0.2022, over 13784.00 frames. ], tot_loss[loss=0.3294, simple_loss=0.3761, pruned_loss=0.1414, over 3811430.76 frames. ], batch size: 136, lr: 2.54e-02, grad_scale: 8.0 +2023-04-01 00:24:48,805 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.022e+02 7.519e+02 9.138e+02 1.115e+03 3.205e+03, threshold=1.828e+03, percent-clipped=8.0 +2023-04-01 00:24:49,111 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16019.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:25:07,118 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9965, 3.7229, 2.0468, 2.8067, 3.1147, 1.7671, 1.2743, 2.0071], + device='cuda:0'), covar=tensor([0.1218, 0.0336, 0.1008, 0.0508, 0.0558, 0.1052, 0.0969, 0.0738], + device='cuda:0'), in_proj_covar=tensor([0.0283, 0.0227, 0.0309, 0.0259, 0.0214, 0.0309, 0.0274, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 00:25:15,405 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 00:25:31,102 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 00:25:34,314 INFO [train.py:903] (0/4) Epoch 3, batch 2400, loss[loss=0.3421, simple_loss=0.3844, pruned_loss=0.1499, over 19574.00 frames. ], tot_loss[loss=0.3309, simple_loss=0.3772, pruned_loss=0.1423, over 3815405.00 frames. ], batch size: 52, lr: 2.54e-02, grad_scale: 8.0 +2023-04-01 00:25:58,796 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5703, 1.2804, 1.5247, 1.8512, 3.0938, 1.3551, 1.9760, 2.9000], + device='cuda:0'), covar=tensor([0.0311, 0.2526, 0.2423, 0.1397, 0.0472, 0.2066, 0.1125, 0.0500], + device='cuda:0'), in_proj_covar=tensor([0.0240, 0.0295, 0.0293, 0.0269, 0.0280, 0.0319, 0.0261, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 00:26:33,074 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16103.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:26:37,324 INFO [train.py:903] (0/4) Epoch 3, batch 2450, loss[loss=0.2919, simple_loss=0.3595, pruned_loss=0.1122, over 19539.00 frames. ], tot_loss[loss=0.3295, simple_loss=0.3761, pruned_loss=0.1414, over 3818734.29 frames. ], batch size: 56, lr: 2.54e-02, grad_scale: 8.0 +2023-04-01 00:26:51,589 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.658e+02 8.374e+02 9.822e+02 1.305e+03 3.634e+03, threshold=1.964e+03, percent-clipped=9.0 +2023-04-01 00:27:02,707 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0 +2023-04-01 00:27:38,195 INFO [train.py:903] (0/4) Epoch 3, batch 2500, loss[loss=0.342, simple_loss=0.3914, pruned_loss=0.1464, over 19791.00 frames. ], tot_loss[loss=0.329, simple_loss=0.3757, pruned_loss=0.1411, over 3808524.45 frames. ], batch size: 56, lr: 2.53e-02, grad_scale: 8.0 +2023-04-01 00:27:50,929 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16167.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:28:01,914 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.87 vs. limit=5.0 +2023-04-01 00:28:14,684 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2159, 1.0077, 1.4770, 1.2109, 1.8109, 1.9349, 1.9583, 0.4611], + device='cuda:0'), covar=tensor([0.1318, 0.2229, 0.1168, 0.1314, 0.0852, 0.0988, 0.0834, 0.2000], + device='cuda:0'), in_proj_covar=tensor([0.0398, 0.0437, 0.0403, 0.0380, 0.0472, 0.0391, 0.0556, 0.0403], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 00:28:40,060 INFO [train.py:903] (0/4) Epoch 3, batch 2550, loss[loss=0.3674, simple_loss=0.41, pruned_loss=0.1624, over 18197.00 frames. ], tot_loss[loss=0.3292, simple_loss=0.3756, pruned_loss=0.1413, over 3812605.52 frames. ], batch size: 83, lr: 2.53e-02, grad_scale: 8.0 +2023-04-01 00:28:56,244 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.521e+02 7.641e+02 9.209e+02 1.283e+03 2.881e+03, threshold=1.842e+03, percent-clipped=1.0 +2023-04-01 00:29:28,552 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=16245.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:29:36,513 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 00:29:44,020 INFO [train.py:903] (0/4) Epoch 3, batch 2600, loss[loss=0.305, simple_loss=0.3676, pruned_loss=0.1212, over 19742.00 frames. ], tot_loss[loss=0.3294, simple_loss=0.3754, pruned_loss=0.1417, over 3822441.43 frames. ], batch size: 54, lr: 2.53e-02, grad_scale: 8.0 +2023-04-01 00:30:46,336 INFO [train.py:903] (0/4) Epoch 3, batch 2650, loss[loss=0.3387, simple_loss=0.3826, pruned_loss=0.1474, over 19605.00 frames. ], tot_loss[loss=0.328, simple_loss=0.3747, pruned_loss=0.1407, over 3822391.54 frames. ], batch size: 61, lr: 2.52e-02, grad_scale: 8.0 +2023-04-01 00:30:47,125 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-01 00:31:00,230 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.667e+02 7.570e+02 8.863e+02 1.283e+03 4.568e+03, threshold=1.773e+03, percent-clipped=9.0 +2023-04-01 00:31:06,930 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 00:31:42,825 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7939, 3.9817, 4.4515, 4.3681, 1.5470, 3.9768, 3.6914, 3.8525], + device='cuda:0'), covar=tensor([0.0635, 0.0676, 0.0431, 0.0300, 0.3292, 0.0257, 0.0348, 0.0944], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0336, 0.0460, 0.0351, 0.0475, 0.0242, 0.0300, 0.0445], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 00:31:47,144 INFO [train.py:903] (0/4) Epoch 3, batch 2700, loss[loss=0.3812, simple_loss=0.4204, pruned_loss=0.171, over 18666.00 frames. ], tot_loss[loss=0.3275, simple_loss=0.3741, pruned_loss=0.1405, over 3823160.39 frames. ], batch size: 74, lr: 2.52e-02, grad_scale: 8.0 +2023-04-01 00:31:51,937 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16360.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:31:54,873 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=16363.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:32:26,204 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16389.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:32:47,343 INFO [train.py:903] (0/4) Epoch 3, batch 2750, loss[loss=0.3611, simple_loss=0.3907, pruned_loss=0.1657, over 19121.00 frames. ], tot_loss[loss=0.3282, simple_loss=0.3745, pruned_loss=0.1409, over 3827958.75 frames. ], batch size: 69, lr: 2.52e-02, grad_scale: 8.0 +2023-04-01 00:33:01,694 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.986e+02 7.672e+02 9.838e+02 1.209e+03 2.463e+03, threshold=1.968e+03, percent-clipped=5.0 +2023-04-01 00:33:35,971 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=16447.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:33:43,140 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.23 vs. limit=5.0 +2023-04-01 00:33:46,965 INFO [train.py:903] (0/4) Epoch 3, batch 2800, loss[loss=0.3561, simple_loss=0.3958, pruned_loss=0.1582, over 13842.00 frames. ], tot_loss[loss=0.33, simple_loss=0.3761, pruned_loss=0.142, over 3815117.50 frames. ], batch size: 136, lr: 2.51e-02, grad_scale: 8.0 +2023-04-01 00:34:03,332 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2777, 2.1080, 1.5887, 1.5211, 1.5640, 1.6154, 0.2566, 1.0196], + device='cuda:0'), covar=tensor([0.0159, 0.0201, 0.0159, 0.0214, 0.0397, 0.0260, 0.0429, 0.0353], + device='cuda:0'), in_proj_covar=tensor([0.0244, 0.0239, 0.0234, 0.0260, 0.0317, 0.0268, 0.0253, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 00:34:13,545 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16478.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:34:20,506 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-04-01 00:34:48,737 INFO [train.py:903] (0/4) Epoch 3, batch 2850, loss[loss=0.3964, simple_loss=0.4278, pruned_loss=0.1825, over 19623.00 frames. ], tot_loss[loss=0.33, simple_loss=0.376, pruned_loss=0.142, over 3823169.80 frames. ], batch size: 57, lr: 2.51e-02, grad_scale: 8.0 +2023-04-01 00:34:54,286 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=16511.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:35:03,282 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.663e+02 7.862e+02 1.093e+03 1.433e+03 3.382e+03, threshold=2.185e+03, percent-clipped=3.0 +2023-04-01 00:35:41,674 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-01 00:35:47,695 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 00:35:49,903 INFO [train.py:903] (0/4) Epoch 3, batch 2900, loss[loss=0.3205, simple_loss=0.373, pruned_loss=0.134, over 19517.00 frames. ], tot_loss[loss=0.3294, simple_loss=0.3753, pruned_loss=0.1417, over 3818731.29 frames. ], batch size: 54, lr: 2.51e-02, grad_scale: 8.0 +2023-04-01 00:35:57,068 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16562.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:36:38,540 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7610, 1.7997, 1.3948, 1.3182, 1.1784, 1.3552, 0.0587, 0.7782], + device='cuda:0'), covar=tensor([0.0183, 0.0188, 0.0125, 0.0164, 0.0412, 0.0223, 0.0345, 0.0290], + device='cuda:0'), in_proj_covar=tensor([0.0250, 0.0242, 0.0240, 0.0261, 0.0318, 0.0272, 0.0255, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 00:36:51,760 INFO [train.py:903] (0/4) Epoch 3, batch 2950, loss[loss=0.3053, simple_loss=0.3473, pruned_loss=0.1316, over 19744.00 frames. ], tot_loss[loss=0.3314, simple_loss=0.377, pruned_loss=0.1429, over 3800576.62 frames. ], batch size: 51, lr: 2.50e-02, grad_scale: 16.0 +2023-04-01 00:37:02,816 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16616.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:37:07,331 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.542e+02 7.271e+02 9.405e+02 1.170e+03 2.853e+03, threshold=1.881e+03, percent-clipped=4.0 +2023-04-01 00:37:15,863 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16626.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:37:33,960 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16641.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:37:52,197 INFO [train.py:903] (0/4) Epoch 3, batch 3000, loss[loss=0.3213, simple_loss=0.3551, pruned_loss=0.1437, over 19386.00 frames. ], tot_loss[loss=0.3294, simple_loss=0.3753, pruned_loss=0.1417, over 3816657.31 frames. ], batch size: 47, lr: 2.50e-02, grad_scale: 16.0 +2023-04-01 00:37:52,198 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 00:38:05,265 INFO [train.py:937] (0/4) Epoch 3, validation: loss=0.231, simple_loss=0.3246, pruned_loss=0.06867, over 944034.00 frames. +2023-04-01 00:38:05,266 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 00:38:08,702 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 00:38:17,985 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-04-01 00:39:07,683 INFO [train.py:903] (0/4) Epoch 3, batch 3050, loss[loss=0.307, simple_loss=0.3578, pruned_loss=0.1281, over 19749.00 frames. ], tot_loss[loss=0.3276, simple_loss=0.3742, pruned_loss=0.1405, over 3822313.90 frames. ], batch size: 51, lr: 2.49e-02, grad_scale: 16.0 +2023-04-01 00:39:22,586 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.950e+02 7.860e+02 1.014e+03 1.267e+03 1.851e+03, threshold=2.027e+03, percent-clipped=0.0 +2023-04-01 00:39:37,069 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.22 vs. limit=5.0 +2023-04-01 00:39:38,697 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=16733.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:39:40,019 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16734.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:40:08,118 INFO [train.py:903] (0/4) Epoch 3, batch 3100, loss[loss=0.2653, simple_loss=0.3265, pruned_loss=0.1021, over 19731.00 frames. ], tot_loss[loss=0.3254, simple_loss=0.3724, pruned_loss=0.1392, over 3827666.25 frames. ], batch size: 45, lr: 2.49e-02, grad_scale: 16.0 +2023-04-01 00:40:11,530 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16759.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:40:31,755 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3915, 1.0599, 1.4025, 0.9617, 2.4798, 2.9878, 2.8444, 3.2114], + device='cuda:0'), covar=tensor([0.1504, 0.2924, 0.3176, 0.2168, 0.0474, 0.0158, 0.0257, 0.0119], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0278, 0.0332, 0.0270, 0.0196, 0.0107, 0.0200, 0.0112], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 00:40:32,044 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-01 00:40:42,530 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4472, 1.3030, 1.2507, 1.7525, 1.3891, 1.7170, 1.8832, 1.5353], + device='cuda:0'), covar=tensor([0.0824, 0.1030, 0.1136, 0.0822, 0.0968, 0.0783, 0.0968, 0.0729], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0283, 0.0271, 0.0310, 0.0314, 0.0262, 0.0290, 0.0245], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 00:41:08,259 INFO [train.py:903] (0/4) Epoch 3, batch 3150, loss[loss=0.2872, simple_loss=0.3352, pruned_loss=0.1196, over 19712.00 frames. ], tot_loss[loss=0.3274, simple_loss=0.3741, pruned_loss=0.1403, over 3820316.70 frames. ], batch size: 46, lr: 2.49e-02, grad_scale: 16.0 +2023-04-01 00:41:20,755 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0756, 1.0971, 1.6337, 1.2653, 2.0792, 1.9863, 2.3024, 0.7791], + device='cuda:0'), covar=tensor([0.1650, 0.2599, 0.1419, 0.1446, 0.1020, 0.1204, 0.1067, 0.2317], + device='cuda:0'), in_proj_covar=tensor([0.0403, 0.0442, 0.0408, 0.0383, 0.0483, 0.0391, 0.0564, 0.0410], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 00:41:21,849 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16818.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:41:22,526 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.096e+02 7.784e+02 9.820e+02 1.270e+03 2.923e+03, threshold=1.964e+03, percent-clipped=2.0 +2023-04-01 00:41:31,802 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 00:41:51,823 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16843.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:41:57,406 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=16848.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:42:07,316 INFO [train.py:903] (0/4) Epoch 3, batch 3200, loss[loss=0.3938, simple_loss=0.4207, pruned_loss=0.1834, over 19614.00 frames. ], tot_loss[loss=0.3275, simple_loss=0.3743, pruned_loss=0.1403, over 3832477.02 frames. ], batch size: 57, lr: 2.48e-02, grad_scale: 8.0 +2023-04-01 00:42:26,199 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.64 vs. limit=5.0 +2023-04-01 00:42:39,227 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=16882.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:42:43,582 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16886.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 00:43:08,575 INFO [train.py:903] (0/4) Epoch 3, batch 3250, loss[loss=0.3453, simple_loss=0.3909, pruned_loss=0.1499, over 19576.00 frames. ], tot_loss[loss=0.3275, simple_loss=0.3744, pruned_loss=0.1403, over 3824923.03 frames. ], batch size: 61, lr: 2.48e-02, grad_scale: 8.0 +2023-04-01 00:43:08,965 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=16907.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:43:24,383 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.206e+02 8.507e+02 1.021e+03 1.288e+03 2.328e+03, threshold=2.042e+03, percent-clipped=1.0 +2023-04-01 00:43:29,307 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=16924.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:43:50,170 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-01 00:44:09,582 INFO [train.py:903] (0/4) Epoch 3, batch 3300, loss[loss=0.2954, simple_loss=0.3522, pruned_loss=0.1193, over 19858.00 frames. ], tot_loss[loss=0.326, simple_loss=0.3733, pruned_loss=0.1393, over 3822830.25 frames. ], batch size: 52, lr: 2.48e-02, grad_scale: 8.0 +2023-04-01 00:44:16,150 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 00:45:10,105 INFO [train.py:903] (0/4) Epoch 3, batch 3350, loss[loss=0.3393, simple_loss=0.3876, pruned_loss=0.1456, over 19306.00 frames. ], tot_loss[loss=0.326, simple_loss=0.3731, pruned_loss=0.1395, over 3816824.54 frames. ], batch size: 66, lr: 2.47e-02, grad_scale: 8.0 +2023-04-01 00:45:24,559 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.093e+02 7.847e+02 9.419e+02 1.175e+03 3.710e+03, threshold=1.884e+03, percent-clipped=5.0 +2023-04-01 00:46:10,035 INFO [train.py:903] (0/4) Epoch 3, batch 3400, loss[loss=0.2959, simple_loss=0.3482, pruned_loss=0.1218, over 19458.00 frames. ], tot_loss[loss=0.3268, simple_loss=0.3737, pruned_loss=0.1399, over 3813284.43 frames. ], batch size: 49, lr: 2.47e-02, grad_scale: 8.0 +2023-04-01 00:46:21,762 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6367, 1.2499, 1.2862, 1.7873, 1.6951, 1.8327, 2.1351, 1.7092], + device='cuda:0'), covar=tensor([0.0882, 0.1268, 0.1378, 0.1124, 0.1023, 0.0998, 0.0930, 0.0752], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0285, 0.0276, 0.0314, 0.0315, 0.0274, 0.0289, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 00:47:08,445 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17104.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:47:12,305 INFO [train.py:903] (0/4) Epoch 3, batch 3450, loss[loss=0.3395, simple_loss=0.3894, pruned_loss=0.1448, over 19440.00 frames. ], tot_loss[loss=0.3269, simple_loss=0.374, pruned_loss=0.1399, over 3800361.30 frames. ], batch size: 70, lr: 2.47e-02, grad_scale: 8.0 +2023-04-01 00:47:14,325 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 00:47:28,182 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.582e+02 8.544e+02 1.014e+03 1.278e+03 1.988e+03, threshold=2.028e+03, percent-clipped=3.0 +2023-04-01 00:47:39,729 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17129.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:48:12,439 INFO [train.py:903] (0/4) Epoch 3, batch 3500, loss[loss=0.3647, simple_loss=0.4108, pruned_loss=0.1593, over 18878.00 frames. ], tot_loss[loss=0.3302, simple_loss=0.3762, pruned_loss=0.1421, over 3790845.00 frames. ], batch size: 74, lr: 2.46e-02, grad_scale: 8.0 +2023-04-01 00:49:12,366 INFO [train.py:903] (0/4) Epoch 3, batch 3550, loss[loss=0.2831, simple_loss=0.3363, pruned_loss=0.115, over 19741.00 frames. ], tot_loss[loss=0.3305, simple_loss=0.3764, pruned_loss=0.1423, over 3798581.34 frames. ], batch size: 51, lr: 2.46e-02, grad_scale: 8.0 +2023-04-01 00:49:12,657 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17207.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:49:26,840 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.377e+02 8.429e+02 1.068e+03 1.302e+03 2.755e+03, threshold=2.137e+03, percent-clipped=4.0 +2023-04-01 00:49:40,870 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=17230.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 00:49:49,121 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-01 00:50:11,407 INFO [train.py:903] (0/4) Epoch 3, batch 3600, loss[loss=0.2778, simple_loss=0.3279, pruned_loss=0.1139, over 19486.00 frames. ], tot_loss[loss=0.3287, simple_loss=0.3754, pruned_loss=0.141, over 3823129.00 frames. ], batch size: 49, lr: 2.46e-02, grad_scale: 8.0 +2023-04-01 00:50:24,836 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=17268.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:50:56,817 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.99 vs. limit=5.0 +2023-04-01 00:51:11,855 INFO [train.py:903] (0/4) Epoch 3, batch 3650, loss[loss=0.3136, simple_loss=0.3722, pruned_loss=0.1275, over 19674.00 frames. ], tot_loss[loss=0.3267, simple_loss=0.3738, pruned_loss=0.1397, over 3820200.22 frames. ], batch size: 58, lr: 2.45e-02, grad_scale: 8.0 +2023-04-01 00:51:27,524 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.083e+02 7.894e+02 9.345e+02 1.119e+03 1.949e+03, threshold=1.869e+03, percent-clipped=0.0 +2023-04-01 00:51:57,327 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17345.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 00:52:12,711 INFO [train.py:903] (0/4) Epoch 3, batch 3700, loss[loss=0.3327, simple_loss=0.3655, pruned_loss=0.1499, over 19749.00 frames. ], tot_loss[loss=0.3235, simple_loss=0.3713, pruned_loss=0.1379, over 3826459.54 frames. ], batch size: 46, lr: 2.45e-02, grad_scale: 8.0 +2023-04-01 00:52:38,311 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4159, 1.2118, 1.6496, 0.8288, 2.5676, 2.9243, 2.6934, 3.0942], + device='cuda:0'), covar=tensor([0.1118, 0.2311, 0.2158, 0.1858, 0.0324, 0.0122, 0.0218, 0.0102], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0276, 0.0325, 0.0270, 0.0192, 0.0107, 0.0199, 0.0113], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 00:52:39,342 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17380.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:52:42,842 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17383.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:53:05,601 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17401.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 00:53:12,843 INFO [train.py:903] (0/4) Epoch 3, batch 3750, loss[loss=0.4477, simple_loss=0.4496, pruned_loss=0.2229, over 13148.00 frames. ], tot_loss[loss=0.3278, simple_loss=0.3744, pruned_loss=0.1406, over 3821082.86 frames. ], batch size: 136, lr: 2.45e-02, grad_scale: 8.0 +2023-04-01 00:53:27,566 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.653e+02 9.192e+02 1.060e+03 1.489e+03 3.397e+03, threshold=2.120e+03, percent-clipped=7.0 +2023-04-01 00:54:12,636 INFO [train.py:903] (0/4) Epoch 3, batch 3800, loss[loss=0.334, simple_loss=0.3815, pruned_loss=0.1433, over 17410.00 frames. ], tot_loss[loss=0.3248, simple_loss=0.3721, pruned_loss=0.1387, over 3819024.50 frames. ], batch size: 101, lr: 2.44e-02, grad_scale: 8.0 +2023-04-01 00:54:23,028 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2653, 1.1637, 1.2958, 0.7903, 2.5237, 2.8232, 2.6974, 3.0739], + device='cuda:0'), covar=tensor([0.1385, 0.2696, 0.2981, 0.2241, 0.0396, 0.0137, 0.0251, 0.0114], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0281, 0.0331, 0.0276, 0.0195, 0.0107, 0.0201, 0.0115], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 00:54:45,634 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 00:55:11,733 INFO [train.py:903] (0/4) Epoch 3, batch 3850, loss[loss=0.307, simple_loss=0.3519, pruned_loss=0.131, over 19737.00 frames. ], tot_loss[loss=0.326, simple_loss=0.3733, pruned_loss=0.1394, over 3829207.18 frames. ], batch size: 45, lr: 2.44e-02, grad_scale: 8.0 +2023-04-01 00:55:28,331 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.037e+02 7.944e+02 9.720e+02 1.209e+03 3.103e+03, threshold=1.944e+03, percent-clipped=2.0 +2023-04-01 00:56:01,956 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4680, 1.0891, 1.5689, 1.1808, 2.5674, 3.3517, 3.2868, 3.6515], + device='cuda:0'), covar=tensor([0.1294, 0.2751, 0.2743, 0.1976, 0.0443, 0.0123, 0.0182, 0.0086], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0280, 0.0326, 0.0274, 0.0191, 0.0107, 0.0200, 0.0113], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 00:56:04,060 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=17551.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:56:12,747 INFO [train.py:903] (0/4) Epoch 3, batch 3900, loss[loss=0.3141, simple_loss=0.3718, pruned_loss=0.1282, over 19784.00 frames. ], tot_loss[loss=0.3265, simple_loss=0.3736, pruned_loss=0.1396, over 3823323.23 frames. ], batch size: 56, lr: 2.44e-02, grad_scale: 8.0 +2023-04-01 00:56:16,667 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4473, 2.1289, 1.6154, 1.7051, 1.5032, 1.6528, 0.2161, 1.0981], + device='cuda:0'), covar=tensor([0.0173, 0.0220, 0.0177, 0.0266, 0.0438, 0.0293, 0.0441, 0.0351], + device='cuda:0'), in_proj_covar=tensor([0.0241, 0.0246, 0.0239, 0.0266, 0.0324, 0.0259, 0.0250, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 00:57:05,241 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17601.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 00:57:11,568 INFO [train.py:903] (0/4) Epoch 3, batch 3950, loss[loss=0.3056, simple_loss=0.3665, pruned_loss=0.1223, over 19588.00 frames. ], tot_loss[loss=0.3269, simple_loss=0.374, pruned_loss=0.1399, over 3817730.71 frames. ], batch size: 61, lr: 2.44e-02, grad_scale: 8.0 +2023-04-01 00:57:18,151 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 00:57:27,263 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.216e+02 7.211e+02 9.089e+02 1.148e+03 2.193e+03, threshold=1.818e+03, percent-clipped=2.0 +2023-04-01 00:57:34,442 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17626.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 00:57:51,660 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17639.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:58:03,602 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7087, 1.7390, 1.4273, 1.4966, 1.3630, 1.5401, 0.5910, 1.2366], + device='cuda:0'), covar=tensor([0.0182, 0.0200, 0.0119, 0.0165, 0.0300, 0.0210, 0.0386, 0.0264], + device='cuda:0'), in_proj_covar=tensor([0.0249, 0.0247, 0.0243, 0.0270, 0.0325, 0.0265, 0.0256, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 00:58:08,354 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.62 vs. limit=2.0 +2023-04-01 00:58:12,126 INFO [train.py:903] (0/4) Epoch 3, batch 4000, loss[loss=0.3975, simple_loss=0.4184, pruned_loss=0.1883, over 14071.00 frames. ], tot_loss[loss=0.3268, simple_loss=0.374, pruned_loss=0.1398, over 3810694.19 frames. ], batch size: 136, lr: 2.43e-02, grad_scale: 8.0 +2023-04-01 00:58:20,391 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17664.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:58:22,671 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17666.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:58:59,486 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 00:59:11,471 INFO [train.py:903] (0/4) Epoch 3, batch 4050, loss[loss=0.299, simple_loss=0.3554, pruned_loss=0.1213, over 19680.00 frames. ], tot_loss[loss=0.3249, simple_loss=0.3731, pruned_loss=0.1384, over 3818966.55 frames. ], batch size: 58, lr: 2.43e-02, grad_scale: 8.0 +2023-04-01 00:59:23,448 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2154, 2.1134, 2.0077, 3.3967, 2.2299, 3.9023, 3.0666, 1.9090], + device='cuda:0'), covar=tensor([0.1307, 0.0956, 0.0519, 0.0625, 0.1257, 0.0223, 0.0856, 0.0811], + device='cuda:0'), in_proj_covar=tensor([0.0478, 0.0445, 0.0449, 0.0594, 0.0527, 0.0358, 0.0548, 0.0446], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 00:59:25,591 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17718.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:59:28,910 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.065e+02 7.398e+02 9.459e+02 1.250e+03 4.446e+03, threshold=1.892e+03, percent-clipped=10.0 +2023-04-01 00:59:32,663 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=17724.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 00:59:57,743 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=17745.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 01:00:01,012 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17748.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:00:02,019 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5619, 3.8864, 4.1129, 4.0730, 1.5098, 3.8068, 3.3780, 3.7209], + device='cuda:0'), covar=tensor([0.0598, 0.0527, 0.0423, 0.0317, 0.3023, 0.0242, 0.0426, 0.0812], + device='cuda:0'), in_proj_covar=tensor([0.0379, 0.0346, 0.0476, 0.0367, 0.0491, 0.0251, 0.0317, 0.0459], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 01:00:10,715 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1099, 1.0064, 1.4719, 1.2191, 1.8558, 1.7764, 1.9609, 0.4815], + device='cuda:0'), covar=tensor([0.1463, 0.2334, 0.1213, 0.1347, 0.0876, 0.1206, 0.0824, 0.2079], + device='cuda:0'), in_proj_covar=tensor([0.0412, 0.0453, 0.0407, 0.0386, 0.0487, 0.0399, 0.0571, 0.0407], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 01:00:12,216 INFO [train.py:903] (0/4) Epoch 3, batch 4100, loss[loss=0.3527, simple_loss=0.3993, pruned_loss=0.153, over 19541.00 frames. ], tot_loss[loss=0.3259, simple_loss=0.3736, pruned_loss=0.139, over 3820790.24 frames. ], batch size: 54, lr: 2.43e-02, grad_scale: 4.0 +2023-04-01 01:00:48,448 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 01:01:11,680 INFO [train.py:903] (0/4) Epoch 3, batch 4150, loss[loss=0.341, simple_loss=0.3869, pruned_loss=0.1475, over 19467.00 frames. ], tot_loss[loss=0.3249, simple_loss=0.373, pruned_loss=0.1384, over 3818844.92 frames. ], batch size: 64, lr: 2.42e-02, grad_scale: 4.0 +2023-04-01 01:01:28,530 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.446e+02 7.784e+02 9.706e+02 1.186e+03 3.618e+03, threshold=1.941e+03, percent-clipped=3.0 +2023-04-01 01:01:49,967 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17839.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:02:10,745 INFO [train.py:903] (0/4) Epoch 3, batch 4200, loss[loss=0.3191, simple_loss=0.3624, pruned_loss=0.1379, over 19846.00 frames. ], tot_loss[loss=0.3252, simple_loss=0.373, pruned_loss=0.1387, over 3808465.95 frames. ], batch size: 52, lr: 2.42e-02, grad_scale: 4.0 +2023-04-01 01:02:14,251 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=17860.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 01:02:14,921 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 01:02:25,250 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0743, 1.2166, 1.3066, 1.6364, 2.6417, 1.2583, 1.8455, 2.6163], + device='cuda:0'), covar=tensor([0.0291, 0.2102, 0.2010, 0.1196, 0.0470, 0.1676, 0.0879, 0.0441], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0303, 0.0295, 0.0273, 0.0281, 0.0318, 0.0272, 0.0278], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:02:38,329 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7613, 1.5007, 1.8183, 1.6387, 2.9127, 4.5434, 4.4721, 4.9366], + device='cuda:0'), covar=tensor([0.1407, 0.2435, 0.2530, 0.1817, 0.0454, 0.0106, 0.0136, 0.0067], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0280, 0.0326, 0.0273, 0.0198, 0.0109, 0.0202, 0.0115], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 01:03:09,259 INFO [train.py:903] (0/4) Epoch 3, batch 4250, loss[loss=0.3471, simple_loss=0.3956, pruned_loss=0.1493, over 18776.00 frames. ], tot_loss[loss=0.3237, simple_loss=0.3718, pruned_loss=0.1378, over 3805527.78 frames. ], batch size: 74, lr: 2.42e-02, grad_scale: 4.0 +2023-04-01 01:03:11,737 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6272, 4.1959, 2.6513, 3.6433, 1.5063, 3.6965, 3.8257, 3.8740], + device='cuda:0'), covar=tensor([0.0546, 0.1072, 0.1816, 0.0736, 0.3272, 0.1022, 0.0572, 0.0788], + device='cuda:0'), in_proj_covar=tensor([0.0316, 0.0292, 0.0337, 0.0271, 0.0339, 0.0293, 0.0243, 0.0281], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:03:26,775 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.782e+02 8.334e+02 9.808e+02 1.259e+03 2.577e+03, threshold=1.962e+03, percent-clipped=5.0 +2023-04-01 01:03:26,832 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 01:03:28,267 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=17922.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:03:37,982 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 01:03:57,655 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=17947.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:04:08,380 INFO [train.py:903] (0/4) Epoch 3, batch 4300, loss[loss=0.3845, simple_loss=0.4141, pruned_loss=0.1775, over 18226.00 frames. ], tot_loss[loss=0.3238, simple_loss=0.3722, pruned_loss=0.1376, over 3804076.37 frames. ], batch size: 84, lr: 2.41e-02, grad_scale: 4.0 +2023-04-01 01:04:35,072 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=17978.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:04:38,680 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5312, 1.7783, 2.1423, 2.7284, 2.2343, 2.2243, 2.1610, 2.6619], + device='cuda:0'), covar=tensor([0.0690, 0.2028, 0.1236, 0.0784, 0.1143, 0.0531, 0.0893, 0.0575], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0383, 0.0287, 0.0256, 0.0317, 0.0267, 0.0274, 0.0243], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:05:00,891 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-18000.pt +2023-04-01 01:05:06,041 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 01:05:10,734 INFO [train.py:903] (0/4) Epoch 3, batch 4350, loss[loss=0.3226, simple_loss=0.3796, pruned_loss=0.1327, over 19682.00 frames. ], tot_loss[loss=0.3214, simple_loss=0.3703, pruned_loss=0.1363, over 3812075.67 frames. ], batch size: 60, lr: 2.41e-02, grad_scale: 4.0 +2023-04-01 01:05:27,020 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.929e+02 7.530e+02 9.485e+02 1.282e+03 2.824e+03, threshold=1.897e+03, percent-clipped=4.0 +2023-04-01 01:06:11,119 INFO [train.py:903] (0/4) Epoch 3, batch 4400, loss[loss=0.3213, simple_loss=0.3565, pruned_loss=0.1431, over 19614.00 frames. ], tot_loss[loss=0.3217, simple_loss=0.3702, pruned_loss=0.1366, over 3813121.93 frames. ], batch size: 50, lr: 2.41e-02, grad_scale: 8.0 +2023-04-01 01:06:17,178 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18062.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:06:32,857 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 01:06:43,596 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 01:06:53,801 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18092.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:06:57,446 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18095.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:07:10,934 INFO [train.py:903] (0/4) Epoch 3, batch 4450, loss[loss=0.3522, simple_loss=0.4007, pruned_loss=0.1519, over 19671.00 frames. ], tot_loss[loss=0.3223, simple_loss=0.371, pruned_loss=0.1368, over 3819376.77 frames. ], batch size: 60, lr: 2.40e-02, grad_scale: 8.0 +2023-04-01 01:07:16,929 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3120, 1.5131, 1.7040, 2.3050, 1.9220, 2.5871, 2.5059, 2.3679], + device='cuda:0'), covar=tensor([0.0737, 0.1173, 0.1226, 0.1293, 0.1237, 0.0716, 0.1131, 0.0715], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0283, 0.0277, 0.0311, 0.0313, 0.0266, 0.0283, 0.0250], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 01:07:21,271 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18116.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 01:07:26,625 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18120.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:07:28,235 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.891e+02 7.625e+02 9.248e+02 1.171e+03 2.408e+03, threshold=1.850e+03, percent-clipped=4.0 +2023-04-01 01:07:33,876 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8297, 3.4772, 2.2481, 3.0844, 1.1545, 3.0833, 3.0453, 3.2860], + device='cuda:0'), covar=tensor([0.0813, 0.1215, 0.2110, 0.0979, 0.3809, 0.1167, 0.0979, 0.0985], + device='cuda:0'), in_proj_covar=tensor([0.0320, 0.0286, 0.0335, 0.0278, 0.0343, 0.0297, 0.0248, 0.0278], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:07:53,433 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18141.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 01:08:05,272 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.80 vs. limit=2.0 +2023-04-01 01:08:11,158 INFO [train.py:903] (0/4) Epoch 3, batch 4500, loss[loss=0.3214, simple_loss=0.3668, pruned_loss=0.1379, over 19743.00 frames. ], tot_loss[loss=0.32, simple_loss=0.369, pruned_loss=0.1355, over 3819553.85 frames. ], batch size: 51, lr: 2.40e-02, grad_scale: 4.0 +2023-04-01 01:08:25,323 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.63 vs. limit=5.0 +2023-04-01 01:08:37,479 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18177.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:08:54,038 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18192.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:08:55,176 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18193.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:09:11,760 INFO [train.py:903] (0/4) Epoch 3, batch 4550, loss[loss=0.3594, simple_loss=0.4072, pruned_loss=0.1558, over 19752.00 frames. ], tot_loss[loss=0.3206, simple_loss=0.3696, pruned_loss=0.1358, over 3812257.02 frames. ], batch size: 63, lr: 2.40e-02, grad_scale: 4.0 +2023-04-01 01:09:12,105 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18207.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:09:18,463 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 01:09:29,444 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.087e+02 7.616e+02 9.596e+02 1.201e+03 2.125e+03, threshold=1.919e+03, percent-clipped=4.0 +2023-04-01 01:09:42,216 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 01:10:11,971 INFO [train.py:903] (0/4) Epoch 3, batch 4600, loss[loss=0.3821, simple_loss=0.4146, pruned_loss=0.1748, over 18114.00 frames. ], tot_loss[loss=0.3212, simple_loss=0.3699, pruned_loss=0.1362, over 3825813.93 frames. ], batch size: 83, lr: 2.39e-02, grad_scale: 4.0 +2023-04-01 01:11:11,634 INFO [train.py:903] (0/4) Epoch 3, batch 4650, loss[loss=0.3231, simple_loss=0.3773, pruned_loss=0.1345, over 19483.00 frames. ], tot_loss[loss=0.3219, simple_loss=0.3704, pruned_loss=0.1367, over 3824816.52 frames. ], batch size: 64, lr: 2.39e-02, grad_scale: 4.0 +2023-04-01 01:11:27,794 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 01:11:28,860 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.638e+02 7.851e+02 9.796e+02 1.308e+03 3.825e+03, threshold=1.959e+03, percent-clipped=6.0 +2023-04-01 01:11:29,035 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18322.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:11:38,749 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 01:12:06,729 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7558, 4.2710, 2.3685, 3.8850, 1.2782, 4.0240, 3.9623, 4.1797], + device='cuda:0'), covar=tensor([0.0473, 0.0915, 0.1991, 0.0622, 0.3470, 0.0780, 0.0591, 0.0706], + device='cuda:0'), in_proj_covar=tensor([0.0310, 0.0281, 0.0324, 0.0268, 0.0332, 0.0283, 0.0238, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:12:10,876 INFO [train.py:903] (0/4) Epoch 3, batch 4700, loss[loss=0.3548, simple_loss=0.3883, pruned_loss=0.1606, over 19494.00 frames. ], tot_loss[loss=0.3235, simple_loss=0.3716, pruned_loss=0.1377, over 3821717.98 frames. ], batch size: 49, lr: 2.39e-02, grad_scale: 4.0 +2023-04-01 01:12:33,343 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 01:13:13,556 INFO [train.py:903] (0/4) Epoch 3, batch 4750, loss[loss=0.3542, simple_loss=0.3978, pruned_loss=0.1553, over 19674.00 frames. ], tot_loss[loss=0.3247, simple_loss=0.3725, pruned_loss=0.1384, over 3827693.85 frames. ], batch size: 58, lr: 2.39e-02, grad_scale: 4.0 +2023-04-01 01:13:31,290 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.455e+02 7.810e+02 9.309e+02 1.222e+03 2.382e+03, threshold=1.862e+03, percent-clipped=4.0 +2023-04-01 01:13:44,348 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18433.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:13:48,810 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18437.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:14:11,429 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 01:14:15,107 INFO [train.py:903] (0/4) Epoch 3, batch 4800, loss[loss=0.3186, simple_loss=0.37, pruned_loss=0.1336, over 19335.00 frames. ], tot_loss[loss=0.323, simple_loss=0.3711, pruned_loss=0.1375, over 3808979.16 frames. ], batch size: 66, lr: 2.38e-02, grad_scale: 8.0 +2023-04-01 01:14:16,595 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18458.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:14:22,279 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18463.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:14:53,523 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18488.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:14:57,073 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4357, 1.3962, 2.5206, 1.6688, 3.1088, 3.0070, 3.5968, 1.3159], + device='cuda:0'), covar=tensor([0.1500, 0.2460, 0.1200, 0.1347, 0.1062, 0.1062, 0.1231, 0.2454], + device='cuda:0'), in_proj_covar=tensor([0.0417, 0.0459, 0.0419, 0.0394, 0.0498, 0.0404, 0.0581, 0.0415], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 01:15:16,093 INFO [train.py:903] (0/4) Epoch 3, batch 4850, loss[loss=0.2963, simple_loss=0.3529, pruned_loss=0.1199, over 19483.00 frames. ], tot_loss[loss=0.3229, simple_loss=0.3716, pruned_loss=0.1371, over 3809939.85 frames. ], batch size: 49, lr: 2.38e-02, grad_scale: 8.0 +2023-04-01 01:15:17,551 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18508.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:15:34,581 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.685e+02 7.295e+02 9.130e+02 1.063e+03 1.681e+03, threshold=1.826e+03, percent-clipped=0.0 +2023-04-01 01:15:38,194 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 01:15:40,614 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18526.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 01:15:52,386 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18536.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:15:53,567 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18537.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:16:00,134 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 01:16:04,624 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 01:16:05,762 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 01:16:14,590 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 01:16:15,642 INFO [train.py:903] (0/4) Epoch 3, batch 4900, loss[loss=0.3714, simple_loss=0.4115, pruned_loss=0.1656, over 19688.00 frames. ], tot_loss[loss=0.3232, simple_loss=0.3718, pruned_loss=0.1373, over 3819998.38 frames. ], batch size: 59, lr: 2.38e-02, grad_scale: 8.0 +2023-04-01 01:16:34,635 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 01:17:14,802 INFO [train.py:903] (0/4) Epoch 3, batch 4950, loss[loss=0.3265, simple_loss=0.3762, pruned_loss=0.1384, over 17509.00 frames. ], tot_loss[loss=0.3241, simple_loss=0.3723, pruned_loss=0.1379, over 3820503.77 frames. ], batch size: 101, lr: 2.37e-02, grad_scale: 4.0 +2023-04-01 01:17:30,520 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 01:17:34,977 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.825e+02 8.688e+02 1.059e+03 1.337e+03 3.400e+03, threshold=2.119e+03, percent-clipped=10.0 +2023-04-01 01:17:52,103 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 01:18:09,818 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18651.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:18:10,944 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18652.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:18:17,099 INFO [train.py:903] (0/4) Epoch 3, batch 5000, loss[loss=0.4049, simple_loss=0.4277, pruned_loss=0.1911, over 14078.00 frames. ], tot_loss[loss=0.3228, simple_loss=0.3715, pruned_loss=0.1371, over 3818691.83 frames. ], batch size: 135, lr: 2.37e-02, grad_scale: 4.0 +2023-04-01 01:18:21,590 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 01:18:32,453 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 01:18:59,277 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18692.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:19:00,335 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18693.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:19:16,433 INFO [train.py:903] (0/4) Epoch 3, batch 5050, loss[loss=0.3191, simple_loss=0.3759, pruned_loss=0.1311, over 19397.00 frames. ], tot_loss[loss=0.3221, simple_loss=0.371, pruned_loss=0.1366, over 3830222.65 frames. ], batch size: 66, lr: 2.37e-02, grad_scale: 4.0 +2023-04-01 01:19:29,215 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18718.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:19:35,283 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.108e+02 7.776e+02 1.028e+03 1.229e+03 3.550e+03, threshold=2.057e+03, percent-clipped=2.0 +2023-04-01 01:19:48,428 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 01:20:16,939 INFO [train.py:903] (0/4) Epoch 3, batch 5100, loss[loss=0.3294, simple_loss=0.3787, pruned_loss=0.14, over 19664.00 frames. ], tot_loss[loss=0.3225, simple_loss=0.3717, pruned_loss=0.1366, over 3824819.00 frames. ], batch size: 55, lr: 2.36e-02, grad_scale: 4.0 +2023-04-01 01:20:24,654 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 01:20:27,859 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 01:20:30,035 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.62 vs. limit=2.0 +2023-04-01 01:20:32,462 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 01:21:19,959 INFO [train.py:903] (0/4) Epoch 3, batch 5150, loss[loss=0.3224, simple_loss=0.3761, pruned_loss=0.1343, over 19732.00 frames. ], tot_loss[loss=0.3194, simple_loss=0.3691, pruned_loss=0.1349, over 3826517.13 frames. ], batch size: 63, lr: 2.36e-02, grad_scale: 4.0 +2023-04-01 01:21:31,925 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 01:21:40,054 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.229e+02 6.924e+02 8.047e+02 1.080e+03 1.932e+03, threshold=1.609e+03, percent-clipped=0.0 +2023-04-01 01:22:06,982 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 01:22:09,583 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.81 vs. limit=5.0 +2023-04-01 01:22:14,743 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18852.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:22:20,356 INFO [train.py:903] (0/4) Epoch 3, batch 5200, loss[loss=0.3627, simple_loss=0.4073, pruned_loss=0.159, over 19603.00 frames. ], tot_loss[loss=0.3204, simple_loss=0.3696, pruned_loss=0.1356, over 3833209.29 frames. ], batch size: 61, lr: 2.36e-02, grad_scale: 8.0 +2023-04-01 01:22:37,386 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 01:22:37,520 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=18870.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 01:22:59,117 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3925, 2.4316, 2.1768, 3.5532, 2.2754, 3.9313, 3.6377, 2.1954], + device='cuda:0'), covar=tensor([0.1319, 0.0882, 0.0493, 0.0543, 0.1194, 0.0222, 0.0735, 0.0751], + device='cuda:0'), in_proj_covar=tensor([0.0486, 0.0454, 0.0458, 0.0608, 0.0534, 0.0372, 0.0555, 0.0452], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 01:23:21,224 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 01:23:23,224 INFO [train.py:903] (0/4) Epoch 3, batch 5250, loss[loss=0.3435, simple_loss=0.3825, pruned_loss=0.1522, over 19657.00 frames. ], tot_loss[loss=0.321, simple_loss=0.37, pruned_loss=0.136, over 3819728.28 frames. ], batch size: 60, lr: 2.36e-02, grad_scale: 4.0 +2023-04-01 01:23:23,663 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18907.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:23:24,823 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=18908.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:23:42,374 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.727e+02 7.388e+02 9.793e+02 1.246e+03 4.620e+03, threshold=1.959e+03, percent-clipped=9.0 +2023-04-01 01:23:53,525 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18932.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:23:54,536 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=18933.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:24:12,744 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=18948.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:24:22,204 INFO [train.py:903] (0/4) Epoch 3, batch 5300, loss[loss=0.285, simple_loss=0.3369, pruned_loss=0.1165, over 19433.00 frames. ], tot_loss[loss=0.3199, simple_loss=0.3691, pruned_loss=0.1353, over 3815059.60 frames. ], batch size: 48, lr: 2.35e-02, grad_scale: 4.0 +2023-04-01 01:24:35,223 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18967.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:24:39,538 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 01:24:56,468 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=18985.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 01:25:22,318 INFO [train.py:903] (0/4) Epoch 3, batch 5350, loss[loss=0.296, simple_loss=0.3509, pruned_loss=0.1206, over 19515.00 frames. ], tot_loss[loss=0.3207, simple_loss=0.3696, pruned_loss=0.1359, over 3820214.36 frames. ], batch size: 54, lr: 2.35e-02, grad_scale: 4.0 +2023-04-01 01:25:42,784 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.207e+02 7.837e+02 1.011e+03 1.314e+03 3.062e+03, threshold=2.023e+03, percent-clipped=6.0 +2023-04-01 01:25:55,450 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 01:25:56,692 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19036.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:26:20,418 INFO [train.py:903] (0/4) Epoch 3, batch 5400, loss[loss=0.2819, simple_loss=0.3343, pruned_loss=0.1148, over 19740.00 frames. ], tot_loss[loss=0.3196, simple_loss=0.3689, pruned_loss=0.1352, over 3819446.33 frames. ], batch size: 51, lr: 2.35e-02, grad_scale: 4.0 +2023-04-01 01:27:22,179 INFO [train.py:903] (0/4) Epoch 3, batch 5450, loss[loss=0.315, simple_loss=0.3697, pruned_loss=0.1302, over 19587.00 frames. ], tot_loss[loss=0.32, simple_loss=0.3693, pruned_loss=0.1353, over 3811951.85 frames. ], batch size: 57, lr: 2.34e-02, grad_scale: 4.0 +2023-04-01 01:27:41,223 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.092e+02 7.941e+02 9.480e+02 1.159e+03 2.761e+03, threshold=1.896e+03, percent-clipped=1.0 +2023-04-01 01:27:49,159 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1281, 3.4775, 1.9783, 2.7210, 2.8199, 1.5439, 1.0911, 1.8073], + device='cuda:0'), covar=tensor([0.0984, 0.0384, 0.0957, 0.0387, 0.0499, 0.1085, 0.1009, 0.0661], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0233, 0.0310, 0.0251, 0.0215, 0.0309, 0.0275, 0.0245], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:28:04,314 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7891, 4.3007, 2.3895, 3.8410, 1.2171, 3.8936, 3.8011, 4.1418], + device='cuda:0'), covar=tensor([0.0490, 0.0993, 0.2004, 0.0631, 0.3715, 0.0820, 0.0699, 0.0709], + device='cuda:0'), in_proj_covar=tensor([0.0315, 0.0299, 0.0346, 0.0280, 0.0348, 0.0295, 0.0252, 0.0284], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:28:08,997 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9741, 1.3378, 0.9188, 1.0306, 1.1721, 0.8227, 0.4774, 1.2179], + device='cuda:0'), covar=tensor([0.0508, 0.0492, 0.1124, 0.0434, 0.0474, 0.1249, 0.0853, 0.0434], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0232, 0.0309, 0.0250, 0.0216, 0.0311, 0.0276, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:28:10,062 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19147.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 01:28:14,848 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19151.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:28:21,132 INFO [train.py:903] (0/4) Epoch 3, batch 5500, loss[loss=0.2894, simple_loss=0.3372, pruned_loss=0.1208, over 19754.00 frames. ], tot_loss[loss=0.3197, simple_loss=0.3691, pruned_loss=0.1351, over 3809092.25 frames. ], batch size: 46, lr: 2.34e-02, grad_scale: 4.0 +2023-04-01 01:28:45,301 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 01:28:58,487 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.4794, 4.0236, 2.5024, 3.6066, 1.2614, 3.5783, 3.6942, 3.8994], + device='cuda:0'), covar=tensor([0.0586, 0.1020, 0.1902, 0.0752, 0.3733, 0.0979, 0.0697, 0.0844], + device='cuda:0'), in_proj_covar=tensor([0.0318, 0.0296, 0.0344, 0.0281, 0.0348, 0.0296, 0.0253, 0.0286], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:29:11,725 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2128, 1.1395, 1.1127, 1.4989, 1.1590, 1.3343, 1.2892, 1.2542], + device='cuda:0'), covar=tensor([0.1089, 0.1327, 0.1440, 0.0945, 0.1084, 0.1109, 0.1160, 0.0887], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0283, 0.0271, 0.0310, 0.0311, 0.0259, 0.0282, 0.0250], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 01:29:20,074 INFO [train.py:903] (0/4) Epoch 3, batch 5550, loss[loss=0.3056, simple_loss=0.3598, pruned_loss=0.1258, over 19857.00 frames. ], tot_loss[loss=0.3199, simple_loss=0.3694, pruned_loss=0.1352, over 3819117.10 frames. ], batch size: 52, lr: 2.34e-02, grad_scale: 4.0 +2023-04-01 01:29:26,556 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3784, 1.0379, 1.3346, 1.2534, 2.1157, 1.1889, 1.7980, 2.0889], + device='cuda:0'), covar=tensor([0.0559, 0.2404, 0.2170, 0.1336, 0.0607, 0.1573, 0.0905, 0.0615], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0310, 0.0303, 0.0278, 0.0290, 0.0316, 0.0276, 0.0282], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:29:27,430 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 01:29:32,666 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-01 01:29:41,442 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19223.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:29:42,151 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.073e+02 7.947e+02 9.800e+02 1.342e+03 2.993e+03, threshold=1.960e+03, percent-clipped=6.0 +2023-04-01 01:30:02,026 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19241.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 01:30:09,705 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19248.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:30:15,624 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 01:30:21,461 INFO [train.py:903] (0/4) Epoch 3, batch 5600, loss[loss=0.3407, simple_loss=0.3873, pruned_loss=0.1471, over 19063.00 frames. ], tot_loss[loss=0.3171, simple_loss=0.367, pruned_loss=0.1336, over 3829164.57 frames. ], batch size: 69, lr: 2.34e-02, grad_scale: 8.0 +2023-04-01 01:30:33,673 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19266.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 01:31:03,988 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19292.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:31:18,606 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-04-01 01:31:22,068 INFO [train.py:903] (0/4) Epoch 3, batch 5650, loss[loss=0.3378, simple_loss=0.3787, pruned_loss=0.1485, over 18220.00 frames. ], tot_loss[loss=0.3167, simple_loss=0.3665, pruned_loss=0.1335, over 3830550.92 frames. ], batch size: 83, lr: 2.33e-02, grad_scale: 8.0 +2023-04-01 01:31:41,038 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.993e+02 7.403e+02 9.102e+02 1.185e+03 3.385e+03, threshold=1.820e+03, percent-clipped=3.0 +2023-04-01 01:32:09,534 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 01:32:15,629 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0431, 2.5271, 1.7809, 2.2705, 1.8391, 1.8328, 0.5436, 2.0725], + device='cuda:0'), covar=tensor([0.0215, 0.0256, 0.0247, 0.0280, 0.0434, 0.0379, 0.0565, 0.0428], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0255, 0.0254, 0.0281, 0.0339, 0.0268, 0.0266, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:32:21,816 INFO [train.py:903] (0/4) Epoch 3, batch 5700, loss[loss=0.2805, simple_loss=0.3319, pruned_loss=0.1145, over 19492.00 frames. ], tot_loss[loss=0.3163, simple_loss=0.3663, pruned_loss=0.1331, over 3833235.61 frames. ], batch size: 49, lr: 2.33e-02, grad_scale: 8.0 +2023-04-01 01:32:38,177 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-04-01 01:32:47,051 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4078, 1.3393, 1.0222, 1.3389, 1.1605, 1.2167, 1.0927, 1.2296], + device='cuda:0'), covar=tensor([0.0893, 0.1187, 0.1351, 0.0842, 0.0993, 0.0615, 0.0984, 0.0775], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0376, 0.0280, 0.0248, 0.0310, 0.0262, 0.0267, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:33:22,521 INFO [train.py:903] (0/4) Epoch 3, batch 5750, loss[loss=0.3201, simple_loss=0.3722, pruned_loss=0.134, over 19677.00 frames. ], tot_loss[loss=0.3174, simple_loss=0.3676, pruned_loss=0.1336, over 3833148.95 frames. ], batch size: 59, lr: 2.33e-02, grad_scale: 8.0 +2023-04-01 01:33:22,872 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19407.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:33:22,942 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19407.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:33:24,898 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 01:33:30,054 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19412.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:33:34,047 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 01:33:39,378 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 01:33:43,561 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.926e+02 7.525e+02 9.538e+02 1.231e+03 3.556e+03, threshold=1.908e+03, percent-clipped=6.0 +2023-04-01 01:33:53,866 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19432.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:34:22,185 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1653, 2.0817, 1.5499, 1.6766, 1.5064, 1.6681, 0.2214, 0.8457], + device='cuda:0'), covar=tensor([0.0203, 0.0190, 0.0143, 0.0189, 0.0410, 0.0246, 0.0481, 0.0368], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0247, 0.0247, 0.0272, 0.0331, 0.0259, 0.0257, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:34:22,889 INFO [train.py:903] (0/4) Epoch 3, batch 5800, loss[loss=0.3447, simple_loss=0.3911, pruned_loss=0.1492, over 19345.00 frames. ], tot_loss[loss=0.3181, simple_loss=0.3679, pruned_loss=0.1341, over 3828606.68 frames. ], batch size: 66, lr: 2.32e-02, grad_scale: 4.0 +2023-04-01 01:34:49,395 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9785, 2.0270, 2.1461, 2.6588, 2.0665, 2.7065, 2.9071, 2.9312], + device='cuda:0'), covar=tensor([0.0584, 0.1025, 0.1023, 0.1234, 0.1240, 0.0801, 0.0853, 0.0532], + device='cuda:0'), in_proj_covar=tensor([0.0245, 0.0275, 0.0266, 0.0300, 0.0307, 0.0249, 0.0274, 0.0244], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 01:35:03,997 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19491.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 01:35:07,671 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.94 vs. limit=5.0 +2023-04-01 01:35:23,138 INFO [train.py:903] (0/4) Epoch 3, batch 5850, loss[loss=0.3044, simple_loss=0.3653, pruned_loss=0.1217, over 19756.00 frames. ], tot_loss[loss=0.3182, simple_loss=0.3673, pruned_loss=0.1346, over 3835829.68 frames. ], batch size: 54, lr: 2.32e-02, grad_scale: 4.0 +2023-04-01 01:35:43,379 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.468e+02 8.354e+02 1.035e+03 1.319e+03 5.609e+03, threshold=2.070e+03, percent-clipped=8.0 +2023-04-01 01:35:54,668 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2784, 1.3122, 1.0830, 1.0120, 0.9411, 1.2592, 0.0115, 0.3904], + device='cuda:0'), covar=tensor([0.0233, 0.0215, 0.0137, 0.0164, 0.0475, 0.0157, 0.0368, 0.0353], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0250, 0.0251, 0.0278, 0.0332, 0.0264, 0.0257, 0.0267], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:36:23,588 INFO [train.py:903] (0/4) Epoch 3, batch 5900, loss[loss=0.2997, simple_loss=0.3558, pruned_loss=0.1218, over 19843.00 frames. ], tot_loss[loss=0.3195, simple_loss=0.3683, pruned_loss=0.1354, over 3825049.82 frames. ], batch size: 52, lr: 2.32e-02, grad_scale: 4.0 +2023-04-01 01:36:26,955 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 01:36:46,207 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 01:37:21,768 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19606.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 01:37:22,519 INFO [train.py:903] (0/4) Epoch 3, batch 5950, loss[loss=0.298, simple_loss=0.3611, pruned_loss=0.1174, over 19607.00 frames. ], tot_loss[loss=0.3196, simple_loss=0.3691, pruned_loss=0.1351, over 3825069.77 frames. ], batch size: 57, lr: 2.32e-02, grad_scale: 4.0 +2023-04-01 01:37:45,645 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.616e+02 6.894e+02 9.030e+02 1.163e+03 3.004e+03, threshold=1.806e+03, percent-clipped=5.0 +2023-04-01 01:37:59,423 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=19636.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:38:24,426 INFO [train.py:903] (0/4) Epoch 3, batch 6000, loss[loss=0.3386, simple_loss=0.3618, pruned_loss=0.1577, over 17772.00 frames. ], tot_loss[loss=0.32, simple_loss=0.3693, pruned_loss=0.1354, over 3821389.79 frames. ], batch size: 39, lr: 2.31e-02, grad_scale: 8.0 +2023-04-01 01:38:24,426 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 01:38:37,335 INFO [train.py:937] (0/4) Epoch 3, validation: loss=0.2218, simple_loss=0.3182, pruned_loss=0.06273, over 944034.00 frames. +2023-04-01 01:38:37,335 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 01:38:45,538 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19663.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:39:15,723 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19688.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:39:37,218 INFO [train.py:903] (0/4) Epoch 3, batch 6050, loss[loss=0.3333, simple_loss=0.3822, pruned_loss=0.1422, over 18322.00 frames. ], tot_loss[loss=0.3192, simple_loss=0.3686, pruned_loss=0.1349, over 3812244.67 frames. ], batch size: 84, lr: 2.31e-02, grad_scale: 8.0 +2023-04-01 01:39:59,308 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.085e+02 7.768e+02 9.451e+02 1.306e+03 2.772e+03, threshold=1.890e+03, percent-clipped=6.0 +2023-04-01 01:40:09,625 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-01 01:40:37,790 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19756.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:40:38,845 INFO [train.py:903] (0/4) Epoch 3, batch 6100, loss[loss=0.3236, simple_loss=0.3629, pruned_loss=0.1422, over 19409.00 frames. ], tot_loss[loss=0.3186, simple_loss=0.3681, pruned_loss=0.1345, over 3809771.56 frames. ], batch size: 48, lr: 2.31e-02, grad_scale: 8.0 +2023-04-01 01:41:38,886 INFO [train.py:903] (0/4) Epoch 3, batch 6150, loss[loss=0.3069, simple_loss=0.3722, pruned_loss=0.1208, over 19285.00 frames. ], tot_loss[loss=0.3204, simple_loss=0.3697, pruned_loss=0.1355, over 3810557.88 frames. ], batch size: 66, lr: 2.30e-02, grad_scale: 8.0 +2023-04-01 01:42:01,425 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.829e+02 8.602e+02 1.123e+03 1.510e+03 2.312e+03, threshold=2.246e+03, percent-clipped=7.0 +2023-04-01 01:42:04,685 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 01:42:38,569 INFO [train.py:903] (0/4) Epoch 3, batch 6200, loss[loss=0.268, simple_loss=0.3224, pruned_loss=0.1068, over 19820.00 frames. ], tot_loss[loss=0.3184, simple_loss=0.3674, pruned_loss=0.1346, over 3810416.10 frames. ], batch size: 47, lr: 2.30e-02, grad_scale: 8.0 +2023-04-01 01:42:46,184 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=19862.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 01:42:56,179 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=19871.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:43:15,863 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=19887.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 01:43:39,798 INFO [train.py:903] (0/4) Epoch 3, batch 6250, loss[loss=0.3011, simple_loss=0.3599, pruned_loss=0.1212, over 19531.00 frames. ], tot_loss[loss=0.3171, simple_loss=0.3667, pruned_loss=0.1338, over 3816534.33 frames. ], batch size: 56, lr: 2.30e-02, grad_scale: 8.0 +2023-04-01 01:44:01,767 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.496e+02 7.100e+02 9.208e+02 1.133e+03 3.490e+03, threshold=1.842e+03, percent-clipped=3.0 +2023-04-01 01:44:09,643 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 01:44:40,497 INFO [train.py:903] (0/4) Epoch 3, batch 6300, loss[loss=0.2416, simple_loss=0.3036, pruned_loss=0.08987, over 19726.00 frames. ], tot_loss[loss=0.3162, simple_loss=0.3661, pruned_loss=0.1331, over 3827202.04 frames. ], batch size: 46, lr: 2.30e-02, grad_scale: 8.0 +2023-04-01 01:44:55,611 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.72 vs. limit=5.0 +2023-04-01 01:45:07,281 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=19980.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:45:21,579 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4106, 0.9346, 1.1764, 1.3121, 2.0735, 1.0499, 1.7849, 2.0331], + device='cuda:0'), covar=tensor([0.0551, 0.2492, 0.2258, 0.1340, 0.0635, 0.1658, 0.0839, 0.0637], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0309, 0.0301, 0.0278, 0.0295, 0.0315, 0.0278, 0.0288], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:45:32,462 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-20000.pt +2023-04-01 01:45:41,421 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1744, 1.9034, 1.8954, 2.2918, 2.1539, 2.0250, 1.9476, 2.0925], + device='cuda:0'), covar=tensor([0.0570, 0.1107, 0.0902, 0.0541, 0.0740, 0.0394, 0.0703, 0.0438], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0381, 0.0288, 0.0245, 0.0310, 0.0261, 0.0270, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:45:42,110 INFO [train.py:903] (0/4) Epoch 3, batch 6350, loss[loss=0.369, simple_loss=0.4038, pruned_loss=0.1671, over 13031.00 frames. ], tot_loss[loss=0.3161, simple_loss=0.3661, pruned_loss=0.133, over 3831010.26 frames. ], batch size: 136, lr: 2.29e-02, grad_scale: 8.0 +2023-04-01 01:46:03,318 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.926e+02 7.424e+02 9.293e+02 1.158e+03 2.968e+03, threshold=1.859e+03, percent-clipped=5.0 +2023-04-01 01:46:20,197 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20038.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:46:42,148 INFO [train.py:903] (0/4) Epoch 3, batch 6400, loss[loss=0.3183, simple_loss=0.381, pruned_loss=0.1278, over 19276.00 frames. ], tot_loss[loss=0.3165, simple_loss=0.3667, pruned_loss=0.1331, over 3837961.01 frames. ], batch size: 66, lr: 2.29e-02, grad_scale: 8.0 +2023-04-01 01:46:47,220 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5892, 1.6182, 1.5125, 2.2912, 1.4614, 1.9080, 1.9860, 1.5093], + device='cuda:0'), covar=tensor([0.1095, 0.0906, 0.0578, 0.0432, 0.0942, 0.0406, 0.0992, 0.0902], + device='cuda:0'), in_proj_covar=tensor([0.0508, 0.0480, 0.0476, 0.0641, 0.0553, 0.0400, 0.0571, 0.0473], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 01:47:29,302 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20095.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:47:43,490 INFO [train.py:903] (0/4) Epoch 3, batch 6450, loss[loss=0.2783, simple_loss=0.3523, pruned_loss=0.1022, over 19654.00 frames. ], tot_loss[loss=0.3143, simple_loss=0.3653, pruned_loss=0.1316, over 3847948.26 frames. ], batch size: 55, lr: 2.29e-02, grad_scale: 8.0 +2023-04-01 01:47:43,803 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20107.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:47:51,714 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-04-01 01:47:53,546 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3169, 3.9165, 2.4082, 3.5171, 1.0447, 3.5337, 3.4346, 3.7387], + device='cuda:0'), covar=tensor([0.0689, 0.1322, 0.2129, 0.0794, 0.4374, 0.1077, 0.0907, 0.1078], + device='cuda:0'), in_proj_covar=tensor([0.0319, 0.0295, 0.0339, 0.0268, 0.0342, 0.0293, 0.0246, 0.0288], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:48:05,562 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.028e+02 7.833e+02 9.380e+02 1.145e+03 2.427e+03, threshold=1.876e+03, percent-clipped=3.0 +2023-04-01 01:48:08,304 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20127.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:48:12,594 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20130.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:48:29,105 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 01:48:39,367 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20152.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:48:42,970 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-01 01:48:44,549 INFO [train.py:903] (0/4) Epoch 3, batch 6500, loss[loss=0.3027, simple_loss=0.373, pruned_loss=0.1162, over 19541.00 frames. ], tot_loss[loss=0.3142, simple_loss=0.3652, pruned_loss=0.1316, over 3834721.63 frames. ], batch size: 56, lr: 2.29e-02, grad_scale: 8.0 +2023-04-01 01:48:52,306 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 01:49:45,353 INFO [train.py:903] (0/4) Epoch 3, batch 6550, loss[loss=0.3194, simple_loss=0.358, pruned_loss=0.1404, over 19382.00 frames. ], tot_loss[loss=0.3137, simple_loss=0.3647, pruned_loss=0.1314, over 3844589.29 frames. ], batch size: 47, lr: 2.28e-02, grad_scale: 8.0 +2023-04-01 01:50:03,598 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20222.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:50:07,448 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.125e+02 7.085e+02 9.596e+02 1.224e+03 2.377e+03, threshold=1.919e+03, percent-clipped=5.0 +2023-04-01 01:50:46,685 INFO [train.py:903] (0/4) Epoch 3, batch 6600, loss[loss=0.3421, simple_loss=0.3935, pruned_loss=0.1453, over 19087.00 frames. ], tot_loss[loss=0.3148, simple_loss=0.3658, pruned_loss=0.1319, over 3837422.34 frames. ], batch size: 69, lr: 2.28e-02, grad_scale: 8.0 +2023-04-01 01:51:19,942 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8441, 1.8616, 2.1142, 3.0267, 2.3464, 2.5398, 2.3354, 2.9620], + device='cuda:0'), covar=tensor([0.0630, 0.1779, 0.1239, 0.0664, 0.1133, 0.0429, 0.0910, 0.0421], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0383, 0.0288, 0.0251, 0.0320, 0.0265, 0.0275, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:51:35,733 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.57 vs. limit=2.0 +2023-04-01 01:51:47,898 INFO [train.py:903] (0/4) Epoch 3, batch 6650, loss[loss=0.3085, simple_loss=0.3685, pruned_loss=0.1243, over 19679.00 frames. ], tot_loss[loss=0.3152, simple_loss=0.3661, pruned_loss=0.1322, over 3839115.63 frames. ], batch size: 55, lr: 2.28e-02, grad_scale: 8.0 +2023-04-01 01:52:10,192 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.507e+02 7.672e+02 8.992e+02 1.144e+03 3.524e+03, threshold=1.798e+03, percent-clipped=4.0 +2023-04-01 01:52:42,304 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20351.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:52:49,537 INFO [train.py:903] (0/4) Epoch 3, batch 6700, loss[loss=0.2793, simple_loss=0.3261, pruned_loss=0.1163, over 19736.00 frames. ], tot_loss[loss=0.3164, simple_loss=0.3669, pruned_loss=0.1329, over 3808837.96 frames. ], batch size: 45, lr: 2.28e-02, grad_scale: 8.0 +2023-04-01 01:53:11,820 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20376.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:53:17,971 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20382.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:53:44,504 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20406.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:53:46,033 INFO [train.py:903] (0/4) Epoch 3, batch 6750, loss[loss=0.3045, simple_loss=0.3616, pruned_loss=0.1237, over 19686.00 frames. ], tot_loss[loss=0.317, simple_loss=0.3669, pruned_loss=0.1335, over 3808230.25 frames. ], batch size: 59, lr: 2.27e-02, grad_scale: 8.0 +2023-04-01 01:54:05,359 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.786e+02 7.667e+02 1.002e+03 1.269e+03 2.908e+03, threshold=2.004e+03, percent-clipped=6.0 +2023-04-01 01:54:17,377 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20436.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:54:26,459 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.02 vs. limit=2.0 +2023-04-01 01:54:34,793 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20451.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:54:41,604 INFO [train.py:903] (0/4) Epoch 3, batch 6800, loss[loss=0.3072, simple_loss=0.3667, pruned_loss=0.1239, over 19558.00 frames. ], tot_loss[loss=0.3151, simple_loss=0.3655, pruned_loss=0.1324, over 3812023.67 frames. ], batch size: 54, lr: 2.27e-02, grad_scale: 8.0 +2023-04-01 01:55:00,655 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20474.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:55:04,509 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6305, 1.6485, 1.9267, 1.7963, 2.6057, 2.8096, 2.8464, 3.0257], + device='cuda:0'), covar=tensor([0.1181, 0.2133, 0.2029, 0.1488, 0.0545, 0.0210, 0.0231, 0.0132], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0276, 0.0322, 0.0264, 0.0192, 0.0106, 0.0196, 0.0119], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 01:55:11,008 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-3.pt +2023-04-01 01:55:26,349 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 01:55:26,796 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 01:55:29,041 INFO [train.py:903] (0/4) Epoch 4, batch 0, loss[loss=0.3576, simple_loss=0.4, pruned_loss=0.1577, over 19750.00 frames. ], tot_loss[loss=0.3576, simple_loss=0.4, pruned_loss=0.1577, over 19750.00 frames. ], batch size: 63, lr: 2.12e-02, grad_scale: 8.0 +2023-04-01 01:55:29,041 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 01:55:40,521 INFO [train.py:937] (0/4) Epoch 4, validation: loss=0.2245, simple_loss=0.3205, pruned_loss=0.06426, over 944034.00 frames. +2023-04-01 01:55:40,521 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 01:55:53,634 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 01:55:55,117 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20497.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:56:27,860 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.490e+02 6.855e+02 8.790e+02 1.166e+03 2.960e+03, threshold=1.758e+03, percent-clipped=3.0 +2023-04-01 01:56:41,009 INFO [train.py:903] (0/4) Epoch 4, batch 50, loss[loss=0.2927, simple_loss=0.3635, pruned_loss=0.1109, over 19534.00 frames. ], tot_loss[loss=0.315, simple_loss=0.3649, pruned_loss=0.1325, over 871116.95 frames. ], batch size: 56, lr: 2.12e-02, grad_scale: 8.0 +2023-04-01 01:56:50,168 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.01 vs. limit=5.0 +2023-04-01 01:56:54,047 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.1088, 5.4775, 2.8471, 4.9270, 1.4283, 5.2270, 5.1771, 5.4814], + device='cuda:0'), covar=tensor([0.0409, 0.0999, 0.1922, 0.0498, 0.3982, 0.0725, 0.0608, 0.0582], + device='cuda:0'), in_proj_covar=tensor([0.0322, 0.0298, 0.0336, 0.0270, 0.0346, 0.0291, 0.0247, 0.0285], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:57:04,734 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4353, 1.1255, 1.3800, 1.7313, 3.0555, 1.2246, 2.1086, 3.1214], + device='cuda:0'), covar=tensor([0.0360, 0.2514, 0.2234, 0.1362, 0.0459, 0.2005, 0.1036, 0.0376], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0308, 0.0307, 0.0280, 0.0293, 0.0315, 0.0276, 0.0284], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:57:14,492 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 01:57:15,796 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20566.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:57:16,005 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20566.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:57:23,251 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.1299, 5.4082, 2.9952, 4.9719, 1.3403, 5.3902, 5.3414, 5.6359], + device='cuda:0'), covar=tensor([0.0357, 0.0915, 0.1507, 0.0467, 0.3511, 0.0567, 0.0426, 0.0503], + device='cuda:0'), in_proj_covar=tensor([0.0319, 0.0294, 0.0332, 0.0266, 0.0341, 0.0287, 0.0244, 0.0282], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 01:57:40,114 INFO [train.py:903] (0/4) Epoch 4, batch 100, loss[loss=0.2696, simple_loss=0.3185, pruned_loss=0.1104, over 18124.00 frames. ], tot_loss[loss=0.3149, simple_loss=0.3644, pruned_loss=0.1328, over 1529178.06 frames. ], batch size: 40, lr: 2.12e-02, grad_scale: 8.0 +2023-04-01 01:57:46,144 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20589.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:57:52,772 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 01:58:05,903 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=20606.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:58:29,320 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.104e+02 7.763e+02 9.275e+02 1.175e+03 2.763e+03, threshold=1.855e+03, percent-clipped=7.0 +2023-04-01 01:58:41,724 INFO [train.py:903] (0/4) Epoch 4, batch 150, loss[loss=0.282, simple_loss=0.3411, pruned_loss=0.1115, over 19742.00 frames. ], tot_loss[loss=0.3099, simple_loss=0.3614, pruned_loss=0.1292, over 2033537.98 frames. ], batch size: 51, lr: 2.11e-02, grad_scale: 8.0 +2023-04-01 01:58:43,271 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0082, 1.8397, 1.3624, 1.3256, 1.6604, 0.8844, 0.6760, 1.5162], + device='cuda:0'), covar=tensor([0.0734, 0.0528, 0.0929, 0.0543, 0.0495, 0.1189, 0.0775, 0.0377], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0238, 0.0308, 0.0248, 0.0211, 0.0306, 0.0269, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 01:59:35,903 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20681.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 01:59:40,169 INFO [train.py:903] (0/4) Epoch 4, batch 200, loss[loss=0.2762, simple_loss=0.3337, pruned_loss=0.1093, over 19740.00 frames. ], tot_loss[loss=0.3103, simple_loss=0.3619, pruned_loss=0.1294, over 2436469.65 frames. ], batch size: 51, lr: 2.11e-02, grad_scale: 8.0 +2023-04-01 01:59:41,292 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 01:59:43,164 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-04-01 02:00:16,602 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1350, 1.1809, 1.9491, 1.3045, 2.3465, 2.3289, 2.6391, 0.8502], + device='cuda:0'), covar=tensor([0.1564, 0.2531, 0.1174, 0.1338, 0.1029, 0.1069, 0.1021, 0.2286], + device='cuda:0'), in_proj_covar=tensor([0.0424, 0.0471, 0.0437, 0.0399, 0.0517, 0.0414, 0.0593, 0.0421], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 02:00:28,682 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.823e+02 7.097e+02 9.184e+02 1.257e+03 2.857e+03, threshold=1.837e+03, percent-clipped=5.0 +2023-04-01 02:00:39,435 INFO [train.py:903] (0/4) Epoch 4, batch 250, loss[loss=0.3275, simple_loss=0.3883, pruned_loss=0.1333, over 19566.00 frames. ], tot_loss[loss=0.3125, simple_loss=0.3636, pruned_loss=0.1307, over 2740885.29 frames. ], batch size: 61, lr: 2.11e-02, grad_scale: 8.0 +2023-04-01 02:00:43,909 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7513, 1.0024, 1.3080, 2.1059, 1.6887, 2.0082, 2.1021, 1.6291], + device='cuda:0'), covar=tensor([0.0796, 0.1373, 0.1371, 0.0987, 0.1163, 0.0756, 0.1135, 0.0876], + device='cuda:0'), in_proj_covar=tensor([0.0243, 0.0276, 0.0267, 0.0295, 0.0305, 0.0248, 0.0268, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 02:00:53,622 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4847, 1.1196, 1.1412, 1.7356, 1.3939, 1.7049, 1.7821, 1.4039], + device='cuda:0'), covar=tensor([0.0745, 0.1161, 0.1304, 0.0813, 0.0967, 0.0690, 0.0881, 0.0789], + device='cuda:0'), in_proj_covar=tensor([0.0242, 0.0274, 0.0266, 0.0294, 0.0304, 0.0247, 0.0266, 0.0240], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 02:00:57,775 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20750.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:01:02,455 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20753.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:01:32,668 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20778.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:01:34,841 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20780.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:01:41,109 INFO [train.py:903] (0/4) Epoch 4, batch 300, loss[loss=0.3494, simple_loss=0.3946, pruned_loss=0.1521, over 19747.00 frames. ], tot_loss[loss=0.3117, simple_loss=0.3635, pruned_loss=0.13, over 2988407.39 frames. ], batch size: 63, lr: 2.11e-02, grad_scale: 8.0 +2023-04-01 02:02:15,290 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-04-01 02:02:25,184 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20822.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:02:29,141 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.541e+02 7.559e+02 9.012e+02 1.206e+03 2.235e+03, threshold=1.802e+03, percent-clipped=6.0 +2023-04-01 02:02:40,179 INFO [train.py:903] (0/4) Epoch 4, batch 350, loss[loss=0.4024, simple_loss=0.4269, pruned_loss=0.1889, over 13341.00 frames. ], tot_loss[loss=0.3122, simple_loss=0.3638, pruned_loss=0.1302, over 3160274.76 frames. ], batch size: 136, lr: 2.10e-02, grad_scale: 8.0 +2023-04-01 02:02:45,648 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 02:02:52,654 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20845.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:02:54,955 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20847.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:03:18,340 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20865.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:03:20,549 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0577, 1.9441, 1.3599, 1.3706, 1.7611, 0.9431, 0.8902, 1.5735], + device='cuda:0'), covar=tensor([0.0631, 0.0364, 0.0875, 0.0422, 0.0325, 0.1025, 0.0621, 0.0329], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0245, 0.0314, 0.0246, 0.0212, 0.0314, 0.0277, 0.0251], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:03:23,857 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20870.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:03:27,190 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3975, 1.6793, 1.8371, 2.4398, 1.8660, 2.6698, 2.9327, 2.6804], + device='cuda:0'), covar=tensor([0.0776, 0.1228, 0.1228, 0.1254, 0.1368, 0.0739, 0.1043, 0.0653], + device='cuda:0'), in_proj_covar=tensor([0.0244, 0.0272, 0.0266, 0.0297, 0.0304, 0.0248, 0.0267, 0.0240], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 02:03:34,713 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3157, 2.4054, 1.5751, 1.7950, 2.1563, 1.1951, 1.1822, 1.5876], + device='cuda:0'), covar=tensor([0.0713, 0.0320, 0.0883, 0.0408, 0.0364, 0.0975, 0.0602, 0.0421], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0241, 0.0312, 0.0244, 0.0211, 0.0310, 0.0273, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:03:40,719 INFO [train.py:903] (0/4) Epoch 4, batch 400, loss[loss=0.3864, simple_loss=0.4201, pruned_loss=0.1764, over 19753.00 frames. ], tot_loss[loss=0.3121, simple_loss=0.3635, pruned_loss=0.1303, over 3307434.16 frames. ], batch size: 63, lr: 2.10e-02, grad_scale: 8.0 +2023-04-01 02:03:52,327 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=20895.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:04:27,878 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.312e+02 7.529e+02 9.870e+02 1.265e+03 2.610e+03, threshold=1.974e+03, percent-clipped=3.0 +2023-04-01 02:04:39,096 INFO [train.py:903] (0/4) Epoch 4, batch 450, loss[loss=0.3524, simple_loss=0.394, pruned_loss=0.1554, over 18233.00 frames. ], tot_loss[loss=0.3127, simple_loss=0.3639, pruned_loss=0.1307, over 3410484.38 frames. ], batch size: 84, lr: 2.10e-02, grad_scale: 8.0 +2023-04-01 02:04:41,848 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=20937.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:04:58,412 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=20950.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:05:02,976 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8396, 4.4872, 2.6621, 3.9905, 1.5237, 4.1128, 3.9366, 3.9894], + device='cuda:0'), covar=tensor([0.0460, 0.0839, 0.1766, 0.0578, 0.3132, 0.0758, 0.0613, 0.0859], + device='cuda:0'), in_proj_covar=tensor([0.0331, 0.0303, 0.0345, 0.0275, 0.0343, 0.0294, 0.0250, 0.0294], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 02:05:12,742 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 02:05:13,091 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=20962.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:05:13,941 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 02:05:40,636 INFO [train.py:903] (0/4) Epoch 4, batch 500, loss[loss=0.3377, simple_loss=0.3928, pruned_loss=0.1413, over 19545.00 frames. ], tot_loss[loss=0.3111, simple_loss=0.3624, pruned_loss=0.1299, over 3507794.81 frames. ], batch size: 56, lr: 2.10e-02, grad_scale: 8.0 +2023-04-01 02:06:27,513 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.944e+02 7.370e+02 9.144e+02 1.191e+03 3.185e+03, threshold=1.829e+03, percent-clipped=4.0 +2023-04-01 02:06:27,865 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8049, 1.5474, 1.5470, 1.8462, 3.3371, 1.2410, 2.2007, 3.4309], + device='cuda:0'), covar=tensor([0.0315, 0.2128, 0.2111, 0.1353, 0.0420, 0.2046, 0.1033, 0.0365], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0304, 0.0301, 0.0282, 0.0295, 0.0316, 0.0281, 0.0285], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:06:40,256 INFO [train.py:903] (0/4) Epoch 4, batch 550, loss[loss=0.2578, simple_loss=0.3262, pruned_loss=0.09475, over 19585.00 frames. ], tot_loss[loss=0.3098, simple_loss=0.3618, pruned_loss=0.1289, over 3585643.90 frames. ], batch size: 52, lr: 2.10e-02, grad_scale: 8.0 +2023-04-01 02:07:08,543 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3497, 2.3799, 1.6128, 1.5574, 2.1035, 1.0518, 1.0411, 1.8452], + device='cuda:0'), covar=tensor([0.0815, 0.0455, 0.0965, 0.0541, 0.0360, 0.1147, 0.0730, 0.0377], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0241, 0.0306, 0.0241, 0.0209, 0.0305, 0.0267, 0.0246], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:07:08,877 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-01 02:07:17,078 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21065.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:07:23,301 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21070.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:07:24,893 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-01 02:07:39,554 INFO [train.py:903] (0/4) Epoch 4, batch 600, loss[loss=0.3488, simple_loss=0.4005, pruned_loss=0.1485, over 19471.00 frames. ], tot_loss[loss=0.3091, simple_loss=0.3613, pruned_loss=0.1285, over 3646176.26 frames. ], batch size: 64, lr: 2.09e-02, grad_scale: 8.0 +2023-04-01 02:08:19,255 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 02:08:23,052 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21121.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:08:27,137 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.199e+02 8.023e+02 1.001e+03 1.305e+03 2.804e+03, threshold=2.003e+03, percent-clipped=3.0 +2023-04-01 02:08:39,444 INFO [train.py:903] (0/4) Epoch 4, batch 650, loss[loss=0.2466, simple_loss=0.3036, pruned_loss=0.09479, over 19737.00 frames. ], tot_loss[loss=0.3076, simple_loss=0.3603, pruned_loss=0.1275, over 3684096.38 frames. ], batch size: 45, lr: 2.09e-02, grad_scale: 8.0 +2023-04-01 02:08:40,602 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21136.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:08:52,206 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21146.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:09:00,276 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21151.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:09:29,101 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21176.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:09:38,635 INFO [train.py:903] (0/4) Epoch 4, batch 700, loss[loss=0.3206, simple_loss=0.3542, pruned_loss=0.1435, over 19385.00 frames. ], tot_loss[loss=0.3071, simple_loss=0.3601, pruned_loss=0.1271, over 3724784.32 frames. ], batch size: 47, lr: 2.09e-02, grad_scale: 8.0 +2023-04-01 02:10:26,834 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.035e+02 7.118e+02 8.929e+02 1.162e+03 2.438e+03, threshold=1.786e+03, percent-clipped=3.0 +2023-04-01 02:10:40,519 INFO [train.py:903] (0/4) Epoch 4, batch 750, loss[loss=0.3277, simple_loss=0.3812, pruned_loss=0.1371, over 19570.00 frames. ], tot_loss[loss=0.3075, simple_loss=0.3605, pruned_loss=0.1273, over 3750877.99 frames. ], batch size: 61, lr: 2.09e-02, grad_scale: 8.0 +2023-04-01 02:11:40,043 INFO [train.py:903] (0/4) Epoch 4, batch 800, loss[loss=0.3132, simple_loss=0.3769, pruned_loss=0.1247, over 19659.00 frames. ], tot_loss[loss=0.3103, simple_loss=0.3625, pruned_loss=0.1291, over 3758970.39 frames. ], batch size: 55, lr: 2.08e-02, grad_scale: 8.0 +2023-04-01 02:11:56,119 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 02:12:25,416 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21321.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:12:30,559 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.379e+02 8.113e+02 1.001e+03 1.207e+03 2.017e+03, threshold=2.002e+03, percent-clipped=2.0 +2023-04-01 02:12:35,494 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21330.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:12:41,604 INFO [train.py:903] (0/4) Epoch 4, batch 850, loss[loss=0.3478, simple_loss=0.3914, pruned_loss=0.152, over 18785.00 frames. ], tot_loss[loss=0.3096, simple_loss=0.3621, pruned_loss=0.1286, over 3767929.43 frames. ], batch size: 74, lr: 2.08e-02, grad_scale: 8.0 +2023-04-01 02:12:42,260 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-01 02:12:54,252 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21346.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:12:54,350 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21346.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:13:30,947 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 02:13:39,983 INFO [train.py:903] (0/4) Epoch 4, batch 900, loss[loss=0.3283, simple_loss=0.367, pruned_loss=0.1448, over 13443.00 frames. ], tot_loss[loss=0.3121, simple_loss=0.3637, pruned_loss=0.1302, over 3769789.89 frames. ], batch size: 137, lr: 2.08e-02, grad_scale: 8.0 +2023-04-01 02:13:42,754 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5024, 1.3380, 1.0761, 1.2711, 1.1594, 1.3206, 1.1058, 1.3336], + device='cuda:0'), covar=tensor([0.0851, 0.1104, 0.1402, 0.0865, 0.0956, 0.0576, 0.1032, 0.0745], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0382, 0.0286, 0.0256, 0.0317, 0.0266, 0.0270, 0.0240], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:14:14,503 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21414.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:14:28,712 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.494e+02 7.656e+02 9.192e+02 1.106e+03 2.022e+03, threshold=1.838e+03, percent-clipped=1.0 +2023-04-01 02:14:40,651 INFO [train.py:903] (0/4) Epoch 4, batch 950, loss[loss=0.2991, simple_loss=0.3484, pruned_loss=0.1249, over 19671.00 frames. ], tot_loss[loss=0.3107, simple_loss=0.3628, pruned_loss=0.1293, over 3791745.30 frames. ], batch size: 53, lr: 2.08e-02, grad_scale: 8.0 +2023-04-01 02:14:43,013 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 02:15:35,413 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21480.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:15:40,648 INFO [train.py:903] (0/4) Epoch 4, batch 1000, loss[loss=0.358, simple_loss=0.395, pruned_loss=0.1605, over 19771.00 frames. ], tot_loss[loss=0.3095, simple_loss=0.3615, pruned_loss=0.1287, over 3801690.28 frames. ], batch size: 54, lr: 2.07e-02, grad_scale: 8.0 +2023-04-01 02:16:29,780 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.128e+02 7.436e+02 9.242e+02 1.292e+03 2.692e+03, threshold=1.848e+03, percent-clipped=7.0 +2023-04-01 02:16:33,308 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 02:16:33,621 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21529.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:16:40,203 INFO [train.py:903] (0/4) Epoch 4, batch 1050, loss[loss=0.2853, simple_loss=0.3447, pruned_loss=0.1129, over 19846.00 frames. ], tot_loss[loss=0.3091, simple_loss=0.3614, pruned_loss=0.1284, over 3814422.65 frames. ], batch size: 52, lr: 2.07e-02, grad_scale: 8.0 +2023-04-01 02:17:14,232 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 02:17:40,177 INFO [train.py:903] (0/4) Epoch 4, batch 1100, loss[loss=0.2963, simple_loss=0.3657, pruned_loss=0.1134, over 19772.00 frames. ], tot_loss[loss=0.3103, simple_loss=0.3625, pruned_loss=0.129, over 3818749.66 frames. ], batch size: 56, lr: 2.07e-02, grad_scale: 8.0 +2023-04-01 02:17:52,953 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21595.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:18:03,973 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2087, 1.3155, 0.9371, 0.9989, 0.9963, 1.1497, 0.0090, 0.3861], + device='cuda:0'), covar=tensor([0.0250, 0.0243, 0.0186, 0.0199, 0.0490, 0.0203, 0.0426, 0.0389], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0262, 0.0258, 0.0285, 0.0338, 0.0275, 0.0264, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 02:18:15,033 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21613.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:18:30,559 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.767e+02 7.921e+02 9.622e+02 1.275e+03 2.981e+03, threshold=1.924e+03, percent-clipped=6.0 +2023-04-01 02:18:36,434 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21631.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:18:43,699 INFO [train.py:903] (0/4) Epoch 4, batch 1150, loss[loss=0.2655, simple_loss=0.307, pruned_loss=0.112, over 19743.00 frames. ], tot_loss[loss=0.3098, simple_loss=0.362, pruned_loss=0.1288, over 3810166.27 frames. ], batch size: 45, lr: 2.07e-02, grad_scale: 8.0 +2023-04-01 02:19:30,748 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21674.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:19:45,550 INFO [train.py:903] (0/4) Epoch 4, batch 1200, loss[loss=0.3354, simple_loss=0.377, pruned_loss=0.1469, over 19726.00 frames. ], tot_loss[loss=0.308, simple_loss=0.3606, pruned_loss=0.1278, over 3810808.14 frames. ], batch size: 63, lr: 2.07e-02, grad_scale: 8.0 +2023-04-01 02:19:51,412 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21690.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:19:54,755 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.5027, 5.0181, 2.9964, 4.3813, 1.4298, 4.5795, 4.5511, 4.8256], + device='cuda:0'), covar=tensor([0.0438, 0.0891, 0.1497, 0.0581, 0.3882, 0.0722, 0.0615, 0.0672], + device='cuda:0'), in_proj_covar=tensor([0.0327, 0.0302, 0.0342, 0.0278, 0.0348, 0.0295, 0.0259, 0.0289], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 02:20:14,889 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 02:20:35,528 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.245e+02 7.515e+02 9.038e+02 1.111e+03 2.454e+03, threshold=1.808e+03, percent-clipped=2.0 +2023-04-01 02:20:45,434 INFO [train.py:903] (0/4) Epoch 4, batch 1250, loss[loss=0.2433, simple_loss=0.303, pruned_loss=0.09183, over 19743.00 frames. ], tot_loss[loss=0.3077, simple_loss=0.3603, pruned_loss=0.1276, over 3800286.15 frames. ], batch size: 47, lr: 2.06e-02, grad_scale: 8.0 +2023-04-01 02:21:39,797 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21781.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:21:44,776 INFO [train.py:903] (0/4) Epoch 4, batch 1300, loss[loss=0.2831, simple_loss=0.3348, pruned_loss=0.1157, over 19291.00 frames. ], tot_loss[loss=0.3055, simple_loss=0.3587, pruned_loss=0.1262, over 3805951.96 frames. ], batch size: 44, lr: 2.06e-02, grad_scale: 8.0 +2023-04-01 02:21:45,186 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21785.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:21:49,660 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21789.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:22:08,377 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0 +2023-04-01 02:22:10,049 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=21805.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:22:15,818 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21810.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:22:34,649 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.842e+02 7.410e+02 9.034e+02 1.121e+03 1.849e+03, threshold=1.807e+03, percent-clipped=1.0 +2023-04-01 02:22:45,118 INFO [train.py:903] (0/4) Epoch 4, batch 1350, loss[loss=0.3001, simple_loss=0.3537, pruned_loss=0.1233, over 19666.00 frames. ], tot_loss[loss=0.3056, simple_loss=0.3589, pruned_loss=0.1262, over 3802811.55 frames. ], batch size: 53, lr: 2.06e-02, grad_scale: 8.0 +2023-04-01 02:23:04,038 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=21851.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:23:23,930 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21868.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:23:33,929 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=21876.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:23:45,407 INFO [train.py:903] (0/4) Epoch 4, batch 1400, loss[loss=0.2864, simple_loss=0.332, pruned_loss=0.1204, over 19025.00 frames. ], tot_loss[loss=0.3072, simple_loss=0.3599, pruned_loss=0.1272, over 3803445.56 frames. ], batch size: 42, lr: 2.06e-02, grad_scale: 8.0 +2023-04-01 02:24:15,793 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.77 vs. limit=5.0 +2023-04-01 02:24:35,708 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.358e+02 7.364e+02 9.517e+02 1.310e+03 2.254e+03, threshold=1.903e+03, percent-clipped=6.0 +2023-04-01 02:24:42,753 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 02:24:46,194 INFO [train.py:903] (0/4) Epoch 4, batch 1450, loss[loss=0.3118, simple_loss=0.3675, pruned_loss=0.1281, over 19690.00 frames. ], tot_loss[loss=0.3059, simple_loss=0.3589, pruned_loss=0.1264, over 3811577.21 frames. ], batch size: 60, lr: 2.05e-02, grad_scale: 8.0 +2023-04-01 02:25:13,206 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21957.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:25:34,255 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=21975.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:25:40,061 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7790, 1.6435, 1.5590, 1.9058, 3.3217, 1.3171, 2.1583, 3.5882], + device='cuda:0'), covar=tensor([0.0322, 0.2119, 0.2010, 0.1360, 0.0426, 0.2035, 0.1128, 0.0271], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0302, 0.0303, 0.0281, 0.0286, 0.0320, 0.0278, 0.0284], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:25:45,923 INFO [train.py:903] (0/4) Epoch 4, batch 1500, loss[loss=0.319, simple_loss=0.3714, pruned_loss=0.1333, over 19673.00 frames. ], tot_loss[loss=0.3069, simple_loss=0.3594, pruned_loss=0.1272, over 3818592.35 frames. ], batch size: 55, lr: 2.05e-02, grad_scale: 8.0 +2023-04-01 02:25:46,306 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0648, 2.7763, 1.8614, 2.1037, 1.8263, 2.1475, 0.7845, 2.1320], + device='cuda:0'), covar=tensor([0.0249, 0.0244, 0.0250, 0.0328, 0.0452, 0.0339, 0.0518, 0.0423], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0264, 0.0264, 0.0288, 0.0347, 0.0276, 0.0265, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 02:25:49,629 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=21988.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:25:50,920 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1697, 1.1374, 1.5704, 1.2874, 2.1774, 2.0975, 2.3312, 0.7726], + device='cuda:0'), covar=tensor([0.1644, 0.2774, 0.1430, 0.1411, 0.1039, 0.1279, 0.1084, 0.2436], + device='cuda:0'), in_proj_covar=tensor([0.0427, 0.0480, 0.0446, 0.0407, 0.0522, 0.0420, 0.0604, 0.0430], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 02:25:55,271 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.2130, 5.5774, 3.0143, 4.9905, 1.4357, 5.4154, 5.4116, 5.5288], + device='cuda:0'), covar=tensor([0.0340, 0.0825, 0.1681, 0.0451, 0.3608, 0.0598, 0.0560, 0.0578], + device='cuda:0'), in_proj_covar=tensor([0.0330, 0.0300, 0.0344, 0.0278, 0.0346, 0.0298, 0.0259, 0.0292], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 02:26:05,040 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-22000.pt +2023-04-01 02:26:20,386 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6991, 1.2899, 1.3808, 2.2332, 1.6375, 1.9355, 2.0198, 1.8931], + device='cuda:0'), covar=tensor([0.0866, 0.1222, 0.1339, 0.0948, 0.1081, 0.0870, 0.1004, 0.0765], + device='cuda:0'), in_proj_covar=tensor([0.0240, 0.0266, 0.0262, 0.0296, 0.0298, 0.0243, 0.0263, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 02:26:36,550 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.570e+02 6.805e+02 9.241e+02 1.170e+03 2.581e+03, threshold=1.848e+03, percent-clipped=2.0 +2023-04-01 02:26:47,124 INFO [train.py:903] (0/4) Epoch 4, batch 1550, loss[loss=0.2669, simple_loss=0.3354, pruned_loss=0.09919, over 19667.00 frames. ], tot_loss[loss=0.3059, simple_loss=0.3586, pruned_loss=0.1266, over 3829033.86 frames. ], batch size: 53, lr: 2.05e-02, grad_scale: 8.0 +2023-04-01 02:27:01,608 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22045.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:27:19,398 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22061.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:27:30,474 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22070.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:27:32,751 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22072.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:27:49,225 INFO [train.py:903] (0/4) Epoch 4, batch 1600, loss[loss=0.3109, simple_loss=0.3691, pruned_loss=0.1264, over 19779.00 frames. ], tot_loss[loss=0.307, simple_loss=0.3596, pruned_loss=0.1272, over 3835901.46 frames. ], batch size: 56, lr: 2.05e-02, grad_scale: 8.0 +2023-04-01 02:27:50,741 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22086.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:27:55,242 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22090.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:28:10,569 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 02:28:37,639 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22125.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:28:39,601 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.648e+02 8.565e+02 1.081e+03 1.346e+03 3.673e+03, threshold=2.162e+03, percent-clipped=6.0 +2023-04-01 02:28:48,843 INFO [train.py:903] (0/4) Epoch 4, batch 1650, loss[loss=0.2888, simple_loss=0.3491, pruned_loss=0.1143, over 19600.00 frames. ], tot_loss[loss=0.3067, simple_loss=0.3593, pruned_loss=0.1271, over 3835287.04 frames. ], batch size: 50, lr: 2.05e-02, grad_scale: 4.0 +2023-04-01 02:29:04,251 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22148.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:29:49,257 INFO [train.py:903] (0/4) Epoch 4, batch 1700, loss[loss=0.3194, simple_loss=0.3738, pruned_loss=0.1325, over 19603.00 frames. ], tot_loss[loss=0.3066, simple_loss=0.3596, pruned_loss=0.1268, over 3841634.57 frames. ], batch size: 61, lr: 2.04e-02, grad_scale: 4.0 +2023-04-01 02:30:00,399 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22194.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:30:23,329 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22212.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:30:27,870 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 02:30:40,332 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.920e+02 6.276e+02 7.753e+02 9.050e+02 1.909e+03, threshold=1.551e+03, percent-clipped=1.0 +2023-04-01 02:30:48,057 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22233.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:30:49,928 INFO [train.py:903] (0/4) Epoch 4, batch 1750, loss[loss=0.3165, simple_loss=0.3718, pruned_loss=0.1307, over 19569.00 frames. ], tot_loss[loss=0.3072, simple_loss=0.36, pruned_loss=0.1272, over 3833302.08 frames. ], batch size: 61, lr: 2.04e-02, grad_scale: 4.0 +2023-04-01 02:30:56,955 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.02 vs. limit=2.0 +2023-04-01 02:30:57,711 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22240.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:31:15,623 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22255.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:31:52,442 INFO [train.py:903] (0/4) Epoch 4, batch 1800, loss[loss=0.3398, simple_loss=0.3878, pruned_loss=0.1459, over 19619.00 frames. ], tot_loss[loss=0.3066, simple_loss=0.3597, pruned_loss=0.1268, over 3825258.49 frames. ], batch size: 57, lr: 2.04e-02, grad_scale: 4.0 +2023-04-01 02:32:43,154 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.733e+02 7.313e+02 8.961e+02 1.128e+03 3.443e+03, threshold=1.792e+03, percent-clipped=8.0 +2023-04-01 02:32:43,412 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22327.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:32:44,458 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22328.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:32:46,332 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 02:32:48,645 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22332.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:32:52,003 INFO [train.py:903] (0/4) Epoch 4, batch 1850, loss[loss=0.264, simple_loss=0.3264, pruned_loss=0.1008, over 19849.00 frames. ], tot_loss[loss=0.3074, simple_loss=0.3605, pruned_loss=0.1271, over 3829521.37 frames. ], batch size: 52, lr: 2.04e-02, grad_scale: 4.0 +2023-04-01 02:33:04,841 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22346.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:33:13,899 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22353.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:33:24,820 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 02:33:36,220 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22371.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:33:51,849 INFO [train.py:903] (0/4) Epoch 4, batch 1900, loss[loss=0.3326, simple_loss=0.3823, pruned_loss=0.1414, over 19747.00 frames. ], tot_loss[loss=0.3069, simple_loss=0.3602, pruned_loss=0.1267, over 3822415.05 frames. ], batch size: 51, lr: 2.03e-02, grad_scale: 4.0 +2023-04-01 02:34:09,721 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 02:34:14,789 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 02:34:15,042 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22403.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:34:39,429 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 02:34:42,867 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.044e+02 7.549e+02 9.520e+02 1.192e+03 3.384e+03, threshold=1.904e+03, percent-clipped=5.0 +2023-04-01 02:34:51,945 INFO [train.py:903] (0/4) Epoch 4, batch 1950, loss[loss=0.283, simple_loss=0.3462, pruned_loss=0.1099, over 19757.00 frames. ], tot_loss[loss=0.3053, simple_loss=0.359, pruned_loss=0.1258, over 3826945.02 frames. ], batch size: 54, lr: 2.03e-02, grad_scale: 4.0 +2023-04-01 02:35:08,637 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22447.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:35:53,487 INFO [train.py:903] (0/4) Epoch 4, batch 2000, loss[loss=0.3054, simple_loss=0.3615, pruned_loss=0.1246, over 19523.00 frames. ], tot_loss[loss=0.3069, simple_loss=0.3601, pruned_loss=0.1268, over 3804381.24 frames. ], batch size: 56, lr: 2.03e-02, grad_scale: 8.0 +2023-04-01 02:36:01,809 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22492.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:36:06,594 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22496.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:36:37,391 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22521.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:36:45,515 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.709e+02 6.735e+02 8.799e+02 1.102e+03 2.294e+03, threshold=1.760e+03, percent-clipped=1.0 +2023-04-01 02:36:46,753 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 02:36:47,091 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4185, 1.1697, 1.5416, 0.9574, 2.5720, 3.1510, 3.0729, 3.3689], + device='cuda:0'), covar=tensor([0.1184, 0.2765, 0.2551, 0.1892, 0.0375, 0.0118, 0.0205, 0.0106], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0280, 0.0323, 0.0260, 0.0192, 0.0110, 0.0201, 0.0124], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 02:36:54,493 INFO [train.py:903] (0/4) Epoch 4, batch 2050, loss[loss=0.3877, simple_loss=0.4286, pruned_loss=0.1734, over 19660.00 frames. ], tot_loss[loss=0.3068, simple_loss=0.3603, pruned_loss=0.1267, over 3795184.52 frames. ], batch size: 55, lr: 2.03e-02, grad_scale: 8.0 +2023-04-01 02:36:58,154 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22538.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:37:04,905 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 02:37:06,035 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 02:37:20,868 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2279, 1.2163, 1.4098, 1.5459, 2.7859, 1.0505, 2.0155, 2.9317], + device='cuda:0'), covar=tensor([0.0422, 0.2544, 0.2358, 0.1521, 0.0539, 0.2237, 0.1035, 0.0446], + device='cuda:0'), in_proj_covar=tensor([0.0268, 0.0308, 0.0304, 0.0278, 0.0292, 0.0320, 0.0283, 0.0289], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:37:27,448 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 02:37:45,369 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22577.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:37:52,351 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22583.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:37:54,201 INFO [train.py:903] (0/4) Epoch 4, batch 2100, loss[loss=0.2733, simple_loss=0.342, pruned_loss=0.1023, over 19715.00 frames. ], tot_loss[loss=0.306, simple_loss=0.3594, pruned_loss=0.1263, over 3809685.80 frames. ], batch size: 59, lr: 2.03e-02, grad_scale: 8.0 +2023-04-01 02:38:10,689 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22599.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:38:21,642 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 02:38:21,978 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22607.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:38:23,279 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22608.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:38:42,450 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 02:38:44,808 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.772e+02 6.954e+02 8.861e+02 1.126e+03 2.028e+03, threshold=1.772e+03, percent-clipped=3.0 +2023-04-01 02:38:53,760 INFO [train.py:903] (0/4) Epoch 4, batch 2150, loss[loss=0.3753, simple_loss=0.4017, pruned_loss=0.1745, over 13370.00 frames. ], tot_loss[loss=0.3062, simple_loss=0.3597, pruned_loss=0.1264, over 3796708.32 frames. ], batch size: 136, lr: 2.02e-02, grad_scale: 8.0 +2023-04-01 02:39:17,732 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22653.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:39:22,223 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22657.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:39:56,339 INFO [train.py:903] (0/4) Epoch 4, batch 2200, loss[loss=0.2792, simple_loss=0.3358, pruned_loss=0.1112, over 19742.00 frames. ], tot_loss[loss=0.3038, simple_loss=0.3578, pruned_loss=0.1249, over 3807163.23 frames. ], batch size: 51, lr: 2.02e-02, grad_scale: 8.0 +2023-04-01 02:40:05,445 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22692.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:40:08,931 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3434, 1.0591, 1.1670, 1.4177, 1.1545, 1.3611, 1.3838, 1.2470], + device='cuda:0'), covar=tensor([0.0852, 0.1190, 0.1154, 0.0764, 0.0874, 0.0819, 0.0828, 0.0733], + device='cuda:0'), in_proj_covar=tensor([0.0248, 0.0266, 0.0264, 0.0301, 0.0300, 0.0249, 0.0264, 0.0238], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 02:40:18,035 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22703.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:40:30,220 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22714.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:40:47,465 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.077e+02 6.866e+02 8.417e+02 1.098e+03 2.160e+03, threshold=1.683e+03, percent-clipped=4.0 +2023-04-01 02:40:49,078 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22728.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:40:56,709 INFO [train.py:903] (0/4) Epoch 4, batch 2250, loss[loss=0.3016, simple_loss=0.3664, pruned_loss=0.1184, over 19701.00 frames. ], tot_loss[loss=0.3049, simple_loss=0.3583, pruned_loss=0.1257, over 3803527.88 frames. ], batch size: 59, lr: 2.02e-02, grad_scale: 8.0 +2023-04-01 02:41:10,258 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=22747.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:41:50,575 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3831, 2.2741, 1.5290, 1.6845, 1.9936, 1.0211, 1.2224, 1.6816], + device='cuda:0'), covar=tensor([0.0771, 0.0441, 0.0965, 0.0534, 0.0370, 0.1095, 0.0659, 0.0413], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0245, 0.0321, 0.0245, 0.0221, 0.0308, 0.0284, 0.0250], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:41:57,020 INFO [train.py:903] (0/4) Epoch 4, batch 2300, loss[loss=0.3809, simple_loss=0.4156, pruned_loss=0.1731, over 19735.00 frames. ], tot_loss[loss=0.3042, simple_loss=0.3576, pruned_loss=0.1254, over 3807093.59 frames. ], batch size: 63, lr: 2.02e-02, grad_scale: 8.0 +2023-04-01 02:42:09,271 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 02:42:31,173 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22812.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:42:47,875 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.632e+02 7.223e+02 9.387e+02 1.200e+03 1.860e+03, threshold=1.877e+03, percent-clipped=8.0 +2023-04-01 02:42:56,332 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.89 vs. limit=5.0 +2023-04-01 02:42:56,885 INFO [train.py:903] (0/4) Epoch 4, batch 2350, loss[loss=0.311, simple_loss=0.3661, pruned_loss=0.1279, over 19569.00 frames. ], tot_loss[loss=0.3056, simple_loss=0.3591, pruned_loss=0.126, over 3810634.02 frames. ], batch size: 61, lr: 2.01e-02, grad_scale: 8.0 +2023-04-01 02:43:06,743 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=22843.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:43:31,035 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=22862.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:43:32,278 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22863.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:43:38,750 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 02:43:54,378 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 02:43:57,673 INFO [train.py:903] (0/4) Epoch 4, batch 2400, loss[loss=0.4557, simple_loss=0.4562, pruned_loss=0.2276, over 12795.00 frames. ], tot_loss[loss=0.305, simple_loss=0.3585, pruned_loss=0.1258, over 3806463.45 frames. ], batch size: 135, lr: 2.01e-02, grad_scale: 8.0 +2023-04-01 02:44:03,301 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22888.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:44:27,618 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22909.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:44:38,976 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0002, 5.3119, 2.6931, 4.6513, 1.3758, 5.2829, 5.2206, 5.3605], + device='cuda:0'), covar=tensor([0.0462, 0.1110, 0.2239, 0.0593, 0.4376, 0.0678, 0.0628, 0.0744], + device='cuda:0'), in_proj_covar=tensor([0.0338, 0.0297, 0.0353, 0.0275, 0.0350, 0.0296, 0.0259, 0.0298], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 02:44:49,611 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.703e+02 7.425e+02 9.466e+02 1.182e+03 3.064e+03, threshold=1.893e+03, percent-clipped=2.0 +2023-04-01 02:44:58,904 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22934.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:44:59,530 INFO [train.py:903] (0/4) Epoch 4, batch 2450, loss[loss=0.3638, simple_loss=0.3977, pruned_loss=0.1649, over 19655.00 frames. ], tot_loss[loss=0.3037, simple_loss=0.3573, pruned_loss=0.1251, over 3812642.06 frames. ], batch size: 53, lr: 2.01e-02, grad_scale: 8.0 +2023-04-01 02:45:14,042 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22948.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:45:40,891 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=22970.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:45:44,395 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22973.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:45:57,738 INFO [train.py:903] (0/4) Epoch 4, batch 2500, loss[loss=0.3713, simple_loss=0.4102, pruned_loss=0.1662, over 19675.00 frames. ], tot_loss[loss=0.3055, simple_loss=0.3583, pruned_loss=0.1264, over 3820946.56 frames. ], batch size: 58, lr: 2.01e-02, grad_scale: 8.0 +2023-04-01 02:46:09,304 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=22995.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:46:16,257 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23001.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:46:46,585 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9893, 1.4490, 1.3976, 2.0866, 1.6363, 2.1178, 2.2532, 1.9836], + device='cuda:0'), covar=tensor([0.0691, 0.0990, 0.1058, 0.0853, 0.0992, 0.0683, 0.0746, 0.0546], + device='cuda:0'), in_proj_covar=tensor([0.0242, 0.0262, 0.0258, 0.0292, 0.0296, 0.0243, 0.0256, 0.0234], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 02:46:46,610 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0394, 1.4605, 1.5239, 1.9867, 1.8188, 1.7421, 1.8473, 1.9000], + device='cuda:0'), covar=tensor([0.0699, 0.1568, 0.1130, 0.0809, 0.1047, 0.0473, 0.0783, 0.0559], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0365, 0.0280, 0.0247, 0.0313, 0.0256, 0.0270, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:46:48,541 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.107e+02 7.308e+02 8.619e+02 1.083e+03 2.930e+03, threshold=1.724e+03, percent-clipped=3.0 +2023-04-01 02:46:57,730 INFO [train.py:903] (0/4) Epoch 4, batch 2550, loss[loss=0.2929, simple_loss=0.3583, pruned_loss=0.1137, over 19789.00 frames. ], tot_loss[loss=0.3063, simple_loss=0.3596, pruned_loss=0.1265, over 3824909.65 frames. ], batch size: 56, lr: 2.01e-02, grad_scale: 8.0 +2023-04-01 02:47:18,940 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.72 vs. limit=2.0 +2023-04-01 02:47:20,074 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-04-01 02:47:43,520 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.73 vs. limit=2.0 +2023-04-01 02:47:49,704 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 02:47:57,382 INFO [train.py:903] (0/4) Epoch 4, batch 2600, loss[loss=0.2917, simple_loss=0.3406, pruned_loss=0.1213, over 19791.00 frames. ], tot_loss[loss=0.3064, simple_loss=0.3596, pruned_loss=0.1266, over 3829727.75 frames. ], batch size: 49, lr: 2.00e-02, grad_scale: 8.0 +2023-04-01 02:48:16,538 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.65 vs. limit=5.0 +2023-04-01 02:48:20,333 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5699, 1.0714, 1.2360, 1.7793, 1.3948, 1.6431, 1.9093, 1.5827], + device='cuda:0'), covar=tensor([0.0843, 0.1205, 0.1229, 0.1009, 0.1077, 0.0797, 0.0856, 0.0737], + device='cuda:0'), in_proj_covar=tensor([0.0243, 0.0265, 0.0259, 0.0294, 0.0300, 0.0244, 0.0262, 0.0237], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 02:48:29,108 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7036, 1.6948, 1.8151, 1.9831, 4.1965, 1.2429, 2.2570, 4.3147], + device='cuda:0'), covar=tensor([0.0238, 0.2251, 0.2367, 0.1465, 0.0441, 0.2230, 0.1248, 0.0289], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0304, 0.0302, 0.0276, 0.0290, 0.0315, 0.0279, 0.0288], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 02:48:33,813 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23116.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:48:36,188 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23118.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:48:42,577 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.04 vs. limit=5.0 +2023-04-01 02:48:46,631 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23126.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:48:48,301 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.619e+02 6.946e+02 8.555e+02 1.074e+03 2.756e+03, threshold=1.711e+03, percent-clipped=6.0 +2023-04-01 02:48:58,041 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-01 02:48:58,589 INFO [train.py:903] (0/4) Epoch 4, batch 2650, loss[loss=0.2745, simple_loss=0.3318, pruned_loss=0.1086, over 19386.00 frames. ], tot_loss[loss=0.3053, simple_loss=0.3589, pruned_loss=0.1259, over 3839529.30 frames. ], batch size: 48, lr: 2.00e-02, grad_scale: 8.0 +2023-04-01 02:49:08,302 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23143.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:49:17,081 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 02:49:21,041 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.95 vs. limit=5.0 +2023-04-01 02:49:22,933 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23156.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:49:43,849 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.86 vs. limit=5.0 +2023-04-01 02:49:58,047 INFO [train.py:903] (0/4) Epoch 4, batch 2700, loss[loss=0.2949, simple_loss=0.357, pruned_loss=0.1164, over 19683.00 frames. ], tot_loss[loss=0.3047, simple_loss=0.3586, pruned_loss=0.1254, over 3840806.07 frames. ], batch size: 59, lr: 2.00e-02, grad_scale: 8.0 +2023-04-01 02:50:00,479 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23187.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:50:48,019 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.628e+02 7.002e+02 8.947e+02 1.091e+03 2.361e+03, threshold=1.789e+03, percent-clipped=7.0 +2023-04-01 02:50:57,145 INFO [train.py:903] (0/4) Epoch 4, batch 2750, loss[loss=0.2727, simple_loss=0.3431, pruned_loss=0.1012, over 19793.00 frames. ], tot_loss[loss=0.3051, simple_loss=0.359, pruned_loss=0.1256, over 3838940.20 frames. ], batch size: 56, lr: 2.00e-02, grad_scale: 8.0 +2023-04-01 02:50:57,412 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23235.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:51:11,551 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23247.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:51:24,729 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2977, 1.1991, 1.8896, 1.3962, 2.3891, 2.2929, 2.6460, 0.9733], + device='cuda:0'), covar=tensor([0.1566, 0.2697, 0.1360, 0.1361, 0.1020, 0.1199, 0.1130, 0.2366], + device='cuda:0'), in_proj_covar=tensor([0.0427, 0.0481, 0.0448, 0.0397, 0.0521, 0.0424, 0.0602, 0.0429], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 02:51:33,819 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-04-01 02:51:39,835 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8830, 4.2802, 4.5873, 4.5300, 1.5092, 4.1630, 3.7780, 4.1883], + device='cuda:0'), covar=tensor([0.0687, 0.0468, 0.0408, 0.0326, 0.3578, 0.0256, 0.0376, 0.0891], + device='cuda:0'), in_proj_covar=tensor([0.0435, 0.0391, 0.0525, 0.0411, 0.0531, 0.0298, 0.0344, 0.0498], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 02:51:41,053 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23271.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:51:56,446 INFO [train.py:903] (0/4) Epoch 4, batch 2800, loss[loss=0.249, simple_loss=0.3078, pruned_loss=0.09514, over 19284.00 frames. ], tot_loss[loss=0.3033, simple_loss=0.3575, pruned_loss=0.1245, over 3838758.42 frames. ], batch size: 44, lr: 2.00e-02, grad_scale: 8.0 +2023-04-01 02:52:17,025 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23302.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:52:45,199 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.435e+02 7.909e+02 1.044e+03 1.347e+03 2.323e+03, threshold=2.087e+03, percent-clipped=7.0 +2023-04-01 02:52:56,802 INFO [train.py:903] (0/4) Epoch 4, batch 2850, loss[loss=0.3148, simple_loss=0.3552, pruned_loss=0.1372, over 19298.00 frames. ], tot_loss[loss=0.3039, simple_loss=0.3578, pruned_loss=0.125, over 3834127.51 frames. ], batch size: 44, lr: 1.99e-02, grad_scale: 8.0 +2023-04-01 02:53:41,821 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23372.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:53:42,859 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4640, 3.8580, 4.0375, 4.0077, 1.4282, 3.6957, 3.1949, 3.5738], + device='cuda:0'), covar=tensor([0.0815, 0.0564, 0.0554, 0.0401, 0.3718, 0.0361, 0.0519, 0.1164], + device='cuda:0'), in_proj_covar=tensor([0.0441, 0.0395, 0.0530, 0.0415, 0.0529, 0.0297, 0.0349, 0.0506], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 02:53:56,271 INFO [train.py:903] (0/4) Epoch 4, batch 2900, loss[loss=0.2593, simple_loss=0.3243, pruned_loss=0.09722, over 19830.00 frames. ], tot_loss[loss=0.3025, simple_loss=0.3567, pruned_loss=0.1242, over 3835338.63 frames. ], batch size: 49, lr: 1.99e-02, grad_scale: 4.0 +2023-04-01 02:53:56,292 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 02:54:10,032 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23397.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:54:45,629 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.212e+02 7.852e+02 1.023e+03 1.284e+03 2.319e+03, threshold=2.047e+03, percent-clipped=2.0 +2023-04-01 02:54:53,634 INFO [train.py:903] (0/4) Epoch 4, batch 2950, loss[loss=0.2823, simple_loss=0.3425, pruned_loss=0.111, over 19758.00 frames. ], tot_loss[loss=0.3042, simple_loss=0.3581, pruned_loss=0.1252, over 3832371.70 frames. ], batch size: 54, lr: 1.99e-02, grad_scale: 4.0 +2023-04-01 02:55:35,105 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23470.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:55:52,223 INFO [train.py:903] (0/4) Epoch 4, batch 3000, loss[loss=0.2851, simple_loss=0.3496, pruned_loss=0.1103, over 19590.00 frames. ], tot_loss[loss=0.3023, simple_loss=0.3567, pruned_loss=0.1239, over 3830742.31 frames. ], batch size: 57, lr: 1.99e-02, grad_scale: 4.0 +2023-04-01 02:55:52,224 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 02:56:05,135 INFO [train.py:937] (0/4) Epoch 4, validation: loss=0.2145, simple_loss=0.3118, pruned_loss=0.05862, over 944034.00 frames. +2023-04-01 02:56:05,136 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 02:56:09,790 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 02:56:32,363 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=23506.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:56:56,208 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23527.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:56:56,886 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.281e+02 6.234e+02 7.977e+02 1.046e+03 2.333e+03, threshold=1.595e+03, percent-clipped=2.0 +2023-04-01 02:57:05,046 INFO [train.py:903] (0/4) Epoch 4, batch 3050, loss[loss=0.2401, simple_loss=0.3042, pruned_loss=0.08794, over 19757.00 frames. ], tot_loss[loss=0.3036, simple_loss=0.3582, pruned_loss=0.1245, over 3826443.00 frames. ], batch size: 46, lr: 1.99e-02, grad_scale: 4.0 +2023-04-01 02:57:26,963 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23552.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:57:28,443 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-01 02:57:33,673 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23558.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:57:53,409 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.01 vs. limit=5.0 +2023-04-01 02:57:58,283 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23579.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:58:04,138 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23583.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:58:06,132 INFO [train.py:903] (0/4) Epoch 4, batch 3100, loss[loss=0.2349, simple_loss=0.3002, pruned_loss=0.08484, over 16407.00 frames. ], tot_loss[loss=0.3016, simple_loss=0.3566, pruned_loss=0.1233, over 3815242.64 frames. ], batch size: 36, lr: 1.98e-02, grad_scale: 4.0 +2023-04-01 02:58:06,485 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23585.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:58:13,277 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23591.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 02:58:55,942 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.118e+02 6.915e+02 8.546e+02 1.092e+03 2.878e+03, threshold=1.709e+03, percent-clipped=7.0 +2023-04-01 02:59:03,925 INFO [train.py:903] (0/4) Epoch 4, batch 3150, loss[loss=0.3385, simple_loss=0.4006, pruned_loss=0.1382, over 19371.00 frames. ], tot_loss[loss=0.3033, simple_loss=0.3577, pruned_loss=0.1245, over 3801547.81 frames. ], batch size: 70, lr: 1.98e-02, grad_scale: 4.0 +2023-04-01 02:59:28,011 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 03:00:02,537 INFO [train.py:903] (0/4) Epoch 4, batch 3200, loss[loss=0.2542, simple_loss=0.3149, pruned_loss=0.09678, over 19405.00 frames. ], tot_loss[loss=0.3046, simple_loss=0.3587, pruned_loss=0.1252, over 3817771.11 frames. ], batch size: 48, lr: 1.98e-02, grad_scale: 8.0 +2023-04-01 03:00:11,296 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-01 03:00:13,184 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23694.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:00:29,404 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23706.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:00:53,668 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.799e+02 7.379e+02 9.197e+02 1.143e+03 1.957e+03, threshold=1.839e+03, percent-clipped=5.0 +2023-04-01 03:01:02,105 INFO [train.py:903] (0/4) Epoch 4, batch 3250, loss[loss=0.3393, simple_loss=0.3826, pruned_loss=0.148, over 17668.00 frames. ], tot_loss[loss=0.3057, simple_loss=0.3593, pruned_loss=0.126, over 3803560.20 frames. ], batch size: 101, lr: 1.98e-02, grad_scale: 8.0 +2023-04-01 03:02:01,901 INFO [train.py:903] (0/4) Epoch 4, batch 3300, loss[loss=0.3276, simple_loss=0.3747, pruned_loss=0.1402, over 19663.00 frames. ], tot_loss[loss=0.3039, simple_loss=0.3575, pruned_loss=0.1251, over 3814808.89 frames. ], batch size: 58, lr: 1.98e-02, grad_scale: 8.0 +2023-04-01 03:02:04,049 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 03:02:54,104 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.057e+02 7.772e+02 9.614e+02 1.210e+03 2.492e+03, threshold=1.923e+03, percent-clipped=5.0 +2023-04-01 03:03:02,120 INFO [train.py:903] (0/4) Epoch 4, batch 3350, loss[loss=0.2692, simple_loss=0.3289, pruned_loss=0.1048, over 19628.00 frames. ], tot_loss[loss=0.3028, simple_loss=0.357, pruned_loss=0.1243, over 3811970.61 frames. ], batch size: 50, lr: 1.97e-02, grad_scale: 8.0 +2023-04-01 03:03:09,291 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23841.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:03:19,200 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=23850.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:03:40,431 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23866.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:03:41,830 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-01 03:04:01,843 INFO [train.py:903] (0/4) Epoch 4, batch 3400, loss[loss=0.3049, simple_loss=0.3564, pruned_loss=0.1267, over 19579.00 frames. ], tot_loss[loss=0.2996, simple_loss=0.3552, pruned_loss=0.1219, over 3818485.13 frames. ], batch size: 52, lr: 1.97e-02, grad_scale: 8.0 +2023-04-01 03:04:42,896 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8304, 1.8684, 2.2066, 2.9748, 2.1741, 2.7937, 2.2382, 2.7349], + device='cuda:0'), covar=tensor([0.0634, 0.1705, 0.1075, 0.0721, 0.1141, 0.0347, 0.0845, 0.0501], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0373, 0.0280, 0.0245, 0.0312, 0.0259, 0.0271, 0.0237], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 03:04:53,666 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.081e+02 7.350e+02 9.318e+02 1.202e+03 2.145e+03, threshold=1.864e+03, percent-clipped=3.0 +2023-04-01 03:05:01,728 INFO [train.py:903] (0/4) Epoch 4, batch 3450, loss[loss=0.3137, simple_loss=0.3703, pruned_loss=0.1286, over 19329.00 frames. ], tot_loss[loss=0.3006, simple_loss=0.356, pruned_loss=0.1226, over 3814321.49 frames. ], batch size: 70, lr: 1.97e-02, grad_scale: 8.0 +2023-04-01 03:05:01,755 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 03:05:22,668 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23950.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:05:36,148 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=23962.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:05:39,417 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=23965.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:05:50,600 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23975.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:06:04,982 INFO [train.py:903] (0/4) Epoch 4, batch 3500, loss[loss=0.3126, simple_loss=0.363, pruned_loss=0.131, over 18805.00 frames. ], tot_loss[loss=0.3027, simple_loss=0.3574, pruned_loss=0.124, over 3815799.85 frames. ], batch size: 74, lr: 1.97e-02, grad_scale: 8.0 +2023-04-01 03:06:07,779 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=23987.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:06:22,278 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-24000.pt +2023-04-01 03:06:58,145 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.373e+02 7.196e+02 8.612e+02 1.120e+03 2.630e+03, threshold=1.722e+03, percent-clipped=5.0 +2023-04-01 03:07:06,277 INFO [train.py:903] (0/4) Epoch 4, batch 3550, loss[loss=0.3369, simple_loss=0.3859, pruned_loss=0.144, over 19572.00 frames. ], tot_loss[loss=0.3014, simple_loss=0.3563, pruned_loss=0.1233, over 3829407.34 frames. ], batch size: 61, lr: 1.97e-02, grad_scale: 8.0 +2023-04-01 03:07:28,425 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24055.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:08:05,265 INFO [train.py:903] (0/4) Epoch 4, batch 3600, loss[loss=0.3536, simple_loss=0.4094, pruned_loss=0.1489, over 19732.00 frames. ], tot_loss[loss=0.3037, simple_loss=0.3576, pruned_loss=0.1249, over 3828935.01 frames. ], batch size: 63, lr: 1.96e-02, grad_scale: 8.0 +2023-04-01 03:08:56,915 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.624e+02 7.148e+02 8.733e+02 1.077e+03 2.339e+03, threshold=1.747e+03, percent-clipped=4.0 +2023-04-01 03:09:04,849 INFO [train.py:903] (0/4) Epoch 4, batch 3650, loss[loss=0.337, simple_loss=0.384, pruned_loss=0.145, over 18748.00 frames. ], tot_loss[loss=0.3026, simple_loss=0.3567, pruned_loss=0.1242, over 3837657.12 frames. ], batch size: 74, lr: 1.96e-02, grad_scale: 8.0 +2023-04-01 03:10:05,500 INFO [train.py:903] (0/4) Epoch 4, batch 3700, loss[loss=0.2673, simple_loss=0.333, pruned_loss=0.1008, over 19686.00 frames. ], tot_loss[loss=0.3006, simple_loss=0.3548, pruned_loss=0.1231, over 3831429.24 frames. ], batch size: 53, lr: 1.96e-02, grad_scale: 4.0 +2023-04-01 03:10:48,446 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24221.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:10:58,949 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.206e+02 7.207e+02 9.640e+02 1.134e+03 2.323e+03, threshold=1.928e+03, percent-clipped=6.0 +2023-04-01 03:11:05,386 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6108, 1.6106, 1.7715, 2.3868, 4.1035, 1.3995, 2.1459, 4.2157], + device='cuda:0'), covar=tensor([0.0307, 0.2455, 0.2311, 0.1252, 0.0442, 0.2169, 0.1348, 0.0278], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0308, 0.0300, 0.0278, 0.0293, 0.0314, 0.0282, 0.0288], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 03:11:07,279 INFO [train.py:903] (0/4) Epoch 4, batch 3750, loss[loss=0.3058, simple_loss=0.3704, pruned_loss=0.1206, over 19562.00 frames. ], tot_loss[loss=0.3005, simple_loss=0.3551, pruned_loss=0.123, over 3827834.00 frames. ], batch size: 61, lr: 1.96e-02, grad_scale: 4.0 +2023-04-01 03:11:20,323 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24246.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:11:20,612 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-04-01 03:12:07,328 INFO [train.py:903] (0/4) Epoch 4, batch 3800, loss[loss=0.2695, simple_loss=0.3206, pruned_loss=0.1092, over 19742.00 frames. ], tot_loss[loss=0.2994, simple_loss=0.3541, pruned_loss=0.1223, over 3826624.73 frames. ], batch size: 45, lr: 1.96e-02, grad_scale: 4.0 +2023-04-01 03:12:38,483 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 03:13:00,196 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.239e+02 7.507e+02 9.076e+02 1.248e+03 3.254e+03, threshold=1.815e+03, percent-clipped=3.0 +2023-04-01 03:13:07,202 INFO [train.py:903] (0/4) Epoch 4, batch 3850, loss[loss=0.2187, simple_loss=0.2845, pruned_loss=0.07643, over 19738.00 frames. ], tot_loss[loss=0.2991, simple_loss=0.3535, pruned_loss=0.1223, over 3835676.77 frames. ], batch size: 46, lr: 1.95e-02, grad_scale: 4.0 +2023-04-01 03:13:12,285 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-04-01 03:14:06,613 INFO [train.py:903] (0/4) Epoch 4, batch 3900, loss[loss=0.3028, simple_loss=0.3679, pruned_loss=0.1189, over 19507.00 frames. ], tot_loss[loss=0.2996, simple_loss=0.3543, pruned_loss=0.1225, over 3829780.51 frames. ], batch size: 64, lr: 1.95e-02, grad_scale: 4.0 +2023-04-01 03:14:25,899 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=24399.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:14:28,371 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1344, 1.2295, 1.0171, 0.9448, 0.8915, 1.0973, 0.0809, 0.4408], + device='cuda:0'), covar=tensor([0.0271, 0.0235, 0.0159, 0.0187, 0.0525, 0.0194, 0.0435, 0.0394], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0271, 0.0268, 0.0294, 0.0357, 0.0286, 0.0276, 0.0286], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 03:14:46,519 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1862, 2.0492, 1.5556, 1.4060, 1.9952, 1.0198, 1.0778, 1.6723], + device='cuda:0'), covar=tensor([0.0673, 0.0377, 0.0704, 0.0495, 0.0339, 0.0931, 0.0576, 0.0348], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0249, 0.0306, 0.0240, 0.0216, 0.0308, 0.0280, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 03:15:00,797 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.463e+02 8.238e+02 9.729e+02 1.230e+03 4.971e+03, threshold=1.946e+03, percent-clipped=9.0 +2023-04-01 03:15:09,421 INFO [train.py:903] (0/4) Epoch 4, batch 3950, loss[loss=0.3706, simple_loss=0.401, pruned_loss=0.1701, over 13081.00 frames. ], tot_loss[loss=0.2991, simple_loss=0.3543, pruned_loss=0.122, over 3820031.08 frames. ], batch size: 136, lr: 1.95e-02, grad_scale: 4.0 +2023-04-01 03:15:17,167 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 03:16:10,702 INFO [train.py:903] (0/4) Epoch 4, batch 4000, loss[loss=0.3239, simple_loss=0.3748, pruned_loss=0.1365, over 19298.00 frames. ], tot_loss[loss=0.2977, simple_loss=0.3534, pruned_loss=0.121, over 3823523.10 frames. ], batch size: 66, lr: 1.95e-02, grad_scale: 8.0 +2023-04-01 03:16:44,991 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=24514.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:16:58,863 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 03:17:03,421 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.288e+02 6.850e+02 8.525e+02 1.041e+03 2.187e+03, threshold=1.705e+03, percent-clipped=1.0 +2023-04-01 03:17:09,918 INFO [train.py:903] (0/4) Epoch 4, batch 4050, loss[loss=0.3466, simple_loss=0.3902, pruned_loss=0.1515, over 19682.00 frames. ], tot_loss[loss=0.3001, simple_loss=0.3554, pruned_loss=0.1224, over 3828205.45 frames. ], batch size: 58, lr: 1.95e-02, grad_scale: 8.0 +2023-04-01 03:17:47,517 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24565.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:18:10,053 INFO [train.py:903] (0/4) Epoch 4, batch 4100, loss[loss=0.2862, simple_loss=0.3483, pruned_loss=0.1121, over 19677.00 frames. ], tot_loss[loss=0.2988, simple_loss=0.3542, pruned_loss=0.1217, over 3831769.32 frames. ], batch size: 55, lr: 1.94e-02, grad_scale: 8.0 +2023-04-01 03:18:12,616 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24587.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:18:49,640 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 03:19:04,403 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.283e+02 6.813e+02 8.989e+02 1.047e+03 2.179e+03, threshold=1.798e+03, percent-clipped=2.0 +2023-04-01 03:19:12,691 INFO [train.py:903] (0/4) Epoch 4, batch 4150, loss[loss=0.2689, simple_loss=0.3426, pruned_loss=0.09759, over 19532.00 frames. ], tot_loss[loss=0.2986, simple_loss=0.3539, pruned_loss=0.1217, over 3831720.21 frames. ], batch size: 54, lr: 1.94e-02, grad_scale: 8.0 +2023-04-01 03:20:13,435 INFO [train.py:903] (0/4) Epoch 4, batch 4200, loss[loss=0.2321, simple_loss=0.2955, pruned_loss=0.08437, over 19717.00 frames. ], tot_loss[loss=0.2973, simple_loss=0.3536, pruned_loss=0.1205, over 3844835.35 frames. ], batch size: 45, lr: 1.94e-02, grad_scale: 8.0 +2023-04-01 03:20:19,920 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 03:21:05,887 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.764e+02 7.220e+02 8.850e+02 1.090e+03 2.101e+03, threshold=1.770e+03, percent-clipped=3.0 +2023-04-01 03:21:12,813 INFO [train.py:903] (0/4) Epoch 4, batch 4250, loss[loss=0.2893, simple_loss=0.3502, pruned_loss=0.1142, over 19795.00 frames. ], tot_loss[loss=0.2986, simple_loss=0.3543, pruned_loss=0.1214, over 3840376.80 frames. ], batch size: 56, lr: 1.94e-02, grad_scale: 8.0 +2023-04-01 03:21:29,776 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 03:21:41,558 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 03:21:56,470 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=24770.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:22:09,998 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2354, 2.1052, 1.7407, 1.6685, 1.4952, 1.7010, 0.4089, 1.0995], + device='cuda:0'), covar=tensor([0.0247, 0.0239, 0.0166, 0.0236, 0.0479, 0.0301, 0.0463, 0.0409], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0274, 0.0265, 0.0291, 0.0354, 0.0278, 0.0269, 0.0280], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 03:22:10,847 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24783.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:22:12,643 INFO [train.py:903] (0/4) Epoch 4, batch 4300, loss[loss=0.3185, simple_loss=0.3704, pruned_loss=0.1333, over 19139.00 frames. ], tot_loss[loss=0.2992, simple_loss=0.3549, pruned_loss=0.1218, over 3824646.93 frames. ], batch size: 69, lr: 1.94e-02, grad_scale: 8.0 +2023-04-01 03:22:26,914 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=24795.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:22:58,045 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-04-01 03:23:06,670 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.009e+02 7.093e+02 8.869e+02 1.163e+03 2.104e+03, threshold=1.774e+03, percent-clipped=1.0 +2023-04-01 03:23:08,973 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 03:23:14,538 INFO [train.py:903] (0/4) Epoch 4, batch 4350, loss[loss=0.3109, simple_loss=0.3624, pruned_loss=0.1297, over 19733.00 frames. ], tot_loss[loss=0.2991, simple_loss=0.3545, pruned_loss=0.1218, over 3831777.76 frames. ], batch size: 51, lr: 1.94e-02, grad_scale: 8.0 +2023-04-01 03:23:52,196 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=24866.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:24:05,092 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.2939, 3.8890, 2.5350, 3.6146, 1.0095, 3.5153, 3.3832, 3.6578], + device='cuda:0'), covar=tensor([0.0639, 0.1145, 0.1729, 0.0689, 0.3894, 0.0904, 0.0700, 0.0846], + device='cuda:0'), in_proj_covar=tensor([0.0343, 0.0302, 0.0356, 0.0280, 0.0356, 0.0306, 0.0266, 0.0302], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 03:24:15,073 INFO [train.py:903] (0/4) Epoch 4, batch 4400, loss[loss=0.3399, simple_loss=0.3886, pruned_loss=0.1456, over 17642.00 frames. ], tot_loss[loss=0.2981, simple_loss=0.3539, pruned_loss=0.1212, over 3827992.48 frames. ], batch size: 101, lr: 1.93e-02, grad_scale: 8.0 +2023-04-01 03:24:40,897 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 03:24:44,302 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=24909.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:24:50,759 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 03:25:09,651 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.331e+02 8.078e+02 9.889e+02 1.280e+03 3.768e+03, threshold=1.978e+03, percent-clipped=10.0 +2023-04-01 03:25:12,010 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=24931.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:25:16,497 INFO [train.py:903] (0/4) Epoch 4, batch 4450, loss[loss=0.3351, simple_loss=0.386, pruned_loss=0.1421, over 19494.00 frames. ], tot_loss[loss=0.2986, simple_loss=0.3543, pruned_loss=0.1214, over 3830332.44 frames. ], batch size: 64, lr: 1.93e-02, grad_scale: 8.0 +2023-04-01 03:26:17,090 INFO [train.py:903] (0/4) Epoch 4, batch 4500, loss[loss=0.3064, simple_loss=0.369, pruned_loss=0.1219, over 19662.00 frames. ], tot_loss[loss=0.2987, simple_loss=0.3546, pruned_loss=0.1214, over 3839765.16 frames. ], batch size: 60, lr: 1.93e-02, grad_scale: 8.0 +2023-04-01 03:26:52,422 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25013.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:27:04,622 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25024.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:27:11,028 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.968e+02 6.473e+02 7.865e+02 1.057e+03 2.211e+03, threshold=1.573e+03, percent-clipped=1.0 +2023-04-01 03:27:16,558 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9238, 0.8168, 0.7761, 1.0799, 0.8249, 0.9418, 1.0433, 0.8972], + device='cuda:0'), covar=tensor([0.0636, 0.0903, 0.0915, 0.0620, 0.0750, 0.0726, 0.0723, 0.0617], + device='cuda:0'), in_proj_covar=tensor([0.0243, 0.0269, 0.0255, 0.0293, 0.0293, 0.0249, 0.0256, 0.0241], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 03:27:18,546 INFO [train.py:903] (0/4) Epoch 4, batch 4550, loss[loss=0.2696, simple_loss=0.3366, pruned_loss=0.1013, over 19854.00 frames. ], tot_loss[loss=0.2971, simple_loss=0.3533, pruned_loss=0.1205, over 3820960.01 frames. ], batch size: 52, lr: 1.93e-02, grad_scale: 8.0 +2023-04-01 03:27:27,109 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 03:27:31,943 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25046.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:27:50,578 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 03:28:18,971 INFO [train.py:903] (0/4) Epoch 4, batch 4600, loss[loss=0.2973, simple_loss=0.3595, pruned_loss=0.1176, over 19687.00 frames. ], tot_loss[loss=0.3001, simple_loss=0.3556, pruned_loss=0.1223, over 3830800.19 frames. ], batch size: 59, lr: 1.93e-02, grad_scale: 8.0 +2023-04-01 03:29:10,640 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=25127.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:29:12,893 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.682e+02 7.418e+02 9.211e+02 1.176e+03 2.853e+03, threshold=1.842e+03, percent-clipped=7.0 +2023-04-01 03:29:20,333 INFO [train.py:903] (0/4) Epoch 4, batch 4650, loss[loss=0.3381, simple_loss=0.3559, pruned_loss=0.1602, over 19769.00 frames. ], tot_loss[loss=0.3, simple_loss=0.3555, pruned_loss=0.1222, over 3820687.13 frames. ], batch size: 47, lr: 1.92e-02, grad_scale: 8.0 +2023-04-01 03:29:37,195 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 03:29:37,785 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 03:29:46,695 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 03:30:19,351 INFO [train.py:903] (0/4) Epoch 4, batch 4700, loss[loss=0.2698, simple_loss=0.3293, pruned_loss=0.1051, over 19411.00 frames. ], tot_loss[loss=0.302, simple_loss=0.3564, pruned_loss=0.1238, over 3818511.98 frames. ], batch size: 48, lr: 1.92e-02, grad_scale: 8.0 +2023-04-01 03:30:42,836 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 03:30:50,627 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=25210.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:31:13,778 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.903e+02 7.587e+02 9.394e+02 1.259e+03 3.233e+03, threshold=1.879e+03, percent-clipped=11.0 +2023-04-01 03:31:21,429 INFO [train.py:903] (0/4) Epoch 4, batch 4750, loss[loss=0.3316, simple_loss=0.3908, pruned_loss=0.1362, over 19684.00 frames. ], tot_loss[loss=0.3023, simple_loss=0.3574, pruned_loss=0.1236, over 3825177.37 frames. ], batch size: 60, lr: 1.92e-02, grad_scale: 8.0 +2023-04-01 03:31:30,697 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25242.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:32:15,372 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25280.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:32:21,436 INFO [train.py:903] (0/4) Epoch 4, batch 4800, loss[loss=0.2885, simple_loss=0.338, pruned_loss=0.1195, over 19580.00 frames. ], tot_loss[loss=0.3004, simple_loss=0.3556, pruned_loss=0.1226, over 3838612.22 frames. ], batch size: 52, lr: 1.92e-02, grad_scale: 8.0 +2023-04-01 03:32:41,259 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25302.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:32:44,297 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=25305.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:33:09,536 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25325.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:33:11,901 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=25327.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:33:13,790 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.433e+02 7.095e+02 8.715e+02 1.261e+03 2.828e+03, threshold=1.743e+03, percent-clipped=4.0 +2023-04-01 03:33:21,568 INFO [train.py:903] (0/4) Epoch 4, batch 4850, loss[loss=0.3444, simple_loss=0.3956, pruned_loss=0.1466, over 19663.00 frames. ], tot_loss[loss=0.3018, simple_loss=0.357, pruned_loss=0.1233, over 3821961.51 frames. ], batch size: 58, lr: 1.92e-02, grad_scale: 8.0 +2023-04-01 03:33:22,957 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8355, 1.8217, 2.2125, 2.8752, 1.8887, 2.7791, 2.8568, 2.5760], + device='cuda:0'), covar=tensor([0.0682, 0.1165, 0.1016, 0.1091, 0.1322, 0.0770, 0.0927, 0.0727], + device='cuda:0'), in_proj_covar=tensor([0.0244, 0.0263, 0.0253, 0.0293, 0.0289, 0.0247, 0.0255, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 03:33:45,829 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 03:33:48,272 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=25357.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:34:04,841 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 03:34:11,320 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 03:34:12,497 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 03:34:21,580 INFO [train.py:903] (0/4) Epoch 4, batch 4900, loss[loss=0.2573, simple_loss=0.3107, pruned_loss=0.102, over 19807.00 frames. ], tot_loss[loss=0.3003, simple_loss=0.3556, pruned_loss=0.1225, over 3825889.90 frames. ], batch size: 48, lr: 1.92e-02, grad_scale: 8.0 +2023-04-01 03:34:21,593 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 03:34:41,692 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 03:35:16,120 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.171e+02 7.094e+02 8.660e+02 1.069e+03 1.655e+03, threshold=1.732e+03, percent-clipped=0.0 +2023-04-01 03:35:23,612 INFO [train.py:903] (0/4) Epoch 4, batch 4950, loss[loss=0.2975, simple_loss=0.3642, pruned_loss=0.1154, over 19369.00 frames. ], tot_loss[loss=0.3002, simple_loss=0.3556, pruned_loss=0.1224, over 3823772.25 frames. ], batch size: 66, lr: 1.91e-02, grad_scale: 8.0 +2023-04-01 03:35:37,177 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25446.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 03:35:41,273 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 03:36:04,439 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 03:36:08,196 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25472.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:36:24,198 INFO [train.py:903] (0/4) Epoch 4, batch 5000, loss[loss=0.2931, simple_loss=0.353, pruned_loss=0.1166, over 19670.00 frames. ], tot_loss[loss=0.3001, simple_loss=0.3555, pruned_loss=0.1223, over 3825799.46 frames. ], batch size: 55, lr: 1.91e-02, grad_scale: 8.0 +2023-04-01 03:36:33,155 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 03:36:40,153 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25498.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:36:44,426 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 03:36:44,672 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8036, 1.2298, 1.4299, 1.6420, 2.4689, 1.2919, 1.7398, 2.4122], + device='cuda:0'), covar=tensor([0.0467, 0.2387, 0.2211, 0.1332, 0.0584, 0.1868, 0.1338, 0.0565], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0306, 0.0306, 0.0280, 0.0297, 0.0316, 0.0283, 0.0290], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 03:36:50,730 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-04-01 03:37:10,957 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=25523.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:37:17,567 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.118e+02 6.823e+02 8.824e+02 1.078e+03 2.588e+03, threshold=1.765e+03, percent-clipped=9.0 +2023-04-01 03:37:24,345 INFO [train.py:903] (0/4) Epoch 4, batch 5050, loss[loss=0.3066, simple_loss=0.3528, pruned_loss=0.1302, over 19816.00 frames. ], tot_loss[loss=0.2998, simple_loss=0.3554, pruned_loss=0.1221, over 3827848.07 frames. ], batch size: 49, lr: 1.91e-02, grad_scale: 8.0 +2023-04-01 03:38:02,129 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 03:38:21,779 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25581.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:38:25,842 INFO [train.py:903] (0/4) Epoch 4, batch 5100, loss[loss=0.404, simple_loss=0.43, pruned_loss=0.189, over 13835.00 frames. ], tot_loss[loss=0.2983, simple_loss=0.3547, pruned_loss=0.1209, over 3827156.78 frames. ], batch size: 135, lr: 1.91e-02, grad_scale: 8.0 +2023-04-01 03:38:33,129 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.85 vs. limit=5.0 +2023-04-01 03:38:36,820 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 03:38:40,932 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 03:38:44,399 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 03:38:52,400 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=25606.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:38:55,771 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3508, 1.1924, 1.4616, 1.2485, 2.6503, 3.4585, 3.4160, 3.7045], + device='cuda:0'), covar=tensor([0.1378, 0.2757, 0.2790, 0.1866, 0.0429, 0.0124, 0.0201, 0.0104], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0278, 0.0317, 0.0256, 0.0194, 0.0113, 0.0202, 0.0127], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 03:39:01,138 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7621, 1.4835, 1.5995, 1.7468, 3.2366, 1.0886, 1.9970, 3.3484], + device='cuda:0'), covar=tensor([0.0296, 0.2290, 0.2218, 0.1422, 0.0485, 0.2300, 0.1281, 0.0383], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0305, 0.0303, 0.0278, 0.0299, 0.0313, 0.0281, 0.0289], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 03:39:19,438 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.903e+02 6.354e+02 8.688e+02 1.168e+03 2.387e+03, threshold=1.738e+03, percent-clipped=4.0 +2023-04-01 03:39:27,222 INFO [train.py:903] (0/4) Epoch 4, batch 5150, loss[loss=0.2788, simple_loss=0.3459, pruned_loss=0.1058, over 18752.00 frames. ], tot_loss[loss=0.2991, simple_loss=0.3557, pruned_loss=0.1213, over 3824835.92 frames. ], batch size: 74, lr: 1.91e-02, grad_scale: 8.0 +2023-04-01 03:39:39,295 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 03:40:12,881 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 03:40:28,104 INFO [train.py:903] (0/4) Epoch 4, batch 5200, loss[loss=0.293, simple_loss=0.3493, pruned_loss=0.1184, over 19598.00 frames. ], tot_loss[loss=0.3001, simple_loss=0.356, pruned_loss=0.1221, over 3829019.86 frames. ], batch size: 52, lr: 1.90e-02, grad_scale: 8.0 +2023-04-01 03:40:42,782 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 03:41:21,034 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=25728.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:41:21,774 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.810e+02 7.304e+02 9.145e+02 1.165e+03 2.884e+03, threshold=1.829e+03, percent-clipped=6.0 +2023-04-01 03:41:25,406 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 03:41:28,709 INFO [train.py:903] (0/4) Epoch 4, batch 5250, loss[loss=0.2993, simple_loss=0.3678, pruned_loss=0.1154, over 19542.00 frames. ], tot_loss[loss=0.2986, simple_loss=0.3552, pruned_loss=0.121, over 3823943.11 frames. ], batch size: 54, lr: 1.90e-02, grad_scale: 8.0 +2023-04-01 03:41:51,406 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=25753.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:42:30,030 INFO [train.py:903] (0/4) Epoch 4, batch 5300, loss[loss=0.2442, simple_loss=0.3064, pruned_loss=0.09095, over 19745.00 frames. ], tot_loss[loss=0.298, simple_loss=0.3547, pruned_loss=0.1206, over 3827317.84 frames. ], batch size: 45, lr: 1.90e-02, grad_scale: 8.0 +2023-04-01 03:42:35,917 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=25790.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 03:42:49,035 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 03:43:11,855 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-04-01 03:43:23,319 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.781e+02 7.857e+02 9.756e+02 1.201e+03 3.803e+03, threshold=1.951e+03, percent-clipped=8.0 +2023-04-01 03:43:31,863 INFO [train.py:903] (0/4) Epoch 4, batch 5350, loss[loss=0.275, simple_loss=0.3213, pruned_loss=0.1144, over 19024.00 frames. ], tot_loss[loss=0.2983, simple_loss=0.3543, pruned_loss=0.1211, over 3839357.51 frames. ], batch size: 42, lr: 1.90e-02, grad_scale: 8.0 +2023-04-01 03:44:04,171 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 03:44:21,907 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=25876.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 03:44:32,189 INFO [train.py:903] (0/4) Epoch 4, batch 5400, loss[loss=0.3103, simple_loss=0.3658, pruned_loss=0.1274, over 19529.00 frames. ], tot_loss[loss=0.2998, simple_loss=0.3553, pruned_loss=0.1221, over 3826137.61 frames. ], batch size: 54, lr: 1.90e-02, grad_scale: 8.0 +2023-04-01 03:44:50,041 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5759, 1.3078, 1.2681, 1.8895, 1.5293, 1.8802, 2.0784, 1.6835], + device='cuda:0'), covar=tensor([0.0890, 0.1059, 0.1179, 0.1056, 0.1128, 0.0746, 0.0967, 0.0731], + device='cuda:0'), in_proj_covar=tensor([0.0243, 0.0260, 0.0251, 0.0284, 0.0287, 0.0242, 0.0250, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 03:44:56,707 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=25905.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 03:45:13,667 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 03:45:22,512 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8523, 1.9339, 1.8616, 2.7736, 1.8946, 2.5038, 2.4745, 1.6785], + device='cuda:0'), covar=tensor([0.1783, 0.1355, 0.0822, 0.0784, 0.1552, 0.0538, 0.1426, 0.1409], + device='cuda:0'), in_proj_covar=tensor([0.0574, 0.0553, 0.0520, 0.0716, 0.0623, 0.0476, 0.0624, 0.0535], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 03:45:26,555 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.180e+02 6.939e+02 8.636e+02 1.056e+03 2.577e+03, threshold=1.727e+03, percent-clipped=2.0 +2023-04-01 03:45:33,333 INFO [train.py:903] (0/4) Epoch 4, batch 5450, loss[loss=0.3229, simple_loss=0.3739, pruned_loss=0.136, over 19723.00 frames. ], tot_loss[loss=0.2982, simple_loss=0.3542, pruned_loss=0.1211, over 3844527.22 frames. ], batch size: 63, lr: 1.90e-02, grad_scale: 8.0 +2023-04-01 03:46:32,923 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-04-01 03:46:34,627 INFO [train.py:903] (0/4) Epoch 4, batch 5500, loss[loss=0.2949, simple_loss=0.3628, pruned_loss=0.1135, over 19708.00 frames. ], tot_loss[loss=0.2976, simple_loss=0.3538, pruned_loss=0.1207, over 3832183.37 frames. ], batch size: 63, lr: 1.89e-02, grad_scale: 4.0 +2023-04-01 03:46:53,477 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-26000.pt +2023-04-01 03:46:58,090 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 03:47:31,758 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.407e+02 7.568e+02 9.013e+02 1.115e+03 1.816e+03, threshold=1.803e+03, percent-clipped=1.0 +2023-04-01 03:47:37,475 INFO [train.py:903] (0/4) Epoch 4, batch 5550, loss[loss=0.2516, simple_loss=0.3276, pruned_loss=0.08782, over 19775.00 frames. ], tot_loss[loss=0.2962, simple_loss=0.3528, pruned_loss=0.1197, over 3832404.31 frames. ], batch size: 54, lr: 1.89e-02, grad_scale: 4.0 +2023-04-01 03:47:45,444 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 03:47:46,216 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.51 vs. limit=5.0 +2023-04-01 03:48:33,399 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 03:48:38,944 INFO [train.py:903] (0/4) Epoch 4, batch 5600, loss[loss=0.3239, simple_loss=0.3742, pruned_loss=0.1369, over 17327.00 frames. ], tot_loss[loss=0.2977, simple_loss=0.3535, pruned_loss=0.121, over 3817834.60 frames. ], batch size: 101, lr: 1.89e-02, grad_scale: 8.0 +2023-04-01 03:48:42,550 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4138, 1.9880, 1.8411, 2.6436, 2.0748, 2.6034, 2.8556, 2.8249], + device='cuda:0'), covar=tensor([0.0758, 0.1033, 0.1045, 0.0958, 0.1007, 0.0713, 0.0920, 0.0572], + device='cuda:0'), in_proj_covar=tensor([0.0242, 0.0258, 0.0250, 0.0285, 0.0285, 0.0238, 0.0251, 0.0233], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 03:48:45,930 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5264, 1.5352, 2.1257, 2.4375, 2.2710, 2.3972, 2.0731, 2.3934], + device='cuda:0'), covar=tensor([0.0671, 0.1757, 0.1107, 0.0771, 0.1010, 0.0372, 0.0827, 0.0515], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0366, 0.0284, 0.0242, 0.0307, 0.0255, 0.0270, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 03:49:06,273 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-04-01 03:49:14,931 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4075, 1.2446, 1.0216, 1.2852, 1.1415, 1.2780, 1.0150, 1.2500], + device='cuda:0'), covar=tensor([0.0886, 0.1017, 0.1349, 0.0828, 0.0982, 0.0519, 0.0978, 0.0729], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0366, 0.0286, 0.0242, 0.0307, 0.0255, 0.0271, 0.0238], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 03:49:22,517 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26120.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:49:34,176 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.637e+02 7.669e+02 9.359e+02 1.114e+03 3.409e+03, threshold=1.872e+03, percent-clipped=3.0 +2023-04-01 03:49:36,813 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7806, 4.2230, 4.3510, 4.3203, 1.5727, 3.9196, 3.5719, 3.9974], + device='cuda:0'), covar=tensor([0.0821, 0.0518, 0.0474, 0.0378, 0.3916, 0.0333, 0.0473, 0.0913], + device='cuda:0'), in_proj_covar=tensor([0.0460, 0.0408, 0.0539, 0.0438, 0.0537, 0.0315, 0.0357, 0.0510], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 03:49:40,064 INFO [train.py:903] (0/4) Epoch 4, batch 5650, loss[loss=0.2838, simple_loss=0.3358, pruned_loss=0.1159, over 19620.00 frames. ], tot_loss[loss=0.2968, simple_loss=0.3526, pruned_loss=0.1205, over 3817420.65 frames. ], batch size: 50, lr: 1.89e-02, grad_scale: 8.0 +2023-04-01 03:50:12,597 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26161.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 03:50:25,371 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 03:50:38,348 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4706, 1.5125, 1.4978, 2.0919, 1.4878, 1.7406, 1.8406, 1.3681], + device='cuda:0'), covar=tensor([0.1430, 0.0988, 0.0662, 0.0504, 0.1089, 0.0477, 0.1197, 0.1118], + device='cuda:0'), in_proj_covar=tensor([0.0577, 0.0556, 0.0517, 0.0711, 0.0622, 0.0475, 0.0629, 0.0536], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 03:50:40,061 INFO [train.py:903] (0/4) Epoch 4, batch 5700, loss[loss=0.3673, simple_loss=0.4038, pruned_loss=0.1654, over 19659.00 frames. ], tot_loss[loss=0.2972, simple_loss=0.3528, pruned_loss=0.1208, over 3817185.92 frames. ], batch size: 58, lr: 1.89e-02, grad_scale: 8.0 +2023-04-01 03:50:42,167 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26186.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:50:42,270 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26186.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 03:51:12,083 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.64 vs. limit=5.0 +2023-04-01 03:51:23,017 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26220.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 03:51:35,339 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.235e+02 7.760e+02 9.507e+02 1.157e+03 2.773e+03, threshold=1.901e+03, percent-clipped=5.0 +2023-04-01 03:51:39,939 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 03:51:40,944 INFO [train.py:903] (0/4) Epoch 4, batch 5750, loss[loss=0.3454, simple_loss=0.3941, pruned_loss=0.1484, over 17591.00 frames. ], tot_loss[loss=0.2979, simple_loss=0.3533, pruned_loss=0.1212, over 3820276.58 frames. ], batch size: 101, lr: 1.89e-02, grad_scale: 8.0 +2023-04-01 03:51:48,679 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 03:51:52,997 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 03:52:02,105 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-01 03:52:20,371 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26268.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:52:40,302 INFO [train.py:903] (0/4) Epoch 4, batch 5800, loss[loss=0.3612, simple_loss=0.3958, pruned_loss=0.1633, over 13247.00 frames. ], tot_loss[loss=0.2993, simple_loss=0.3541, pruned_loss=0.1223, over 3833241.76 frames. ], batch size: 137, lr: 1.88e-02, grad_scale: 8.0 +2023-04-01 03:52:44,287 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-01 03:53:36,644 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.499e+02 6.787e+02 8.542e+02 1.114e+03 2.576e+03, threshold=1.708e+03, percent-clipped=4.0 +2023-04-01 03:53:41,254 INFO [train.py:903] (0/4) Epoch 4, batch 5850, loss[loss=0.303, simple_loss=0.3588, pruned_loss=0.1236, over 19766.00 frames. ], tot_loss[loss=0.2981, simple_loss=0.3534, pruned_loss=0.1214, over 3830760.55 frames. ], batch size: 54, lr: 1.88e-02, grad_scale: 4.0 +2023-04-01 03:53:41,584 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26335.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 03:54:40,799 INFO [train.py:903] (0/4) Epoch 4, batch 5900, loss[loss=0.2841, simple_loss=0.351, pruned_loss=0.1086, over 19688.00 frames. ], tot_loss[loss=0.2973, simple_loss=0.3528, pruned_loss=0.1209, over 3820920.44 frames. ], batch size: 60, lr: 1.88e-02, grad_scale: 4.0 +2023-04-01 03:54:41,856 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 03:55:03,873 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 03:55:37,697 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.672e+02 6.673e+02 8.574e+02 1.115e+03 3.080e+03, threshold=1.715e+03, percent-clipped=4.0 +2023-04-01 03:55:39,799 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26432.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:55:43,006 INFO [train.py:903] (0/4) Epoch 4, batch 5950, loss[loss=0.286, simple_loss=0.3487, pruned_loss=0.1116, over 19382.00 frames. ], tot_loss[loss=0.2961, simple_loss=0.3516, pruned_loss=0.1203, over 3831208.68 frames. ], batch size: 70, lr: 1.88e-02, grad_scale: 4.0 +2023-04-01 03:56:17,508 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26464.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:56:43,853 INFO [train.py:903] (0/4) Epoch 4, batch 6000, loss[loss=0.2806, simple_loss=0.3484, pruned_loss=0.1063, over 17078.00 frames. ], tot_loss[loss=0.2951, simple_loss=0.3507, pruned_loss=0.1198, over 3820541.46 frames. ], batch size: 101, lr: 1.88e-02, grad_scale: 8.0 +2023-04-01 03:56:43,854 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 03:56:57,354 INFO [train.py:937] (0/4) Epoch 4, validation: loss=0.2103, simple_loss=0.3081, pruned_loss=0.05622, over 944034.00 frames. +2023-04-01 03:56:57,355 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 03:57:04,045 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-01 03:57:51,996 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26530.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:57:52,992 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.403e+02 7.085e+02 8.722e+02 1.145e+03 2.334e+03, threshold=1.744e+03, percent-clipped=4.0 +2023-04-01 03:57:54,583 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1003, 1.1228, 1.4465, 1.2384, 1.9359, 1.7555, 2.0140, 0.5065], + device='cuda:0'), covar=tensor([0.1568, 0.2585, 0.1423, 0.1406, 0.0931, 0.1374, 0.0892, 0.2402], + device='cuda:0'), in_proj_covar=tensor([0.0431, 0.0485, 0.0453, 0.0403, 0.0529, 0.0431, 0.0602, 0.0429], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 03:57:55,656 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4820, 1.2974, 1.4458, 0.9545, 2.6946, 3.2165, 2.9711, 3.3698], + device='cuda:0'), covar=tensor([0.1220, 0.2688, 0.2747, 0.2082, 0.0359, 0.0127, 0.0225, 0.0120], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0282, 0.0316, 0.0258, 0.0196, 0.0114, 0.0201, 0.0128], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 03:57:57,465 INFO [train.py:903] (0/4) Epoch 4, batch 6050, loss[loss=0.2713, simple_loss=0.3281, pruned_loss=0.1073, over 19614.00 frames. ], tot_loss[loss=0.2957, simple_loss=0.3514, pruned_loss=0.12, over 3813706.27 frames. ], batch size: 50, lr: 1.87e-02, grad_scale: 8.0 +2023-04-01 03:58:49,947 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26579.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:58:57,306 INFO [train.py:903] (0/4) Epoch 4, batch 6100, loss[loss=0.2498, simple_loss=0.3115, pruned_loss=0.09399, over 19468.00 frames. ], tot_loss[loss=0.2968, simple_loss=0.352, pruned_loss=0.1208, over 3796992.65 frames. ], batch size: 49, lr: 1.87e-02, grad_scale: 8.0 +2023-04-01 03:59:04,716 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26591.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 03:59:29,873 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26612.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 03:59:34,651 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26616.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 03:59:51,598 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.600e+02 6.768e+02 7.963e+02 1.032e+03 2.370e+03, threshold=1.593e+03, percent-clipped=5.0 +2023-04-01 03:59:56,256 INFO [train.py:903] (0/4) Epoch 4, batch 6150, loss[loss=0.267, simple_loss=0.3295, pruned_loss=0.1023, over 19862.00 frames. ], tot_loss[loss=0.2973, simple_loss=0.3529, pruned_loss=0.1208, over 3795416.12 frames. ], batch size: 52, lr: 1.87e-02, grad_scale: 8.0 +2023-04-01 04:00:08,621 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26645.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:00:23,799 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 04:00:54,796 INFO [train.py:903] (0/4) Epoch 4, batch 6200, loss[loss=0.2996, simple_loss=0.3615, pruned_loss=0.1189, over 19543.00 frames. ], tot_loss[loss=0.2976, simple_loss=0.3532, pruned_loss=0.121, over 3810278.98 frames. ], batch size: 56, lr: 1.87e-02, grad_scale: 8.0 +2023-04-01 04:01:45,232 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26727.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:01:51,166 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.725e+02 7.340e+02 8.907e+02 1.200e+03 2.630e+03, threshold=1.781e+03, percent-clipped=5.0 +2023-04-01 04:01:55,635 INFO [train.py:903] (0/4) Epoch 4, batch 6250, loss[loss=0.2551, simple_loss=0.313, pruned_loss=0.09861, over 16908.00 frames. ], tot_loss[loss=0.2984, simple_loss=0.3535, pruned_loss=0.1217, over 3800926.31 frames. ], batch size: 37, lr: 1.87e-02, grad_scale: 8.0 +2023-04-01 04:02:24,768 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 04:02:44,566 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=26776.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:02:55,315 INFO [train.py:903] (0/4) Epoch 4, batch 6300, loss[loss=0.3254, simple_loss=0.382, pruned_loss=0.1344, over 19523.00 frames. ], tot_loss[loss=0.298, simple_loss=0.3536, pruned_loss=0.1212, over 3820141.44 frames. ], batch size: 54, lr: 1.87e-02, grad_scale: 8.0 +2023-04-01 04:03:06,891 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4142, 2.3065, 2.0975, 3.3164, 2.3592, 3.6747, 3.1733, 2.1813], + device='cuda:0'), covar=tensor([0.1703, 0.1298, 0.0740, 0.0900, 0.1607, 0.0406, 0.1161, 0.1143], + device='cuda:0'), in_proj_covar=tensor([0.0584, 0.0567, 0.0531, 0.0731, 0.0632, 0.0483, 0.0636, 0.0539], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 04:03:49,878 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.898e+02 7.473e+02 9.358e+02 1.215e+03 3.413e+03, threshold=1.872e+03, percent-clipped=4.0 +2023-04-01 04:03:54,564 INFO [train.py:903] (0/4) Epoch 4, batch 6350, loss[loss=0.2715, simple_loss=0.3248, pruned_loss=0.1091, over 18669.00 frames. ], tot_loss[loss=0.2991, simple_loss=0.3549, pruned_loss=0.1216, over 3818788.94 frames. ], batch size: 41, lr: 1.86e-02, grad_scale: 8.0 +2023-04-01 04:03:54,968 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26835.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:04:26,060 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26860.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:04:55,432 INFO [train.py:903] (0/4) Epoch 4, batch 6400, loss[loss=0.3205, simple_loss=0.3794, pruned_loss=0.1308, over 19725.00 frames. ], tot_loss[loss=0.2976, simple_loss=0.3539, pruned_loss=0.1206, over 3823879.77 frames. ], batch size: 63, lr: 1.86e-02, grad_scale: 8.0 +2023-04-01 04:05:05,504 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=26891.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:05:16,536 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26901.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:05:41,164 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26922.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 04:05:45,906 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=26926.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:05:52,842 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.488e+02 6.991e+02 8.678e+02 1.023e+03 2.915e+03, threshold=1.736e+03, percent-clipped=2.0 +2023-04-01 04:05:58,078 INFO [train.py:903] (0/4) Epoch 4, batch 6450, loss[loss=0.3224, simple_loss=0.3643, pruned_loss=0.1402, over 19588.00 frames. ], tot_loss[loss=0.2952, simple_loss=0.3519, pruned_loss=0.1193, over 3824607.43 frames. ], batch size: 52, lr: 1.86e-02, grad_scale: 8.0 +2023-04-01 04:06:36,874 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=26968.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:06:39,715 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 04:06:56,677 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=26983.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:06:58,602 INFO [train.py:903] (0/4) Epoch 4, batch 6500, loss[loss=0.3, simple_loss=0.3591, pruned_loss=0.1204, over 19736.00 frames. ], tot_loss[loss=0.2956, simple_loss=0.3519, pruned_loss=0.1196, over 3821405.89 frames. ], batch size: 63, lr: 1.86e-02, grad_scale: 8.0 +2023-04-01 04:07:03,006 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 04:07:25,281 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=27008.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:07:42,483 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-04-01 04:07:53,581 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.310e+02 7.867e+02 9.982e+02 1.245e+03 2.621e+03, threshold=1.996e+03, percent-clipped=6.0 +2023-04-01 04:07:57,612 INFO [train.py:903] (0/4) Epoch 4, batch 6550, loss[loss=0.3415, simple_loss=0.387, pruned_loss=0.148, over 19668.00 frames. ], tot_loss[loss=0.2977, simple_loss=0.3531, pruned_loss=0.1212, over 3830637.31 frames. ], batch size: 59, lr: 1.86e-02, grad_scale: 8.0 +2023-04-01 04:08:18,787 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4003, 2.1206, 1.5620, 1.5026, 2.0201, 1.0783, 1.1277, 1.6296], + device='cuda:0'), covar=tensor([0.0597, 0.0453, 0.0853, 0.0485, 0.0321, 0.0998, 0.0647, 0.0356], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0255, 0.0314, 0.0236, 0.0213, 0.0308, 0.0284, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 04:08:34,242 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4464, 1.3110, 1.5902, 1.2290, 2.7082, 3.4532, 3.3370, 3.7114], + device='cuda:0'), covar=tensor([0.1353, 0.2775, 0.2685, 0.1930, 0.0379, 0.0154, 0.0186, 0.0096], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0279, 0.0314, 0.0259, 0.0194, 0.0114, 0.0202, 0.0130], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 04:08:57,147 INFO [train.py:903] (0/4) Epoch 4, batch 6600, loss[loss=0.3072, simple_loss=0.3697, pruned_loss=0.1223, over 19601.00 frames. ], tot_loss[loss=0.296, simple_loss=0.3518, pruned_loss=0.1201, over 3827811.62 frames. ], batch size: 57, lr: 1.86e-02, grad_scale: 8.0 +2023-04-01 04:09:48,364 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6050, 1.2855, 1.8946, 1.4465, 2.9408, 4.2356, 4.4123, 4.8713], + device='cuda:0'), covar=tensor([0.1342, 0.2847, 0.2538, 0.1854, 0.0398, 0.0132, 0.0146, 0.0079], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0277, 0.0314, 0.0257, 0.0194, 0.0114, 0.0199, 0.0129], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0001], + device='cuda:0') +2023-04-01 04:09:53,358 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.938e+02 7.635e+02 9.605e+02 1.185e+03 2.942e+03, threshold=1.921e+03, percent-clipped=6.0 +2023-04-01 04:09:58,812 INFO [train.py:903] (0/4) Epoch 4, batch 6650, loss[loss=0.2567, simple_loss=0.3182, pruned_loss=0.09759, over 19613.00 frames. ], tot_loss[loss=0.2973, simple_loss=0.3531, pruned_loss=0.1208, over 3822160.13 frames. ], batch size: 50, lr: 1.85e-02, grad_scale: 8.0 +2023-04-01 04:10:13,141 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27147.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:10:27,427 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0174, 1.5680, 1.5290, 2.1529, 1.7559, 1.8859, 1.6443, 1.9639], + device='cuda:0'), covar=tensor([0.0725, 0.1532, 0.1220, 0.0689, 0.1076, 0.0421, 0.0937, 0.0546], + device='cuda:0'), in_proj_covar=tensor([0.0249, 0.0367, 0.0279, 0.0238, 0.0304, 0.0253, 0.0270, 0.0232], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 04:10:35,294 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=27166.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:10:42,750 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=27172.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:10:58,824 INFO [train.py:903] (0/4) Epoch 4, batch 6700, loss[loss=0.2831, simple_loss=0.3457, pruned_loss=0.1103, over 19676.00 frames. ], tot_loss[loss=0.299, simple_loss=0.3543, pruned_loss=0.1219, over 3822559.75 frames. ], batch size: 53, lr: 1.85e-02, grad_scale: 8.0 +2023-04-01 04:11:52,644 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.097e+02 7.210e+02 9.176e+02 1.266e+03 4.477e+03, threshold=1.835e+03, percent-clipped=7.0 +2023-04-01 04:11:57,026 INFO [train.py:903] (0/4) Epoch 4, batch 6750, loss[loss=0.2724, simple_loss=0.3342, pruned_loss=0.1053, over 19612.00 frames. ], tot_loss[loss=0.3003, simple_loss=0.3555, pruned_loss=0.1225, over 3810126.95 frames. ], batch size: 50, lr: 1.85e-02, grad_scale: 8.0 +2023-04-01 04:12:31,956 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=27266.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 04:12:52,954 INFO [train.py:903] (0/4) Epoch 4, batch 6800, loss[loss=0.3137, simple_loss=0.3741, pruned_loss=0.1266, over 19661.00 frames. ], tot_loss[loss=0.3001, simple_loss=0.3553, pruned_loss=0.1225, over 3818227.78 frames. ], batch size: 58, lr: 1.85e-02, grad_scale: 8.0 +2023-04-01 04:13:21,175 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-4.pt +2023-04-01 04:13:37,539 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 04:13:37,988 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 04:13:40,505 INFO [train.py:903] (0/4) Epoch 5, batch 0, loss[loss=0.3105, simple_loss=0.3681, pruned_loss=0.1265, over 19743.00 frames. ], tot_loss[loss=0.3105, simple_loss=0.3681, pruned_loss=0.1265, over 19743.00 frames. ], batch size: 63, lr: 1.72e-02, grad_scale: 8.0 +2023-04-01 04:13:40,505 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 04:13:52,268 INFO [train.py:937] (0/4) Epoch 5, validation: loss=0.2121, simple_loss=0.3102, pruned_loss=0.05704, over 944034.00 frames. +2023-04-01 04:13:52,269 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 04:13:52,407 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=27312.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:13:55,669 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5892, 4.2059, 2.5719, 3.8282, 1.2790, 3.9287, 3.7847, 3.9840], + device='cuda:0'), covar=tensor([0.0563, 0.1020, 0.1732, 0.0674, 0.3582, 0.0691, 0.0685, 0.0780], + device='cuda:0'), in_proj_covar=tensor([0.0336, 0.0303, 0.0352, 0.0280, 0.0347, 0.0293, 0.0263, 0.0297], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 04:14:04,636 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 04:14:13,183 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-01 04:14:16,053 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 5.218e+02 7.861e+02 9.859e+02 1.236e+03 2.711e+03, threshold=1.972e+03, percent-clipped=3.0 +2023-04-01 04:14:52,474 INFO [train.py:903] (0/4) Epoch 5, batch 50, loss[loss=0.2489, simple_loss=0.3167, pruned_loss=0.09053, over 19836.00 frames. ], tot_loss[loss=0.2968, simple_loss=0.3529, pruned_loss=0.1203, over 869766.43 frames. ], batch size: 52, lr: 1.72e-02, grad_scale: 8.0 +2023-04-01 04:15:15,084 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27381.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 04:15:22,381 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-04-01 04:15:26,076 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 04:15:26,406 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1268, 0.9638, 1.0425, 1.4986, 1.2129, 1.2013, 1.2868, 1.2365], + device='cuda:0'), covar=tensor([0.0965, 0.1256, 0.1234, 0.0719, 0.0854, 0.0946, 0.0920, 0.0851], + device='cuda:0'), in_proj_covar=tensor([0.0238, 0.0254, 0.0250, 0.0288, 0.0280, 0.0233, 0.0247, 0.0230], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 04:15:53,859 INFO [train.py:903] (0/4) Epoch 5, batch 100, loss[loss=0.2809, simple_loss=0.346, pruned_loss=0.1079, over 19626.00 frames. ], tot_loss[loss=0.2911, simple_loss=0.3487, pruned_loss=0.1167, over 1523886.11 frames. ], batch size: 61, lr: 1.72e-02, grad_scale: 8.0 +2023-04-01 04:16:05,293 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 04:16:11,123 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27427.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:16:15,199 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.977e+02 6.953e+02 8.679e+02 1.081e+03 2.199e+03, threshold=1.736e+03, percent-clipped=1.0 +2023-04-01 04:16:53,736 INFO [train.py:903] (0/4) Epoch 5, batch 150, loss[loss=0.2876, simple_loss=0.3417, pruned_loss=0.1167, over 19719.00 frames. ], tot_loss[loss=0.2896, simple_loss=0.3472, pruned_loss=0.116, over 2049724.30 frames. ], batch size: 51, lr: 1.72e-02, grad_scale: 8.0 +2023-04-01 04:17:52,379 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=27510.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:17:54,460 INFO [train.py:903] (0/4) Epoch 5, batch 200, loss[loss=0.3076, simple_loss=0.3657, pruned_loss=0.1247, over 19655.00 frames. ], tot_loss[loss=0.2934, simple_loss=0.35, pruned_loss=0.1184, over 2433482.82 frames. ], batch size: 55, lr: 1.72e-02, grad_scale: 8.0 +2023-04-01 04:17:54,476 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 04:18:05,479 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.84 vs. limit=5.0 +2023-04-01 04:18:19,387 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.602e+02 6.870e+02 8.382e+02 1.064e+03 2.606e+03, threshold=1.676e+03, percent-clipped=3.0 +2023-04-01 04:18:40,863 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8646, 1.4554, 1.5631, 2.1078, 1.9206, 1.8279, 2.1149, 1.9031], + device='cuda:0'), covar=tensor([0.0560, 0.0847, 0.0855, 0.0718, 0.0734, 0.0713, 0.0693, 0.0578], + device='cuda:0'), in_proj_covar=tensor([0.0238, 0.0254, 0.0250, 0.0289, 0.0283, 0.0234, 0.0245, 0.0232], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 04:18:43,141 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2688, 2.0823, 1.5722, 1.4403, 1.9298, 1.0365, 1.1731, 1.5793], + device='cuda:0'), covar=tensor([0.0643, 0.0487, 0.0859, 0.0463, 0.0372, 0.0994, 0.0555, 0.0364], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0260, 0.0322, 0.0241, 0.0221, 0.0312, 0.0283, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 04:18:56,503 INFO [train.py:903] (0/4) Epoch 5, batch 250, loss[loss=0.289, simple_loss=0.3518, pruned_loss=0.1131, over 19659.00 frames. ], tot_loss[loss=0.2933, simple_loss=0.3506, pruned_loss=0.118, over 2749329.30 frames. ], batch size: 58, lr: 1.71e-02, grad_scale: 4.0 +2023-04-01 04:19:51,634 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-04-01 04:19:58,944 INFO [train.py:903] (0/4) Epoch 5, batch 300, loss[loss=0.301, simple_loss=0.3693, pruned_loss=0.1163, over 19680.00 frames. ], tot_loss[loss=0.2933, simple_loss=0.3504, pruned_loss=0.1181, over 2988419.38 frames. ], batch size: 59, lr: 1.71e-02, grad_scale: 4.0 +2023-04-01 04:20:15,393 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=27625.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:20:22,963 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.497e+02 6.581e+02 8.607e+02 1.103e+03 1.922e+03, threshold=1.721e+03, percent-clipped=6.0 +2023-04-01 04:20:28,958 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27637.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 04:20:43,959 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-04-01 04:21:01,020 INFO [train.py:903] (0/4) Epoch 5, batch 350, loss[loss=0.3139, simple_loss=0.3806, pruned_loss=0.1235, over 19705.00 frames. ], tot_loss[loss=0.2921, simple_loss=0.3491, pruned_loss=0.1175, over 3185611.28 frames. ], batch size: 59, lr: 1.71e-02, grad_scale: 4.0 +2023-04-01 04:21:01,367 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=27662.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 04:21:07,197 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 04:21:11,013 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8634, 1.3655, 0.9820, 1.0107, 1.2194, 0.8618, 0.8606, 1.2238], + device='cuda:0'), covar=tensor([0.0440, 0.0535, 0.0870, 0.0426, 0.0370, 0.0955, 0.0517, 0.0314], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0258, 0.0313, 0.0238, 0.0217, 0.0314, 0.0282, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 04:21:26,386 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27683.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:21:58,719 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=27708.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:22:02,931 INFO [train.py:903] (0/4) Epoch 5, batch 400, loss[loss=0.2833, simple_loss=0.3436, pruned_loss=0.1115, over 19674.00 frames. ], tot_loss[loss=0.2903, simple_loss=0.3479, pruned_loss=0.1164, over 3337691.28 frames. ], batch size: 60, lr: 1.71e-02, grad_scale: 8.0 +2023-04-01 04:22:27,976 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.948e+02 7.305e+02 9.017e+02 1.065e+03 1.815e+03, threshold=1.803e+03, percent-clipped=3.0 +2023-04-01 04:22:28,315 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=27732.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 04:23:04,213 INFO [train.py:903] (0/4) Epoch 5, batch 450, loss[loss=0.2879, simple_loss=0.3505, pruned_loss=0.1126, over 19617.00 frames. ], tot_loss[loss=0.2894, simple_loss=0.3474, pruned_loss=0.1157, over 3446463.61 frames. ], batch size: 57, lr: 1.71e-02, grad_scale: 8.0 +2023-04-01 04:23:46,004 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 04:23:47,123 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 04:24:07,122 INFO [train.py:903] (0/4) Epoch 5, batch 500, loss[loss=0.2766, simple_loss=0.3345, pruned_loss=0.1094, over 19805.00 frames. ], tot_loss[loss=0.2885, simple_loss=0.3469, pruned_loss=0.1151, over 3546593.16 frames. ], batch size: 49, lr: 1.71e-02, grad_scale: 8.0 +2023-04-01 04:24:29,747 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4185, 1.0357, 1.2053, 1.2842, 2.0598, 1.1252, 1.7961, 2.0968], + device='cuda:0'), covar=tensor([0.0599, 0.2676, 0.2511, 0.1446, 0.0776, 0.1787, 0.0951, 0.0661], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0310, 0.0309, 0.0285, 0.0304, 0.0316, 0.0283, 0.0297], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 04:24:31,792 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.519e+02 6.150e+02 8.318e+02 1.057e+03 1.987e+03, threshold=1.664e+03, percent-clipped=1.0 +2023-04-01 04:25:11,321 INFO [train.py:903] (0/4) Epoch 5, batch 550, loss[loss=0.3533, simple_loss=0.3968, pruned_loss=0.1549, over 19539.00 frames. ], tot_loss[loss=0.2897, simple_loss=0.3481, pruned_loss=0.1156, over 3616510.61 frames. ], batch size: 54, lr: 1.71e-02, grad_scale: 8.0 +2023-04-01 04:25:35,615 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=27881.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:26:07,938 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=27906.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:26:14,562 INFO [train.py:903] (0/4) Epoch 5, batch 600, loss[loss=0.3217, simple_loss=0.3821, pruned_loss=0.1306, over 19500.00 frames. ], tot_loss[loss=0.2873, simple_loss=0.3463, pruned_loss=0.1141, over 3676515.85 frames. ], batch size: 64, lr: 1.70e-02, grad_scale: 8.0 +2023-04-01 04:26:38,769 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.822e+02 6.595e+02 8.388e+02 1.023e+03 2.578e+03, threshold=1.678e+03, percent-clipped=3.0 +2023-04-01 04:27:02,826 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 04:27:17,957 INFO [train.py:903] (0/4) Epoch 5, batch 650, loss[loss=0.3042, simple_loss=0.3685, pruned_loss=0.12, over 19486.00 frames. ], tot_loss[loss=0.2883, simple_loss=0.3472, pruned_loss=0.1147, over 3718564.89 frames. ], batch size: 64, lr: 1.70e-02, grad_scale: 8.0 +2023-04-01 04:27:35,648 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7672, 1.8882, 1.7566, 2.6062, 1.6660, 2.3672, 2.2830, 1.7406], + device='cuda:0'), covar=tensor([0.1575, 0.1255, 0.0770, 0.0618, 0.1469, 0.0531, 0.1268, 0.1218], + device='cuda:0'), in_proj_covar=tensor([0.0593, 0.0576, 0.0533, 0.0733, 0.0639, 0.0494, 0.0649, 0.0549], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 04:27:36,863 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1816, 2.4232, 2.1378, 3.5797, 2.3228, 3.7066, 3.3254, 2.2417], + device='cuda:0'), covar=tensor([0.1918, 0.1330, 0.0732, 0.0763, 0.1764, 0.0429, 0.1075, 0.1139], + device='cuda:0'), in_proj_covar=tensor([0.0592, 0.0575, 0.0532, 0.0733, 0.0638, 0.0493, 0.0648, 0.0548], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 04:28:05,333 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-28000.pt +2023-04-01 04:28:20,088 INFO [train.py:903] (0/4) Epoch 5, batch 700, loss[loss=0.3808, simple_loss=0.4172, pruned_loss=0.1722, over 18848.00 frames. ], tot_loss[loss=0.2914, simple_loss=0.3499, pruned_loss=0.1164, over 3746228.73 frames. ], batch size: 74, lr: 1.70e-02, grad_scale: 8.0 +2023-04-01 04:28:47,045 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.733e+02 7.486e+02 9.333e+02 1.140e+03 2.488e+03, threshold=1.867e+03, percent-clipped=5.0 +2023-04-01 04:28:48,570 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0661, 3.4503, 3.6345, 3.6013, 1.4828, 3.2787, 2.9679, 3.2668], + device='cuda:0'), covar=tensor([0.1099, 0.0946, 0.0591, 0.0525, 0.3845, 0.0539, 0.0595, 0.1085], + device='cuda:0'), in_proj_covar=tensor([0.0477, 0.0426, 0.0567, 0.0450, 0.0552, 0.0323, 0.0367, 0.0525], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 04:29:25,959 INFO [train.py:903] (0/4) Epoch 5, batch 750, loss[loss=0.2616, simple_loss=0.3135, pruned_loss=0.1049, over 19369.00 frames. ], tot_loss[loss=0.2916, simple_loss=0.3494, pruned_loss=0.1169, over 3745790.62 frames. ], batch size: 47, lr: 1.70e-02, grad_scale: 8.0 +2023-04-01 04:29:43,610 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28076.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 04:30:27,878 INFO [train.py:903] (0/4) Epoch 5, batch 800, loss[loss=0.2572, simple_loss=0.3155, pruned_loss=0.09949, over 19402.00 frames. ], tot_loss[loss=0.2908, simple_loss=0.3488, pruned_loss=0.1164, over 3769501.47 frames. ], batch size: 48, lr: 1.70e-02, grad_scale: 8.0 +2023-04-01 04:30:48,172 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 04:30:52,981 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.190e+02 6.427e+02 8.567e+02 1.032e+03 2.729e+03, threshold=1.713e+03, percent-clipped=3.0 +2023-04-01 04:31:32,201 INFO [train.py:903] (0/4) Epoch 5, batch 850, loss[loss=0.305, simple_loss=0.3653, pruned_loss=0.1223, over 19658.00 frames. ], tot_loss[loss=0.2905, simple_loss=0.3484, pruned_loss=0.1163, over 3779897.05 frames. ], batch size: 58, lr: 1.70e-02, grad_scale: 8.0 +2023-04-01 04:31:55,401 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9746, 2.0102, 1.7643, 2.7777, 1.7629, 2.5573, 2.5649, 1.8991], + device='cuda:0'), covar=tensor([0.1680, 0.1264, 0.0774, 0.0824, 0.1636, 0.0565, 0.1275, 0.1173], + device='cuda:0'), in_proj_covar=tensor([0.0590, 0.0574, 0.0530, 0.0733, 0.0633, 0.0496, 0.0639, 0.0544], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 04:32:08,717 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28191.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 04:32:29,192 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 04:32:33,625 INFO [train.py:903] (0/4) Epoch 5, batch 900, loss[loss=0.2655, simple_loss=0.3393, pruned_loss=0.09585, over 19662.00 frames. ], tot_loss[loss=0.2898, simple_loss=0.348, pruned_loss=0.1158, over 3794012.63 frames. ], batch size: 58, lr: 1.69e-02, grad_scale: 8.0 +2023-04-01 04:32:59,335 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.597e+02 6.938e+02 8.196e+02 1.125e+03 2.658e+03, threshold=1.639e+03, percent-clipped=4.0 +2023-04-01 04:33:17,411 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28247.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:33:36,689 INFO [train.py:903] (0/4) Epoch 5, batch 950, loss[loss=0.2932, simple_loss=0.3591, pruned_loss=0.1136, over 19288.00 frames. ], tot_loss[loss=0.2909, simple_loss=0.3486, pruned_loss=0.1166, over 3787412.55 frames. ], batch size: 70, lr: 1.69e-02, grad_scale: 8.0 +2023-04-01 04:33:42,366 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 04:33:49,413 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28273.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 04:34:28,478 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28305.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:34:36,190 INFO [train.py:903] (0/4) Epoch 5, batch 1000, loss[loss=0.2746, simple_loss=0.3388, pruned_loss=0.1052, over 19831.00 frames. ], tot_loss[loss=0.2915, simple_loss=0.349, pruned_loss=0.117, over 3792616.61 frames. ], batch size: 52, lr: 1.69e-02, grad_scale: 8.0 +2023-04-01 04:34:51,484 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.93 vs. limit=5.0 +2023-04-01 04:34:59,302 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.517e+02 7.143e+02 8.888e+02 1.138e+03 2.880e+03, threshold=1.778e+03, percent-clipped=9.0 +2023-04-01 04:35:30,262 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 04:35:36,616 INFO [train.py:903] (0/4) Epoch 5, batch 1050, loss[loss=0.2889, simple_loss=0.3353, pruned_loss=0.1213, over 19709.00 frames. ], tot_loss[loss=0.2918, simple_loss=0.349, pruned_loss=0.1173, over 3799515.43 frames. ], batch size: 45, lr: 1.69e-02, grad_scale: 8.0 +2023-04-01 04:36:09,501 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 04:36:36,350 INFO [train.py:903] (0/4) Epoch 5, batch 1100, loss[loss=0.3588, simple_loss=0.405, pruned_loss=0.1563, over 19525.00 frames. ], tot_loss[loss=0.2915, simple_loss=0.3485, pruned_loss=0.1172, over 3813379.45 frames. ], batch size: 54, lr: 1.69e-02, grad_scale: 8.0 +2023-04-01 04:36:56,391 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2162, 1.1930, 1.4312, 1.3238, 1.8175, 1.8652, 1.8936, 0.4491], + device='cuda:0'), covar=tensor([0.1849, 0.2944, 0.1716, 0.1472, 0.1143, 0.1535, 0.1115, 0.2849], + device='cuda:0'), in_proj_covar=tensor([0.0441, 0.0495, 0.0467, 0.0408, 0.0538, 0.0439, 0.0616, 0.0430], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 04:37:01,573 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.535e+02 6.917e+02 8.996e+02 1.190e+03 3.192e+03, threshold=1.799e+03, percent-clipped=6.0 +2023-04-01 04:37:19,643 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28447.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 04:37:37,666 INFO [train.py:903] (0/4) Epoch 5, batch 1150, loss[loss=0.3022, simple_loss=0.3611, pruned_loss=0.1217, over 19614.00 frames. ], tot_loss[loss=0.2881, simple_loss=0.3464, pruned_loss=0.115, over 3827959.53 frames. ], batch size: 57, lr: 1.69e-02, grad_scale: 8.0 +2023-04-01 04:37:50,278 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28472.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 04:38:11,081 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7895, 1.8923, 1.8533, 2.9047, 2.0918, 2.5928, 3.0533, 2.8570], + device='cuda:0'), covar=tensor([0.0651, 0.1013, 0.1102, 0.0948, 0.1050, 0.0724, 0.0899, 0.0581], + device='cuda:0'), in_proj_covar=tensor([0.0232, 0.0254, 0.0245, 0.0279, 0.0280, 0.0237, 0.0242, 0.0229], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0004, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 04:38:29,188 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28504.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:38:36,122 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28510.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:38:38,083 INFO [train.py:903] (0/4) Epoch 5, batch 1200, loss[loss=0.2677, simple_loss=0.3284, pruned_loss=0.1035, over 19725.00 frames. ], tot_loss[loss=0.289, simple_loss=0.3475, pruned_loss=0.1153, over 3835695.38 frames. ], batch size: 51, lr: 1.69e-02, grad_scale: 8.0 +2023-04-01 04:39:01,763 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.620e+02 6.666e+02 7.999e+02 1.012e+03 1.920e+03, threshold=1.600e+03, percent-clipped=0.0 +2023-04-01 04:39:12,237 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 04:39:37,132 INFO [train.py:903] (0/4) Epoch 5, batch 1250, loss[loss=0.2976, simple_loss=0.3596, pruned_loss=0.1178, over 19402.00 frames. ], tot_loss[loss=0.2898, simple_loss=0.3482, pruned_loss=0.1157, over 3835710.92 frames. ], batch size: 66, lr: 1.68e-02, grad_scale: 8.0 +2023-04-01 04:40:13,037 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28591.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:40:37,856 INFO [train.py:903] (0/4) Epoch 5, batch 1300, loss[loss=0.2711, simple_loss=0.3428, pruned_loss=0.09967, over 19635.00 frames. ], tot_loss[loss=0.2897, simple_loss=0.348, pruned_loss=0.1157, over 3832412.32 frames. ], batch size: 57, lr: 1.68e-02, grad_scale: 8.0 +2023-04-01 04:40:43,789 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28617.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 04:41:03,192 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.211e+02 7.072e+02 8.596e+02 1.219e+03 1.879e+03, threshold=1.719e+03, percent-clipped=8.0 +2023-04-01 04:41:11,313 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9512, 2.8228, 1.7350, 2.1352, 1.7966, 2.0941, 0.6564, 2.1611], + device='cuda:0'), covar=tensor([0.0274, 0.0253, 0.0321, 0.0388, 0.0494, 0.0359, 0.0602, 0.0450], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0277, 0.0275, 0.0296, 0.0360, 0.0277, 0.0273, 0.0281], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 04:41:23,365 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28649.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:41:39,858 INFO [train.py:903] (0/4) Epoch 5, batch 1350, loss[loss=0.337, simple_loss=0.3765, pruned_loss=0.1488, over 13235.00 frames. ], tot_loss[loss=0.2915, simple_loss=0.3492, pruned_loss=0.1169, over 3803158.54 frames. ], batch size: 136, lr: 1.68e-02, grad_scale: 8.0 +2023-04-01 04:42:34,039 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28706.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:42:40,549 INFO [train.py:903] (0/4) Epoch 5, batch 1400, loss[loss=0.3283, simple_loss=0.3767, pruned_loss=0.1399, over 17588.00 frames. ], tot_loss[loss=0.2917, simple_loss=0.3494, pruned_loss=0.117, over 3801432.51 frames. ], batch size: 101, lr: 1.68e-02, grad_scale: 8.0 +2023-04-01 04:43:01,030 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5345, 3.9484, 4.1822, 4.1392, 1.5472, 3.7636, 3.3865, 3.7335], + device='cuda:0'), covar=tensor([0.0940, 0.0626, 0.0470, 0.0408, 0.3648, 0.0375, 0.0526, 0.0931], + device='cuda:0'), in_proj_covar=tensor([0.0495, 0.0445, 0.0587, 0.0466, 0.0564, 0.0337, 0.0379, 0.0543], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 04:43:04,133 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.747e+02 6.792e+02 9.169e+02 1.129e+03 1.829e+03, threshold=1.834e+03, percent-clipped=1.0 +2023-04-01 04:43:04,516 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28732.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 04:43:40,836 INFO [train.py:903] (0/4) Epoch 5, batch 1450, loss[loss=0.2367, simple_loss=0.3009, pruned_loss=0.08624, over 19774.00 frames. ], tot_loss[loss=0.2896, simple_loss=0.3476, pruned_loss=0.1158, over 3805596.52 frames. ], batch size: 47, lr: 1.68e-02, grad_scale: 8.0 +2023-04-01 04:43:43,185 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 04:43:43,500 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28764.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:44:41,296 INFO [train.py:903] (0/4) Epoch 5, batch 1500, loss[loss=0.2887, simple_loss=0.3435, pruned_loss=0.1169, over 19829.00 frames. ], tot_loss[loss=0.2887, simple_loss=0.3469, pruned_loss=0.1152, over 3816649.33 frames. ], batch size: 52, lr: 1.68e-02, grad_scale: 8.0 +2023-04-01 04:44:55,667 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4897, 1.1857, 1.6021, 1.0492, 2.4713, 2.7454, 2.6754, 2.8837], + device='cuda:0'), covar=tensor([0.1089, 0.2624, 0.2445, 0.1898, 0.0442, 0.0304, 0.0246, 0.0163], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0281, 0.0320, 0.0256, 0.0197, 0.0116, 0.0206, 0.0138], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 04:45:06,145 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.730e+02 6.912e+02 8.562e+02 1.022e+03 2.509e+03, threshold=1.712e+03, percent-clipped=1.0 +2023-04-01 04:45:24,295 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28848.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:45:31,721 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=28854.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:45:32,286 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-01 04:45:42,352 INFO [train.py:903] (0/4) Epoch 5, batch 1550, loss[loss=0.2712, simple_loss=0.324, pruned_loss=0.1093, over 19314.00 frames. ], tot_loss[loss=0.2876, simple_loss=0.3458, pruned_loss=0.1147, over 3817385.99 frames. ], batch size: 44, lr: 1.68e-02, grad_scale: 8.0 +2023-04-01 04:46:02,329 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=28879.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:46:41,474 INFO [train.py:903] (0/4) Epoch 5, batch 1600, loss[loss=0.3224, simple_loss=0.3709, pruned_loss=0.137, over 19511.00 frames. ], tot_loss[loss=0.2894, simple_loss=0.3473, pruned_loss=0.1157, over 3802716.21 frames. ], batch size: 54, lr: 1.67e-02, grad_scale: 8.0 +2023-04-01 04:47:01,760 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8626, 1.9110, 1.8148, 2.6817, 1.7100, 2.4593, 2.4792, 1.7734], + device='cuda:0'), covar=tensor([0.1643, 0.1307, 0.0754, 0.0819, 0.1575, 0.0564, 0.1263, 0.1285], + device='cuda:0'), in_proj_covar=tensor([0.0605, 0.0591, 0.0544, 0.0751, 0.0651, 0.0507, 0.0656, 0.0561], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 04:47:04,538 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.940e+02 6.984e+02 8.420e+02 1.102e+03 2.946e+03, threshold=1.684e+03, percent-clipped=4.0 +2023-04-01 04:47:04,565 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 04:47:41,583 INFO [train.py:903] (0/4) Epoch 5, batch 1650, loss[loss=0.3094, simple_loss=0.3661, pruned_loss=0.1264, over 18828.00 frames. ], tot_loss[loss=0.2892, simple_loss=0.3471, pruned_loss=0.1156, over 3812322.11 frames. ], batch size: 74, lr: 1.67e-02, grad_scale: 4.0 +2023-04-01 04:47:42,028 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28962.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:47:43,145 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28963.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:47:49,892 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=28969.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:48:12,452 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=28987.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:48:13,613 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=28988.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 04:48:42,351 INFO [train.py:903] (0/4) Epoch 5, batch 1700, loss[loss=0.3343, simple_loss=0.3788, pruned_loss=0.1449, over 17310.00 frames. ], tot_loss[loss=0.2885, simple_loss=0.3463, pruned_loss=0.1154, over 3817670.04 frames. ], batch size: 101, lr: 1.67e-02, grad_scale: 4.0 +2023-04-01 04:48:43,895 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29013.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 04:48:51,927 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29020.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:48:52,070 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29020.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:49:08,766 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.635e+02 6.699e+02 8.227e+02 1.083e+03 2.721e+03, threshold=1.645e+03, percent-clipped=4.0 +2023-04-01 04:49:16,951 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2218, 2.1881, 2.1709, 2.9890, 2.3831, 2.9687, 2.7354, 1.9956], + device='cuda:0'), covar=tensor([0.1401, 0.1105, 0.0596, 0.0635, 0.1136, 0.0365, 0.0946, 0.1010], + device='cuda:0'), in_proj_covar=tensor([0.0599, 0.0589, 0.0536, 0.0743, 0.0645, 0.0502, 0.0657, 0.0557], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 04:49:21,047 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 04:49:22,493 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29045.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:49:36,747 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.48 vs. limit=5.0 +2023-04-01 04:49:40,680 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9745, 2.0119, 1.8856, 2.7413, 1.7938, 2.7814, 2.5119, 1.8195], + device='cuda:0'), covar=tensor([0.1696, 0.1293, 0.0757, 0.0800, 0.1574, 0.0480, 0.1385, 0.1281], + device='cuda:0'), in_proj_covar=tensor([0.0594, 0.0584, 0.0531, 0.0737, 0.0640, 0.0498, 0.0651, 0.0554], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 04:49:42,474 INFO [train.py:903] (0/4) Epoch 5, batch 1750, loss[loss=0.284, simple_loss=0.3596, pruned_loss=0.1042, over 19291.00 frames. ], tot_loss[loss=0.2869, simple_loss=0.3451, pruned_loss=0.1144, over 3814772.84 frames. ], batch size: 66, lr: 1.67e-02, grad_scale: 4.0 +2023-04-01 04:50:26,370 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-04-01 04:50:43,924 INFO [train.py:903] (0/4) Epoch 5, batch 1800, loss[loss=0.2576, simple_loss=0.3325, pruned_loss=0.09142, over 19766.00 frames. ], tot_loss[loss=0.2865, simple_loss=0.3442, pruned_loss=0.1143, over 3815482.89 frames. ], batch size: 63, lr: 1.67e-02, grad_scale: 4.0 +2023-04-01 04:50:59,177 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0545, 1.1806, 1.5576, 0.8160, 2.4901, 2.8511, 2.6955, 3.0803], + device='cuda:0'), covar=tensor([0.1423, 0.2811, 0.2724, 0.2003, 0.0382, 0.0144, 0.0262, 0.0145], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0278, 0.0314, 0.0253, 0.0197, 0.0115, 0.0205, 0.0136], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 04:51:07,718 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.740e+02 6.050e+02 7.849e+02 1.011e+03 2.328e+03, threshold=1.570e+03, percent-clipped=7.0 +2023-04-01 04:51:39,754 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 04:51:44,191 INFO [train.py:903] (0/4) Epoch 5, batch 1850, loss[loss=0.2514, simple_loss=0.3181, pruned_loss=0.09233, over 19500.00 frames. ], tot_loss[loss=0.2864, simple_loss=0.3446, pruned_loss=0.1141, over 3807896.22 frames. ], batch size: 49, lr: 1.67e-02, grad_scale: 4.0 +2023-04-01 04:52:17,660 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 04:52:19,834 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0 +2023-04-01 04:52:43,728 INFO [train.py:903] (0/4) Epoch 5, batch 1900, loss[loss=0.3243, simple_loss=0.3824, pruned_loss=0.1331, over 19699.00 frames. ], tot_loss[loss=0.2886, simple_loss=0.3464, pruned_loss=0.1153, over 3818231.57 frames. ], batch size: 59, lr: 1.67e-02, grad_scale: 4.0 +2023-04-01 04:52:53,184 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29219.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:52:58,359 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=29223.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:53:01,679 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29225.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:53:02,418 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 04:53:07,799 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 04:53:11,315 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.676e+02 6.988e+02 9.073e+02 1.156e+03 1.890e+03, threshold=1.815e+03, percent-clipped=5.0 +2023-04-01 04:53:23,884 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29244.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:53:28,884 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 04:53:30,457 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29250.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:53:35,730 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29254.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:53:44,967 INFO [train.py:903] (0/4) Epoch 5, batch 1950, loss[loss=0.2492, simple_loss=0.3005, pruned_loss=0.09895, over 19751.00 frames. ], tot_loss[loss=0.289, simple_loss=0.3467, pruned_loss=0.1157, over 3819574.64 frames. ], batch size: 46, lr: 1.67e-02, grad_scale: 4.0 +2023-04-01 04:54:46,885 INFO [train.py:903] (0/4) Epoch 5, batch 2000, loss[loss=0.2637, simple_loss=0.3229, pruned_loss=0.1022, over 19828.00 frames. ], tot_loss[loss=0.2886, simple_loss=0.3465, pruned_loss=0.1154, over 3805304.89 frames. ], batch size: 52, lr: 1.66e-02, grad_scale: 8.0 +2023-04-01 04:55:10,279 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.142e+02 7.102e+02 9.141e+02 1.135e+03 3.050e+03, threshold=1.828e+03, percent-clipped=2.0 +2023-04-01 04:55:10,958 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-04-01 04:55:16,857 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29338.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:55:43,166 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 04:55:46,566 INFO [train.py:903] (0/4) Epoch 5, batch 2050, loss[loss=0.2711, simple_loss=0.3391, pruned_loss=0.1015, over 19578.00 frames. ], tot_loss[loss=0.2871, simple_loss=0.3455, pruned_loss=0.1144, over 3802581.34 frames. ], batch size: 61, lr: 1.66e-02, grad_scale: 8.0 +2023-04-01 04:55:49,002 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=29364.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:55:59,870 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 04:56:00,835 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 04:56:23,069 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 04:56:37,161 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1789, 1.2074, 1.5533, 1.2709, 2.1913, 1.9146, 2.3501, 0.7804], + device='cuda:0'), covar=tensor([0.1648, 0.2723, 0.1464, 0.1428, 0.1056, 0.1396, 0.1104, 0.2576], + device='cuda:0'), in_proj_covar=tensor([0.0440, 0.0502, 0.0472, 0.0405, 0.0538, 0.0434, 0.0620, 0.0436], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 04:56:45,838 INFO [train.py:903] (0/4) Epoch 5, batch 2100, loss[loss=0.3063, simple_loss=0.3601, pruned_loss=0.1263, over 13473.00 frames. ], tot_loss[loss=0.2902, simple_loss=0.3481, pruned_loss=0.1162, over 3795799.01 frames. ], batch size: 136, lr: 1.66e-02, grad_scale: 8.0 +2023-04-01 04:57:07,818 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-04-01 04:57:12,304 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.101e+02 7.145e+02 9.347e+02 1.270e+03 4.921e+03, threshold=1.869e+03, percent-clipped=10.0 +2023-04-01 04:57:13,529 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 04:57:20,913 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1587, 2.1065, 1.5188, 1.2266, 1.9421, 0.8394, 1.1743, 1.7254], + device='cuda:0'), covar=tensor([0.0616, 0.0339, 0.0696, 0.0530, 0.0298, 0.1017, 0.0489, 0.0267], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0260, 0.0306, 0.0236, 0.0212, 0.0306, 0.0280, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 04:57:34,281 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 04:57:46,686 INFO [train.py:903] (0/4) Epoch 5, batch 2150, loss[loss=0.2506, simple_loss=0.3074, pruned_loss=0.09697, over 19781.00 frames. ], tot_loss[loss=0.2916, simple_loss=0.3493, pruned_loss=0.117, over 3797948.58 frames. ], batch size: 47, lr: 1.66e-02, grad_scale: 8.0 +2023-04-01 04:58:08,901 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29479.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 04:58:48,389 INFO [train.py:903] (0/4) Epoch 5, batch 2200, loss[loss=0.2935, simple_loss=0.3532, pruned_loss=0.1168, over 17541.00 frames. ], tot_loss[loss=0.2918, simple_loss=0.3494, pruned_loss=0.1171, over 3793764.44 frames. ], batch size: 101, lr: 1.66e-02, grad_scale: 8.0 +2023-04-01 04:59:11,936 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.753e+02 6.760e+02 8.285e+02 1.117e+03 1.782e+03, threshold=1.657e+03, percent-clipped=0.0 +2023-04-01 04:59:47,491 INFO [train.py:903] (0/4) Epoch 5, batch 2250, loss[loss=0.3097, simple_loss=0.37, pruned_loss=0.1247, over 19544.00 frames. ], tot_loss[loss=0.2903, simple_loss=0.3484, pruned_loss=0.1161, over 3807165.73 frames. ], batch size: 56, lr: 1.66e-02, grad_scale: 8.0 +2023-04-01 04:59:55,522 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29569.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:00:28,197 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29594.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:00:31,586 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5070, 1.1138, 1.1800, 1.6222, 1.3272, 1.5800, 1.7592, 1.4357], + device='cuda:0'), covar=tensor([0.0833, 0.1159, 0.1173, 0.0910, 0.0909, 0.0839, 0.0919, 0.0811], + device='cuda:0'), in_proj_covar=tensor([0.0234, 0.0253, 0.0245, 0.0283, 0.0277, 0.0230, 0.0240, 0.0229], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 05:00:32,399 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=29598.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:00:48,024 INFO [train.py:903] (0/4) Epoch 5, batch 2300, loss[loss=0.2865, simple_loss=0.3357, pruned_loss=0.1186, over 19828.00 frames. ], tot_loss[loss=0.289, simple_loss=0.3475, pruned_loss=0.1152, over 3816865.51 frames. ], batch size: 52, lr: 1.66e-02, grad_scale: 8.0 +2023-04-01 05:00:56,315 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29619.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:01:03,670 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 05:01:15,152 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.694e+02 6.380e+02 7.654e+02 9.572e+02 1.859e+03, threshold=1.531e+03, percent-clipped=2.0 +2023-04-01 05:01:48,834 INFO [train.py:903] (0/4) Epoch 5, batch 2350, loss[loss=0.3285, simple_loss=0.3691, pruned_loss=0.1439, over 13536.00 frames. ], tot_loss[loss=0.2884, simple_loss=0.3471, pruned_loss=0.1148, over 3800785.09 frames. ], batch size: 136, lr: 1.65e-02, grad_scale: 8.0 +2023-04-01 05:02:11,351 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29680.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:02:28,866 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 05:02:45,269 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 05:02:49,427 INFO [train.py:903] (0/4) Epoch 5, batch 2400, loss[loss=0.261, simple_loss=0.3324, pruned_loss=0.0948, over 19590.00 frames. ], tot_loss[loss=0.2877, simple_loss=0.3467, pruned_loss=0.1143, over 3807691.22 frames. ], batch size: 57, lr: 1.65e-02, grad_scale: 8.0 +2023-04-01 05:02:50,597 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=29713.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:03:12,596 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.376e+02 6.843e+02 7.889e+02 1.103e+03 3.246e+03, threshold=1.578e+03, percent-clipped=5.0 +2023-04-01 05:03:15,272 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29735.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:03:46,774 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29760.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:03:48,490 INFO [train.py:903] (0/4) Epoch 5, batch 2450, loss[loss=0.3431, simple_loss=0.3765, pruned_loss=0.1548, over 18074.00 frames. ], tot_loss[loss=0.2888, simple_loss=0.3477, pruned_loss=0.115, over 3821359.58 frames. ], batch size: 83, lr: 1.65e-02, grad_scale: 8.0 +2023-04-01 05:04:48,603 INFO [train.py:903] (0/4) Epoch 5, batch 2500, loss[loss=0.2989, simple_loss=0.3524, pruned_loss=0.1227, over 19586.00 frames. ], tot_loss[loss=0.2883, simple_loss=0.3475, pruned_loss=0.1145, over 3819633.86 frames. ], batch size: 52, lr: 1.65e-02, grad_scale: 8.0 +2023-04-01 05:04:57,774 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=29820.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:05:14,587 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.471e+02 6.790e+02 8.500e+02 1.086e+03 2.138e+03, threshold=1.700e+03, percent-clipped=3.0 +2023-04-01 05:05:44,700 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.58 vs. limit=5.0 +2023-04-01 05:05:48,271 INFO [train.py:903] (0/4) Epoch 5, batch 2550, loss[loss=0.3128, simple_loss=0.3669, pruned_loss=0.1293, over 19762.00 frames. ], tot_loss[loss=0.2881, simple_loss=0.3472, pruned_loss=0.1145, over 3818443.76 frames. ], batch size: 54, lr: 1.65e-02, grad_scale: 8.0 +2023-04-01 05:06:24,469 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-04-01 05:06:34,350 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4263, 1.4698, 2.0021, 1.5995, 3.0083, 2.5029, 2.9848, 1.3823], + device='cuda:0'), covar=tensor([0.1596, 0.2744, 0.1549, 0.1294, 0.1168, 0.1266, 0.1415, 0.2678], + device='cuda:0'), in_proj_covar=tensor([0.0439, 0.0504, 0.0472, 0.0403, 0.0546, 0.0435, 0.0620, 0.0437], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:06:40,389 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 05:06:48,588 INFO [train.py:903] (0/4) Epoch 5, batch 2600, loss[loss=0.2857, simple_loss=0.3529, pruned_loss=0.1093, over 19536.00 frames. ], tot_loss[loss=0.2893, simple_loss=0.3483, pruned_loss=0.1151, over 3821463.48 frames. ], batch size: 56, lr: 1.65e-02, grad_scale: 8.0 +2023-04-01 05:06:50,697 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=29913.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:07:13,516 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.137e+02 6.882e+02 9.175e+02 1.239e+03 1.828e+03, threshold=1.835e+03, percent-clipped=5.0 +2023-04-01 05:07:50,213 INFO [train.py:903] (0/4) Epoch 5, batch 2650, loss[loss=0.2624, simple_loss=0.3279, pruned_loss=0.09843, over 19599.00 frames. ], tot_loss[loss=0.2877, simple_loss=0.3471, pruned_loss=0.1142, over 3816453.55 frames. ], batch size: 52, lr: 1.65e-02, grad_scale: 8.0 +2023-04-01 05:07:58,269 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=29969.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:08:08,357 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 05:08:29,809 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=29994.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:08:36,443 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-30000.pt +2023-04-01 05:08:50,701 INFO [train.py:903] (0/4) Epoch 5, batch 2700, loss[loss=0.279, simple_loss=0.3511, pruned_loss=0.1034, over 19666.00 frames. ], tot_loss[loss=0.2871, simple_loss=0.3462, pruned_loss=0.114, over 3809458.75 frames. ], batch size: 55, lr: 1.64e-02, grad_scale: 8.0 +2023-04-01 05:09:04,697 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30024.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:09:10,274 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30028.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:09:17,306 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.255e+02 6.674e+02 8.375e+02 9.964e+02 1.932e+03, threshold=1.675e+03, percent-clipped=1.0 +2023-04-01 05:09:49,950 INFO [train.py:903] (0/4) Epoch 5, batch 2750, loss[loss=0.2557, simple_loss=0.3157, pruned_loss=0.09792, over 19481.00 frames. ], tot_loss[loss=0.2854, simple_loss=0.3446, pruned_loss=0.1131, over 3812991.49 frames. ], batch size: 49, lr: 1.64e-02, grad_scale: 4.0 +2023-04-01 05:10:20,401 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-01 05:10:50,796 INFO [train.py:903] (0/4) Epoch 5, batch 2800, loss[loss=0.3156, simple_loss=0.3746, pruned_loss=0.1283, over 19612.00 frames. ], tot_loss[loss=0.2863, simple_loss=0.3455, pruned_loss=0.1136, over 3809537.60 frames. ], batch size: 57, lr: 1.64e-02, grad_scale: 8.0 +2023-04-01 05:11:17,048 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.279e+02 7.041e+02 9.072e+02 1.191e+03 2.188e+03, threshold=1.814e+03, percent-clipped=6.0 +2023-04-01 05:11:22,801 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30139.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:11:27,364 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30143.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:11:51,983 INFO [train.py:903] (0/4) Epoch 5, batch 2850, loss[loss=0.339, simple_loss=0.3767, pruned_loss=0.1507, over 13278.00 frames. ], tot_loss[loss=0.2857, simple_loss=0.3446, pruned_loss=0.1134, over 3805378.40 frames. ], batch size: 136, lr: 1.64e-02, grad_scale: 4.0 +2023-04-01 05:11:54,299 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30164.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:12:17,089 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5855, 1.4401, 1.4292, 1.7271, 1.6095, 1.4764, 1.4463, 1.5247], + device='cuda:0'), covar=tensor([0.0681, 0.1026, 0.0944, 0.0544, 0.0732, 0.0427, 0.0745, 0.0488], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0371, 0.0287, 0.0246, 0.0309, 0.0255, 0.0276, 0.0234], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:12:51,657 INFO [train.py:903] (0/4) Epoch 5, batch 2900, loss[loss=0.2719, simple_loss=0.3423, pruned_loss=0.1008, over 19611.00 frames. ], tot_loss[loss=0.2857, simple_loss=0.3449, pruned_loss=0.1132, over 3808450.81 frames. ], batch size: 57, lr: 1.64e-02, grad_scale: 4.0 +2023-04-01 05:12:51,692 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 05:12:53,196 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7352, 1.6057, 2.1218, 1.8696, 3.0575, 2.4519, 3.0932, 1.9770], + device='cuda:0'), covar=tensor([0.1514, 0.2617, 0.1499, 0.1324, 0.1004, 0.1314, 0.1183, 0.2190], + device='cuda:0'), in_proj_covar=tensor([0.0445, 0.0508, 0.0480, 0.0408, 0.0559, 0.0443, 0.0630, 0.0445], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:13:09,289 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30227.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:13:14,407 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7500, 1.8334, 1.4977, 1.3298, 1.3386, 1.4811, 0.1145, 0.7930], + device='cuda:0'), covar=tensor([0.0242, 0.0222, 0.0168, 0.0233, 0.0542, 0.0241, 0.0480, 0.0409], + device='cuda:0'), in_proj_covar=tensor([0.0286, 0.0281, 0.0280, 0.0301, 0.0365, 0.0289, 0.0275, 0.0289], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 05:13:20,144 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.023e+02 7.237e+02 9.091e+02 1.153e+03 2.755e+03, threshold=1.818e+03, percent-clipped=7.0 +2023-04-01 05:13:51,429 INFO [train.py:903] (0/4) Epoch 5, batch 2950, loss[loss=0.3084, simple_loss=0.3621, pruned_loss=0.1273, over 18854.00 frames. ], tot_loss[loss=0.2865, simple_loss=0.3457, pruned_loss=0.1136, over 3804197.01 frames. ], batch size: 74, lr: 1.64e-02, grad_scale: 4.0 +2023-04-01 05:14:12,808 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30279.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:14:18,203 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30284.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:14:46,074 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30309.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:14:50,603 INFO [train.py:903] (0/4) Epoch 5, batch 3000, loss[loss=0.2502, simple_loss=0.307, pruned_loss=0.09668, over 19351.00 frames. ], tot_loss[loss=0.2879, simple_loss=0.3462, pruned_loss=0.1148, over 3805226.02 frames. ], batch size: 47, lr: 1.64e-02, grad_scale: 4.0 +2023-04-01 05:14:50,604 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 05:15:03,127 INFO [train.py:937] (0/4) Epoch 5, validation: loss=0.2047, simple_loss=0.3034, pruned_loss=0.05296, over 944034.00 frames. +2023-04-01 05:15:03,128 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 05:15:05,715 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 05:15:33,552 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.690e+02 7.155e+02 8.736e+02 1.085e+03 2.346e+03, threshold=1.747e+03, percent-clipped=4.0 +2023-04-01 05:16:06,357 INFO [train.py:903] (0/4) Epoch 5, batch 3050, loss[loss=0.3837, simple_loss=0.4022, pruned_loss=0.1826, over 13340.00 frames. ], tot_loss[loss=0.2871, simple_loss=0.3461, pruned_loss=0.1141, over 3814815.78 frames. ], batch size: 136, lr: 1.64e-02, grad_scale: 4.0 +2023-04-01 05:16:26,373 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30378.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:16:45,458 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30395.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:16:57,139 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7618, 1.5919, 1.7898, 1.9661, 4.1926, 1.1254, 2.2088, 4.2016], + device='cuda:0'), covar=tensor([0.0313, 0.2367, 0.2439, 0.1509, 0.0511, 0.2417, 0.1370, 0.0324], + device='cuda:0'), in_proj_covar=tensor([0.0289, 0.0313, 0.0314, 0.0290, 0.0311, 0.0319, 0.0292, 0.0306], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:17:07,570 INFO [train.py:903] (0/4) Epoch 5, batch 3100, loss[loss=0.3571, simple_loss=0.3942, pruned_loss=0.16, over 19506.00 frames. ], tot_loss[loss=0.2875, simple_loss=0.3465, pruned_loss=0.1143, over 3812366.00 frames. ], batch size: 64, lr: 1.63e-02, grad_scale: 4.0 +2023-04-01 05:17:17,061 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30420.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:17:33,697 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.705e+02 6.847e+02 8.274e+02 1.001e+03 3.134e+03, threshold=1.655e+03, percent-clipped=2.0 +2023-04-01 05:18:06,709 INFO [train.py:903] (0/4) Epoch 5, batch 3150, loss[loss=0.2443, simple_loss=0.3158, pruned_loss=0.0864, over 19586.00 frames. ], tot_loss[loss=0.288, simple_loss=0.3461, pruned_loss=0.115, over 3803169.13 frames. ], batch size: 52, lr: 1.63e-02, grad_scale: 4.0 +2023-04-01 05:18:34,302 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 05:18:37,220 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30487.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:19:06,289 INFO [train.py:903] (0/4) Epoch 5, batch 3200, loss[loss=0.314, simple_loss=0.3613, pruned_loss=0.1333, over 13467.00 frames. ], tot_loss[loss=0.288, simple_loss=0.3463, pruned_loss=0.1149, over 3799671.60 frames. ], batch size: 135, lr: 1.63e-02, grad_scale: 8.0 +2023-04-01 05:19:30,541 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.93 vs. limit=5.0 +2023-04-01 05:19:35,546 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.065e+02 7.005e+02 8.588e+02 1.128e+03 3.335e+03, threshold=1.718e+03, percent-clipped=13.0 +2023-04-01 05:19:35,978 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30535.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:20:04,445 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30560.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:20:06,327 INFO [train.py:903] (0/4) Epoch 5, batch 3250, loss[loss=0.3371, simple_loss=0.3846, pruned_loss=0.1447, over 19530.00 frames. ], tot_loss[loss=0.288, simple_loss=0.3462, pruned_loss=0.1149, over 3813924.97 frames. ], batch size: 54, lr: 1.63e-02, grad_scale: 8.0 +2023-04-01 05:20:19,561 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30571.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:20:35,356 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-01 05:20:55,331 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30602.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:20:59,221 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7880, 1.2637, 1.3153, 1.6822, 1.5120, 1.4844, 1.4009, 1.6293], + device='cuda:0'), covar=tensor([0.0790, 0.1253, 0.1241, 0.0803, 0.0936, 0.0470, 0.0912, 0.0612], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0364, 0.0278, 0.0235, 0.0301, 0.0245, 0.0266, 0.0230], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:21:08,680 INFO [train.py:903] (0/4) Epoch 5, batch 3300, loss[loss=0.3505, simple_loss=0.3958, pruned_loss=0.1526, over 19621.00 frames. ], tot_loss[loss=0.2883, simple_loss=0.3467, pruned_loss=0.115, over 3803473.17 frames. ], batch size: 60, lr: 1.63e-02, grad_scale: 8.0 +2023-04-01 05:21:16,490 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 05:21:35,516 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.601e+02 6.699e+02 7.755e+02 9.198e+02 2.155e+03, threshold=1.551e+03, percent-clipped=1.0 +2023-04-01 05:21:35,952 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1503, 1.9232, 1.4671, 1.2267, 1.7765, 0.9799, 1.0818, 1.6175], + device='cuda:0'), covar=tensor([0.0503, 0.0413, 0.0811, 0.0482, 0.0307, 0.0949, 0.0415, 0.0247], + device='cuda:0'), in_proj_covar=tensor([0.0284, 0.0271, 0.0316, 0.0242, 0.0220, 0.0313, 0.0284, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:21:46,021 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1331, 2.2549, 2.0340, 3.3365, 2.0865, 3.4102, 3.0317, 2.0125], + device='cuda:0'), covar=tensor([0.2118, 0.1645, 0.0853, 0.0957, 0.2116, 0.0539, 0.1549, 0.1439], + device='cuda:0'), in_proj_covar=tensor([0.0618, 0.0615, 0.0552, 0.0770, 0.0663, 0.0527, 0.0676, 0.0574], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:22:09,093 INFO [train.py:903] (0/4) Epoch 5, batch 3350, loss[loss=0.2812, simple_loss=0.3496, pruned_loss=0.1064, over 19673.00 frames. ], tot_loss[loss=0.2863, simple_loss=0.3452, pruned_loss=0.1136, over 3806883.00 frames. ], batch size: 60, lr: 1.63e-02, grad_scale: 8.0 +2023-04-01 05:22:21,622 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30673.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:22:38,263 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30686.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:23:09,773 INFO [train.py:903] (0/4) Epoch 5, batch 3400, loss[loss=0.2587, simple_loss=0.3268, pruned_loss=0.09533, over 19668.00 frames. ], tot_loss[loss=0.2871, simple_loss=0.3455, pruned_loss=0.1143, over 3786438.40 frames. ], batch size: 53, lr: 1.63e-02, grad_scale: 8.0 +2023-04-01 05:23:22,328 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=30722.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:23:35,530 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8730, 1.5194, 1.5103, 2.0717, 1.8105, 1.5827, 1.6608, 1.8630], + device='cuda:0'), covar=tensor([0.0791, 0.1692, 0.1242, 0.0843, 0.1079, 0.0506, 0.0885, 0.0601], + device='cuda:0'), in_proj_covar=tensor([0.0250, 0.0361, 0.0277, 0.0234, 0.0300, 0.0243, 0.0266, 0.0230], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:23:39,931 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.254e+02 6.893e+02 8.724e+02 1.073e+03 2.213e+03, threshold=1.745e+03, percent-clipped=7.0 +2023-04-01 05:24:11,822 INFO [train.py:903] (0/4) Epoch 5, batch 3450, loss[loss=0.3161, simple_loss=0.3719, pruned_loss=0.1301, over 19361.00 frames. ], tot_loss[loss=0.2877, simple_loss=0.3461, pruned_loss=0.1146, over 3789220.74 frames. ], batch size: 66, lr: 1.63e-02, grad_scale: 8.0 +2023-04-01 05:24:15,158 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 05:24:27,413 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1122, 1.8830, 1.4235, 1.2293, 1.7165, 1.0510, 0.9878, 1.5825], + device='cuda:0'), covar=tensor([0.0538, 0.0422, 0.0752, 0.0482, 0.0285, 0.0796, 0.0498, 0.0265], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0265, 0.0310, 0.0240, 0.0218, 0.0306, 0.0279, 0.0250], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:25:04,229 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-04-01 05:25:13,755 INFO [train.py:903] (0/4) Epoch 5, batch 3500, loss[loss=0.2628, simple_loss=0.3412, pruned_loss=0.09218, over 19342.00 frames. ], tot_loss[loss=0.2849, simple_loss=0.344, pruned_loss=0.1129, over 3797961.42 frames. ], batch size: 66, lr: 1.62e-02, grad_scale: 8.0 +2023-04-01 05:25:39,223 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.495e+02 7.015e+02 8.195e+02 1.097e+03 2.546e+03, threshold=1.639e+03, percent-clipped=5.0 +2023-04-01 05:25:41,894 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=30837.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:25:50,112 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9499, 1.2865, 0.9746, 0.8705, 1.1988, 0.9110, 0.7146, 1.2422], + device='cuda:0'), covar=tensor([0.0516, 0.0596, 0.0932, 0.0518, 0.0400, 0.0941, 0.0636, 0.0337], + device='cuda:0'), in_proj_covar=tensor([0.0286, 0.0270, 0.0316, 0.0244, 0.0219, 0.0312, 0.0287, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:26:09,305 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30858.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:26:13,372 INFO [train.py:903] (0/4) Epoch 5, batch 3550, loss[loss=0.2835, simple_loss=0.3441, pruned_loss=0.1114, over 19297.00 frames. ], tot_loss[loss=0.2868, simple_loss=0.3458, pruned_loss=0.1139, over 3807538.47 frames. ], batch size: 66, lr: 1.62e-02, grad_scale: 8.0 +2023-04-01 05:26:37,866 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30883.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:27:13,895 INFO [train.py:903] (0/4) Epoch 5, batch 3600, loss[loss=0.3155, simple_loss=0.3622, pruned_loss=0.1344, over 19785.00 frames. ], tot_loss[loss=0.2868, simple_loss=0.3461, pruned_loss=0.1138, over 3822350.32 frames. ], batch size: 56, lr: 1.62e-02, grad_scale: 8.0 +2023-04-01 05:27:24,174 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30921.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:27:33,968 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2788, 1.2361, 1.4722, 0.7677, 2.3102, 2.8822, 2.6568, 2.9574], + device='cuda:0'), covar=tensor([0.1245, 0.2782, 0.2823, 0.2026, 0.0420, 0.0172, 0.0245, 0.0160], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0280, 0.0318, 0.0252, 0.0196, 0.0119, 0.0205, 0.0136], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 05:27:43,240 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.778e+02 6.769e+02 8.289e+02 1.066e+03 2.218e+03, threshold=1.658e+03, percent-clipped=4.0 +2023-04-01 05:27:51,217 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=30942.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:27:52,165 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30943.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:28:02,186 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=30952.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:28:13,490 INFO [train.py:903] (0/4) Epoch 5, batch 3650, loss[loss=0.3178, simple_loss=0.3689, pruned_loss=0.1334, over 19378.00 frames. ], tot_loss[loss=0.2869, simple_loss=0.346, pruned_loss=0.1139, over 3816240.54 frames. ], batch size: 70, lr: 1.62e-02, grad_scale: 8.0 +2023-04-01 05:28:20,113 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=30967.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:29:14,323 INFO [train.py:903] (0/4) Epoch 5, batch 3700, loss[loss=0.2991, simple_loss=0.354, pruned_loss=0.1221, over 19507.00 frames. ], tot_loss[loss=0.2881, simple_loss=0.3468, pruned_loss=0.1147, over 3820951.34 frames. ], batch size: 64, lr: 1.62e-02, grad_scale: 8.0 +2023-04-01 05:29:21,016 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31017.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:29:27,635 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1534, 1.2490, 1.7453, 1.4553, 2.1039, 1.7826, 2.1325, 0.7708], + device='cuda:0'), covar=tensor([0.1845, 0.2878, 0.1479, 0.1448, 0.1153, 0.1648, 0.1369, 0.2816], + device='cuda:0'), in_proj_covar=tensor([0.0443, 0.0504, 0.0481, 0.0406, 0.0549, 0.0442, 0.0625, 0.0446], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:29:34,011 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31029.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 05:29:40,640 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.405e+02 7.099e+02 8.962e+02 1.140e+03 3.223e+03, threshold=1.792e+03, percent-clipped=9.0 +2023-04-01 05:30:15,308 INFO [train.py:903] (0/4) Epoch 5, batch 3750, loss[loss=0.2809, simple_loss=0.3482, pruned_loss=0.1068, over 19297.00 frames. ], tot_loss[loss=0.286, simple_loss=0.345, pruned_loss=0.1135, over 3820974.14 frames. ], batch size: 66, lr: 1.62e-02, grad_scale: 8.0 +2023-04-01 05:30:53,657 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31093.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:31:15,743 INFO [train.py:903] (0/4) Epoch 5, batch 3800, loss[loss=0.3161, simple_loss=0.3737, pruned_loss=0.1292, over 19331.00 frames. ], tot_loss[loss=0.2858, simple_loss=0.3451, pruned_loss=0.1133, over 3816063.96 frames. ], batch size: 70, lr: 1.62e-02, grad_scale: 8.0 +2023-04-01 05:31:22,810 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31118.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:31:40,724 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31132.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:31:44,605 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.073e+02 5.856e+02 8.467e+02 1.115e+03 2.554e+03, threshold=1.693e+03, percent-clipped=5.0 +2023-04-01 05:31:46,154 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5026, 1.9325, 1.4782, 1.5266, 1.8248, 1.2371, 1.2991, 1.5559], + device='cuda:0'), covar=tensor([0.0499, 0.0396, 0.0596, 0.0402, 0.0262, 0.0648, 0.0428, 0.0308], + device='cuda:0'), in_proj_covar=tensor([0.0286, 0.0271, 0.0316, 0.0240, 0.0221, 0.0312, 0.0283, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:31:49,130 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 05:32:15,367 INFO [train.py:903] (0/4) Epoch 5, batch 3850, loss[loss=0.2521, simple_loss=0.3112, pruned_loss=0.09647, over 19039.00 frames. ], tot_loss[loss=0.2847, simple_loss=0.3441, pruned_loss=0.1126, over 3807614.54 frames. ], batch size: 42, lr: 1.61e-02, grad_scale: 8.0 +2023-04-01 05:32:34,617 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.60 vs. limit=5.0 +2023-04-01 05:33:19,169 INFO [train.py:903] (0/4) Epoch 5, batch 3900, loss[loss=0.2309, simple_loss=0.2945, pruned_loss=0.0837, over 19791.00 frames. ], tot_loss[loss=0.283, simple_loss=0.3428, pruned_loss=0.1116, over 3813112.51 frames. ], batch size: 47, lr: 1.61e-02, grad_scale: 8.0 +2023-04-01 05:33:45,289 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.653e+02 6.709e+02 8.069e+02 1.055e+03 2.198e+03, threshold=1.614e+03, percent-clipped=3.0 +2023-04-01 05:34:18,959 INFO [train.py:903] (0/4) Epoch 5, batch 3950, loss[loss=0.338, simple_loss=0.3897, pruned_loss=0.1432, over 19768.00 frames. ], tot_loss[loss=0.2822, simple_loss=0.3422, pruned_loss=0.1111, over 3821302.60 frames. ], batch size: 54, lr: 1.61e-02, grad_scale: 4.0 +2023-04-01 05:34:22,387 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31265.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:34:24,564 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 05:34:47,793 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31287.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:34:52,307 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8611, 1.8755, 1.8015, 2.8736, 1.9424, 2.6013, 2.6065, 1.8093], + device='cuda:0'), covar=tensor([0.2125, 0.1696, 0.0894, 0.0905, 0.1864, 0.0684, 0.1599, 0.1631], + device='cuda:0'), in_proj_covar=tensor([0.0622, 0.0613, 0.0553, 0.0767, 0.0659, 0.0529, 0.0676, 0.0580], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:35:00,482 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31296.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:35:18,263 INFO [train.py:903] (0/4) Epoch 5, batch 4000, loss[loss=0.3295, simple_loss=0.3778, pruned_loss=0.1406, over 19587.00 frames. ], tot_loss[loss=0.284, simple_loss=0.3435, pruned_loss=0.1122, over 3822176.49 frames. ], batch size: 57, lr: 1.61e-02, grad_scale: 8.0 +2023-04-01 05:35:48,960 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.459e+02 7.061e+02 8.767e+02 1.081e+03 2.366e+03, threshold=1.753e+03, percent-clipped=7.0 +2023-04-01 05:36:06,106 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 05:36:18,325 INFO [train.py:903] (0/4) Epoch 5, batch 4050, loss[loss=0.257, simple_loss=0.3173, pruned_loss=0.09833, over 19788.00 frames. ], tot_loss[loss=0.2844, simple_loss=0.3439, pruned_loss=0.1125, over 3811102.74 frames. ], batch size: 48, lr: 1.61e-02, grad_scale: 4.0 +2023-04-01 05:36:26,545 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1702, 2.7727, 1.9032, 2.1287, 1.8087, 2.2451, 0.7252, 2.0487], + device='cuda:0'), covar=tensor([0.0272, 0.0231, 0.0293, 0.0384, 0.0471, 0.0402, 0.0579, 0.0483], + device='cuda:0'), in_proj_covar=tensor([0.0285, 0.0280, 0.0282, 0.0301, 0.0375, 0.0293, 0.0277, 0.0292], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 05:36:33,369 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31373.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 05:36:42,165 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31380.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:36:51,300 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31388.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:37:04,926 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4055, 2.0526, 1.8279, 1.6676, 1.4530, 1.7660, 0.3343, 1.2594], + device='cuda:0'), covar=tensor([0.0225, 0.0268, 0.0191, 0.0306, 0.0531, 0.0297, 0.0511, 0.0383], + device='cuda:0'), in_proj_covar=tensor([0.0285, 0.0282, 0.0283, 0.0303, 0.0377, 0.0294, 0.0278, 0.0292], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 05:37:07,163 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31402.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:37:18,594 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31411.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:37:19,401 INFO [train.py:903] (0/4) Epoch 5, batch 4100, loss[loss=0.2531, simple_loss=0.3228, pruned_loss=0.09167, over 19568.00 frames. ], tot_loss[loss=0.2842, simple_loss=0.344, pruned_loss=0.1122, over 3812082.64 frames. ], batch size: 52, lr: 1.61e-02, grad_scale: 4.0 +2023-04-01 05:37:20,890 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31413.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:37:37,684 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31427.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:37:48,918 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.057e+02 5.964e+02 7.397e+02 9.605e+02 3.908e+03, threshold=1.479e+03, percent-clipped=5.0 +2023-04-01 05:37:53,727 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 05:38:20,925 INFO [train.py:903] (0/4) Epoch 5, batch 4150, loss[loss=0.2722, simple_loss=0.3379, pruned_loss=0.1032, over 19326.00 frames. ], tot_loss[loss=0.2837, simple_loss=0.3438, pruned_loss=0.1118, over 3804753.61 frames. ], batch size: 66, lr: 1.61e-02, grad_scale: 4.0 +2023-04-01 05:38:26,188 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0 +2023-04-01 05:38:50,977 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31488.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 05:39:16,850 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31509.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 05:39:20,061 INFO [train.py:903] (0/4) Epoch 5, batch 4200, loss[loss=0.3089, simple_loss=0.3588, pruned_loss=0.1295, over 19495.00 frames. ], tot_loss[loss=0.2818, simple_loss=0.3421, pruned_loss=0.1107, over 3818550.60 frames. ], batch size: 49, lr: 1.61e-02, grad_scale: 4.0 +2023-04-01 05:39:23,674 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 05:39:50,936 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.017e+02 6.856e+02 8.101e+02 1.045e+03 3.023e+03, threshold=1.620e+03, percent-clipped=6.0 +2023-04-01 05:40:15,021 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1381, 3.9624, 2.3595, 2.6180, 3.4605, 1.7373, 1.1993, 2.0181], + device='cuda:0'), covar=tensor([0.0778, 0.0270, 0.0626, 0.0492, 0.0255, 0.0893, 0.0853, 0.0507], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0266, 0.0315, 0.0235, 0.0217, 0.0309, 0.0282, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:40:19,241 INFO [train.py:903] (0/4) Epoch 5, batch 4250, loss[loss=0.36, simple_loss=0.3999, pruned_loss=0.16, over 17512.00 frames. ], tot_loss[loss=0.2812, simple_loss=0.3417, pruned_loss=0.1104, over 3814178.00 frames. ], batch size: 101, lr: 1.60e-02, grad_scale: 4.0 +2023-04-01 05:40:35,452 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 05:40:44,450 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9209, 1.9596, 1.8888, 2.8817, 1.9795, 2.6925, 2.5977, 1.8178], + device='cuda:0'), covar=tensor([0.1926, 0.1602, 0.0820, 0.0831, 0.1765, 0.0601, 0.1473, 0.1423], + device='cuda:0'), in_proj_covar=tensor([0.0619, 0.0612, 0.0553, 0.0768, 0.0656, 0.0529, 0.0681, 0.0579], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:40:46,345 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 05:41:20,811 INFO [train.py:903] (0/4) Epoch 5, batch 4300, loss[loss=0.2632, simple_loss=0.3352, pruned_loss=0.0956, over 18784.00 frames. ], tot_loss[loss=0.2811, simple_loss=0.3413, pruned_loss=0.1105, over 3805214.41 frames. ], batch size: 74, lr: 1.60e-02, grad_scale: 4.0 +2023-04-01 05:41:35,094 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6873, 1.3626, 1.4526, 1.7424, 1.6125, 1.5634, 1.4655, 1.6434], + device='cuda:0'), covar=tensor([0.0856, 0.1591, 0.1231, 0.0825, 0.1063, 0.0508, 0.1024, 0.0654], + device='cuda:0'), in_proj_covar=tensor([0.0248, 0.0363, 0.0277, 0.0233, 0.0300, 0.0240, 0.0266, 0.0226], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:41:51,080 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31636.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:41:51,772 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.826e+02 7.144e+02 8.425e+02 1.091e+03 2.021e+03, threshold=1.685e+03, percent-clipped=5.0 +2023-04-01 05:42:13,881 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 05:42:18,582 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31658.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:42:21,711 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31661.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:42:22,335 INFO [train.py:903] (0/4) Epoch 5, batch 4350, loss[loss=0.3268, simple_loss=0.3696, pruned_loss=0.142, over 19852.00 frames. ], tot_loss[loss=0.2816, simple_loss=0.3416, pruned_loss=0.1109, over 3811683.91 frames. ], batch size: 52, lr: 1.60e-02, grad_scale: 4.0 +2023-04-01 05:42:28,415 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31667.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:42:39,497 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7764, 3.0913, 3.2436, 3.2479, 1.1474, 3.0211, 2.7490, 2.9262], + device='cuda:0'), covar=tensor([0.1050, 0.0790, 0.0798, 0.0602, 0.3789, 0.0490, 0.0658, 0.1279], + device='cuda:0'), in_proj_covar=tensor([0.0505, 0.0443, 0.0590, 0.0490, 0.0575, 0.0358, 0.0388, 0.0553], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 05:42:45,917 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31683.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:42:53,936 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-01 05:42:57,763 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31692.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:43:03,543 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.64 vs. limit=5.0 +2023-04-01 05:43:22,083 INFO [train.py:903] (0/4) Epoch 5, batch 4400, loss[loss=0.2299, simple_loss=0.2916, pruned_loss=0.08404, over 19345.00 frames. ], tot_loss[loss=0.283, simple_loss=0.3423, pruned_loss=0.1118, over 3795660.72 frames. ], batch size: 47, lr: 1.60e-02, grad_scale: 8.0 +2023-04-01 05:43:28,260 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6387, 2.5273, 1.8408, 1.8585, 1.7893, 2.0484, 0.9740, 1.9411], + device='cuda:0'), covar=tensor([0.0289, 0.0280, 0.0278, 0.0415, 0.0468, 0.0379, 0.0536, 0.0430], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0281, 0.0279, 0.0301, 0.0375, 0.0293, 0.0275, 0.0289], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 05:43:33,977 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31722.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:43:44,927 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 05:43:53,319 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.183e+02 7.191e+02 8.879e+02 1.082e+03 1.961e+03, threshold=1.776e+03, percent-clipped=1.0 +2023-04-01 05:43:55,417 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 05:44:02,701 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=31744.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 05:44:22,696 INFO [train.py:903] (0/4) Epoch 5, batch 4450, loss[loss=0.3098, simple_loss=0.3623, pruned_loss=0.1287, over 19308.00 frames. ], tot_loss[loss=0.2846, simple_loss=0.344, pruned_loss=0.1126, over 3809518.99 frames. ], batch size: 66, lr: 1.60e-02, grad_scale: 8.0 +2023-04-01 05:44:30,926 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=31769.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 05:44:32,872 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31771.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:44:40,335 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8777, 1.9054, 1.9099, 3.0206, 1.9674, 2.9336, 2.8146, 1.9819], + device='cuda:0'), covar=tensor([0.2213, 0.1788, 0.0897, 0.0939, 0.2061, 0.0649, 0.1530, 0.1539], + device='cuda:0'), in_proj_covar=tensor([0.0622, 0.0613, 0.0554, 0.0770, 0.0664, 0.0533, 0.0678, 0.0584], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:45:16,689 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31806.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:45:23,178 INFO [train.py:903] (0/4) Epoch 5, batch 4500, loss[loss=0.2606, simple_loss=0.3176, pruned_loss=0.1018, over 19386.00 frames. ], tot_loss[loss=0.2837, simple_loss=0.3433, pruned_loss=0.1121, over 3818752.52 frames. ], batch size: 47, lr: 1.60e-02, grad_scale: 8.0 +2023-04-01 05:45:31,277 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-01 05:45:53,690 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.229e+02 6.666e+02 8.776e+02 1.149e+03 2.550e+03, threshold=1.755e+03, percent-clipped=7.0 +2023-04-01 05:45:54,070 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9237, 1.9385, 1.4796, 1.4583, 1.3348, 1.4176, 0.4228, 1.1533], + device='cuda:0'), covar=tensor([0.0324, 0.0322, 0.0281, 0.0375, 0.0630, 0.0427, 0.0602, 0.0471], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0280, 0.0279, 0.0301, 0.0374, 0.0293, 0.0273, 0.0289], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 05:45:58,550 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31841.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:46:12,844 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=31853.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 05:46:24,490 INFO [train.py:903] (0/4) Epoch 5, batch 4550, loss[loss=0.3264, simple_loss=0.3684, pruned_loss=0.1422, over 19680.00 frames. ], tot_loss[loss=0.285, simple_loss=0.3441, pruned_loss=0.1129, over 3817643.72 frames. ], batch size: 59, lr: 1.60e-02, grad_scale: 8.0 +2023-04-01 05:46:30,257 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 05:46:51,956 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 05:46:52,338 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31886.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:47:04,882 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-04-01 05:47:05,643 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31896.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:47:25,666 INFO [train.py:903] (0/4) Epoch 5, batch 4600, loss[loss=0.2546, simple_loss=0.3163, pruned_loss=0.09647, over 19778.00 frames. ], tot_loss[loss=0.2835, simple_loss=0.3427, pruned_loss=0.1121, over 3808094.05 frames. ], batch size: 47, lr: 1.60e-02, grad_scale: 8.0 +2023-04-01 05:47:49,549 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31933.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:47:55,745 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.582e+02 6.775e+02 7.918e+02 9.900e+02 3.222e+03, threshold=1.584e+03, percent-clipped=3.0 +2023-04-01 05:48:25,674 INFO [train.py:903] (0/4) Epoch 5, batch 4650, loss[loss=0.3316, simple_loss=0.3948, pruned_loss=0.1341, over 19532.00 frames. ], tot_loss[loss=0.2846, simple_loss=0.3439, pruned_loss=0.1126, over 3801359.44 frames. ], batch size: 56, lr: 1.60e-02, grad_scale: 8.0 +2023-04-01 05:48:32,740 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=31968.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:48:32,807 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=31968.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 05:48:38,525 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1823, 1.1167, 1.4164, 1.3063, 1.7523, 1.7126, 1.7949, 0.4639], + device='cuda:0'), covar=tensor([0.1645, 0.2903, 0.1514, 0.1453, 0.1046, 0.1550, 0.1013, 0.2757], + device='cuda:0'), in_proj_covar=tensor([0.0440, 0.0502, 0.0477, 0.0405, 0.0549, 0.0445, 0.0618, 0.0440], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:48:41,065 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 05:48:52,756 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 05:49:11,288 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-32000.pt +2023-04-01 05:49:24,516 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9770, 1.8734, 2.5161, 2.5418, 2.6163, 2.5770, 2.2385, 2.8624], + device='cuda:0'), covar=tensor([0.0598, 0.1656, 0.1015, 0.0807, 0.0966, 0.0382, 0.0856, 0.0419], + device='cuda:0'), in_proj_covar=tensor([0.0249, 0.0361, 0.0279, 0.0237, 0.0299, 0.0239, 0.0265, 0.0229], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:49:25,340 INFO [train.py:903] (0/4) Epoch 5, batch 4700, loss[loss=0.2451, simple_loss=0.3085, pruned_loss=0.09084, over 19389.00 frames. ], tot_loss[loss=0.2866, simple_loss=0.3456, pruned_loss=0.1138, over 3797962.10 frames. ], batch size: 47, lr: 1.59e-02, grad_scale: 8.0 +2023-04-01 05:49:48,441 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 05:49:56,506 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.386e+02 6.540e+02 8.637e+02 1.061e+03 2.519e+03, threshold=1.727e+03, percent-clipped=5.0 +2023-04-01 05:50:07,060 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4850, 0.9888, 1.1838, 1.2467, 2.1242, 0.9571, 1.8825, 2.1201], + device='cuda:0'), covar=tensor([0.0561, 0.2554, 0.2412, 0.1459, 0.0741, 0.1898, 0.0906, 0.0603], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0312, 0.0310, 0.0282, 0.0299, 0.0310, 0.0285, 0.0300], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:50:27,317 INFO [train.py:903] (0/4) Epoch 5, batch 4750, loss[loss=0.3271, simple_loss=0.3796, pruned_loss=0.1373, over 19468.00 frames. ], tot_loss[loss=0.2852, simple_loss=0.3451, pruned_loss=0.1126, over 3807124.78 frames. ], batch size: 64, lr: 1.59e-02, grad_scale: 8.0 +2023-04-01 05:50:32,825 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32066.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:50:38,990 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7144, 1.7536, 1.7379, 2.3977, 1.4491, 2.2222, 2.2367, 1.8003], + device='cuda:0'), covar=tensor([0.1859, 0.1550, 0.0860, 0.0841, 0.1888, 0.0693, 0.1558, 0.1449], + device='cuda:0'), in_proj_covar=tensor([0.0626, 0.0615, 0.0550, 0.0774, 0.0662, 0.0532, 0.0680, 0.0581], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:51:28,010 INFO [train.py:903] (0/4) Epoch 5, batch 4800, loss[loss=0.2863, simple_loss=0.3501, pruned_loss=0.1112, over 19291.00 frames. ], tot_loss[loss=0.2848, simple_loss=0.3447, pruned_loss=0.1125, over 3812084.94 frames. ], batch size: 66, lr: 1.59e-02, grad_scale: 8.0 +2023-04-01 05:51:57,642 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.741e+02 7.051e+02 8.763e+02 1.042e+03 3.094e+03, threshold=1.753e+03, percent-clipped=4.0 +2023-04-01 05:52:04,790 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32142.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:52:14,227 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32150.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:52:27,936 INFO [train.py:903] (0/4) Epoch 5, batch 4850, loss[loss=0.3365, simple_loss=0.3924, pruned_loss=0.1403, over 19789.00 frames. ], tot_loss[loss=0.2857, simple_loss=0.345, pruned_loss=0.1132, over 3808102.86 frames. ], batch size: 56, lr: 1.59e-02, grad_scale: 8.0 +2023-04-01 05:52:34,156 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32167.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:52:45,383 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8456, 1.8583, 1.7826, 2.7296, 1.7855, 2.6957, 2.4674, 1.8779], + device='cuda:0'), covar=tensor([0.2037, 0.1623, 0.0881, 0.0901, 0.1843, 0.0625, 0.1627, 0.1455], + device='cuda:0'), in_proj_covar=tensor([0.0625, 0.0614, 0.0550, 0.0770, 0.0659, 0.0533, 0.0679, 0.0578], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 05:52:51,460 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32181.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:52:54,108 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 05:52:56,527 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32185.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:53:12,906 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 05:53:17,712 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 05:53:18,683 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 05:53:26,876 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 05:53:27,946 INFO [train.py:903] (0/4) Epoch 5, batch 4900, loss[loss=0.2623, simple_loss=0.3342, pruned_loss=0.09516, over 19672.00 frames. ], tot_loss[loss=0.2842, simple_loss=0.3441, pruned_loss=0.1121, over 3816462.20 frames. ], batch size: 55, lr: 1.59e-02, grad_scale: 8.0 +2023-04-01 05:53:44,427 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32224.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 05:53:48,965 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 05:53:59,194 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.307e+02 6.585e+02 7.912e+02 1.039e+03 2.328e+03, threshold=1.582e+03, percent-clipped=1.0 +2023-04-01 05:54:02,588 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32240.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:54:13,268 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32249.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 05:54:29,207 INFO [train.py:903] (0/4) Epoch 5, batch 4950, loss[loss=0.2544, simple_loss=0.3192, pruned_loss=0.09485, over 19731.00 frames. ], tot_loss[loss=0.282, simple_loss=0.3424, pruned_loss=0.1108, over 3834760.16 frames. ], batch size: 51, lr: 1.59e-02, grad_scale: 8.0 +2023-04-01 05:54:33,924 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32265.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:54:48,143 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32277.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:54:49,213 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 05:55:12,831 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 05:55:15,535 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32300.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:55:30,488 INFO [train.py:903] (0/4) Epoch 5, batch 5000, loss[loss=0.3158, simple_loss=0.3624, pruned_loss=0.1346, over 19780.00 frames. ], tot_loss[loss=0.2818, simple_loss=0.3423, pruned_loss=0.1106, over 3841079.65 frames. ], batch size: 54, lr: 1.59e-02, grad_scale: 8.0 +2023-04-01 05:55:30,639 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=32312.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:55:40,413 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 05:55:50,535 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 05:55:58,317 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.834e+02 7.030e+02 8.789e+02 1.165e+03 2.289e+03, threshold=1.758e+03, percent-clipped=7.0 +2023-04-01 05:56:21,773 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32355.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:56:29,553 INFO [train.py:903] (0/4) Epoch 5, batch 5050, loss[loss=0.2821, simple_loss=0.3422, pruned_loss=0.111, over 18825.00 frames. ], tot_loss[loss=0.2817, simple_loss=0.3425, pruned_loss=0.1105, over 3843172.70 frames. ], batch size: 74, lr: 1.59e-02, grad_scale: 8.0 +2023-04-01 05:57:05,115 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 05:57:07,784 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32392.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:57:29,759 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-01 05:57:30,263 INFO [train.py:903] (0/4) Epoch 5, batch 5100, loss[loss=0.246, simple_loss=0.3234, pruned_loss=0.08426, over 19312.00 frames. ], tot_loss[loss=0.2807, simple_loss=0.3415, pruned_loss=0.11, over 3833550.38 frames. ], batch size: 66, lr: 1.58e-02, grad_scale: 8.0 +2023-04-01 05:57:33,080 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8318, 1.3332, 1.5079, 1.8075, 1.5315, 1.4967, 1.5854, 1.7232], + device='cuda:0'), covar=tensor([0.0815, 0.1432, 0.1222, 0.0873, 0.1092, 0.0513, 0.0946, 0.0622], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0359, 0.0282, 0.0240, 0.0304, 0.0242, 0.0268, 0.0229], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:57:40,591 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 05:57:44,911 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 05:57:51,245 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 05:57:51,520 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=32427.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:58:02,216 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.811e+02 6.879e+02 8.267e+02 1.044e+03 2.791e+03, threshold=1.653e+03, percent-clipped=3.0 +2023-04-01 05:58:02,636 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32437.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:58:16,439 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2711, 2.9769, 2.2167, 2.7656, 0.9610, 2.8020, 2.7326, 2.8216], + device='cuda:0'), covar=tensor([0.0910, 0.1264, 0.1765, 0.0882, 0.3578, 0.1040, 0.0879, 0.1055], + device='cuda:0'), in_proj_covar=tensor([0.0347, 0.0310, 0.0360, 0.0284, 0.0352, 0.0301, 0.0286, 0.0308], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 05:58:32,010 INFO [train.py:903] (0/4) Epoch 5, batch 5150, loss[loss=0.3915, simple_loss=0.4173, pruned_loss=0.1829, over 16980.00 frames. ], tot_loss[loss=0.2802, simple_loss=0.3411, pruned_loss=0.1096, over 3828750.57 frames. ], batch size: 100, lr: 1.58e-02, grad_scale: 8.0 +2023-04-01 05:58:32,371 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32462.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:58:36,864 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7639, 1.3494, 1.4075, 1.7625, 1.4676, 1.4950, 1.5665, 1.6083], + device='cuda:0'), covar=tensor([0.0840, 0.1627, 0.1376, 0.0861, 0.1280, 0.0519, 0.0887, 0.0684], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0362, 0.0283, 0.0241, 0.0307, 0.0241, 0.0266, 0.0230], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 05:58:45,282 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 05:59:19,198 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 05:59:33,452 INFO [train.py:903] (0/4) Epoch 5, batch 5200, loss[loss=0.29, simple_loss=0.353, pruned_loss=0.1135, over 19545.00 frames. ], tot_loss[loss=0.2813, simple_loss=0.3422, pruned_loss=0.1103, over 3823153.25 frames. ], batch size: 64, lr: 1.58e-02, grad_scale: 8.0 +2023-04-01 05:59:43,879 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32521.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 05:59:45,953 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 05:59:50,748 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2119, 1.2051, 1.7921, 1.4569, 2.6233, 2.2410, 2.8556, 1.0200], + device='cuda:0'), covar=tensor([0.1942, 0.3397, 0.1831, 0.1558, 0.1308, 0.1579, 0.1376, 0.3100], + device='cuda:0'), in_proj_covar=tensor([0.0448, 0.0507, 0.0479, 0.0407, 0.0554, 0.0448, 0.0622, 0.0437], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 06:00:02,763 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.318e+02 6.175e+02 7.903e+02 1.065e+03 1.799e+03, threshold=1.581e+03, percent-clipped=1.0 +2023-04-01 06:00:15,185 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32546.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:00:26,285 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32556.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:00:29,346 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 06:00:32,702 INFO [train.py:903] (0/4) Epoch 5, batch 5250, loss[loss=0.2604, simple_loss=0.3204, pruned_loss=0.1002, over 19767.00 frames. ], tot_loss[loss=0.2813, simple_loss=0.3422, pruned_loss=0.1102, over 3810555.16 frames. ], batch size: 48, lr: 1.58e-02, grad_scale: 8.0 +2023-04-01 06:00:34,028 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5467, 4.1443, 2.4262, 3.6938, 1.0073, 3.6864, 3.6947, 3.9362], + device='cuda:0'), covar=tensor([0.0604, 0.1151, 0.1938, 0.0738, 0.4346, 0.0848, 0.0785, 0.0649], + device='cuda:0'), in_proj_covar=tensor([0.0352, 0.0314, 0.0365, 0.0287, 0.0359, 0.0305, 0.0290, 0.0313], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 06:00:55,222 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32581.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:01:32,127 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32611.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:01:32,913 INFO [train.py:903] (0/4) Epoch 5, batch 5300, loss[loss=0.3912, simple_loss=0.4085, pruned_loss=0.1869, over 13114.00 frames. ], tot_loss[loss=0.2818, simple_loss=0.3427, pruned_loss=0.1104, over 3814768.54 frames. ], batch size: 135, lr: 1.58e-02, grad_scale: 8.0 +2023-04-01 06:01:52,345 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 06:02:04,545 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32636.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:02:05,310 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.788e+02 6.583e+02 8.041e+02 1.027e+03 2.106e+03, threshold=1.608e+03, percent-clipped=4.0 +2023-04-01 06:02:17,833 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32648.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:02:19,376 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.59 vs. limit=2.0 +2023-04-01 06:02:34,326 INFO [train.py:903] (0/4) Epoch 5, batch 5350, loss[loss=0.2793, simple_loss=0.3498, pruned_loss=0.1044, over 19715.00 frames. ], tot_loss[loss=0.282, simple_loss=0.343, pruned_loss=0.1104, over 3816177.52 frames. ], batch size: 63, lr: 1.58e-02, grad_scale: 8.0 +2023-04-01 06:02:48,887 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32673.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:02:59,137 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2013, 1.3076, 1.0761, 0.9801, 1.0455, 1.0901, 0.0441, 0.4126], + device='cuda:0'), covar=tensor([0.0263, 0.0256, 0.0159, 0.0212, 0.0514, 0.0209, 0.0491, 0.0444], + device='cuda:0'), in_proj_covar=tensor([0.0285, 0.0293, 0.0286, 0.0310, 0.0378, 0.0301, 0.0281, 0.0298], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 06:03:00,320 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=32683.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:03:06,684 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 06:03:31,090 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=32708.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:03:35,125 INFO [train.py:903] (0/4) Epoch 5, batch 5400, loss[loss=0.2889, simple_loss=0.3504, pruned_loss=0.1137, over 19656.00 frames. ], tot_loss[loss=0.2814, simple_loss=0.3425, pruned_loss=0.1102, over 3822800.26 frames. ], batch size: 58, lr: 1.58e-02, grad_scale: 8.0 +2023-04-01 06:03:56,047 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.58 vs. limit=2.0 +2023-04-01 06:04:03,209 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.192e+02 6.673e+02 8.431e+02 1.064e+03 2.658e+03, threshold=1.686e+03, percent-clipped=8.0 +2023-04-01 06:04:34,478 INFO [train.py:903] (0/4) Epoch 5, batch 5450, loss[loss=0.2657, simple_loss=0.3382, pruned_loss=0.09658, over 19650.00 frames. ], tot_loss[loss=0.2818, simple_loss=0.3429, pruned_loss=0.1103, over 3828300.53 frames. ], batch size: 55, lr: 1.58e-02, grad_scale: 8.0 +2023-04-01 06:05:34,683 INFO [train.py:903] (0/4) Epoch 5, batch 5500, loss[loss=0.2552, simple_loss=0.3236, pruned_loss=0.09338, over 19845.00 frames. ], tot_loss[loss=0.2831, simple_loss=0.3438, pruned_loss=0.1113, over 3814501.96 frames. ], batch size: 52, lr: 1.57e-02, grad_scale: 8.0 +2023-04-01 06:05:57,569 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 06:06:05,437 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.564e+02 6.283e+02 7.782e+02 1.000e+03 2.107e+03, threshold=1.556e+03, percent-clipped=4.0 +2023-04-01 06:06:34,470 INFO [train.py:903] (0/4) Epoch 5, batch 5550, loss[loss=0.2518, simple_loss=0.308, pruned_loss=0.09779, over 19380.00 frames. ], tot_loss[loss=0.2816, simple_loss=0.3422, pruned_loss=0.1105, over 3817854.19 frames. ], batch size: 47, lr: 1.57e-02, grad_scale: 8.0 +2023-04-01 06:06:40,877 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 06:06:55,829 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.96 vs. limit=2.0 +2023-04-01 06:07:29,976 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 06:07:36,771 INFO [train.py:903] (0/4) Epoch 5, batch 5600, loss[loss=0.2705, simple_loss=0.347, pruned_loss=0.09698, over 19663.00 frames. ], tot_loss[loss=0.2818, simple_loss=0.3421, pruned_loss=0.1108, over 3813794.84 frames. ], batch size: 55, lr: 1.57e-02, grad_scale: 8.0 +2023-04-01 06:07:51,836 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2432, 1.2192, 1.9418, 1.3978, 3.1261, 2.4630, 3.2136, 1.4218], + device='cuda:0'), covar=tensor([0.1948, 0.3366, 0.1807, 0.1618, 0.1337, 0.1595, 0.1613, 0.2982], + device='cuda:0'), in_proj_covar=tensor([0.0446, 0.0505, 0.0482, 0.0407, 0.0554, 0.0446, 0.0628, 0.0444], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 06:08:06,606 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.479e+02 7.188e+02 9.159e+02 1.163e+03 2.158e+03, threshold=1.832e+03, percent-clipped=9.0 +2023-04-01 06:08:23,129 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1150, 1.1916, 1.7132, 1.3026, 2.3716, 1.9238, 2.4219, 0.9262], + device='cuda:0'), covar=tensor([0.1797, 0.2969, 0.1515, 0.1484, 0.1175, 0.1511, 0.1239, 0.2774], + device='cuda:0'), in_proj_covar=tensor([0.0443, 0.0500, 0.0479, 0.0403, 0.0550, 0.0444, 0.0622, 0.0439], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 06:08:38,293 INFO [train.py:903] (0/4) Epoch 5, batch 5650, loss[loss=0.24, simple_loss=0.3101, pruned_loss=0.08498, over 19576.00 frames. ], tot_loss[loss=0.281, simple_loss=0.341, pruned_loss=0.1105, over 3819943.04 frames. ], batch size: 52, lr: 1.57e-02, grad_scale: 8.0 +2023-04-01 06:08:54,135 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.63 vs. limit=2.0 +2023-04-01 06:09:04,595 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=32984.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:09:24,980 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 06:09:38,288 INFO [train.py:903] (0/4) Epoch 5, batch 5700, loss[loss=0.2765, simple_loss=0.3438, pruned_loss=0.1046, over 19153.00 frames. ], tot_loss[loss=0.279, simple_loss=0.3394, pruned_loss=0.1093, over 3838881.64 frames. ], batch size: 69, lr: 1.57e-02, grad_scale: 8.0 +2023-04-01 06:09:42,493 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-01 06:10:09,259 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.491e+02 7.117e+02 8.783e+02 1.086e+03 2.576e+03, threshold=1.757e+03, percent-clipped=4.0 +2023-04-01 06:10:38,536 INFO [train.py:903] (0/4) Epoch 5, batch 5750, loss[loss=0.3329, simple_loss=0.3761, pruned_loss=0.1449, over 19680.00 frames. ], tot_loss[loss=0.2784, simple_loss=0.3388, pruned_loss=0.109, over 3830787.62 frames. ], batch size: 58, lr: 1.57e-02, grad_scale: 8.0 +2023-04-01 06:10:39,678 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 06:10:47,556 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 06:10:52,572 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 06:11:01,624 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1624, 1.1946, 1.7812, 1.4164, 2.4057, 2.1933, 2.6559, 0.9932], + device='cuda:0'), covar=tensor([0.1898, 0.3097, 0.1634, 0.1516, 0.1236, 0.1427, 0.1306, 0.2840], + device='cuda:0'), in_proj_covar=tensor([0.0442, 0.0496, 0.0476, 0.0402, 0.0549, 0.0437, 0.0617, 0.0437], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 06:11:09,105 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33087.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:11:40,202 INFO [train.py:903] (0/4) Epoch 5, batch 5800, loss[loss=0.2751, simple_loss=0.3298, pruned_loss=0.1102, over 19452.00 frames. ], tot_loss[loss=0.2796, simple_loss=0.34, pruned_loss=0.1096, over 3831952.07 frames. ], batch size: 49, lr: 1.57e-02, grad_scale: 8.0 +2023-04-01 06:12:08,910 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.601e+02 7.070e+02 8.520e+02 1.129e+03 2.712e+03, threshold=1.704e+03, percent-clipped=8.0 +2023-04-01 06:12:40,678 INFO [train.py:903] (0/4) Epoch 5, batch 5850, loss[loss=0.2867, simple_loss=0.3495, pruned_loss=0.112, over 19329.00 frames. ], tot_loss[loss=0.2797, simple_loss=0.3403, pruned_loss=0.1096, over 3837955.63 frames. ], batch size: 66, lr: 1.57e-02, grad_scale: 8.0 +2023-04-01 06:12:53,479 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33173.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:13:20,550 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8022, 1.9554, 1.7386, 3.1154, 1.9195, 2.6980, 2.5600, 1.6589], + device='cuda:0'), covar=tensor([0.2173, 0.1635, 0.0886, 0.0899, 0.2032, 0.0695, 0.1650, 0.1619], + device='cuda:0'), in_proj_covar=tensor([0.0642, 0.0629, 0.0559, 0.0790, 0.0675, 0.0556, 0.0695, 0.0593], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 06:13:40,936 INFO [train.py:903] (0/4) Epoch 5, batch 5900, loss[loss=0.2763, simple_loss=0.3352, pruned_loss=0.1087, over 19548.00 frames. ], tot_loss[loss=0.28, simple_loss=0.3404, pruned_loss=0.1098, over 3839407.43 frames. ], batch size: 56, lr: 1.57e-02, grad_scale: 8.0 +2023-04-01 06:13:43,338 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 06:14:04,496 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 06:14:11,990 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.345e+02 7.066e+02 8.762e+02 1.130e+03 2.300e+03, threshold=1.752e+03, percent-clipped=6.0 +2023-04-01 06:14:41,680 INFO [train.py:903] (0/4) Epoch 5, batch 5950, loss[loss=0.2787, simple_loss=0.3498, pruned_loss=0.1037, over 19663.00 frames. ], tot_loss[loss=0.2791, simple_loss=0.34, pruned_loss=0.1091, over 3838346.93 frames. ], batch size: 60, lr: 1.56e-02, grad_scale: 8.0 +2023-04-01 06:15:08,525 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2381, 1.1993, 1.5474, 0.8232, 2.3930, 2.9488, 2.7377, 3.1184], + device='cuda:0'), covar=tensor([0.1356, 0.2977, 0.2806, 0.2024, 0.0434, 0.0158, 0.0260, 0.0153], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0284, 0.0316, 0.0251, 0.0200, 0.0118, 0.0204, 0.0144], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 06:15:43,948 INFO [train.py:903] (0/4) Epoch 5, batch 6000, loss[loss=0.2485, simple_loss=0.3155, pruned_loss=0.09073, over 19733.00 frames. ], tot_loss[loss=0.2793, simple_loss=0.3403, pruned_loss=0.1091, over 3813988.18 frames. ], batch size: 51, lr: 1.56e-02, grad_scale: 8.0 +2023-04-01 06:15:43,949 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 06:15:56,881 INFO [train.py:937] (0/4) Epoch 5, validation: loss=0.203, simple_loss=0.3017, pruned_loss=0.05213, over 944034.00 frames. +2023-04-01 06:15:56,882 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 06:16:18,149 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=33328.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:16:28,853 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.050e+02 6.893e+02 8.503e+02 1.056e+03 1.945e+03, threshold=1.701e+03, percent-clipped=4.0 +2023-04-01 06:16:59,410 INFO [train.py:903] (0/4) Epoch 5, batch 6050, loss[loss=0.2692, simple_loss=0.334, pruned_loss=0.1022, over 18111.00 frames. ], tot_loss[loss=0.2796, simple_loss=0.3402, pruned_loss=0.1095, over 3822805.39 frames. ], batch size: 83, lr: 1.56e-02, grad_scale: 16.0 +2023-04-01 06:18:00,558 INFO [train.py:903] (0/4) Epoch 5, batch 6100, loss[loss=0.2741, simple_loss=0.33, pruned_loss=0.1091, over 19742.00 frames. ], tot_loss[loss=0.2802, simple_loss=0.3404, pruned_loss=0.11, over 3823670.95 frames. ], batch size: 51, lr: 1.56e-02, grad_scale: 8.0 +2023-04-01 06:18:21,940 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6523, 4.1484, 2.7370, 3.7971, 1.4339, 3.5991, 3.8553, 3.9311], + device='cuda:0'), covar=tensor([0.0626, 0.1303, 0.1941, 0.0694, 0.3769, 0.1131, 0.0814, 0.1373], + device='cuda:0'), in_proj_covar=tensor([0.0354, 0.0318, 0.0371, 0.0289, 0.0360, 0.0312, 0.0294, 0.0323], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 06:18:23,019 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=33431.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:18:28,996 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.55 vs. limit=2.0 +2023-04-01 06:18:32,606 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.702e+02 6.262e+02 7.464e+02 9.853e+02 2.581e+03, threshold=1.493e+03, percent-clipped=2.0 +2023-04-01 06:18:39,251 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33443.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:19:00,969 INFO [train.py:903] (0/4) Epoch 5, batch 6150, loss[loss=0.3074, simple_loss=0.3638, pruned_loss=0.1255, over 19675.00 frames. ], tot_loss[loss=0.2813, simple_loss=0.3412, pruned_loss=0.1107, over 3813347.69 frames. ], batch size: 60, lr: 1.56e-02, grad_scale: 8.0 +2023-04-01 06:19:29,653 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 06:19:35,805 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33490.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:20:01,182 INFO [train.py:903] (0/4) Epoch 5, batch 6200, loss[loss=0.3141, simple_loss=0.3622, pruned_loss=0.133, over 19353.00 frames. ], tot_loss[loss=0.281, simple_loss=0.3406, pruned_loss=0.1106, over 3811727.77 frames. ], batch size: 66, lr: 1.56e-02, grad_scale: 8.0 +2023-04-01 06:20:08,862 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=33517.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:20:25,364 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3069, 3.0327, 2.0008, 2.8064, 0.9413, 2.7691, 2.7653, 2.9169], + device='cuda:0'), covar=tensor([0.0907, 0.1199, 0.1982, 0.0864, 0.3525, 0.1079, 0.0990, 0.1049], + device='cuda:0'), in_proj_covar=tensor([0.0349, 0.0308, 0.0364, 0.0285, 0.0350, 0.0303, 0.0287, 0.0315], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 06:20:34,185 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.213e+02 6.728e+02 8.677e+02 1.165e+03 2.777e+03, threshold=1.735e+03, percent-clipped=13.0 +2023-04-01 06:20:40,236 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1705, 2.1664, 1.6627, 1.5820, 1.5233, 1.7180, 0.2701, 1.0382], + device='cuda:0'), covar=tensor([0.0226, 0.0254, 0.0204, 0.0249, 0.0492, 0.0313, 0.0479, 0.0414], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0279, 0.0282, 0.0298, 0.0372, 0.0293, 0.0272, 0.0291], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 06:20:43,552 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33546.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:21:03,399 INFO [train.py:903] (0/4) Epoch 5, batch 6250, loss[loss=0.3276, simple_loss=0.3748, pruned_loss=0.1402, over 13160.00 frames. ], tot_loss[loss=0.2813, simple_loss=0.3413, pruned_loss=0.1107, over 3799152.21 frames. ], batch size: 136, lr: 1.56e-02, grad_scale: 8.0 +2023-04-01 06:21:31,243 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 06:22:03,940 INFO [train.py:903] (0/4) Epoch 5, batch 6300, loss[loss=0.2872, simple_loss=0.3541, pruned_loss=0.1101, over 17669.00 frames. ], tot_loss[loss=0.283, simple_loss=0.3423, pruned_loss=0.1118, over 3795507.24 frames. ], batch size: 101, lr: 1.56e-02, grad_scale: 8.0 +2023-04-01 06:22:28,455 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33632.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:22:35,563 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.303e+02 6.435e+02 8.030e+02 9.840e+02 2.632e+03, threshold=1.606e+03, percent-clipped=3.0 +2023-04-01 06:23:04,054 INFO [train.py:903] (0/4) Epoch 5, batch 6350, loss[loss=0.2287, simple_loss=0.303, pruned_loss=0.07724, over 19848.00 frames. ], tot_loss[loss=0.2826, simple_loss=0.3419, pruned_loss=0.1117, over 3782971.73 frames. ], batch size: 52, lr: 1.56e-02, grad_scale: 8.0 +2023-04-01 06:23:45,940 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33695.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:23:50,615 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33699.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:24:05,044 INFO [train.py:903] (0/4) Epoch 5, batch 6400, loss[loss=0.2338, simple_loss=0.2949, pruned_loss=0.08638, over 19718.00 frames. ], tot_loss[loss=0.281, simple_loss=0.3411, pruned_loss=0.1104, over 3807214.67 frames. ], batch size: 46, lr: 1.55e-02, grad_scale: 8.0 +2023-04-01 06:24:20,318 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33724.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:24:37,366 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.704e+02 6.874e+02 8.420e+02 1.031e+03 3.616e+03, threshold=1.684e+03, percent-clipped=3.0 +2023-04-01 06:25:05,851 INFO [train.py:903] (0/4) Epoch 5, batch 6450, loss[loss=0.2677, simple_loss=0.3165, pruned_loss=0.1094, over 19414.00 frames. ], tot_loss[loss=0.2809, simple_loss=0.3409, pruned_loss=0.1104, over 3813682.68 frames. ], batch size: 48, lr: 1.55e-02, grad_scale: 8.0 +2023-04-01 06:25:18,508 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6566, 1.2772, 1.5654, 1.2107, 2.5275, 3.1528, 3.0194, 3.2942], + device='cuda:0'), covar=tensor([0.1180, 0.2931, 0.2928, 0.1929, 0.0460, 0.0301, 0.0215, 0.0141], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0282, 0.0316, 0.0250, 0.0198, 0.0116, 0.0203, 0.0143], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 06:25:48,207 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 06:25:54,084 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33802.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:26:06,364 INFO [train.py:903] (0/4) Epoch 5, batch 6500, loss[loss=0.2924, simple_loss=0.3566, pruned_loss=0.114, over 19608.00 frames. ], tot_loss[loss=0.2816, simple_loss=0.3418, pruned_loss=0.1107, over 3827457.61 frames. ], batch size: 61, lr: 1.55e-02, grad_scale: 8.0 +2023-04-01 06:26:12,175 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 06:26:24,758 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33827.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:26:32,491 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=33834.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:26:36,709 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.072e+02 6.680e+02 8.171e+02 1.132e+03 2.519e+03, threshold=1.634e+03, percent-clipped=6.0 +2023-04-01 06:27:07,202 INFO [train.py:903] (0/4) Epoch 5, batch 6550, loss[loss=0.2908, simple_loss=0.3543, pruned_loss=0.1137, over 19665.00 frames. ], tot_loss[loss=0.2814, simple_loss=0.341, pruned_loss=0.1109, over 3830240.58 frames. ], batch size: 58, lr: 1.55e-02, grad_scale: 8.0 +2023-04-01 06:27:38,512 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=33888.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:27:59,331 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8556, 1.3310, 0.9729, 0.9653, 1.2214, 0.9053, 0.7505, 1.2269], + device='cuda:0'), covar=tensor([0.0442, 0.0542, 0.0944, 0.0419, 0.0355, 0.0945, 0.0480, 0.0343], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0269, 0.0312, 0.0237, 0.0222, 0.0303, 0.0285, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 06:28:06,987 INFO [train.py:903] (0/4) Epoch 5, batch 6600, loss[loss=0.3068, simple_loss=0.3652, pruned_loss=0.1242, over 19386.00 frames. ], tot_loss[loss=0.2808, simple_loss=0.3409, pruned_loss=0.1104, over 3836026.27 frames. ], batch size: 70, lr: 1.55e-02, grad_scale: 4.0 +2023-04-01 06:28:08,463 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=33913.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:28:40,229 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.407e+02 7.502e+02 9.137e+02 1.060e+03 2.817e+03, threshold=1.827e+03, percent-clipped=6.0 +2023-04-01 06:28:52,881 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=33949.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:29:01,830 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=33957.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:29:09,036 INFO [train.py:903] (0/4) Epoch 5, batch 6650, loss[loss=0.2891, simple_loss=0.3573, pruned_loss=0.1105, over 19673.00 frames. ], tot_loss[loss=0.2808, simple_loss=0.341, pruned_loss=0.1104, over 3814120.65 frames. ], batch size: 55, lr: 1.55e-02, grad_scale: 4.0 +2023-04-01 06:29:54,080 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-34000.pt +2023-04-01 06:30:10,052 INFO [train.py:903] (0/4) Epoch 5, batch 6700, loss[loss=0.2135, simple_loss=0.2814, pruned_loss=0.07279, over 19747.00 frames. ], tot_loss[loss=0.279, simple_loss=0.3395, pruned_loss=0.1092, over 3828593.57 frames. ], batch size: 48, lr: 1.55e-02, grad_scale: 4.0 +2023-04-01 06:30:40,326 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.004e+02 7.257e+02 9.131e+02 1.100e+03 2.314e+03, threshold=1.826e+03, percent-clipped=7.0 +2023-04-01 06:30:40,470 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34039.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:31:06,265 INFO [train.py:903] (0/4) Epoch 5, batch 6750, loss[loss=0.3088, simple_loss=0.3718, pruned_loss=0.1229, over 19141.00 frames. ], tot_loss[loss=0.2826, simple_loss=0.3425, pruned_loss=0.1113, over 3826430.96 frames. ], batch size: 69, lr: 1.55e-02, grad_scale: 4.0 +2023-04-01 06:31:06,502 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34062.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:31:22,417 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3951, 1.2132, 1.2788, 1.7773, 1.3886, 1.6579, 1.7325, 1.4586], + device='cuda:0'), covar=tensor([0.0891, 0.1102, 0.1176, 0.0859, 0.0986, 0.0780, 0.0859, 0.0752], + device='cuda:0'), in_proj_covar=tensor([0.0232, 0.0251, 0.0249, 0.0278, 0.0268, 0.0233, 0.0231, 0.0224], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 06:32:02,677 INFO [train.py:903] (0/4) Epoch 5, batch 6800, loss[loss=0.3043, simple_loss=0.3548, pruned_loss=0.1269, over 19764.00 frames. ], tot_loss[loss=0.2822, simple_loss=0.3425, pruned_loss=0.111, over 3827199.72 frames. ], batch size: 54, lr: 1.55e-02, grad_scale: 8.0 +2023-04-01 06:32:30,647 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.271e+02 6.317e+02 7.808e+02 9.230e+02 1.582e+03, threshold=1.562e+03, percent-clipped=0.0 +2023-04-01 06:32:32,097 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-5.pt +2023-04-01 06:32:48,125 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 06:32:48,569 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 06:32:50,739 INFO [train.py:903] (0/4) Epoch 6, batch 0, loss[loss=0.2949, simple_loss=0.3453, pruned_loss=0.1222, over 19600.00 frames. ], tot_loss[loss=0.2949, simple_loss=0.3453, pruned_loss=0.1222, over 19600.00 frames. ], batch size: 52, lr: 1.44e-02, grad_scale: 8.0 +2023-04-01 06:32:50,740 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 06:33:02,092 INFO [train.py:937] (0/4) Epoch 6, validation: loss=0.2022, simple_loss=0.3015, pruned_loss=0.05149, over 944034.00 frames. +2023-04-01 06:33:02,093 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 06:33:15,327 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 06:33:20,318 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34154.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:33:30,302 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34163.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:34:03,526 INFO [train.py:903] (0/4) Epoch 6, batch 50, loss[loss=0.2153, simple_loss=0.2815, pruned_loss=0.07455, over 19754.00 frames. ], tot_loss[loss=0.2796, simple_loss=0.3403, pruned_loss=0.1095, over 870008.59 frames. ], batch size: 47, lr: 1.44e-02, grad_scale: 8.0 +2023-04-01 06:34:22,182 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34205.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:34:40,711 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 06:34:53,240 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34230.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:35:05,224 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.721e+02 5.756e+02 7.149e+02 1.025e+03 3.166e+03, threshold=1.430e+03, percent-clipped=7.0 +2023-04-01 06:35:06,289 INFO [train.py:903] (0/4) Epoch 6, batch 100, loss[loss=0.3104, simple_loss=0.3727, pruned_loss=0.124, over 19688.00 frames. ], tot_loss[loss=0.2769, simple_loss=0.3385, pruned_loss=0.1076, over 1538250.44 frames. ], batch size: 59, lr: 1.44e-02, grad_scale: 8.0 +2023-04-01 06:35:18,596 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 06:35:23,478 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7170, 0.8430, 0.9827, 0.9929, 1.5366, 0.7876, 1.3455, 1.5794], + device='cuda:0'), covar=tensor([0.0451, 0.1692, 0.1575, 0.0975, 0.0512, 0.1331, 0.0840, 0.0401], + device='cuda:0'), in_proj_covar=tensor([0.0285, 0.0311, 0.0313, 0.0286, 0.0306, 0.0316, 0.0283, 0.0301], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 06:35:26,668 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.2916, 3.9396, 2.5049, 3.5125, 1.2537, 3.5252, 3.6275, 3.7331], + device='cuda:0'), covar=tensor([0.0591, 0.0948, 0.1728, 0.0648, 0.3306, 0.0745, 0.0685, 0.0715], + device='cuda:0'), in_proj_covar=tensor([0.0349, 0.0313, 0.0371, 0.0287, 0.0353, 0.0304, 0.0287, 0.0315], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 06:36:06,452 INFO [train.py:903] (0/4) Epoch 6, batch 150, loss[loss=0.2661, simple_loss=0.3325, pruned_loss=0.09987, over 19382.00 frames. ], tot_loss[loss=0.2786, simple_loss=0.3398, pruned_loss=0.1087, over 2051831.82 frames. ], batch size: 70, lr: 1.44e-02, grad_scale: 4.0 +2023-04-01 06:36:19,454 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34301.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:37:08,887 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.164e+02 6.478e+02 8.283e+02 9.993e+02 1.951e+03, threshold=1.657e+03, percent-clipped=7.0 +2023-04-01 06:37:08,905 INFO [train.py:903] (0/4) Epoch 6, batch 200, loss[loss=0.2655, simple_loss=0.3302, pruned_loss=0.1004, over 19485.00 frames. ], tot_loss[loss=0.2752, simple_loss=0.3374, pruned_loss=0.1065, over 2451775.14 frames. ], batch size: 49, lr: 1.44e-02, grad_scale: 4.0 +2023-04-01 06:37:08,924 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 06:37:44,159 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2050, 1.3006, 1.6500, 1.3905, 2.1163, 2.0101, 2.2916, 0.7456], + device='cuda:0'), covar=tensor([0.1878, 0.3079, 0.1620, 0.1520, 0.1127, 0.1592, 0.1259, 0.2921], + device='cuda:0'), in_proj_covar=tensor([0.0450, 0.0503, 0.0485, 0.0411, 0.0553, 0.0448, 0.0627, 0.0444], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 06:38:12,102 INFO [train.py:903] (0/4) Epoch 6, batch 250, loss[loss=0.2783, simple_loss=0.3457, pruned_loss=0.1055, over 18358.00 frames. ], tot_loss[loss=0.2755, simple_loss=0.3371, pruned_loss=0.1069, over 2753470.04 frames. ], batch size: 83, lr: 1.44e-02, grad_scale: 4.0 +2023-04-01 06:38:33,066 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34406.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:38:38,003 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34410.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:38:38,438 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.90 vs. limit=5.0 +2023-04-01 06:38:44,813 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.2451, 3.8757, 2.4699, 3.5495, 1.1028, 3.4491, 3.5226, 3.6990], + device='cuda:0'), covar=tensor([0.0705, 0.1088, 0.1938, 0.0749, 0.3919, 0.1048, 0.0843, 0.0959], + device='cuda:0'), in_proj_covar=tensor([0.0350, 0.0310, 0.0369, 0.0286, 0.0355, 0.0308, 0.0286, 0.0315], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 06:38:44,979 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34416.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:39:08,874 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34435.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:39:14,113 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.775e+02 6.949e+02 8.663e+02 1.115e+03 2.860e+03, threshold=1.733e+03, percent-clipped=3.0 +2023-04-01 06:39:14,131 INFO [train.py:903] (0/4) Epoch 6, batch 300, loss[loss=0.2992, simple_loss=0.3601, pruned_loss=0.1191, over 19666.00 frames. ], tot_loss[loss=0.2757, simple_loss=0.3372, pruned_loss=0.1071, over 2997294.16 frames. ], batch size: 60, lr: 1.44e-02, grad_scale: 4.0 +2023-04-01 06:39:15,636 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3432, 1.0575, 1.2471, 1.2147, 2.0073, 0.9211, 1.6921, 2.0062], + device='cuda:0'), covar=tensor([0.0585, 0.2509, 0.2351, 0.1448, 0.0764, 0.2038, 0.1048, 0.0640], + device='cuda:0'), in_proj_covar=tensor([0.0290, 0.0316, 0.0320, 0.0288, 0.0311, 0.0318, 0.0289, 0.0307], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 06:40:17,215 INFO [train.py:903] (0/4) Epoch 6, batch 350, loss[loss=0.224, simple_loss=0.284, pruned_loss=0.08205, over 19322.00 frames. ], tot_loss[loss=0.2756, simple_loss=0.3372, pruned_loss=0.107, over 3189702.66 frames. ], batch size: 44, lr: 1.43e-02, grad_scale: 4.0 +2023-04-01 06:40:22,965 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 06:40:37,839 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34507.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:40:55,542 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34521.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:41:18,579 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.691e+02 6.755e+02 8.258e+02 9.979e+02 1.871e+03, threshold=1.652e+03, percent-clipped=1.0 +2023-04-01 06:41:18,598 INFO [train.py:903] (0/4) Epoch 6, batch 400, loss[loss=0.278, simple_loss=0.3448, pruned_loss=0.1056, over 19763.00 frames. ], tot_loss[loss=0.2758, simple_loss=0.3375, pruned_loss=0.1071, over 3325758.11 frames. ], batch size: 54, lr: 1.43e-02, grad_scale: 8.0 +2023-04-01 06:42:20,652 INFO [train.py:903] (0/4) Epoch 6, batch 450, loss[loss=0.2641, simple_loss=0.3276, pruned_loss=0.1003, over 19748.00 frames. ], tot_loss[loss=0.2792, simple_loss=0.3397, pruned_loss=0.1093, over 3430988.89 frames. ], batch size: 51, lr: 1.43e-02, grad_scale: 8.0 +2023-04-01 06:42:48,992 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34611.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:42:54,487 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 06:42:55,450 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 06:43:01,502 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=34622.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:43:23,817 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.432e+02 6.970e+02 8.269e+02 1.071e+03 2.551e+03, threshold=1.654e+03, percent-clipped=6.0 +2023-04-01 06:43:23,840 INFO [train.py:903] (0/4) Epoch 6, batch 500, loss[loss=0.2721, simple_loss=0.3431, pruned_loss=0.1006, over 19774.00 frames. ], tot_loss[loss=0.2789, simple_loss=0.3398, pruned_loss=0.109, over 3524004.18 frames. ], batch size: 54, lr: 1.43e-02, grad_scale: 8.0 +2023-04-01 06:43:24,103 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34640.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:43:29,888 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34644.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:44:03,457 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34672.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:44:27,130 INFO [train.py:903] (0/4) Epoch 6, batch 550, loss[loss=0.2612, simple_loss=0.328, pruned_loss=0.09714, over 19732.00 frames. ], tot_loss[loss=0.2794, simple_loss=0.3401, pruned_loss=0.1093, over 3585220.77 frames. ], batch size: 51, lr: 1.43e-02, grad_scale: 8.0 +2023-04-01 06:44:37,128 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34697.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:45:17,136 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=34728.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 06:45:31,829 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.620e+02 6.191e+02 8.093e+02 9.820e+02 1.880e+03, threshold=1.619e+03, percent-clipped=2.0 +2023-04-01 06:45:31,848 INFO [train.py:903] (0/4) Epoch 6, batch 600, loss[loss=0.2841, simple_loss=0.3398, pruned_loss=0.1143, over 19737.00 frames. ], tot_loss[loss=0.2786, simple_loss=0.34, pruned_loss=0.1086, over 3644102.81 frames. ], batch size: 51, lr: 1.43e-02, grad_scale: 8.0 +2023-04-01 06:46:13,316 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 06:46:15,091 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-04-01 06:46:20,218 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34777.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:46:32,653 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.98 vs. limit=5.0 +2023-04-01 06:46:35,474 INFO [train.py:903] (0/4) Epoch 6, batch 650, loss[loss=0.2711, simple_loss=0.3201, pruned_loss=0.111, over 19813.00 frames. ], tot_loss[loss=0.2774, simple_loss=0.339, pruned_loss=0.1079, over 3690163.13 frames. ], batch size: 48, lr: 1.43e-02, grad_scale: 8.0 +2023-04-01 06:46:51,790 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34802.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:47:38,631 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.675e+02 6.280e+02 8.587e+02 1.153e+03 3.497e+03, threshold=1.717e+03, percent-clipped=9.0 +2023-04-01 06:47:38,650 INFO [train.py:903] (0/4) Epoch 6, batch 700, loss[loss=0.2729, simple_loss=0.3419, pruned_loss=0.102, over 17385.00 frames. ], tot_loss[loss=0.2769, simple_loss=0.3386, pruned_loss=0.1076, over 3723561.96 frames. ], batch size: 101, lr: 1.43e-02, grad_scale: 8.0 +2023-04-01 06:48:27,730 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=34878.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:48:43,535 INFO [train.py:903] (0/4) Epoch 6, batch 750, loss[loss=0.2742, simple_loss=0.3264, pruned_loss=0.111, over 17781.00 frames. ], tot_loss[loss=0.276, simple_loss=0.3379, pruned_loss=0.1071, over 3746373.60 frames. ], batch size: 39, lr: 1.43e-02, grad_scale: 8.0 +2023-04-01 06:49:00,037 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=34903.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:49:06,815 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4672, 1.2782, 1.5122, 1.0503, 2.5349, 3.5133, 3.2750, 3.6232], + device='cuda:0'), covar=tensor([0.1321, 0.2849, 0.2719, 0.1900, 0.0417, 0.0165, 0.0191, 0.0123], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0281, 0.0311, 0.0246, 0.0198, 0.0118, 0.0201, 0.0145], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 06:49:45,092 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.784e+02 6.254e+02 7.861e+02 1.094e+03 2.828e+03, threshold=1.572e+03, percent-clipped=5.0 +2023-04-01 06:49:45,111 INFO [train.py:903] (0/4) Epoch 6, batch 800, loss[loss=0.2831, simple_loss=0.3436, pruned_loss=0.1113, over 17467.00 frames. ], tot_loss[loss=0.2747, simple_loss=0.3363, pruned_loss=0.1066, over 3762699.73 frames. ], batch size: 101, lr: 1.43e-02, grad_scale: 8.0 +2023-04-01 06:50:02,472 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 06:50:03,702 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34955.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:50:21,726 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-01 06:50:41,616 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34984.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:50:46,214 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=34988.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:50:48,468 INFO [train.py:903] (0/4) Epoch 6, batch 850, loss[loss=0.3071, simple_loss=0.366, pruned_loss=0.1241, over 18867.00 frames. ], tot_loss[loss=0.2762, simple_loss=0.3378, pruned_loss=0.1073, over 3769992.58 frames. ], batch size: 74, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 06:51:32,723 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1318, 1.2032, 1.7023, 1.4273, 2.3693, 2.0647, 2.5437, 0.9043], + device='cuda:0'), covar=tensor([0.1873, 0.2914, 0.1529, 0.1374, 0.1112, 0.1482, 0.1204, 0.2728], + device='cuda:0'), in_proj_covar=tensor([0.0448, 0.0511, 0.0488, 0.0412, 0.0556, 0.0449, 0.0625, 0.0448], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 06:51:42,520 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 06:51:49,529 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.027e+02 6.223e+02 7.935e+02 9.772e+02 2.166e+03, threshold=1.587e+03, percent-clipped=2.0 +2023-04-01 06:51:49,548 INFO [train.py:903] (0/4) Epoch 6, batch 900, loss[loss=0.2866, simple_loss=0.3417, pruned_loss=0.1157, over 19297.00 frames. ], tot_loss[loss=0.2769, simple_loss=0.3385, pruned_loss=0.1077, over 3780519.94 frames. ], batch size: 44, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 06:51:51,218 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8357, 1.5882, 1.5523, 1.6248, 3.3750, 1.0646, 2.1531, 3.5447], + device='cuda:0'), covar=tensor([0.0299, 0.2155, 0.2117, 0.1428, 0.0540, 0.2268, 0.1185, 0.0307], + device='cuda:0'), in_proj_covar=tensor([0.0291, 0.0314, 0.0313, 0.0284, 0.0310, 0.0315, 0.0290, 0.0305], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 06:52:28,231 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35070.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:52:30,301 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35072.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 06:52:51,734 INFO [train.py:903] (0/4) Epoch 6, batch 950, loss[loss=0.3201, simple_loss=0.3684, pruned_loss=0.1359, over 19400.00 frames. ], tot_loss[loss=0.2764, simple_loss=0.3382, pruned_loss=0.1073, over 3792512.54 frames. ], batch size: 70, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 06:52:57,559 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 06:53:04,298 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35099.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:53:09,177 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35103.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:53:18,080 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35110.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:53:55,217 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.390e+02 7.194e+02 8.589e+02 1.083e+03 2.096e+03, threshold=1.718e+03, percent-clipped=5.0 +2023-04-01 06:53:55,236 INFO [train.py:903] (0/4) Epoch 6, batch 1000, loss[loss=0.3145, simple_loss=0.3641, pruned_loss=0.1324, over 19676.00 frames. ], tot_loss[loss=0.2752, simple_loss=0.3371, pruned_loss=0.1067, over 3800089.39 frames. ], batch size: 55, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 06:54:48,390 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 06:54:52,210 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35187.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 06:54:55,179 INFO [train.py:903] (0/4) Epoch 6, batch 1050, loss[loss=0.2696, simple_loss=0.3343, pruned_loss=0.1024, over 17466.00 frames. ], tot_loss[loss=0.2771, simple_loss=0.3381, pruned_loss=0.1081, over 3795982.80 frames. ], batch size: 101, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 06:55:31,003 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 06:55:49,745 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-04-01 06:55:57,165 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.102e+02 7.066e+02 8.619e+02 1.238e+03 3.302e+03, threshold=1.724e+03, percent-clipped=8.0 +2023-04-01 06:55:57,184 INFO [train.py:903] (0/4) Epoch 6, batch 1100, loss[loss=0.227, simple_loss=0.2912, pruned_loss=0.08138, over 19736.00 frames. ], tot_loss[loss=0.2784, simple_loss=0.3391, pruned_loss=0.1088, over 3804929.15 frames. ], batch size: 46, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 06:56:09,825 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9621, 2.0232, 2.0482, 1.9972, 4.4279, 1.1071, 2.4709, 4.5797], + device='cuda:0'), covar=tensor([0.0242, 0.2129, 0.2257, 0.1465, 0.0503, 0.2404, 0.1175, 0.0247], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0315, 0.0317, 0.0286, 0.0310, 0.0315, 0.0288, 0.0307], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 06:56:48,011 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.95 vs. limit=5.0 +2023-04-01 06:56:59,593 INFO [train.py:903] (0/4) Epoch 6, batch 1150, loss[loss=0.2856, simple_loss=0.3508, pruned_loss=0.1102, over 19756.00 frames. ], tot_loss[loss=0.279, simple_loss=0.3397, pruned_loss=0.1092, over 3792027.51 frames. ], batch size: 54, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 06:57:45,474 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35326.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:58:04,197 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.615e+02 6.124e+02 7.469e+02 9.050e+02 1.884e+03, threshold=1.494e+03, percent-clipped=1.0 +2023-04-01 06:58:04,215 INFO [train.py:903] (0/4) Epoch 6, batch 1200, loss[loss=0.3346, simple_loss=0.3898, pruned_loss=0.1397, over 19749.00 frames. ], tot_loss[loss=0.2768, simple_loss=0.3382, pruned_loss=0.1077, over 3812562.66 frames. ], batch size: 63, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 06:58:17,318 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35351.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:58:23,057 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35355.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:58:27,691 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35359.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:58:32,182 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35363.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 06:58:36,603 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 06:58:55,103 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35380.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:58:56,253 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1323, 1.0579, 1.7143, 1.2693, 2.7931, 2.1528, 2.8553, 1.1406], + device='cuda:0'), covar=tensor([0.2066, 0.3449, 0.1871, 0.1788, 0.1113, 0.1637, 0.1332, 0.2995], + device='cuda:0'), in_proj_covar=tensor([0.0447, 0.0508, 0.0494, 0.0413, 0.0562, 0.0451, 0.0626, 0.0452], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 06:58:59,588 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35384.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 06:59:05,734 INFO [train.py:903] (0/4) Epoch 6, batch 1250, loss[loss=0.3123, simple_loss=0.3612, pruned_loss=0.1317, over 19654.00 frames. ], tot_loss[loss=0.2774, simple_loss=0.339, pruned_loss=0.1079, over 3802460.86 frames. ], batch size: 55, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 06:59:09,384 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35392.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:00:08,109 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.399e+02 6.685e+02 8.523e+02 1.077e+03 2.432e+03, threshold=1.705e+03, percent-clipped=5.0 +2023-04-01 07:00:08,128 INFO [train.py:903] (0/4) Epoch 6, batch 1300, loss[loss=0.2523, simple_loss=0.3078, pruned_loss=0.0984, over 19356.00 frames. ], tot_loss[loss=0.276, simple_loss=0.3376, pruned_loss=0.1071, over 3821045.00 frames. ], batch size: 44, lr: 1.42e-02, grad_scale: 8.0 +2023-04-01 07:00:12,135 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35443.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 07:00:26,225 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35454.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:00:43,964 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35468.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 07:01:11,031 INFO [train.py:903] (0/4) Epoch 6, batch 1350, loss[loss=0.2646, simple_loss=0.3208, pruned_loss=0.1042, over 19749.00 frames. ], tot_loss[loss=0.2757, simple_loss=0.3375, pruned_loss=0.107, over 3814628.61 frames. ], batch size: 45, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:01:40,785 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6233, 1.2039, 1.8972, 1.9182, 2.7367, 4.6752, 4.4846, 4.8875], + device='cuda:0'), covar=tensor([0.1324, 0.3113, 0.2746, 0.1525, 0.0514, 0.0103, 0.0165, 0.0075], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0285, 0.0312, 0.0249, 0.0201, 0.0122, 0.0202, 0.0146], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 07:02:13,019 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.489e+02 6.783e+02 8.495e+02 1.071e+03 2.340e+03, threshold=1.699e+03, percent-clipped=3.0 +2023-04-01 07:02:13,038 INFO [train.py:903] (0/4) Epoch 6, batch 1400, loss[loss=0.2458, simple_loss=0.3095, pruned_loss=0.09105, over 19837.00 frames. ], tot_loss[loss=0.2745, simple_loss=0.3369, pruned_loss=0.106, over 3829642.53 frames. ], batch size: 52, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:02:22,777 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35548.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:02:47,214 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35569.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:03:13,362 INFO [train.py:903] (0/4) Epoch 6, batch 1450, loss[loss=0.3225, simple_loss=0.3847, pruned_loss=0.1301, over 19273.00 frames. ], tot_loss[loss=0.2747, simple_loss=0.3368, pruned_loss=0.1063, over 3833781.17 frames. ], batch size: 66, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:03:13,400 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 07:03:52,454 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4591, 1.1485, 1.2935, 1.2538, 2.1236, 0.9181, 1.7394, 2.0783], + device='cuda:0'), covar=tensor([0.0592, 0.2396, 0.2373, 0.1358, 0.0831, 0.1934, 0.0952, 0.0625], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0313, 0.0316, 0.0286, 0.0313, 0.0312, 0.0286, 0.0305], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:04:15,923 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.374e+02 6.609e+02 8.520e+02 1.101e+03 2.891e+03, threshold=1.704e+03, percent-clipped=3.0 +2023-04-01 07:04:15,942 INFO [train.py:903] (0/4) Epoch 6, batch 1500, loss[loss=0.2596, simple_loss=0.3294, pruned_loss=0.09491, over 19667.00 frames. ], tot_loss[loss=0.2761, simple_loss=0.338, pruned_loss=0.1071, over 3830828.13 frames. ], batch size: 55, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:05:17,132 INFO [train.py:903] (0/4) Epoch 6, batch 1550, loss[loss=0.2228, simple_loss=0.2852, pruned_loss=0.08023, over 19393.00 frames. ], tot_loss[loss=0.2763, simple_loss=0.3378, pruned_loss=0.1074, over 3820694.60 frames. ], batch size: 47, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:05:39,881 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35707.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 07:06:16,826 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35736.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:06:22,436 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.104e+02 6.450e+02 9.026e+02 1.093e+03 2.835e+03, threshold=1.805e+03, percent-clipped=5.0 +2023-04-01 07:06:22,455 INFO [train.py:903] (0/4) Epoch 6, batch 1600, loss[loss=0.2656, simple_loss=0.335, pruned_loss=0.09806, over 18698.00 frames. ], tot_loss[loss=0.2762, simple_loss=0.3376, pruned_loss=0.1074, over 3802520.94 frames. ], batch size: 74, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:06:24,317 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.87 vs. limit=5.0 +2023-04-01 07:06:44,234 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 07:06:53,836 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7941, 1.7255, 1.8735, 2.0315, 4.2691, 1.1052, 2.1699, 4.1515], + device='cuda:0'), covar=tensor([0.0265, 0.2372, 0.2425, 0.1408, 0.0496, 0.2320, 0.1280, 0.0321], + device='cuda:0'), in_proj_covar=tensor([0.0294, 0.0310, 0.0313, 0.0286, 0.0308, 0.0309, 0.0283, 0.0302], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:07:23,507 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35789.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:07:24,377 INFO [train.py:903] (0/4) Epoch 6, batch 1650, loss[loss=0.2001, simple_loss=0.2788, pruned_loss=0.06064, over 19754.00 frames. ], tot_loss[loss=0.2755, simple_loss=0.3369, pruned_loss=0.1071, over 3804936.83 frames. ], batch size: 45, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:07:35,338 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5286, 1.5905, 1.6385, 2.0808, 1.3475, 1.7335, 1.9865, 1.5695], + device='cuda:0'), covar=tensor([0.2354, 0.1795, 0.1031, 0.0987, 0.1994, 0.0877, 0.2167, 0.1819], + device='cuda:0'), in_proj_covar=tensor([0.0659, 0.0649, 0.0575, 0.0807, 0.0686, 0.0568, 0.0701, 0.0612], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 07:07:50,596 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-01 07:07:51,783 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-04-01 07:08:05,697 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35822.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 07:08:09,351 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=35825.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:08:13,994 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=35829.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:08:27,430 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.226e+02 6.593e+02 7.720e+02 9.793e+02 2.227e+03, threshold=1.544e+03, percent-clipped=1.0 +2023-04-01 07:08:27,449 INFO [train.py:903] (0/4) Epoch 6, batch 1700, loss[loss=0.2993, simple_loss=0.3526, pruned_loss=0.123, over 13590.00 frames. ], tot_loss[loss=0.2757, simple_loss=0.3374, pruned_loss=0.107, over 3780200.08 frames. ], batch size: 136, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:08:38,889 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=35850.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:08:40,027 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=35851.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:09:06,051 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 07:09:25,824 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-01 07:09:29,406 INFO [train.py:903] (0/4) Epoch 6, batch 1750, loss[loss=0.2388, simple_loss=0.3045, pruned_loss=0.0866, over 19802.00 frames. ], tot_loss[loss=0.276, simple_loss=0.3379, pruned_loss=0.107, over 3802380.32 frames. ], batch size: 48, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:09:31,972 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=35892.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:10:33,848 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.141e+02 6.695e+02 8.372e+02 1.116e+03 2.634e+03, threshold=1.674e+03, percent-clipped=7.0 +2023-04-01 07:10:33,867 INFO [train.py:903] (0/4) Epoch 6, batch 1800, loss[loss=0.2817, simple_loss=0.3358, pruned_loss=0.1138, over 19730.00 frames. ], tot_loss[loss=0.2764, simple_loss=0.3381, pruned_loss=0.1073, over 3795439.27 frames. ], batch size: 51, lr: 1.41e-02, grad_scale: 8.0 +2023-04-01 07:11:31,918 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 07:11:36,738 INFO [train.py:903] (0/4) Epoch 6, batch 1850, loss[loss=0.2675, simple_loss=0.3322, pruned_loss=0.1014, over 19521.00 frames. ], tot_loss[loss=0.2764, simple_loss=0.3381, pruned_loss=0.1074, over 3787836.74 frames. ], batch size: 56, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:11:49,650 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-36000.pt +2023-04-01 07:11:59,150 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36007.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:12:11,431 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 07:12:40,879 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.024e+02 7.330e+02 9.053e+02 1.086e+03 1.723e+03, threshold=1.811e+03, percent-clipped=2.0 +2023-04-01 07:12:40,903 INFO [train.py:903] (0/4) Epoch 6, batch 1900, loss[loss=0.2588, simple_loss=0.336, pruned_loss=0.09076, over 19785.00 frames. ], tot_loss[loss=0.2761, simple_loss=0.3381, pruned_loss=0.1071, over 3797805.02 frames. ], batch size: 56, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:12:57,334 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 07:13:04,088 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 07:13:27,625 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 07:13:29,151 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36078.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 07:13:42,466 INFO [train.py:903] (0/4) Epoch 6, batch 1950, loss[loss=0.2786, simple_loss=0.3483, pruned_loss=0.1044, over 19533.00 frames. ], tot_loss[loss=0.2769, simple_loss=0.3393, pruned_loss=0.1073, over 3800999.77 frames. ], batch size: 56, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:13:49,847 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6417, 1.3629, 2.1245, 1.6022, 3.1758, 4.8250, 4.6551, 5.0497], + device='cuda:0'), covar=tensor([0.1322, 0.2777, 0.2463, 0.1654, 0.0367, 0.0124, 0.0135, 0.0060], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0283, 0.0317, 0.0248, 0.0204, 0.0124, 0.0203, 0.0148], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 07:14:00,163 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36103.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 07:14:04,984 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36107.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:14:21,453 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.55 vs. limit=2.0 +2023-04-01 07:14:28,121 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1521, 1.2101, 1.5131, 1.2795, 2.1903, 1.8451, 2.3296, 0.7321], + device='cuda:0'), covar=tensor([0.1826, 0.2954, 0.1594, 0.1505, 0.1066, 0.1658, 0.1116, 0.2810], + device='cuda:0'), in_proj_covar=tensor([0.0459, 0.0519, 0.0499, 0.0420, 0.0567, 0.0456, 0.0636, 0.0455], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 07:14:35,237 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36132.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:14:36,138 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36133.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:14:36,296 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7678, 4.3003, 4.5781, 4.5166, 1.5458, 4.2073, 3.6066, 4.1082], + device='cuda:0'), covar=tensor([0.1120, 0.0557, 0.0464, 0.0455, 0.4258, 0.0369, 0.0558, 0.0981], + device='cuda:0'), in_proj_covar=tensor([0.0544, 0.0478, 0.0630, 0.0524, 0.0604, 0.0389, 0.0403, 0.0594], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 07:14:45,007 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.894e+02 6.352e+02 7.868e+02 9.806e+02 1.510e+03, threshold=1.574e+03, percent-clipped=0.0 +2023-04-01 07:14:45,036 INFO [train.py:903] (0/4) Epoch 6, batch 2000, loss[loss=0.2709, simple_loss=0.3405, pruned_loss=0.1006, over 19674.00 frames. ], tot_loss[loss=0.2764, simple_loss=0.3389, pruned_loss=0.107, over 3800910.30 frames. ], batch size: 58, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:15:24,944 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36173.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:15:42,582 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 07:15:46,082 INFO [train.py:903] (0/4) Epoch 6, batch 2050, loss[loss=0.263, simple_loss=0.3349, pruned_loss=0.09549, over 19570.00 frames. ], tot_loss[loss=0.2759, simple_loss=0.3383, pruned_loss=0.1067, over 3817871.99 frames. ], batch size: 61, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:15:57,096 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36199.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:16:01,825 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 07:16:03,077 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 07:16:22,967 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 07:16:23,197 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2411, 2.9654, 2.0227, 2.7492, 1.0044, 2.7620, 2.6454, 2.7744], + device='cuda:0'), covar=tensor([0.1033, 0.1470, 0.2103, 0.0875, 0.3543, 0.1044, 0.0942, 0.1063], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0321, 0.0377, 0.0296, 0.0360, 0.0317, 0.0297, 0.0331], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 07:16:37,655 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8725, 1.2817, 0.9748, 0.9572, 1.1847, 0.8756, 0.6082, 1.2321], + device='cuda:0'), covar=tensor([0.0485, 0.0535, 0.0902, 0.0428, 0.0380, 0.0981, 0.0649, 0.0403], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0276, 0.0310, 0.0240, 0.0228, 0.0309, 0.0287, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:16:47,774 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.109e+02 6.690e+02 8.798e+02 1.173e+03 2.442e+03, threshold=1.760e+03, percent-clipped=12.0 +2023-04-01 07:16:47,792 INFO [train.py:903] (0/4) Epoch 6, batch 2100, loss[loss=0.3079, simple_loss=0.3623, pruned_loss=0.1268, over 19316.00 frames. ], tot_loss[loss=0.2765, simple_loss=0.339, pruned_loss=0.107, over 3818599.85 frames. ], batch size: 70, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:16:57,882 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36248.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:17:17,819 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36263.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:17:18,609 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 07:17:39,999 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 07:17:49,959 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36288.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:17:50,005 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36288.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:17:51,848 INFO [train.py:903] (0/4) Epoch 6, batch 2150, loss[loss=0.3113, simple_loss=0.3486, pruned_loss=0.137, over 19764.00 frames. ], tot_loss[loss=0.2764, simple_loss=0.3389, pruned_loss=0.107, over 3807505.26 frames. ], batch size: 45, lr: 1.40e-02, grad_scale: 16.0 +2023-04-01 07:18:39,882 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6531, 1.2110, 1.2657, 1.8989, 1.6086, 1.9132, 2.0942, 1.6879], + device='cuda:0'), covar=tensor([0.0883, 0.1197, 0.1269, 0.1070, 0.1099, 0.0820, 0.0968, 0.0793], + device='cuda:0'), in_proj_covar=tensor([0.0231, 0.0247, 0.0241, 0.0275, 0.0269, 0.0230, 0.0226, 0.0220], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 07:18:53,906 INFO [train.py:903] (0/4) Epoch 6, batch 2200, loss[loss=0.2385, simple_loss=0.3012, pruned_loss=0.08784, over 19802.00 frames. ], tot_loss[loss=0.277, simple_loss=0.3391, pruned_loss=0.1074, over 3800101.49 frames. ], batch size: 48, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:18:55,064 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.621e+02 6.300e+02 8.031e+02 1.073e+03 2.013e+03, threshold=1.606e+03, percent-clipped=1.0 +2023-04-01 07:19:57,016 INFO [train.py:903] (0/4) Epoch 6, batch 2250, loss[loss=0.2491, simple_loss=0.3084, pruned_loss=0.09493, over 19740.00 frames. ], tot_loss[loss=0.2747, simple_loss=0.3375, pruned_loss=0.1059, over 3818150.56 frames. ], batch size: 51, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:20:40,675 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-01 07:20:58,383 INFO [train.py:903] (0/4) Epoch 6, batch 2300, loss[loss=0.2359, simple_loss=0.3079, pruned_loss=0.08198, over 19679.00 frames. ], tot_loss[loss=0.2739, simple_loss=0.3367, pruned_loss=0.1056, over 3814499.37 frames. ], batch size: 60, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:20:59,563 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.730e+02 6.689e+02 7.482e+02 9.822e+02 1.768e+03, threshold=1.496e+03, percent-clipped=2.0 +2023-04-01 07:21:12,179 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8568, 4.3072, 2.5176, 3.8222, 0.9972, 3.8455, 4.0424, 4.1764], + device='cuda:0'), covar=tensor([0.0532, 0.1097, 0.2077, 0.0680, 0.4112, 0.0933, 0.0668, 0.0898], + device='cuda:0'), in_proj_covar=tensor([0.0362, 0.0321, 0.0375, 0.0295, 0.0358, 0.0319, 0.0294, 0.0329], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 07:21:12,407 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3889, 2.2394, 1.7972, 1.7406, 1.6275, 1.9006, 0.2608, 1.1921], + device='cuda:0'), covar=tensor([0.0261, 0.0267, 0.0237, 0.0349, 0.0528, 0.0373, 0.0616, 0.0509], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0300, 0.0297, 0.0316, 0.0384, 0.0307, 0.0288, 0.0302], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 07:21:14,333 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 07:22:00,468 INFO [train.py:903] (0/4) Epoch 6, batch 2350, loss[loss=0.2261, simple_loss=0.2968, pruned_loss=0.07767, over 19132.00 frames. ], tot_loss[loss=0.2718, simple_loss=0.3352, pruned_loss=0.1042, over 3830803.61 frames. ], batch size: 42, lr: 1.40e-02, grad_scale: 8.0 +2023-04-01 07:22:19,324 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36504.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:22:43,460 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 07:22:49,414 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36529.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:23:00,577 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 07:23:02,880 INFO [train.py:903] (0/4) Epoch 6, batch 2400, loss[loss=0.2696, simple_loss=0.344, pruned_loss=0.09758, over 19645.00 frames. ], tot_loss[loss=0.2708, simple_loss=0.3347, pruned_loss=0.1035, over 3824590.02 frames. ], batch size: 58, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:23:04,017 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.742e+02 5.696e+02 7.249e+02 9.157e+02 1.479e+03, threshold=1.450e+03, percent-clipped=0.0 +2023-04-01 07:23:07,411 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36543.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:23:08,776 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36544.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:23:39,050 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36569.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:24:06,969 INFO [train.py:903] (0/4) Epoch 6, batch 2450, loss[loss=0.301, simple_loss=0.3593, pruned_loss=0.1213, over 17419.00 frames. ], tot_loss[loss=0.272, simple_loss=0.336, pruned_loss=0.104, over 3813282.99 frames. ], batch size: 101, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:24:27,654 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-04-01 07:24:46,725 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36623.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:24:51,039 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36626.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:24:56,976 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36630.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:25:07,680 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.57 vs. limit=5.0 +2023-04-01 07:25:07,911 INFO [train.py:903] (0/4) Epoch 6, batch 2500, loss[loss=0.3053, simple_loss=0.3671, pruned_loss=0.1218, over 19611.00 frames. ], tot_loss[loss=0.2728, simple_loss=0.3362, pruned_loss=0.1047, over 3816080.04 frames. ], batch size: 57, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:25:09,078 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.995e+02 6.265e+02 8.006e+02 9.274e+02 1.564e+03, threshold=1.601e+03, percent-clipped=1.0 +2023-04-01 07:25:30,299 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=36658.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:25:39,443 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1352, 0.9929, 0.9895, 1.2265, 1.1470, 1.2251, 1.2990, 1.2347], + device='cuda:0'), covar=tensor([0.0956, 0.1116, 0.1225, 0.0748, 0.0840, 0.0899, 0.0870, 0.0771], + device='cuda:0'), in_proj_covar=tensor([0.0227, 0.0244, 0.0235, 0.0268, 0.0259, 0.0226, 0.0227, 0.0215], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 07:26:09,592 INFO [train.py:903] (0/4) Epoch 6, batch 2550, loss[loss=0.3042, simple_loss=0.359, pruned_loss=0.1248, over 19687.00 frames. ], tot_loss[loss=0.2723, simple_loss=0.3358, pruned_loss=0.1044, over 3822704.08 frames. ], batch size: 58, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:26:44,788 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36718.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 07:26:46,975 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0939, 1.2428, 1.7440, 1.4583, 2.7608, 4.4060, 4.4446, 4.9526], + device='cuda:0'), covar=tensor([0.1548, 0.3051, 0.2894, 0.1867, 0.0480, 0.0143, 0.0132, 0.0062], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0282, 0.0312, 0.0247, 0.0200, 0.0122, 0.0200, 0.0148], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 07:26:53,889 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.9934, 5.3749, 2.8385, 4.7716, 0.9231, 5.0821, 5.2173, 5.3478], + device='cuda:0'), covar=tensor([0.0419, 0.0861, 0.1974, 0.0494, 0.4471, 0.0721, 0.0530, 0.0659], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0314, 0.0372, 0.0291, 0.0353, 0.0313, 0.0294, 0.0327], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 07:27:05,101 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 07:27:10,912 INFO [train.py:903] (0/4) Epoch 6, batch 2600, loss[loss=0.2353, simple_loss=0.2996, pruned_loss=0.08548, over 19393.00 frames. ], tot_loss[loss=0.2733, simple_loss=0.3367, pruned_loss=0.1049, over 3814954.94 frames. ], batch size: 48, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:27:12,677 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.869e+02 6.492e+02 8.253e+02 1.085e+03 2.742e+03, threshold=1.651e+03, percent-clipped=10.0 +2023-04-01 07:28:14,515 INFO [train.py:903] (0/4) Epoch 6, batch 2650, loss[loss=0.2894, simple_loss=0.3578, pruned_loss=0.1105, over 19787.00 frames. ], tot_loss[loss=0.273, simple_loss=0.3363, pruned_loss=0.1048, over 3808836.02 frames. ], batch size: 56, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:28:37,569 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 07:28:54,596 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36823.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 07:29:05,797 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8114, 1.3479, 0.9570, 0.9248, 1.2166, 0.8750, 0.8523, 1.1787], + device='cuda:0'), covar=tensor([0.0488, 0.0490, 0.0881, 0.0456, 0.0400, 0.0980, 0.0497, 0.0332], + device='cuda:0'), in_proj_covar=tensor([0.0283, 0.0278, 0.0312, 0.0240, 0.0226, 0.0312, 0.0282, 0.0250], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:29:16,460 INFO [train.py:903] (0/4) Epoch 6, batch 2700, loss[loss=0.3397, simple_loss=0.3861, pruned_loss=0.1467, over 19145.00 frames. ], tot_loss[loss=0.2733, simple_loss=0.3367, pruned_loss=0.1049, over 3807229.39 frames. ], batch size: 69, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:29:17,596 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.977e+02 6.342e+02 7.427e+02 9.812e+02 2.890e+03, threshold=1.485e+03, percent-clipped=2.0 +2023-04-01 07:30:18,807 INFO [train.py:903] (0/4) Epoch 6, batch 2750, loss[loss=0.2964, simple_loss=0.3553, pruned_loss=0.1187, over 19774.00 frames. ], tot_loss[loss=0.2741, simple_loss=0.3373, pruned_loss=0.1054, over 3809799.63 frames. ], batch size: 56, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:30:51,769 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=36914.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:31:04,563 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9885, 1.0089, 1.4046, 0.4349, 2.2833, 2.3972, 2.1139, 2.5395], + device='cuda:0'), covar=tensor([0.1405, 0.3241, 0.2992, 0.2208, 0.0407, 0.0202, 0.0372, 0.0225], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0284, 0.0312, 0.0247, 0.0202, 0.0123, 0.0202, 0.0152], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 07:31:09,568 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36929.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:31:22,223 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=36939.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:31:23,026 INFO [train.py:903] (0/4) Epoch 6, batch 2800, loss[loss=0.212, simple_loss=0.2825, pruned_loss=0.07072, over 15677.00 frames. ], tot_loss[loss=0.2725, simple_loss=0.3361, pruned_loss=0.1045, over 3808613.15 frames. ], batch size: 34, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:31:24,219 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.898e+02 7.168e+02 8.701e+02 1.243e+03 3.330e+03, threshold=1.740e+03, percent-clipped=17.0 +2023-04-01 07:31:56,914 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36967.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:32:00,390 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36970.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:32:06,257 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=36974.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:32:08,504 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=36976.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:32:26,414 INFO [train.py:903] (0/4) Epoch 6, batch 2850, loss[loss=0.2361, simple_loss=0.3139, pruned_loss=0.07917, over 19653.00 frames. ], tot_loss[loss=0.2723, simple_loss=0.3361, pruned_loss=0.1042, over 3808058.09 frames. ], batch size: 55, lr: 1.39e-02, grad_scale: 8.0 +2023-04-01 07:32:34,757 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2345, 2.1485, 2.1274, 3.4530, 2.0058, 3.2954, 2.9926, 2.0480], + device='cuda:0'), covar=tensor([0.2522, 0.1991, 0.0920, 0.1098, 0.2574, 0.0738, 0.1745, 0.1750], + device='cuda:0'), in_proj_covar=tensor([0.0661, 0.0653, 0.0573, 0.0804, 0.0685, 0.0567, 0.0701, 0.0604], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 07:32:48,220 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1751, 3.7086, 3.8085, 3.7852, 1.2379, 3.5354, 3.1335, 3.4546], + device='cuda:0'), covar=tensor([0.1051, 0.0572, 0.0550, 0.0529, 0.4183, 0.0461, 0.0595, 0.1063], + device='cuda:0'), in_proj_covar=tensor([0.0546, 0.0471, 0.0641, 0.0527, 0.0601, 0.0395, 0.0410, 0.0601], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 07:32:52,096 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-04-01 07:33:05,605 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37022.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:33:05,639 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3086, 1.3390, 1.3500, 1.6020, 2.8726, 0.9789, 1.8894, 2.9745], + device='cuda:0'), covar=tensor([0.0367, 0.2274, 0.2343, 0.1331, 0.0552, 0.2197, 0.1250, 0.0393], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0315, 0.0315, 0.0290, 0.0307, 0.0314, 0.0289, 0.0305], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:33:28,652 INFO [train.py:903] (0/4) Epoch 6, batch 2900, loss[loss=0.2901, simple_loss=0.3524, pruned_loss=0.1139, over 19349.00 frames. ], tot_loss[loss=0.2733, simple_loss=0.3371, pruned_loss=0.1047, over 3811091.80 frames. ], batch size: 66, lr: 1.38e-02, grad_scale: 8.0 +2023-04-01 07:33:28,669 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 07:33:28,987 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6781, 4.1221, 4.3932, 4.3212, 1.5587, 3.9718, 3.5548, 3.9777], + device='cuda:0'), covar=tensor([0.1027, 0.0659, 0.0558, 0.0480, 0.4297, 0.0446, 0.0578, 0.1073], + device='cuda:0'), in_proj_covar=tensor([0.0546, 0.0470, 0.0639, 0.0526, 0.0598, 0.0394, 0.0407, 0.0598], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 07:33:29,873 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.185e+02 6.105e+02 7.947e+02 1.025e+03 2.308e+03, threshold=1.589e+03, percent-clipped=2.0 +2023-04-01 07:33:55,100 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37062.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 07:33:58,580 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8741, 1.4930, 1.4229, 1.8322, 1.6849, 1.7091, 1.5447, 1.7760], + device='cuda:0'), covar=tensor([0.0751, 0.1270, 0.1205, 0.0817, 0.1006, 0.0419, 0.0929, 0.0562], + device='cuda:0'), in_proj_covar=tensor([0.0248, 0.0355, 0.0279, 0.0235, 0.0299, 0.0238, 0.0263, 0.0222], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:34:20,465 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37082.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:34:24,021 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37085.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:34:28,758 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37089.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:34:29,562 INFO [train.py:903] (0/4) Epoch 6, batch 2950, loss[loss=0.2339, simple_loss=0.3096, pruned_loss=0.07912, over 19762.00 frames. ], tot_loss[loss=0.274, simple_loss=0.3376, pruned_loss=0.1052, over 3796960.27 frames. ], batch size: 51, lr: 1.38e-02, grad_scale: 8.0 +2023-04-01 07:34:58,745 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2737, 3.6637, 3.7832, 3.7945, 1.4793, 3.5239, 3.1811, 3.4470], + device='cuda:0'), covar=tensor([0.1085, 0.0764, 0.0637, 0.0513, 0.3981, 0.0504, 0.0574, 0.1193], + device='cuda:0'), in_proj_covar=tensor([0.0558, 0.0482, 0.0651, 0.0532, 0.0607, 0.0398, 0.0414, 0.0610], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 07:35:31,165 INFO [train.py:903] (0/4) Epoch 6, batch 3000, loss[loss=0.2682, simple_loss=0.336, pruned_loss=0.1002, over 19772.00 frames. ], tot_loss[loss=0.2743, simple_loss=0.3378, pruned_loss=0.1055, over 3811803.86 frames. ], batch size: 54, lr: 1.38e-02, grad_scale: 8.0 +2023-04-01 07:35:31,166 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 07:35:43,641 INFO [train.py:937] (0/4) Epoch 6, validation: loss=0.1968, simple_loss=0.2962, pruned_loss=0.04867, over 944034.00 frames. +2023-04-01 07:35:43,644 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 07:35:44,844 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.497e+02 6.001e+02 7.289e+02 9.626e+02 1.809e+03, threshold=1.458e+03, percent-clipped=5.0 +2023-04-01 07:35:48,627 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 07:36:18,742 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37167.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 07:36:30,242 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37177.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 07:36:33,604 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8723, 1.7837, 1.8134, 2.0787, 4.2306, 1.2529, 2.5143, 4.2966], + device='cuda:0'), covar=tensor([0.0308, 0.2445, 0.2345, 0.1391, 0.0530, 0.2206, 0.1132, 0.0303], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0318, 0.0318, 0.0290, 0.0312, 0.0315, 0.0290, 0.0309], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:36:45,878 INFO [train.py:903] (0/4) Epoch 6, batch 3050, loss[loss=0.2593, simple_loss=0.3128, pruned_loss=0.1029, over 19784.00 frames. ], tot_loss[loss=0.2749, simple_loss=0.3376, pruned_loss=0.1061, over 3800805.14 frames. ], batch size: 47, lr: 1.38e-02, grad_scale: 8.0 +2023-04-01 07:37:34,453 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7676, 1.8115, 1.5042, 1.3523, 1.2771, 1.4530, 0.0883, 0.6458], + device='cuda:0'), covar=tensor([0.0221, 0.0238, 0.0165, 0.0220, 0.0527, 0.0230, 0.0504, 0.0478], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0298, 0.0295, 0.0312, 0.0384, 0.0308, 0.0283, 0.0303], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 07:37:34,884 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-04-01 07:37:48,506 INFO [train.py:903] (0/4) Epoch 6, batch 3100, loss[loss=0.2859, simple_loss=0.3519, pruned_loss=0.11, over 18784.00 frames. ], tot_loss[loss=0.2744, simple_loss=0.3375, pruned_loss=0.1056, over 3810487.45 frames. ], batch size: 74, lr: 1.38e-02, grad_scale: 8.0 +2023-04-01 07:37:49,789 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.764e+02 6.699e+02 8.375e+02 1.038e+03 2.239e+03, threshold=1.675e+03, percent-clipped=7.0 +2023-04-01 07:38:28,933 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37273.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:38:41,342 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37282.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 07:38:46,565 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37286.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:38:50,728 INFO [train.py:903] (0/4) Epoch 6, batch 3150, loss[loss=0.2407, simple_loss=0.3038, pruned_loss=0.08884, over 19486.00 frames. ], tot_loss[loss=0.2739, simple_loss=0.3377, pruned_loss=0.105, over 3815780.66 frames. ], batch size: 49, lr: 1.38e-02, grad_scale: 8.0 +2023-04-01 07:38:52,157 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7189, 4.0965, 4.3609, 4.3189, 1.5653, 3.9210, 3.5817, 3.9849], + device='cuda:0'), covar=tensor([0.1105, 0.0713, 0.0491, 0.0488, 0.4257, 0.0465, 0.0542, 0.0976], + device='cuda:0'), in_proj_covar=tensor([0.0549, 0.0472, 0.0642, 0.0531, 0.0604, 0.0392, 0.0407, 0.0598], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 07:38:54,763 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-01 07:39:13,511 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 07:39:27,509 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37320.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:39:29,898 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7845, 1.4239, 1.6355, 2.2437, 1.9494, 1.7165, 2.1008, 1.7152], + device='cuda:0'), covar=tensor([0.0927, 0.1442, 0.1179, 0.0920, 0.1041, 0.1179, 0.1041, 0.1008], + device='cuda:0'), in_proj_covar=tensor([0.0228, 0.0243, 0.0237, 0.0273, 0.0262, 0.0229, 0.0226, 0.0218], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 07:39:49,074 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37338.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:39:51,054 INFO [train.py:903] (0/4) Epoch 6, batch 3200, loss[loss=0.2729, simple_loss=0.3382, pruned_loss=0.1038, over 17331.00 frames. ], tot_loss[loss=0.2726, simple_loss=0.3362, pruned_loss=0.1045, over 3807847.46 frames. ], batch size: 101, lr: 1.38e-02, grad_scale: 8.0 +2023-04-01 07:39:52,147 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.326e+02 6.253e+02 8.171e+02 9.916e+02 1.975e+03, threshold=1.634e+03, percent-clipped=4.0 +2023-04-01 07:39:52,616 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37341.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:39:58,078 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37345.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:40:00,253 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6965, 1.4816, 1.2789, 1.5172, 3.1305, 0.9462, 2.0880, 3.3614], + device='cuda:0'), covar=tensor([0.0381, 0.2410, 0.2626, 0.1526, 0.0597, 0.2431, 0.1248, 0.0349], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0318, 0.0317, 0.0294, 0.0313, 0.0316, 0.0291, 0.0310], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:40:20,405 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37363.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:40:24,434 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37366.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:40:24,668 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37366.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:40:29,132 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37370.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:40:42,587 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37382.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:40:50,639 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37388.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:40:52,610 INFO [train.py:903] (0/4) Epoch 6, batch 3250, loss[loss=0.2818, simple_loss=0.3517, pruned_loss=0.1059, over 19534.00 frames. ], tot_loss[loss=0.2724, simple_loss=0.3359, pruned_loss=0.1044, over 3827124.21 frames. ], batch size: 56, lr: 1.38e-02, grad_scale: 4.0 +2023-04-01 07:41:30,770 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9504, 1.3554, 1.0202, 0.9216, 1.2019, 0.8352, 0.8273, 1.2695], + device='cuda:0'), covar=tensor([0.0403, 0.0478, 0.0891, 0.0433, 0.0361, 0.0933, 0.0460, 0.0289], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0276, 0.0309, 0.0241, 0.0221, 0.0310, 0.0280, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:41:45,866 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37433.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 07:41:48,020 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37435.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:41:53,540 INFO [train.py:903] (0/4) Epoch 6, batch 3300, loss[loss=0.2632, simple_loss=0.3308, pruned_loss=0.09778, over 19608.00 frames. ], tot_loss[loss=0.2716, simple_loss=0.3352, pruned_loss=0.104, over 3843328.06 frames. ], batch size: 52, lr: 1.38e-02, grad_scale: 4.0 +2023-04-01 07:41:57,421 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.613e+02 6.108e+02 8.102e+02 1.002e+03 3.053e+03, threshold=1.620e+03, percent-clipped=3.0 +2023-04-01 07:42:01,121 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 07:42:17,161 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37458.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 07:42:44,790 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37481.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:42:56,307 INFO [train.py:903] (0/4) Epoch 6, batch 3350, loss[loss=0.2963, simple_loss=0.3533, pruned_loss=0.1197, over 18263.00 frames. ], tot_loss[loss=0.2722, simple_loss=0.3351, pruned_loss=0.1046, over 3841669.93 frames. ], batch size: 83, lr: 1.38e-02, grad_scale: 4.0 +2023-04-01 07:43:56,162 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37538.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 07:43:57,998 INFO [train.py:903] (0/4) Epoch 6, batch 3400, loss[loss=0.2812, simple_loss=0.3492, pruned_loss=0.1066, over 19493.00 frames. ], tot_loss[loss=0.2714, simple_loss=0.3346, pruned_loss=0.1041, over 3856386.85 frames. ], batch size: 64, lr: 1.38e-02, grad_scale: 4.0 +2023-04-01 07:44:00,246 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.822e+02 6.377e+02 8.364e+02 1.096e+03 2.128e+03, threshold=1.673e+03, percent-clipped=5.0 +2023-04-01 07:44:26,239 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37563.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 07:44:59,262 INFO [train.py:903] (0/4) Epoch 6, batch 3450, loss[loss=0.2476, simple_loss=0.3152, pruned_loss=0.09002, over 19611.00 frames. ], tot_loss[loss=0.2726, simple_loss=0.3356, pruned_loss=0.1048, over 3827651.45 frames. ], batch size: 50, lr: 1.37e-02, grad_scale: 4.0 +2023-04-01 07:45:07,201 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 07:45:12,647 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-01 07:45:39,230 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8611, 3.4834, 2.2152, 3.2294, 0.8188, 3.2051, 3.2031, 3.3197], + device='cuda:0'), covar=tensor([0.0824, 0.1142, 0.2227, 0.0787, 0.4022, 0.0971, 0.0852, 0.1051], + device='cuda:0'), in_proj_covar=tensor([0.0366, 0.0324, 0.0384, 0.0294, 0.0360, 0.0319, 0.0297, 0.0332], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 07:45:49,554 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37630.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:46:02,045 INFO [train.py:903] (0/4) Epoch 6, batch 3500, loss[loss=0.242, simple_loss=0.3032, pruned_loss=0.09035, over 19759.00 frames. ], tot_loss[loss=0.2718, simple_loss=0.3351, pruned_loss=0.1043, over 3831427.40 frames. ], batch size: 46, lr: 1.37e-02, grad_scale: 4.0 +2023-04-01 07:46:04,580 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.712e+02 6.340e+02 8.060e+02 1.060e+03 3.220e+03, threshold=1.612e+03, percent-clipped=3.0 +2023-04-01 07:46:07,389 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37644.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:46:23,301 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5914, 1.8802, 1.6922, 2.2032, 1.8951, 2.5498, 2.7438, 2.3249], + device='cuda:0'), covar=tensor([0.0667, 0.0949, 0.1088, 0.1150, 0.1022, 0.0734, 0.0861, 0.0677], + device='cuda:0'), in_proj_covar=tensor([0.0226, 0.0244, 0.0237, 0.0274, 0.0262, 0.0230, 0.0226, 0.0219], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005, 0.0004], + device='cuda:0') +2023-04-01 07:46:38,285 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37669.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:46:54,379 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9784, 1.0726, 1.2748, 1.4772, 2.5517, 0.9827, 1.8945, 2.6750], + device='cuda:0'), covar=tensor([0.0490, 0.2764, 0.2522, 0.1437, 0.0673, 0.2314, 0.1083, 0.0473], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0316, 0.0311, 0.0287, 0.0311, 0.0312, 0.0288, 0.0304], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:47:04,856 INFO [train.py:903] (0/4) Epoch 6, batch 3550, loss[loss=0.3101, simple_loss=0.3836, pruned_loss=0.1184, over 18738.00 frames. ], tot_loss[loss=0.2718, simple_loss=0.3351, pruned_loss=0.1042, over 3842628.93 frames. ], batch size: 74, lr: 1.37e-02, grad_scale: 4.0 +2023-04-01 07:47:07,128 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37691.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:47:12,717 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7490, 0.7652, 0.8909, 0.9637, 1.5533, 0.7546, 1.4222, 1.5215], + device='cuda:0'), covar=tensor([0.0459, 0.1812, 0.1704, 0.1002, 0.0562, 0.1379, 0.1094, 0.0479], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0318, 0.0315, 0.0288, 0.0312, 0.0313, 0.0288, 0.0305], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 07:47:34,876 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37716.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:47:48,282 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=37726.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:48:03,026 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=37737.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:48:05,908 INFO [train.py:903] (0/4) Epoch 6, batch 3600, loss[loss=0.3157, simple_loss=0.3694, pruned_loss=0.131, over 19770.00 frames. ], tot_loss[loss=0.2738, simple_loss=0.3363, pruned_loss=0.1057, over 3839293.78 frames. ], batch size: 63, lr: 1.37e-02, grad_scale: 8.0 +2023-04-01 07:48:06,563 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-01 07:48:08,231 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.206e+02 7.040e+02 8.572e+02 1.227e+03 4.209e+03, threshold=1.714e+03, percent-clipped=12.0 +2023-04-01 07:48:12,136 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37745.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:48:33,220 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=37762.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:49:08,254 INFO [train.py:903] (0/4) Epoch 6, batch 3650, loss[loss=0.274, simple_loss=0.3451, pruned_loss=0.1015, over 19519.00 frames. ], tot_loss[loss=0.2738, simple_loss=0.3363, pruned_loss=0.1056, over 3828778.13 frames. ], batch size: 56, lr: 1.37e-02, grad_scale: 8.0 +2023-04-01 07:50:08,368 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4020, 2.9863, 1.9386, 2.4126, 2.3429, 2.6269, 0.7658, 2.2270], + device='cuda:0'), covar=tensor([0.0326, 0.0270, 0.0328, 0.0399, 0.0513, 0.0408, 0.0688, 0.0479], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0297, 0.0295, 0.0312, 0.0384, 0.0305, 0.0282, 0.0298], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 07:50:10,246 INFO [train.py:903] (0/4) Epoch 6, batch 3700, loss[loss=0.2759, simple_loss=0.3401, pruned_loss=0.1058, over 18738.00 frames. ], tot_loss[loss=0.2757, simple_loss=0.3379, pruned_loss=0.1067, over 3809110.81 frames. ], batch size: 74, lr: 1.37e-02, grad_scale: 8.0 +2023-04-01 07:50:11,756 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=37841.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:50:12,527 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.880e+02 5.969e+02 7.272e+02 1.002e+03 1.787e+03, threshold=1.454e+03, percent-clipped=1.0 +2023-04-01 07:50:27,705 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-01 07:51:00,200 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6557, 4.0853, 4.3095, 4.2933, 1.6164, 3.9488, 3.5321, 3.9581], + device='cuda:0'), covar=tensor([0.0990, 0.0649, 0.0453, 0.0463, 0.3985, 0.0413, 0.0512, 0.0913], + device='cuda:0'), in_proj_covar=tensor([0.0551, 0.0473, 0.0651, 0.0534, 0.0606, 0.0398, 0.0411, 0.0603], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 07:51:13,478 INFO [train.py:903] (0/4) Epoch 6, batch 3750, loss[loss=0.3025, simple_loss=0.3551, pruned_loss=0.1249, over 12912.00 frames. ], tot_loss[loss=0.2757, simple_loss=0.3382, pruned_loss=0.1066, over 3812285.83 frames. ], batch size: 135, lr: 1.37e-02, grad_scale: 8.0 +2023-04-01 07:52:16,185 INFO [train.py:903] (0/4) Epoch 6, batch 3800, loss[loss=0.3125, simple_loss=0.3657, pruned_loss=0.1296, over 17195.00 frames. ], tot_loss[loss=0.2746, simple_loss=0.3377, pruned_loss=0.1058, over 3812551.00 frames. ], batch size: 101, lr: 1.37e-02, grad_scale: 8.0 +2023-04-01 07:52:18,422 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.239e+02 5.762e+02 7.572e+02 9.178e+02 2.007e+03, threshold=1.514e+03, percent-clipped=4.0 +2023-04-01 07:52:47,795 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 07:53:13,676 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.57 vs. limit=2.0 +2023-04-01 07:53:17,773 INFO [train.py:903] (0/4) Epoch 6, batch 3850, loss[loss=0.2365, simple_loss=0.31, pruned_loss=0.08154, over 19735.00 frames. ], tot_loss[loss=0.273, simple_loss=0.3364, pruned_loss=0.1048, over 3821760.03 frames. ], batch size: 51, lr: 1.37e-02, grad_scale: 8.0 +2023-04-01 07:53:25,802 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=37997.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:53:30,386 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-38000.pt +2023-04-01 07:53:32,719 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=38001.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:54:03,070 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=38026.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:54:20,062 INFO [train.py:903] (0/4) Epoch 6, batch 3900, loss[loss=0.3228, simple_loss=0.3735, pruned_loss=0.136, over 13042.00 frames. ], tot_loss[loss=0.2738, simple_loss=0.3367, pruned_loss=0.1055, over 3816061.04 frames. ], batch size: 136, lr: 1.37e-02, grad_scale: 8.0 +2023-04-01 07:54:22,365 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.032e+02 6.773e+02 7.927e+02 9.711e+02 2.220e+03, threshold=1.585e+03, percent-clipped=5.0 +2023-04-01 07:54:59,167 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-01 07:55:24,014 INFO [train.py:903] (0/4) Epoch 6, batch 3950, loss[loss=0.2846, simple_loss=0.3513, pruned_loss=0.1089, over 18335.00 frames. ], tot_loss[loss=0.2735, simple_loss=0.3365, pruned_loss=0.1053, over 3810103.93 frames. ], batch size: 83, lr: 1.37e-02, grad_scale: 8.0 +2023-04-01 07:55:27,718 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 07:55:33,054 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=38097.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:55:35,397 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8174, 1.8812, 1.8444, 2.5563, 1.8445, 2.4054, 2.3785, 1.7767], + device='cuda:0'), covar=tensor([0.2254, 0.1765, 0.0972, 0.1151, 0.2017, 0.0869, 0.1929, 0.1761], + device='cuda:0'), in_proj_covar=tensor([0.0673, 0.0671, 0.0585, 0.0828, 0.0699, 0.0593, 0.0717, 0.0626], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 07:56:03,022 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=38122.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 07:56:25,972 INFO [train.py:903] (0/4) Epoch 6, batch 4000, loss[loss=0.2878, simple_loss=0.3603, pruned_loss=0.1076, over 19499.00 frames. ], tot_loss[loss=0.2717, simple_loss=0.3351, pruned_loss=0.1041, over 3808104.52 frames. ], batch size: 64, lr: 1.37e-02, grad_scale: 8.0 +2023-04-01 07:56:28,272 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.559e+02 6.005e+02 7.523e+02 9.814e+02 1.567e+03, threshold=1.505e+03, percent-clipped=0.0 +2023-04-01 07:56:52,235 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-04-01 07:57:10,251 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 07:57:25,998 INFO [train.py:903] (0/4) Epoch 6, batch 4050, loss[loss=0.2336, simple_loss=0.2925, pruned_loss=0.08735, over 19777.00 frames. ], tot_loss[loss=0.271, simple_loss=0.3348, pruned_loss=0.1036, over 3819049.21 frames. ], batch size: 45, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 07:58:23,646 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.77 vs. limit=2.0 +2023-04-01 07:58:27,533 INFO [train.py:903] (0/4) Epoch 6, batch 4100, loss[loss=0.2737, simple_loss=0.3294, pruned_loss=0.1091, over 19612.00 frames. ], tot_loss[loss=0.2702, simple_loss=0.3338, pruned_loss=0.1032, over 3817225.32 frames. ], batch size: 50, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 07:58:30,599 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.078e+02 6.486e+02 7.817e+02 9.790e+02 2.532e+03, threshold=1.563e+03, percent-clipped=8.0 +2023-04-01 07:59:04,836 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 07:59:30,609 INFO [train.py:903] (0/4) Epoch 6, batch 4150, loss[loss=0.2572, simple_loss=0.3333, pruned_loss=0.09058, over 19552.00 frames. ], tot_loss[loss=0.27, simple_loss=0.3337, pruned_loss=0.1032, over 3824076.63 frames. ], batch size: 61, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 08:00:34,887 INFO [train.py:903] (0/4) Epoch 6, batch 4200, loss[loss=0.2961, simple_loss=0.3561, pruned_loss=0.1181, over 19375.00 frames. ], tot_loss[loss=0.2702, simple_loss=0.3338, pruned_loss=0.1033, over 3828444.48 frames. ], batch size: 70, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 08:00:36,308 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=38341.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:00:37,311 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.522e+02 6.104e+02 7.462e+02 9.650e+02 2.123e+03, threshold=1.492e+03, percent-clipped=5.0 +2023-04-01 08:00:38,135 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.68 vs. limit=5.0 +2023-04-01 08:00:39,355 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 08:01:21,315 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=38378.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:01:34,728 INFO [train.py:903] (0/4) Epoch 6, batch 4250, loss[loss=0.3687, simple_loss=0.4032, pruned_loss=0.167, over 19195.00 frames. ], tot_loss[loss=0.2699, simple_loss=0.3338, pruned_loss=0.103, over 3827846.27 frames. ], batch size: 69, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 08:01:48,255 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 08:02:01,581 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 08:02:34,849 INFO [train.py:903] (0/4) Epoch 6, batch 4300, loss[loss=0.236, simple_loss=0.3071, pruned_loss=0.0825, over 19857.00 frames. ], tot_loss[loss=0.2677, simple_loss=0.3317, pruned_loss=0.1019, over 3836522.57 frames. ], batch size: 52, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 08:02:37,125 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.280e+02 6.582e+02 8.580e+02 1.078e+03 2.349e+03, threshold=1.716e+03, percent-clipped=8.0 +2023-04-01 08:02:56,002 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=38456.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:03:23,344 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6681, 3.3634, 2.4917, 3.0429, 1.5548, 3.0229, 3.0083, 3.1277], + device='cuda:0'), covar=tensor([0.0811, 0.1039, 0.1719, 0.0719, 0.2733, 0.0940, 0.0870, 0.1110], + device='cuda:0'), in_proj_covar=tensor([0.0367, 0.0320, 0.0379, 0.0291, 0.0358, 0.0317, 0.0302, 0.0334], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 08:03:27,717 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 08:03:35,923 INFO [train.py:903] (0/4) Epoch 6, batch 4350, loss[loss=0.2436, simple_loss=0.3104, pruned_loss=0.08844, over 19478.00 frames. ], tot_loss[loss=0.2684, simple_loss=0.3322, pruned_loss=0.1022, over 3828948.62 frames. ], batch size: 49, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 08:04:10,315 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2759, 3.7058, 3.8542, 3.8674, 1.3758, 3.5775, 3.2033, 3.4930], + device='cuda:0'), covar=tensor([0.1056, 0.0717, 0.0555, 0.0474, 0.4502, 0.0530, 0.0579, 0.1070], + device='cuda:0'), in_proj_covar=tensor([0.0560, 0.0479, 0.0657, 0.0528, 0.0616, 0.0398, 0.0414, 0.0608], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 08:04:33,255 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5994, 1.3202, 1.4414, 1.6668, 3.1830, 0.9578, 1.9721, 3.2801], + device='cuda:0'), covar=tensor([0.0326, 0.2226, 0.2180, 0.1296, 0.0533, 0.2222, 0.1188, 0.0313], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0316, 0.0319, 0.0292, 0.0316, 0.0311, 0.0291, 0.0310], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 08:04:40,411 INFO [train.py:903] (0/4) Epoch 6, batch 4400, loss[loss=0.2841, simple_loss=0.3547, pruned_loss=0.1068, over 19643.00 frames. ], tot_loss[loss=0.2687, simple_loss=0.3324, pruned_loss=0.1025, over 3831713.98 frames. ], batch size: 55, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 08:04:40,965 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-01 08:04:42,536 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.411e+02 6.588e+02 8.122e+02 1.160e+03 2.348e+03, threshold=1.624e+03, percent-clipped=4.0 +2023-04-01 08:05:05,910 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 08:05:13,676 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 08:05:30,775 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-04-01 08:05:40,416 INFO [train.py:903] (0/4) Epoch 6, batch 4450, loss[loss=0.2691, simple_loss=0.3393, pruned_loss=0.0995, over 19787.00 frames. ], tot_loss[loss=0.2686, simple_loss=0.3326, pruned_loss=0.1023, over 3838033.36 frames. ], batch size: 56, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 08:05:42,850 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3933, 1.1390, 1.3426, 0.8557, 2.3495, 2.9864, 2.7568, 3.1858], + device='cuda:0'), covar=tensor([0.1318, 0.3123, 0.3217, 0.2168, 0.0462, 0.0143, 0.0256, 0.0164], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0283, 0.0316, 0.0248, 0.0201, 0.0128, 0.0203, 0.0155], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:06:42,162 INFO [train.py:903] (0/4) Epoch 6, batch 4500, loss[loss=0.2729, simple_loss=0.3339, pruned_loss=0.1059, over 19595.00 frames. ], tot_loss[loss=0.2678, simple_loss=0.3318, pruned_loss=0.1019, over 3817444.34 frames. ], batch size: 50, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 08:06:44,523 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.035e+02 6.139e+02 7.598e+02 9.442e+02 2.713e+03, threshold=1.520e+03, percent-clipped=3.0 +2023-04-01 08:07:28,071 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3929, 1.2397, 1.7113, 1.5634, 3.0603, 4.2514, 4.4076, 4.8280], + device='cuda:0'), covar=tensor([0.1508, 0.3234, 0.3046, 0.1914, 0.0437, 0.0242, 0.0131, 0.0081], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0286, 0.0318, 0.0250, 0.0202, 0.0130, 0.0204, 0.0156], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:07:42,587 INFO [train.py:903] (0/4) Epoch 6, batch 4550, loss[loss=0.2745, simple_loss=0.3459, pruned_loss=0.1015, over 19682.00 frames. ], tot_loss[loss=0.2677, simple_loss=0.3315, pruned_loss=0.1019, over 3825459.51 frames. ], batch size: 55, lr: 1.36e-02, grad_scale: 8.0 +2023-04-01 08:07:53,123 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 08:08:10,876 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=38712.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:08:17,336 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 08:08:22,019 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=38722.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:08:41,573 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=38737.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:08:45,393 INFO [train.py:903] (0/4) Epoch 6, batch 4600, loss[loss=0.2638, simple_loss=0.3383, pruned_loss=0.09462, over 19712.00 frames. ], tot_loss[loss=0.2684, simple_loss=0.3323, pruned_loss=0.1023, over 3816978.42 frames. ], batch size: 59, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:08:47,715 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.208e+02 6.588e+02 8.068e+02 1.040e+03 1.807e+03, threshold=1.614e+03, percent-clipped=3.0 +2023-04-01 08:09:22,048 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-01 08:09:45,876 INFO [train.py:903] (0/4) Epoch 6, batch 4650, loss[loss=0.3364, simple_loss=0.3844, pruned_loss=0.1442, over 17481.00 frames. ], tot_loss[loss=0.2682, simple_loss=0.3324, pruned_loss=0.102, over 3819536.90 frames. ], batch size: 101, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:10:01,541 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 08:10:10,760 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0725, 5.3447, 2.8666, 4.6591, 1.6195, 5.3028, 5.1947, 5.3958], + device='cuda:0'), covar=tensor([0.0388, 0.0843, 0.1712, 0.0491, 0.3242, 0.0559, 0.0612, 0.0660], + device='cuda:0'), in_proj_covar=tensor([0.0370, 0.0324, 0.0385, 0.0291, 0.0357, 0.0319, 0.0299, 0.0335], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 08:10:11,707 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 08:10:43,233 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=38837.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:10:46,234 INFO [train.py:903] (0/4) Epoch 6, batch 4700, loss[loss=0.2784, simple_loss=0.3392, pruned_loss=0.1088, over 19769.00 frames. ], tot_loss[loss=0.2701, simple_loss=0.3337, pruned_loss=0.1033, over 3819478.03 frames. ], batch size: 54, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:10:46,872 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-01 08:10:47,718 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=38841.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:10:48,620 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.118e+02 6.701e+02 8.528e+02 1.085e+03 2.106e+03, threshold=1.706e+03, percent-clipped=3.0 +2023-04-01 08:11:07,249 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 08:11:46,444 INFO [train.py:903] (0/4) Epoch 6, batch 4750, loss[loss=0.209, simple_loss=0.2809, pruned_loss=0.06851, over 19778.00 frames. ], tot_loss[loss=0.2695, simple_loss=0.3335, pruned_loss=0.1028, over 3819821.92 frames. ], batch size: 48, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:12:40,567 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8125, 1.2752, 1.3400, 1.4215, 2.4426, 0.9406, 1.7694, 2.5296], + device='cuda:0'), covar=tensor([0.0426, 0.2230, 0.2223, 0.1328, 0.0598, 0.2022, 0.0974, 0.0442], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0309, 0.0313, 0.0289, 0.0309, 0.0308, 0.0288, 0.0305], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 08:12:47,940 INFO [train.py:903] (0/4) Epoch 6, batch 4800, loss[loss=0.2965, simple_loss=0.3538, pruned_loss=0.1196, over 19660.00 frames. ], tot_loss[loss=0.2721, simple_loss=0.335, pruned_loss=0.1046, over 3816529.18 frames. ], batch size: 55, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:12:52,286 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.945e+02 6.641e+02 7.688e+02 1.063e+03 2.770e+03, threshold=1.538e+03, percent-clipped=4.0 +2023-04-01 08:13:49,887 INFO [train.py:903] (0/4) Epoch 6, batch 4850, loss[loss=0.2539, simple_loss=0.3267, pruned_loss=0.09052, over 19650.00 frames. ], tot_loss[loss=0.2723, simple_loss=0.3352, pruned_loss=0.1047, over 3817537.86 frames. ], batch size: 55, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:13:56,377 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0442, 2.8849, 1.8359, 2.0030, 1.8112, 2.1486, 0.4811, 1.9351], + device='cuda:0'), covar=tensor([0.0354, 0.0337, 0.0400, 0.0655, 0.0623, 0.0643, 0.0810, 0.0683], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0298, 0.0300, 0.0318, 0.0383, 0.0312, 0.0286, 0.0301], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:14:14,736 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 08:14:17,462 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4501, 1.1492, 1.6376, 1.3473, 2.8848, 3.6859, 3.4916, 3.9114], + device='cuda:0'), covar=tensor([0.1392, 0.3214, 0.3001, 0.1897, 0.0406, 0.0158, 0.0190, 0.0125], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0283, 0.0316, 0.0246, 0.0201, 0.0129, 0.0201, 0.0153], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:14:36,514 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 08:14:42,310 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 08:14:42,351 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 08:14:51,529 INFO [train.py:903] (0/4) Epoch 6, batch 4900, loss[loss=0.3132, simple_loss=0.3757, pruned_loss=0.1253, over 19591.00 frames. ], tot_loss[loss=0.2709, simple_loss=0.3342, pruned_loss=0.1038, over 3814380.34 frames. ], batch size: 57, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:14:51,635 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 08:14:55,127 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.255e+02 6.820e+02 8.455e+02 1.082e+03 3.554e+03, threshold=1.691e+03, percent-clipped=3.0 +2023-04-01 08:15:13,710 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 08:15:52,286 INFO [train.py:903] (0/4) Epoch 6, batch 4950, loss[loss=0.2342, simple_loss=0.3073, pruned_loss=0.08055, over 19624.00 frames. ], tot_loss[loss=0.2705, simple_loss=0.3342, pruned_loss=0.1034, over 3821751.62 frames. ], batch size: 50, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:15:56,199 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39093.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:16:11,576 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 08:16:27,947 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39118.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:16:34,752 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 08:16:54,830 INFO [train.py:903] (0/4) Epoch 6, batch 5000, loss[loss=0.2884, simple_loss=0.3434, pruned_loss=0.1168, over 19766.00 frames. ], tot_loss[loss=0.271, simple_loss=0.3343, pruned_loss=0.1039, over 3819567.90 frames. ], batch size: 54, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:16:58,439 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8621, 1.3645, 1.4505, 1.5278, 3.3435, 0.9245, 2.1391, 3.4949], + device='cuda:0'), covar=tensor([0.0340, 0.2416, 0.2401, 0.1631, 0.0603, 0.2516, 0.1173, 0.0356], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0317, 0.0320, 0.0297, 0.0318, 0.0316, 0.0296, 0.0314], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 08:16:59,190 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.473e+02 6.699e+02 7.802e+02 1.019e+03 2.317e+03, threshold=1.560e+03, percent-clipped=4.0 +2023-04-01 08:17:05,631 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 08:17:14,754 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 08:17:51,526 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=39185.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:17:57,114 INFO [train.py:903] (0/4) Epoch 6, batch 5050, loss[loss=0.2343, simple_loss=0.3173, pruned_loss=0.0756, over 19538.00 frames. ], tot_loss[loss=0.2709, simple_loss=0.334, pruned_loss=0.1039, over 3801850.12 frames. ], batch size: 56, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:18:09,742 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2618, 2.9288, 2.0184, 2.7103, 0.8665, 2.7880, 2.6748, 2.7989], + device='cuda:0'), covar=tensor([0.1092, 0.1484, 0.2211, 0.0965, 0.4134, 0.1175, 0.1024, 0.1398], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0318, 0.0381, 0.0290, 0.0356, 0.0316, 0.0297, 0.0332], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 08:18:12,171 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39203.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:18:29,793 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 08:18:57,179 INFO [train.py:903] (0/4) Epoch 6, batch 5100, loss[loss=0.3398, simple_loss=0.3908, pruned_loss=0.1444, over 19736.00 frames. ], tot_loss[loss=0.2714, simple_loss=0.3345, pruned_loss=0.1041, over 3790503.68 frames. ], batch size: 63, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:19:00,429 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.152e+02 6.811e+02 8.395e+02 1.062e+03 1.934e+03, threshold=1.679e+03, percent-clipped=3.0 +2023-04-01 08:19:05,462 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 08:19:08,950 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 08:19:13,355 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 08:19:57,721 INFO [train.py:903] (0/4) Epoch 6, batch 5150, loss[loss=0.2622, simple_loss=0.3376, pruned_loss=0.09339, over 19484.00 frames. ], tot_loss[loss=0.2722, simple_loss=0.3354, pruned_loss=0.1045, over 3783939.13 frames. ], batch size: 64, lr: 1.35e-02, grad_scale: 8.0 +2023-04-01 08:20:08,431 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 08:20:11,734 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39300.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:20:13,849 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39302.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:20:43,363 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 08:21:00,837 INFO [train.py:903] (0/4) Epoch 6, batch 5200, loss[loss=0.2354, simple_loss=0.296, pruned_loss=0.0874, over 19776.00 frames. ], tot_loss[loss=0.2716, simple_loss=0.3348, pruned_loss=0.1042, over 3788830.75 frames. ], batch size: 48, lr: 1.34e-02, grad_scale: 8.0 +2023-04-01 08:21:04,236 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.495e+02 6.191e+02 7.598e+02 1.014e+03 2.218e+03, threshold=1.520e+03, percent-clipped=2.0 +2023-04-01 08:21:14,285 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 08:21:45,102 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8355, 1.8853, 1.9244, 2.8513, 1.9646, 2.7117, 2.4115, 1.7405], + device='cuda:0'), covar=tensor([0.2725, 0.2163, 0.1157, 0.1241, 0.2317, 0.0878, 0.2247, 0.1998], + device='cuda:0'), in_proj_covar=tensor([0.0679, 0.0675, 0.0581, 0.0825, 0.0703, 0.0587, 0.0718, 0.0627], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 08:21:56,492 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 08:22:03,032 INFO [train.py:903] (0/4) Epoch 6, batch 5250, loss[loss=0.2622, simple_loss=0.3218, pruned_loss=0.1013, over 19586.00 frames. ], tot_loss[loss=0.2716, simple_loss=0.3347, pruned_loss=0.1042, over 3780214.77 frames. ], batch size: 52, lr: 1.34e-02, grad_scale: 8.0 +2023-04-01 08:23:05,659 INFO [train.py:903] (0/4) Epoch 6, batch 5300, loss[loss=0.2522, simple_loss=0.3088, pruned_loss=0.09775, over 19765.00 frames. ], tot_loss[loss=0.2719, simple_loss=0.3349, pruned_loss=0.1045, over 3777307.38 frames. ], batch size: 46, lr: 1.34e-02, grad_scale: 8.0 +2023-04-01 08:23:10,367 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.313e+02 6.384e+02 7.596e+02 9.799e+02 2.153e+03, threshold=1.519e+03, percent-clipped=7.0 +2023-04-01 08:23:22,443 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 08:23:56,530 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39481.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:24:06,722 INFO [train.py:903] (0/4) Epoch 6, batch 5350, loss[loss=0.2665, simple_loss=0.3354, pruned_loss=0.09882, over 19286.00 frames. ], tot_loss[loss=0.2704, simple_loss=0.3338, pruned_loss=0.1035, over 3797394.48 frames. ], batch size: 66, lr: 1.34e-02, grad_scale: 4.0 +2023-04-01 08:24:41,850 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 08:25:07,433 INFO [train.py:903] (0/4) Epoch 6, batch 5400, loss[loss=0.2628, simple_loss=0.3343, pruned_loss=0.09562, over 19737.00 frames. ], tot_loss[loss=0.2704, simple_loss=0.3339, pruned_loss=0.1034, over 3811293.06 frames. ], batch size: 63, lr: 1.34e-02, grad_scale: 4.0 +2023-04-01 08:25:12,677 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.51 vs. limit=5.0 +2023-04-01 08:25:15,198 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.754e+02 6.102e+02 7.285e+02 1.015e+03 2.320e+03, threshold=1.457e+03, percent-clipped=6.0 +2023-04-01 08:25:18,913 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=39547.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:25:30,309 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39556.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:26:00,334 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39581.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:26:08,227 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3048, 1.3171, 1.1534, 1.0928, 0.9875, 1.1702, 0.2397, 0.6103], + device='cuda:0'), covar=tensor([0.0206, 0.0222, 0.0134, 0.0171, 0.0377, 0.0206, 0.0419, 0.0341], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0302, 0.0299, 0.0315, 0.0385, 0.0312, 0.0288, 0.0301], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:26:13,207 INFO [train.py:903] (0/4) Epoch 6, batch 5450, loss[loss=0.2536, simple_loss=0.3228, pruned_loss=0.09222, over 19573.00 frames. ], tot_loss[loss=0.27, simple_loss=0.3336, pruned_loss=0.1032, over 3797010.86 frames. ], batch size: 52, lr: 1.34e-02, grad_scale: 4.0 +2023-04-01 08:26:44,785 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.2986, 3.8542, 2.4724, 3.4900, 1.2031, 3.4414, 3.5910, 3.6198], + device='cuda:0'), covar=tensor([0.0682, 0.1097, 0.2032, 0.0771, 0.3616, 0.1063, 0.0767, 0.1143], + device='cuda:0'), in_proj_covar=tensor([0.0373, 0.0330, 0.0386, 0.0297, 0.0365, 0.0322, 0.0304, 0.0337], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 08:27:13,643 INFO [train.py:903] (0/4) Epoch 6, batch 5500, loss[loss=0.2592, simple_loss=0.3267, pruned_loss=0.09586, over 19607.00 frames. ], tot_loss[loss=0.2711, simple_loss=0.3343, pruned_loss=0.104, over 3802067.58 frames. ], batch size: 57, lr: 1.34e-02, grad_scale: 4.0 +2023-04-01 08:27:18,176 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.238e+02 6.796e+02 8.775e+02 1.086e+03 2.226e+03, threshold=1.755e+03, percent-clipped=13.0 +2023-04-01 08:27:20,740 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=39646.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:27:33,892 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39657.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 08:27:36,852 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 08:27:40,434 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39662.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:28:14,678 INFO [train.py:903] (0/4) Epoch 6, batch 5550, loss[loss=0.3542, simple_loss=0.3998, pruned_loss=0.1543, over 19347.00 frames. ], tot_loss[loss=0.2713, simple_loss=0.3342, pruned_loss=0.1042, over 3818516.33 frames. ], batch size: 66, lr: 1.34e-02, grad_scale: 4.0 +2023-04-01 08:28:21,847 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 08:28:28,790 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5588, 3.7775, 4.3017, 4.4376, 1.5142, 3.9960, 3.4074, 3.5410], + device='cuda:0'), covar=tensor([0.1814, 0.1541, 0.1144, 0.1004, 0.6558, 0.1124, 0.1098, 0.2178], + device='cuda:0'), in_proj_covar=tensor([0.0562, 0.0482, 0.0655, 0.0540, 0.0613, 0.0407, 0.0413, 0.0607], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 08:28:30,190 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.23 vs. limit=5.0 +2023-04-01 08:28:47,144 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0120, 1.1198, 1.3937, 0.8455, 2.4467, 2.9709, 2.7697, 3.1696], + device='cuda:0'), covar=tensor([0.1445, 0.3062, 0.2905, 0.2036, 0.0386, 0.0164, 0.0246, 0.0155], + device='cuda:0'), in_proj_covar=tensor([0.0250, 0.0283, 0.0313, 0.0247, 0.0198, 0.0128, 0.0201, 0.0152], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:28:50,403 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0293, 1.9530, 1.6550, 1.5375, 1.5291, 1.7933, 0.3283, 0.8354], + device='cuda:0'), covar=tensor([0.0278, 0.0270, 0.0183, 0.0293, 0.0564, 0.0298, 0.0535, 0.0480], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0300, 0.0301, 0.0314, 0.0385, 0.0313, 0.0288, 0.0303], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:29:11,948 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 08:29:15,371 INFO [train.py:903] (0/4) Epoch 6, batch 5600, loss[loss=0.2505, simple_loss=0.3122, pruned_loss=0.09433, over 19604.00 frames. ], tot_loss[loss=0.2704, simple_loss=0.3336, pruned_loss=0.1035, over 3818374.08 frames. ], batch size: 50, lr: 1.34e-02, grad_scale: 8.0 +2023-04-01 08:29:20,720 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.863e+02 6.221e+02 7.656e+02 9.358e+02 1.388e+03, threshold=1.531e+03, percent-clipped=0.0 +2023-04-01 08:29:43,080 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39761.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:30:11,505 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39785.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:30:19,675 INFO [train.py:903] (0/4) Epoch 6, batch 5650, loss[loss=0.2737, simple_loss=0.3354, pruned_loss=0.1059, over 19091.00 frames. ], tot_loss[loss=0.2717, simple_loss=0.3349, pruned_loss=0.1043, over 3798645.60 frames. ], batch size: 69, lr: 1.34e-02, grad_scale: 8.0 +2023-04-01 08:31:01,144 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=39825.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:31:08,882 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 08:31:12,745 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2451, 1.2879, 1.7170, 1.4279, 2.3755, 1.9589, 2.5020, 0.9755], + device='cuda:0'), covar=tensor([0.1856, 0.3124, 0.1682, 0.1532, 0.1149, 0.1602, 0.1201, 0.2992], + device='cuda:0'), in_proj_covar=tensor([0.0441, 0.0516, 0.0503, 0.0406, 0.0564, 0.0447, 0.0629, 0.0453], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 08:31:13,795 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=39833.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:31:17,320 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7302, 4.2177, 2.6063, 3.7129, 1.1912, 3.9899, 3.8692, 4.2038], + device='cuda:0'), covar=tensor([0.0576, 0.1014, 0.1884, 0.0749, 0.3883, 0.0772, 0.0708, 0.0797], + device='cuda:0'), in_proj_covar=tensor([0.0367, 0.0323, 0.0377, 0.0290, 0.0357, 0.0310, 0.0297, 0.0327], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:31:21,778 INFO [train.py:903] (0/4) Epoch 6, batch 5700, loss[loss=0.2696, simple_loss=0.3391, pruned_loss=0.1001, over 19671.00 frames. ], tot_loss[loss=0.2706, simple_loss=0.3344, pruned_loss=0.1034, over 3809636.40 frames. ], batch size: 59, lr: 1.34e-02, grad_scale: 8.0 +2023-04-01 08:31:26,540 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.066e+02 6.807e+02 8.639e+02 1.032e+03 2.369e+03, threshold=1.728e+03, percent-clipped=2.0 +2023-04-01 08:31:54,453 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6038, 1.6170, 1.7412, 2.1136, 1.2637, 1.7576, 2.1019, 1.6858], + device='cuda:0'), covar=tensor([0.2343, 0.1872, 0.0978, 0.1024, 0.2110, 0.0934, 0.2152, 0.1735], + device='cuda:0'), in_proj_covar=tensor([0.0681, 0.0675, 0.0581, 0.0825, 0.0706, 0.0591, 0.0718, 0.0622], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 08:32:21,661 INFO [train.py:903] (0/4) Epoch 6, batch 5750, loss[loss=0.3201, simple_loss=0.3709, pruned_loss=0.1346, over 19649.00 frames. ], tot_loss[loss=0.272, simple_loss=0.3357, pruned_loss=0.1042, over 3811446.69 frames. ], batch size: 58, lr: 1.34e-02, grad_scale: 8.0 +2023-04-01 08:32:23,996 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 08:32:30,902 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 08:32:36,528 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 08:32:42,524 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1942, 2.1088, 1.7132, 1.5766, 1.4237, 1.7384, 0.3401, 0.9821], + device='cuda:0'), covar=tensor([0.0279, 0.0275, 0.0216, 0.0306, 0.0629, 0.0365, 0.0577, 0.0491], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0303, 0.0301, 0.0316, 0.0388, 0.0313, 0.0291, 0.0305], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:32:58,453 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=39918.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:33:23,262 INFO [train.py:903] (0/4) Epoch 6, batch 5800, loss[loss=0.3066, simple_loss=0.3772, pruned_loss=0.118, over 19532.00 frames. ], tot_loss[loss=0.2726, simple_loss=0.3361, pruned_loss=0.1046, over 3790608.99 frames. ], batch size: 56, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:33:23,652 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=39940.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:33:28,185 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=39943.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:33:29,042 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.561e+02 6.519e+02 7.840e+02 9.394e+02 2.455e+03, threshold=1.568e+03, percent-clipped=4.0 +2023-04-01 08:34:26,995 INFO [train.py:903] (0/4) Epoch 6, batch 5850, loss[loss=0.2115, simple_loss=0.2776, pruned_loss=0.07268, over 17258.00 frames. ], tot_loss[loss=0.2706, simple_loss=0.3348, pruned_loss=0.1032, over 3806277.60 frames. ], batch size: 38, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:34:38,529 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-40000.pt +2023-04-01 08:34:40,698 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40001.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 08:34:46,871 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-01 08:34:59,466 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40017.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:35:28,999 INFO [train.py:903] (0/4) Epoch 6, batch 5900, loss[loss=0.2872, simple_loss=0.3525, pruned_loss=0.111, over 17582.00 frames. ], tot_loss[loss=0.2707, simple_loss=0.3345, pruned_loss=0.1034, over 3821386.66 frames. ], batch size: 101, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:35:31,453 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 08:35:31,825 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40042.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:35:33,716 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.177e+02 6.123e+02 7.695e+02 9.772e+02 2.844e+03, threshold=1.539e+03, percent-clipped=4.0 +2023-04-01 08:35:53,054 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 08:36:21,308 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2952, 2.1856, 1.5867, 1.3646, 2.0814, 1.0404, 1.1504, 1.7745], + device='cuda:0'), covar=tensor([0.0846, 0.0562, 0.0949, 0.0604, 0.0389, 0.1118, 0.0673, 0.0425], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0280, 0.0315, 0.0237, 0.0226, 0.0310, 0.0288, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 08:36:30,103 INFO [train.py:903] (0/4) Epoch 6, batch 5950, loss[loss=0.2536, simple_loss=0.3232, pruned_loss=0.092, over 19680.00 frames. ], tot_loss[loss=0.2713, simple_loss=0.3352, pruned_loss=0.1037, over 3814684.34 frames. ], batch size: 60, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:37:03,174 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40116.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 08:37:14,645 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0384, 5.0349, 5.9808, 5.8770, 1.6708, 5.4932, 4.7202, 5.4770], + device='cuda:0'), covar=tensor([0.1098, 0.0598, 0.0390, 0.0403, 0.4976, 0.0366, 0.0489, 0.0854], + device='cuda:0'), in_proj_covar=tensor([0.0564, 0.0481, 0.0657, 0.0541, 0.0613, 0.0408, 0.0417, 0.0604], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 08:37:18,121 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40129.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:37:27,905 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.33 vs. limit=5.0 +2023-04-01 08:37:30,832 INFO [train.py:903] (0/4) Epoch 6, batch 6000, loss[loss=0.2782, simple_loss=0.3396, pruned_loss=0.1084, over 19597.00 frames. ], tot_loss[loss=0.2697, simple_loss=0.3334, pruned_loss=0.103, over 3808929.07 frames. ], batch size: 52, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:37:30,833 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 08:37:43,221 INFO [train.py:937] (0/4) Epoch 6, validation: loss=0.1955, simple_loss=0.2951, pruned_loss=0.04789, over 944034.00 frames. +2023-04-01 08:37:43,222 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 08:37:47,775 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.228e+02 6.298e+02 7.514e+02 9.544e+02 1.960e+03, threshold=1.503e+03, percent-clipped=1.0 +2023-04-01 08:38:29,636 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40177.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:38:44,844 INFO [train.py:903] (0/4) Epoch 6, batch 6050, loss[loss=0.2819, simple_loss=0.3516, pruned_loss=0.106, over 19776.00 frames. ], tot_loss[loss=0.2703, simple_loss=0.3339, pruned_loss=0.1033, over 3790955.51 frames. ], batch size: 56, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:38:45,115 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7996, 1.4222, 1.6246, 2.2918, 2.1608, 1.8466, 2.0790, 1.7958], + device='cuda:0'), covar=tensor([0.0673, 0.1009, 0.0856, 0.0623, 0.0628, 0.0737, 0.0765, 0.0701], + device='cuda:0'), in_proj_covar=tensor([0.0220, 0.0238, 0.0234, 0.0267, 0.0255, 0.0223, 0.0221, 0.0211], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 08:38:53,724 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40196.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:39:14,895 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40213.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:39:24,383 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40221.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:39:43,586 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40236.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:39:48,081 INFO [train.py:903] (0/4) Epoch 6, batch 6100, loss[loss=0.3143, simple_loss=0.3691, pruned_loss=0.1298, over 19696.00 frames. ], tot_loss[loss=0.2713, simple_loss=0.3345, pruned_loss=0.1041, over 3790016.00 frames. ], batch size: 59, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:39:54,269 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.428e+02 6.367e+02 7.728e+02 1.144e+03 2.582e+03, threshold=1.546e+03, percent-clipped=10.0 +2023-04-01 08:39:54,682 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40244.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:40:51,268 INFO [train.py:903] (0/4) Epoch 6, batch 6150, loss[loss=0.3003, simple_loss=0.3591, pruned_loss=0.1208, over 19086.00 frames. ], tot_loss[loss=0.2695, simple_loss=0.3329, pruned_loss=0.1031, over 3803434.61 frames. ], batch size: 69, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:40:51,598 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40290.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:40:54,070 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40292.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:40:57,798 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0540, 3.6975, 2.1068, 1.9708, 3.1943, 1.6837, 1.0637, 1.8739], + device='cuda:0'), covar=tensor([0.1004, 0.0347, 0.0817, 0.0634, 0.0446, 0.1002, 0.0975, 0.0613], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0280, 0.0314, 0.0234, 0.0223, 0.0308, 0.0284, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 08:41:19,454 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 08:41:36,484 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.1108, 3.7209, 2.5632, 3.3660, 1.2373, 3.4596, 3.4484, 3.6414], + device='cuda:0'), covar=tensor([0.0745, 0.1205, 0.1685, 0.0728, 0.3482, 0.0768, 0.0733, 0.0810], + device='cuda:0'), in_proj_covar=tensor([0.0374, 0.0327, 0.0384, 0.0297, 0.0363, 0.0316, 0.0304, 0.0335], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 08:41:52,313 INFO [train.py:903] (0/4) Epoch 6, batch 6200, loss[loss=0.3088, simple_loss=0.3687, pruned_loss=0.1245, over 19463.00 frames. ], tot_loss[loss=0.2708, simple_loss=0.334, pruned_loss=0.1039, over 3795635.29 frames. ], batch size: 64, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:41:57,219 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.101e+02 6.810e+02 8.218e+02 1.008e+03 2.334e+03, threshold=1.644e+03, percent-clipped=5.0 +2023-04-01 08:42:01,794 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40348.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:42:33,815 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40372.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 08:42:54,272 INFO [train.py:903] (0/4) Epoch 6, batch 6250, loss[loss=0.3002, simple_loss=0.359, pruned_loss=0.1207, over 19374.00 frames. ], tot_loss[loss=0.2697, simple_loss=0.3328, pruned_loss=0.1033, over 3784074.75 frames. ], batch size: 70, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:43:04,605 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40397.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 08:43:27,278 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 08:43:58,137 INFO [train.py:903] (0/4) Epoch 6, batch 6300, loss[loss=0.2328, simple_loss=0.3077, pruned_loss=0.07892, over 19659.00 frames. ], tot_loss[loss=0.2689, simple_loss=0.3324, pruned_loss=0.1027, over 3795742.07 frames. ], batch size: 55, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:44:03,775 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.716e+02 6.389e+02 8.265e+02 1.061e+03 2.633e+03, threshold=1.653e+03, percent-clipped=7.0 +2023-04-01 08:44:07,500 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0305, 1.4844, 1.6677, 1.9657, 1.9132, 1.8196, 1.5791, 2.0643], + device='cuda:0'), covar=tensor([0.0825, 0.1673, 0.1247, 0.0930, 0.1127, 0.0489, 0.1152, 0.0571], + device='cuda:0'), in_proj_covar=tensor([0.0245, 0.0360, 0.0284, 0.0237, 0.0300, 0.0247, 0.0273, 0.0230], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 08:45:00,760 INFO [train.py:903] (0/4) Epoch 6, batch 6350, loss[loss=0.2621, simple_loss=0.3354, pruned_loss=0.09443, over 19595.00 frames. ], tot_loss[loss=0.2694, simple_loss=0.3331, pruned_loss=0.1029, over 3818275.81 frames. ], batch size: 61, lr: 1.33e-02, grad_scale: 8.0 +2023-04-01 08:45:12,818 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40500.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:45:24,298 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40510.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:45:45,004 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40525.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:46:02,598 INFO [train.py:903] (0/4) Epoch 6, batch 6400, loss[loss=0.2614, simple_loss=0.3336, pruned_loss=0.09462, over 19695.00 frames. ], tot_loss[loss=0.2696, simple_loss=0.3333, pruned_loss=0.1029, over 3811925.79 frames. ], batch size: 59, lr: 1.32e-02, grad_scale: 8.0 +2023-04-01 08:46:07,242 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.716e+02 6.528e+02 8.039e+02 1.077e+03 1.980e+03, threshold=1.608e+03, percent-clipped=3.0 +2023-04-01 08:46:12,185 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40548.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:46:23,189 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40557.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:46:44,906 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40573.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:46:46,462 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-01 08:46:52,950 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40580.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:46:53,173 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7779, 1.3718, 1.3638, 2.0119, 1.6258, 2.0495, 2.1364, 1.9714], + device='cuda:0'), covar=tensor([0.0796, 0.1010, 0.1116, 0.0933, 0.0986, 0.0735, 0.0919, 0.0618], + device='cuda:0'), in_proj_covar=tensor([0.0222, 0.0238, 0.0236, 0.0266, 0.0260, 0.0221, 0.0220, 0.0213], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 08:47:04,384 INFO [train.py:903] (0/4) Epoch 6, batch 6450, loss[loss=0.3428, simple_loss=0.3849, pruned_loss=0.1503, over 13550.00 frames. ], tot_loss[loss=0.2691, simple_loss=0.333, pruned_loss=0.1026, over 3810708.29 frames. ], batch size: 136, lr: 1.32e-02, grad_scale: 8.0 +2023-04-01 08:47:38,283 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.43 vs. limit=5.0 +2023-04-01 08:47:50,040 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 08:47:59,669 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40634.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:48:07,614 INFO [train.py:903] (0/4) Epoch 6, batch 6500, loss[loss=0.2574, simple_loss=0.3283, pruned_loss=0.09325, over 19705.00 frames. ], tot_loss[loss=0.2689, simple_loss=0.333, pruned_loss=0.1024, over 3823882.48 frames. ], batch size: 59, lr: 1.32e-02, grad_scale: 8.0 +2023-04-01 08:48:13,023 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.640e+02 6.039e+02 7.833e+02 1.001e+03 2.233e+03, threshold=1.567e+03, percent-clipped=5.0 +2023-04-01 08:48:15,437 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 08:48:44,317 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40669.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:48:47,930 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40672.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:48:55,147 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-01 08:49:12,721 INFO [train.py:903] (0/4) Epoch 6, batch 6550, loss[loss=0.2927, simple_loss=0.3554, pruned_loss=0.115, over 19697.00 frames. ], tot_loss[loss=0.2681, simple_loss=0.3322, pruned_loss=0.102, over 3826250.00 frames. ], batch size: 59, lr: 1.32e-02, grad_scale: 8.0 +2023-04-01 08:49:15,168 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40692.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:49:18,782 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40695.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:50:15,129 INFO [train.py:903] (0/4) Epoch 6, batch 6600, loss[loss=0.2952, simple_loss=0.3502, pruned_loss=0.1201, over 19673.00 frames. ], tot_loss[loss=0.2687, simple_loss=0.3326, pruned_loss=0.1024, over 3826215.81 frames. ], batch size: 53, lr: 1.32e-02, grad_scale: 8.0 +2023-04-01 08:50:19,765 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.418e+02 6.030e+02 7.716e+02 9.913e+02 2.888e+03, threshold=1.543e+03, percent-clipped=3.0 +2023-04-01 08:50:26,104 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40749.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:51:17,189 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.76 vs. limit=5.0 +2023-04-01 08:51:17,593 INFO [train.py:903] (0/4) Epoch 6, batch 6650, loss[loss=0.222, simple_loss=0.2945, pruned_loss=0.07475, over 14488.00 frames. ], tot_loss[loss=0.2679, simple_loss=0.3319, pruned_loss=0.1019, over 3820213.69 frames. ], batch size: 32, lr: 1.32e-02, grad_scale: 8.0 +2023-04-01 08:51:40,497 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40807.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:51:55,839 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-01 08:52:19,377 INFO [train.py:903] (0/4) Epoch 6, batch 6700, loss[loss=0.2793, simple_loss=0.3492, pruned_loss=0.1047, over 19612.00 frames. ], tot_loss[loss=0.2688, simple_loss=0.3327, pruned_loss=0.1025, over 3811805.76 frames. ], batch size: 57, lr: 1.32e-02, grad_scale: 8.0 +2023-04-01 08:52:24,141 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.981e+02 6.315e+02 8.385e+02 9.952e+02 2.559e+03, threshold=1.677e+03, percent-clipped=5.0 +2023-04-01 08:52:39,403 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=40854.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:53:13,414 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40883.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:53:21,378 INFO [train.py:903] (0/4) Epoch 6, batch 6750, loss[loss=0.2474, simple_loss=0.325, pruned_loss=0.08485, over 19606.00 frames. ], tot_loss[loss=0.2678, simple_loss=0.3319, pruned_loss=0.1018, over 3817664.94 frames. ], batch size: 57, lr: 1.32e-02, grad_scale: 8.0 +2023-04-01 08:53:27,466 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7288, 1.4535, 1.4481, 1.6036, 3.2668, 1.0454, 2.1922, 3.5138], + device='cuda:0'), covar=tensor([0.0335, 0.2296, 0.2355, 0.1541, 0.0619, 0.2304, 0.1199, 0.0291], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0310, 0.0315, 0.0292, 0.0315, 0.0310, 0.0289, 0.0310], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 08:53:42,286 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3453, 2.2236, 1.7488, 1.7487, 1.6253, 1.7154, 0.3554, 1.0697], + device='cuda:0'), covar=tensor([0.0250, 0.0290, 0.0260, 0.0384, 0.0591, 0.0400, 0.0646, 0.0524], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0305, 0.0300, 0.0322, 0.0394, 0.0318, 0.0293, 0.0308], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 08:53:53,173 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=40918.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:53:54,385 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9507, 1.5949, 1.5255, 1.9344, 1.8713, 1.8102, 1.5774, 1.8741], + device='cuda:0'), covar=tensor([0.0843, 0.1462, 0.1346, 0.0828, 0.1037, 0.0389, 0.1030, 0.0617], + device='cuda:0'), in_proj_covar=tensor([0.0246, 0.0358, 0.0283, 0.0236, 0.0299, 0.0241, 0.0270, 0.0229], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 08:54:04,563 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40928.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:54:14,788 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0 +2023-04-01 08:54:17,527 INFO [train.py:903] (0/4) Epoch 6, batch 6800, loss[loss=0.3028, simple_loss=0.3598, pruned_loss=0.1229, over 19757.00 frames. ], tot_loss[loss=0.2693, simple_loss=0.333, pruned_loss=0.1027, over 3815441.02 frames. ], batch size: 56, lr: 1.32e-02, grad_scale: 8.0 +2023-04-01 08:54:23,021 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.567e+02 5.879e+02 7.609e+02 1.019e+03 2.150e+03, threshold=1.522e+03, percent-clipped=4.0 +2023-04-01 08:54:31,380 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=40951.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:54:33,645 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40953.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:54:49,204 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-6.pt +2023-04-01 08:55:04,819 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 08:55:06,219 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 08:55:08,788 INFO [train.py:903] (0/4) Epoch 7, batch 0, loss[loss=0.2292, simple_loss=0.297, pruned_loss=0.08067, over 19368.00 frames. ], tot_loss[loss=0.2292, simple_loss=0.297, pruned_loss=0.08067, over 19368.00 frames. ], batch size: 47, lr: 1.24e-02, grad_scale: 8.0 +2023-04-01 08:55:08,789 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 08:55:20,409 INFO [train.py:937] (0/4) Epoch 7, validation: loss=0.1957, simple_loss=0.2957, pruned_loss=0.04779, over 944034.00 frames. +2023-04-01 08:55:20,410 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 08:55:21,930 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=40969.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:55:30,928 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=40976.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:55:32,930 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 08:56:04,635 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41005.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:56:08,594 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-04-01 08:56:15,085 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4435, 2.4248, 1.7218, 1.7241, 2.4053, 1.2779, 1.1095, 1.7916], + device='cuda:0'), covar=tensor([0.0790, 0.0499, 0.0849, 0.0574, 0.0340, 0.0983, 0.0726, 0.0411], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0278, 0.0314, 0.0237, 0.0225, 0.0309, 0.0283, 0.0251], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 08:56:15,981 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41013.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:56:18,316 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41015.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:56:22,012 INFO [train.py:903] (0/4) Epoch 7, batch 50, loss[loss=0.329, simple_loss=0.3885, pruned_loss=0.1347, over 19318.00 frames. ], tot_loss[loss=0.266, simple_loss=0.3305, pruned_loss=0.1008, over 855737.61 frames. ], batch size: 70, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 08:56:36,108 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41030.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:56:51,947 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.786e+02 6.089e+02 7.435e+02 1.027e+03 3.072e+03, threshold=1.487e+03, percent-clipped=7.0 +2023-04-01 08:56:56,506 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 08:57:17,945 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41063.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:57:18,085 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41063.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:57:23,454 INFO [train.py:903] (0/4) Epoch 7, batch 100, loss[loss=0.2237, simple_loss=0.2929, pruned_loss=0.07727, over 19579.00 frames. ], tot_loss[loss=0.2666, simple_loss=0.3309, pruned_loss=0.1011, over 1515414.81 frames. ], batch size: 52, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 08:57:34,761 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 08:57:46,196 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41088.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:58:24,483 INFO [train.py:903] (0/4) Epoch 7, batch 150, loss[loss=0.2514, simple_loss=0.3286, pruned_loss=0.08712, over 19778.00 frames. ], tot_loss[loss=0.2662, simple_loss=0.3308, pruned_loss=0.1008, over 2029761.39 frames. ], batch size: 56, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 08:58:26,862 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41120.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:58:36,370 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41128.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 08:58:56,746 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.181e+02 6.219e+02 8.190e+02 1.094e+03 2.901e+03, threshold=1.638e+03, percent-clipped=4.0 +2023-04-01 08:59:22,647 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 08:59:23,718 INFO [train.py:903] (0/4) Epoch 7, batch 200, loss[loss=0.2711, simple_loss=0.3368, pruned_loss=0.1027, over 19477.00 frames. ], tot_loss[loss=0.2653, simple_loss=0.3295, pruned_loss=0.1006, over 2439607.47 frames. ], batch size: 64, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 08:59:44,056 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5662, 1.5046, 2.0517, 1.6808, 3.2274, 4.7499, 4.6865, 5.0720], + device='cuda:0'), covar=tensor([0.1353, 0.2821, 0.2508, 0.1683, 0.0390, 0.0097, 0.0130, 0.0059], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0284, 0.0314, 0.0247, 0.0205, 0.0131, 0.0202, 0.0156], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 09:00:27,762 INFO [train.py:903] (0/4) Epoch 7, batch 250, loss[loss=0.2649, simple_loss=0.3363, pruned_loss=0.09678, over 19660.00 frames. ], tot_loss[loss=0.2678, simple_loss=0.3317, pruned_loss=0.102, over 2746449.77 frames. ], batch size: 55, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 09:00:30,418 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.43 vs. limit=5.0 +2023-04-01 09:00:37,150 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.11 vs. limit=2.0 +2023-04-01 09:00:37,984 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41225.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:00:40,037 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41227.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:00:59,430 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.194e+02 6.866e+02 8.894e+02 1.080e+03 3.290e+03, threshold=1.779e+03, percent-clipped=6.0 +2023-04-01 09:01:06,541 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41250.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:01:22,636 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41262.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:01:30,752 INFO [train.py:903] (0/4) Epoch 7, batch 300, loss[loss=0.2936, simple_loss=0.3514, pruned_loss=0.1179, over 19777.00 frames. ], tot_loss[loss=0.2678, simple_loss=0.3315, pruned_loss=0.102, over 2969771.38 frames. ], batch size: 56, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 09:02:31,489 INFO [train.py:903] (0/4) Epoch 7, batch 350, loss[loss=0.2336, simple_loss=0.309, pruned_loss=0.07912, over 19760.00 frames. ], tot_loss[loss=0.2677, simple_loss=0.3316, pruned_loss=0.1019, over 3160259.60 frames. ], batch size: 54, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 09:02:34,017 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 09:03:02,033 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41342.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:03:04,935 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.733e+02 5.797e+02 7.460e+02 9.435e+02 2.818e+03, threshold=1.492e+03, percent-clipped=3.0 +2023-04-01 09:03:22,751 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41359.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:03:32,953 INFO [train.py:903] (0/4) Epoch 7, batch 400, loss[loss=0.3551, simple_loss=0.3975, pruned_loss=0.1563, over 19372.00 frames. ], tot_loss[loss=0.2667, simple_loss=0.3309, pruned_loss=0.1012, over 3306103.28 frames. ], batch size: 70, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 09:03:43,454 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41377.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:03:54,220 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41384.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:04:20,757 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41407.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:04:23,150 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41409.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:04:34,709 INFO [train.py:903] (0/4) Epoch 7, batch 450, loss[loss=0.2456, simple_loss=0.3163, pruned_loss=0.08739, over 19669.00 frames. ], tot_loss[loss=0.2662, simple_loss=0.3307, pruned_loss=0.1009, over 3426697.45 frames. ], batch size: 53, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 09:05:02,709 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 09:05:03,835 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 09:05:06,101 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.663e+02 5.824e+02 7.719e+02 9.807e+02 3.448e+03, threshold=1.544e+03, percent-clipped=7.0 +2023-04-01 09:05:30,639 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=41464.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:05:37,396 INFO [train.py:903] (0/4) Epoch 7, batch 500, loss[loss=0.2757, simple_loss=0.341, pruned_loss=0.1052, over 19533.00 frames. ], tot_loss[loss=0.2659, simple_loss=0.3302, pruned_loss=0.1008, over 3528891.33 frames. ], batch size: 54, lr: 1.23e-02, grad_scale: 16.0 +2023-04-01 09:05:44,880 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41474.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:06:38,625 INFO [train.py:903] (0/4) Epoch 7, batch 550, loss[loss=0.2175, simple_loss=0.2908, pruned_loss=0.07207, over 19863.00 frames. ], tot_loss[loss=0.2634, simple_loss=0.3286, pruned_loss=0.09911, over 3591518.01 frames. ], batch size: 52, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 09:06:43,702 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41522.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:07:09,851 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.859e+02 6.032e+02 7.411e+02 9.275e+02 1.625e+03, threshold=1.482e+03, percent-clipped=1.0 +2023-04-01 09:07:37,920 INFO [train.py:903] (0/4) Epoch 7, batch 600, loss[loss=0.2376, simple_loss=0.2974, pruned_loss=0.08887, over 19012.00 frames. ], tot_loss[loss=0.2647, simple_loss=0.3297, pruned_loss=0.09982, over 3635467.65 frames. ], batch size: 42, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 09:07:50,892 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=41579.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:08:13,417 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 09:08:16,131 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41598.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:08:39,327 INFO [train.py:903] (0/4) Epoch 7, batch 650, loss[loss=0.2614, simple_loss=0.3214, pruned_loss=0.1006, over 19686.00 frames. ], tot_loss[loss=0.265, simple_loss=0.3302, pruned_loss=0.09987, over 3683613.09 frames. ], batch size: 53, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 09:08:45,330 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41623.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:09:00,381 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41633.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:09:13,503 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.335e+02 5.719e+02 7.418e+02 1.065e+03 4.334e+03, threshold=1.484e+03, percent-clipped=7.0 +2023-04-01 09:09:28,655 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41658.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:09:39,246 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-04-01 09:09:40,774 INFO [train.py:903] (0/4) Epoch 7, batch 700, loss[loss=0.2636, simple_loss=0.3318, pruned_loss=0.09768, over 19670.00 frames. ], tot_loss[loss=0.2665, simple_loss=0.3317, pruned_loss=0.1006, over 3708762.52 frames. ], batch size: 55, lr: 1.23e-02, grad_scale: 8.0 +2023-04-01 09:09:48,552 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41672.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:10:11,474 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-04-01 09:10:37,099 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-04-01 09:10:45,014 INFO [train.py:903] (0/4) Epoch 7, batch 750, loss[loss=0.2282, simple_loss=0.2973, pruned_loss=0.07954, over 19405.00 frames. ], tot_loss[loss=0.2641, simple_loss=0.3298, pruned_loss=0.09925, over 3735532.36 frames. ], batch size: 48, lr: 1.22e-02, grad_scale: 8.0 +2023-04-01 09:10:59,407 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41730.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:11:15,964 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.723e+02 5.751e+02 6.886e+02 8.721e+02 1.519e+03, threshold=1.377e+03, percent-clipped=2.0 +2023-04-01 09:11:31,830 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41755.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:11:46,763 INFO [train.py:903] (0/4) Epoch 7, batch 800, loss[loss=0.2523, simple_loss=0.3091, pruned_loss=0.09769, over 19765.00 frames. ], tot_loss[loss=0.2644, simple_loss=0.3296, pruned_loss=0.09957, over 3746218.48 frames. ], batch size: 48, lr: 1.22e-02, grad_scale: 8.0 +2023-04-01 09:11:56,187 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 09:11:58,683 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41778.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:12:31,056 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41803.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:12:42,769 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0037, 2.0319, 2.1381, 3.1339, 2.0834, 3.1686, 2.9584, 2.0407], + device='cuda:0'), covar=tensor([0.3035, 0.2422, 0.1093, 0.1442, 0.2880, 0.0893, 0.2089, 0.1991], + device='cuda:0'), in_proj_covar=tensor([0.0702, 0.0694, 0.0598, 0.0845, 0.0712, 0.0606, 0.0730, 0.0643], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 09:12:48,118 INFO [train.py:903] (0/4) Epoch 7, batch 850, loss[loss=0.2689, simple_loss=0.3424, pruned_loss=0.09777, over 19519.00 frames. ], tot_loss[loss=0.2642, simple_loss=0.3296, pruned_loss=0.09936, over 3759810.44 frames. ], batch size: 54, lr: 1.22e-02, grad_scale: 8.0 +2023-04-01 09:13:10,238 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=41835.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:13:23,186 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.000e+02 6.514e+02 8.083e+02 9.646e+02 1.896e+03, threshold=1.617e+03, percent-clipped=5.0 +2023-04-01 09:13:38,252 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 09:13:40,926 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=41860.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:13:41,462 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-04-01 09:13:50,188 INFO [train.py:903] (0/4) Epoch 7, batch 900, loss[loss=0.2823, simple_loss=0.3571, pruned_loss=0.1037, over 19602.00 frames. ], tot_loss[loss=0.2629, simple_loss=0.3282, pruned_loss=0.09874, over 3774396.47 frames. ], batch size: 57, lr: 1.22e-02, grad_scale: 8.0 +2023-04-01 09:14:29,634 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2268, 1.3242, 1.1589, 0.9781, 1.0480, 1.0728, 0.0246, 0.3403], + device='cuda:0'), covar=tensor([0.0305, 0.0329, 0.0195, 0.0259, 0.0652, 0.0258, 0.0576, 0.0527], + device='cuda:0'), in_proj_covar=tensor([0.0296, 0.0303, 0.0295, 0.0323, 0.0390, 0.0317, 0.0287, 0.0304], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 09:14:49,701 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2284, 1.1676, 1.6596, 0.9685, 2.5506, 3.2971, 3.0816, 3.5139], + device='cuda:0'), covar=tensor([0.1363, 0.3059, 0.2721, 0.2079, 0.0415, 0.0149, 0.0206, 0.0142], + device='cuda:0'), in_proj_covar=tensor([0.0250, 0.0279, 0.0308, 0.0245, 0.0201, 0.0133, 0.0200, 0.0156], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 09:14:51,573 INFO [train.py:903] (0/4) Epoch 7, batch 950, loss[loss=0.3015, simple_loss=0.3598, pruned_loss=0.1216, over 19647.00 frames. ], tot_loss[loss=0.2628, simple_loss=0.3282, pruned_loss=0.09873, over 3790516.15 frames. ], batch size: 55, lr: 1.22e-02, grad_scale: 4.0 +2023-04-01 09:14:55,086 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 09:15:01,482 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41924.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:15:02,799 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0218, 3.4948, 2.0082, 2.2640, 3.1490, 1.6768, 0.9925, 1.8036], + device='cuda:0'), covar=tensor([0.0859, 0.0346, 0.0814, 0.0538, 0.0356, 0.0859, 0.0933, 0.0573], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0280, 0.0310, 0.0237, 0.0223, 0.0305, 0.0287, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:15:26,244 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.318e+02 6.505e+02 7.519e+02 9.487e+02 1.757e+03, threshold=1.504e+03, percent-clipped=1.0 +2023-04-01 09:15:53,965 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=41966.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 09:15:55,891 INFO [train.py:903] (0/4) Epoch 7, batch 1000, loss[loss=0.2283, simple_loss=0.2976, pruned_loss=0.07953, over 19708.00 frames. ], tot_loss[loss=0.2635, simple_loss=0.3291, pruned_loss=0.09896, over 3800275.68 frames. ], batch size: 51, lr: 1.22e-02, grad_scale: 4.0 +2023-04-01 09:16:36,144 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-42000.pt +2023-04-01 09:16:45,390 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8400, 1.6410, 1.3821, 1.8801, 1.9480, 1.6574, 1.5414, 1.8613], + device='cuda:0'), covar=tensor([0.0832, 0.1302, 0.1256, 0.0735, 0.0882, 0.0462, 0.0956, 0.0547], + device='cuda:0'), in_proj_covar=tensor([0.0246, 0.0361, 0.0289, 0.0238, 0.0303, 0.0248, 0.0274, 0.0231], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:16:48,575 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 09:16:56,929 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=42016.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:16:59,085 INFO [train.py:903] (0/4) Epoch 7, batch 1050, loss[loss=0.2816, simple_loss=0.3379, pruned_loss=0.1127, over 19730.00 frames. ], tot_loss[loss=0.2633, simple_loss=0.3286, pruned_loss=0.099, over 3799701.93 frames. ], batch size: 51, lr: 1.22e-02, grad_scale: 4.0 +2023-04-01 09:17:14,962 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6102, 1.2850, 1.3712, 1.8659, 1.4541, 1.7907, 1.7951, 1.5937], + device='cuda:0'), covar=tensor([0.0837, 0.1050, 0.1115, 0.0858, 0.0913, 0.0734, 0.0895, 0.0745], + device='cuda:0'), in_proj_covar=tensor([0.0219, 0.0234, 0.0234, 0.0265, 0.0255, 0.0217, 0.0218, 0.0211], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0004, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 09:17:19,836 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.11 vs. limit=2.0 +2023-04-01 09:17:29,374 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 09:17:33,746 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.193e+02 5.894e+02 7.129e+02 9.584e+02 2.561e+03, threshold=1.426e+03, percent-clipped=5.0 +2023-04-01 09:18:00,554 INFO [train.py:903] (0/4) Epoch 7, batch 1100, loss[loss=0.2932, simple_loss=0.3314, pruned_loss=0.1275, over 19758.00 frames. ], tot_loss[loss=0.2638, simple_loss=0.329, pruned_loss=0.09931, over 3788807.01 frames. ], batch size: 45, lr: 1.22e-02, grad_scale: 4.0 +2023-04-01 09:18:29,521 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7109, 1.2305, 1.4246, 1.5119, 3.1947, 0.9379, 2.1290, 3.4541], + device='cuda:0'), covar=tensor([0.0369, 0.2576, 0.2426, 0.1679, 0.0636, 0.2498, 0.1210, 0.0316], + device='cuda:0'), in_proj_covar=tensor([0.0305, 0.0318, 0.0325, 0.0297, 0.0320, 0.0317, 0.0299, 0.0317], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:19:03,370 INFO [train.py:903] (0/4) Epoch 7, batch 1150, loss[loss=0.3056, simple_loss=0.3605, pruned_loss=0.1254, over 18784.00 frames. ], tot_loss[loss=0.2628, simple_loss=0.3288, pruned_loss=0.09844, over 3797883.45 frames. ], batch size: 74, lr: 1.22e-02, grad_scale: 4.0 +2023-04-01 09:19:21,126 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42131.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:19:37,737 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.517e+02 5.892e+02 7.369e+02 1.011e+03 1.805e+03, threshold=1.474e+03, percent-clipped=4.0 +2023-04-01 09:20:05,775 INFO [train.py:903] (0/4) Epoch 7, batch 1200, loss[loss=0.2189, simple_loss=0.2996, pruned_loss=0.06906, over 19666.00 frames. ], tot_loss[loss=0.2629, simple_loss=0.3287, pruned_loss=0.09857, over 3800539.35 frames. ], batch size: 53, lr: 1.22e-02, grad_scale: 8.0 +2023-04-01 09:20:30,599 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 09:21:08,017 INFO [train.py:903] (0/4) Epoch 7, batch 1250, loss[loss=0.248, simple_loss=0.313, pruned_loss=0.09154, over 19613.00 frames. ], tot_loss[loss=0.2641, simple_loss=0.3298, pruned_loss=0.09923, over 3805921.59 frames. ], batch size: 50, lr: 1.22e-02, grad_scale: 8.0 +2023-04-01 09:21:43,467 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.288e+02 6.237e+02 7.824e+02 1.034e+03 2.254e+03, threshold=1.565e+03, percent-clipped=6.0 +2023-04-01 09:22:09,623 INFO [train.py:903] (0/4) Epoch 7, batch 1300, loss[loss=0.2346, simple_loss=0.2935, pruned_loss=0.08782, over 19715.00 frames. ], tot_loss[loss=0.2641, simple_loss=0.3292, pruned_loss=0.09953, over 3810905.47 frames. ], batch size: 45, lr: 1.22e-02, grad_scale: 8.0 +2023-04-01 09:22:09,768 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=42268.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:22:53,544 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9262, 1.9692, 1.9384, 2.8286, 1.8432, 2.6346, 2.5640, 1.8565], + device='cuda:0'), covar=tensor([0.2550, 0.2105, 0.1059, 0.1270, 0.2433, 0.0872, 0.2057, 0.1938], + device='cuda:0'), in_proj_covar=tensor([0.0698, 0.0700, 0.0595, 0.0843, 0.0716, 0.0608, 0.0729, 0.0641], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 09:23:02,502 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=42310.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 09:23:12,318 INFO [train.py:903] (0/4) Epoch 7, batch 1350, loss[loss=0.2828, simple_loss=0.3447, pruned_loss=0.1104, over 19690.00 frames. ], tot_loss[loss=0.2646, simple_loss=0.3296, pruned_loss=0.09979, over 3812727.46 frames. ], batch size: 59, lr: 1.22e-02, grad_scale: 8.0 +2023-04-01 09:23:47,351 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.606e+02 6.578e+02 7.819e+02 1.024e+03 2.032e+03, threshold=1.564e+03, percent-clipped=3.0 +2023-04-01 09:24:15,699 INFO [train.py:903] (0/4) Epoch 7, batch 1400, loss[loss=0.2475, simple_loss=0.3198, pruned_loss=0.08757, over 19775.00 frames. ], tot_loss[loss=0.264, simple_loss=0.3287, pruned_loss=0.09965, over 3815492.25 frames. ], batch size: 54, lr: 1.22e-02, grad_scale: 8.0 +2023-04-01 09:24:34,374 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42383.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:24:39,136 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42387.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:24:50,678 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42397.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:25:10,405 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42412.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:25:13,363 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 09:25:17,936 INFO [train.py:903] (0/4) Epoch 7, batch 1450, loss[loss=0.249, simple_loss=0.3266, pruned_loss=0.08568, over 19676.00 frames. ], tot_loss[loss=0.2649, simple_loss=0.3297, pruned_loss=0.1, over 3818201.49 frames. ], batch size: 53, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:25:26,410 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42425.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 09:25:53,249 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.957e+02 5.854e+02 7.466e+02 9.791e+02 2.115e+03, threshold=1.493e+03, percent-clipped=4.0 +2023-04-01 09:26:19,589 INFO [train.py:903] (0/4) Epoch 7, batch 1500, loss[loss=0.2311, simple_loss=0.303, pruned_loss=0.07959, over 19573.00 frames. ], tot_loss[loss=0.2635, simple_loss=0.3288, pruned_loss=0.09906, over 3813539.50 frames. ], batch size: 52, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:27:01,855 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1953, 2.1632, 2.2564, 3.3981, 2.0868, 3.4757, 3.0516, 2.2332], + device='cuda:0'), covar=tensor([0.2779, 0.2327, 0.0976, 0.1323, 0.2813, 0.0799, 0.2021, 0.1862], + device='cuda:0'), in_proj_covar=tensor([0.0690, 0.0691, 0.0587, 0.0827, 0.0707, 0.0605, 0.0725, 0.0632], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 09:27:04,040 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5121, 3.6738, 3.9722, 3.9518, 2.2723, 3.6572, 3.4388, 3.6953], + device='cuda:0'), covar=tensor([0.0935, 0.1885, 0.0520, 0.0511, 0.3056, 0.0688, 0.0488, 0.0875], + device='cuda:0'), in_proj_covar=tensor([0.0569, 0.0495, 0.0668, 0.0552, 0.0622, 0.0422, 0.0426, 0.0618], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 09:27:15,281 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8410, 1.9996, 2.0101, 2.7697, 2.4520, 2.3390, 1.9214, 2.5788], + device='cuda:0'), covar=tensor([0.0591, 0.1574, 0.1197, 0.0780, 0.1041, 0.0401, 0.0995, 0.0525], + device='cuda:0'), in_proj_covar=tensor([0.0241, 0.0355, 0.0285, 0.0235, 0.0301, 0.0245, 0.0265, 0.0232], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:27:20,493 INFO [train.py:903] (0/4) Epoch 7, batch 1550, loss[loss=0.2355, simple_loss=0.3049, pruned_loss=0.08304, over 19620.00 frames. ], tot_loss[loss=0.2634, simple_loss=0.3287, pruned_loss=0.09909, over 3813860.23 frames. ], batch size: 50, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:27:29,090 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42524.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:27:55,725 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.300e+02 6.427e+02 8.071e+02 9.919e+02 2.182e+03, threshold=1.614e+03, percent-clipped=7.0 +2023-04-01 09:28:23,017 INFO [train.py:903] (0/4) Epoch 7, batch 1600, loss[loss=0.2562, simple_loss=0.3294, pruned_loss=0.09146, over 18266.00 frames. ], tot_loss[loss=0.2628, simple_loss=0.3286, pruned_loss=0.09855, over 3818172.66 frames. ], batch size: 83, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:28:41,292 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 09:29:02,627 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7272, 1.7847, 1.4480, 1.3777, 1.2601, 1.4702, 0.0777, 0.7043], + device='cuda:0'), covar=tensor([0.0291, 0.0296, 0.0201, 0.0253, 0.0584, 0.0289, 0.0528, 0.0461], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0303, 0.0299, 0.0323, 0.0388, 0.0318, 0.0285, 0.0300], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 09:29:24,654 INFO [train.py:903] (0/4) Epoch 7, batch 1650, loss[loss=0.2908, simple_loss=0.3549, pruned_loss=0.1133, over 18083.00 frames. ], tot_loss[loss=0.2649, simple_loss=0.3301, pruned_loss=0.09986, over 3818016.60 frames. ], batch size: 83, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:29:50,117 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42639.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:29:59,743 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.059e+02 6.175e+02 7.945e+02 9.824e+02 2.630e+03, threshold=1.589e+03, percent-clipped=4.0 +2023-04-01 09:30:23,042 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42664.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:30:26,305 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42667.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:30:27,187 INFO [train.py:903] (0/4) Epoch 7, batch 1700, loss[loss=0.3161, simple_loss=0.3593, pruned_loss=0.1364, over 13250.00 frames. ], tot_loss[loss=0.2649, simple_loss=0.3304, pruned_loss=0.09974, over 3800739.93 frames. ], batch size: 135, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:30:42,949 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=42681.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 09:31:02,744 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 09:31:15,437 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=42706.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 09:31:29,069 INFO [train.py:903] (0/4) Epoch 7, batch 1750, loss[loss=0.2539, simple_loss=0.3184, pruned_loss=0.09468, over 19560.00 frames. ], tot_loss[loss=0.2628, simple_loss=0.3283, pruned_loss=0.09869, over 3807044.88 frames. ], batch size: 61, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:31:59,553 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=42741.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:32:05,278 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.512e+02 5.682e+02 7.177e+02 9.179e+02 1.731e+03, threshold=1.435e+03, percent-clipped=1.0 +2023-04-01 09:32:23,200 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7380, 3.1538, 3.2128, 3.2238, 1.2177, 3.0349, 2.6738, 2.8972], + device='cuda:0'), covar=tensor([0.1303, 0.0844, 0.0755, 0.0725, 0.4202, 0.0725, 0.0797, 0.1388], + device='cuda:0'), in_proj_covar=tensor([0.0567, 0.0491, 0.0669, 0.0551, 0.0621, 0.0424, 0.0425, 0.0625], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 09:32:33,267 INFO [train.py:903] (0/4) Epoch 7, batch 1800, loss[loss=0.2503, simple_loss=0.3019, pruned_loss=0.09934, over 19791.00 frames. ], tot_loss[loss=0.261, simple_loss=0.3267, pruned_loss=0.09761, over 3825896.45 frames. ], batch size: 48, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:33:27,633 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 09:33:35,071 INFO [train.py:903] (0/4) Epoch 7, batch 1850, loss[loss=0.3018, simple_loss=0.3487, pruned_loss=0.1274, over 19537.00 frames. ], tot_loss[loss=0.2625, simple_loss=0.3277, pruned_loss=0.09872, over 3809231.93 frames. ], batch size: 54, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:33:54,700 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8582, 1.3184, 1.0094, 0.9141, 1.1681, 0.8701, 0.6660, 1.1730], + device='cuda:0'), covar=tensor([0.0479, 0.0628, 0.0927, 0.0489, 0.0391, 0.0953, 0.0586, 0.0340], + device='cuda:0'), in_proj_covar=tensor([0.0284, 0.0284, 0.0321, 0.0242, 0.0226, 0.0316, 0.0289, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:34:04,789 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 09:34:09,013 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.550e+02 6.596e+02 7.909e+02 1.066e+03 2.536e+03, threshold=1.582e+03, percent-clipped=10.0 +2023-04-01 09:34:22,419 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42856.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:34:37,124 INFO [train.py:903] (0/4) Epoch 7, batch 1900, loss[loss=0.3461, simple_loss=0.3804, pruned_loss=0.1559, over 13355.00 frames. ], tot_loss[loss=0.2637, simple_loss=0.3283, pruned_loss=0.09957, over 3816761.41 frames. ], batch size: 136, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:34:37,261 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=42868.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:34:48,032 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42877.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:34:51,096 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 09:34:55,747 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 09:35:14,054 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=42897.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:35:21,945 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 09:35:38,556 INFO [train.py:903] (0/4) Epoch 7, batch 1950, loss[loss=0.2348, simple_loss=0.3015, pruned_loss=0.08405, over 19785.00 frames. ], tot_loss[loss=0.2635, simple_loss=0.3278, pruned_loss=0.09961, over 3813222.16 frames. ], batch size: 49, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:36:15,248 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.499e+02 6.682e+02 8.244e+02 9.665e+02 2.689e+03, threshold=1.649e+03, percent-clipped=3.0 +2023-04-01 09:36:41,126 INFO [train.py:903] (0/4) Epoch 7, batch 2000, loss[loss=0.228, simple_loss=0.2949, pruned_loss=0.08053, over 19752.00 frames. ], tot_loss[loss=0.2646, simple_loss=0.329, pruned_loss=0.1001, over 3807590.14 frames. ], batch size: 46, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:37:00,285 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=42983.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:37:35,121 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=43011.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:37:36,087 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 09:37:43,657 INFO [train.py:903] (0/4) Epoch 7, batch 2050, loss[loss=0.296, simple_loss=0.3552, pruned_loss=0.1184, over 19599.00 frames. ], tot_loss[loss=0.2642, simple_loss=0.3287, pruned_loss=0.09983, over 3802105.23 frames. ], batch size: 57, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:37:56,189 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 09:37:57,386 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 09:38:08,467 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3183, 1.5366, 1.8391, 2.3393, 1.6305, 2.4190, 2.4697, 2.0899], + device='cuda:0'), covar=tensor([0.0760, 0.1057, 0.1081, 0.1147, 0.1092, 0.0733, 0.0899, 0.0673], + device='cuda:0'), in_proj_covar=tensor([0.0224, 0.0236, 0.0232, 0.0264, 0.0254, 0.0221, 0.0216, 0.0211], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 09:38:17,401 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.956e+02 6.149e+02 7.669e+02 9.586e+02 2.177e+03, threshold=1.534e+03, percent-clipped=1.0 +2023-04-01 09:38:18,627 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 09:38:46,694 INFO [train.py:903] (0/4) Epoch 7, batch 2100, loss[loss=0.224, simple_loss=0.2995, pruned_loss=0.07426, over 19863.00 frames. ], tot_loss[loss=0.2641, simple_loss=0.329, pruned_loss=0.09959, over 3799807.50 frames. ], batch size: 52, lr: 1.21e-02, grad_scale: 8.0 +2023-04-01 09:39:13,037 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 09:39:16,682 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5357, 1.0174, 1.2409, 1.2098, 2.1529, 0.9509, 1.8936, 2.1405], + device='cuda:0'), covar=tensor([0.0514, 0.2491, 0.2408, 0.1388, 0.0750, 0.1889, 0.0870, 0.0612], + device='cuda:0'), in_proj_covar=tensor([0.0307, 0.0316, 0.0326, 0.0295, 0.0322, 0.0318, 0.0298, 0.0320], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:39:36,358 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 09:39:41,425 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43112.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:39:48,067 INFO [train.py:903] (0/4) Epoch 7, batch 2150, loss[loss=0.2291, simple_loss=0.3042, pruned_loss=0.07698, over 19784.00 frames. ], tot_loss[loss=0.2642, simple_loss=0.3291, pruned_loss=0.09965, over 3799549.49 frames. ], batch size: 54, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:39:57,615 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43126.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:40:12,343 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43137.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:40:23,158 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.978e+02 6.527e+02 7.673e+02 9.951e+02 2.226e+03, threshold=1.535e+03, percent-clipped=3.0 +2023-04-01 09:40:40,206 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-04-01 09:40:49,521 INFO [train.py:903] (0/4) Epoch 7, batch 2200, loss[loss=0.2411, simple_loss=0.306, pruned_loss=0.08807, over 19578.00 frames. ], tot_loss[loss=0.2646, simple_loss=0.3296, pruned_loss=0.09985, over 3797253.78 frames. ], batch size: 52, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:41:16,781 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2714, 1.1038, 1.5290, 1.0996, 2.7959, 3.4899, 3.3282, 3.6916], + device='cuda:0'), covar=tensor([0.1403, 0.3143, 0.2951, 0.1921, 0.0399, 0.0131, 0.0205, 0.0150], + device='cuda:0'), in_proj_covar=tensor([0.0250, 0.0284, 0.0314, 0.0247, 0.0203, 0.0134, 0.0204, 0.0161], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 09:41:53,420 INFO [train.py:903] (0/4) Epoch 7, batch 2250, loss[loss=0.2364, simple_loss=0.3098, pruned_loss=0.08152, over 19588.00 frames. ], tot_loss[loss=0.262, simple_loss=0.3273, pruned_loss=0.09839, over 3824165.71 frames. ], batch size: 52, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:41:57,890 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=43221.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:42:19,712 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43239.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:42:21,846 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=43241.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:42:27,686 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.233e+02 6.206e+02 7.577e+02 9.251e+02 2.641e+03, threshold=1.515e+03, percent-clipped=5.0 +2023-04-01 09:42:31,784 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3485, 1.5061, 1.9941, 1.7022, 3.1447, 2.4610, 3.4515, 1.3243], + device='cuda:0'), covar=tensor([0.2186, 0.3685, 0.2235, 0.1686, 0.1416, 0.1808, 0.1503, 0.3324], + device='cuda:0'), in_proj_covar=tensor([0.0462, 0.0523, 0.0524, 0.0418, 0.0576, 0.0464, 0.0639, 0.0459], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 09:42:36,529 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3129, 2.3930, 1.6940, 1.4469, 2.1413, 1.1479, 1.1942, 1.8141], + device='cuda:0'), covar=tensor([0.0806, 0.0484, 0.0792, 0.0616, 0.0371, 0.0916, 0.0673, 0.0369], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0280, 0.0318, 0.0239, 0.0222, 0.0313, 0.0281, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:42:52,514 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43264.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:42:56,677 INFO [train.py:903] (0/4) Epoch 7, batch 2300, loss[loss=0.2577, simple_loss=0.3267, pruned_loss=0.09435, over 19607.00 frames. ], tot_loss[loss=0.2636, simple_loss=0.3291, pruned_loss=0.09911, over 3818590.85 frames. ], batch size: 57, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:43:10,467 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 09:43:59,263 INFO [train.py:903] (0/4) Epoch 7, batch 2350, loss[loss=0.2508, simple_loss=0.3213, pruned_loss=0.09012, over 19541.00 frames. ], tot_loss[loss=0.2631, simple_loss=0.3287, pruned_loss=0.09873, over 3824561.40 frames. ], batch size: 54, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:44:07,555 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8446, 1.9462, 1.9852, 2.8001, 1.7964, 2.6112, 2.5611, 1.9433], + device='cuda:0'), covar=tensor([0.2883, 0.2263, 0.1142, 0.1250, 0.2616, 0.0953, 0.2252, 0.1998], + device='cuda:0'), in_proj_covar=tensor([0.0696, 0.0692, 0.0599, 0.0837, 0.0713, 0.0611, 0.0729, 0.0635], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 09:44:20,934 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43336.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:44:34,241 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.645e+02 6.151e+02 7.487e+02 9.533e+02 1.563e+03, threshold=1.497e+03, percent-clipped=2.0 +2023-04-01 09:44:43,148 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 09:44:46,804 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43356.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:44:59,568 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 09:45:00,596 INFO [train.py:903] (0/4) Epoch 7, batch 2400, loss[loss=0.292, simple_loss=0.3492, pruned_loss=0.1174, over 13339.00 frames. ], tot_loss[loss=0.2617, simple_loss=0.3277, pruned_loss=0.09789, over 3828688.68 frames. ], batch size: 136, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:45:20,051 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43382.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:45:49,711 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43407.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:46:04,429 INFO [train.py:903] (0/4) Epoch 7, batch 2450, loss[loss=0.2203, simple_loss=0.2912, pruned_loss=0.07468, over 19462.00 frames. ], tot_loss[loss=0.2603, simple_loss=0.3264, pruned_loss=0.0971, over 3832083.30 frames. ], batch size: 49, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:46:38,146 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.412e+02 5.802e+02 7.669e+02 8.855e+02 2.284e+03, threshold=1.534e+03, percent-clipped=5.0 +2023-04-01 09:46:43,558 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.73 vs. limit=2.0 +2023-04-01 09:47:02,390 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2652, 1.5534, 1.9293, 1.7575, 2.9712, 4.3552, 4.3754, 4.7712], + device='cuda:0'), covar=tensor([0.1465, 0.2806, 0.2762, 0.1779, 0.0452, 0.0245, 0.0158, 0.0101], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0286, 0.0316, 0.0250, 0.0205, 0.0135, 0.0205, 0.0161], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 09:47:06,555 INFO [train.py:903] (0/4) Epoch 7, batch 2500, loss[loss=0.2946, simple_loss=0.3525, pruned_loss=0.1183, over 18139.00 frames. ], tot_loss[loss=0.2603, simple_loss=0.3265, pruned_loss=0.09702, over 3831669.11 frames. ], batch size: 83, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:47:38,462 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-04-01 09:48:09,515 INFO [train.py:903] (0/4) Epoch 7, batch 2550, loss[loss=0.2445, simple_loss=0.3153, pruned_loss=0.08681, over 19741.00 frames. ], tot_loss[loss=0.2601, simple_loss=0.3264, pruned_loss=0.09691, over 3818273.08 frames. ], batch size: 51, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:48:25,709 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43532.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:48:44,537 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.143e+02 6.070e+02 7.288e+02 8.830e+02 1.707e+03, threshold=1.458e+03, percent-clipped=2.0 +2023-04-01 09:49:01,566 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2723, 1.0931, 1.4986, 1.1419, 2.4588, 3.1671, 2.9964, 3.4794], + device='cuda:0'), covar=tensor([0.1561, 0.4311, 0.3892, 0.2173, 0.0531, 0.0219, 0.0303, 0.0183], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0284, 0.0315, 0.0247, 0.0202, 0.0133, 0.0203, 0.0160], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 09:49:05,669 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 09:49:10,238 INFO [train.py:903] (0/4) Epoch 7, batch 2600, loss[loss=0.2454, simple_loss=0.318, pruned_loss=0.08638, over 19664.00 frames. ], tot_loss[loss=0.261, simple_loss=0.3269, pruned_loss=0.09756, over 3803463.67 frames. ], batch size: 55, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:49:41,285 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43592.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:49:59,882 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.03 vs. limit=5.0 +2023-04-01 09:50:05,431 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=43612.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:50:12,240 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43617.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:50:12,993 INFO [train.py:903] (0/4) Epoch 7, batch 2650, loss[loss=0.2923, simple_loss=0.3491, pruned_loss=0.1177, over 18136.00 frames. ], tot_loss[loss=0.2624, simple_loss=0.3277, pruned_loss=0.09852, over 3786872.03 frames. ], batch size: 83, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:50:28,013 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8786, 1.8480, 1.6248, 1.4931, 1.4205, 1.6362, 0.3136, 0.9189], + device='cuda:0'), covar=tensor([0.0282, 0.0305, 0.0187, 0.0294, 0.0523, 0.0305, 0.0592, 0.0472], + device='cuda:0'), in_proj_covar=tensor([0.0305, 0.0310, 0.0303, 0.0329, 0.0396, 0.0322, 0.0291, 0.0306], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 09:50:35,411 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 09:50:38,352 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=43637.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:50:49,495 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.166e+02 6.243e+02 7.316e+02 9.618e+02 1.411e+03, threshold=1.463e+03, percent-clipped=0.0 +2023-04-01 09:51:16,495 INFO [train.py:903] (0/4) Epoch 7, batch 2700, loss[loss=0.2258, simple_loss=0.2953, pruned_loss=0.07811, over 19826.00 frames. ], tot_loss[loss=0.2618, simple_loss=0.3271, pruned_loss=0.0982, over 3806138.69 frames. ], batch size: 49, lr: 1.20e-02, grad_scale: 4.0 +2023-04-01 09:51:25,752 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1857, 1.1837, 1.1623, 1.3256, 1.0496, 1.3424, 1.3425, 1.2918], + device='cuda:0'), covar=tensor([0.0901, 0.1007, 0.1112, 0.0738, 0.0851, 0.0839, 0.0824, 0.0769], + device='cuda:0'), in_proj_covar=tensor([0.0228, 0.0241, 0.0236, 0.0271, 0.0258, 0.0222, 0.0219, 0.0216], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 09:51:26,106 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.64 vs. limit=5.0 +2023-04-01 09:51:44,012 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-04-01 09:52:07,225 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9840, 1.9439, 2.2594, 2.9938, 2.7096, 2.5192, 1.8708, 2.8029], + device='cuda:0'), covar=tensor([0.0668, 0.1707, 0.1213, 0.0737, 0.1030, 0.0398, 0.1164, 0.0533], + device='cuda:0'), in_proj_covar=tensor([0.0246, 0.0357, 0.0291, 0.0237, 0.0304, 0.0247, 0.0271, 0.0234], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:52:19,231 INFO [train.py:903] (0/4) Epoch 7, batch 2750, loss[loss=0.3068, simple_loss=0.3712, pruned_loss=0.1212, over 19481.00 frames. ], tot_loss[loss=0.262, simple_loss=0.3272, pruned_loss=0.09838, over 3807823.33 frames. ], batch size: 64, lr: 1.20e-02, grad_scale: 4.0 +2023-04-01 09:52:43,620 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4993, 1.0553, 1.2399, 1.2369, 2.1726, 0.9072, 1.7188, 2.2609], + device='cuda:0'), covar=tensor([0.0560, 0.2507, 0.2403, 0.1409, 0.0741, 0.1949, 0.1019, 0.0528], + device='cuda:0'), in_proj_covar=tensor([0.0305, 0.0313, 0.0322, 0.0291, 0.0319, 0.0312, 0.0297, 0.0314], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:52:55,474 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.581e+02 6.869e+02 8.299e+02 1.091e+03 2.331e+03, threshold=1.660e+03, percent-clipped=8.0 +2023-04-01 09:53:20,556 INFO [train.py:903] (0/4) Epoch 7, batch 2800, loss[loss=0.2485, simple_loss=0.3091, pruned_loss=0.09395, over 19771.00 frames. ], tot_loss[loss=0.2617, simple_loss=0.3273, pruned_loss=0.09805, over 3819070.34 frames. ], batch size: 46, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:53:35,880 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3235, 1.0920, 1.5845, 1.3593, 2.7110, 3.7593, 3.6050, 4.0154], + device='cuda:0'), covar=tensor([0.1410, 0.2995, 0.2700, 0.1887, 0.0458, 0.0148, 0.0179, 0.0123], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0283, 0.0311, 0.0245, 0.0200, 0.0132, 0.0203, 0.0160], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0001, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 09:54:07,559 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43805.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:54:22,954 INFO [train.py:903] (0/4) Epoch 7, batch 2850, loss[loss=0.2555, simple_loss=0.3141, pruned_loss=0.09842, over 19389.00 frames. ], tot_loss[loss=0.261, simple_loss=0.3264, pruned_loss=0.09786, over 3813589.94 frames. ], batch size: 48, lr: 1.20e-02, grad_scale: 8.0 +2023-04-01 09:54:59,107 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.801e+02 6.316e+02 8.520e+02 1.005e+03 1.613e+03, threshold=1.704e+03, percent-clipped=0.0 +2023-04-01 09:55:02,921 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5345, 1.2564, 1.3742, 1.2253, 2.1986, 0.8769, 1.8417, 2.3325], + device='cuda:0'), covar=tensor([0.0518, 0.2202, 0.2149, 0.1332, 0.0763, 0.1794, 0.0849, 0.0467], + device='cuda:0'), in_proj_covar=tensor([0.0310, 0.0315, 0.0326, 0.0296, 0.0323, 0.0316, 0.0300, 0.0319], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:55:25,665 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 09:55:26,287 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 09:55:26,752 INFO [train.py:903] (0/4) Epoch 7, batch 2900, loss[loss=0.1859, simple_loss=0.263, pruned_loss=0.05437, over 19346.00 frames. ], tot_loss[loss=0.2606, simple_loss=0.326, pruned_loss=0.09756, over 3800226.60 frames. ], batch size: 47, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 09:55:36,388 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=43876.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:56:28,085 INFO [train.py:903] (0/4) Epoch 7, batch 2950, loss[loss=0.2729, simple_loss=0.3487, pruned_loss=0.09857, over 19486.00 frames. ], tot_loss[loss=0.2592, simple_loss=0.3251, pruned_loss=0.09668, over 3813732.83 frames. ], batch size: 64, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 09:56:39,830 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43927.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:57:04,995 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.608e+02 5.923e+02 7.218e+02 9.278e+02 2.092e+03, threshold=1.444e+03, percent-clipped=1.0 +2023-04-01 09:57:30,521 INFO [train.py:903] (0/4) Epoch 7, batch 3000, loss[loss=0.2041, simple_loss=0.279, pruned_loss=0.06457, over 19350.00 frames. ], tot_loss[loss=0.259, simple_loss=0.3247, pruned_loss=0.09664, over 3825992.05 frames. ], batch size: 47, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 09:57:30,521 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 09:57:43,098 INFO [train.py:937] (0/4) Epoch 7, validation: loss=0.1917, simple_loss=0.2919, pruned_loss=0.04574, over 944034.00 frames. +2023-04-01 09:57:43,099 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 09:57:49,866 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 09:57:54,933 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=43977.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:58:08,008 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5423, 2.5078, 1.6863, 1.6407, 2.2339, 1.2321, 1.2965, 1.8624], + device='cuda:0'), covar=tensor([0.0794, 0.0511, 0.0859, 0.0576, 0.0326, 0.0963, 0.0624, 0.0342], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0277, 0.0315, 0.0237, 0.0224, 0.0305, 0.0276, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 09:58:13,799 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=43991.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 09:58:24,506 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-44000.pt +2023-04-01 09:58:47,326 INFO [train.py:903] (0/4) Epoch 7, batch 3050, loss[loss=0.2485, simple_loss=0.3162, pruned_loss=0.09036, over 19623.00 frames. ], tot_loss[loss=0.2598, simple_loss=0.3255, pruned_loss=0.09704, over 3830704.28 frames. ], batch size: 50, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 09:59:24,086 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.748e+02 5.923e+02 7.306e+02 8.819e+02 1.422e+03, threshold=1.461e+03, percent-clipped=0.0 +2023-04-01 09:59:50,240 INFO [train.py:903] (0/4) Epoch 7, batch 3100, loss[loss=0.2248, simple_loss=0.2931, pruned_loss=0.07825, over 19737.00 frames. ], tot_loss[loss=0.2595, simple_loss=0.3253, pruned_loss=0.09688, over 3836292.04 frames. ], batch size: 46, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 10:00:17,741 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2767, 1.4203, 1.8988, 1.5341, 2.7470, 2.2410, 2.8759, 1.1758], + device='cuda:0'), covar=tensor([0.1814, 0.3031, 0.1678, 0.1413, 0.1129, 0.1479, 0.1265, 0.2835], + device='cuda:0'), in_proj_covar=tensor([0.0457, 0.0524, 0.0518, 0.0409, 0.0562, 0.0457, 0.0631, 0.0460], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 10:00:50,927 INFO [train.py:903] (0/4) Epoch 7, batch 3150, loss[loss=0.251, simple_loss=0.3169, pruned_loss=0.09253, over 18132.00 frames. ], tot_loss[loss=0.2617, simple_loss=0.3272, pruned_loss=0.09808, over 3826725.25 frames. ], batch size: 83, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 10:01:18,570 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 10:01:26,086 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.269e+02 6.470e+02 8.010e+02 1.018e+03 2.357e+03, threshold=1.602e+03, percent-clipped=4.0 +2023-04-01 10:01:28,521 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=44149.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:01:31,187 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-04-01 10:01:51,333 INFO [train.py:903] (0/4) Epoch 7, batch 3200, loss[loss=0.3008, simple_loss=0.3608, pruned_loss=0.1204, over 19658.00 frames. ], tot_loss[loss=0.261, simple_loss=0.3264, pruned_loss=0.09784, over 3820600.36 frames. ], batch size: 60, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 10:02:48,450 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1765, 1.2082, 1.8477, 1.5039, 2.8476, 4.4916, 4.4553, 5.0130], + device='cuda:0'), covar=tensor([0.1531, 0.3138, 0.2704, 0.1813, 0.0456, 0.0151, 0.0134, 0.0069], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0284, 0.0312, 0.0247, 0.0202, 0.0133, 0.0204, 0.0162], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 10:02:51,382 INFO [train.py:903] (0/4) Epoch 7, batch 3250, loss[loss=0.2708, simple_loss=0.3398, pruned_loss=0.1009, over 19612.00 frames. ], tot_loss[loss=0.261, simple_loss=0.3263, pruned_loss=0.09787, over 3818231.07 frames. ], batch size: 61, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 10:03:27,995 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.707e+02 6.408e+02 8.261e+02 1.024e+03 1.757e+03, threshold=1.652e+03, percent-clipped=4.0 +2023-04-01 10:03:28,411 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44247.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:03:49,182 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44264.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:03:53,343 INFO [train.py:903] (0/4) Epoch 7, batch 3300, loss[loss=0.2544, simple_loss=0.3372, pruned_loss=0.08579, over 19686.00 frames. ], tot_loss[loss=0.2614, simple_loss=0.3271, pruned_loss=0.09784, over 3812602.91 frames. ], batch size: 60, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 10:03:59,175 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=44271.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:04:00,222 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 10:04:00,594 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=44272.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:04:57,173 INFO [train.py:903] (0/4) Epoch 7, batch 3350, loss[loss=0.2655, simple_loss=0.33, pruned_loss=0.1005, over 19592.00 frames. ], tot_loss[loss=0.2616, simple_loss=0.3273, pruned_loss=0.09799, over 3812840.87 frames. ], batch size: 52, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 10:05:00,795 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=44321.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:05:32,241 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.423e+02 6.124e+02 7.865e+02 1.014e+03 2.362e+03, threshold=1.573e+03, percent-clipped=3.0 +2023-04-01 10:05:58,470 INFO [train.py:903] (0/4) Epoch 7, batch 3400, loss[loss=0.2457, simple_loss=0.302, pruned_loss=0.09473, over 19740.00 frames. ], tot_loss[loss=0.262, simple_loss=0.3276, pruned_loss=0.09823, over 3821664.52 frames. ], batch size: 47, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 10:06:05,548 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.1970, 5.1069, 6.0423, 5.9473, 1.8888, 5.6138, 4.8794, 5.5583], + device='cuda:0'), covar=tensor([0.1033, 0.0567, 0.0375, 0.0331, 0.4616, 0.0255, 0.0424, 0.0900], + device='cuda:0'), in_proj_covar=tensor([0.0575, 0.0507, 0.0682, 0.0555, 0.0631, 0.0426, 0.0432, 0.0627], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 10:06:20,848 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44386.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:07:01,356 INFO [train.py:903] (0/4) Epoch 7, batch 3450, loss[loss=0.1958, simple_loss=0.2644, pruned_loss=0.06353, over 19406.00 frames. ], tot_loss[loss=0.2622, simple_loss=0.3277, pruned_loss=0.09829, over 3827351.05 frames. ], batch size: 47, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 10:07:06,920 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 10:07:09,602 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5636, 1.9089, 2.1405, 2.4774, 1.8124, 2.3844, 2.7259, 2.5316], + device='cuda:0'), covar=tensor([0.0738, 0.0961, 0.1011, 0.1049, 0.1080, 0.0789, 0.0876, 0.0617], + device='cuda:0'), in_proj_covar=tensor([0.0226, 0.0238, 0.0234, 0.0269, 0.0253, 0.0220, 0.0218, 0.0213], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 10:07:20,235 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3758, 1.2360, 1.6826, 1.2925, 2.7292, 3.7173, 3.5220, 3.9222], + device='cuda:0'), covar=tensor([0.1371, 0.2940, 0.2688, 0.1809, 0.0430, 0.0157, 0.0197, 0.0137], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0284, 0.0314, 0.0248, 0.0203, 0.0134, 0.0202, 0.0162], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 10:07:25,067 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=44436.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:07:39,986 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.210e+02 5.624e+02 7.209e+02 9.228e+02 1.461e+03, threshold=1.442e+03, percent-clipped=0.0 +2023-04-01 10:08:03,657 INFO [train.py:903] (0/4) Epoch 7, batch 3500, loss[loss=0.2566, simple_loss=0.3309, pruned_loss=0.09112, over 19595.00 frames. ], tot_loss[loss=0.2622, simple_loss=0.3278, pruned_loss=0.09825, over 3833364.64 frames. ], batch size: 61, lr: 1.19e-02, grad_scale: 4.0 +2023-04-01 10:09:07,864 INFO [train.py:903] (0/4) Epoch 7, batch 3550, loss[loss=0.2428, simple_loss=0.3013, pruned_loss=0.09217, over 19741.00 frames. ], tot_loss[loss=0.2622, simple_loss=0.3278, pruned_loss=0.0983, over 3821433.01 frames. ], batch size: 47, lr: 1.19e-02, grad_scale: 4.0 +2023-04-01 10:09:10,459 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44520.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:09:40,592 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44545.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:09:40,714 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=44545.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:09:43,711 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.445e+02 5.736e+02 7.487e+02 9.426e+02 2.431e+03, threshold=1.497e+03, percent-clipped=5.0 +2023-04-01 10:09:52,445 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3335, 2.2680, 1.4902, 1.3022, 1.8331, 1.0478, 1.1874, 1.8154], + device='cuda:0'), covar=tensor([0.0695, 0.0472, 0.0876, 0.0595, 0.0441, 0.1009, 0.0622, 0.0358], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0280, 0.0317, 0.0242, 0.0224, 0.0314, 0.0288, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 10:10:10,023 INFO [train.py:903] (0/4) Epoch 7, batch 3600, loss[loss=0.2376, simple_loss=0.3195, pruned_loss=0.07787, over 19417.00 frames. ], tot_loss[loss=0.2634, simple_loss=0.3289, pruned_loss=0.09898, over 3816922.01 frames. ], batch size: 70, lr: 1.19e-02, grad_scale: 8.0 +2023-04-01 10:10:10,221 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=44568.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:11:12,083 INFO [train.py:903] (0/4) Epoch 7, batch 3650, loss[loss=0.2774, simple_loss=0.3333, pruned_loss=0.1107, over 19673.00 frames. ], tot_loss[loss=0.2635, simple_loss=0.3288, pruned_loss=0.09914, over 3817038.23 frames. ], batch size: 53, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:11:42,351 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44642.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:11:49,797 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.538e+02 6.222e+02 7.769e+02 9.647e+02 2.431e+03, threshold=1.554e+03, percent-clipped=4.0 +2023-04-01 10:12:12,892 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=44667.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:12:13,721 INFO [train.py:903] (0/4) Epoch 7, batch 3700, loss[loss=0.2468, simple_loss=0.3197, pruned_loss=0.08696, over 19676.00 frames. ], tot_loss[loss=0.2622, simple_loss=0.3276, pruned_loss=0.09838, over 3820726.15 frames. ], batch size: 53, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:12:37,014 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.2332, 5.5419, 3.1483, 4.8847, 1.0866, 5.4465, 5.4247, 5.7870], + device='cuda:0'), covar=tensor([0.0397, 0.0855, 0.1778, 0.0620, 0.4135, 0.0658, 0.0628, 0.0687], + device='cuda:0'), in_proj_covar=tensor([0.0380, 0.0337, 0.0393, 0.0296, 0.0362, 0.0322, 0.0307, 0.0341], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 10:12:45,598 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=44692.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:12:49,113 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.67 vs. limit=2.0 +2023-04-01 10:13:16,171 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=44717.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:13:16,975 INFO [train.py:903] (0/4) Epoch 7, batch 3750, loss[loss=0.2902, simple_loss=0.3567, pruned_loss=0.1119, over 19662.00 frames. ], tot_loss[loss=0.2629, simple_loss=0.3281, pruned_loss=0.09889, over 3815828.18 frames. ], batch size: 58, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:13:53,510 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.417e+02 6.149e+02 7.382e+02 9.468e+02 1.650e+03, threshold=1.476e+03, percent-clipped=2.0 +2023-04-01 10:14:18,151 INFO [train.py:903] (0/4) Epoch 7, batch 3800, loss[loss=0.2692, simple_loss=0.3331, pruned_loss=0.1027, over 19768.00 frames. ], tot_loss[loss=0.2623, simple_loss=0.3281, pruned_loss=0.09823, over 3819726.90 frames. ], batch size: 54, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:14:50,696 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 10:14:59,337 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0102, 1.2062, 1.3164, 1.5552, 2.6515, 1.0061, 1.8605, 2.7381], + device='cuda:0'), covar=tensor([0.0429, 0.2399, 0.2542, 0.1327, 0.0584, 0.2098, 0.1161, 0.0423], + device='cuda:0'), in_proj_covar=tensor([0.0305, 0.0314, 0.0325, 0.0292, 0.0318, 0.0311, 0.0294, 0.0314], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 10:15:06,260 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1506, 1.2105, 1.4395, 1.3070, 1.8481, 1.8339, 1.8775, 0.4860], + device='cuda:0'), covar=tensor([0.1997, 0.3489, 0.1915, 0.1614, 0.1169, 0.1721, 0.1176, 0.3274], + device='cuda:0'), in_proj_covar=tensor([0.0460, 0.0531, 0.0526, 0.0416, 0.0572, 0.0460, 0.0636, 0.0464], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 10:15:19,311 INFO [train.py:903] (0/4) Epoch 7, batch 3850, loss[loss=0.2544, simple_loss=0.3287, pruned_loss=0.09007, over 19427.00 frames. ], tot_loss[loss=0.2638, simple_loss=0.3292, pruned_loss=0.09916, over 3800593.85 frames. ], batch size: 70, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:15:57,069 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.826e+02 6.658e+02 8.285e+02 1.095e+03 3.075e+03, threshold=1.657e+03, percent-clipped=10.0 +2023-04-01 10:16:20,993 INFO [train.py:903] (0/4) Epoch 7, batch 3900, loss[loss=0.2736, simple_loss=0.3219, pruned_loss=0.1126, over 19798.00 frames. ], tot_loss[loss=0.2628, simple_loss=0.3287, pruned_loss=0.0984, over 3806541.64 frames. ], batch size: 46, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:16:38,265 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2536, 1.0768, 1.1939, 1.3637, 0.9875, 1.3120, 1.3082, 1.2865], + device='cuda:0'), covar=tensor([0.0888, 0.1092, 0.1135, 0.0738, 0.0935, 0.0808, 0.0873, 0.0804], + device='cuda:0'), in_proj_covar=tensor([0.0222, 0.0234, 0.0231, 0.0263, 0.0250, 0.0217, 0.0215, 0.0211], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 10:16:48,321 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=44889.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:17:15,621 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=44912.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:17:24,388 INFO [train.py:903] (0/4) Epoch 7, batch 3950, loss[loss=0.2769, simple_loss=0.3358, pruned_loss=0.1091, over 19682.00 frames. ], tot_loss[loss=0.2621, simple_loss=0.3284, pruned_loss=0.09791, over 3819823.07 frames. ], batch size: 60, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:17:29,146 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 10:18:00,741 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.332e+02 5.655e+02 7.377e+02 9.527e+02 2.304e+03, threshold=1.475e+03, percent-clipped=3.0 +2023-04-01 10:18:26,676 INFO [train.py:903] (0/4) Epoch 7, batch 4000, loss[loss=0.2419, simple_loss=0.302, pruned_loss=0.09084, over 19807.00 frames. ], tot_loss[loss=0.2613, simple_loss=0.3276, pruned_loss=0.09754, over 3826813.76 frames. ], batch size: 48, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:19:02,821 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6426, 1.6500, 1.9211, 1.7585, 2.6097, 2.2173, 2.6974, 1.5310], + device='cuda:0'), covar=tensor([0.1426, 0.2511, 0.1442, 0.1167, 0.0944, 0.1238, 0.0914, 0.2323], + device='cuda:0'), in_proj_covar=tensor([0.0460, 0.0530, 0.0526, 0.0416, 0.0572, 0.0464, 0.0633, 0.0461], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 10:19:07,590 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9022, 4.2888, 4.6060, 4.5664, 1.6025, 4.2179, 3.6736, 4.2118], + device='cuda:0'), covar=tensor([0.1108, 0.0648, 0.0451, 0.0454, 0.4438, 0.0441, 0.0582, 0.1027], + device='cuda:0'), in_proj_covar=tensor([0.0588, 0.0514, 0.0691, 0.0566, 0.0643, 0.0432, 0.0439, 0.0645], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 10:19:11,966 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45004.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:19:15,207 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 10:19:27,795 INFO [train.py:903] (0/4) Epoch 7, batch 4050, loss[loss=0.2598, simple_loss=0.3389, pruned_loss=0.09034, over 18158.00 frames. ], tot_loss[loss=0.262, simple_loss=0.3282, pruned_loss=0.09788, over 3825474.80 frames. ], batch size: 83, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:19:39,276 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45027.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:20:05,075 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.186e+02 6.019e+02 6.941e+02 8.524e+02 1.564e+03, threshold=1.388e+03, percent-clipped=1.0 +2023-04-01 10:20:28,979 INFO [train.py:903] (0/4) Epoch 7, batch 4100, loss[loss=0.2757, simple_loss=0.3424, pruned_loss=0.1045, over 18379.00 frames. ], tot_loss[loss=0.2614, simple_loss=0.3276, pruned_loss=0.09758, over 3820558.25 frames. ], batch size: 84, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:20:36,276 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45074.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:21:05,600 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 10:21:07,064 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45099.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:21:30,598 INFO [train.py:903] (0/4) Epoch 7, batch 4150, loss[loss=0.2802, simple_loss=0.3422, pruned_loss=0.1091, over 19588.00 frames. ], tot_loss[loss=0.2626, simple_loss=0.3284, pruned_loss=0.09837, over 3818469.22 frames. ], batch size: 61, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:22:07,170 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.699e+02 6.430e+02 8.235e+02 1.004e+03 2.607e+03, threshold=1.647e+03, percent-clipped=6.0 +2023-04-01 10:22:14,699 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2030, 1.1815, 1.4488, 1.3283, 1.8059, 1.7351, 1.7742, 0.5392], + device='cuda:0'), covar=tensor([0.2332, 0.3901, 0.2150, 0.1879, 0.1376, 0.2100, 0.1333, 0.3646], + device='cuda:0'), in_proj_covar=tensor([0.0464, 0.0536, 0.0528, 0.0417, 0.0575, 0.0466, 0.0640, 0.0464], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 10:22:32,969 INFO [train.py:903] (0/4) Epoch 7, batch 4200, loss[loss=0.2593, simple_loss=0.3142, pruned_loss=0.1023, over 16826.00 frames. ], tot_loss[loss=0.262, simple_loss=0.328, pruned_loss=0.09802, over 3815836.31 frames. ], batch size: 37, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:22:38,319 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 10:22:41,074 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9776, 1.9379, 1.5872, 1.4995, 1.2489, 1.4029, 0.2992, 0.9298], + device='cuda:0'), covar=tensor([0.0481, 0.0478, 0.0334, 0.0512, 0.0991, 0.0631, 0.0830, 0.0733], + device='cuda:0'), in_proj_covar=tensor([0.0305, 0.0309, 0.0297, 0.0326, 0.0402, 0.0320, 0.0287, 0.0305], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 10:23:33,350 INFO [train.py:903] (0/4) Epoch 7, batch 4250, loss[loss=0.2186, simple_loss=0.303, pruned_loss=0.06715, over 19772.00 frames. ], tot_loss[loss=0.2607, simple_loss=0.327, pruned_loss=0.09722, over 3821559.65 frames. ], batch size: 56, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:23:49,629 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 10:24:01,331 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 10:24:12,490 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.934e+02 6.110e+02 8.094e+02 1.072e+03 2.309e+03, threshold=1.619e+03, percent-clipped=5.0 +2023-04-01 10:24:27,303 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45260.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:24:37,360 INFO [train.py:903] (0/4) Epoch 7, batch 4300, loss[loss=0.2226, simple_loss=0.2891, pruned_loss=0.07806, over 19755.00 frames. ], tot_loss[loss=0.2603, simple_loss=0.3264, pruned_loss=0.09707, over 3823591.97 frames. ], batch size: 45, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:24:57,127 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45283.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:24:59,375 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45285.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:25:27,023 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45308.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:25:32,122 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 10:25:39,637 INFO [train.py:903] (0/4) Epoch 7, batch 4350, loss[loss=0.2672, simple_loss=0.3355, pruned_loss=0.09948, over 19773.00 frames. ], tot_loss[loss=0.2606, simple_loss=0.3265, pruned_loss=0.09735, over 3824994.49 frames. ], batch size: 54, lr: 1.18e-02, grad_scale: 8.0 +2023-04-01 10:25:54,093 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.18 vs. limit=5.0 +2023-04-01 10:26:16,052 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.214e+02 6.479e+02 7.493e+02 1.002e+03 2.532e+03, threshold=1.499e+03, percent-clipped=5.0 +2023-04-01 10:26:39,963 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45366.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 10:26:42,861 INFO [train.py:903] (0/4) Epoch 7, batch 4400, loss[loss=0.2536, simple_loss=0.3226, pruned_loss=0.09231, over 17330.00 frames. ], tot_loss[loss=0.2605, simple_loss=0.3264, pruned_loss=0.09726, over 3834003.73 frames. ], batch size: 101, lr: 1.17e-02, grad_scale: 8.0 +2023-04-01 10:27:06,046 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 10:27:14,898 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 10:27:44,661 INFO [train.py:903] (0/4) Epoch 7, batch 4450, loss[loss=0.2676, simple_loss=0.332, pruned_loss=0.1016, over 19685.00 frames. ], tot_loss[loss=0.2614, simple_loss=0.3271, pruned_loss=0.09781, over 3824260.80 frames. ], batch size: 58, lr: 1.17e-02, grad_scale: 4.0 +2023-04-01 10:27:44,795 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=45418.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:28:16,040 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=45443.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:28:22,922 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.630e+02 6.151e+02 7.769e+02 9.586e+02 4.695e+03, threshold=1.554e+03, percent-clipped=4.0 +2023-04-01 10:28:46,109 INFO [train.py:903] (0/4) Epoch 7, batch 4500, loss[loss=0.2726, simple_loss=0.3429, pruned_loss=0.1011, over 19656.00 frames. ], tot_loss[loss=0.2615, simple_loss=0.3272, pruned_loss=0.09793, over 3830585.80 frames. ], batch size: 60, lr: 1.17e-02, grad_scale: 4.0 +2023-04-01 10:28:46,489 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5278, 1.3879, 1.1581, 1.4758, 1.4644, 1.2118, 1.1352, 1.3463], + device='cuda:0'), covar=tensor([0.1076, 0.1527, 0.1737, 0.1128, 0.1289, 0.0939, 0.1449, 0.1019], + device='cuda:0'), in_proj_covar=tensor([0.0240, 0.0350, 0.0281, 0.0232, 0.0294, 0.0239, 0.0263, 0.0227], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 10:29:48,542 INFO [train.py:903] (0/4) Epoch 7, batch 4550, loss[loss=0.3161, simple_loss=0.3609, pruned_loss=0.1356, over 12993.00 frames. ], tot_loss[loss=0.2622, simple_loss=0.328, pruned_loss=0.09817, over 3835636.78 frames. ], batch size: 136, lr: 1.17e-02, grad_scale: 4.0 +2023-04-01 10:30:00,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 10:30:09,079 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45533.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:30:23,722 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 10:30:27,043 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.783e+02 6.251e+02 7.662e+02 9.726e+02 2.453e+03, threshold=1.532e+03, percent-clipped=6.0 +2023-04-01 10:30:38,114 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45558.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:30:42,540 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.1832, 5.5791, 2.8843, 4.7279, 1.4248, 5.3322, 5.4591, 5.5225], + device='cuda:0'), covar=tensor([0.0368, 0.0723, 0.1658, 0.0505, 0.3451, 0.0499, 0.0549, 0.0519], + device='cuda:0'), in_proj_covar=tensor([0.0378, 0.0328, 0.0386, 0.0291, 0.0353, 0.0317, 0.0304, 0.0335], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 10:30:51,057 INFO [train.py:903] (0/4) Epoch 7, batch 4600, loss[loss=0.2318, simple_loss=0.3118, pruned_loss=0.07588, over 18779.00 frames. ], tot_loss[loss=0.2625, simple_loss=0.3287, pruned_loss=0.09814, over 3835483.07 frames. ], batch size: 74, lr: 1.17e-02, grad_scale: 4.0 +2023-04-01 10:31:21,632 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6762, 1.7496, 1.8623, 2.5266, 1.5905, 2.2130, 2.2702, 1.8555], + device='cuda:0'), covar=tensor([0.2746, 0.2164, 0.1168, 0.1196, 0.2384, 0.1009, 0.2412, 0.1937], + device='cuda:0'), in_proj_covar=tensor([0.0711, 0.0709, 0.0600, 0.0851, 0.0728, 0.0622, 0.0743, 0.0646], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 10:31:48,241 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-04-01 10:31:54,302 INFO [train.py:903] (0/4) Epoch 7, batch 4650, loss[loss=0.3372, simple_loss=0.3781, pruned_loss=0.1481, over 13505.00 frames. ], tot_loss[loss=0.2623, simple_loss=0.3285, pruned_loss=0.09811, over 3820551.00 frames. ], batch size: 135, lr: 1.17e-02, grad_scale: 4.0 +2023-04-01 10:31:57,361 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-04-01 10:32:11,875 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 10:32:24,508 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 10:32:33,275 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.198e+02 5.859e+02 7.403e+02 8.847e+02 2.429e+03, threshold=1.481e+03, percent-clipped=2.0 +2023-04-01 10:32:55,949 INFO [train.py:903] (0/4) Epoch 7, batch 4700, loss[loss=0.2416, simple_loss=0.3209, pruned_loss=0.08117, over 19745.00 frames. ], tot_loss[loss=0.2599, simple_loss=0.3263, pruned_loss=0.09676, over 3839251.87 frames. ], batch size: 63, lr: 1.17e-02, grad_scale: 4.0 +2023-04-01 10:33:09,339 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-01 10:33:18,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 10:33:47,800 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=45710.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 10:33:58,622 INFO [train.py:903] (0/4) Epoch 7, batch 4750, loss[loss=0.2532, simple_loss=0.3272, pruned_loss=0.08964, over 19708.00 frames. ], tot_loss[loss=0.2615, simple_loss=0.3276, pruned_loss=0.09773, over 3828241.97 frames. ], batch size: 59, lr: 1.17e-02, grad_scale: 4.0 +2023-04-01 10:34:01,206 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45720.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:34:11,839 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1067, 2.1178, 1.6130, 1.6019, 1.3310, 1.5526, 0.3515, 1.0136], + device='cuda:0'), covar=tensor([0.0415, 0.0401, 0.0345, 0.0542, 0.0853, 0.0645, 0.0764, 0.0700], + device='cuda:0'), in_proj_covar=tensor([0.0311, 0.0313, 0.0310, 0.0332, 0.0404, 0.0327, 0.0288, 0.0316], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 10:34:36,015 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.623e+02 6.576e+02 8.008e+02 9.322e+02 2.457e+03, threshold=1.602e+03, percent-clipped=6.0 +2023-04-01 10:35:01,470 INFO [train.py:903] (0/4) Epoch 7, batch 4800, loss[loss=0.3273, simple_loss=0.379, pruned_loss=0.1377, over 19288.00 frames. ], tot_loss[loss=0.2614, simple_loss=0.3271, pruned_loss=0.09785, over 3821779.53 frames. ], batch size: 66, lr: 1.17e-02, grad_scale: 8.0 +2023-04-01 10:35:14,193 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-01 10:35:26,434 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45789.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:35:29,614 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45792.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:35:30,839 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8671, 4.2525, 4.4646, 4.4419, 1.5318, 4.1510, 3.7721, 4.1251], + device='cuda:0'), covar=tensor([0.1079, 0.0577, 0.0500, 0.0477, 0.4518, 0.0437, 0.0464, 0.0985], + device='cuda:0'), in_proj_covar=tensor([0.0574, 0.0504, 0.0681, 0.0557, 0.0633, 0.0428, 0.0433, 0.0632], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 10:35:37,407 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6550, 2.0757, 2.0885, 2.9151, 2.0105, 2.5695, 2.6005, 2.5190], + device='cuda:0'), covar=tensor([0.0615, 0.0862, 0.0916, 0.0891, 0.0963, 0.0676, 0.0924, 0.0591], + device='cuda:0'), in_proj_covar=tensor([0.0220, 0.0237, 0.0233, 0.0264, 0.0253, 0.0220, 0.0217, 0.0215], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 10:35:58,068 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45814.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:35:58,111 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=45814.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:36:02,109 INFO [train.py:903] (0/4) Epoch 7, batch 4850, loss[loss=0.2647, simple_loss=0.3358, pruned_loss=0.09681, over 19655.00 frames. ], tot_loss[loss=0.2617, simple_loss=0.3274, pruned_loss=0.09805, over 3815419.51 frames. ], batch size: 58, lr: 1.17e-02, grad_scale: 8.0 +2023-04-01 10:36:10,635 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=45825.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 10:36:23,726 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-04-01 10:36:25,100 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 10:36:27,427 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=45839.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:36:40,692 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.989e+02 6.823e+02 8.718e+02 1.115e+03 2.265e+03, threshold=1.744e+03, percent-clipped=6.0 +2023-04-01 10:36:48,786 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 10:36:54,254 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 10:36:54,274 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 10:37:03,554 INFO [train.py:903] (0/4) Epoch 7, batch 4900, loss[loss=0.2458, simple_loss=0.3195, pruned_loss=0.08604, over 19581.00 frames. ], tot_loss[loss=0.2617, simple_loss=0.3273, pruned_loss=0.09807, over 3814466.42 frames. ], batch size: 61, lr: 1.17e-02, grad_scale: 8.0 +2023-04-01 10:37:04,798 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 10:37:14,483 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=45877.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:37:26,132 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 10:38:05,113 INFO [train.py:903] (0/4) Epoch 7, batch 4950, loss[loss=0.2553, simple_loss=0.3376, pruned_loss=0.08651, over 19684.00 frames. ], tot_loss[loss=0.2597, simple_loss=0.3254, pruned_loss=0.097, over 3820836.96 frames. ], batch size: 59, lr: 1.17e-02, grad_scale: 8.0 +2023-04-01 10:38:24,786 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 10:38:44,273 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.370e+02 6.277e+02 7.738e+02 9.356e+02 2.304e+03, threshold=1.548e+03, percent-clipped=1.0 +2023-04-01 10:38:44,961 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.75 vs. limit=5.0 +2023-04-01 10:38:47,739 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 10:39:09,686 INFO [train.py:903] (0/4) Epoch 7, batch 5000, loss[loss=0.3003, simple_loss=0.3546, pruned_loss=0.123, over 19620.00 frames. ], tot_loss[loss=0.2588, simple_loss=0.3246, pruned_loss=0.09647, over 3821574.96 frames. ], batch size: 57, lr: 1.17e-02, grad_scale: 8.0 +2023-04-01 10:39:18,583 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8555, 1.5180, 1.7615, 2.2958, 1.9381, 1.8409, 2.0600, 1.8816], + device='cuda:0'), covar=tensor([0.0863, 0.1370, 0.1098, 0.0925, 0.0996, 0.1039, 0.1053, 0.0859], + device='cuda:0'), in_proj_covar=tensor([0.0219, 0.0235, 0.0234, 0.0264, 0.0252, 0.0217, 0.0217, 0.0213], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 10:39:20,559 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 10:39:30,928 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 10:39:46,818 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-46000.pt +2023-04-01 10:40:12,336 INFO [train.py:903] (0/4) Epoch 7, batch 5050, loss[loss=0.2046, simple_loss=0.2694, pruned_loss=0.06991, over 19763.00 frames. ], tot_loss[loss=0.2598, simple_loss=0.3255, pruned_loss=0.097, over 3825625.76 frames. ], batch size: 46, lr: 1.17e-02, grad_scale: 8.0 +2023-04-01 10:40:49,612 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 10:40:51,923 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.897e+02 6.208e+02 7.894e+02 9.516e+02 2.052e+03, threshold=1.579e+03, percent-clipped=3.0 +2023-04-01 10:41:08,544 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46064.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:41:13,137 INFO [train.py:903] (0/4) Epoch 7, batch 5100, loss[loss=0.2315, simple_loss=0.289, pruned_loss=0.08694, over 19734.00 frames. ], tot_loss[loss=0.2604, simple_loss=0.3261, pruned_loss=0.09732, over 3805449.43 frames. ], batch size: 46, lr: 1.17e-02, grad_scale: 4.0 +2023-04-01 10:41:24,913 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 10:41:28,151 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 10:41:28,587 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46081.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 10:41:32,521 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 10:42:00,011 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46106.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 10:42:08,110 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4077, 1.2754, 1.7839, 1.7545, 3.2806, 4.6451, 4.5617, 5.0696], + device='cuda:0'), covar=tensor([0.1459, 0.3148, 0.2831, 0.1666, 0.0381, 0.0139, 0.0132, 0.0078], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0284, 0.0317, 0.0244, 0.0204, 0.0133, 0.0203, 0.0164], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 10:42:13,351 INFO [train.py:903] (0/4) Epoch 7, batch 5150, loss[loss=0.2199, simple_loss=0.2871, pruned_loss=0.07633, over 19369.00 frames. ], tot_loss[loss=0.2594, simple_loss=0.3255, pruned_loss=0.09664, over 3813036.06 frames. ], batch size: 47, lr: 1.17e-02, grad_scale: 4.0 +2023-04-01 10:42:26,341 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 10:42:37,776 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46136.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:42:53,637 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.315e+02 6.102e+02 7.228e+02 8.853e+02 1.806e+03, threshold=1.446e+03, percent-clipped=2.0 +2023-04-01 10:43:00,541 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 10:43:15,668 INFO [train.py:903] (0/4) Epoch 7, batch 5200, loss[loss=0.329, simple_loss=0.3859, pruned_loss=0.1361, over 19648.00 frames. ], tot_loss[loss=0.2597, simple_loss=0.3259, pruned_loss=0.0968, over 3817572.12 frames. ], batch size: 55, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:43:30,858 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46179.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:43:31,678 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 10:44:18,557 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 10:44:19,699 INFO [train.py:903] (0/4) Epoch 7, batch 5250, loss[loss=0.2576, simple_loss=0.3293, pruned_loss=0.09296, over 19510.00 frames. ], tot_loss[loss=0.2602, simple_loss=0.3266, pruned_loss=0.09693, over 3815877.90 frames. ], batch size: 54, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:44:23,422 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46221.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:44:52,954 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0903, 1.4594, 1.7086, 1.9731, 1.8766, 1.8003, 1.9556, 1.9574], + device='cuda:0'), covar=tensor([0.0739, 0.1600, 0.1153, 0.0856, 0.1046, 0.0446, 0.0794, 0.0537], + device='cuda:0'), in_proj_covar=tensor([0.0244, 0.0356, 0.0285, 0.0240, 0.0299, 0.0242, 0.0272, 0.0228], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 10:44:58,429 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.945e+02 5.998e+02 7.245e+02 8.826e+02 1.462e+03, threshold=1.449e+03, percent-clipped=1.0 +2023-04-01 10:45:00,912 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46251.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:45:21,289 INFO [train.py:903] (0/4) Epoch 7, batch 5300, loss[loss=0.2665, simple_loss=0.3406, pruned_loss=0.09619, over 19676.00 frames. ], tot_loss[loss=0.2602, simple_loss=0.3265, pruned_loss=0.09695, over 3803668.98 frames. ], batch size: 55, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:45:34,743 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-04-01 10:45:39,539 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 10:45:42,180 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4170, 1.7646, 2.0036, 2.2439, 1.7149, 2.5021, 2.6796, 2.3445], + device='cuda:0'), covar=tensor([0.0731, 0.1031, 0.1051, 0.1211, 0.1145, 0.0732, 0.0912, 0.0681], + device='cuda:0'), in_proj_covar=tensor([0.0218, 0.0234, 0.0234, 0.0263, 0.0253, 0.0219, 0.0216, 0.0212], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 10:45:53,407 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46294.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:46:22,052 INFO [train.py:903] (0/4) Epoch 7, batch 5350, loss[loss=0.3002, simple_loss=0.365, pruned_loss=0.1177, over 19280.00 frames. ], tot_loss[loss=0.259, simple_loss=0.3255, pruned_loss=0.09628, over 3795295.15 frames. ], batch size: 66, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:46:46,793 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46336.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:47:00,313 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 10:47:03,770 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.963e+02 5.681e+02 7.157e+02 9.578e+02 3.754e+03, threshold=1.431e+03, percent-clipped=4.0 +2023-04-01 10:47:12,116 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2475, 2.1627, 1.7767, 1.6413, 1.5403, 1.7705, 0.3111, 1.0640], + device='cuda:0'), covar=tensor([0.0354, 0.0369, 0.0280, 0.0439, 0.0728, 0.0472, 0.0718, 0.0645], + device='cuda:0'), in_proj_covar=tensor([0.0308, 0.0307, 0.0304, 0.0328, 0.0397, 0.0317, 0.0287, 0.0310], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 10:47:26,532 INFO [train.py:903] (0/4) Epoch 7, batch 5400, loss[loss=0.276, simple_loss=0.3437, pruned_loss=0.1041, over 19658.00 frames. ], tot_loss[loss=0.2586, simple_loss=0.3256, pruned_loss=0.0958, over 3799717.83 frames. ], batch size: 55, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:47:37,298 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.96 vs. limit=5.0 +2023-04-01 10:48:29,996 INFO [train.py:903] (0/4) Epoch 7, batch 5450, loss[loss=0.2864, simple_loss=0.358, pruned_loss=0.1074, over 19534.00 frames. ], tot_loss[loss=0.2603, simple_loss=0.3269, pruned_loss=0.09686, over 3808009.56 frames. ], batch size: 56, lr: 1.16e-02, grad_scale: 4.0 +2023-04-01 10:48:35,958 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0174, 5.1011, 5.8762, 5.7648, 1.6711, 5.5020, 4.7131, 5.4373], + device='cuda:0'), covar=tensor([0.1065, 0.0615, 0.0443, 0.0356, 0.5100, 0.0304, 0.0417, 0.0935], + device='cuda:0'), in_proj_covar=tensor([0.0578, 0.0511, 0.0691, 0.0573, 0.0642, 0.0438, 0.0437, 0.0644], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 10:48:49,874 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46435.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:49:10,731 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.600e+02 7.102e+02 8.361e+02 1.036e+03 2.875e+03, threshold=1.672e+03, percent-clipped=7.0 +2023-04-01 10:49:23,094 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46460.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:49:31,708 INFO [train.py:903] (0/4) Epoch 7, batch 5500, loss[loss=0.2413, simple_loss=0.3184, pruned_loss=0.0821, over 19801.00 frames. ], tot_loss[loss=0.2606, simple_loss=0.327, pruned_loss=0.09706, over 3824087.74 frames. ], batch size: 56, lr: 1.16e-02, grad_scale: 4.0 +2023-04-01 10:49:59,136 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 10:50:20,916 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46507.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:50:33,168 INFO [train.py:903] (0/4) Epoch 7, batch 5550, loss[loss=0.2811, simple_loss=0.3465, pruned_loss=0.1078, over 19587.00 frames. ], tot_loss[loss=0.2608, simple_loss=0.3272, pruned_loss=0.09726, over 3822511.84 frames. ], batch size: 61, lr: 1.16e-02, grad_scale: 4.0 +2023-04-01 10:50:43,607 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 10:50:51,456 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46532.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:51:02,836 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-01 10:51:15,055 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.541e+02 5.815e+02 7.044e+02 9.146e+02 3.032e+03, threshold=1.409e+03, percent-clipped=3.0 +2023-04-01 10:51:31,073 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 10:51:35,768 INFO [train.py:903] (0/4) Epoch 7, batch 5600, loss[loss=0.2036, simple_loss=0.2739, pruned_loss=0.0666, over 19575.00 frames. ], tot_loss[loss=0.261, simple_loss=0.327, pruned_loss=0.09753, over 3816646.87 frames. ], batch size: 52, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:52:06,759 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=46592.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:52:32,928 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46613.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:52:38,706 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=46617.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:52:40,323 INFO [train.py:903] (0/4) Epoch 7, batch 5650, loss[loss=0.2686, simple_loss=0.3348, pruned_loss=0.1012, over 19679.00 frames. ], tot_loss[loss=0.2634, simple_loss=0.3289, pruned_loss=0.09897, over 3810613.24 frames. ], batch size: 53, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:52:42,899 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46620.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:52:58,471 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.98 vs. limit=5.0 +2023-04-01 10:53:03,523 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46638.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:53:20,360 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.719e+02 5.976e+02 7.674e+02 9.559e+02 1.706e+03, threshold=1.535e+03, percent-clipped=4.0 +2023-04-01 10:53:28,315 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 10:53:42,094 INFO [train.py:903] (0/4) Epoch 7, batch 5700, loss[loss=0.26, simple_loss=0.3212, pruned_loss=0.09942, over 18598.00 frames. ], tot_loss[loss=0.2628, simple_loss=0.3286, pruned_loss=0.0985, over 3819362.90 frames. ], batch size: 41, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:53:53,874 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5742, 1.2517, 1.7518, 1.6710, 3.1137, 4.4411, 4.4370, 4.8649], + device='cuda:0'), covar=tensor([0.1486, 0.3307, 0.3078, 0.1841, 0.0430, 0.0159, 0.0154, 0.0087], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0284, 0.0315, 0.0245, 0.0205, 0.0135, 0.0203, 0.0168], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 10:54:43,275 INFO [train.py:903] (0/4) Epoch 7, batch 5750, loss[loss=0.2756, simple_loss=0.3359, pruned_loss=0.1077, over 19797.00 frames. ], tot_loss[loss=0.262, simple_loss=0.3277, pruned_loss=0.09815, over 3830201.43 frames. ], batch size: 56, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:54:45,654 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 10:54:55,173 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 10:55:00,701 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 10:55:25,496 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.168e+02 6.148e+02 7.021e+02 8.191e+02 1.564e+03, threshold=1.404e+03, percent-clipped=1.0 +2023-04-01 10:55:28,126 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=46753.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:55:28,163 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9588, 1.6468, 1.5604, 2.0728, 1.9236, 1.7314, 1.6632, 1.9295], + device='cuda:0'), covar=tensor([0.0815, 0.1603, 0.1363, 0.0898, 0.1120, 0.0499, 0.1044, 0.0583], + device='cuda:0'), in_proj_covar=tensor([0.0244, 0.0355, 0.0284, 0.0237, 0.0297, 0.0241, 0.0269, 0.0230], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 10:55:45,406 INFO [train.py:903] (0/4) Epoch 7, batch 5800, loss[loss=0.2839, simple_loss=0.352, pruned_loss=0.1078, over 19590.00 frames. ], tot_loss[loss=0.2622, simple_loss=0.328, pruned_loss=0.09824, over 3825341.56 frames. ], batch size: 61, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:56:49,193 INFO [train.py:903] (0/4) Epoch 7, batch 5850, loss[loss=0.228, simple_loss=0.3097, pruned_loss=0.07321, over 19658.00 frames. ], tot_loss[loss=0.2625, simple_loss=0.3282, pruned_loss=0.0984, over 3827360.83 frames. ], batch size: 53, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:57:02,082 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=46828.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:57:29,390 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.050e+02 5.967e+02 7.374e+02 9.303e+02 2.879e+03, threshold=1.475e+03, percent-clipped=6.0 +2023-04-01 10:57:51,527 INFO [train.py:903] (0/4) Epoch 7, batch 5900, loss[loss=0.2802, simple_loss=0.3469, pruned_loss=0.1068, over 19580.00 frames. ], tot_loss[loss=0.2619, simple_loss=0.3276, pruned_loss=0.09806, over 3827620.04 frames. ], batch size: 61, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:57:57,284 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 10:58:16,945 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 10:58:52,935 INFO [train.py:903] (0/4) Epoch 7, batch 5950, loss[loss=0.2172, simple_loss=0.2991, pruned_loss=0.06768, over 19760.00 frames. ], tot_loss[loss=0.2617, simple_loss=0.3278, pruned_loss=0.09778, over 3826114.27 frames. ], batch size: 54, lr: 1.16e-02, grad_scale: 8.0 +2023-04-01 10:59:34,059 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.877e+02 6.644e+02 7.805e+02 9.690e+02 1.794e+03, threshold=1.561e+03, percent-clipped=4.0 +2023-04-01 10:59:40,943 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46957.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:59:48,548 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=46964.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 10:59:52,255 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.33 vs. limit=5.0 +2023-04-01 10:59:52,799 INFO [train.py:903] (0/4) Epoch 7, batch 6000, loss[loss=0.2656, simple_loss=0.3359, pruned_loss=0.09761, over 19791.00 frames. ], tot_loss[loss=0.2622, simple_loss=0.3281, pruned_loss=0.09813, over 3829656.07 frames. ], batch size: 56, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 10:59:52,799 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 11:00:05,285 INFO [train.py:937] (0/4) Epoch 7, validation: loss=0.1903, simple_loss=0.2902, pruned_loss=0.04516, over 944034.00 frames. +2023-04-01 11:00:05,286 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 11:00:57,546 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47009.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:01:09,015 INFO [train.py:903] (0/4) Epoch 7, batch 6050, loss[loss=0.2508, simple_loss=0.3263, pruned_loss=0.08768, over 19699.00 frames. ], tot_loss[loss=0.2594, simple_loss=0.3257, pruned_loss=0.09654, over 3830388.48 frames. ], batch size: 60, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:01:30,910 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47034.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:01:49,944 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.317e+02 6.018e+02 7.577e+02 1.005e+03 2.434e+03, threshold=1.515e+03, percent-clipped=7.0 +2023-04-01 11:02:13,575 INFO [train.py:903] (0/4) Epoch 7, batch 6100, loss[loss=0.2146, simple_loss=0.2804, pruned_loss=0.07436, over 19729.00 frames. ], tot_loss[loss=0.2597, simple_loss=0.326, pruned_loss=0.09669, over 3808883.05 frames. ], batch size: 45, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:02:18,714 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47072.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:02:26,714 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47079.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:02:41,737 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9548, 4.4122, 4.7608, 4.6991, 1.7983, 4.3815, 3.8023, 4.3944], + device='cuda:0'), covar=tensor([0.1117, 0.0572, 0.0416, 0.0399, 0.4123, 0.0381, 0.0519, 0.0879], + device='cuda:0'), in_proj_covar=tensor([0.0584, 0.0511, 0.0694, 0.0574, 0.0647, 0.0446, 0.0440, 0.0655], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 11:03:15,755 INFO [train.py:903] (0/4) Epoch 7, batch 6150, loss[loss=0.2524, simple_loss=0.3336, pruned_loss=0.0856, over 19789.00 frames. ], tot_loss[loss=0.2588, simple_loss=0.3253, pruned_loss=0.09613, over 3811389.92 frames. ], batch size: 56, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:03:36,425 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4520, 2.2813, 1.5952, 1.4074, 1.9358, 1.0987, 1.1432, 1.7004], + device='cuda:0'), covar=tensor([0.0768, 0.0502, 0.0764, 0.0588, 0.0449, 0.0953, 0.0672, 0.0406], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0283, 0.0314, 0.0239, 0.0226, 0.0311, 0.0287, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 11:03:41,636 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 11:03:56,564 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.178e+02 5.986e+02 7.602e+02 9.492e+02 2.168e+03, threshold=1.520e+03, percent-clipped=5.0 +2023-04-01 11:04:01,805 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.60 vs. limit=5.0 +2023-04-01 11:04:15,667 INFO [train.py:903] (0/4) Epoch 7, batch 6200, loss[loss=0.2791, simple_loss=0.3399, pruned_loss=0.1092, over 19678.00 frames. ], tot_loss[loss=0.2585, simple_loss=0.325, pruned_loss=0.09603, over 3813528.68 frames. ], batch size: 60, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:04:20,338 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=47172.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:05:15,435 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8507, 1.5414, 1.6016, 2.0737, 1.6954, 2.1902, 2.2248, 2.0079], + device='cuda:0'), covar=tensor([0.0769, 0.1087, 0.1026, 0.0892, 0.0937, 0.0714, 0.0856, 0.0647], + device='cuda:0'), in_proj_covar=tensor([0.0218, 0.0238, 0.0233, 0.0264, 0.0253, 0.0219, 0.0213, 0.0212], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 11:05:17,523 INFO [train.py:903] (0/4) Epoch 7, batch 6250, loss[loss=0.2622, simple_loss=0.3315, pruned_loss=0.09644, over 19537.00 frames. ], tot_loss[loss=0.2573, simple_loss=0.3239, pruned_loss=0.09536, over 3827812.66 frames. ], batch size: 56, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:05:47,058 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 11:05:57,375 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.252e+02 6.379e+02 7.672e+02 9.726e+02 2.182e+03, threshold=1.534e+03, percent-clipped=2.0 +2023-04-01 11:06:08,287 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47259.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:06:15,454 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0370, 3.5947, 2.0138, 1.5262, 3.0171, 1.3608, 1.1780, 2.1377], + device='cuda:0'), covar=tensor([0.0738, 0.0231, 0.0610, 0.0634, 0.0322, 0.0907, 0.0766, 0.0397], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0285, 0.0317, 0.0241, 0.0226, 0.0314, 0.0288, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 11:06:19,223 INFO [train.py:903] (0/4) Epoch 7, batch 6300, loss[loss=0.2228, simple_loss=0.2949, pruned_loss=0.07537, over 19837.00 frames. ], tot_loss[loss=0.2572, simple_loss=0.324, pruned_loss=0.0952, over 3843584.07 frames. ], batch size: 52, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:06:43,112 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47287.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:07:21,227 INFO [train.py:903] (0/4) Epoch 7, batch 6350, loss[loss=0.2103, simple_loss=0.288, pruned_loss=0.06635, over 19474.00 frames. ], tot_loss[loss=0.2569, simple_loss=0.3239, pruned_loss=0.09496, over 3839197.58 frames. ], batch size: 49, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:07:33,170 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47328.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:07:36,026 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.45 vs. limit=2.0 +2023-04-01 11:07:41,269 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47335.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:07:49,423 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.25 vs. limit=5.0 +2023-04-01 11:08:02,949 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.061e+02 5.885e+02 7.911e+02 1.072e+03 3.322e+03, threshold=1.582e+03, percent-clipped=7.0 +2023-04-01 11:08:05,647 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47353.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:08:13,857 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47360.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:08:22,753 INFO [train.py:903] (0/4) Epoch 7, batch 6400, loss[loss=0.266, simple_loss=0.3363, pruned_loss=0.09787, over 19668.00 frames. ], tot_loss[loss=0.2567, simple_loss=0.3238, pruned_loss=0.09482, over 3845430.45 frames. ], batch size: 58, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:08:48,782 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.37 vs. limit=5.0 +2023-04-01 11:09:00,195 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47397.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:09:25,569 INFO [train.py:903] (0/4) Epoch 7, batch 6450, loss[loss=0.2811, simple_loss=0.3522, pruned_loss=0.105, over 19677.00 frames. ], tot_loss[loss=0.2587, simple_loss=0.3257, pruned_loss=0.09583, over 3823431.72 frames. ], batch size: 55, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:10:05,689 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.425e+02 6.283e+02 7.588e+02 1.008e+03 1.535e+03, threshold=1.518e+03, percent-clipped=0.0 +2023-04-01 11:10:08,031 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 11:10:26,639 INFO [train.py:903] (0/4) Epoch 7, batch 6500, loss[loss=0.237, simple_loss=0.3141, pruned_loss=0.0799, over 19662.00 frames. ], tot_loss[loss=0.257, simple_loss=0.3241, pruned_loss=0.09496, over 3813782.07 frames. ], batch size: 58, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:10:29,886 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 11:11:27,822 INFO [train.py:903] (0/4) Epoch 7, batch 6550, loss[loss=0.2614, simple_loss=0.3323, pruned_loss=0.09524, over 18127.00 frames. ], tot_loss[loss=0.2582, simple_loss=0.3252, pruned_loss=0.09563, over 3803273.80 frames. ], batch size: 83, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:11:58,213 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47543.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:12:10,160 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.342e+02 6.347e+02 8.071e+02 1.082e+03 2.174e+03, threshold=1.614e+03, percent-clipped=4.0 +2023-04-01 11:12:29,838 INFO [train.py:903] (0/4) Epoch 7, batch 6600, loss[loss=0.2858, simple_loss=0.3489, pruned_loss=0.1113, over 19527.00 frames. ], tot_loss[loss=0.2588, simple_loss=0.3257, pruned_loss=0.09591, over 3816702.64 frames. ], batch size: 56, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:12:30,228 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47568.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:13:00,157 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-04-01 11:13:13,378 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=47603.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:13:31,696 INFO [train.py:903] (0/4) Epoch 7, batch 6650, loss[loss=0.2592, simple_loss=0.3323, pruned_loss=0.09308, over 19496.00 frames. ], tot_loss[loss=0.2594, simple_loss=0.3262, pruned_loss=0.09628, over 3825756.89 frames. ], batch size: 64, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:14:11,831 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.526e+02 6.649e+02 7.986e+02 1.008e+03 1.623e+03, threshold=1.597e+03, percent-clipped=0.0 +2023-04-01 11:14:28,114 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2161, 1.1180, 1.1502, 1.2663, 0.9256, 1.3354, 1.3781, 1.1774], + device='cuda:0'), covar=tensor([0.0893, 0.1043, 0.1094, 0.0770, 0.0904, 0.0826, 0.0823, 0.0789], + device='cuda:0'), in_proj_covar=tensor([0.0220, 0.0237, 0.0234, 0.0267, 0.0251, 0.0221, 0.0216, 0.0213], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 11:14:32,252 INFO [train.py:903] (0/4) Epoch 7, batch 6700, loss[loss=0.2952, simple_loss=0.351, pruned_loss=0.1197, over 19674.00 frames. ], tot_loss[loss=0.2591, simple_loss=0.3263, pruned_loss=0.09595, over 3824288.28 frames. ], batch size: 53, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:15:30,802 INFO [train.py:903] (0/4) Epoch 7, batch 6750, loss[loss=0.28, simple_loss=0.3317, pruned_loss=0.1142, over 19463.00 frames. ], tot_loss[loss=0.2587, simple_loss=0.3257, pruned_loss=0.09582, over 3818778.44 frames. ], batch size: 49, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:15:31,149 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47718.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:15:31,676 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-04-01 11:15:52,143 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.01 vs. limit=2.0 +2023-04-01 11:15:58,012 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=47741.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:16:09,116 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.932e+02 5.831e+02 7.200e+02 8.445e+02 1.747e+03, threshold=1.440e+03, percent-clipped=3.0 +2023-04-01 11:16:15,147 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47756.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:16:18,707 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6240, 2.3582, 1.7582, 1.6939, 2.0410, 1.2860, 1.3961, 1.8332], + device='cuda:0'), covar=tensor([0.0690, 0.0448, 0.0714, 0.0509, 0.0465, 0.0889, 0.0605, 0.0425], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0282, 0.0314, 0.0241, 0.0230, 0.0310, 0.0286, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 11:16:28,694 INFO [train.py:903] (0/4) Epoch 7, batch 6800, loss[loss=0.2133, simple_loss=0.2824, pruned_loss=0.07207, over 19756.00 frames. ], tot_loss[loss=0.258, simple_loss=0.3253, pruned_loss=0.0954, over 3822335.85 frames. ], batch size: 47, lr: 1.15e-02, grad_scale: 8.0 +2023-04-01 11:16:33,961 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.10 vs. limit=2.0 +2023-04-01 11:16:59,276 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-7.pt +2023-04-01 11:17:14,720 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 11:17:15,773 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 11:17:18,393 INFO [train.py:903] (0/4) Epoch 8, batch 0, loss[loss=0.2561, simple_loss=0.332, pruned_loss=0.09009, over 19671.00 frames. ], tot_loss[loss=0.2561, simple_loss=0.332, pruned_loss=0.09009, over 19671.00 frames. ], batch size: 59, lr: 1.08e-02, grad_scale: 8.0 +2023-04-01 11:17:18,394 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 11:17:30,977 INFO [train.py:937] (0/4) Epoch 8, validation: loss=0.1916, simple_loss=0.2915, pruned_loss=0.0458, over 944034.00 frames. +2023-04-01 11:17:30,978 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 11:17:41,959 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 11:17:43,286 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47806.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:18:07,516 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=47827.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 11:18:15,587 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5316, 1.6156, 1.7261, 2.1369, 1.3676, 1.6799, 2.0452, 1.6525], + device='cuda:0'), covar=tensor([0.2619, 0.2063, 0.1097, 0.1194, 0.2478, 0.1162, 0.2516, 0.1961], + device='cuda:0'), in_proj_covar=tensor([0.0713, 0.0720, 0.0605, 0.0849, 0.0731, 0.0631, 0.0750, 0.0652], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:18:32,756 INFO [train.py:903] (0/4) Epoch 8, batch 50, loss[loss=0.2463, simple_loss=0.3236, pruned_loss=0.08452, over 19543.00 frames. ], tot_loss[loss=0.2651, simple_loss=0.3315, pruned_loss=0.0993, over 868754.44 frames. ], batch size: 56, lr: 1.08e-02, grad_scale: 8.0 +2023-04-01 11:18:38,615 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.560e+02 5.925e+02 7.453e+02 9.478e+02 2.348e+03, threshold=1.491e+03, percent-clipped=8.0 +2023-04-01 11:18:44,797 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=47856.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:19:06,101 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 11:19:32,492 INFO [train.py:903] (0/4) Epoch 8, batch 100, loss[loss=0.3056, simple_loss=0.366, pruned_loss=0.1226, over 19295.00 frames. ], tot_loss[loss=0.2618, simple_loss=0.3276, pruned_loss=0.09801, over 1530344.41 frames. ], batch size: 66, lr: 1.08e-02, grad_scale: 8.0 +2023-04-01 11:19:42,598 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 11:20:32,672 INFO [train.py:903] (0/4) Epoch 8, batch 150, loss[loss=0.2677, simple_loss=0.3445, pruned_loss=0.0955, over 19674.00 frames. ], tot_loss[loss=0.2622, simple_loss=0.3287, pruned_loss=0.09788, over 2035841.21 frames. ], batch size: 58, lr: 1.08e-02, grad_scale: 8.0 +2023-04-01 11:20:38,423 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.629e+02 6.208e+02 7.626e+02 9.440e+02 2.273e+03, threshold=1.525e+03, percent-clipped=3.0 +2023-04-01 11:20:38,856 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2618, 2.9356, 1.9394, 2.0277, 1.8987, 2.4042, 0.7259, 2.0499], + device='cuda:0'), covar=tensor([0.0296, 0.0315, 0.0407, 0.0561, 0.0636, 0.0487, 0.0720, 0.0623], + device='cuda:0'), in_proj_covar=tensor([0.0311, 0.0314, 0.0307, 0.0325, 0.0398, 0.0322, 0.0285, 0.0311], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 11:21:07,721 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=47974.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:21:29,220 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 11:21:32,695 INFO [train.py:903] (0/4) Epoch 8, batch 200, loss[loss=0.2513, simple_loss=0.3199, pruned_loss=0.09131, over 17501.00 frames. ], tot_loss[loss=0.2609, simple_loss=0.3271, pruned_loss=0.09732, over 2422774.94 frames. ], batch size: 101, lr: 1.08e-02, grad_scale: 8.0 +2023-04-01 11:21:36,553 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=47999.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:21:37,382 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-48000.pt +2023-04-01 11:22:15,558 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1211, 1.2279, 1.6910, 1.2982, 2.7300, 3.5729, 3.3441, 3.8082], + device='cuda:0'), covar=tensor([0.1530, 0.3190, 0.2871, 0.2014, 0.0467, 0.0150, 0.0224, 0.0153], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0284, 0.0317, 0.0247, 0.0208, 0.0136, 0.0204, 0.0172], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 11:22:35,824 INFO [train.py:903] (0/4) Epoch 8, batch 250, loss[loss=0.2497, simple_loss=0.3271, pruned_loss=0.08615, over 19525.00 frames. ], tot_loss[loss=0.2584, simple_loss=0.3258, pruned_loss=0.09553, over 2743390.47 frames. ], batch size: 56, lr: 1.08e-02, grad_scale: 8.0 +2023-04-01 11:22:42,343 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.557e+02 6.073e+02 7.344e+02 8.973e+02 2.163e+03, threshold=1.469e+03, percent-clipped=4.0 +2023-04-01 11:23:30,358 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48090.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:23:31,723 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2216, 1.2806, 1.6916, 1.4071, 2.6467, 2.1609, 2.8278, 1.1478], + device='cuda:0'), covar=tensor([0.1988, 0.3461, 0.1929, 0.1623, 0.1348, 0.1789, 0.1401, 0.3192], + device='cuda:0'), in_proj_covar=tensor([0.0455, 0.0533, 0.0530, 0.0416, 0.0572, 0.0467, 0.0624, 0.0459], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:23:36,839 INFO [train.py:903] (0/4) Epoch 8, batch 300, loss[loss=0.2957, simple_loss=0.3642, pruned_loss=0.1136, over 19696.00 frames. ], tot_loss[loss=0.2584, simple_loss=0.3257, pruned_loss=0.09555, over 2992738.77 frames. ], batch size: 59, lr: 1.08e-02, grad_scale: 8.0 +2023-04-01 11:23:41,627 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48100.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:23:55,920 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48112.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:23:59,587 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2900, 1.3280, 1.7482, 1.4860, 2.6836, 2.2515, 2.8190, 1.0948], + device='cuda:0'), covar=tensor([0.1915, 0.3512, 0.1942, 0.1589, 0.1305, 0.1642, 0.1444, 0.3179], + device='cuda:0'), in_proj_covar=tensor([0.0458, 0.0536, 0.0532, 0.0418, 0.0575, 0.0468, 0.0629, 0.0461], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:24:14,370 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1325, 2.1329, 2.2066, 3.3030, 2.2397, 3.2909, 2.8210, 1.9113], + device='cuda:0'), covar=tensor([0.3031, 0.2552, 0.1107, 0.1397, 0.2961, 0.0973, 0.2331, 0.2171], + device='cuda:0'), in_proj_covar=tensor([0.0721, 0.0723, 0.0607, 0.0852, 0.0734, 0.0632, 0.0748, 0.0657], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:24:27,759 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48137.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:24:37,683 INFO [train.py:903] (0/4) Epoch 8, batch 350, loss[loss=0.2285, simple_loss=0.307, pruned_loss=0.07501, over 19686.00 frames. ], tot_loss[loss=0.2558, simple_loss=0.3235, pruned_loss=0.09402, over 3185719.24 frames. ], batch size: 59, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:24:39,936 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 11:24:42,101 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48150.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:24:43,051 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.994e+02 6.192e+02 7.149e+02 9.949e+02 1.629e+03, threshold=1.430e+03, percent-clipped=6.0 +2023-04-01 11:25:08,554 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48171.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 11:25:22,139 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9245, 0.7280, 0.9645, 1.3702, 1.0350, 0.8681, 1.1388, 0.8721], + device='cuda:0'), covar=tensor([0.1078, 0.1839, 0.1319, 0.0718, 0.0903, 0.1395, 0.1036, 0.1133], + device='cuda:0'), in_proj_covar=tensor([0.0217, 0.0235, 0.0232, 0.0262, 0.0246, 0.0217, 0.0211, 0.0209], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 11:25:33,638 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3381, 1.2622, 1.7732, 1.6584, 2.8980, 4.3132, 4.4153, 4.8663], + device='cuda:0'), covar=tensor([0.1551, 0.3349, 0.3038, 0.1835, 0.0503, 0.0186, 0.0160, 0.0105], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0284, 0.0313, 0.0244, 0.0207, 0.0136, 0.0202, 0.0171], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 11:25:37,730 INFO [train.py:903] (0/4) Epoch 8, batch 400, loss[loss=0.3206, simple_loss=0.3825, pruned_loss=0.1293, over 19727.00 frames. ], tot_loss[loss=0.2577, simple_loss=0.325, pruned_loss=0.09516, over 3321570.49 frames. ], batch size: 63, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:25:57,754 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5934, 3.2057, 2.3990, 2.4150, 2.1031, 2.5762, 0.9838, 2.3571], + device='cuda:0'), covar=tensor([0.0345, 0.0272, 0.0347, 0.0508, 0.0615, 0.0479, 0.0673, 0.0545], + device='cuda:0'), in_proj_covar=tensor([0.0309, 0.0311, 0.0306, 0.0324, 0.0397, 0.0318, 0.0285, 0.0308], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 11:26:03,257 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5821, 4.1320, 2.5801, 3.6910, 0.9616, 3.7039, 3.8366, 4.0098], + device='cuda:0'), covar=tensor([0.0567, 0.0969, 0.1812, 0.0635, 0.3782, 0.0730, 0.0700, 0.0781], + device='cuda:0'), in_proj_covar=tensor([0.0390, 0.0340, 0.0395, 0.0297, 0.0364, 0.0322, 0.0316, 0.0348], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 11:26:03,419 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48215.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:26:13,139 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7399, 1.8391, 1.9307, 2.6215, 1.6740, 2.3531, 2.4565, 1.9008], + device='cuda:0'), covar=tensor([0.2857, 0.2367, 0.1173, 0.1319, 0.2663, 0.1112, 0.2348, 0.2062], + device='cuda:0'), in_proj_covar=tensor([0.0726, 0.0730, 0.0612, 0.0861, 0.0737, 0.0638, 0.0756, 0.0660], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:26:14,296 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0303, 2.1313, 2.1705, 3.0596, 2.2464, 2.9332, 2.8722, 2.0702], + device='cuda:0'), covar=tensor([0.2937, 0.2254, 0.1101, 0.1483, 0.2636, 0.1054, 0.2093, 0.1898], + device='cuda:0'), in_proj_covar=tensor([0.0726, 0.0730, 0.0612, 0.0861, 0.0737, 0.0638, 0.0756, 0.0660], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:26:21,192 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48230.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 11:26:38,938 INFO [train.py:903] (0/4) Epoch 8, batch 450, loss[loss=0.2047, simple_loss=0.2813, pruned_loss=0.06401, over 19490.00 frames. ], tot_loss[loss=0.2571, simple_loss=0.3242, pruned_loss=0.09499, over 3437183.67 frames. ], batch size: 49, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:26:45,631 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.339e+02 5.626e+02 7.034e+02 8.568e+02 1.629e+03, threshold=1.407e+03, percent-clipped=1.0 +2023-04-01 11:26:47,795 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1754, 1.1790, 1.5371, 0.9445, 2.3804, 3.0207, 2.7281, 3.1826], + device='cuda:0'), covar=tensor([0.1425, 0.3112, 0.2862, 0.2043, 0.0442, 0.0169, 0.0261, 0.0186], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0283, 0.0314, 0.0244, 0.0207, 0.0135, 0.0202, 0.0171], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 11:27:03,656 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48265.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:27:04,763 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1852, 1.2867, 1.5507, 1.2662, 2.6765, 3.4298, 3.2571, 3.6593], + device='cuda:0'), covar=tensor([0.1466, 0.3007, 0.2819, 0.1954, 0.0458, 0.0202, 0.0193, 0.0144], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0283, 0.0314, 0.0244, 0.0207, 0.0135, 0.0202, 0.0171], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 11:27:05,778 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6085, 4.7275, 5.2726, 5.2245, 2.0665, 4.8434, 4.3266, 4.9109], + device='cuda:0'), covar=tensor([0.1027, 0.0886, 0.0451, 0.0398, 0.4208, 0.0406, 0.0441, 0.0884], + device='cuda:0'), in_proj_covar=tensor([0.0591, 0.0515, 0.0700, 0.0588, 0.0645, 0.0447, 0.0446, 0.0647], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 11:27:11,123 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 11:27:12,255 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 11:27:27,259 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48286.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 11:27:41,080 INFO [train.py:903] (0/4) Epoch 8, batch 500, loss[loss=0.2983, simple_loss=0.3695, pruned_loss=0.1135, over 19096.00 frames. ], tot_loss[loss=0.2572, simple_loss=0.324, pruned_loss=0.09515, over 3522595.75 frames. ], batch size: 69, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:28:23,295 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-04-01 11:28:37,096 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48342.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:28:41,343 INFO [train.py:903] (0/4) Epoch 8, batch 550, loss[loss=0.2565, simple_loss=0.321, pruned_loss=0.09603, over 19757.00 frames. ], tot_loss[loss=0.2555, simple_loss=0.3231, pruned_loss=0.09394, over 3602439.07 frames. ], batch size: 47, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:28:47,050 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.262e+02 6.228e+02 7.563e+02 9.337e+02 1.593e+03, threshold=1.513e+03, percent-clipped=3.0 +2023-04-01 11:28:56,627 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5632, 1.9121, 1.9675, 2.5969, 2.3899, 2.3101, 2.1056, 2.4602], + device='cuda:0'), covar=tensor([0.0820, 0.1858, 0.1300, 0.0931, 0.1246, 0.0447, 0.1008, 0.0618], + device='cuda:0'), in_proj_covar=tensor([0.0245, 0.0354, 0.0285, 0.0241, 0.0301, 0.0243, 0.0273, 0.0233], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 11:29:42,852 INFO [train.py:903] (0/4) Epoch 8, batch 600, loss[loss=0.2307, simple_loss=0.2927, pruned_loss=0.08436, over 19755.00 frames. ], tot_loss[loss=0.2549, simple_loss=0.3227, pruned_loss=0.09356, over 3650258.95 frames. ], batch size: 46, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:29:49,152 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1538, 2.1270, 2.2518, 3.2982, 2.1804, 3.0796, 2.8895, 2.0998], + device='cuda:0'), covar=tensor([0.2976, 0.2479, 0.1095, 0.1490, 0.3053, 0.1100, 0.2251, 0.2072], + device='cuda:0'), in_proj_covar=tensor([0.0721, 0.0725, 0.0605, 0.0852, 0.0730, 0.0630, 0.0746, 0.0656], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:30:05,034 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48415.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:30:26,079 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 11:30:31,032 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48434.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:30:39,383 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48441.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:30:44,669 INFO [train.py:903] (0/4) Epoch 8, batch 650, loss[loss=0.2626, simple_loss=0.3395, pruned_loss=0.09281, over 19797.00 frames. ], tot_loss[loss=0.2553, simple_loss=0.3229, pruned_loss=0.09389, over 3689233.15 frames. ], batch size: 56, lr: 1.07e-02, grad_scale: 16.0 +2023-04-01 11:30:50,371 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.061e+02 6.183e+02 7.491e+02 9.829e+02 2.830e+03, threshold=1.498e+03, percent-clipped=3.0 +2023-04-01 11:31:17,303 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48471.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:31:45,468 INFO [train.py:903] (0/4) Epoch 8, batch 700, loss[loss=0.2844, simple_loss=0.3604, pruned_loss=0.1042, over 19492.00 frames. ], tot_loss[loss=0.2556, simple_loss=0.3234, pruned_loss=0.09389, over 3720679.03 frames. ], batch size: 64, lr: 1.07e-02, grad_scale: 16.0 +2023-04-01 11:31:45,812 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48496.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:32:04,038 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2510, 1.2391, 1.4399, 1.3467, 1.6974, 1.8039, 1.8365, 0.5300], + device='cuda:0'), covar=tensor([0.1807, 0.3114, 0.1861, 0.1502, 0.1240, 0.1671, 0.1139, 0.3013], + device='cuda:0'), in_proj_covar=tensor([0.0463, 0.0540, 0.0535, 0.0421, 0.0577, 0.0469, 0.0635, 0.0466], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:32:17,584 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48521.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:32:44,030 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48542.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 11:32:48,881 INFO [train.py:903] (0/4) Epoch 8, batch 750, loss[loss=0.2744, simple_loss=0.3338, pruned_loss=0.1075, over 19843.00 frames. ], tot_loss[loss=0.2592, simple_loss=0.3265, pruned_loss=0.09592, over 3737391.59 frames. ], batch size: 52, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:32:49,288 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48546.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:32:52,552 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48549.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:32:55,669 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.581e+02 6.054e+02 7.636e+02 9.380e+02 1.990e+03, threshold=1.527e+03, percent-clipped=3.0 +2023-04-01 11:33:10,158 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6900, 1.3310, 1.3122, 1.8072, 1.3553, 1.8194, 1.8563, 1.6114], + device='cuda:0'), covar=tensor([0.0774, 0.1023, 0.1069, 0.0865, 0.0905, 0.0759, 0.0821, 0.0706], + device='cuda:0'), in_proj_covar=tensor([0.0219, 0.0234, 0.0231, 0.0261, 0.0248, 0.0218, 0.0211, 0.0208], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 11:33:13,733 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48567.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 11:33:17,070 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3534, 1.1767, 1.3047, 1.6079, 2.9130, 0.9244, 2.0578, 3.1524], + device='cuda:0'), covar=tensor([0.0425, 0.2731, 0.2752, 0.1518, 0.0685, 0.2528, 0.1086, 0.0355], + device='cuda:0'), in_proj_covar=tensor([0.0314, 0.0321, 0.0328, 0.0301, 0.0325, 0.0321, 0.0298, 0.0321], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 11:33:21,435 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48574.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 11:33:49,547 INFO [train.py:903] (0/4) Epoch 8, batch 800, loss[loss=0.3181, simple_loss=0.3648, pruned_loss=0.1357, over 19482.00 frames. ], tot_loss[loss=0.2581, simple_loss=0.3252, pruned_loss=0.09553, over 3753021.61 frames. ], batch size: 64, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:34:02,654 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 11:34:19,554 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8945, 1.9803, 1.9505, 2.7610, 1.8479, 2.3572, 2.3780, 1.8783], + device='cuda:0'), covar=tensor([0.2596, 0.2061, 0.1080, 0.1182, 0.2400, 0.1035, 0.2392, 0.2024], + device='cuda:0'), in_proj_covar=tensor([0.0722, 0.0727, 0.0609, 0.0856, 0.0731, 0.0634, 0.0755, 0.0661], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:34:29,987 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2204, 1.3974, 1.7611, 1.4788, 2.7133, 2.2744, 2.8502, 1.1658], + device='cuda:0'), covar=tensor([0.2126, 0.3488, 0.2053, 0.1615, 0.1348, 0.1753, 0.1415, 0.3305], + device='cuda:0'), in_proj_covar=tensor([0.0457, 0.0534, 0.0530, 0.0417, 0.0574, 0.0468, 0.0628, 0.0460], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:34:34,562 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4390, 1.7428, 2.1718, 2.6067, 1.9255, 2.8630, 2.6741, 2.4081], + device='cuda:0'), covar=tensor([0.0676, 0.0968, 0.0916, 0.1023, 0.1011, 0.0578, 0.0784, 0.0629], + device='cuda:0'), in_proj_covar=tensor([0.0219, 0.0232, 0.0230, 0.0261, 0.0248, 0.0215, 0.0209, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 11:34:48,687 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.50 vs. limit=2.0 +2023-04-01 11:34:51,085 INFO [train.py:903] (0/4) Epoch 8, batch 850, loss[loss=0.2625, simple_loss=0.3366, pruned_loss=0.09419, over 19615.00 frames. ], tot_loss[loss=0.2564, simple_loss=0.3244, pruned_loss=0.09423, over 3788448.76 frames. ], batch size: 57, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:34:57,942 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.799e+02 5.922e+02 7.936e+02 9.993e+02 1.897e+03, threshold=1.587e+03, percent-clipped=5.0 +2023-04-01 11:35:39,764 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 11:35:39,886 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48686.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:35:43,325 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48689.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 11:35:50,882 INFO [train.py:903] (0/4) Epoch 8, batch 900, loss[loss=0.2524, simple_loss=0.3233, pruned_loss=0.09076, over 19684.00 frames. ], tot_loss[loss=0.2568, simple_loss=0.3245, pruned_loss=0.09455, over 3798392.15 frames. ], batch size: 59, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:36:54,671 INFO [train.py:903] (0/4) Epoch 8, batch 950, loss[loss=0.1965, simple_loss=0.268, pruned_loss=0.06253, over 19325.00 frames. ], tot_loss[loss=0.2552, simple_loss=0.323, pruned_loss=0.09369, over 3798843.27 frames. ], batch size: 44, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:36:56,564 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 11:36:57,657 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9003, 3.5084, 2.3382, 3.3094, 0.8886, 3.1806, 3.2912, 3.3406], + device='cuda:0'), covar=tensor([0.0929, 0.1399, 0.2242, 0.0876, 0.4388, 0.1017, 0.0963, 0.1185], + device='cuda:0'), in_proj_covar=tensor([0.0395, 0.0342, 0.0401, 0.0301, 0.0366, 0.0329, 0.0318, 0.0352], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 11:37:03,057 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.996e+02 5.931e+02 7.048e+02 8.289e+02 1.665e+03, threshold=1.410e+03, percent-clipped=1.0 +2023-04-01 11:37:11,345 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48759.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:37:42,371 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=48785.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:37:57,225 INFO [train.py:903] (0/4) Epoch 8, batch 1000, loss[loss=0.3041, simple_loss=0.3631, pruned_loss=0.1226, over 19590.00 frames. ], tot_loss[loss=0.2549, simple_loss=0.3229, pruned_loss=0.09351, over 3796899.09 frames. ], batch size: 61, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:38:02,369 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8384, 2.6869, 1.7131, 1.8827, 1.4795, 1.9260, 0.7719, 1.8848], + device='cuda:0'), covar=tensor([0.0472, 0.0487, 0.0573, 0.0789, 0.1098, 0.0846, 0.0911, 0.0925], + device='cuda:0'), in_proj_covar=tensor([0.0317, 0.0318, 0.0317, 0.0334, 0.0409, 0.0329, 0.0296, 0.0316], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 11:38:03,518 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48801.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:38:04,675 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3897, 1.2164, 1.1656, 1.4744, 1.3014, 1.3057, 1.1953, 1.3904], + device='cuda:0'), covar=tensor([0.0713, 0.1077, 0.0924, 0.0571, 0.0834, 0.0400, 0.0865, 0.0511], + device='cuda:0'), in_proj_covar=tensor([0.0242, 0.0353, 0.0280, 0.0236, 0.0299, 0.0238, 0.0270, 0.0228], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 11:38:08,245 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48805.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:38:16,351 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1863, 1.3219, 1.7377, 1.4245, 2.6247, 2.1202, 2.7875, 1.0238], + device='cuda:0'), covar=tensor([0.2092, 0.3599, 0.2034, 0.1709, 0.1299, 0.1788, 0.1403, 0.3383], + device='cuda:0'), in_proj_covar=tensor([0.0454, 0.0530, 0.0528, 0.0417, 0.0570, 0.0463, 0.0625, 0.0458], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:38:40,109 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48830.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:38:50,179 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 11:38:51,451 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48839.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:38:59,281 INFO [train.py:903] (0/4) Epoch 8, batch 1050, loss[loss=0.2166, simple_loss=0.2765, pruned_loss=0.07839, over 19726.00 frames. ], tot_loss[loss=0.2559, simple_loss=0.3237, pruned_loss=0.09411, over 3810327.51 frames. ], batch size: 45, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:39:06,227 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.557e+02 5.500e+02 6.561e+02 8.180e+02 1.521e+03, threshold=1.312e+03, percent-clipped=1.0 +2023-04-01 11:39:11,093 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=48856.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:39:31,750 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 11:39:33,597 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-04-01 11:39:34,233 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48874.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:39:58,991 INFO [train.py:903] (0/4) Epoch 8, batch 1100, loss[loss=0.2363, simple_loss=0.3148, pruned_loss=0.07885, over 19761.00 frames. ], tot_loss[loss=0.2561, simple_loss=0.3239, pruned_loss=0.09417, over 3815582.05 frames. ], batch size: 54, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:40:04,016 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=48900.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:41:00,104 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=48945.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 11:41:00,823 INFO [train.py:903] (0/4) Epoch 8, batch 1150, loss[loss=0.2628, simple_loss=0.3351, pruned_loss=0.09522, over 19124.00 frames. ], tot_loss[loss=0.2551, simple_loss=0.3228, pruned_loss=0.09368, over 3794413.60 frames. ], batch size: 69, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:41:09,121 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.525e+02 5.943e+02 6.952e+02 8.882e+02 1.618e+03, threshold=1.390e+03, percent-clipped=5.0 +2023-04-01 11:41:20,470 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8882, 4.3073, 4.6122, 4.5677, 1.7097, 4.2367, 3.6415, 4.1817], + device='cuda:0'), covar=tensor([0.1256, 0.0686, 0.0532, 0.0502, 0.4705, 0.0546, 0.0589, 0.0998], + device='cuda:0'), in_proj_covar=tensor([0.0593, 0.0519, 0.0705, 0.0593, 0.0656, 0.0448, 0.0449, 0.0654], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 11:41:30,935 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=48970.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 11:42:04,106 INFO [train.py:903] (0/4) Epoch 8, batch 1200, loss[loss=0.2683, simple_loss=0.3397, pruned_loss=0.09844, over 19281.00 frames. ], tot_loss[loss=0.2548, simple_loss=0.3224, pruned_loss=0.09353, over 3795196.09 frames. ], batch size: 70, lr: 1.07e-02, grad_scale: 8.0 +2023-04-01 11:42:32,694 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 11:43:05,011 INFO [train.py:903] (0/4) Epoch 8, batch 1250, loss[loss=0.2889, simple_loss=0.353, pruned_loss=0.1124, over 18261.00 frames. ], tot_loss[loss=0.255, simple_loss=0.3231, pruned_loss=0.09349, over 3816939.23 frames. ], batch size: 83, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:43:11,736 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.529e+02 6.811e+02 8.521e+02 1.008e+03 2.064e+03, threshold=1.704e+03, percent-clipped=4.0 +2023-04-01 11:43:17,993 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49057.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:43:50,278 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49082.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:44:06,032 INFO [train.py:903] (0/4) Epoch 8, batch 1300, loss[loss=0.2725, simple_loss=0.346, pruned_loss=0.09953, over 19607.00 frames. ], tot_loss[loss=0.2541, simple_loss=0.3226, pruned_loss=0.09282, over 3820243.68 frames. ], batch size: 61, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:44:49,025 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49130.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:45:07,324 INFO [train.py:903] (0/4) Epoch 8, batch 1350, loss[loss=0.231, simple_loss=0.3006, pruned_loss=0.0807, over 19416.00 frames. ], tot_loss[loss=0.2551, simple_loss=0.3229, pruned_loss=0.0936, over 3813764.20 frames. ], batch size: 48, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:45:16,537 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.771e+02 5.835e+02 7.092e+02 8.908e+02 2.388e+03, threshold=1.418e+03, percent-clipped=3.0 +2023-04-01 11:45:20,471 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49155.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:45:22,451 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49156.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:45:26,736 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9690, 3.5839, 1.7816, 2.0614, 2.9568, 1.6652, 1.2236, 1.9177], + device='cuda:0'), covar=tensor([0.1115, 0.0403, 0.1048, 0.0696, 0.0470, 0.1015, 0.0954, 0.0675], + device='cuda:0'), in_proj_covar=tensor([0.0283, 0.0292, 0.0318, 0.0245, 0.0233, 0.0311, 0.0286, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 11:45:52,070 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49181.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:45:54,116 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49183.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:46:10,824 INFO [train.py:903] (0/4) Epoch 8, batch 1400, loss[loss=0.2528, simple_loss=0.3159, pruned_loss=0.09483, over 19836.00 frames. ], tot_loss[loss=0.2535, simple_loss=0.3214, pruned_loss=0.09278, over 3832131.90 frames. ], batch size: 52, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:46:17,525 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49200.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:47:01,994 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.95 vs. limit=5.0 +2023-04-01 11:47:12,312 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 11:47:13,437 INFO [train.py:903] (0/4) Epoch 8, batch 1450, loss[loss=0.2417, simple_loss=0.3199, pruned_loss=0.0817, over 19529.00 frames. ], tot_loss[loss=0.2543, simple_loss=0.3223, pruned_loss=0.09319, over 3822123.91 frames. ], batch size: 54, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:47:17,037 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3016, 1.2079, 1.3747, 1.3628, 1.7368, 1.9206, 1.8313, 0.4781], + device='cuda:0'), covar=tensor([0.1851, 0.3314, 0.1959, 0.1615, 0.1220, 0.1637, 0.1165, 0.3226], + device='cuda:0'), in_proj_covar=tensor([0.0458, 0.0532, 0.0530, 0.0415, 0.0574, 0.0463, 0.0628, 0.0462], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:47:19,901 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.194e+02 6.215e+02 8.146e+02 9.729e+02 2.293e+03, threshold=1.629e+03, percent-clipped=2.0 +2023-04-01 11:47:22,550 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49254.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:47:56,333 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5605, 2.4153, 1.5912, 1.5793, 2.1207, 1.2756, 1.3362, 1.7487], + device='cuda:0'), covar=tensor([0.0845, 0.0545, 0.0975, 0.0607, 0.0479, 0.1001, 0.0673, 0.0490], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0290, 0.0317, 0.0242, 0.0230, 0.0310, 0.0283, 0.0251], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 11:48:14,505 INFO [train.py:903] (0/4) Epoch 8, batch 1500, loss[loss=0.24, simple_loss=0.3011, pruned_loss=0.08945, over 19397.00 frames. ], tot_loss[loss=0.2552, simple_loss=0.3232, pruned_loss=0.09363, over 3808271.37 frames. ], batch size: 48, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:48:16,780 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49298.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:48:38,575 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49315.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:49:12,447 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3627, 1.3135, 1.3583, 1.3333, 2.9425, 0.8847, 2.0360, 3.0522], + device='cuda:0'), covar=tensor([0.0429, 0.2632, 0.2583, 0.1743, 0.0642, 0.2589, 0.1185, 0.0403], + device='cuda:0'), in_proj_covar=tensor([0.0314, 0.0323, 0.0330, 0.0301, 0.0328, 0.0320, 0.0301, 0.0324], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 11:49:14,530 INFO [train.py:903] (0/4) Epoch 8, batch 1550, loss[loss=0.2735, simple_loss=0.3438, pruned_loss=0.1015, over 19296.00 frames. ], tot_loss[loss=0.2555, simple_loss=0.3231, pruned_loss=0.09388, over 3808338.54 frames. ], batch size: 66, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:49:23,152 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.230e+02 6.392e+02 7.844e+02 9.464e+02 1.840e+03, threshold=1.569e+03, percent-clipped=1.0 +2023-04-01 11:49:24,615 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1226, 0.9430, 1.0729, 1.4649, 1.1198, 1.2469, 1.3410, 1.1103], + device='cuda:0'), covar=tensor([0.1126, 0.1448, 0.1347, 0.0764, 0.0936, 0.1118, 0.0971, 0.1018], + device='cuda:0'), in_proj_covar=tensor([0.0220, 0.0235, 0.0233, 0.0265, 0.0252, 0.0222, 0.0212, 0.0209], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 11:50:17,464 INFO [train.py:903] (0/4) Epoch 8, batch 1600, loss[loss=0.2454, simple_loss=0.3187, pruned_loss=0.08602, over 19653.00 frames. ], tot_loss[loss=0.2544, simple_loss=0.3223, pruned_loss=0.09323, over 3807148.77 frames. ], batch size: 55, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:50:38,475 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 11:50:49,074 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0840, 0.9799, 1.0719, 1.3299, 0.9968, 1.2740, 1.3447, 1.1732], + device='cuda:0'), covar=tensor([0.1087, 0.1222, 0.1237, 0.0857, 0.0942, 0.0927, 0.0916, 0.0887], + device='cuda:0'), in_proj_covar=tensor([0.0220, 0.0234, 0.0233, 0.0264, 0.0252, 0.0223, 0.0212, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 11:51:12,181 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49440.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:51:20,002 INFO [train.py:903] (0/4) Epoch 8, batch 1650, loss[loss=0.2821, simple_loss=0.348, pruned_loss=0.1081, over 18814.00 frames. ], tot_loss[loss=0.2552, simple_loss=0.3228, pruned_loss=0.09377, over 3806726.42 frames. ], batch size: 74, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:51:24,784 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49450.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 11:51:26,741 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.384e+02 6.230e+02 7.478e+02 9.229e+02 3.510e+03, threshold=1.496e+03, percent-clipped=3.0 +2023-04-01 11:52:16,281 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6497, 4.1625, 4.3918, 4.3954, 1.5816, 4.1349, 3.5642, 3.9753], + device='cuda:0'), covar=tensor([0.1231, 0.0659, 0.0519, 0.0494, 0.4807, 0.0480, 0.0577, 0.1185], + device='cuda:0'), in_proj_covar=tensor([0.0606, 0.0537, 0.0718, 0.0602, 0.0671, 0.0462, 0.0457, 0.0671], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 11:52:21,802 INFO [train.py:903] (0/4) Epoch 8, batch 1700, loss[loss=0.2219, simple_loss=0.2897, pruned_loss=0.07704, over 19769.00 frames. ], tot_loss[loss=0.2542, simple_loss=0.322, pruned_loss=0.09317, over 3813406.26 frames. ], batch size: 47, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:53:02,523 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 11:53:03,215 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-01 11:53:06,461 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4739, 1.4809, 1.9306, 1.6147, 2.6744, 2.5049, 2.7763, 1.3517], + device='cuda:0'), covar=tensor([0.1854, 0.3314, 0.1906, 0.1558, 0.1321, 0.1480, 0.1457, 0.3012], + device='cuda:0'), in_proj_covar=tensor([0.0461, 0.0538, 0.0534, 0.0418, 0.0578, 0.0467, 0.0631, 0.0465], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 11:53:23,395 INFO [train.py:903] (0/4) Epoch 8, batch 1750, loss[loss=0.2861, simple_loss=0.3483, pruned_loss=0.112, over 19662.00 frames. ], tot_loss[loss=0.2541, simple_loss=0.3219, pruned_loss=0.09311, over 3816672.21 frames. ], batch size: 60, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:53:26,506 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-01 11:53:31,457 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.581e+02 6.147e+02 7.390e+02 1.012e+03 1.809e+03, threshold=1.478e+03, percent-clipped=6.0 +2023-04-01 11:53:35,100 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49554.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:53:50,020 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.61 vs. limit=2.0 +2023-04-01 11:53:56,591 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49571.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:53:58,883 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7691, 1.9804, 2.3223, 2.1154, 3.0641, 3.6364, 3.6674, 3.9199], + device='cuda:0'), covar=tensor([0.1226, 0.2268, 0.2094, 0.1688, 0.0681, 0.0358, 0.0167, 0.0158], + device='cuda:0'), in_proj_covar=tensor([0.0249, 0.0282, 0.0311, 0.0245, 0.0205, 0.0135, 0.0203, 0.0169], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 11:54:06,005 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49579.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:54:27,936 INFO [train.py:903] (0/4) Epoch 8, batch 1800, loss[loss=0.3099, simple_loss=0.368, pruned_loss=0.1259, over 17390.00 frames. ], tot_loss[loss=0.2538, simple_loss=0.3214, pruned_loss=0.09303, over 3821521.18 frames. ], batch size: 101, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:54:28,351 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49596.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:54:31,405 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49598.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:55:25,420 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 11:55:30,998 INFO [train.py:903] (0/4) Epoch 8, batch 1850, loss[loss=0.3566, simple_loss=0.3963, pruned_loss=0.1585, over 12901.00 frames. ], tot_loss[loss=0.2552, simple_loss=0.3225, pruned_loss=0.09399, over 3822947.29 frames. ], batch size: 136, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:55:38,000 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.438e+02 5.845e+02 7.519e+02 8.649e+02 2.522e+03, threshold=1.504e+03, percent-clipped=4.0 +2023-04-01 11:56:02,470 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 11:56:31,214 INFO [train.py:903] (0/4) Epoch 8, batch 1900, loss[loss=0.2664, simple_loss=0.3415, pruned_loss=0.09567, over 19543.00 frames. ], tot_loss[loss=0.2565, simple_loss=0.3236, pruned_loss=0.09472, over 3829276.81 frames. ], batch size: 56, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:56:48,649 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 11:56:52,377 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49713.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:56:54,379 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 11:57:18,982 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 11:57:33,019 INFO [train.py:903] (0/4) Epoch 8, batch 1950, loss[loss=0.2406, simple_loss=0.3218, pruned_loss=0.07972, over 19542.00 frames. ], tot_loss[loss=0.256, simple_loss=0.3235, pruned_loss=0.09429, over 3823350.01 frames. ], batch size: 54, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:57:40,110 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.492e+02 5.402e+02 6.624e+02 8.916e+02 2.925e+03, threshold=1.325e+03, percent-clipped=4.0 +2023-04-01 11:58:20,344 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49784.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:58:20,492 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=49784.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 11:58:26,588 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2485, 1.6154, 2.1675, 2.8779, 1.8641, 2.2389, 2.6203, 2.2988], + device='cuda:0'), covar=tensor([0.0763, 0.0984, 0.0855, 0.0832, 0.1000, 0.0757, 0.0886, 0.0620], + device='cuda:0'), in_proj_covar=tensor([0.0216, 0.0231, 0.0229, 0.0261, 0.0248, 0.0217, 0.0210, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 11:58:32,981 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=49794.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 11:58:35,922 INFO [train.py:903] (0/4) Epoch 8, batch 2000, loss[loss=0.2441, simple_loss=0.3247, pruned_loss=0.08178, over 19788.00 frames. ], tot_loss[loss=0.2562, simple_loss=0.3238, pruned_loss=0.09426, over 3823424.87 frames. ], batch size: 56, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:59:35,609 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 11:59:38,067 INFO [train.py:903] (0/4) Epoch 8, batch 2050, loss[loss=0.1924, simple_loss=0.2778, pruned_loss=0.05356, over 19576.00 frames. ], tot_loss[loss=0.2547, simple_loss=0.322, pruned_loss=0.09373, over 3820674.95 frames. ], batch size: 52, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 11:59:45,925 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.480e+02 5.411e+02 7.156e+02 9.098e+02 3.444e+03, threshold=1.431e+03, percent-clipped=9.0 +2023-04-01 11:59:53,811 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 11:59:55,135 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 12:00:17,460 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 12:00:39,973 INFO [train.py:903] (0/4) Epoch 8, batch 2100, loss[loss=0.2579, simple_loss=0.3288, pruned_loss=0.09344, over 19770.00 frames. ], tot_loss[loss=0.2551, simple_loss=0.3224, pruned_loss=0.09386, over 3827346.29 frames. ], batch size: 56, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 12:00:43,806 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49899.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:00:55,217 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=49909.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 12:01:09,476 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 12:01:32,556 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 12:01:41,833 INFO [train.py:903] (0/4) Epoch 8, batch 2150, loss[loss=0.3029, simple_loss=0.364, pruned_loss=0.1209, over 19425.00 frames. ], tot_loss[loss=0.2541, simple_loss=0.3213, pruned_loss=0.09348, over 3819410.12 frames. ], batch size: 70, lr: 1.06e-02, grad_scale: 8.0 +2023-04-01 12:01:48,292 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.004e+02 5.997e+02 7.086e+02 8.659e+02 2.224e+03, threshold=1.417e+03, percent-clipped=8.0 +2023-04-01 12:02:03,058 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7751, 1.8887, 1.9971, 2.7504, 1.8151, 2.5972, 2.4897, 1.8896], + device='cuda:0'), covar=tensor([0.2958, 0.2374, 0.1150, 0.1279, 0.2564, 0.0981, 0.2411, 0.2195], + device='cuda:0'), in_proj_covar=tensor([0.0722, 0.0732, 0.0610, 0.0856, 0.0728, 0.0633, 0.0756, 0.0655], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 12:02:05,611 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-01 12:02:12,033 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=49969.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:02:40,672 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=49994.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:02:43,585 INFO [train.py:903] (0/4) Epoch 8, batch 2200, loss[loss=0.2399, simple_loss=0.3066, pruned_loss=0.08658, over 19854.00 frames. ], tot_loss[loss=0.2546, simple_loss=0.3214, pruned_loss=0.09391, over 3815483.83 frames. ], batch size: 52, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:02:49,452 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-50000.pt +2023-04-01 12:03:48,609 INFO [train.py:903] (0/4) Epoch 8, batch 2250, loss[loss=0.245, simple_loss=0.3113, pruned_loss=0.08937, over 19475.00 frames. ], tot_loss[loss=0.2553, simple_loss=0.3224, pruned_loss=0.09414, over 3809943.17 frames. ], batch size: 49, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:03:55,476 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.410e+02 6.238e+02 7.806e+02 1.014e+03 2.092e+03, threshold=1.561e+03, percent-clipped=8.0 +2023-04-01 12:04:51,015 INFO [train.py:903] (0/4) Epoch 8, batch 2300, loss[loss=0.2523, simple_loss=0.3094, pruned_loss=0.09756, over 16052.00 frames. ], tot_loss[loss=0.2573, simple_loss=0.3242, pruned_loss=0.09521, over 3810612.82 frames. ], batch size: 35, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:05:04,996 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 12:05:31,266 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=50128.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:05:52,929 INFO [train.py:903] (0/4) Epoch 8, batch 2350, loss[loss=0.2654, simple_loss=0.3351, pruned_loss=0.09786, over 19531.00 frames. ], tot_loss[loss=0.2577, simple_loss=0.3248, pruned_loss=0.09528, over 3824001.65 frames. ], batch size: 54, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:05:53,240 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50146.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:05:59,870 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.785e+02 5.938e+02 7.771e+02 9.106e+02 1.869e+03, threshold=1.554e+03, percent-clipped=2.0 +2023-04-01 12:06:03,818 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50155.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:06:18,096 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50165.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 12:06:31,230 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-01 12:06:36,669 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50180.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:06:38,683 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 12:06:47,995 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50190.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:06:54,683 INFO [train.py:903] (0/4) Epoch 8, batch 2400, loss[loss=0.2611, simple_loss=0.328, pruned_loss=0.0971, over 19598.00 frames. ], tot_loss[loss=0.2573, simple_loss=0.3241, pruned_loss=0.09526, over 3823104.74 frames. ], batch size: 57, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:06:54,692 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 12:07:20,904 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50215.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 12:07:53,620 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50243.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:07:58,090 INFO [train.py:903] (0/4) Epoch 8, batch 2450, loss[loss=0.3001, simple_loss=0.3659, pruned_loss=0.1172, over 19307.00 frames. ], tot_loss[loss=0.2584, simple_loss=0.3249, pruned_loss=0.09596, over 3796436.39 frames. ], batch size: 66, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:08:05,247 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.846e+02 6.354e+02 7.339e+02 9.160e+02 2.255e+03, threshold=1.468e+03, percent-clipped=3.0 +2023-04-01 12:09:00,867 INFO [train.py:903] (0/4) Epoch 8, batch 2500, loss[loss=0.3442, simple_loss=0.3805, pruned_loss=0.154, over 13582.00 frames. ], tot_loss[loss=0.258, simple_loss=0.325, pruned_loss=0.09551, over 3771460.92 frames. ], batch size: 136, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:10:03,030 INFO [train.py:903] (0/4) Epoch 8, batch 2550, loss[loss=0.2742, simple_loss=0.3452, pruned_loss=0.1016, over 19308.00 frames. ], tot_loss[loss=0.256, simple_loss=0.3239, pruned_loss=0.09405, over 3795804.26 frames. ], batch size: 66, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:10:09,527 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.329e+02 5.479e+02 6.803e+02 8.076e+02 1.672e+03, threshold=1.361e+03, percent-clipped=2.0 +2023-04-01 12:10:59,275 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 12:11:05,094 INFO [train.py:903] (0/4) Epoch 8, batch 2600, loss[loss=0.2451, simple_loss=0.3005, pruned_loss=0.09483, over 17389.00 frames. ], tot_loss[loss=0.2551, simple_loss=0.3228, pruned_loss=0.09368, over 3807753.41 frames. ], batch size: 38, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:12:09,324 INFO [train.py:903] (0/4) Epoch 8, batch 2650, loss[loss=0.2902, simple_loss=0.356, pruned_loss=0.1122, over 19789.00 frames. ], tot_loss[loss=0.255, simple_loss=0.323, pruned_loss=0.09356, over 3814469.00 frames. ], batch size: 56, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:12:15,976 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.098e+02 6.836e+02 8.198e+02 1.046e+03 1.620e+03, threshold=1.640e+03, percent-clipped=8.0 +2023-04-01 12:12:27,569 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 12:12:34,601 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5904, 4.0299, 4.2218, 4.2098, 1.5506, 3.9275, 3.4424, 3.8762], + device='cuda:0'), covar=tensor([0.1107, 0.0607, 0.0500, 0.0491, 0.4634, 0.0523, 0.0602, 0.0992], + device='cuda:0'), in_proj_covar=tensor([0.0592, 0.0524, 0.0707, 0.0597, 0.0661, 0.0454, 0.0445, 0.0657], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 12:13:04,636 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=50490.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:13:09,322 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0384, 5.3595, 2.8515, 4.5727, 1.0759, 5.3231, 5.2825, 5.5728], + device='cuda:0'), covar=tensor([0.0417, 0.0923, 0.1946, 0.0611, 0.4099, 0.0517, 0.0571, 0.0819], + device='cuda:0'), in_proj_covar=tensor([0.0389, 0.0338, 0.0401, 0.0293, 0.0361, 0.0322, 0.0313, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 12:13:11,330 INFO [train.py:903] (0/4) Epoch 8, batch 2700, loss[loss=0.2632, simple_loss=0.3145, pruned_loss=0.1059, over 14693.00 frames. ], tot_loss[loss=0.2542, simple_loss=0.322, pruned_loss=0.0932, over 3814877.03 frames. ], batch size: 32, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:13:16,508 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50499.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:13:46,656 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50524.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:14:15,300 INFO [train.py:903] (0/4) Epoch 8, batch 2750, loss[loss=0.2798, simple_loss=0.3463, pruned_loss=0.1066, over 19652.00 frames. ], tot_loss[loss=0.2539, simple_loss=0.322, pruned_loss=0.09292, over 3830869.18 frames. ], batch size: 55, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:14:20,713 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4837, 1.5298, 1.6174, 1.9786, 1.3537, 1.7122, 1.8825, 1.6042], + device='cuda:0'), covar=tensor([0.2505, 0.2032, 0.1065, 0.1057, 0.2208, 0.1004, 0.2399, 0.1917], + device='cuda:0'), in_proj_covar=tensor([0.0731, 0.0738, 0.0617, 0.0859, 0.0740, 0.0642, 0.0768, 0.0662], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 12:14:23,786 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.883e+02 5.912e+02 7.181e+02 9.047e+02 1.864e+03, threshold=1.436e+03, percent-clipped=1.0 +2023-04-01 12:14:29,751 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3194, 1.6380, 2.0255, 2.5035, 1.8829, 2.9105, 2.6727, 2.3312], + device='cuda:0'), covar=tensor([0.0662, 0.0939, 0.0908, 0.0998, 0.0967, 0.0572, 0.0800, 0.0639], + device='cuda:0'), in_proj_covar=tensor([0.0216, 0.0233, 0.0231, 0.0260, 0.0247, 0.0218, 0.0210, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 12:14:30,887 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=50559.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:14:37,732 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50564.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:15:18,012 INFO [train.py:903] (0/4) Epoch 8, batch 2800, loss[loss=0.3304, simple_loss=0.3659, pruned_loss=0.1474, over 13579.00 frames. ], tot_loss[loss=0.2555, simple_loss=0.3232, pruned_loss=0.09389, over 3805480.39 frames. ], batch size: 135, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:15:29,724 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50605.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:16:21,990 INFO [train.py:903] (0/4) Epoch 8, batch 2850, loss[loss=0.2282, simple_loss=0.291, pruned_loss=0.08274, over 19751.00 frames. ], tot_loss[loss=0.2556, simple_loss=0.3232, pruned_loss=0.09394, over 3813513.95 frames. ], batch size: 47, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:16:31,181 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.810e+02 5.617e+02 7.073e+02 8.787e+02 1.544e+03, threshold=1.415e+03, percent-clipped=2.0 +2023-04-01 12:16:57,575 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=50674.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:17:25,989 INFO [train.py:903] (0/4) Epoch 8, batch 2900, loss[loss=0.2277, simple_loss=0.2864, pruned_loss=0.08447, over 19287.00 frames. ], tot_loss[loss=0.2538, simple_loss=0.3218, pruned_loss=0.09291, over 3825853.60 frames. ], batch size: 44, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:17:26,079 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 12:18:29,372 INFO [train.py:903] (0/4) Epoch 8, batch 2950, loss[loss=0.247, simple_loss=0.316, pruned_loss=0.08903, over 19864.00 frames. ], tot_loss[loss=0.2528, simple_loss=0.3213, pruned_loss=0.09214, over 3837620.02 frames. ], batch size: 52, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:18:37,517 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.796e+02 6.074e+02 7.951e+02 1.027e+03 2.467e+03, threshold=1.590e+03, percent-clipped=7.0 +2023-04-01 12:19:31,424 INFO [train.py:903] (0/4) Epoch 8, batch 3000, loss[loss=0.3076, simple_loss=0.3625, pruned_loss=0.1263, over 19272.00 frames. ], tot_loss[loss=0.2548, simple_loss=0.3228, pruned_loss=0.09342, over 3834783.41 frames. ], batch size: 66, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:19:31,425 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 12:19:38,491 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1338, 1.2677, 1.5660, 0.9745, 2.2626, 2.8823, 2.6386, 3.0138], + device='cuda:0'), covar=tensor([0.1579, 0.3070, 0.2863, 0.2203, 0.0484, 0.0244, 0.0287, 0.0205], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0287, 0.0316, 0.0249, 0.0209, 0.0137, 0.0207, 0.0174], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 12:19:44,060 INFO [train.py:937] (0/4) Epoch 8, validation: loss=0.1875, simple_loss=0.2879, pruned_loss=0.04358, over 944034.00 frames. +2023-04-01 12:19:44,061 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 12:19:46,424 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 12:19:49,186 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50800.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:20:21,695 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50826.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 12:20:45,735 INFO [train.py:903] (0/4) Epoch 8, batch 3050, loss[loss=0.286, simple_loss=0.346, pruned_loss=0.113, over 17624.00 frames. ], tot_loss[loss=0.2555, simple_loss=0.3236, pruned_loss=0.09374, over 3830890.95 frames. ], batch size: 101, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:20:55,183 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.189e+02 5.734e+02 7.199e+02 9.163e+02 1.650e+03, threshold=1.440e+03, percent-clipped=2.0 +2023-04-01 12:21:06,927 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50861.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:21:22,874 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2141, 1.1662, 1.4937, 1.3795, 2.0964, 1.9193, 2.1007, 0.8601], + device='cuda:0'), covar=tensor([0.2404, 0.4147, 0.2362, 0.2030, 0.1458, 0.2048, 0.1565, 0.3620], + device='cuda:0'), in_proj_covar=tensor([0.0467, 0.0540, 0.0537, 0.0422, 0.0576, 0.0471, 0.0637, 0.0468], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 12:21:24,278 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.76 vs. limit=2.0 +2023-04-01 12:21:36,759 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50886.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:21:44,445 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2653, 1.3631, 1.4056, 1.5206, 2.7764, 0.8505, 2.2081, 3.0608], + device='cuda:0'), covar=tensor([0.0457, 0.2368, 0.2608, 0.1465, 0.0752, 0.2370, 0.1098, 0.0395], + device='cuda:0'), in_proj_covar=tensor([0.0321, 0.0326, 0.0340, 0.0302, 0.0334, 0.0320, 0.0308, 0.0328], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 12:21:49,959 INFO [train.py:903] (0/4) Epoch 8, batch 3100, loss[loss=0.2044, simple_loss=0.2836, pruned_loss=0.0626, over 19648.00 frames. ], tot_loss[loss=0.2558, simple_loss=0.3236, pruned_loss=0.09405, over 3818211.99 frames. ], batch size: 55, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:22:04,733 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=50908.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:22:32,054 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=50930.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:22:52,354 INFO [train.py:903] (0/4) Epoch 8, batch 3150, loss[loss=0.2501, simple_loss=0.3291, pruned_loss=0.08553, over 18462.00 frames. ], tot_loss[loss=0.2556, simple_loss=0.3234, pruned_loss=0.09392, over 3815959.08 frames. ], batch size: 84, lr: 1.05e-02, grad_scale: 8.0 +2023-04-01 12:23:00,494 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.154e+02 5.915e+02 7.023e+02 8.955e+02 1.571e+03, threshold=1.405e+03, percent-clipped=4.0 +2023-04-01 12:23:03,163 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=50955.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 12:23:14,706 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=50964.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:23:20,378 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 12:23:32,483 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.39 vs. limit=5.0 +2023-04-01 12:23:35,052 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2996, 2.9797, 2.0954, 2.7824, 0.9532, 2.8117, 2.7671, 2.8543], + device='cuda:0'), covar=tensor([0.1069, 0.1446, 0.2094, 0.0836, 0.3609, 0.1106, 0.0942, 0.1235], + device='cuda:0'), in_proj_covar=tensor([0.0397, 0.0345, 0.0402, 0.0294, 0.0365, 0.0328, 0.0323, 0.0359], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 12:23:54,314 INFO [train.py:903] (0/4) Epoch 8, batch 3200, loss[loss=0.2546, simple_loss=0.3262, pruned_loss=0.09146, over 19403.00 frames. ], tot_loss[loss=0.2554, simple_loss=0.3232, pruned_loss=0.0938, over 3827178.92 frames. ], batch size: 48, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:24:30,128 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51023.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:24:57,559 INFO [train.py:903] (0/4) Epoch 8, batch 3250, loss[loss=0.2321, simple_loss=0.3076, pruned_loss=0.07827, over 19617.00 frames. ], tot_loss[loss=0.2541, simple_loss=0.3222, pruned_loss=0.09299, over 3831501.40 frames. ], batch size: 50, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:25:03,797 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7347, 1.4924, 1.6295, 1.6922, 3.2800, 1.0824, 2.3107, 3.5924], + device='cuda:0'), covar=tensor([0.0423, 0.2416, 0.2410, 0.1567, 0.0697, 0.2390, 0.1157, 0.0271], + device='cuda:0'), in_proj_covar=tensor([0.0322, 0.0322, 0.0335, 0.0301, 0.0329, 0.0319, 0.0307, 0.0325], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 12:25:05,780 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.083e+02 6.087e+02 7.883e+02 9.942e+02 3.174e+03, threshold=1.577e+03, percent-clipped=7.0 +2023-04-01 12:25:40,489 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51080.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:26:00,812 INFO [train.py:903] (0/4) Epoch 8, batch 3300, loss[loss=0.3066, simple_loss=0.3557, pruned_loss=0.1288, over 13878.00 frames. ], tot_loss[loss=0.2544, simple_loss=0.3223, pruned_loss=0.09329, over 3821833.78 frames. ], batch size: 135, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:26:01,209 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6180, 1.6854, 1.4275, 1.3344, 1.2090, 1.4368, 0.3970, 0.9040], + device='cuda:0'), covar=tensor([0.0287, 0.0304, 0.0226, 0.0289, 0.0580, 0.0328, 0.0597, 0.0497], + device='cuda:0'), in_proj_covar=tensor([0.0323, 0.0320, 0.0320, 0.0336, 0.0413, 0.0336, 0.0297, 0.0317], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 12:26:08,900 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 12:26:17,151 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6990, 1.1079, 1.3344, 1.5275, 2.9610, 0.8619, 2.2607, 3.3566], + device='cuda:0'), covar=tensor([0.0521, 0.3562, 0.3306, 0.1892, 0.1090, 0.3025, 0.1329, 0.0516], + device='cuda:0'), in_proj_covar=tensor([0.0322, 0.0324, 0.0335, 0.0301, 0.0331, 0.0321, 0.0307, 0.0326], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 12:27:02,456 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51144.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:27:04,793 INFO [train.py:903] (0/4) Epoch 8, batch 3350, loss[loss=0.2615, simple_loss=0.3131, pruned_loss=0.105, over 19289.00 frames. ], tot_loss[loss=0.2548, simple_loss=0.3224, pruned_loss=0.09367, over 3809564.79 frames. ], batch size: 44, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:27:12,715 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.478e+02 5.942e+02 7.377e+02 9.279e+02 2.136e+03, threshold=1.475e+03, percent-clipped=2.0 +2023-04-01 12:27:34,023 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51170.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:28:00,636 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51190.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:28:07,181 INFO [train.py:903] (0/4) Epoch 8, batch 3400, loss[loss=0.252, simple_loss=0.3275, pruned_loss=0.08826, over 19781.00 frames. ], tot_loss[loss=0.2558, simple_loss=0.3234, pruned_loss=0.09411, over 3810066.47 frames. ], batch size: 56, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:29:10,666 INFO [train.py:903] (0/4) Epoch 8, batch 3450, loss[loss=0.2108, simple_loss=0.2816, pruned_loss=0.07001, over 19855.00 frames. ], tot_loss[loss=0.2556, simple_loss=0.3227, pruned_loss=0.09419, over 3801831.88 frames. ], batch size: 52, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:29:16,207 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 12:29:16,430 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0014, 5.3962, 2.8562, 4.6654, 1.5852, 5.2869, 5.2587, 5.3854], + device='cuda:0'), covar=tensor([0.0370, 0.0791, 0.1801, 0.0586, 0.3424, 0.0521, 0.0564, 0.0680], + device='cuda:0'), in_proj_covar=tensor([0.0399, 0.0346, 0.0405, 0.0300, 0.0373, 0.0330, 0.0323, 0.0361], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 12:29:18,575 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.243e+02 6.280e+02 7.477e+02 9.686e+02 1.820e+03, threshold=1.495e+03, percent-clipped=3.0 +2023-04-01 12:29:27,081 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51259.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:29:52,057 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51279.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:30:00,073 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51285.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:30:12,716 INFO [train.py:903] (0/4) Epoch 8, batch 3500, loss[loss=0.2763, simple_loss=0.3242, pruned_loss=0.1142, over 18970.00 frames. ], tot_loss[loss=0.2568, simple_loss=0.3238, pruned_loss=0.09484, over 3810367.64 frames. ], batch size: 42, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:30:25,469 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51304.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:30:26,781 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8631, 1.3501, 1.0113, 0.9786, 1.1594, 0.8688, 0.8269, 1.2539], + device='cuda:0'), covar=tensor([0.0472, 0.0695, 0.0916, 0.0509, 0.0436, 0.1071, 0.0566, 0.0361], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0291, 0.0314, 0.0242, 0.0230, 0.0309, 0.0283, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 12:30:30,172 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51308.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:31:19,058 INFO [train.py:903] (0/4) Epoch 8, batch 3550, loss[loss=0.3049, simple_loss=0.3658, pruned_loss=0.122, over 19137.00 frames. ], tot_loss[loss=0.2557, simple_loss=0.323, pruned_loss=0.09415, over 3817276.46 frames. ], batch size: 69, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:31:27,385 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.001e+02 5.209e+02 6.396e+02 8.561e+02 1.899e+03, threshold=1.279e+03, percent-clipped=3.0 +2023-04-01 12:32:00,660 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.38 vs. limit=5.0 +2023-04-01 12:32:21,070 INFO [train.py:903] (0/4) Epoch 8, batch 3600, loss[loss=0.2513, simple_loss=0.3214, pruned_loss=0.09064, over 19765.00 frames. ], tot_loss[loss=0.2549, simple_loss=0.3224, pruned_loss=0.09369, over 3830063.64 frames. ], batch size: 54, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:32:26,101 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51400.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:32:38,763 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51410.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:32:54,976 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51423.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:32:56,810 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51424.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:33:16,720 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0658, 2.2885, 2.4197, 2.4235, 1.0859, 2.2249, 2.0590, 2.2131], + device='cuda:0'), covar=tensor([0.1180, 0.1872, 0.0659, 0.0646, 0.3329, 0.0936, 0.0621, 0.1023], + device='cuda:0'), in_proj_covar=tensor([0.0598, 0.0527, 0.0711, 0.0595, 0.0659, 0.0460, 0.0450, 0.0655], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 12:33:22,342 INFO [train.py:903] (0/4) Epoch 8, batch 3650, loss[loss=0.22, simple_loss=0.2913, pruned_loss=0.07437, over 19842.00 frames. ], tot_loss[loss=0.2564, simple_loss=0.3239, pruned_loss=0.09439, over 3828372.37 frames. ], batch size: 52, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:33:27,496 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8842, 2.0119, 2.0222, 3.0091, 1.9314, 2.7124, 2.4732, 1.9289], + device='cuda:0'), covar=tensor([0.2939, 0.2206, 0.1118, 0.1228, 0.2603, 0.1026, 0.2463, 0.2098], + device='cuda:0'), in_proj_covar=tensor([0.0724, 0.0733, 0.0611, 0.0855, 0.0728, 0.0641, 0.0748, 0.0660], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 12:33:31,510 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.306e+02 6.415e+02 7.779e+02 9.918e+02 2.619e+03, threshold=1.556e+03, percent-clipped=14.0 +2023-04-01 12:34:24,464 INFO [train.py:903] (0/4) Epoch 8, batch 3700, loss[loss=0.2273, simple_loss=0.2956, pruned_loss=0.07945, over 19726.00 frames. ], tot_loss[loss=0.2554, simple_loss=0.323, pruned_loss=0.09392, over 3830706.75 frames. ], batch size: 46, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:34:46,880 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51513.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:34:49,336 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51515.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:35:12,246 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51534.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:35:18,924 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51539.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:35:20,178 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51540.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:35:21,369 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51541.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:35:27,845 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8121, 1.5983, 1.4842, 1.8574, 1.6965, 1.5543, 1.6151, 1.7046], + device='cuda:0'), covar=tensor([0.0877, 0.1492, 0.1317, 0.0849, 0.1020, 0.0489, 0.1037, 0.0658], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0356, 0.0290, 0.0238, 0.0299, 0.0242, 0.0271, 0.0236], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 12:35:28,605 INFO [train.py:903] (0/4) Epoch 8, batch 3750, loss[loss=0.2637, simple_loss=0.3341, pruned_loss=0.09665, over 17543.00 frames. ], tot_loss[loss=0.2559, simple_loss=0.3237, pruned_loss=0.09407, over 3830990.94 frames. ], batch size: 101, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:35:36,643 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.985e+02 5.907e+02 7.282e+02 9.270e+02 2.268e+03, threshold=1.456e+03, percent-clipped=4.0 +2023-04-01 12:35:52,197 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51566.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 12:36:30,191 INFO [train.py:903] (0/4) Epoch 8, batch 3800, loss[loss=0.2366, simple_loss=0.2991, pruned_loss=0.08698, over 19732.00 frames. ], tot_loss[loss=0.2547, simple_loss=0.3229, pruned_loss=0.0933, over 3829307.21 frames. ], batch size: 45, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:37:02,478 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 12:37:31,376 INFO [train.py:903] (0/4) Epoch 8, batch 3850, loss[loss=0.2703, simple_loss=0.3479, pruned_loss=0.09631, over 19296.00 frames. ], tot_loss[loss=0.2554, simple_loss=0.3236, pruned_loss=0.0936, over 3821238.67 frames. ], batch size: 70, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:37:35,032 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51649.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 12:37:40,054 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.568e+02 6.155e+02 7.716e+02 1.023e+03 2.199e+03, threshold=1.543e+03, percent-clipped=8.0 +2023-04-01 12:38:13,457 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51679.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:38:33,778 INFO [train.py:903] (0/4) Epoch 8, batch 3900, loss[loss=0.2513, simple_loss=0.332, pruned_loss=0.08525, over 19616.00 frames. ], tot_loss[loss=0.2531, simple_loss=0.3217, pruned_loss=0.09219, over 3830347.93 frames. ], batch size: 57, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:38:44,978 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51704.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:39:33,677 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51744.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:39:37,156 INFO [train.py:903] (0/4) Epoch 8, batch 3950, loss[loss=0.2199, simple_loss=0.2861, pruned_loss=0.07685, over 19757.00 frames. ], tot_loss[loss=0.254, simple_loss=0.3227, pruned_loss=0.09266, over 3827851.33 frames. ], batch size: 46, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:39:38,716 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3886, 2.2468, 1.5988, 1.5179, 2.1098, 1.2605, 1.1184, 1.8647], + device='cuda:0'), covar=tensor([0.0831, 0.0594, 0.0832, 0.0609, 0.0373, 0.0976, 0.0705, 0.0385], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0286, 0.0310, 0.0237, 0.0226, 0.0305, 0.0279, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 12:39:41,705 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 12:39:45,225 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.212e+02 5.815e+02 7.280e+02 9.203e+02 2.422e+03, threshold=1.456e+03, percent-clipped=4.0 +2023-04-01 12:39:46,646 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51754.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:39:52,176 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-01 12:39:59,557 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=51765.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:40:37,935 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51795.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:40:38,730 INFO [train.py:903] (0/4) Epoch 8, batch 4000, loss[loss=0.254, simple_loss=0.33, pruned_loss=0.08902, over 19661.00 frames. ], tot_loss[loss=0.2551, simple_loss=0.3238, pruned_loss=0.09323, over 3819679.77 frames. ], batch size: 59, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:41:09,869 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51820.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:41:27,810 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 12:41:29,436 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7052, 2.1719, 1.7386, 1.6283, 2.0758, 1.5807, 1.4985, 1.7909], + device='cuda:0'), covar=tensor([0.0599, 0.0440, 0.0563, 0.0462, 0.0364, 0.0696, 0.0483, 0.0375], + device='cuda:0'), in_proj_covar=tensor([0.0283, 0.0290, 0.0314, 0.0239, 0.0227, 0.0308, 0.0282, 0.0251], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 12:41:41,697 INFO [train.py:903] (0/4) Epoch 8, batch 4050, loss[loss=0.2824, simple_loss=0.3434, pruned_loss=0.1107, over 18211.00 frames. ], tot_loss[loss=0.2546, simple_loss=0.3231, pruned_loss=0.09303, over 3816072.80 frames. ], batch size: 83, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:41:45,837 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-04-01 12:41:50,762 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.576e+02 5.742e+02 7.614e+02 9.901e+02 2.045e+03, threshold=1.523e+03, percent-clipped=5.0 +2023-04-01 12:41:55,471 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=51857.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:41:58,910 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51859.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:42:11,936 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51869.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:42:14,876 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.03 vs. limit=2.0 +2023-04-01 12:42:43,856 INFO [train.py:903] (0/4) Epoch 8, batch 4100, loss[loss=0.2979, simple_loss=0.3596, pruned_loss=0.1181, over 19112.00 frames. ], tot_loss[loss=0.2533, simple_loss=0.3218, pruned_loss=0.09237, over 3822449.97 frames. ], batch size: 69, lr: 1.04e-02, grad_scale: 8.0 +2023-04-01 12:42:56,230 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=51905.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 12:43:21,448 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 12:43:26,437 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=51930.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 12:43:47,771 INFO [train.py:903] (0/4) Epoch 8, batch 4150, loss[loss=0.2462, simple_loss=0.3189, pruned_loss=0.08673, over 19287.00 frames. ], tot_loss[loss=0.2525, simple_loss=0.3215, pruned_loss=0.09179, over 3840447.65 frames. ], batch size: 66, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:43:56,804 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.129e+02 6.343e+02 7.798e+02 9.790e+02 2.215e+03, threshold=1.560e+03, percent-clipped=4.0 +2023-04-01 12:44:19,713 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=51972.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:44:50,689 INFO [train.py:903] (0/4) Epoch 8, batch 4200, loss[loss=0.1953, simple_loss=0.2673, pruned_loss=0.06169, over 19719.00 frames. ], tot_loss[loss=0.2516, simple_loss=0.3204, pruned_loss=0.0914, over 3845691.57 frames. ], batch size: 46, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:44:55,600 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-52000.pt +2023-04-01 12:44:57,654 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 12:45:28,968 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-01 12:45:51,380 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5030, 1.2279, 1.1038, 1.4174, 1.2167, 1.2770, 1.0508, 1.3447], + device='cuda:0'), covar=tensor([0.0867, 0.1104, 0.1326, 0.0738, 0.0883, 0.0528, 0.1178, 0.0683], + device='cuda:0'), in_proj_covar=tensor([0.0247, 0.0350, 0.0283, 0.0234, 0.0295, 0.0239, 0.0268, 0.0232], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 12:45:53,317 INFO [train.py:903] (0/4) Epoch 8, batch 4250, loss[loss=0.2422, simple_loss=0.3041, pruned_loss=0.09013, over 19495.00 frames. ], tot_loss[loss=0.2515, simple_loss=0.3202, pruned_loss=0.09144, over 3829915.22 frames. ], batch size: 49, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:46:01,321 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.676e+02 5.455e+02 6.472e+02 8.916e+02 2.597e+03, threshold=1.294e+03, percent-clipped=4.0 +2023-04-01 12:46:08,431 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52058.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:46:11,491 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 12:46:21,927 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 12:46:55,489 INFO [train.py:903] (0/4) Epoch 8, batch 4300, loss[loss=0.2305, simple_loss=0.2997, pruned_loss=0.08067, over 19815.00 frames. ], tot_loss[loss=0.2523, simple_loss=0.321, pruned_loss=0.09175, over 3813540.92 frames. ], batch size: 48, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:47:14,054 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52109.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:47:22,315 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52115.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:47:34,074 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52125.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:47:50,788 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 12:47:53,279 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52140.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:47:57,575 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9836, 1.9291, 1.5798, 1.4983, 1.2551, 1.4901, 0.5042, 1.0276], + device='cuda:0'), covar=tensor([0.0476, 0.0459, 0.0357, 0.0550, 0.0904, 0.0619, 0.0794, 0.0709], + device='cuda:0'), in_proj_covar=tensor([0.0318, 0.0316, 0.0315, 0.0332, 0.0412, 0.0331, 0.0296, 0.0314], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 12:48:00,440 INFO [train.py:903] (0/4) Epoch 8, batch 4350, loss[loss=0.2383, simple_loss=0.3153, pruned_loss=0.08068, over 19658.00 frames. ], tot_loss[loss=0.2511, simple_loss=0.3197, pruned_loss=0.09122, over 3823152.68 frames. ], batch size: 55, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:48:06,421 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52150.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:48:09,409 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.508e+02 5.672e+02 7.291e+02 9.101e+02 1.997e+03, threshold=1.458e+03, percent-clipped=8.0 +2023-04-01 12:48:29,737 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7569, 4.1450, 4.4258, 4.3994, 1.6732, 4.1385, 3.6774, 4.0292], + device='cuda:0'), covar=tensor([0.1128, 0.0724, 0.0534, 0.0476, 0.4551, 0.0485, 0.0559, 0.1069], + device='cuda:0'), in_proj_covar=tensor([0.0610, 0.0532, 0.0720, 0.0610, 0.0672, 0.0467, 0.0457, 0.0670], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 12:49:03,039 INFO [train.py:903] (0/4) Epoch 8, batch 4400, loss[loss=0.3244, simple_loss=0.3718, pruned_loss=0.1385, over 19728.00 frames. ], tot_loss[loss=0.2521, simple_loss=0.3202, pruned_loss=0.09201, over 3814588.33 frames. ], batch size: 63, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:49:21,164 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52211.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:49:26,769 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 12:49:37,191 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4790, 1.5034, 1.8690, 1.9161, 2.9348, 3.9305, 3.9469, 4.3911], + device='cuda:0'), covar=tensor([0.1494, 0.2969, 0.2865, 0.1730, 0.0539, 0.0316, 0.0176, 0.0127], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0286, 0.0317, 0.0247, 0.0207, 0.0138, 0.0205, 0.0175], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 12:49:38,214 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 12:49:38,565 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52224.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:49:43,352 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52228.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:50:05,229 INFO [train.py:903] (0/4) Epoch 8, batch 4450, loss[loss=0.2032, simple_loss=0.2784, pruned_loss=0.06404, over 19464.00 frames. ], tot_loss[loss=0.2512, simple_loss=0.3199, pruned_loss=0.09129, over 3822211.52 frames. ], batch size: 49, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:50:13,302 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.227e+02 5.880e+02 7.086e+02 8.839e+02 1.936e+03, threshold=1.417e+03, percent-clipped=3.0 +2023-04-01 12:50:13,713 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52253.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:51:06,387 INFO [train.py:903] (0/4) Epoch 8, batch 4500, loss[loss=0.2962, simple_loss=0.3578, pruned_loss=0.1173, over 19722.00 frames. ], tot_loss[loss=0.2502, simple_loss=0.3193, pruned_loss=0.0906, over 3826568.00 frames. ], batch size: 63, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:51:50,107 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52330.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:52:10,352 INFO [train.py:903] (0/4) Epoch 8, batch 4550, loss[loss=0.2086, simple_loss=0.2789, pruned_loss=0.06913, over 19134.00 frames. ], tot_loss[loss=0.2488, simple_loss=0.3179, pruned_loss=0.08989, over 3819991.34 frames. ], batch size: 42, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:52:18,695 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.675e+02 5.922e+02 7.010e+02 8.869e+02 1.679e+03, threshold=1.402e+03, percent-clipped=2.0 +2023-04-01 12:52:18,736 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 12:52:25,128 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-04-01 12:52:41,946 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 12:53:11,447 INFO [train.py:903] (0/4) Epoch 8, batch 4600, loss[loss=0.2326, simple_loss=0.308, pruned_loss=0.07863, over 19590.00 frames. ], tot_loss[loss=0.2495, simple_loss=0.3183, pruned_loss=0.09038, over 3817668.43 frames. ], batch size: 52, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:53:18,524 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52402.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:54:12,886 INFO [train.py:903] (0/4) Epoch 8, batch 4650, loss[loss=0.2488, simple_loss=0.3158, pruned_loss=0.09087, over 19610.00 frames. ], tot_loss[loss=0.2488, simple_loss=0.3178, pruned_loss=0.08994, over 3816246.66 frames. ], batch size: 52, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:54:21,262 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.655e+02 5.664e+02 6.903e+02 8.285e+02 1.576e+03, threshold=1.381e+03, percent-clipped=2.0 +2023-04-01 12:54:30,516 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 12:54:42,682 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 12:54:56,955 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52480.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:55:15,223 INFO [train.py:903] (0/4) Epoch 8, batch 4700, loss[loss=0.3027, simple_loss=0.3553, pruned_loss=0.1251, over 13451.00 frames. ], tot_loss[loss=0.2494, simple_loss=0.3184, pruned_loss=0.09019, over 3809243.98 frames. ], batch size: 138, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:55:27,863 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52505.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:55:39,729 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 12:55:43,532 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52517.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:56:18,368 INFO [train.py:903] (0/4) Epoch 8, batch 4750, loss[loss=0.2577, simple_loss=0.3362, pruned_loss=0.08957, over 19544.00 frames. ], tot_loss[loss=0.2494, simple_loss=0.3186, pruned_loss=0.09009, over 3812161.28 frames. ], batch size: 56, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:56:29,690 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.555e+02 6.316e+02 7.348e+02 9.529e+02 1.491e+03, threshold=1.470e+03, percent-clipped=3.0 +2023-04-01 12:56:31,055 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52555.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:56:52,281 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52573.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 12:57:22,243 INFO [train.py:903] (0/4) Epoch 8, batch 4800, loss[loss=0.2585, simple_loss=0.3374, pruned_loss=0.08975, over 18084.00 frames. ], tot_loss[loss=0.2508, simple_loss=0.3195, pruned_loss=0.09104, over 3811695.84 frames. ], batch size: 83, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:58:22,763 INFO [train.py:903] (0/4) Epoch 8, batch 4850, loss[loss=0.2485, simple_loss=0.3244, pruned_loss=0.08634, over 19747.00 frames. ], tot_loss[loss=0.2507, simple_loss=0.3192, pruned_loss=0.09106, over 3810103.56 frames. ], batch size: 63, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:58:32,082 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.583e+02 6.021e+02 7.604e+02 9.872e+02 2.114e+03, threshold=1.521e+03, percent-clipped=8.0 +2023-04-01 12:58:45,992 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 12:58:52,918 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52670.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:58:55,093 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2826, 1.3176, 1.4224, 1.5062, 2.8168, 0.9940, 2.1323, 3.1068], + device='cuda:0'), covar=tensor([0.0520, 0.2498, 0.2563, 0.1582, 0.0812, 0.2311, 0.1119, 0.0350], + device='cuda:0'), in_proj_covar=tensor([0.0325, 0.0324, 0.0335, 0.0303, 0.0329, 0.0319, 0.0308, 0.0325], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 12:58:58,258 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52674.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 12:59:08,333 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 12:59:14,079 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 12:59:14,106 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 12:59:23,232 INFO [train.py:903] (0/4) Epoch 8, batch 4900, loss[loss=0.2377, simple_loss=0.2947, pruned_loss=0.09036, over 19775.00 frames. ], tot_loss[loss=0.2515, simple_loss=0.32, pruned_loss=0.09154, over 3811413.77 frames. ], batch size: 48, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 12:59:24,415 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 12:59:44,290 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 13:00:22,669 INFO [train.py:903] (0/4) Epoch 8, batch 4950, loss[loss=0.2352, simple_loss=0.297, pruned_loss=0.08667, over 19770.00 frames. ], tot_loss[loss=0.2513, simple_loss=0.3195, pruned_loss=0.0915, over 3811592.74 frames. ], batch size: 48, lr: 1.03e-02, grad_scale: 8.0 +2023-04-01 13:00:35,713 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.348e+02 6.374e+02 8.178e+02 1.048e+03 2.702e+03, threshold=1.636e+03, percent-clipped=11.0 +2023-04-01 13:00:39,449 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4681, 1.2865, 1.4285, 1.5627, 2.9647, 0.8317, 2.1968, 3.2745], + device='cuda:0'), covar=tensor([0.0447, 0.2522, 0.2553, 0.1513, 0.0786, 0.2546, 0.1218, 0.0334], + device='cuda:0'), in_proj_covar=tensor([0.0329, 0.0326, 0.0338, 0.0305, 0.0332, 0.0324, 0.0311, 0.0329], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:00:40,338 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 13:00:53,291 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=52770.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:00:56,997 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52773.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:01:04,907 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 13:01:14,501 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3456, 2.2396, 1.7754, 1.6425, 1.5323, 1.7911, 0.3337, 1.0351], + device='cuda:0'), covar=tensor([0.0302, 0.0326, 0.0268, 0.0429, 0.0687, 0.0450, 0.0693, 0.0634], + device='cuda:0'), in_proj_covar=tensor([0.0317, 0.0317, 0.0313, 0.0331, 0.0407, 0.0328, 0.0295, 0.0312], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 13:01:17,589 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=52789.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:01:19,453 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.67 vs. limit=5.0 +2023-04-01 13:01:26,610 INFO [train.py:903] (0/4) Epoch 8, batch 5000, loss[loss=0.2345, simple_loss=0.3136, pruned_loss=0.0777, over 19695.00 frames. ], tot_loss[loss=0.2516, simple_loss=0.32, pruned_loss=0.09162, over 3802334.20 frames. ], batch size: 53, lr: 1.03e-02, grad_scale: 4.0 +2023-04-01 13:01:29,398 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52798.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:01:35,982 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 13:01:47,278 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 13:02:28,523 INFO [train.py:903] (0/4) Epoch 8, batch 5050, loss[loss=0.24, simple_loss=0.3149, pruned_loss=0.08251, over 19664.00 frames. ], tot_loss[loss=0.25, simple_loss=0.3187, pruned_loss=0.09059, over 3807730.58 frames. ], batch size: 53, lr: 1.03e-02, grad_scale: 4.0 +2023-04-01 13:02:39,042 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.635e+02 5.648e+02 7.062e+02 8.811e+02 1.795e+03, threshold=1.412e+03, percent-clipped=2.0 +2023-04-01 13:03:04,997 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 13:03:15,058 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-01 13:03:30,501 INFO [train.py:903] (0/4) Epoch 8, batch 5100, loss[loss=0.2386, simple_loss=0.3042, pruned_loss=0.08652, over 18573.00 frames. ], tot_loss[loss=0.2507, simple_loss=0.3194, pruned_loss=0.09102, over 3807128.65 frames. ], batch size: 41, lr: 1.03e-02, grad_scale: 4.0 +2023-04-01 13:03:30,934 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0299, 2.0294, 1.6371, 1.6131, 1.4291, 1.5969, 0.1901, 0.7642], + device='cuda:0'), covar=tensor([0.0332, 0.0350, 0.0252, 0.0365, 0.0779, 0.0426, 0.0732, 0.0656], + device='cuda:0'), in_proj_covar=tensor([0.0322, 0.0321, 0.0318, 0.0335, 0.0413, 0.0331, 0.0297, 0.0315], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 13:03:41,003 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 13:03:45,397 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 13:03:48,754 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9053, 1.7089, 1.5522, 1.9708, 1.9011, 1.7151, 1.7442, 1.8633], + device='cuda:0'), covar=tensor([0.0937, 0.1622, 0.1394, 0.0992, 0.1211, 0.0516, 0.1049, 0.0652], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0358, 0.0291, 0.0240, 0.0303, 0.0245, 0.0277, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:03:50,849 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 13:03:59,135 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=52917.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 13:04:09,616 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=52926.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:04:32,212 INFO [train.py:903] (0/4) Epoch 8, batch 5150, loss[loss=0.2546, simple_loss=0.3313, pruned_loss=0.08893, over 19779.00 frames. ], tot_loss[loss=0.2491, simple_loss=0.318, pruned_loss=0.0901, over 3813150.31 frames. ], batch size: 56, lr: 1.03e-02, grad_scale: 4.0 +2023-04-01 13:04:40,975 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=52951.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:04:45,910 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.391e+02 6.060e+02 7.983e+02 1.041e+03 2.368e+03, threshold=1.597e+03, percent-clipped=6.0 +2023-04-01 13:04:47,177 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 13:04:55,784 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8763, 1.5932, 1.4155, 2.1438, 1.6493, 2.2041, 2.2651, 1.9187], + device='cuda:0'), covar=tensor([0.0716, 0.0864, 0.0966, 0.0748, 0.0903, 0.0585, 0.0700, 0.0573], + device='cuda:0'), in_proj_covar=tensor([0.0219, 0.0235, 0.0233, 0.0260, 0.0249, 0.0219, 0.0210, 0.0210], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 13:05:19,594 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 13:05:37,037 INFO [train.py:903] (0/4) Epoch 8, batch 5200, loss[loss=0.2281, simple_loss=0.2981, pruned_loss=0.07909, over 16082.00 frames. ], tot_loss[loss=0.2512, simple_loss=0.32, pruned_loss=0.09116, over 3806822.66 frames. ], batch size: 35, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:05:51,328 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 13:05:51,705 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3885, 1.3847, 1.8960, 1.4918, 2.6282, 2.1172, 2.6833, 1.1890], + device='cuda:0'), covar=tensor([0.1901, 0.3195, 0.1750, 0.1534, 0.1175, 0.1574, 0.1187, 0.2967], + device='cuda:0'), in_proj_covar=tensor([0.0468, 0.0538, 0.0542, 0.0416, 0.0576, 0.0472, 0.0637, 0.0469], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 13:06:21,619 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53032.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 13:06:36,718 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 13:06:38,303 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53045.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:06:39,061 INFO [train.py:903] (0/4) Epoch 8, batch 5250, loss[loss=0.2108, simple_loss=0.2818, pruned_loss=0.06993, over 19760.00 frames. ], tot_loss[loss=0.2509, simple_loss=0.3197, pruned_loss=0.09102, over 3807935.48 frames. ], batch size: 46, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:06:49,012 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.283e+02 5.973e+02 7.081e+02 8.822e+02 3.028e+03, threshold=1.416e+03, percent-clipped=2.0 +2023-04-01 13:07:08,270 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53070.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:07:13,746 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53074.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:07:39,281 INFO [train.py:903] (0/4) Epoch 8, batch 5300, loss[loss=0.2492, simple_loss=0.323, pruned_loss=0.08771, over 19542.00 frames. ], tot_loss[loss=0.2516, simple_loss=0.3205, pruned_loss=0.09138, over 3820378.49 frames. ], batch size: 56, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:07:57,420 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 13:08:03,077 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53114.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:08:25,885 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-04-01 13:08:41,249 INFO [train.py:903] (0/4) Epoch 8, batch 5350, loss[loss=0.2053, simple_loss=0.2711, pruned_loss=0.06975, over 19790.00 frames. ], tot_loss[loss=0.2515, simple_loss=0.3204, pruned_loss=0.09123, over 3818400.06 frames. ], batch size: 47, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:08:52,775 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.210e+02 6.176e+02 7.478e+02 9.376e+02 1.338e+03, threshold=1.496e+03, percent-clipped=0.0 +2023-04-01 13:09:14,589 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7935, 1.5740, 1.4038, 1.6445, 1.5345, 1.4994, 1.5465, 1.5863], + device='cuda:0'), covar=tensor([0.0848, 0.1310, 0.1430, 0.0978, 0.1207, 0.0765, 0.1143, 0.0837], + device='cuda:0'), in_proj_covar=tensor([0.0250, 0.0354, 0.0287, 0.0240, 0.0299, 0.0242, 0.0273, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:09:18,557 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 13:09:44,002 INFO [train.py:903] (0/4) Epoch 8, batch 5400, loss[loss=0.2692, simple_loss=0.3288, pruned_loss=0.1048, over 19507.00 frames. ], tot_loss[loss=0.2516, simple_loss=0.3205, pruned_loss=0.09135, over 3802611.57 frames. ], batch size: 64, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:10:24,037 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53229.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:10:47,209 INFO [train.py:903] (0/4) Epoch 8, batch 5450, loss[loss=0.211, simple_loss=0.2837, pruned_loss=0.06911, over 19721.00 frames. ], tot_loss[loss=0.2511, simple_loss=0.3198, pruned_loss=0.09122, over 3811099.17 frames. ], batch size: 51, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:10:57,345 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.737e+02 5.768e+02 7.150e+02 9.218e+02 2.127e+03, threshold=1.430e+03, percent-clipped=3.0 +2023-04-01 13:11:02,279 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53259.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:11:40,463 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53288.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 13:11:48,913 INFO [train.py:903] (0/4) Epoch 8, batch 5500, loss[loss=0.2646, simple_loss=0.3165, pruned_loss=0.1063, over 19424.00 frames. ], tot_loss[loss=0.2522, simple_loss=0.3208, pruned_loss=0.09185, over 3806684.69 frames. ], batch size: 43, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:12:11,184 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53313.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 13:12:16,567 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 13:12:23,256 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53322.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:12:34,450 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1614, 1.0838, 1.4589, 0.8226, 2.3178, 3.0249, 2.7939, 3.2349], + device='cuda:0'), covar=tensor([0.1443, 0.3264, 0.2960, 0.2185, 0.0459, 0.0169, 0.0234, 0.0173], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0287, 0.0315, 0.0247, 0.0207, 0.0139, 0.0205, 0.0174], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 13:12:35,592 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53333.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:12:50,196 INFO [train.py:903] (0/4) Epoch 8, batch 5550, loss[loss=0.2696, simple_loss=0.3393, pruned_loss=0.09992, over 18467.00 frames. ], tot_loss[loss=0.2525, simple_loss=0.3211, pruned_loss=0.09195, over 3814016.63 frames. ], batch size: 84, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:12:59,710 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 13:13:03,078 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.436e+02 6.394e+02 7.821e+02 9.803e+02 2.197e+03, threshold=1.564e+03, percent-clipped=2.0 +2023-04-01 13:13:40,357 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=53385.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:13:50,338 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 13:13:53,916 INFO [train.py:903] (0/4) Epoch 8, batch 5600, loss[loss=0.2418, simple_loss=0.3186, pruned_loss=0.08248, over 19678.00 frames. ], tot_loss[loss=0.2529, simple_loss=0.3217, pruned_loss=0.09212, over 3809015.22 frames. ], batch size: 59, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:14:21,324 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53418.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:14:57,495 INFO [train.py:903] (0/4) Epoch 8, batch 5650, loss[loss=0.2711, simple_loss=0.3318, pruned_loss=0.1052, over 19665.00 frames. ], tot_loss[loss=0.2545, simple_loss=0.3227, pruned_loss=0.09313, over 3799226.30 frames. ], batch size: 60, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:15:07,839 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.837e+02 5.961e+02 7.306e+02 9.270e+02 2.985e+03, threshold=1.461e+03, percent-clipped=1.0 +2023-04-01 13:15:45,799 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 13:15:46,228 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53485.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:15:58,761 INFO [train.py:903] (0/4) Epoch 8, batch 5700, loss[loss=0.2297, simple_loss=0.3106, pruned_loss=0.07437, over 19676.00 frames. ], tot_loss[loss=0.2546, simple_loss=0.3232, pruned_loss=0.09307, over 3798425.28 frames. ], batch size: 60, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:16:15,331 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53510.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:16:44,952 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53533.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:16:59,704 INFO [train.py:903] (0/4) Epoch 8, batch 5750, loss[loss=0.2562, simple_loss=0.3234, pruned_loss=0.09447, over 19531.00 frames. ], tot_loss[loss=0.2522, simple_loss=0.3215, pruned_loss=0.0914, over 3803214.10 frames. ], batch size: 54, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:17:00,958 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 13:17:10,455 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 13:17:11,664 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.475e+02 5.699e+02 6.647e+02 8.184e+02 1.829e+03, threshold=1.329e+03, percent-clipped=1.0 +2023-04-01 13:17:15,820 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 13:18:02,113 INFO [train.py:903] (0/4) Epoch 8, batch 5800, loss[loss=0.2668, simple_loss=0.3316, pruned_loss=0.101, over 18904.00 frames. ], tot_loss[loss=0.2523, simple_loss=0.3216, pruned_loss=0.09147, over 3811813.88 frames. ], batch size: 74, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:18:12,767 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53603.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:19:04,522 INFO [train.py:903] (0/4) Epoch 8, batch 5850, loss[loss=0.2541, simple_loss=0.33, pruned_loss=0.08907, over 19489.00 frames. ], tot_loss[loss=0.2518, simple_loss=0.3212, pruned_loss=0.09113, over 3811327.19 frames. ], batch size: 64, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:19:15,051 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.702e+02 6.371e+02 7.746e+02 9.220e+02 2.993e+03, threshold=1.549e+03, percent-clipped=10.0 +2023-04-01 13:19:27,604 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53666.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:19:41,999 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53677.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:20:05,080 INFO [train.py:903] (0/4) Epoch 8, batch 5900, loss[loss=0.2596, simple_loss=0.3364, pruned_loss=0.09143, over 19746.00 frames. ], tot_loss[loss=0.2517, simple_loss=0.3214, pruned_loss=0.09104, over 3806240.67 frames. ], batch size: 63, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:20:09,559 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 13:20:30,235 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 13:20:32,887 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53718.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:20:47,010 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=53729.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:21:06,831 INFO [train.py:903] (0/4) Epoch 8, batch 5950, loss[loss=0.2163, simple_loss=0.2945, pruned_loss=0.06907, over 19609.00 frames. ], tot_loss[loss=0.2505, simple_loss=0.3201, pruned_loss=0.09042, over 3818894.35 frames. ], batch size: 50, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:21:10,425 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1594, 3.8304, 2.1701, 2.1822, 3.3184, 2.0075, 1.4639, 1.9358], + device='cuda:0'), covar=tensor([0.0965, 0.0301, 0.0793, 0.0653, 0.0376, 0.0941, 0.0823, 0.0587], + device='cuda:0'), in_proj_covar=tensor([0.0283, 0.0293, 0.0319, 0.0245, 0.0228, 0.0319, 0.0291, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:21:19,050 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.652e+02 5.886e+02 7.195e+02 1.025e+03 2.007e+03, threshold=1.439e+03, percent-clipped=3.0 +2023-04-01 13:21:50,757 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53781.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:22:00,077 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53789.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:22:04,770 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53792.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:22:09,975 INFO [train.py:903] (0/4) Epoch 8, batch 6000, loss[loss=0.2806, simple_loss=0.3371, pruned_loss=0.1121, over 19598.00 frames. ], tot_loss[loss=0.2513, simple_loss=0.3205, pruned_loss=0.0911, over 3797427.81 frames. ], batch size: 52, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:22:09,975 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 13:22:22,632 INFO [train.py:937] (0/4) Epoch 8, validation: loss=0.1864, simple_loss=0.2865, pruned_loss=0.04314, over 944034.00 frames. +2023-04-01 13:22:22,633 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 13:22:34,008 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.11 vs. limit=2.0 +2023-04-01 13:22:48,334 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53814.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:23:06,483 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.1188, 5.4760, 3.1817, 4.6940, 1.1380, 5.2220, 5.3378, 5.4438], + device='cuda:0'), covar=tensor([0.0415, 0.0851, 0.1664, 0.0613, 0.3929, 0.0561, 0.0681, 0.0884], + device='cuda:0'), in_proj_covar=tensor([0.0408, 0.0348, 0.0410, 0.0305, 0.0373, 0.0337, 0.0329, 0.0367], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 13:23:10,013 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0335, 1.7773, 1.5947, 1.9944, 1.8593, 1.8659, 1.6498, 1.9883], + device='cuda:0'), covar=tensor([0.0850, 0.1501, 0.1307, 0.0977, 0.1188, 0.0482, 0.1109, 0.0617], + device='cuda:0'), in_proj_covar=tensor([0.0248, 0.0355, 0.0289, 0.0238, 0.0298, 0.0241, 0.0272, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:23:24,801 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=53844.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:23:26,621 INFO [train.py:903] (0/4) Epoch 8, batch 6050, loss[loss=0.2853, simple_loss=0.3504, pruned_loss=0.1101, over 19444.00 frames. ], tot_loss[loss=0.2509, simple_loss=0.3203, pruned_loss=0.09081, over 3794606.88 frames. ], batch size: 64, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:23:39,145 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.317e+02 5.583e+02 7.013e+02 9.917e+02 2.418e+03, threshold=1.403e+03, percent-clipped=8.0 +2023-04-01 13:24:06,029 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6279, 2.4973, 1.6751, 1.5264, 2.2547, 1.3073, 1.3266, 1.8556], + device='cuda:0'), covar=tensor([0.0803, 0.0543, 0.0953, 0.0663, 0.0403, 0.1115, 0.0705, 0.0456], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0292, 0.0314, 0.0244, 0.0227, 0.0316, 0.0290, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:24:30,159 INFO [train.py:903] (0/4) Epoch 8, batch 6100, loss[loss=0.2404, simple_loss=0.3254, pruned_loss=0.07776, over 19683.00 frames. ], tot_loss[loss=0.2501, simple_loss=0.3197, pruned_loss=0.0902, over 3790984.85 frames. ], batch size: 58, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:25:31,558 INFO [train.py:903] (0/4) Epoch 8, batch 6150, loss[loss=0.2354, simple_loss=0.3063, pruned_loss=0.08229, over 19748.00 frames. ], tot_loss[loss=0.2513, simple_loss=0.3206, pruned_loss=0.09102, over 3796070.61 frames. ], batch size: 51, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:25:42,179 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.923e+02 5.703e+02 7.303e+02 8.849e+02 1.874e+03, threshold=1.461e+03, percent-clipped=4.0 +2023-04-01 13:25:56,449 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 13:26:07,872 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=53974.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:26:33,363 INFO [train.py:903] (0/4) Epoch 8, batch 6200, loss[loss=0.2609, simple_loss=0.329, pruned_loss=0.09644, over 19742.00 frames. ], tot_loss[loss=0.2509, simple_loss=0.3198, pruned_loss=0.09097, over 3808490.00 frames. ], batch size: 63, lr: 1.02e-02, grad_scale: 8.0 +2023-04-01 13:26:35,345 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-04-01 13:26:37,419 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=53999.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:26:39,237 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-54000.pt +2023-04-01 13:27:25,486 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=54037.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:27:37,604 INFO [train.py:903] (0/4) Epoch 8, batch 6250, loss[loss=0.2861, simple_loss=0.3467, pruned_loss=0.1128, over 17600.00 frames. ], tot_loss[loss=0.2493, simple_loss=0.3185, pruned_loss=0.09005, over 3816522.50 frames. ], batch size: 101, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:27:39,096 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0161, 1.2045, 1.3174, 1.3595, 2.5677, 0.9142, 1.9109, 2.8518], + device='cuda:0'), covar=tensor([0.0521, 0.2534, 0.2633, 0.1609, 0.0820, 0.2325, 0.1201, 0.0391], + device='cuda:0'), in_proj_covar=tensor([0.0325, 0.0323, 0.0334, 0.0304, 0.0330, 0.0319, 0.0311, 0.0324], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:27:40,341 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=54048.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:27:49,335 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.888e+02 5.723e+02 7.068e+02 9.572e+02 2.133e+03, threshold=1.414e+03, percent-clipped=6.0 +2023-04-01 13:27:58,109 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=54062.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:28:04,584 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 13:28:09,556 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6237, 1.3801, 1.4585, 1.9444, 1.5197, 1.8512, 2.0097, 1.7811], + device='cuda:0'), covar=tensor([0.0811, 0.1015, 0.1032, 0.0844, 0.0933, 0.0756, 0.0838, 0.0660], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0228, 0.0226, 0.0254, 0.0243, 0.0211, 0.0206, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 13:28:10,804 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=54073.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:28:40,884 INFO [train.py:903] (0/4) Epoch 8, batch 6300, loss[loss=0.2241, simple_loss=0.2884, pruned_loss=0.07991, over 19749.00 frames. ], tot_loss[loss=0.2492, simple_loss=0.3185, pruned_loss=0.08994, over 3817370.86 frames. ], batch size: 46, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:28:45,594 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=54100.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:29:16,207 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=54125.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:29:41,755 INFO [train.py:903] (0/4) Epoch 8, batch 6350, loss[loss=0.2223, simple_loss=0.294, pruned_loss=0.07534, over 19338.00 frames. ], tot_loss[loss=0.2498, simple_loss=0.3189, pruned_loss=0.09028, over 3831023.47 frames. ], batch size: 44, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:29:52,040 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.072e+02 6.055e+02 7.520e+02 8.667e+02 2.456e+03, threshold=1.504e+03, percent-clipped=3.0 +2023-04-01 13:30:19,246 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2418, 3.6905, 3.8605, 3.8280, 1.3341, 3.6045, 3.1747, 3.5752], + device='cuda:0'), covar=tensor([0.1280, 0.0779, 0.0614, 0.0598, 0.4635, 0.0548, 0.0656, 0.1122], + device='cuda:0'), in_proj_covar=tensor([0.0617, 0.0543, 0.0734, 0.0611, 0.0675, 0.0479, 0.0463, 0.0677], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 13:30:43,420 INFO [train.py:903] (0/4) Epoch 8, batch 6400, loss[loss=0.2391, simple_loss=0.306, pruned_loss=0.08612, over 19847.00 frames. ], tot_loss[loss=0.2501, simple_loss=0.3191, pruned_loss=0.09052, over 3836734.88 frames. ], batch size: 52, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:31:45,983 INFO [train.py:903] (0/4) Epoch 8, batch 6450, loss[loss=0.2204, simple_loss=0.2817, pruned_loss=0.07955, over 19774.00 frames. ], tot_loss[loss=0.2497, simple_loss=0.3188, pruned_loss=0.09035, over 3819923.89 frames. ], batch size: 47, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:31:58,335 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.099e+02 5.571e+02 6.761e+02 8.500e+02 1.702e+03, threshold=1.352e+03, percent-clipped=3.0 +2023-04-01 13:32:30,248 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 13:32:35,376 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-01 13:32:48,754 INFO [train.py:903] (0/4) Epoch 8, batch 6500, loss[loss=0.247, simple_loss=0.3248, pruned_loss=0.08458, over 19673.00 frames. ], tot_loss[loss=0.2482, simple_loss=0.3176, pruned_loss=0.08938, over 3832084.01 frames. ], batch size: 55, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:32:54,434 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 13:33:50,179 INFO [train.py:903] (0/4) Epoch 8, batch 6550, loss[loss=0.2514, simple_loss=0.3239, pruned_loss=0.08949, over 19680.00 frames. ], tot_loss[loss=0.25, simple_loss=0.3193, pruned_loss=0.09036, over 3832635.75 frames. ], batch size: 59, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:33:53,971 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3118, 1.7554, 1.9057, 2.6661, 1.8620, 2.6935, 2.5880, 2.3405], + device='cuda:0'), covar=tensor([0.0726, 0.0924, 0.0988, 0.0857, 0.0966, 0.0583, 0.0840, 0.0597], + device='cuda:0'), in_proj_covar=tensor([0.0219, 0.0234, 0.0232, 0.0256, 0.0246, 0.0216, 0.0207, 0.0209], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 13:34:00,552 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.237e+02 6.329e+02 8.151e+02 1.084e+03 2.341e+03, threshold=1.630e+03, percent-clipped=12.0 +2023-04-01 13:34:26,067 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.63 vs. limit=2.0 +2023-04-01 13:34:51,138 INFO [train.py:903] (0/4) Epoch 8, batch 6600, loss[loss=0.2179, simple_loss=0.2938, pruned_loss=0.07098, over 19765.00 frames. ], tot_loss[loss=0.2488, simple_loss=0.318, pruned_loss=0.08977, over 3840995.77 frames. ], batch size: 48, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:35:16,706 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.71 vs. limit=5.0 +2023-04-01 13:35:19,566 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9629, 1.4766, 1.5217, 1.9522, 1.5790, 1.7631, 1.6290, 1.8073], + device='cuda:0'), covar=tensor([0.0820, 0.1576, 0.1270, 0.0915, 0.1165, 0.0453, 0.1003, 0.0615], + device='cuda:0'), in_proj_covar=tensor([0.0249, 0.0359, 0.0290, 0.0241, 0.0300, 0.0244, 0.0270, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:35:32,053 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-04-01 13:35:46,650 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2607, 1.3633, 1.6561, 1.4236, 2.5229, 2.0686, 2.4921, 0.9753], + device='cuda:0'), covar=tensor([0.2041, 0.3471, 0.2042, 0.1719, 0.1171, 0.1832, 0.1366, 0.3345], + device='cuda:0'), in_proj_covar=tensor([0.0467, 0.0545, 0.0549, 0.0420, 0.0573, 0.0474, 0.0640, 0.0474], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 13:35:53,837 INFO [train.py:903] (0/4) Epoch 8, batch 6650, loss[loss=0.2365, simple_loss=0.3025, pruned_loss=0.0852, over 19740.00 frames. ], tot_loss[loss=0.2486, simple_loss=0.3179, pruned_loss=0.08969, over 3829148.19 frames. ], batch size: 47, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:36:04,915 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.746e+02 5.865e+02 7.808e+02 1.010e+03 1.907e+03, threshold=1.562e+03, percent-clipped=2.0 +2023-04-01 13:36:55,565 INFO [train.py:903] (0/4) Epoch 8, batch 6700, loss[loss=0.2175, simple_loss=0.2804, pruned_loss=0.07735, over 19742.00 frames. ], tot_loss[loss=0.2483, simple_loss=0.3179, pruned_loss=0.0894, over 3823104.86 frames. ], batch size: 46, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:37:52,318 INFO [train.py:903] (0/4) Epoch 8, batch 6750, loss[loss=0.2348, simple_loss=0.3125, pruned_loss=0.0786, over 19589.00 frames. ], tot_loss[loss=0.2492, simple_loss=0.3182, pruned_loss=0.09015, over 3809203.82 frames. ], batch size: 52, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:38:03,632 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.992e+02 6.380e+02 7.224e+02 9.353e+02 2.017e+03, threshold=1.445e+03, percent-clipped=3.0 +2023-04-01 13:38:12,145 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-01 13:38:50,509 INFO [train.py:903] (0/4) Epoch 8, batch 6800, loss[loss=0.2128, simple_loss=0.2887, pruned_loss=0.06844, over 19750.00 frames. ], tot_loss[loss=0.249, simple_loss=0.3179, pruned_loss=0.09001, over 3821022.05 frames. ], batch size: 54, lr: 1.01e-02, grad_scale: 8.0 +2023-04-01 13:39:13,885 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.44 vs. limit=5.0 +2023-04-01 13:39:19,159 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-8.pt +2023-04-01 13:39:34,794 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 13:39:35,256 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 13:39:38,465 INFO [train.py:903] (0/4) Epoch 9, batch 0, loss[loss=0.2728, simple_loss=0.3405, pruned_loss=0.1026, over 19748.00 frames. ], tot_loss[loss=0.2728, simple_loss=0.3405, pruned_loss=0.1026, over 19748.00 frames. ], batch size: 63, lr: 9.56e-03, grad_scale: 8.0 +2023-04-01 13:39:38,465 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 13:39:49,509 INFO [train.py:937] (0/4) Epoch 9, validation: loss=0.1866, simple_loss=0.2872, pruned_loss=0.04294, over 944034.00 frames. +2023-04-01 13:39:49,510 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 13:40:03,807 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 13:40:28,239 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.879e+02 5.500e+02 7.208e+02 8.930e+02 1.459e+03, threshold=1.442e+03, percent-clipped=1.0 +2023-04-01 13:40:51,457 INFO [train.py:903] (0/4) Epoch 9, batch 50, loss[loss=0.2627, simple_loss=0.3248, pruned_loss=0.1004, over 13579.00 frames. ], tot_loss[loss=0.2575, simple_loss=0.3249, pruned_loss=0.09508, over 855116.27 frames. ], batch size: 136, lr: 9.55e-03, grad_scale: 8.0 +2023-04-01 13:41:02,921 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=54682.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 13:41:26,435 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 13:41:53,043 INFO [train.py:903] (0/4) Epoch 9, batch 100, loss[loss=0.2272, simple_loss=0.3062, pruned_loss=0.07409, over 17393.00 frames. ], tot_loss[loss=0.2523, simple_loss=0.3213, pruned_loss=0.09162, over 1527801.42 frames. ], batch size: 101, lr: 9.55e-03, grad_scale: 8.0 +2023-04-01 13:42:05,367 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 13:42:31,167 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.400e+02 5.564e+02 7.190e+02 9.001e+02 2.877e+03, threshold=1.438e+03, percent-clipped=2.0 +2023-04-01 13:42:53,291 INFO [train.py:903] (0/4) Epoch 9, batch 150, loss[loss=0.2723, simple_loss=0.3338, pruned_loss=0.1054, over 17383.00 frames. ], tot_loss[loss=0.2505, simple_loss=0.32, pruned_loss=0.0905, over 2050338.23 frames. ], batch size: 101, lr: 9.54e-03, grad_scale: 16.0 +2023-04-01 13:43:53,898 INFO [train.py:903] (0/4) Epoch 9, batch 200, loss[loss=0.2331, simple_loss=0.3024, pruned_loss=0.08186, over 19847.00 frames. ], tot_loss[loss=0.2514, simple_loss=0.3201, pruned_loss=0.09137, over 2443493.41 frames. ], batch size: 52, lr: 9.54e-03, grad_scale: 8.0 +2023-04-01 13:43:56,317 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 13:44:03,669 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6873, 2.0618, 2.3307, 2.7385, 2.4713, 2.3372, 2.2070, 2.8269], + device='cuda:0'), covar=tensor([0.0689, 0.1618, 0.1180, 0.0796, 0.1155, 0.0419, 0.0972, 0.0492], + device='cuda:0'), in_proj_covar=tensor([0.0247, 0.0355, 0.0290, 0.0239, 0.0297, 0.0244, 0.0269, 0.0234], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:44:36,456 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.193e+02 5.883e+02 7.738e+02 9.204e+02 1.688e+03, threshold=1.548e+03, percent-clipped=2.0 +2023-04-01 13:44:47,188 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6533, 1.5575, 1.5894, 1.5862, 3.1477, 1.1040, 2.1764, 3.5139], + device='cuda:0'), covar=tensor([0.0441, 0.2441, 0.2465, 0.1732, 0.0713, 0.2361, 0.1236, 0.0294], + device='cuda:0'), in_proj_covar=tensor([0.0332, 0.0330, 0.0339, 0.0309, 0.0339, 0.0323, 0.0313, 0.0334], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:44:57,170 INFO [train.py:903] (0/4) Epoch 9, batch 250, loss[loss=0.2198, simple_loss=0.2847, pruned_loss=0.07748, over 19720.00 frames. ], tot_loss[loss=0.2494, simple_loss=0.3183, pruned_loss=0.09024, over 2755074.32 frames. ], batch size: 46, lr: 9.54e-03, grad_scale: 8.0 +2023-04-01 13:45:06,563 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=54880.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:45:39,327 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4040, 2.1970, 1.6614, 1.3776, 1.9761, 1.2474, 1.3560, 1.8358], + device='cuda:0'), covar=tensor([0.0748, 0.0530, 0.0788, 0.0585, 0.0459, 0.0941, 0.0585, 0.0353], + device='cuda:0'), in_proj_covar=tensor([0.0285, 0.0290, 0.0321, 0.0241, 0.0228, 0.0310, 0.0287, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:45:57,845 INFO [train.py:903] (0/4) Epoch 9, batch 300, loss[loss=0.2799, simple_loss=0.3539, pruned_loss=0.1029, over 18771.00 frames. ], tot_loss[loss=0.2488, simple_loss=0.318, pruned_loss=0.08978, over 2988915.34 frames. ], batch size: 74, lr: 9.53e-03, grad_scale: 8.0 +2023-04-01 13:46:39,713 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.328e+02 5.660e+02 7.032e+02 9.457e+02 2.087e+03, threshold=1.406e+03, percent-clipped=3.0 +2023-04-01 13:47:01,351 INFO [train.py:903] (0/4) Epoch 9, batch 350, loss[loss=0.2461, simple_loss=0.3155, pruned_loss=0.08836, over 19682.00 frames. ], tot_loss[loss=0.2472, simple_loss=0.3168, pruned_loss=0.0888, over 3185257.41 frames. ], batch size: 53, lr: 9.53e-03, grad_scale: 8.0 +2023-04-01 13:47:07,236 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 13:48:03,059 INFO [train.py:903] (0/4) Epoch 9, batch 400, loss[loss=0.2455, simple_loss=0.3186, pruned_loss=0.08624, over 19796.00 frames. ], tot_loss[loss=0.2483, simple_loss=0.3178, pruned_loss=0.08938, over 3336028.45 frames. ], batch size: 56, lr: 9.52e-03, grad_scale: 8.0 +2023-04-01 13:48:06,563 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=55026.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 13:48:38,439 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.97 vs. limit=2.0 +2023-04-01 13:48:44,522 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.195e+02 5.719e+02 7.898e+02 1.018e+03 2.327e+03, threshold=1.580e+03, percent-clipped=4.0 +2023-04-01 13:49:04,321 INFO [train.py:903] (0/4) Epoch 9, batch 450, loss[loss=0.2825, simple_loss=0.3463, pruned_loss=0.1094, over 19344.00 frames. ], tot_loss[loss=0.2487, simple_loss=0.318, pruned_loss=0.08971, over 3436635.60 frames. ], batch size: 70, lr: 9.52e-03, grad_scale: 8.0 +2023-04-01 13:49:42,336 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 13:49:43,530 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 13:50:00,608 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3962, 2.1573, 1.5938, 1.4145, 2.0027, 1.1610, 1.2115, 1.7096], + device='cuda:0'), covar=tensor([0.0806, 0.0573, 0.0890, 0.0632, 0.0397, 0.1023, 0.0640, 0.0414], + device='cuda:0'), in_proj_covar=tensor([0.0287, 0.0290, 0.0321, 0.0242, 0.0229, 0.0312, 0.0286, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 13:50:01,631 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2090, 1.1889, 1.6595, 0.9712, 2.5369, 3.2899, 3.0428, 3.4794], + device='cuda:0'), covar=tensor([0.1530, 0.3315, 0.2920, 0.2148, 0.0472, 0.0183, 0.0229, 0.0174], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0288, 0.0317, 0.0249, 0.0209, 0.0141, 0.0206, 0.0176], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 13:50:07,774 INFO [train.py:903] (0/4) Epoch 9, batch 500, loss[loss=0.199, simple_loss=0.2758, pruned_loss=0.06109, over 19421.00 frames. ], tot_loss[loss=0.2497, simple_loss=0.3191, pruned_loss=0.09014, over 3525361.88 frames. ], batch size: 48, lr: 9.51e-03, grad_scale: 8.0 +2023-04-01 13:50:30,907 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55141.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 13:50:47,945 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.663e+02 6.183e+02 7.261e+02 8.870e+02 1.589e+03, threshold=1.452e+03, percent-clipped=1.0 +2023-04-01 13:51:00,576 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7641, 4.2124, 2.7273, 3.6935, 1.2086, 4.0215, 4.0087, 4.2316], + device='cuda:0'), covar=tensor([0.0579, 0.0978, 0.1965, 0.0810, 0.3898, 0.0773, 0.0716, 0.1028], + device='cuda:0'), in_proj_covar=tensor([0.0406, 0.0347, 0.0412, 0.0307, 0.0373, 0.0340, 0.0329, 0.0364], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 13:51:01,627 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55166.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:51:04,366 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-01 13:51:10,560 INFO [train.py:903] (0/4) Epoch 9, batch 550, loss[loss=0.2744, simple_loss=0.3479, pruned_loss=0.1004, over 19696.00 frames. ], tot_loss[loss=0.2497, simple_loss=0.319, pruned_loss=0.09014, over 3593621.64 frames. ], batch size: 59, lr: 9.51e-03, grad_scale: 8.0 +2023-04-01 13:52:14,550 INFO [train.py:903] (0/4) Epoch 9, batch 600, loss[loss=0.2774, simple_loss=0.3371, pruned_loss=0.1088, over 17418.00 frames. ], tot_loss[loss=0.2498, simple_loss=0.3193, pruned_loss=0.09016, over 3645097.06 frames. ], batch size: 101, lr: 9.51e-03, grad_scale: 8.0 +2023-04-01 13:52:15,824 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=55224.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:52:21,725 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55229.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:52:47,918 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3428, 1.3467, 1.6400, 1.2953, 2.4795, 3.0857, 2.9041, 3.2319], + device='cuda:0'), covar=tensor([0.1537, 0.3350, 0.3133, 0.2212, 0.0682, 0.0293, 0.0300, 0.0261], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0290, 0.0319, 0.0250, 0.0211, 0.0142, 0.0208, 0.0177], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 13:52:55,194 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.376e+02 5.838e+02 6.783e+02 8.586e+02 3.812e+03, threshold=1.357e+03, percent-clipped=5.0 +2023-04-01 13:52:57,862 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2546, 3.8044, 3.8924, 3.8476, 1.4844, 3.5870, 3.2426, 3.5895], + device='cuda:0'), covar=tensor([0.1231, 0.0632, 0.0576, 0.0610, 0.4390, 0.0626, 0.0614, 0.1053], + device='cuda:0'), in_proj_covar=tensor([0.0619, 0.0543, 0.0732, 0.0613, 0.0668, 0.0476, 0.0462, 0.0671], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 13:52:58,753 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 13:53:13,356 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2247, 2.1160, 1.7660, 1.6450, 1.5550, 1.7364, 0.3807, 0.9628], + device='cuda:0'), covar=tensor([0.0304, 0.0328, 0.0237, 0.0398, 0.0703, 0.0380, 0.0681, 0.0612], + device='cuda:0'), in_proj_covar=tensor([0.0319, 0.0313, 0.0313, 0.0333, 0.0408, 0.0334, 0.0296, 0.0315], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 13:53:16,399 INFO [train.py:903] (0/4) Epoch 9, batch 650, loss[loss=0.2864, simple_loss=0.3492, pruned_loss=0.1117, over 19450.00 frames. ], tot_loss[loss=0.2488, simple_loss=0.318, pruned_loss=0.08978, over 3685869.05 frames. ], batch size: 70, lr: 9.50e-03, grad_scale: 4.0 +2023-04-01 13:54:19,294 INFO [train.py:903] (0/4) Epoch 9, batch 700, loss[loss=0.2387, simple_loss=0.3006, pruned_loss=0.08837, over 19383.00 frames. ], tot_loss[loss=0.2481, simple_loss=0.3178, pruned_loss=0.08922, over 3720194.94 frames. ], batch size: 47, lr: 9.50e-03, grad_scale: 4.0 +2023-04-01 13:54:41,883 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55339.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:55:02,469 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.468e+02 5.553e+02 6.808e+02 9.255e+02 2.546e+03, threshold=1.362e+03, percent-clipped=4.0 +2023-04-01 13:55:22,970 INFO [train.py:903] (0/4) Epoch 9, batch 750, loss[loss=0.2268, simple_loss=0.2944, pruned_loss=0.0796, over 19754.00 frames. ], tot_loss[loss=0.249, simple_loss=0.3183, pruned_loss=0.08983, over 3736593.00 frames. ], batch size: 48, lr: 9.49e-03, grad_scale: 4.0 +2023-04-01 13:55:53,610 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55397.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 13:56:25,969 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55422.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 13:56:26,675 INFO [train.py:903] (0/4) Epoch 9, batch 800, loss[loss=0.2532, simple_loss=0.325, pruned_loss=0.09074, over 19776.00 frames. ], tot_loss[loss=0.2483, simple_loss=0.3181, pruned_loss=0.08931, over 3765973.51 frames. ], batch size: 56, lr: 9.49e-03, grad_scale: 8.0 +2023-04-01 13:56:41,973 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 13:57:07,881 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.102e+02 5.631e+02 7.179e+02 9.586e+02 1.610e+03, threshold=1.436e+03, percent-clipped=4.0 +2023-04-01 13:57:28,780 INFO [train.py:903] (0/4) Epoch 9, batch 850, loss[loss=0.3247, simple_loss=0.3647, pruned_loss=0.1424, over 13200.00 frames. ], tot_loss[loss=0.2462, simple_loss=0.3163, pruned_loss=0.08805, over 3779019.43 frames. ], batch size: 135, lr: 9.48e-03, grad_scale: 8.0 +2023-04-01 13:58:10,412 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9386, 1.9701, 2.0861, 2.8454, 1.9112, 2.6632, 2.5580, 1.9066], + device='cuda:0'), covar=tensor([0.3121, 0.2661, 0.1248, 0.1407, 0.2831, 0.1120, 0.2638, 0.2322], + device='cuda:0'), in_proj_covar=tensor([0.0743, 0.0756, 0.0627, 0.0872, 0.0751, 0.0658, 0.0774, 0.0677], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 13:58:14,346 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=55510.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:58:22,713 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 13:58:29,605 INFO [train.py:903] (0/4) Epoch 9, batch 900, loss[loss=0.2578, simple_loss=0.3332, pruned_loss=0.09123, over 19681.00 frames. ], tot_loss[loss=0.2472, simple_loss=0.3173, pruned_loss=0.08855, over 3797810.27 frames. ], batch size: 58, lr: 9.48e-03, grad_scale: 8.0 +2023-04-01 13:59:12,471 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.962e+02 5.974e+02 7.143e+02 8.825e+02 2.413e+03, threshold=1.429e+03, percent-clipped=6.0 +2023-04-01 13:59:32,019 INFO [train.py:903] (0/4) Epoch 9, batch 950, loss[loss=0.2901, simple_loss=0.3613, pruned_loss=0.1095, over 19295.00 frames. ], tot_loss[loss=0.2467, simple_loss=0.3168, pruned_loss=0.08832, over 3810386.40 frames. ], batch size: 66, lr: 9.48e-03, grad_scale: 8.0 +2023-04-01 13:59:32,202 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=55573.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 13:59:37,619 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 14:00:01,046 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55595.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:00:32,376 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55620.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:00:35,409 INFO [train.py:903] (0/4) Epoch 9, batch 1000, loss[loss=0.2506, simple_loss=0.3136, pruned_loss=0.09382, over 19855.00 frames. ], tot_loss[loss=0.2481, simple_loss=0.3179, pruned_loss=0.08913, over 3814548.98 frames. ], batch size: 52, lr: 9.47e-03, grad_scale: 8.0 +2023-04-01 14:00:37,956 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55625.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:01:18,081 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.465e+02 5.596e+02 6.834e+02 8.838e+02 1.578e+03, threshold=1.367e+03, percent-clipped=2.0 +2023-04-01 14:01:29,803 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 14:01:39,305 INFO [train.py:903] (0/4) Epoch 9, batch 1050, loss[loss=0.2519, simple_loss=0.3241, pruned_loss=0.08978, over 19368.00 frames. ], tot_loss[loss=0.2506, simple_loss=0.3197, pruned_loss=0.09068, over 3800873.01 frames. ], batch size: 66, lr: 9.47e-03, grad_scale: 8.0 +2023-04-01 14:01:57,352 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=55688.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:02:10,954 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 14:02:42,621 INFO [train.py:903] (0/4) Epoch 9, batch 1100, loss[loss=0.2528, simple_loss=0.3299, pruned_loss=0.08785, over 19579.00 frames. ], tot_loss[loss=0.2499, simple_loss=0.319, pruned_loss=0.09036, over 3796415.37 frames. ], batch size: 61, lr: 9.46e-03, grad_scale: 8.0 +2023-04-01 14:02:54,472 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2926, 1.3372, 1.6378, 1.4343, 2.2377, 2.1346, 2.3057, 0.7409], + device='cuda:0'), covar=tensor([0.1823, 0.3261, 0.1823, 0.1418, 0.1162, 0.1556, 0.1201, 0.3190], + device='cuda:0'), in_proj_covar=tensor([0.0476, 0.0552, 0.0557, 0.0425, 0.0579, 0.0478, 0.0641, 0.0477], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 14:03:25,902 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.690e+02 6.007e+02 7.412e+02 9.315e+02 2.515e+03, threshold=1.482e+03, percent-clipped=6.0 +2023-04-01 14:03:36,838 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55766.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:03:45,596 INFO [train.py:903] (0/4) Epoch 9, batch 1150, loss[loss=0.1871, simple_loss=0.2635, pruned_loss=0.05533, over 18592.00 frames. ], tot_loss[loss=0.2509, simple_loss=0.32, pruned_loss=0.09091, over 3786183.29 frames. ], batch size: 41, lr: 9.46e-03, grad_scale: 8.0 +2023-04-01 14:04:16,530 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-01 14:04:50,174 INFO [train.py:903] (0/4) Epoch 9, batch 1200, loss[loss=0.313, simple_loss=0.3684, pruned_loss=0.1288, over 19607.00 frames. ], tot_loss[loss=0.2505, simple_loss=0.3199, pruned_loss=0.09055, over 3800885.11 frames. ], batch size: 61, lr: 9.45e-03, grad_scale: 8.0 +2023-04-01 14:04:55,602 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-01 14:05:19,005 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 14:05:31,643 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.284e+02 5.926e+02 7.645e+02 1.010e+03 3.329e+03, threshold=1.529e+03, percent-clipped=6.0 +2023-04-01 14:05:44,721 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=55866.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:05:53,923 INFO [train.py:903] (0/4) Epoch 9, batch 1250, loss[loss=0.287, simple_loss=0.3515, pruned_loss=0.1112, over 19329.00 frames. ], tot_loss[loss=0.2504, simple_loss=0.3195, pruned_loss=0.09061, over 3806453.25 frames. ], batch size: 66, lr: 9.45e-03, grad_scale: 8.0 +2023-04-01 14:06:03,230 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55881.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:06:17,962 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6045, 1.3188, 1.6743, 1.2943, 2.7526, 3.7544, 3.5673, 4.0322], + device='cuda:0'), covar=tensor([0.1174, 0.3050, 0.2852, 0.1980, 0.0445, 0.0150, 0.0184, 0.0133], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0286, 0.0315, 0.0248, 0.0207, 0.0141, 0.0206, 0.0176], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 14:06:35,442 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55906.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:06:55,962 INFO [train.py:903] (0/4) Epoch 9, batch 1300, loss[loss=0.2123, simple_loss=0.2946, pruned_loss=0.06501, over 19539.00 frames. ], tot_loss[loss=0.2499, simple_loss=0.3189, pruned_loss=0.09049, over 3803745.90 frames. ], batch size: 56, lr: 9.45e-03, grad_scale: 8.0 +2023-04-01 14:07:07,846 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.94 vs. limit=5.0 +2023-04-01 14:07:23,915 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=55944.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:07:39,624 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.793e+02 5.793e+02 6.909e+02 8.321e+02 2.022e+03, threshold=1.382e+03, percent-clipped=2.0 +2023-04-01 14:07:54,312 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=55969.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:07:58,594 INFO [train.py:903] (0/4) Epoch 9, batch 1350, loss[loss=0.2457, simple_loss=0.3228, pruned_loss=0.08429, over 19647.00 frames. ], tot_loss[loss=0.2484, simple_loss=0.3176, pruned_loss=0.08956, over 3808471.22 frames. ], batch size: 58, lr: 9.44e-03, grad_scale: 8.0 +2023-04-01 14:08:33,446 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-56000.pt +2023-04-01 14:08:50,498 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-01 14:09:02,607 INFO [train.py:903] (0/4) Epoch 9, batch 1400, loss[loss=0.2211, simple_loss=0.3079, pruned_loss=0.06711, over 19078.00 frames. ], tot_loss[loss=0.2486, simple_loss=0.318, pruned_loss=0.08965, over 3815240.61 frames. ], batch size: 69, lr: 9.44e-03, grad_scale: 8.0 +2023-04-01 14:09:47,083 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.547e+02 5.536e+02 7.314e+02 8.995e+02 2.483e+03, threshold=1.463e+03, percent-clipped=9.0 +2023-04-01 14:10:07,183 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 14:10:08,161 INFO [train.py:903] (0/4) Epoch 9, batch 1450, loss[loss=0.2641, simple_loss=0.3308, pruned_loss=0.09871, over 19729.00 frames. ], tot_loss[loss=0.249, simple_loss=0.3181, pruned_loss=0.08995, over 3818585.15 frames. ], batch size: 51, lr: 9.43e-03, grad_scale: 8.0 +2023-04-01 14:10:33,652 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9667, 4.3247, 4.5994, 4.5816, 1.5198, 4.1827, 3.7586, 4.2154], + device='cuda:0'), covar=tensor([0.1196, 0.0680, 0.0522, 0.0493, 0.5028, 0.0546, 0.0624, 0.1029], + device='cuda:0'), in_proj_covar=tensor([0.0615, 0.0543, 0.0739, 0.0621, 0.0680, 0.0486, 0.0467, 0.0675], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 14:10:55,227 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=56110.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:11:11,143 INFO [train.py:903] (0/4) Epoch 9, batch 1500, loss[loss=0.264, simple_loss=0.3403, pruned_loss=0.09382, over 19676.00 frames. ], tot_loss[loss=0.2487, simple_loss=0.3178, pruned_loss=0.08981, over 3834083.05 frames. ], batch size: 55, lr: 9.43e-03, grad_scale: 8.0 +2023-04-01 14:11:29,349 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8267, 1.3171, 1.4181, 1.5967, 3.2806, 0.9442, 2.2205, 3.4938], + device='cuda:0'), covar=tensor([0.0375, 0.2747, 0.2768, 0.1684, 0.0745, 0.2664, 0.1250, 0.0337], + device='cuda:0'), in_proj_covar=tensor([0.0329, 0.0328, 0.0339, 0.0308, 0.0333, 0.0325, 0.0315, 0.0334], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 14:11:52,401 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.776e+02 6.008e+02 7.164e+02 9.066e+02 2.093e+03, threshold=1.433e+03, percent-clipped=3.0 +2023-04-01 14:12:01,104 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56163.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:12:12,299 INFO [train.py:903] (0/4) Epoch 9, batch 1550, loss[loss=0.2258, simple_loss=0.3129, pruned_loss=0.06933, over 19775.00 frames. ], tot_loss[loss=0.2514, simple_loss=0.3202, pruned_loss=0.09134, over 3824422.48 frames. ], batch size: 56, lr: 9.43e-03, grad_scale: 8.0 +2023-04-01 14:12:26,264 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7393, 4.2322, 4.4163, 4.3976, 1.5633, 4.1318, 3.5724, 4.0535], + device='cuda:0'), covar=tensor([0.1281, 0.0638, 0.0525, 0.0522, 0.5049, 0.0530, 0.0631, 0.0995], + device='cuda:0'), in_proj_covar=tensor([0.0608, 0.0538, 0.0731, 0.0615, 0.0678, 0.0479, 0.0460, 0.0671], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0001, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 14:12:59,851 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=56210.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:13:15,311 INFO [train.py:903] (0/4) Epoch 9, batch 1600, loss[loss=0.2815, simple_loss=0.348, pruned_loss=0.1075, over 18064.00 frames. ], tot_loss[loss=0.2503, simple_loss=0.3194, pruned_loss=0.09057, over 3827840.75 frames. ], batch size: 83, lr: 9.42e-03, grad_scale: 8.0 +2023-04-01 14:13:18,936 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=56225.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:13:41,614 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 14:13:59,884 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.177e+02 5.781e+02 6.769e+02 8.617e+02 2.222e+03, threshold=1.354e+03, percent-clipped=2.0 +2023-04-01 14:14:19,016 INFO [train.py:903] (0/4) Epoch 9, batch 1650, loss[loss=0.363, simple_loss=0.3912, pruned_loss=0.1674, over 13009.00 frames. ], tot_loss[loss=0.2502, simple_loss=0.319, pruned_loss=0.09068, over 3799348.09 frames. ], batch size: 136, lr: 9.42e-03, grad_scale: 4.0 +2023-04-01 14:14:21,806 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6393, 1.6658, 1.8070, 1.7299, 4.0809, 1.0362, 2.2956, 4.2888], + device='cuda:0'), covar=tensor([0.0349, 0.2449, 0.2429, 0.1685, 0.0653, 0.2639, 0.1421, 0.0219], + device='cuda:0'), in_proj_covar=tensor([0.0324, 0.0321, 0.0335, 0.0301, 0.0328, 0.0322, 0.0311, 0.0330], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 14:15:20,633 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1609, 2.0700, 1.7288, 1.6105, 1.4940, 1.6804, 0.3499, 1.0008], + device='cuda:0'), covar=tensor([0.0360, 0.0399, 0.0297, 0.0448, 0.0845, 0.0459, 0.0747, 0.0701], + device='cuda:0'), in_proj_covar=tensor([0.0317, 0.0317, 0.0316, 0.0333, 0.0409, 0.0335, 0.0295, 0.0317], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 14:15:22,541 INFO [train.py:903] (0/4) Epoch 9, batch 1700, loss[loss=0.2302, simple_loss=0.302, pruned_loss=0.0792, over 19480.00 frames. ], tot_loss[loss=0.2486, simple_loss=0.3179, pruned_loss=0.0897, over 3802636.48 frames. ], batch size: 49, lr: 9.41e-03, grad_scale: 4.0 +2023-04-01 14:15:25,173 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=56325.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:16:02,378 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 14:16:05,921 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.642e+02 5.619e+02 6.768e+02 8.904e+02 2.101e+03, threshold=1.354e+03, percent-clipped=6.0 +2023-04-01 14:16:24,627 INFO [train.py:903] (0/4) Epoch 9, batch 1750, loss[loss=0.2459, simple_loss=0.3167, pruned_loss=0.08756, over 19588.00 frames. ], tot_loss[loss=0.2483, simple_loss=0.3177, pruned_loss=0.08943, over 3812679.59 frames. ], batch size: 52, lr: 9.41e-03, grad_scale: 4.0 +2023-04-01 14:17:24,011 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 14:17:26,706 INFO [train.py:903] (0/4) Epoch 9, batch 1800, loss[loss=0.2306, simple_loss=0.304, pruned_loss=0.07856, over 19699.00 frames. ], tot_loss[loss=0.2494, simple_loss=0.3187, pruned_loss=0.09008, over 3807014.92 frames. ], batch size: 53, lr: 9.40e-03, grad_scale: 4.0 +2023-04-01 14:18:09,955 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.374e+02 5.839e+02 7.002e+02 8.564e+02 1.629e+03, threshold=1.400e+03, percent-clipped=1.0 +2023-04-01 14:18:25,592 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 14:18:29,985 INFO [train.py:903] (0/4) Epoch 9, batch 1850, loss[loss=0.2745, simple_loss=0.3417, pruned_loss=0.1036, over 19778.00 frames. ], tot_loss[loss=0.2484, simple_loss=0.318, pruned_loss=0.08933, over 3806344.92 frames. ], batch size: 54, lr: 9.40e-03, grad_scale: 4.0 +2023-04-01 14:18:40,529 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56481.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:19:02,326 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 14:19:10,212 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56506.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:19:12,018 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=56507.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:19:32,999 INFO [train.py:903] (0/4) Epoch 9, batch 1900, loss[loss=0.2685, simple_loss=0.3335, pruned_loss=0.1017, over 19600.00 frames. ], tot_loss[loss=0.2479, simple_loss=0.3176, pruned_loss=0.08909, over 3801894.86 frames. ], batch size: 61, lr: 9.40e-03, grad_scale: 4.0 +2023-04-01 14:19:34,870 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.34 vs. limit=5.0 +2023-04-01 14:19:45,165 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3160, 2.2217, 1.8405, 1.7076, 1.6559, 1.7373, 0.3368, 1.0608], + device='cuda:0'), covar=tensor([0.0290, 0.0324, 0.0287, 0.0396, 0.0646, 0.0479, 0.0762, 0.0611], + device='cuda:0'), in_proj_covar=tensor([0.0315, 0.0314, 0.0311, 0.0332, 0.0405, 0.0331, 0.0292, 0.0314], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 14:19:48,396 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 14:19:54,885 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 14:19:55,172 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56541.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 14:20:16,633 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.412e+02 5.665e+02 6.760e+02 8.403e+02 1.758e+03, threshold=1.352e+03, percent-clipped=2.0 +2023-04-01 14:20:19,007 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 14:20:35,965 INFO [train.py:903] (0/4) Epoch 9, batch 1950, loss[loss=0.2074, simple_loss=0.2773, pruned_loss=0.06878, over 19476.00 frames. ], tot_loss[loss=0.2478, simple_loss=0.3177, pruned_loss=0.08892, over 3813467.00 frames. ], batch size: 49, lr: 9.39e-03, grad_scale: 4.0 +2023-04-01 14:20:46,754 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56581.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:21:18,991 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56606.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:21:22,317 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7621, 1.4107, 1.9054, 1.6848, 4.1855, 0.8221, 2.3487, 4.4082], + device='cuda:0'), covar=tensor([0.0353, 0.2559, 0.2504, 0.1657, 0.0710, 0.2774, 0.1414, 0.0250], + device='cuda:0'), in_proj_covar=tensor([0.0325, 0.0322, 0.0335, 0.0301, 0.0331, 0.0319, 0.0310, 0.0329], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 14:21:38,550 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=56622.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:21:39,347 INFO [train.py:903] (0/4) Epoch 9, batch 2000, loss[loss=0.2105, simple_loss=0.272, pruned_loss=0.07452, over 19717.00 frames. ], tot_loss[loss=0.2457, simple_loss=0.3162, pruned_loss=0.08756, over 3823197.60 frames. ], batch size: 46, lr: 9.39e-03, grad_scale: 8.0 +2023-04-01 14:21:41,563 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.95 vs. limit=5.0 +2023-04-01 14:22:22,879 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.621e+02 5.506e+02 7.081e+02 9.090e+02 3.144e+03, threshold=1.416e+03, percent-clipped=7.0 +2023-04-01 14:22:36,118 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 14:22:42,524 INFO [train.py:903] (0/4) Epoch 9, batch 2050, loss[loss=0.2323, simple_loss=0.3041, pruned_loss=0.08018, over 19561.00 frames. ], tot_loss[loss=0.2474, simple_loss=0.317, pruned_loss=0.08889, over 3793298.77 frames. ], batch size: 52, lr: 9.38e-03, grad_scale: 8.0 +2023-04-01 14:22:56,339 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 14:22:57,521 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 14:23:19,236 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 14:23:44,278 INFO [train.py:903] (0/4) Epoch 9, batch 2100, loss[loss=0.2633, simple_loss=0.3403, pruned_loss=0.09318, over 19313.00 frames. ], tot_loss[loss=0.2477, simple_loss=0.3173, pruned_loss=0.08911, over 3805234.24 frames. ], batch size: 66, lr: 9.38e-03, grad_scale: 8.0 +2023-04-01 14:24:12,024 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 14:24:29,325 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.187e+02 5.573e+02 6.906e+02 8.990e+02 1.566e+03, threshold=1.381e+03, percent-clipped=3.0 +2023-04-01 14:24:35,337 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 14:24:48,316 INFO [train.py:903] (0/4) Epoch 9, batch 2150, loss[loss=0.2298, simple_loss=0.3059, pruned_loss=0.07686, over 19690.00 frames. ], tot_loss[loss=0.2471, simple_loss=0.3169, pruned_loss=0.08871, over 3814966.00 frames. ], batch size: 55, lr: 9.38e-03, grad_scale: 8.0 +2023-04-01 14:25:48,735 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56821.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:25:50,905 INFO [train.py:903] (0/4) Epoch 9, batch 2200, loss[loss=0.2575, simple_loss=0.3308, pruned_loss=0.09207, over 19299.00 frames. ], tot_loss[loss=0.2472, simple_loss=0.317, pruned_loss=0.08865, over 3814714.40 frames. ], batch size: 70, lr: 9.37e-03, grad_scale: 8.0 +2023-04-01 14:26:36,144 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.021e+02 5.854e+02 7.300e+02 9.690e+02 2.298e+03, threshold=1.460e+03, percent-clipped=5.0 +2023-04-01 14:26:46,346 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.53 vs. limit=5.0 +2023-04-01 14:26:57,258 INFO [train.py:903] (0/4) Epoch 9, batch 2250, loss[loss=0.2457, simple_loss=0.3027, pruned_loss=0.09428, over 19394.00 frames. ], tot_loss[loss=0.2473, simple_loss=0.3166, pruned_loss=0.089, over 3810859.83 frames. ], batch size: 48, lr: 9.37e-03, grad_scale: 8.0 +2023-04-01 14:27:03,664 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=56878.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:27:11,764 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=56885.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 14:27:34,064 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=56903.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:27:49,936 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3308, 3.9545, 2.6246, 3.5173, 1.1669, 3.6842, 3.7502, 3.7874], + device='cuda:0'), covar=tensor([0.0699, 0.1061, 0.1933, 0.0757, 0.3575, 0.0786, 0.0708, 0.1056], + device='cuda:0'), in_proj_covar=tensor([0.0402, 0.0343, 0.0408, 0.0299, 0.0369, 0.0336, 0.0327, 0.0363], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 14:28:00,647 INFO [train.py:903] (0/4) Epoch 9, batch 2300, loss[loss=0.2368, simple_loss=0.315, pruned_loss=0.07934, over 19547.00 frames. ], tot_loss[loss=0.2462, simple_loss=0.3159, pruned_loss=0.08826, over 3805294.34 frames. ], batch size: 56, lr: 9.36e-03, grad_scale: 8.0 +2023-04-01 14:28:05,878 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=56926.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:28:13,784 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 14:28:22,612 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9737, 2.0912, 1.9292, 1.9565, 4.3951, 1.0600, 2.3577, 4.7625], + device='cuda:0'), covar=tensor([0.0320, 0.2387, 0.2560, 0.1646, 0.0655, 0.2746, 0.1448, 0.0228], + device='cuda:0'), in_proj_covar=tensor([0.0332, 0.0325, 0.0341, 0.0304, 0.0334, 0.0322, 0.0312, 0.0333], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 14:28:46,769 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.182e+02 5.809e+02 7.207e+02 9.233e+02 1.673e+03, threshold=1.441e+03, percent-clipped=4.0 +2023-04-01 14:29:05,085 INFO [train.py:903] (0/4) Epoch 9, batch 2350, loss[loss=0.2579, simple_loss=0.3468, pruned_loss=0.08449, over 19670.00 frames. ], tot_loss[loss=0.2452, simple_loss=0.3154, pruned_loss=0.08753, over 3813269.83 frames. ], batch size: 55, lr: 9.36e-03, grad_scale: 8.0 +2023-04-01 14:29:40,063 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57000.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 14:29:46,435 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 14:29:53,601 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57011.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:29:55,742 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57013.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:30:01,352 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 14:30:07,034 INFO [train.py:903] (0/4) Epoch 9, batch 2400, loss[loss=0.2254, simple_loss=0.2955, pruned_loss=0.07763, over 19828.00 frames. ], tot_loss[loss=0.2469, simple_loss=0.3165, pruned_loss=0.08863, over 3806512.14 frames. ], batch size: 52, lr: 9.36e-03, grad_scale: 8.0 +2023-04-01 14:30:46,603 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-01 14:30:51,617 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.434e+02 5.803e+02 7.402e+02 8.988e+02 1.700e+03, threshold=1.480e+03, percent-clipped=2.0 +2023-04-01 14:31:11,540 INFO [train.py:903] (0/4) Epoch 9, batch 2450, loss[loss=0.2846, simple_loss=0.3426, pruned_loss=0.1133, over 13565.00 frames. ], tot_loss[loss=0.247, simple_loss=0.3164, pruned_loss=0.08874, over 3801453.23 frames. ], batch size: 135, lr: 9.35e-03, grad_scale: 8.0 +2023-04-01 14:31:32,269 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3325, 2.0800, 1.8515, 1.6979, 1.6301, 1.8284, 0.3272, 1.1781], + device='cuda:0'), covar=tensor([0.0361, 0.0368, 0.0320, 0.0502, 0.0772, 0.0486, 0.0796, 0.0630], + device='cuda:0'), in_proj_covar=tensor([0.0322, 0.0321, 0.0319, 0.0335, 0.0416, 0.0335, 0.0296, 0.0317], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 14:31:59,444 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2121, 1.3816, 1.8908, 1.6300, 3.1572, 4.6309, 4.5462, 4.9907], + device='cuda:0'), covar=tensor([0.1550, 0.3096, 0.2799, 0.1817, 0.0415, 0.0137, 0.0128, 0.0086], + device='cuda:0'), in_proj_covar=tensor([0.0248, 0.0285, 0.0316, 0.0245, 0.0205, 0.0143, 0.0203, 0.0177], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 14:32:09,327 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.01 vs. limit=5.0 +2023-04-01 14:32:15,744 INFO [train.py:903] (0/4) Epoch 9, batch 2500, loss[loss=0.2143, simple_loss=0.2927, pruned_loss=0.0679, over 19662.00 frames. ], tot_loss[loss=0.2475, simple_loss=0.3171, pruned_loss=0.08898, over 3811082.28 frames. ], batch size: 53, lr: 9.35e-03, grad_scale: 8.0 +2023-04-01 14:32:17,608 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 14:33:00,908 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.384e+02 5.325e+02 6.954e+02 9.918e+02 1.981e+03, threshold=1.391e+03, percent-clipped=3.0 +2023-04-01 14:33:09,578 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57165.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:33:19,761 INFO [train.py:903] (0/4) Epoch 9, batch 2550, loss[loss=0.1913, simple_loss=0.2599, pruned_loss=0.06134, over 18661.00 frames. ], tot_loss[loss=0.2467, simple_loss=0.3164, pruned_loss=0.08849, over 3813615.54 frames. ], batch size: 41, lr: 9.34e-03, grad_scale: 8.0 +2023-04-01 14:33:59,329 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57204.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:34:13,711 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57215.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:34:15,579 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 14:34:22,474 INFO [train.py:903] (0/4) Epoch 9, batch 2600, loss[loss=0.2556, simple_loss=0.3244, pruned_loss=0.09347, over 19763.00 frames. ], tot_loss[loss=0.2478, simple_loss=0.3175, pruned_loss=0.08908, over 3819799.05 frames. ], batch size: 54, lr: 9.34e-03, grad_scale: 8.0 +2023-04-01 14:35:05,258 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57256.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 14:35:07,156 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.285e+02 5.816e+02 6.932e+02 7.774e+02 1.592e+03, threshold=1.386e+03, percent-clipped=3.0 +2023-04-01 14:35:11,174 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9833, 1.9191, 1.6983, 1.4932, 1.3490, 1.4520, 0.3536, 0.7807], + device='cuda:0'), covar=tensor([0.0360, 0.0352, 0.0241, 0.0353, 0.0799, 0.0430, 0.0673, 0.0663], + device='cuda:0'), in_proj_covar=tensor([0.0317, 0.0318, 0.0316, 0.0331, 0.0413, 0.0333, 0.0295, 0.0316], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 14:35:12,154 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57262.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:35:22,400 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57270.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:35:25,692 INFO [train.py:903] (0/4) Epoch 9, batch 2650, loss[loss=0.354, simple_loss=0.3917, pruned_loss=0.1581, over 13558.00 frames. ], tot_loss[loss=0.248, simple_loss=0.3178, pruned_loss=0.08913, over 3818283.65 frames. ], batch size: 135, lr: 9.34e-03, grad_scale: 8.0 +2023-04-01 14:35:36,443 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57280.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:35:37,701 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57281.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 14:35:45,902 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 14:36:31,185 INFO [train.py:903] (0/4) Epoch 9, batch 2700, loss[loss=0.247, simple_loss=0.3089, pruned_loss=0.09255, over 19785.00 frames. ], tot_loss[loss=0.2482, simple_loss=0.3177, pruned_loss=0.08931, over 3814027.31 frames. ], batch size: 49, lr: 9.33e-03, grad_scale: 8.0 +2023-04-01 14:37:12,070 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57355.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:37:14,460 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57357.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:37:15,518 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.587e+02 5.592e+02 7.209e+02 8.892e+02 3.755e+03, threshold=1.442e+03, percent-clipped=7.0 +2023-04-01 14:37:33,721 INFO [train.py:903] (0/4) Epoch 9, batch 2750, loss[loss=0.2409, simple_loss=0.3075, pruned_loss=0.0872, over 19623.00 frames. ], tot_loss[loss=0.2481, simple_loss=0.3176, pruned_loss=0.08935, over 3805133.71 frames. ], batch size: 50, lr: 9.33e-03, grad_scale: 8.0 +2023-04-01 14:37:49,356 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57385.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:38:30,625 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57416.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:38:38,520 INFO [train.py:903] (0/4) Epoch 9, batch 2800, loss[loss=0.2514, simple_loss=0.3283, pruned_loss=0.08727, over 19576.00 frames. ], tot_loss[loss=0.2459, simple_loss=0.3156, pruned_loss=0.08807, over 3806582.36 frames. ], batch size: 61, lr: 9.32e-03, grad_scale: 8.0 +2023-04-01 14:39:01,670 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5894, 1.2599, 1.2067, 1.4362, 1.2467, 1.3771, 1.1751, 1.3978], + device='cuda:0'), covar=tensor([0.0904, 0.1072, 0.1332, 0.0843, 0.1024, 0.0546, 0.1170, 0.0705], + device='cuda:0'), in_proj_covar=tensor([0.0249, 0.0352, 0.0288, 0.0237, 0.0297, 0.0243, 0.0270, 0.0231], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 14:39:11,834 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9973, 1.9321, 1.6569, 1.5242, 1.3637, 1.5247, 0.2789, 0.7960], + device='cuda:0'), covar=tensor([0.0311, 0.0355, 0.0242, 0.0390, 0.0775, 0.0441, 0.0733, 0.0646], + device='cuda:0'), in_proj_covar=tensor([0.0316, 0.0318, 0.0315, 0.0330, 0.0412, 0.0333, 0.0293, 0.0316], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 14:39:23,160 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.000e+02 5.525e+02 6.599e+02 8.446e+02 1.316e+03, threshold=1.320e+03, percent-clipped=0.0 +2023-04-01 14:39:38,392 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57470.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:39:40,784 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57472.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:39:42,543 INFO [train.py:903] (0/4) Epoch 9, batch 2850, loss[loss=0.231, simple_loss=0.3038, pruned_loss=0.07917, over 19617.00 frames. ], tot_loss[loss=0.245, simple_loss=0.315, pruned_loss=0.08743, over 3815584.55 frames. ], batch size: 50, lr: 9.32e-03, grad_scale: 8.0 +2023-04-01 14:39:51,249 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57480.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:40:33,749 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57513.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:40:37,262 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9616, 4.2555, 4.5727, 4.5702, 1.5878, 4.2930, 3.7380, 4.2295], + device='cuda:0'), covar=tensor([0.1045, 0.0782, 0.0512, 0.0460, 0.4900, 0.0489, 0.0580, 0.0962], + device='cuda:0'), in_proj_covar=tensor([0.0618, 0.0546, 0.0734, 0.0612, 0.0681, 0.0485, 0.0463, 0.0681], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 14:40:41,921 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 14:40:46,553 INFO [train.py:903] (0/4) Epoch 9, batch 2900, loss[loss=0.2743, simple_loss=0.3421, pruned_loss=0.1033, over 19285.00 frames. ], tot_loss[loss=0.2463, simple_loss=0.3161, pruned_loss=0.08827, over 3831751.65 frames. ], batch size: 66, lr: 9.32e-03, grad_scale: 8.0 +2023-04-01 14:41:03,475 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57536.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:41:16,926 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57548.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:41:30,791 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.787e+02 5.829e+02 7.475e+02 8.920e+02 2.516e+03, threshold=1.495e+03, percent-clipped=6.0 +2023-04-01 14:41:32,218 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57559.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:41:34,862 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57561.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:41:45,174 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5333, 2.3409, 1.6185, 1.5827, 2.1214, 1.2251, 1.2371, 1.7904], + device='cuda:0'), covar=tensor([0.0824, 0.0532, 0.0920, 0.0668, 0.0418, 0.1031, 0.0695, 0.0442], + device='cuda:0'), in_proj_covar=tensor([0.0287, 0.0291, 0.0320, 0.0239, 0.0230, 0.0316, 0.0287, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 14:41:49,407 INFO [train.py:903] (0/4) Epoch 9, batch 2950, loss[loss=0.2253, simple_loss=0.2979, pruned_loss=0.07636, over 19615.00 frames. ], tot_loss[loss=0.2461, simple_loss=0.3159, pruned_loss=0.08812, over 3830173.48 frames. ], batch size: 50, lr: 9.31e-03, grad_scale: 8.0 +2023-04-01 14:42:32,112 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57606.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:42:53,540 INFO [train.py:903] (0/4) Epoch 9, batch 3000, loss[loss=0.2761, simple_loss=0.3435, pruned_loss=0.1044, over 18790.00 frames. ], tot_loss[loss=0.2471, simple_loss=0.3171, pruned_loss=0.08858, over 3833178.54 frames. ], batch size: 74, lr: 9.31e-03, grad_scale: 8.0 +2023-04-01 14:42:53,541 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 14:43:06,212 INFO [train.py:937] (0/4) Epoch 9, validation: loss=0.1831, simple_loss=0.2838, pruned_loss=0.04122, over 944034.00 frames. +2023-04-01 14:43:06,213 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 14:43:08,521 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 14:43:28,364 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57641.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:43:50,440 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.712e+02 6.061e+02 7.914e+02 9.800e+02 2.087e+03, threshold=1.583e+03, percent-clipped=4.0 +2023-04-01 14:43:56,627 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57663.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:44:00,204 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57666.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:44:08,767 INFO [train.py:903] (0/4) Epoch 9, batch 3050, loss[loss=0.2529, simple_loss=0.3053, pruned_loss=0.1003, over 19758.00 frames. ], tot_loss[loss=0.2469, simple_loss=0.3169, pruned_loss=0.0885, over 3838069.73 frames. ], batch size: 47, lr: 9.30e-03, grad_scale: 8.0 +2023-04-01 14:44:10,222 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57674.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:44:10,359 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8256, 1.9021, 1.9843, 2.8572, 1.8868, 2.5277, 2.3449, 1.7805], + device='cuda:0'), covar=tensor([0.3404, 0.2842, 0.1361, 0.1637, 0.3265, 0.1350, 0.3138, 0.2602], + device='cuda:0'), in_proj_covar=tensor([0.0746, 0.0759, 0.0623, 0.0872, 0.0748, 0.0663, 0.0770, 0.0675], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 14:45:09,113 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57721.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:45:11,162 INFO [train.py:903] (0/4) Epoch 9, batch 3100, loss[loss=0.2279, simple_loss=0.2884, pruned_loss=0.08373, over 19763.00 frames. ], tot_loss[loss=0.2458, simple_loss=0.3161, pruned_loss=0.08777, over 3836410.49 frames. ], batch size: 47, lr: 9.30e-03, grad_scale: 8.0 +2023-04-01 14:45:16,040 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57726.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:45:18,345 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57728.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:45:47,443 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=57751.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:45:47,546 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57751.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:45:49,783 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57753.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:45:54,924 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.860e+02 5.906e+02 7.408e+02 1.029e+03 2.368e+03, threshold=1.482e+03, percent-clipped=3.0 +2023-04-01 14:45:57,424 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57760.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:46:14,158 INFO [train.py:903] (0/4) Epoch 9, batch 3150, loss[loss=0.2323, simple_loss=0.3186, pruned_loss=0.073, over 19655.00 frames. ], tot_loss[loss=0.2442, simple_loss=0.3147, pruned_loss=0.08684, over 3844818.68 frames. ], batch size: 58, lr: 9.30e-03, grad_scale: 8.0 +2023-04-01 14:46:42,120 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 14:47:14,703 INFO [train.py:903] (0/4) Epoch 9, batch 3200, loss[loss=0.2428, simple_loss=0.3113, pruned_loss=0.08709, over 19780.00 frames. ], tot_loss[loss=0.2446, simple_loss=0.3149, pruned_loss=0.08711, over 3855103.13 frames. ], batch size: 54, lr: 9.29e-03, grad_scale: 8.0 +2023-04-01 14:47:15,996 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57824.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:47:55,929 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=57857.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:47:56,961 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.598e+02 5.791e+02 7.100e+02 9.261e+02 4.038e+03, threshold=1.420e+03, percent-clipped=7.0 +2023-04-01 14:48:14,915 INFO [train.py:903] (0/4) Epoch 9, batch 3250, loss[loss=0.3447, simple_loss=0.386, pruned_loss=0.1517, over 13715.00 frames. ], tot_loss[loss=0.2461, simple_loss=0.3164, pruned_loss=0.08791, over 3835449.25 frames. ], batch size: 136, lr: 9.29e-03, grad_scale: 8.0 +2023-04-01 14:48:18,561 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57875.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:49:12,232 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57919.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:49:16,425 INFO [train.py:903] (0/4) Epoch 9, batch 3300, loss[loss=0.2632, simple_loss=0.333, pruned_loss=0.09668, over 19504.00 frames. ], tot_loss[loss=0.2449, simple_loss=0.3156, pruned_loss=0.08714, over 3834217.00 frames. ], batch size: 64, lr: 9.28e-03, grad_scale: 8.0 +2023-04-01 14:49:23,979 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 14:49:25,456 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57930.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:49:36,421 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57939.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:49:43,152 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57944.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:49:56,563 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=57955.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:49:59,470 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.480e+02 5.470e+02 6.783e+02 8.234e+02 1.579e+03, threshold=1.357e+03, percent-clipped=1.0 +2023-04-01 14:50:17,131 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=57972.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:50:17,931 INFO [train.py:903] (0/4) Epoch 9, batch 3350, loss[loss=0.2585, simple_loss=0.331, pruned_loss=0.09301, over 19618.00 frames. ], tot_loss[loss=0.2468, simple_loss=0.3171, pruned_loss=0.08825, over 3834085.96 frames. ], batch size: 61, lr: 9.28e-03, grad_scale: 8.0 +2023-04-01 14:50:22,855 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=57977.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:50:24,996 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.6508, 5.0819, 2.8263, 4.4547, 1.4122, 4.7503, 4.9263, 5.0677], + device='cuda:0'), covar=tensor([0.0402, 0.0829, 0.1994, 0.0616, 0.3558, 0.0707, 0.0611, 0.0920], + device='cuda:0'), in_proj_covar=tensor([0.0411, 0.0346, 0.0414, 0.0312, 0.0367, 0.0341, 0.0331, 0.0368], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 14:50:34,088 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8758, 2.7881, 1.7529, 1.8153, 1.6892, 2.0403, 0.9245, 1.8950], + device='cuda:0'), covar=tensor([0.0655, 0.0521, 0.0608, 0.0994, 0.1035, 0.1125, 0.0959, 0.0902], + device='cuda:0'), in_proj_covar=tensor([0.0315, 0.0321, 0.0317, 0.0334, 0.0408, 0.0335, 0.0297, 0.0318], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 14:50:50,894 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-58000.pt +2023-04-01 14:50:54,314 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58002.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:51:10,709 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.77 vs. limit=5.0 +2023-04-01 14:51:20,068 INFO [train.py:903] (0/4) Epoch 9, batch 3400, loss[loss=0.2546, simple_loss=0.3256, pruned_loss=0.09184, over 17489.00 frames. ], tot_loss[loss=0.2471, simple_loss=0.3173, pruned_loss=0.08848, over 3842500.26 frames. ], batch size: 101, lr: 9.28e-03, grad_scale: 8.0 +2023-04-01 14:51:35,092 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-01 14:52:02,957 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.171e+02 5.309e+02 6.592e+02 8.930e+02 1.711e+03, threshold=1.318e+03, percent-clipped=4.0 +2023-04-01 14:52:20,108 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7234, 1.4343, 1.4288, 2.2292, 1.8005, 2.1667, 2.2027, 1.8404], + device='cuda:0'), covar=tensor([0.0845, 0.1007, 0.1070, 0.0784, 0.0844, 0.0681, 0.0878, 0.0642], + device='cuda:0'), in_proj_covar=tensor([0.0217, 0.0232, 0.0229, 0.0256, 0.0243, 0.0216, 0.0204, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 14:52:20,816 INFO [train.py:903] (0/4) Epoch 9, batch 3450, loss[loss=0.262, simple_loss=0.3216, pruned_loss=0.1012, over 19386.00 frames. ], tot_loss[loss=0.2481, simple_loss=0.3184, pruned_loss=0.08891, over 3840396.91 frames. ], batch size: 48, lr: 9.27e-03, grad_scale: 8.0 +2023-04-01 14:52:25,124 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 14:52:48,534 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=58095.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:53:22,370 INFO [train.py:903] (0/4) Epoch 9, batch 3500, loss[loss=0.2946, simple_loss=0.3505, pruned_loss=0.1193, over 18643.00 frames. ], tot_loss[loss=0.2472, simple_loss=0.3175, pruned_loss=0.08848, over 3831508.95 frames. ], batch size: 74, lr: 9.27e-03, grad_scale: 8.0 +2023-04-01 14:53:25,170 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9497, 4.3835, 4.6906, 4.6725, 1.5725, 4.3035, 3.8056, 4.2933], + device='cuda:0'), covar=tensor([0.1197, 0.0704, 0.0522, 0.0475, 0.5225, 0.0487, 0.0560, 0.1031], + device='cuda:0'), in_proj_covar=tensor([0.0624, 0.0550, 0.0750, 0.0620, 0.0686, 0.0491, 0.0467, 0.0685], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 14:53:33,956 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58131.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:53:36,489 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-01 14:54:03,568 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58156.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:54:05,462 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.510e+02 6.015e+02 7.407e+02 9.414e+02 2.837e+03, threshold=1.481e+03, percent-clipped=3.0 +2023-04-01 14:54:06,979 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58159.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:54:08,244 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2292, 1.2020, 1.4066, 1.3634, 1.8790, 1.7930, 1.8310, 0.5052], + device='cuda:0'), covar=tensor([0.2033, 0.3498, 0.2121, 0.1616, 0.1187, 0.1830, 0.1196, 0.3418], + device='cuda:0'), in_proj_covar=tensor([0.0468, 0.0548, 0.0558, 0.0422, 0.0580, 0.0473, 0.0638, 0.0474], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 14:54:10,314 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58162.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:54:24,272 INFO [train.py:903] (0/4) Epoch 9, batch 3550, loss[loss=0.2166, simple_loss=0.2867, pruned_loss=0.07327, over 19499.00 frames. ], tot_loss[loss=0.2466, simple_loss=0.3171, pruned_loss=0.08811, over 3837666.04 frames. ], batch size: 49, lr: 9.26e-03, grad_scale: 8.0 +2023-04-01 14:54:37,410 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0148, 2.0679, 1.7159, 1.5354, 1.3182, 1.5451, 0.4324, 0.8775], + device='cuda:0'), covar=tensor([0.0601, 0.0512, 0.0367, 0.0577, 0.1011, 0.0658, 0.0855, 0.0872], + device='cuda:0'), in_proj_covar=tensor([0.0318, 0.0322, 0.0319, 0.0337, 0.0409, 0.0334, 0.0299, 0.0316], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 14:54:50,570 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58195.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:55:08,378 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58210.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:55:20,402 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58220.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:55:24,266 INFO [train.py:903] (0/4) Epoch 9, batch 3600, loss[loss=0.2729, simple_loss=0.3449, pruned_loss=0.1005, over 19783.00 frames. ], tot_loss[loss=0.2492, simple_loss=0.3191, pruned_loss=0.08963, over 3830008.84 frames. ], batch size: 56, lr: 9.26e-03, grad_scale: 8.0 +2023-04-01 14:55:30,332 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58228.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:56:01,038 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58253.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 14:56:06,153 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.641e+02 5.487e+02 6.500e+02 7.997e+02 2.924e+03, threshold=1.300e+03, percent-clipped=2.0 +2023-04-01 14:56:23,768 INFO [train.py:903] (0/4) Epoch 9, batch 3650, loss[loss=0.2572, simple_loss=0.3262, pruned_loss=0.09412, over 19362.00 frames. ], tot_loss[loss=0.2479, simple_loss=0.3179, pruned_loss=0.08893, over 3829797.39 frames. ], batch size: 66, lr: 9.26e-03, grad_scale: 16.0 +2023-04-01 14:57:24,454 INFO [train.py:903] (0/4) Epoch 9, batch 3700, loss[loss=0.2085, simple_loss=0.2837, pruned_loss=0.06662, over 19716.00 frames. ], tot_loss[loss=0.247, simple_loss=0.3173, pruned_loss=0.08833, over 3838350.19 frames. ], batch size: 46, lr: 9.25e-03, grad_scale: 8.0 +2023-04-01 14:57:59,834 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0 +2023-04-01 14:58:07,749 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.190e+02 6.060e+02 7.736e+02 9.843e+02 2.060e+03, threshold=1.547e+03, percent-clipped=9.0 +2023-04-01 14:58:23,948 INFO [train.py:903] (0/4) Epoch 9, batch 3750, loss[loss=0.2229, simple_loss=0.3015, pruned_loss=0.07218, over 19597.00 frames. ], tot_loss[loss=0.2475, simple_loss=0.3177, pruned_loss=0.08871, over 3830777.63 frames. ], batch size: 52, lr: 9.25e-03, grad_scale: 8.0 +2023-04-01 14:59:24,635 INFO [train.py:903] (0/4) Epoch 9, batch 3800, loss[loss=0.2303, simple_loss=0.3069, pruned_loss=0.07679, over 19837.00 frames. ], tot_loss[loss=0.2482, simple_loss=0.318, pruned_loss=0.08916, over 3814045.02 frames. ], batch size: 52, lr: 9.24e-03, grad_scale: 8.0 +2023-04-01 14:59:54,350 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 15:00:08,584 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.327e+02 5.230e+02 6.601e+02 8.580e+02 1.875e+03, threshold=1.320e+03, percent-clipped=3.0 +2023-04-01 15:00:11,128 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2376, 1.3891, 2.0137, 1.5316, 3.2006, 2.4375, 3.5661, 1.6517], + device='cuda:0'), covar=tensor([0.2269, 0.3850, 0.2275, 0.1727, 0.1406, 0.1857, 0.1486, 0.3278], + device='cuda:0'), in_proj_covar=tensor([0.0475, 0.0554, 0.0568, 0.0426, 0.0584, 0.0480, 0.0643, 0.0477], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 15:00:17,587 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58466.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:00:24,958 INFO [train.py:903] (0/4) Epoch 9, batch 3850, loss[loss=0.2259, simple_loss=0.2988, pruned_loss=0.07647, over 19484.00 frames. ], tot_loss[loss=0.2475, simple_loss=0.318, pruned_loss=0.08853, over 3800789.67 frames. ], batch size: 49, lr: 9.24e-03, grad_scale: 8.0 +2023-04-01 15:00:46,676 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58491.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:01:01,846 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=58503.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:01:06,094 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=58506.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:01:25,313 INFO [train.py:903] (0/4) Epoch 9, batch 3900, loss[loss=0.2485, simple_loss=0.3123, pruned_loss=0.09238, over 19486.00 frames. ], tot_loss[loss=0.247, simple_loss=0.3176, pruned_loss=0.0882, over 3802377.38 frames. ], batch size: 49, lr: 9.24e-03, grad_scale: 8.0 +2023-04-01 15:02:08,183 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.315e+02 6.109e+02 7.308e+02 8.933e+02 2.200e+03, threshold=1.462e+03, percent-clipped=5.0 +2023-04-01 15:02:26,008 INFO [train.py:903] (0/4) Epoch 9, batch 3950, loss[loss=0.2994, simple_loss=0.3547, pruned_loss=0.1221, over 13913.00 frames. ], tot_loss[loss=0.2462, simple_loss=0.3172, pruned_loss=0.08763, over 3805361.74 frames. ], batch size: 136, lr: 9.23e-03, grad_scale: 8.0 +2023-04-01 15:02:28,084 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 15:02:51,957 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5937, 1.2196, 1.1727, 1.4573, 1.1682, 1.3113, 1.1712, 1.3926], + device='cuda:0'), covar=tensor([0.0904, 0.1106, 0.1451, 0.0816, 0.1053, 0.0562, 0.1217, 0.0707], + device='cuda:0'), in_proj_covar=tensor([0.0248, 0.0351, 0.0290, 0.0239, 0.0298, 0.0242, 0.0270, 0.0232], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 15:03:21,977 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58618.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:03:25,464 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=58621.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:03:28,118 INFO [train.py:903] (0/4) Epoch 9, batch 4000, loss[loss=0.2379, simple_loss=0.3003, pruned_loss=0.08777, over 19814.00 frames. ], tot_loss[loss=0.2457, simple_loss=0.3164, pruned_loss=0.08751, over 3798179.84 frames. ], batch size: 48, lr: 9.23e-03, grad_scale: 8.0 +2023-04-01 15:04:10,731 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.392e+02 5.642e+02 6.845e+02 8.578e+02 2.087e+03, threshold=1.369e+03, percent-clipped=3.0 +2023-04-01 15:04:10,793 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 15:04:27,023 INFO [train.py:903] (0/4) Epoch 9, batch 4050, loss[loss=0.263, simple_loss=0.3345, pruned_loss=0.09573, over 19666.00 frames. ], tot_loss[loss=0.246, simple_loss=0.3169, pruned_loss=0.08754, over 3801731.60 frames. ], batch size: 55, lr: 9.22e-03, grad_scale: 8.0 +2023-04-01 15:04:59,250 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58699.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:05:10,974 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.94 vs. limit=5.0 +2023-04-01 15:05:28,158 INFO [train.py:903] (0/4) Epoch 9, batch 4100, loss[loss=0.2634, simple_loss=0.3387, pruned_loss=0.094, over 19322.00 frames. ], tot_loss[loss=0.2459, simple_loss=0.3168, pruned_loss=0.0875, over 3814478.86 frames. ], batch size: 66, lr: 9.22e-03, grad_scale: 8.0 +2023-04-01 15:06:03,196 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 15:06:11,190 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.329e+02 5.572e+02 6.953e+02 8.904e+02 1.888e+03, threshold=1.391e+03, percent-clipped=3.0 +2023-04-01 15:06:28,257 INFO [train.py:903] (0/4) Epoch 9, batch 4150, loss[loss=0.2496, simple_loss=0.3263, pruned_loss=0.08644, over 19744.00 frames. ], tot_loss[loss=0.247, simple_loss=0.3178, pruned_loss=0.08813, over 3828558.34 frames. ], batch size: 63, lr: 9.22e-03, grad_scale: 8.0 +2023-04-01 15:06:29,678 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0 +2023-04-01 15:06:54,404 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58793.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:07:30,861 INFO [train.py:903] (0/4) Epoch 9, batch 4200, loss[loss=0.2026, simple_loss=0.2701, pruned_loss=0.06753, over 19329.00 frames. ], tot_loss[loss=0.2462, simple_loss=0.3171, pruned_loss=0.08765, over 3825966.08 frames. ], batch size: 44, lr: 9.21e-03, grad_scale: 8.0 +2023-04-01 15:07:34,201 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 15:08:14,670 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.798e+02 5.991e+02 7.229e+02 9.145e+02 1.621e+03, threshold=1.446e+03, percent-clipped=3.0 +2023-04-01 15:08:32,533 INFO [train.py:903] (0/4) Epoch 9, batch 4250, loss[loss=0.2026, simple_loss=0.2899, pruned_loss=0.0576, over 19677.00 frames. ], tot_loss[loss=0.2468, simple_loss=0.3175, pruned_loss=0.08804, over 3830023.58 frames. ], batch size: 58, lr: 9.21e-03, grad_scale: 8.0 +2023-04-01 15:08:34,169 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58874.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:08:37,487 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=58877.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:08:48,673 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 15:08:59,995 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 15:09:03,830 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58899.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:09:07,328 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=58902.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:09:32,792 INFO [train.py:903] (0/4) Epoch 9, batch 4300, loss[loss=0.2298, simple_loss=0.2825, pruned_loss=0.08855, over 18679.00 frames. ], tot_loss[loss=0.2475, simple_loss=0.3176, pruned_loss=0.08866, over 3814861.25 frames. ], batch size: 41, lr: 9.20e-03, grad_scale: 8.0 +2023-04-01 15:09:34,185 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=58924.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:10:17,294 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.360e+02 5.584e+02 7.321e+02 9.114e+02 2.155e+03, threshold=1.464e+03, percent-clipped=5.0 +2023-04-01 15:10:27,340 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 15:10:33,830 INFO [train.py:903] (0/4) Epoch 9, batch 4350, loss[loss=0.2784, simple_loss=0.3423, pruned_loss=0.1072, over 19590.00 frames. ], tot_loss[loss=0.2463, simple_loss=0.3167, pruned_loss=0.08794, over 3819278.21 frames. ], batch size: 52, lr: 9.20e-03, grad_scale: 8.0 +2023-04-01 15:10:43,759 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-01 15:11:10,113 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6021, 4.1434, 2.7215, 3.7403, 1.2679, 3.9142, 3.8770, 4.0675], + device='cuda:0'), covar=tensor([0.0695, 0.0968, 0.2003, 0.0701, 0.3743, 0.0804, 0.0815, 0.0951], + device='cuda:0'), in_proj_covar=tensor([0.0410, 0.0350, 0.0415, 0.0311, 0.0369, 0.0344, 0.0335, 0.0372], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 15:11:34,348 INFO [train.py:903] (0/4) Epoch 9, batch 4400, loss[loss=0.2542, simple_loss=0.3315, pruned_loss=0.08842, over 19689.00 frames. ], tot_loss[loss=0.2457, simple_loss=0.3164, pruned_loss=0.08755, over 3819875.03 frames. ], batch size: 59, lr: 9.20e-03, grad_scale: 8.0 +2023-04-01 15:11:58,694 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 15:11:58,793 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=59043.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:12:08,639 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 15:12:18,355 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.517e+02 5.626e+02 6.682e+02 1.014e+03 2.125e+03, threshold=1.336e+03, percent-clipped=6.0 +2023-04-01 15:12:35,839 INFO [train.py:903] (0/4) Epoch 9, batch 4450, loss[loss=0.208, simple_loss=0.2805, pruned_loss=0.06772, over 15188.00 frames. ], tot_loss[loss=0.2453, simple_loss=0.3157, pruned_loss=0.08748, over 3814682.92 frames. ], batch size: 33, lr: 9.19e-03, grad_scale: 8.0 +2023-04-01 15:13:05,750 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-01 15:13:36,625 INFO [train.py:903] (0/4) Epoch 9, batch 4500, loss[loss=0.2212, simple_loss=0.3031, pruned_loss=0.06964, over 19649.00 frames. ], tot_loss[loss=0.2456, simple_loss=0.3163, pruned_loss=0.08747, over 3816116.54 frames. ], batch size: 58, lr: 9.19e-03, grad_scale: 8.0 +2023-04-01 15:13:54,290 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=59137.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:14:17,431 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-01 15:14:20,682 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59158.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:14:21,457 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.665e+02 5.638e+02 7.209e+02 9.049e+02 2.430e+03, threshold=1.442e+03, percent-clipped=5.0 +2023-04-01 15:14:38,158 INFO [train.py:903] (0/4) Epoch 9, batch 4550, loss[loss=0.2406, simple_loss=0.3082, pruned_loss=0.08651, over 19742.00 frames. ], tot_loss[loss=0.246, simple_loss=0.3165, pruned_loss=0.08778, over 3800857.58 frames. ], batch size: 51, lr: 9.19e-03, grad_scale: 8.0 +2023-04-01 15:14:46,071 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 15:14:49,756 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59182.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:15:11,382 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 15:15:39,255 INFO [train.py:903] (0/4) Epoch 9, batch 4600, loss[loss=0.281, simple_loss=0.344, pruned_loss=0.109, over 19744.00 frames. ], tot_loss[loss=0.2452, simple_loss=0.3155, pruned_loss=0.08742, over 3805114.06 frames. ], batch size: 63, lr: 9.18e-03, grad_scale: 8.0 +2023-04-01 15:16:12,646 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-01 15:16:15,388 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59252.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:16:24,166 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.716e+02 5.555e+02 6.855e+02 8.371e+02 1.742e+03, threshold=1.371e+03, percent-clipped=2.0 +2023-04-01 15:16:35,108 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=59268.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:16:37,663 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1650, 1.2224, 1.1570, 0.9804, 1.0027, 1.0412, 0.0463, 0.3435], + device='cuda:0'), covar=tensor([0.0401, 0.0400, 0.0253, 0.0306, 0.0813, 0.0344, 0.0784, 0.0708], + device='cuda:0'), in_proj_covar=tensor([0.0322, 0.0322, 0.0321, 0.0337, 0.0413, 0.0339, 0.0298, 0.0318], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 15:16:40,775 INFO [train.py:903] (0/4) Epoch 9, batch 4650, loss[loss=0.2049, simple_loss=0.2712, pruned_loss=0.06929, over 19738.00 frames. ], tot_loss[loss=0.2462, simple_loss=0.3165, pruned_loss=0.08797, over 3798430.91 frames. ], batch size: 45, lr: 9.18e-03, grad_scale: 8.0 +2023-04-01 15:16:56,382 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 15:17:09,162 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 15:17:41,937 INFO [train.py:903] (0/4) Epoch 9, batch 4700, loss[loss=0.2554, simple_loss=0.318, pruned_loss=0.09638, over 18122.00 frames. ], tot_loss[loss=0.2445, simple_loss=0.315, pruned_loss=0.08703, over 3803124.02 frames. ], batch size: 83, lr: 9.17e-03, grad_scale: 8.0 +2023-04-01 15:18:03,983 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 15:18:13,119 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7942, 1.3109, 1.4025, 2.2234, 1.8225, 1.9220, 2.0341, 1.7967], + device='cuda:0'), covar=tensor([0.0866, 0.1235, 0.1163, 0.0935, 0.0884, 0.0881, 0.0877, 0.0763], + device='cuda:0'), in_proj_covar=tensor([0.0216, 0.0230, 0.0228, 0.0257, 0.0240, 0.0213, 0.0202, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 15:18:25,859 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.928e+02 5.586e+02 6.716e+02 8.168e+02 1.499e+03, threshold=1.343e+03, percent-clipped=1.0 +2023-04-01 15:18:32,956 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9833, 2.1486, 2.1864, 1.7965, 4.4888, 0.9714, 2.3465, 4.7774], + device='cuda:0'), covar=tensor([0.0323, 0.2216, 0.2130, 0.1727, 0.0726, 0.2610, 0.1305, 0.0194], + device='cuda:0'), in_proj_covar=tensor([0.0338, 0.0329, 0.0338, 0.0311, 0.0343, 0.0326, 0.0315, 0.0338], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 15:18:41,885 INFO [train.py:903] (0/4) Epoch 9, batch 4750, loss[loss=0.2211, simple_loss=0.285, pruned_loss=0.07864, over 19408.00 frames. ], tot_loss[loss=0.2437, simple_loss=0.3144, pruned_loss=0.08653, over 3802910.17 frames. ], batch size: 48, lr: 9.17e-03, grad_scale: 8.0 +2023-04-01 15:18:54,722 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59383.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:19:32,444 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59414.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:19:43,259 INFO [train.py:903] (0/4) Epoch 9, batch 4800, loss[loss=0.3011, simple_loss=0.3587, pruned_loss=0.1217, over 17691.00 frames. ], tot_loss[loss=0.2446, simple_loss=0.3152, pruned_loss=0.08702, over 3795917.05 frames. ], batch size: 101, lr: 9.17e-03, grad_scale: 8.0 +2023-04-01 15:19:44,867 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.28 vs. limit=5.0 +2023-04-01 15:19:58,756 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1223, 3.6508, 2.1442, 2.2601, 3.1919, 1.7115, 1.3561, 1.9723], + device='cuda:0'), covar=tensor([0.1086, 0.0377, 0.0845, 0.0621, 0.0424, 0.0974, 0.0860, 0.0678], + device='cuda:0'), in_proj_covar=tensor([0.0287, 0.0293, 0.0319, 0.0242, 0.0230, 0.0315, 0.0287, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 15:20:03,392 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59439.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:20:27,560 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.770e+02 5.762e+02 6.893e+02 8.818e+02 1.836e+03, threshold=1.379e+03, percent-clipped=7.0 +2023-04-01 15:20:43,943 INFO [train.py:903] (0/4) Epoch 9, batch 4850, loss[loss=0.2228, simple_loss=0.2986, pruned_loss=0.07351, over 19542.00 frames. ], tot_loss[loss=0.2466, simple_loss=0.3166, pruned_loss=0.08827, over 3797619.60 frames. ], batch size: 56, lr: 9.16e-03, grad_scale: 8.0 +2023-04-01 15:21:08,915 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 15:21:27,115 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59508.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:21:29,004 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 15:21:35,282 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 15:21:36,464 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 15:21:42,242 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59520.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:21:45,372 INFO [train.py:903] (0/4) Epoch 9, batch 4900, loss[loss=0.2344, simple_loss=0.3048, pruned_loss=0.08202, over 19676.00 frames. ], tot_loss[loss=0.2462, simple_loss=0.3165, pruned_loss=0.08798, over 3805920.55 frames. ], batch size: 60, lr: 9.16e-03, grad_scale: 8.0 +2023-04-01 15:21:46,550 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 15:21:49,052 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=59526.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:21:56,482 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.50 vs. limit=2.0 +2023-04-01 15:21:56,952 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59533.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:22:05,289 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 15:22:29,457 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.724e+02 5.274e+02 6.528e+02 8.023e+02 2.606e+03, threshold=1.306e+03, percent-clipped=3.0 +2023-04-01 15:22:45,478 INFO [train.py:903] (0/4) Epoch 9, batch 4950, loss[loss=0.2761, simple_loss=0.3404, pruned_loss=0.1059, over 19536.00 frames. ], tot_loss[loss=0.2454, simple_loss=0.3158, pruned_loss=0.08747, over 3812576.57 frames. ], batch size: 54, lr: 9.15e-03, grad_scale: 8.0 +2023-04-01 15:23:01,068 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 15:23:24,590 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 15:23:44,258 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5613, 2.3526, 1.7025, 1.5771, 2.1132, 1.3986, 1.2132, 1.8185], + device='cuda:0'), covar=tensor([0.0815, 0.0563, 0.0929, 0.0636, 0.0451, 0.1009, 0.0728, 0.0447], + device='cuda:0'), in_proj_covar=tensor([0.0287, 0.0291, 0.0318, 0.0241, 0.0230, 0.0314, 0.0286, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 15:23:46,154 INFO [train.py:903] (0/4) Epoch 9, batch 5000, loss[loss=0.2008, simple_loss=0.2799, pruned_loss=0.0608, over 19612.00 frames. ], tot_loss[loss=0.2444, simple_loss=0.3151, pruned_loss=0.08685, over 3810664.66 frames. ], batch size: 50, lr: 9.15e-03, grad_scale: 4.0 +2023-04-01 15:23:53,576 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 15:24:04,788 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 15:24:06,356 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59639.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:24:08,576 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59641.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:24:30,504 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.264e+02 5.818e+02 6.890e+02 9.185e+02 2.943e+03, threshold=1.378e+03, percent-clipped=3.0 +2023-04-01 15:24:35,487 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59664.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:24:46,097 INFO [train.py:903] (0/4) Epoch 9, batch 5050, loss[loss=0.2329, simple_loss=0.3139, pruned_loss=0.0759, over 19541.00 frames. ], tot_loss[loss=0.2447, simple_loss=0.3152, pruned_loss=0.08706, over 3821939.47 frames. ], batch size: 56, lr: 9.15e-03, grad_scale: 4.0 +2023-04-01 15:25:21,712 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 15:25:47,327 INFO [train.py:903] (0/4) Epoch 9, batch 5100, loss[loss=0.257, simple_loss=0.3291, pruned_loss=0.09244, over 19277.00 frames. ], tot_loss[loss=0.2434, simple_loss=0.3143, pruned_loss=0.0863, over 3826981.09 frames. ], batch size: 66, lr: 9.14e-03, grad_scale: 4.0 +2023-04-01 15:25:56,483 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 15:25:59,753 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 15:26:01,325 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1963, 1.8160, 1.3813, 1.1242, 1.6423, 1.0634, 1.1133, 1.7131], + device='cuda:0'), covar=tensor([0.0669, 0.0576, 0.0960, 0.0625, 0.0377, 0.1086, 0.0540, 0.0321], + device='cuda:0'), in_proj_covar=tensor([0.0289, 0.0294, 0.0321, 0.0242, 0.0231, 0.0317, 0.0288, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 15:26:05,154 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 15:26:33,075 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.640e+02 5.962e+02 7.316e+02 9.170e+02 1.645e+03, threshold=1.463e+03, percent-clipped=5.0 +2023-04-01 15:26:47,508 INFO [train.py:903] (0/4) Epoch 9, batch 5150, loss[loss=0.3467, simple_loss=0.3804, pruned_loss=0.1565, over 13305.00 frames. ], tot_loss[loss=0.2435, simple_loss=0.3141, pruned_loss=0.08648, over 3822087.77 frames. ], batch size: 136, lr: 9.14e-03, grad_scale: 4.0 +2023-04-01 15:26:55,208 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9835, 1.1868, 1.2790, 1.4008, 2.3055, 0.9739, 2.0659, 2.7291], + device='cuda:0'), covar=tensor([0.0647, 0.2884, 0.2835, 0.1689, 0.1146, 0.2477, 0.1107, 0.0483], + device='cuda:0'), in_proj_covar=tensor([0.0338, 0.0327, 0.0337, 0.0310, 0.0340, 0.0324, 0.0316, 0.0337], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 15:26:58,279 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 15:27:04,620 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3616, 1.4649, 1.8424, 1.5550, 3.0354, 2.4569, 3.3630, 1.5730], + device='cuda:0'), covar=tensor([0.2126, 0.3610, 0.2320, 0.1723, 0.1466, 0.1815, 0.1525, 0.3269], + device='cuda:0'), in_proj_covar=tensor([0.0476, 0.0555, 0.0570, 0.0425, 0.0588, 0.0477, 0.0641, 0.0476], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 15:27:32,217 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 15:27:49,828 INFO [train.py:903] (0/4) Epoch 9, batch 5200, loss[loss=0.2532, simple_loss=0.3305, pruned_loss=0.08798, over 19762.00 frames. ], tot_loss[loss=0.2442, simple_loss=0.3151, pruned_loss=0.08664, over 3820771.39 frames. ], batch size: 54, lr: 9.14e-03, grad_scale: 8.0 +2023-04-01 15:27:55,803 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0519, 1.1867, 1.4185, 1.4299, 2.6553, 0.9407, 1.9019, 2.7910], + device='cuda:0'), covar=tensor([0.0542, 0.2620, 0.2539, 0.1547, 0.0743, 0.2414, 0.1196, 0.0395], + device='cuda:0'), in_proj_covar=tensor([0.0338, 0.0328, 0.0338, 0.0310, 0.0338, 0.0324, 0.0316, 0.0337], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 15:27:59,743 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 15:28:34,608 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.759e+02 5.545e+02 6.726e+02 8.723e+02 1.623e+03, threshold=1.345e+03, percent-clipped=2.0 +2023-04-01 15:28:39,319 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=59864.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:28:41,618 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 15:28:41,857 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59866.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:28:50,361 INFO [train.py:903] (0/4) Epoch 9, batch 5250, loss[loss=0.2788, simple_loss=0.3465, pruned_loss=0.1056, over 17232.00 frames. ], tot_loss[loss=0.2445, simple_loss=0.3151, pruned_loss=0.08694, over 3813929.01 frames. ], batch size: 101, lr: 9.13e-03, grad_scale: 8.0 +2023-04-01 15:29:19,337 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=59897.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:29:31,279 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=59907.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:29:50,067 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=59922.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:29:50,797 INFO [train.py:903] (0/4) Epoch 9, batch 5300, loss[loss=0.2126, simple_loss=0.2883, pruned_loss=0.06847, over 19761.00 frames. ], tot_loss[loss=0.2448, simple_loss=0.3157, pruned_loss=0.08697, over 3821033.25 frames. ], batch size: 54, lr: 9.13e-03, grad_scale: 8.0 +2023-04-01 15:30:04,494 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 15:30:36,750 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.396e+02 5.525e+02 7.204e+02 8.995e+02 2.228e+03, threshold=1.441e+03, percent-clipped=7.0 +2023-04-01 15:30:50,970 INFO [train.py:903] (0/4) Epoch 9, batch 5350, loss[loss=0.2695, simple_loss=0.3393, pruned_loss=0.09981, over 19695.00 frames. ], tot_loss[loss=0.244, simple_loss=0.3148, pruned_loss=0.08662, over 3830951.69 frames. ], batch size: 59, lr: 9.12e-03, grad_scale: 8.0 +2023-04-01 15:30:58,318 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=59979.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:31:24,018 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 15:31:25,304 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-60000.pt +2023-04-01 15:31:52,148 INFO [train.py:903] (0/4) Epoch 9, batch 5400, loss[loss=0.2181, simple_loss=0.2822, pruned_loss=0.07704, over 19317.00 frames. ], tot_loss[loss=0.2445, simple_loss=0.315, pruned_loss=0.08698, over 3831000.48 frames. ], batch size: 44, lr: 9.12e-03, grad_scale: 8.0 +2023-04-01 15:32:16,759 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2206, 1.2785, 1.1671, 1.0183, 1.0118, 1.0934, 0.0893, 0.3763], + device='cuda:0'), covar=tensor([0.0411, 0.0400, 0.0246, 0.0308, 0.0918, 0.0349, 0.0762, 0.0685], + device='cuda:0'), in_proj_covar=tensor([0.0322, 0.0316, 0.0318, 0.0332, 0.0412, 0.0334, 0.0296, 0.0315], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 15:32:34,990 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60058.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:32:37,000 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.978e+02 5.587e+02 7.112e+02 9.365e+02 1.948e+03, threshold=1.422e+03, percent-clipped=3.0 +2023-04-01 15:32:49,966 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60070.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:32:53,196 INFO [train.py:903] (0/4) Epoch 9, batch 5450, loss[loss=0.2087, simple_loss=0.2854, pruned_loss=0.06606, over 19829.00 frames. ], tot_loss[loss=0.2444, simple_loss=0.3155, pruned_loss=0.08668, over 3833944.16 frames. ], batch size: 52, lr: 9.12e-03, grad_scale: 8.0 +2023-04-01 15:33:04,727 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-04-01 15:33:06,639 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60083.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:33:21,780 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60096.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:33:40,910 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-04-01 15:33:55,434 INFO [train.py:903] (0/4) Epoch 9, batch 5500, loss[loss=0.2664, simple_loss=0.3482, pruned_loss=0.09232, over 19766.00 frames. ], tot_loss[loss=0.2461, simple_loss=0.3166, pruned_loss=0.08777, over 3817592.77 frames. ], batch size: 56, lr: 9.11e-03, grad_scale: 8.0 +2023-04-01 15:34:17,030 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 15:34:40,127 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.116e+02 5.485e+02 6.858e+02 8.862e+02 1.983e+03, threshold=1.372e+03, percent-clipped=4.0 +2023-04-01 15:34:56,047 INFO [train.py:903] (0/4) Epoch 9, batch 5550, loss[loss=0.2834, simple_loss=0.3506, pruned_loss=0.1081, over 19709.00 frames. ], tot_loss[loss=0.246, simple_loss=0.3168, pruned_loss=0.08756, over 3815427.34 frames. ], batch size: 59, lr: 9.11e-03, grad_scale: 8.0 +2023-04-01 15:34:59,783 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2197, 3.6975, 3.8344, 3.8373, 1.4420, 3.5896, 3.1343, 3.5251], + device='cuda:0'), covar=tensor([0.1306, 0.0877, 0.0628, 0.0641, 0.4858, 0.0756, 0.0690, 0.1085], + device='cuda:0'), in_proj_covar=tensor([0.0623, 0.0559, 0.0743, 0.0629, 0.0683, 0.0497, 0.0456, 0.0691], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 15:35:01,731 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 15:35:42,126 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60210.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:35:48,978 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60216.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:35:49,884 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 15:35:57,517 INFO [train.py:903] (0/4) Epoch 9, batch 5600, loss[loss=0.2381, simple_loss=0.3168, pruned_loss=0.0797, over 18023.00 frames. ], tot_loss[loss=0.2458, simple_loss=0.3163, pruned_loss=0.08759, over 3818509.95 frames. ], batch size: 83, lr: 9.11e-03, grad_scale: 8.0 +2023-04-01 15:36:12,194 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60235.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:36:16,416 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-04-01 15:36:31,693 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60251.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:36:41,623 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.287e+02 5.903e+02 7.092e+02 9.595e+02 1.671e+03, threshold=1.418e+03, percent-clipped=4.0 +2023-04-01 15:36:42,004 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60260.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:36:58,107 INFO [train.py:903] (0/4) Epoch 9, batch 5650, loss[loss=0.2456, simple_loss=0.3207, pruned_loss=0.08525, over 19775.00 frames. ], tot_loss[loss=0.2446, simple_loss=0.3153, pruned_loss=0.08691, over 3829739.05 frames. ], batch size: 56, lr: 9.10e-03, grad_scale: 8.0 +2023-04-01 15:37:38,577 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7903, 4.3664, 2.8725, 3.8309, 0.8904, 4.0202, 4.0770, 4.2250], + device='cuda:0'), covar=tensor([0.0595, 0.0950, 0.1692, 0.0719, 0.3871, 0.0751, 0.0705, 0.0784], + device='cuda:0'), in_proj_covar=tensor([0.0410, 0.0347, 0.0418, 0.0304, 0.0368, 0.0344, 0.0332, 0.0369], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0002, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 15:37:45,043 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 15:38:00,337 INFO [train.py:903] (0/4) Epoch 9, batch 5700, loss[loss=0.2289, simple_loss=0.3027, pruned_loss=0.07755, over 19760.00 frames. ], tot_loss[loss=0.2441, simple_loss=0.315, pruned_loss=0.08658, over 3826637.72 frames. ], batch size: 54, lr: 9.10e-03, grad_scale: 8.0 +2023-04-01 15:38:02,934 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60325.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:38:45,031 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.384e+02 5.631e+02 6.670e+02 7.834e+02 2.342e+03, threshold=1.334e+03, percent-clipped=2.0 +2023-04-01 15:38:52,986 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60366.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:38:59,661 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 15:39:00,754 INFO [train.py:903] (0/4) Epoch 9, batch 5750, loss[loss=0.2503, simple_loss=0.3337, pruned_loss=0.08347, over 19338.00 frames. ], tot_loss[loss=0.2445, simple_loss=0.3151, pruned_loss=0.08694, over 3818424.55 frames. ], batch size: 66, lr: 9.09e-03, grad_scale: 8.0 +2023-04-01 15:39:07,411 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 15:39:12,833 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 15:39:36,133 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60402.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:39:50,843 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60414.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:40:00,918 INFO [train.py:903] (0/4) Epoch 9, batch 5800, loss[loss=0.2347, simple_loss=0.303, pruned_loss=0.08325, over 19589.00 frames. ], tot_loss[loss=0.2461, simple_loss=0.3165, pruned_loss=0.08784, over 3812031.70 frames. ], batch size: 52, lr: 9.09e-03, grad_scale: 8.0 +2023-04-01 15:40:06,314 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60427.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:40:12,303 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-01 15:40:22,430 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60440.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:40:45,425 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.333e+02 6.303e+02 7.771e+02 9.979e+02 2.257e+03, threshold=1.554e+03, percent-clipped=12.0 +2023-04-01 15:41:01,366 INFO [train.py:903] (0/4) Epoch 9, batch 5850, loss[loss=0.2025, simple_loss=0.2708, pruned_loss=0.06715, over 19093.00 frames. ], tot_loss[loss=0.2467, simple_loss=0.3167, pruned_loss=0.08833, over 3822375.09 frames. ], batch size: 42, lr: 9.09e-03, grad_scale: 8.0 +2023-04-01 15:41:55,996 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60517.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:42:02,422 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 15:42:03,580 INFO [train.py:903] (0/4) Epoch 9, batch 5900, loss[loss=0.2128, simple_loss=0.2797, pruned_loss=0.0729, over 19737.00 frames. ], tot_loss[loss=0.2453, simple_loss=0.3156, pruned_loss=0.08751, over 3801626.76 frames. ], batch size: 47, lr: 9.08e-03, grad_scale: 8.0 +2023-04-01 15:42:10,666 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60529.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:42:22,843 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 15:42:25,414 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60542.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:42:42,845 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60555.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:42:46,108 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5662, 1.1188, 1.1898, 2.1503, 1.4605, 1.5129, 1.7756, 1.6011], + device='cuda:0'), covar=tensor([0.0972, 0.1432, 0.1259, 0.0856, 0.1128, 0.1060, 0.1053, 0.0915], + device='cuda:0'), in_proj_covar=tensor([0.0218, 0.0232, 0.0234, 0.0259, 0.0247, 0.0218, 0.0204, 0.0210], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 15:42:47,829 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.985e+02 5.460e+02 7.799e+02 9.723e+02 2.713e+03, threshold=1.560e+03, percent-clipped=1.0 +2023-04-01 15:42:47,991 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=60560.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:43:03,509 INFO [train.py:903] (0/4) Epoch 9, batch 5950, loss[loss=0.2425, simple_loss=0.3123, pruned_loss=0.08631, over 19587.00 frames. ], tot_loss[loss=0.2462, simple_loss=0.3162, pruned_loss=0.08813, over 3809454.58 frames. ], batch size: 52, lr: 9.08e-03, grad_scale: 8.0 +2023-04-01 15:43:10,771 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8736, 1.3285, 1.0708, 0.9026, 1.1589, 0.9945, 0.8339, 1.2596], + device='cuda:0'), covar=tensor([0.0489, 0.0619, 0.0910, 0.0519, 0.0428, 0.1010, 0.0531, 0.0376], + device='cuda:0'), in_proj_covar=tensor([0.0290, 0.0298, 0.0326, 0.0245, 0.0233, 0.0324, 0.0292, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 15:43:13,134 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60581.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:43:20,760 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5647, 1.7135, 2.0807, 1.6301, 3.0619, 4.8904, 4.7084, 5.2105], + device='cuda:0'), covar=tensor([0.1384, 0.2810, 0.2682, 0.1764, 0.0442, 0.0133, 0.0135, 0.0101], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0288, 0.0317, 0.0245, 0.0209, 0.0145, 0.0202, 0.0183], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 15:43:45,240 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60606.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:44:03,655 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60622.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:44:04,413 INFO [train.py:903] (0/4) Epoch 9, batch 6000, loss[loss=0.2033, simple_loss=0.2847, pruned_loss=0.06092, over 19692.00 frames. ], tot_loss[loss=0.2471, simple_loss=0.3172, pruned_loss=0.08852, over 3790258.44 frames. ], batch size: 53, lr: 9.08e-03, grad_scale: 8.0 +2023-04-01 15:44:04,413 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 15:44:16,875 INFO [train.py:937] (0/4) Epoch 9, validation: loss=0.1828, simple_loss=0.2835, pruned_loss=0.04105, over 944034.00 frames. +2023-04-01 15:44:16,876 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 15:44:46,052 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60647.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:45:02,309 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.543e+02 5.785e+02 7.329e+02 9.590e+02 1.552e+03, threshold=1.466e+03, percent-clipped=0.0 +2023-04-01 15:45:17,224 INFO [train.py:903] (0/4) Epoch 9, batch 6050, loss[loss=0.2707, simple_loss=0.3372, pruned_loss=0.1021, over 17232.00 frames. ], tot_loss[loss=0.2454, simple_loss=0.3157, pruned_loss=0.08759, over 3797462.23 frames. ], batch size: 101, lr: 9.07e-03, grad_scale: 8.0 +2023-04-01 15:45:19,785 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=60675.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:46:18,277 INFO [train.py:903] (0/4) Epoch 9, batch 6100, loss[loss=0.2006, simple_loss=0.2703, pruned_loss=0.06539, over 19720.00 frames. ], tot_loss[loss=0.2434, simple_loss=0.3139, pruned_loss=0.08642, over 3807160.13 frames. ], batch size: 45, lr: 9.07e-03, grad_scale: 8.0 +2023-04-01 15:46:34,561 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=60736.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:47:03,004 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.337e+02 5.339e+02 6.367e+02 8.531e+02 1.806e+03, threshold=1.273e+03, percent-clipped=4.0 +2023-04-01 15:47:18,986 INFO [train.py:903] (0/4) Epoch 9, batch 6150, loss[loss=0.2569, simple_loss=0.3262, pruned_loss=0.0938, over 19656.00 frames. ], tot_loss[loss=0.2452, simple_loss=0.3151, pruned_loss=0.0876, over 3803873.55 frames. ], batch size: 55, lr: 9.06e-03, grad_scale: 8.0 +2023-04-01 15:47:19,435 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60773.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:47:33,339 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60785.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:47:44,095 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 15:47:48,612 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60798.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:47:48,641 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60798.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:48:03,536 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60810.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:48:04,567 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60811.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:48:18,466 INFO [train.py:903] (0/4) Epoch 9, batch 6200, loss[loss=0.2601, simple_loss=0.3366, pruned_loss=0.09176, over 19645.00 frames. ], tot_loss[loss=0.2456, simple_loss=0.3155, pruned_loss=0.08786, over 3802121.18 frames. ], batch size: 55, lr: 9.06e-03, grad_scale: 8.0 +2023-04-01 15:48:18,860 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60823.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:48:34,296 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60836.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:49:03,644 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.706e+02 5.965e+02 7.614e+02 9.330e+02 2.107e+03, threshold=1.523e+03, percent-clipped=6.0 +2023-04-01 15:49:19,598 INFO [train.py:903] (0/4) Epoch 9, batch 6250, loss[loss=0.2578, simple_loss=0.331, pruned_loss=0.09225, over 19658.00 frames. ], tot_loss[loss=0.2461, simple_loss=0.3164, pruned_loss=0.08787, over 3810409.55 frames. ], batch size: 55, lr: 9.06e-03, grad_scale: 8.0 +2023-04-01 15:49:49,664 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 15:50:19,515 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7568, 1.7730, 1.6269, 1.5314, 1.3720, 1.5801, 0.8950, 1.1673], + device='cuda:0'), covar=tensor([0.0300, 0.0333, 0.0210, 0.0309, 0.0567, 0.0387, 0.0613, 0.0479], + device='cuda:0'), in_proj_covar=tensor([0.0324, 0.0319, 0.0321, 0.0334, 0.0415, 0.0339, 0.0298, 0.0318], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 15:50:20,300 INFO [train.py:903] (0/4) Epoch 9, batch 6300, loss[loss=0.3091, simple_loss=0.3601, pruned_loss=0.1291, over 12418.00 frames. ], tot_loss[loss=0.2466, simple_loss=0.3169, pruned_loss=0.08817, over 3793135.10 frames. ], batch size: 135, lr: 9.05e-03, grad_scale: 8.0 +2023-04-01 15:50:30,434 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=60931.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:51:01,525 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=60956.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:51:05,846 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.132e+02 5.456e+02 6.999e+02 8.896e+02 1.665e+03, threshold=1.400e+03, percent-clipped=1.0 +2023-04-01 15:51:21,557 INFO [train.py:903] (0/4) Epoch 9, batch 6350, loss[loss=0.2857, simple_loss=0.3466, pruned_loss=0.1125, over 19737.00 frames. ], tot_loss[loss=0.2459, simple_loss=0.3163, pruned_loss=0.08774, over 3791256.64 frames. ], batch size: 63, lr: 9.05e-03, grad_scale: 8.0 +2023-04-01 15:52:05,028 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2997, 3.7421, 3.8250, 3.8545, 1.5135, 3.5816, 3.1919, 3.5113], + device='cuda:0'), covar=tensor([0.1207, 0.0868, 0.0601, 0.0561, 0.4687, 0.0663, 0.0635, 0.1045], + device='cuda:0'), in_proj_covar=tensor([0.0629, 0.0564, 0.0756, 0.0629, 0.0693, 0.0501, 0.0466, 0.0698], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 15:52:22,396 INFO [train.py:903] (0/4) Epoch 9, batch 6400, loss[loss=0.187, simple_loss=0.2703, pruned_loss=0.0519, over 19767.00 frames. ], tot_loss[loss=0.2469, simple_loss=0.3176, pruned_loss=0.08809, over 3798433.20 frames. ], batch size: 54, lr: 9.05e-03, grad_scale: 8.0 +2023-04-01 15:52:47,690 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4470, 1.1805, 1.2753, 1.6472, 1.2999, 1.5813, 1.6979, 1.4926], + device='cuda:0'), covar=tensor([0.0884, 0.1122, 0.1122, 0.0842, 0.0972, 0.0880, 0.0811, 0.0773], + device='cuda:0'), in_proj_covar=tensor([0.0216, 0.0230, 0.0232, 0.0257, 0.0245, 0.0216, 0.0203, 0.0210], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 15:53:07,177 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.915e+02 6.196e+02 7.384e+02 8.672e+02 1.804e+03, threshold=1.477e+03, percent-clipped=4.0 +2023-04-01 15:53:21,075 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.64 vs. limit=2.0 +2023-04-01 15:53:22,488 INFO [train.py:903] (0/4) Epoch 9, batch 6450, loss[loss=0.2646, simple_loss=0.3347, pruned_loss=0.09728, over 19575.00 frames. ], tot_loss[loss=0.2478, simple_loss=0.3181, pruned_loss=0.08871, over 3806168.18 frames. ], batch size: 61, lr: 9.04e-03, grad_scale: 8.0 +2023-04-01 15:53:30,332 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=61079.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:53:31,420 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=61080.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:54:05,086 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 15:54:22,079 INFO [train.py:903] (0/4) Epoch 9, batch 6500, loss[loss=0.2146, simple_loss=0.2838, pruned_loss=0.07273, over 19735.00 frames. ], tot_loss[loss=0.2477, simple_loss=0.318, pruned_loss=0.08866, over 3804547.33 frames. ], batch size: 51, lr: 9.04e-03, grad_scale: 8.0 +2023-04-01 15:54:27,555 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 15:55:05,286 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5106, 1.8999, 1.9102, 1.9675, 4.0086, 1.2779, 2.5624, 4.2072], + device='cuda:0'), covar=tensor([0.0386, 0.2161, 0.2234, 0.1485, 0.0616, 0.2242, 0.1161, 0.0221], + device='cuda:0'), in_proj_covar=tensor([0.0337, 0.0327, 0.0338, 0.0310, 0.0338, 0.0323, 0.0317, 0.0337], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 15:55:06,067 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.712e+02 6.706e+02 8.302e+02 1.012e+03 2.679e+03, threshold=1.660e+03, percent-clipped=7.0 +2023-04-01 15:55:22,000 INFO [train.py:903] (0/4) Epoch 9, batch 6550, loss[loss=0.2727, simple_loss=0.3437, pruned_loss=0.1009, over 19706.00 frames. ], tot_loss[loss=0.2471, simple_loss=0.3178, pruned_loss=0.08826, over 3810855.61 frames. ], batch size: 59, lr: 9.03e-03, grad_scale: 8.0 +2023-04-01 15:55:50,226 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=61195.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 15:56:10,802 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-01 15:56:24,270 INFO [train.py:903] (0/4) Epoch 9, batch 6600, loss[loss=0.2033, simple_loss=0.2704, pruned_loss=0.06812, over 19822.00 frames. ], tot_loss[loss=0.2459, simple_loss=0.3167, pruned_loss=0.08757, over 3813269.83 frames. ], batch size: 49, lr: 9.03e-03, grad_scale: 8.0 +2023-04-01 15:57:09,253 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.208e+02 5.654e+02 6.784e+02 8.392e+02 1.741e+03, threshold=1.357e+03, percent-clipped=1.0 +2023-04-01 15:57:24,976 INFO [train.py:903] (0/4) Epoch 9, batch 6650, loss[loss=0.2541, simple_loss=0.3377, pruned_loss=0.08523, over 19700.00 frames. ], tot_loss[loss=0.2455, simple_loss=0.3162, pruned_loss=0.08735, over 3814270.16 frames. ], batch size: 59, lr: 9.03e-03, grad_scale: 8.0 +2023-04-01 15:57:51,114 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-01 15:58:25,582 INFO [train.py:903] (0/4) Epoch 9, batch 6700, loss[loss=0.2363, simple_loss=0.317, pruned_loss=0.07778, over 19751.00 frames. ], tot_loss[loss=0.2462, simple_loss=0.3169, pruned_loss=0.08774, over 3819265.96 frames. ], batch size: 63, lr: 9.02e-03, grad_scale: 8.0 +2023-04-01 15:59:08,505 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.478e+02 5.683e+02 7.501e+02 9.618e+02 2.603e+03, threshold=1.500e+03, percent-clipped=7.0 +2023-04-01 15:59:23,018 INFO [train.py:903] (0/4) Epoch 9, batch 6750, loss[loss=0.2047, simple_loss=0.286, pruned_loss=0.06163, over 19597.00 frames. ], tot_loss[loss=0.2445, simple_loss=0.3155, pruned_loss=0.08676, over 3830871.09 frames. ], batch size: 50, lr: 9.02e-03, grad_scale: 8.0 +2023-04-01 15:59:43,448 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3252, 2.3793, 1.8173, 1.5653, 2.1847, 1.3998, 1.2919, 1.8372], + device='cuda:0'), covar=tensor([0.0914, 0.0587, 0.0770, 0.0652, 0.0386, 0.0986, 0.0681, 0.0397], + device='cuda:0'), in_proj_covar=tensor([0.0288, 0.0298, 0.0321, 0.0244, 0.0232, 0.0322, 0.0287, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:00:18,968 INFO [train.py:903] (0/4) Epoch 9, batch 6800, loss[loss=0.2503, simple_loss=0.3255, pruned_loss=0.08755, over 19678.00 frames. ], tot_loss[loss=0.2447, simple_loss=0.3153, pruned_loss=0.08703, over 3814591.80 frames. ], batch size: 60, lr: 9.02e-03, grad_scale: 8.0 +2023-04-01 16:00:19,084 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=61423.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:00:40,936 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-04-01 16:00:48,222 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-9.pt +2023-04-01 16:01:03,773 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 16:01:04,209 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 16:01:06,621 INFO [train.py:903] (0/4) Epoch 10, batch 0, loss[loss=0.2248, simple_loss=0.3153, pruned_loss=0.06717, over 19770.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.3153, pruned_loss=0.06717, over 19770.00 frames. ], batch size: 56, lr: 8.57e-03, grad_scale: 8.0 +2023-04-01 16:01:06,622 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 16:01:17,500 INFO [train.py:937] (0/4) Epoch 10, validation: loss=0.1825, simple_loss=0.2836, pruned_loss=0.04072, over 944034.00 frames. +2023-04-01 16:01:17,501 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 16:01:17,949 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61451.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:01:27,607 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.284e+02 5.689e+02 6.760e+02 8.116e+02 1.440e+03, threshold=1.352e+03, percent-clipped=0.0 +2023-04-01 16:01:29,706 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 16:01:47,126 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61476.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:02:17,429 INFO [train.py:903] (0/4) Epoch 10, batch 50, loss[loss=0.2402, simple_loss=0.3189, pruned_loss=0.08078, over 17954.00 frames. ], tot_loss[loss=0.249, simple_loss=0.319, pruned_loss=0.08951, over 861154.67 frames. ], batch size: 83, lr: 8.57e-03, grad_scale: 8.0 +2023-04-01 16:02:50,337 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 16:02:55,416 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9352, 1.3663, 1.0568, 1.0308, 1.2150, 1.0116, 1.0569, 1.2362], + device='cuda:0'), covar=tensor([0.0487, 0.0704, 0.1002, 0.0521, 0.0460, 0.1134, 0.0472, 0.0407], + device='cuda:0'), in_proj_covar=tensor([0.0290, 0.0300, 0.0324, 0.0245, 0.0234, 0.0326, 0.0289, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:03:03,257 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=61538.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:03:18,483 INFO [train.py:903] (0/4) Epoch 10, batch 100, loss[loss=0.2552, simple_loss=0.3334, pruned_loss=0.08846, over 19140.00 frames. ], tot_loss[loss=0.2488, simple_loss=0.3179, pruned_loss=0.08988, over 1503468.67 frames. ], batch size: 69, lr: 8.57e-03, grad_scale: 8.0 +2023-04-01 16:03:24,178 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 16:03:29,318 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.995e+02 6.266e+02 7.755e+02 9.384e+02 2.029e+03, threshold=1.551e+03, percent-clipped=6.0 +2023-04-01 16:04:19,612 INFO [train.py:903] (0/4) Epoch 10, batch 150, loss[loss=0.2296, simple_loss=0.2913, pruned_loss=0.08394, over 19361.00 frames. ], tot_loss[loss=0.2447, simple_loss=0.315, pruned_loss=0.08717, over 2022825.03 frames. ], batch size: 47, lr: 8.56e-03, grad_scale: 16.0 +2023-04-01 16:05:12,384 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 16:05:20,103 INFO [train.py:903] (0/4) Epoch 10, batch 200, loss[loss=0.25, simple_loss=0.3252, pruned_loss=0.08736, over 19541.00 frames. ], tot_loss[loss=0.2426, simple_loss=0.3132, pruned_loss=0.08604, over 2424694.62 frames. ], batch size: 54, lr: 8.56e-03, grad_scale: 8.0 +2023-04-01 16:05:32,331 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.478e+02 5.325e+02 6.934e+02 9.117e+02 1.602e+03, threshold=1.387e+03, percent-clipped=3.0 +2023-04-01 16:05:34,685 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6916, 2.4432, 2.0330, 2.8614, 2.7278, 2.3746, 2.3204, 2.5244], + device='cuda:0'), covar=tensor([0.0743, 0.1398, 0.1296, 0.0829, 0.1019, 0.0414, 0.0912, 0.0543], + device='cuda:0'), in_proj_covar=tensor([0.0250, 0.0354, 0.0289, 0.0238, 0.0297, 0.0243, 0.0274, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:05:35,096 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-04-01 16:05:45,459 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8887, 5.0673, 5.7451, 5.7284, 2.1443, 5.4255, 4.6875, 5.3106], + device='cuda:0'), covar=tensor([0.1284, 0.0671, 0.0490, 0.0463, 0.4820, 0.0546, 0.0513, 0.1073], + device='cuda:0'), in_proj_covar=tensor([0.0632, 0.0563, 0.0751, 0.0633, 0.0694, 0.0506, 0.0465, 0.0701], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 16:06:20,989 INFO [train.py:903] (0/4) Epoch 10, batch 250, loss[loss=0.264, simple_loss=0.3393, pruned_loss=0.09432, over 19582.00 frames. ], tot_loss[loss=0.242, simple_loss=0.3126, pruned_loss=0.08569, over 2749852.63 frames. ], batch size: 61, lr: 8.56e-03, grad_scale: 8.0 +2023-04-01 16:07:19,641 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=61750.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:07:20,496 INFO [train.py:903] (0/4) Epoch 10, batch 300, loss[loss=0.1969, simple_loss=0.2643, pruned_loss=0.06475, over 19734.00 frames. ], tot_loss[loss=0.2429, simple_loss=0.3135, pruned_loss=0.08614, over 2990418.77 frames. ], batch size: 46, lr: 8.55e-03, grad_scale: 8.0 +2023-04-01 16:07:32,774 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.696e+02 5.563e+02 6.785e+02 8.281e+02 1.821e+03, threshold=1.357e+03, percent-clipped=1.0 +2023-04-01 16:07:33,018 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=61761.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:08:12,876 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=61794.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:08:20,712 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 16:08:21,872 INFO [train.py:903] (0/4) Epoch 10, batch 350, loss[loss=0.218, simple_loss=0.2947, pruned_loss=0.07068, over 19547.00 frames. ], tot_loss[loss=0.2401, simple_loss=0.3111, pruned_loss=0.08458, over 3176511.28 frames. ], batch size: 54, lr: 8.55e-03, grad_scale: 8.0 +2023-04-01 16:08:44,854 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=61819.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:09:23,122 INFO [train.py:903] (0/4) Epoch 10, batch 400, loss[loss=0.2494, simple_loss=0.3232, pruned_loss=0.08786, over 19581.00 frames. ], tot_loss[loss=0.2409, simple_loss=0.3114, pruned_loss=0.08516, over 3322295.20 frames. ], batch size: 52, lr: 8.54e-03, grad_scale: 8.0 +2023-04-01 16:09:36,146 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.162e+02 5.438e+02 6.846e+02 8.745e+02 2.106e+03, threshold=1.369e+03, percent-clipped=7.0 +2023-04-01 16:10:26,712 INFO [train.py:903] (0/4) Epoch 10, batch 450, loss[loss=0.2516, simple_loss=0.3285, pruned_loss=0.08734, over 19679.00 frames. ], tot_loss[loss=0.2419, simple_loss=0.3123, pruned_loss=0.08578, over 3434267.60 frames. ], batch size: 60, lr: 8.54e-03, grad_scale: 8.0 +2023-04-01 16:10:29,708 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-01 16:10:50,966 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 16:10:51,004 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 16:11:05,170 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=61932.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:11:19,241 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0882, 5.1018, 5.8658, 5.8179, 1.8979, 5.5153, 4.7241, 5.3879], + device='cuda:0'), covar=tensor([0.1158, 0.0682, 0.0495, 0.0447, 0.5102, 0.0446, 0.0495, 0.1073], + device='cuda:0'), in_proj_covar=tensor([0.0624, 0.0559, 0.0743, 0.0631, 0.0687, 0.0500, 0.0460, 0.0690], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 16:11:27,993 INFO [train.py:903] (0/4) Epoch 10, batch 500, loss[loss=0.2202, simple_loss=0.3171, pruned_loss=0.06165, over 19609.00 frames. ], tot_loss[loss=0.2411, simple_loss=0.3121, pruned_loss=0.08505, over 3528203.21 frames. ], batch size: 57, lr: 8.54e-03, grad_scale: 8.0 +2023-04-01 16:11:30,101 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-04-01 16:11:39,884 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.771e+02 5.473e+02 6.472e+02 7.882e+02 1.512e+03, threshold=1.294e+03, percent-clipped=2.0 +2023-04-01 16:12:28,789 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-62000.pt +2023-04-01 16:12:30,817 INFO [train.py:903] (0/4) Epoch 10, batch 550, loss[loss=0.2303, simple_loss=0.3121, pruned_loss=0.07427, over 19695.00 frames. ], tot_loss[loss=0.2399, simple_loss=0.3112, pruned_loss=0.08424, over 3610131.51 frames. ], batch size: 59, lr: 8.53e-03, grad_scale: 8.0 +2023-04-01 16:13:32,123 INFO [train.py:903] (0/4) Epoch 10, batch 600, loss[loss=0.2354, simple_loss=0.3148, pruned_loss=0.07801, over 19765.00 frames. ], tot_loss[loss=0.2404, simple_loss=0.3115, pruned_loss=0.08462, over 3661987.19 frames. ], batch size: 54, lr: 8.53e-03, grad_scale: 8.0 +2023-04-01 16:13:42,210 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-04-01 16:13:46,035 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.499e+02 5.366e+02 7.096e+02 8.541e+02 1.663e+03, threshold=1.419e+03, percent-clipped=2.0 +2023-04-01 16:14:11,911 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 16:14:26,432 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62094.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:14:35,312 INFO [train.py:903] (0/4) Epoch 10, batch 650, loss[loss=0.2312, simple_loss=0.3019, pruned_loss=0.08022, over 19732.00 frames. ], tot_loss[loss=0.2405, simple_loss=0.3115, pruned_loss=0.08471, over 3712516.32 frames. ], batch size: 51, lr: 8.53e-03, grad_scale: 8.0 +2023-04-01 16:14:40,022 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62105.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:15:24,559 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62140.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:15:39,368 INFO [train.py:903] (0/4) Epoch 10, batch 700, loss[loss=0.2378, simple_loss=0.306, pruned_loss=0.08482, over 19484.00 frames. ], tot_loss[loss=0.2403, simple_loss=0.3114, pruned_loss=0.08459, over 3735832.33 frames. ], batch size: 49, lr: 8.52e-03, grad_scale: 8.0 +2023-04-01 16:15:51,155 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.410e+02 5.986e+02 7.007e+02 9.230e+02 2.462e+03, threshold=1.401e+03, percent-clipped=6.0 +2023-04-01 16:15:51,631 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1899, 1.5088, 2.1686, 1.6801, 3.0655, 2.1328, 3.1401, 1.4320], + device='cuda:0'), covar=tensor([0.2363, 0.3917, 0.2145, 0.1704, 0.1422, 0.2154, 0.1696, 0.3634], + device='cuda:0'), in_proj_covar=tensor([0.0476, 0.0557, 0.0575, 0.0426, 0.0585, 0.0480, 0.0639, 0.0475], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 16:16:41,939 INFO [train.py:903] (0/4) Epoch 10, batch 750, loss[loss=0.2091, simple_loss=0.2943, pruned_loss=0.06199, over 19681.00 frames. ], tot_loss[loss=0.2416, simple_loss=0.313, pruned_loss=0.08516, over 3764743.73 frames. ], batch size: 53, lr: 8.52e-03, grad_scale: 8.0 +2023-04-01 16:16:51,056 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62209.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:17:04,546 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62220.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:17:11,148 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2987, 1.3070, 1.4780, 1.7032, 2.8623, 1.2017, 2.0650, 3.0500], + device='cuda:0'), covar=tensor([0.0460, 0.2461, 0.2557, 0.1468, 0.0722, 0.2211, 0.1276, 0.0378], + device='cuda:0'), in_proj_covar=tensor([0.0338, 0.0326, 0.0337, 0.0311, 0.0338, 0.0322, 0.0317, 0.0334], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:17:35,232 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.81 vs. limit=5.0 +2023-04-01 16:17:42,573 INFO [train.py:903] (0/4) Epoch 10, batch 800, loss[loss=0.2658, simple_loss=0.3403, pruned_loss=0.09561, over 19532.00 frames. ], tot_loss[loss=0.2429, simple_loss=0.314, pruned_loss=0.08586, over 3778246.86 frames. ], batch size: 56, lr: 8.52e-03, grad_scale: 8.0 +2023-04-01 16:17:54,855 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.040e+02 5.237e+02 6.930e+02 8.487e+02 1.526e+03, threshold=1.386e+03, percent-clipped=2.0 +2023-04-01 16:17:58,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 16:18:14,952 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62276.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:18:32,371 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62291.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:18:43,709 INFO [train.py:903] (0/4) Epoch 10, batch 850, loss[loss=0.2393, simple_loss=0.3035, pruned_loss=0.08754, over 18269.00 frames. ], tot_loss[loss=0.2428, simple_loss=0.314, pruned_loss=0.08583, over 3803257.89 frames. ], batch size: 40, lr: 8.51e-03, grad_scale: 8.0 +2023-04-01 16:19:37,716 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 16:19:45,735 INFO [train.py:903] (0/4) Epoch 10, batch 900, loss[loss=0.2421, simple_loss=0.3164, pruned_loss=0.08386, over 19590.00 frames. ], tot_loss[loss=0.2417, simple_loss=0.3129, pruned_loss=0.08525, over 3818284.60 frames. ], batch size: 61, lr: 8.51e-03, grad_scale: 8.0 +2023-04-01 16:19:59,161 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.268e+02 6.023e+02 7.076e+02 9.770e+02 2.916e+03, threshold=1.415e+03, percent-clipped=7.0 +2023-04-01 16:20:16,734 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4901, 1.5552, 1.5739, 1.9819, 1.4668, 1.8381, 1.7939, 1.5146], + device='cuda:0'), covar=tensor([0.2438, 0.2117, 0.1068, 0.1132, 0.2140, 0.0955, 0.2223, 0.1801], + device='cuda:0'), in_proj_covar=tensor([0.0762, 0.0769, 0.0629, 0.0879, 0.0759, 0.0674, 0.0774, 0.0687], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 16:20:35,892 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62391.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:20:48,986 INFO [train.py:903] (0/4) Epoch 10, batch 950, loss[loss=0.207, simple_loss=0.2888, pruned_loss=0.06261, over 19730.00 frames. ], tot_loss[loss=0.2416, simple_loss=0.313, pruned_loss=0.08511, over 3816894.26 frames. ], batch size: 51, lr: 8.51e-03, grad_scale: 8.0 +2023-04-01 16:20:50,166 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 16:20:58,641 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62409.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:21:50,230 INFO [train.py:903] (0/4) Epoch 10, batch 1000, loss[loss=0.3064, simple_loss=0.3574, pruned_loss=0.1277, over 13658.00 frames. ], tot_loss[loss=0.2436, simple_loss=0.3146, pruned_loss=0.08627, over 3801219.44 frames. ], batch size: 137, lr: 8.50e-03, grad_scale: 8.0 +2023-04-01 16:22:01,606 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.402e+02 5.469e+02 6.659e+02 8.311e+02 1.987e+03, threshold=1.332e+03, percent-clipped=4.0 +2023-04-01 16:22:07,600 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62465.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:22:21,491 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62476.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:22:31,351 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62484.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:22:38,723 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62490.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:22:43,055 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 16:22:51,163 INFO [train.py:903] (0/4) Epoch 10, batch 1050, loss[loss=0.2157, simple_loss=0.2957, pruned_loss=0.06785, over 19657.00 frames. ], tot_loss[loss=0.2418, simple_loss=0.313, pruned_loss=0.08526, over 3820405.46 frames. ], batch size: 55, lr: 8.50e-03, grad_scale: 8.0 +2023-04-01 16:22:51,588 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62501.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:23:23,823 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 16:23:54,501 INFO [train.py:903] (0/4) Epoch 10, batch 1100, loss[loss=0.3099, simple_loss=0.377, pruned_loss=0.1214, over 18283.00 frames. ], tot_loss[loss=0.2407, simple_loss=0.3122, pruned_loss=0.08462, over 3838327.00 frames. ], batch size: 83, lr: 8.50e-03, grad_scale: 8.0 +2023-04-01 16:24:07,777 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.238e+02 5.569e+02 6.927e+02 9.101e+02 1.941e+03, threshold=1.385e+03, percent-clipped=3.0 +2023-04-01 16:24:10,206 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62563.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:24:54,352 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62599.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:24:56,403 INFO [train.py:903] (0/4) Epoch 10, batch 1150, loss[loss=0.3073, simple_loss=0.3607, pruned_loss=0.127, over 19281.00 frames. ], tot_loss[loss=0.2426, simple_loss=0.3135, pruned_loss=0.08586, over 3840477.87 frames. ], batch size: 66, lr: 8.49e-03, grad_scale: 8.0 +2023-04-01 16:24:59,681 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62603.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:25:15,631 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2402, 2.1044, 1.6472, 1.3326, 1.9179, 1.2423, 1.1120, 1.7504], + device='cuda:0'), covar=tensor([0.0682, 0.0598, 0.0846, 0.0667, 0.0366, 0.1013, 0.0576, 0.0357], + device='cuda:0'), in_proj_covar=tensor([0.0287, 0.0296, 0.0321, 0.0239, 0.0231, 0.0325, 0.0285, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:25:37,687 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62635.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:25:47,587 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-04-01 16:25:53,850 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62647.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:25:58,140 INFO [train.py:903] (0/4) Epoch 10, batch 1200, loss[loss=0.2512, simple_loss=0.3149, pruned_loss=0.09373, over 19467.00 frames. ], tot_loss[loss=0.2431, simple_loss=0.3136, pruned_loss=0.0863, over 3827660.32 frames. ], batch size: 49, lr: 8.49e-03, grad_scale: 8.0 +2023-04-01 16:26:09,506 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.384e+02 5.581e+02 6.902e+02 9.588e+02 2.703e+03, threshold=1.380e+03, percent-clipped=8.0 +2023-04-01 16:26:24,950 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62672.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:26:32,533 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 16:26:39,683 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62684.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:26:50,277 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.66 vs. limit=5.0 +2023-04-01 16:26:52,183 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5652, 2.0731, 1.9930, 2.6186, 2.3407, 2.0103, 2.1173, 2.4940], + device='cuda:0'), covar=tensor([0.0772, 0.1649, 0.1278, 0.0818, 0.1106, 0.0501, 0.1018, 0.0588], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0355, 0.0288, 0.0241, 0.0297, 0.0244, 0.0273, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:26:53,328 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1275, 1.7052, 1.5734, 2.0776, 1.8773, 1.7538, 1.7238, 1.9685], + device='cuda:0'), covar=tensor([0.0834, 0.1554, 0.1421, 0.0971, 0.1210, 0.0525, 0.1069, 0.0687], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0355, 0.0288, 0.0241, 0.0297, 0.0244, 0.0273, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:26:59,723 INFO [train.py:903] (0/4) Epoch 10, batch 1250, loss[loss=0.2549, simple_loss=0.3178, pruned_loss=0.09598, over 19659.00 frames. ], tot_loss[loss=0.2418, simple_loss=0.3129, pruned_loss=0.08537, over 3835740.93 frames. ], batch size: 53, lr: 8.49e-03, grad_scale: 4.0 +2023-04-01 16:27:08,208 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4447, 1.9618, 2.0252, 2.9459, 2.1764, 2.4848, 2.6954, 2.4558], + device='cuda:0'), covar=tensor([0.0683, 0.0923, 0.0978, 0.0810, 0.0805, 0.0754, 0.0820, 0.0641], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0229, 0.0227, 0.0255, 0.0239, 0.0215, 0.0199, 0.0208], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 16:28:00,861 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62750.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:28:01,690 INFO [train.py:903] (0/4) Epoch 10, batch 1300, loss[loss=0.2447, simple_loss=0.3206, pruned_loss=0.08442, over 18130.00 frames. ], tot_loss[loss=0.2429, simple_loss=0.3141, pruned_loss=0.08588, over 3822021.37 frames. ], batch size: 83, lr: 8.48e-03, grad_scale: 4.0 +2023-04-01 16:28:05,030 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62753.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:28:16,588 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.133e+02 5.064e+02 6.596e+02 8.862e+02 1.920e+03, threshold=1.319e+03, percent-clipped=1.0 +2023-04-01 16:28:51,844 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=62791.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:29:05,061 INFO [train.py:903] (0/4) Epoch 10, batch 1350, loss[loss=0.2245, simple_loss=0.292, pruned_loss=0.07853, over 19762.00 frames. ], tot_loss[loss=0.2434, simple_loss=0.3143, pruned_loss=0.0863, over 3809231.39 frames. ], batch size: 47, lr: 8.48e-03, grad_scale: 4.0 +2023-04-01 16:29:39,490 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-04-01 16:30:07,945 INFO [train.py:903] (0/4) Epoch 10, batch 1400, loss[loss=0.2637, simple_loss=0.3397, pruned_loss=0.09389, over 19713.00 frames. ], tot_loss[loss=0.2426, simple_loss=0.3136, pruned_loss=0.08579, over 3814397.52 frames. ], batch size: 63, lr: 8.48e-03, grad_scale: 4.0 +2023-04-01 16:30:13,090 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=62855.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:30:20,939 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.159e+02 5.591e+02 6.749e+02 8.217e+02 1.554e+03, threshold=1.350e+03, percent-clipped=4.0 +2023-04-01 16:30:27,994 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=62868.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:30:44,858 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=62880.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:30:47,426 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-01 16:31:07,169 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 16:31:09,388 INFO [train.py:903] (0/4) Epoch 10, batch 1450, loss[loss=0.2699, simple_loss=0.3371, pruned_loss=0.1014, over 19584.00 frames. ], tot_loss[loss=0.2424, simple_loss=0.3135, pruned_loss=0.08565, over 3811674.18 frames. ], batch size: 61, lr: 8.47e-03, grad_scale: 4.0 +2023-04-01 16:31:16,513 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62907.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:32:07,384 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=62947.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:32:11,887 INFO [train.py:903] (0/4) Epoch 10, batch 1500, loss[loss=0.2442, simple_loss=0.3213, pruned_loss=0.08352, over 18083.00 frames. ], tot_loss[loss=0.2426, simple_loss=0.3138, pruned_loss=0.08565, over 3802877.86 frames. ], batch size: 84, lr: 8.47e-03, grad_scale: 4.0 +2023-04-01 16:32:27,768 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.889e+02 5.675e+02 6.883e+02 8.252e+02 2.690e+03, threshold=1.377e+03, percent-clipped=4.0 +2023-04-01 16:33:13,784 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2936, 3.0186, 2.2030, 2.7784, 0.8778, 2.8771, 2.7925, 2.8810], + device='cuda:0'), covar=tensor([0.1065, 0.1302, 0.1919, 0.0898, 0.3500, 0.1009, 0.0954, 0.1293], + device='cuda:0'), in_proj_covar=tensor([0.0421, 0.0356, 0.0425, 0.0312, 0.0372, 0.0352, 0.0345, 0.0377], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 16:33:17,108 INFO [train.py:903] (0/4) Epoch 10, batch 1550, loss[loss=0.287, simple_loss=0.3559, pruned_loss=0.109, over 19092.00 frames. ], tot_loss[loss=0.2419, simple_loss=0.3134, pruned_loss=0.08517, over 3816228.04 frames. ], batch size: 69, lr: 8.47e-03, grad_scale: 4.0 +2023-04-01 16:33:23,634 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63006.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:33:40,047 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-01 16:33:42,939 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63022.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:33:49,615 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63028.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:33:53,291 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63031.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:34:20,244 INFO [train.py:903] (0/4) Epoch 10, batch 1600, loss[loss=0.2799, simple_loss=0.3346, pruned_loss=0.1126, over 19674.00 frames. ], tot_loss[loss=0.2408, simple_loss=0.3123, pruned_loss=0.08463, over 3824701.91 frames. ], batch size: 53, lr: 8.46e-03, grad_scale: 8.0 +2023-04-01 16:34:33,009 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.609e+02 5.367e+02 6.713e+02 8.573e+02 1.582e+03, threshold=1.343e+03, percent-clipped=2.0 +2023-04-01 16:34:33,423 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63062.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:34:42,293 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 16:35:21,351 INFO [train.py:903] (0/4) Epoch 10, batch 1650, loss[loss=0.2368, simple_loss=0.3144, pruned_loss=0.07964, over 18179.00 frames. ], tot_loss[loss=0.2418, simple_loss=0.313, pruned_loss=0.08534, over 3832436.42 frames. ], batch size: 83, lr: 8.46e-03, grad_scale: 8.0 +2023-04-01 16:35:51,856 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63124.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:35:58,416 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6905, 1.4881, 1.5206, 1.5356, 3.1868, 1.0536, 2.3964, 3.6218], + device='cuda:0'), covar=tensor([0.0435, 0.2392, 0.2463, 0.1697, 0.0728, 0.2457, 0.1132, 0.0234], + device='cuda:0'), in_proj_covar=tensor([0.0342, 0.0328, 0.0342, 0.0314, 0.0341, 0.0327, 0.0321, 0.0343], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:36:05,480 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63135.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:36:14,724 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63143.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:36:21,904 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63149.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:36:23,875 INFO [train.py:903] (0/4) Epoch 10, batch 1700, loss[loss=0.2675, simple_loss=0.3361, pruned_loss=0.0995, over 19788.00 frames. ], tot_loss[loss=0.2415, simple_loss=0.3125, pruned_loss=0.08519, over 3832597.78 frames. ], batch size: 56, lr: 8.46e-03, grad_scale: 8.0 +2023-04-01 16:36:38,455 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.334e+02 5.819e+02 7.178e+02 9.040e+02 2.117e+03, threshold=1.436e+03, percent-clipped=7.0 +2023-04-01 16:37:04,086 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 16:37:28,665 INFO [train.py:903] (0/4) Epoch 10, batch 1750, loss[loss=0.2403, simple_loss=0.3109, pruned_loss=0.08486, over 19604.00 frames. ], tot_loss[loss=0.241, simple_loss=0.3122, pruned_loss=0.08494, over 3835918.06 frames. ], batch size: 50, lr: 8.45e-03, grad_scale: 4.0 +2023-04-01 16:38:14,984 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63239.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:38:30,205 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63250.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:38:30,878 INFO [train.py:903] (0/4) Epoch 10, batch 1800, loss[loss=0.2316, simple_loss=0.2926, pruned_loss=0.08532, over 19347.00 frames. ], tot_loss[loss=0.241, simple_loss=0.312, pruned_loss=0.08495, over 3817969.38 frames. ], batch size: 47, lr: 8.45e-03, grad_scale: 4.0 +2023-04-01 16:38:44,564 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.600e+02 5.870e+02 7.659e+02 9.293e+02 2.596e+03, threshold=1.532e+03, percent-clipped=8.0 +2023-04-01 16:39:03,354 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63278.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:39:03,697 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.84 vs. limit=5.0 +2023-04-01 16:39:26,422 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8020, 3.3604, 1.8450, 1.6731, 3.0979, 1.4700, 1.0948, 2.0012], + device='cuda:0'), covar=tensor([0.1056, 0.0392, 0.0845, 0.0798, 0.0479, 0.1126, 0.0900, 0.0575], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0301, 0.0329, 0.0248, 0.0236, 0.0326, 0.0291, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:39:30,786 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 16:39:32,968 INFO [train.py:903] (0/4) Epoch 10, batch 1850, loss[loss=0.1908, simple_loss=0.2612, pruned_loss=0.06021, over 19075.00 frames. ], tot_loss[loss=0.2411, simple_loss=0.3124, pruned_loss=0.08485, over 3810662.97 frames. ], batch size: 42, lr: 8.45e-03, grad_scale: 4.0 +2023-04-01 16:39:35,801 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63303.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:39:54,102 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63318.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:40:10,177 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 16:40:19,588 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0911, 1.6632, 1.6132, 2.0450, 1.8795, 1.7609, 1.6405, 2.0718], + device='cuda:0'), covar=tensor([0.0816, 0.1507, 0.1372, 0.0905, 0.1116, 0.0500, 0.1098, 0.0547], + device='cuda:0'), in_proj_covar=tensor([0.0250, 0.0354, 0.0289, 0.0239, 0.0296, 0.0244, 0.0275, 0.0238], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 16:40:26,743 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63343.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:40:35,819 INFO [train.py:903] (0/4) Epoch 10, batch 1900, loss[loss=0.2081, simple_loss=0.2928, pruned_loss=0.06171, over 19771.00 frames. ], tot_loss[loss=0.2421, simple_loss=0.3135, pruned_loss=0.08531, over 3807755.32 frames. ], batch size: 56, lr: 8.44e-03, grad_scale: 4.0 +2023-04-01 16:40:52,657 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.343e+02 5.562e+02 7.038e+02 8.618e+02 1.834e+03, threshold=1.408e+03, percent-clipped=3.0 +2023-04-01 16:40:55,976 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 16:41:01,876 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 16:41:24,930 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 16:41:37,239 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63399.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:41:40,406 INFO [train.py:903] (0/4) Epoch 10, batch 1950, loss[loss=0.2575, simple_loss=0.3316, pruned_loss=0.09167, over 19524.00 frames. ], tot_loss[loss=0.2424, simple_loss=0.3134, pruned_loss=0.08568, over 3810458.08 frames. ], batch size: 64, lr: 8.44e-03, grad_scale: 4.0 +2023-04-01 16:42:10,547 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63424.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:42:27,969 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63438.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:42:31,447 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63440.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:42:44,768 INFO [train.py:903] (0/4) Epoch 10, batch 2000, loss[loss=0.2836, simple_loss=0.347, pruned_loss=0.1101, over 19743.00 frames. ], tot_loss[loss=0.2414, simple_loss=0.3127, pruned_loss=0.08503, over 3808631.63 frames. ], batch size: 63, lr: 8.44e-03, grad_scale: 8.0 +2023-04-01 16:43:00,250 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.731e+02 5.221e+02 6.641e+02 8.776e+02 2.044e+03, threshold=1.328e+03, percent-clipped=3.0 +2023-04-01 16:43:43,937 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 16:43:47,379 INFO [train.py:903] (0/4) Epoch 10, batch 2050, loss[loss=0.2263, simple_loss=0.3079, pruned_loss=0.07237, over 19496.00 frames. ], tot_loss[loss=0.2419, simple_loss=0.3131, pruned_loss=0.08538, over 3801173.95 frames. ], batch size: 64, lr: 8.43e-03, grad_scale: 8.0 +2023-04-01 16:43:53,614 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63506.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:44:03,935 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 16:44:05,931 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 16:44:26,404 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63531.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:44:28,433 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 16:44:50,581 INFO [train.py:903] (0/4) Epoch 10, batch 2100, loss[loss=0.2863, simple_loss=0.3581, pruned_loss=0.1073, over 19331.00 frames. ], tot_loss[loss=0.2418, simple_loss=0.3132, pruned_loss=0.08518, over 3799589.35 frames. ], batch size: 66, lr: 8.43e-03, grad_scale: 8.0 +2023-04-01 16:45:06,405 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.339e+02 5.645e+02 7.348e+02 9.688e+02 2.351e+03, threshold=1.470e+03, percent-clipped=4.0 +2023-04-01 16:45:26,369 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 16:45:32,492 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63583.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:45:37,346 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63587.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:45:46,623 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 16:45:54,834 INFO [train.py:903] (0/4) Epoch 10, batch 2150, loss[loss=0.2231, simple_loss=0.3019, pruned_loss=0.0722, over 19516.00 frames. ], tot_loss[loss=0.2431, simple_loss=0.3142, pruned_loss=0.08597, over 3803232.44 frames. ], batch size: 56, lr: 8.43e-03, grad_scale: 8.0 +2023-04-01 16:46:00,127 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63604.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:46:58,775 INFO [train.py:903] (0/4) Epoch 10, batch 2200, loss[loss=0.1835, simple_loss=0.2572, pruned_loss=0.05488, over 19759.00 frames. ], tot_loss[loss=0.2426, simple_loss=0.3134, pruned_loss=0.08586, over 3787790.90 frames. ], batch size: 45, lr: 8.42e-03, grad_scale: 8.0 +2023-04-01 16:47:13,920 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.200e+02 5.616e+02 6.875e+02 8.680e+02 1.983e+03, threshold=1.375e+03, percent-clipped=4.0 +2023-04-01 16:47:36,985 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9896, 2.2053, 2.3051, 2.3160, 1.0020, 2.1515, 1.9912, 2.1172], + device='cuda:0'), covar=tensor([0.1083, 0.1735, 0.0647, 0.0660, 0.3534, 0.0830, 0.0562, 0.0978], + device='cuda:0'), in_proj_covar=tensor([0.0635, 0.0575, 0.0767, 0.0642, 0.0704, 0.0512, 0.0469, 0.0703], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 16:47:50,343 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-04-01 16:47:59,060 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3360, 1.4201, 1.7967, 1.5684, 2.6710, 2.3088, 2.9261, 1.2183], + device='cuda:0'), covar=tensor([0.2139, 0.3709, 0.2163, 0.1729, 0.1397, 0.1715, 0.1384, 0.3465], + device='cuda:0'), in_proj_covar=tensor([0.0480, 0.0564, 0.0583, 0.0428, 0.0587, 0.0483, 0.0646, 0.0482], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 16:48:00,173 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63698.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:48:03,389 INFO [train.py:903] (0/4) Epoch 10, batch 2250, loss[loss=0.222, simple_loss=0.2891, pruned_loss=0.07748, over 19379.00 frames. ], tot_loss[loss=0.242, simple_loss=0.313, pruned_loss=0.08547, over 3801664.35 frames. ], batch size: 48, lr: 8.42e-03, grad_scale: 8.0 +2023-04-01 16:48:38,898 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7352, 3.8856, 4.1997, 4.2016, 2.4646, 3.9136, 3.6576, 3.9713], + device='cuda:0'), covar=tensor([0.1038, 0.2305, 0.0541, 0.0481, 0.3461, 0.0764, 0.0496, 0.0823], + device='cuda:0'), in_proj_covar=tensor([0.0634, 0.0575, 0.0767, 0.0639, 0.0704, 0.0512, 0.0470, 0.0701], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 16:48:44,201 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.50 vs. limit=2.0 +2023-04-01 16:48:51,732 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3263, 1.4083, 1.7469, 1.4638, 3.0792, 2.4786, 3.4651, 1.5667], + device='cuda:0'), covar=tensor([0.2290, 0.3820, 0.2404, 0.1899, 0.1399, 0.1767, 0.1349, 0.3257], + device='cuda:0'), in_proj_covar=tensor([0.0479, 0.0560, 0.0581, 0.0427, 0.0583, 0.0481, 0.0643, 0.0481], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 16:49:04,920 INFO [train.py:903] (0/4) Epoch 10, batch 2300, loss[loss=0.2349, simple_loss=0.3143, pruned_loss=0.07781, over 19669.00 frames. ], tot_loss[loss=0.2426, simple_loss=0.3137, pruned_loss=0.08568, over 3812151.80 frames. ], batch size: 55, lr: 8.42e-03, grad_scale: 4.0 +2023-04-01 16:49:18,691 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 16:49:23,050 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.591e+02 5.837e+02 6.965e+02 8.642e+02 2.205e+03, threshold=1.393e+03, percent-clipped=3.0 +2023-04-01 16:49:45,517 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63782.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:49:47,898 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63784.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:50:09,267 INFO [train.py:903] (0/4) Epoch 10, batch 2350, loss[loss=0.3091, simple_loss=0.3597, pruned_loss=0.1292, over 13099.00 frames. ], tot_loss[loss=0.2414, simple_loss=0.3131, pruned_loss=0.08485, over 3822079.33 frames. ], batch size: 135, lr: 8.41e-03, grad_scale: 4.0 +2023-04-01 16:50:44,281 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63828.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:50:50,877 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 16:51:10,038 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 16:51:11,956 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-04-01 16:51:13,622 INFO [train.py:903] (0/4) Epoch 10, batch 2400, loss[loss=0.2174, simple_loss=0.294, pruned_loss=0.0704, over 19764.00 frames. ], tot_loss[loss=0.2411, simple_loss=0.3125, pruned_loss=0.08485, over 3836030.40 frames. ], batch size: 54, lr: 8.41e-03, grad_scale: 8.0 +2023-04-01 16:51:29,387 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63863.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:51:30,159 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.763e+02 5.915e+02 7.022e+02 9.244e+02 2.907e+03, threshold=1.404e+03, percent-clipped=10.0 +2023-04-01 16:51:55,405 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7415, 1.8116, 2.0581, 2.4885, 1.6429, 2.2478, 2.3492, 1.9599], + device='cuda:0'), covar=tensor([0.3370, 0.2749, 0.1367, 0.1440, 0.2936, 0.1336, 0.3050, 0.2393], + device='cuda:0'), in_proj_covar=tensor([0.0772, 0.0780, 0.0638, 0.0882, 0.0763, 0.0683, 0.0781, 0.0695], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 16:52:03,543 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.64 vs. limit=2.0 +2023-04-01 16:52:12,814 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63897.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:52:15,904 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=63899.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:52:17,970 INFO [train.py:903] (0/4) Epoch 10, batch 2450, loss[loss=0.2536, simple_loss=0.3303, pruned_loss=0.08849, over 17447.00 frames. ], tot_loss[loss=0.2403, simple_loss=0.3121, pruned_loss=0.08425, over 3837621.75 frames. ], batch size: 101, lr: 8.41e-03, grad_scale: 8.0 +2023-04-01 16:52:55,521 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63931.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:52:59,271 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=63934.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:53:16,593 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=63948.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:53:19,963 INFO [train.py:903] (0/4) Epoch 10, batch 2500, loss[loss=0.2166, simple_loss=0.2987, pruned_loss=0.06725, over 19577.00 frames. ], tot_loss[loss=0.2393, simple_loss=0.3114, pruned_loss=0.08364, over 3826870.46 frames. ], batch size: 52, lr: 8.40e-03, grad_scale: 8.0 +2023-04-01 16:53:23,766 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=63954.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:53:35,922 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.566e+02 5.261e+02 6.826e+02 9.233e+02 1.687e+03, threshold=1.365e+03, percent-clipped=6.0 +2023-04-01 16:53:56,230 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=63979.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:54:00,224 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.37 vs. limit=2.0 +2023-04-01 16:54:21,195 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-64000.pt +2023-04-01 16:54:23,068 INFO [train.py:903] (0/4) Epoch 10, batch 2550, loss[loss=0.2505, simple_loss=0.3307, pruned_loss=0.08518, over 19590.00 frames. ], tot_loss[loss=0.24, simple_loss=0.3118, pruned_loss=0.08409, over 3829920.57 frames. ], batch size: 57, lr: 8.40e-03, grad_scale: 8.0 +2023-04-01 16:55:13,541 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64041.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:55:15,690 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 16:55:20,769 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64046.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:55:26,459 INFO [train.py:903] (0/4) Epoch 10, batch 2600, loss[loss=0.244, simple_loss=0.3167, pruned_loss=0.08568, over 19683.00 frames. ], tot_loss[loss=0.241, simple_loss=0.3124, pruned_loss=0.0848, over 3812273.05 frames. ], batch size: 59, lr: 8.40e-03, grad_scale: 8.0 +2023-04-01 16:55:41,492 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64063.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:55:42,249 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.625e+02 5.188e+02 6.411e+02 7.864e+02 1.888e+03, threshold=1.282e+03, percent-clipped=1.0 +2023-04-01 16:56:27,597 INFO [train.py:903] (0/4) Epoch 10, batch 2650, loss[loss=0.298, simple_loss=0.3608, pruned_loss=0.1176, over 19516.00 frames. ], tot_loss[loss=0.2434, simple_loss=0.3143, pruned_loss=0.0863, over 3800690.99 frames. ], batch size: 54, lr: 8.39e-03, grad_scale: 8.0 +2023-04-01 16:56:45,343 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 16:57:23,814 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6894, 4.1818, 4.4348, 4.4187, 1.4674, 4.1120, 3.6069, 4.1294], + device='cuda:0'), covar=tensor([0.1375, 0.0765, 0.0573, 0.0573, 0.5506, 0.0578, 0.0613, 0.1083], + device='cuda:0'), in_proj_covar=tensor([0.0640, 0.0579, 0.0767, 0.0639, 0.0713, 0.0517, 0.0475, 0.0706], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 16:57:29,267 INFO [train.py:903] (0/4) Epoch 10, batch 2700, loss[loss=0.2551, simple_loss=0.3252, pruned_loss=0.09254, over 17625.00 frames. ], tot_loss[loss=0.2432, simple_loss=0.314, pruned_loss=0.0862, over 3804523.25 frames. ], batch size: 101, lr: 8.39e-03, grad_scale: 8.0 +2023-04-01 16:57:32,906 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64153.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:57:35,308 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64155.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:57:45,502 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.698e+02 5.891e+02 6.747e+02 8.779e+02 3.257e+03, threshold=1.349e+03, percent-clipped=11.0 +2023-04-01 16:57:56,898 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64172.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:58:04,097 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64178.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:58:07,373 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64180.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:58:33,076 INFO [train.py:903] (0/4) Epoch 10, batch 2750, loss[loss=0.2737, simple_loss=0.3411, pruned_loss=0.1032, over 18209.00 frames. ], tot_loss[loss=0.2412, simple_loss=0.3124, pruned_loss=0.08501, over 3803178.35 frames. ], batch size: 83, lr: 8.39e-03, grad_scale: 8.0 +2023-04-01 16:58:41,438 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64207.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:58:57,441 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2154, 1.2427, 1.6131, 1.3926, 2.5941, 2.0486, 2.7639, 1.0840], + device='cuda:0'), covar=tensor([0.2473, 0.4116, 0.2419, 0.1957, 0.1412, 0.1997, 0.1384, 0.3751], + device='cuda:0'), in_proj_covar=tensor([0.0481, 0.0564, 0.0583, 0.0428, 0.0587, 0.0483, 0.0646, 0.0484], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 16:59:36,101 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64250.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 16:59:36,916 INFO [train.py:903] (0/4) Epoch 10, batch 2800, loss[loss=0.2448, simple_loss=0.314, pruned_loss=0.08779, over 17223.00 frames. ], tot_loss[loss=0.2396, simple_loss=0.3112, pruned_loss=0.08403, over 3815804.52 frames. ], batch size: 101, lr: 8.38e-03, grad_scale: 8.0 +2023-04-01 16:59:52,863 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.388e+02 5.243e+02 6.695e+02 7.923e+02 2.040e+03, threshold=1.339e+03, percent-clipped=2.0 +2023-04-01 17:00:10,647 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64278.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:00:13,398 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3627, 1.3995, 1.6535, 1.5209, 2.4942, 2.1329, 2.5263, 1.0234], + device='cuda:0'), covar=tensor([0.2130, 0.3762, 0.2259, 0.1733, 0.1367, 0.1820, 0.1396, 0.3589], + device='cuda:0'), in_proj_covar=tensor([0.0479, 0.0564, 0.0582, 0.0427, 0.0587, 0.0481, 0.0644, 0.0483], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 17:00:21,802 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64287.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:00:40,323 INFO [train.py:903] (0/4) Epoch 10, batch 2850, loss[loss=0.2077, simple_loss=0.283, pruned_loss=0.06619, over 19700.00 frames. ], tot_loss[loss=0.239, simple_loss=0.3107, pruned_loss=0.08358, over 3821818.90 frames. ], batch size: 53, lr: 8.38e-03, grad_scale: 8.0 +2023-04-01 17:00:41,952 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64302.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:00:59,518 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3622, 1.8155, 1.6233, 2.5711, 1.9083, 2.4996, 2.6904, 2.5406], + device='cuda:0'), covar=tensor([0.0691, 0.0900, 0.0980, 0.0979, 0.0942, 0.0693, 0.0783, 0.0571], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0223, 0.0223, 0.0252, 0.0236, 0.0212, 0.0199, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 17:01:03,234 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64319.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:01:06,434 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64322.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:01:11,307 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5474, 1.2037, 1.1279, 1.3458, 1.1694, 1.1983, 0.9967, 1.2814], + device='cuda:0'), covar=tensor([0.0984, 0.1194, 0.1520, 0.0948, 0.1152, 0.0776, 0.1480, 0.0857], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0352, 0.0289, 0.0239, 0.0299, 0.0243, 0.0275, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 17:01:12,442 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64327.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:01:35,570 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64344.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:01:39,732 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 17:01:43,220 INFO [train.py:903] (0/4) Epoch 10, batch 2900, loss[loss=0.2243, simple_loss=0.2857, pruned_loss=0.08142, over 19793.00 frames. ], tot_loss[loss=0.2383, simple_loss=0.3104, pruned_loss=0.08316, over 3829573.24 frames. ], batch size: 49, lr: 8.38e-03, grad_scale: 8.0 +2023-04-01 17:01:58,206 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64363.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:01:58,969 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.152e+02 6.015e+02 7.349e+02 1.031e+03 2.008e+03, threshold=1.470e+03, percent-clipped=12.0 +2023-04-01 17:02:15,485 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3721, 1.2904, 1.4322, 1.5945, 2.9133, 1.1496, 2.0592, 3.1794], + device='cuda:0'), covar=tensor([0.0418, 0.2729, 0.2734, 0.1645, 0.0751, 0.2324, 0.1275, 0.0330], + device='cuda:0'), in_proj_covar=tensor([0.0339, 0.0331, 0.0345, 0.0315, 0.0339, 0.0328, 0.0321, 0.0343], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 17:02:26,618 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64385.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:02:36,405 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64393.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:02:46,691 INFO [train.py:903] (0/4) Epoch 10, batch 2950, loss[loss=0.2665, simple_loss=0.3404, pruned_loss=0.09631, over 19670.00 frames. ], tot_loss[loss=0.239, simple_loss=0.311, pruned_loss=0.08346, over 3833240.30 frames. ], batch size: 58, lr: 8.37e-03, grad_scale: 8.0 +2023-04-01 17:03:06,095 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.75 vs. limit=5.0 +2023-04-01 17:03:48,251 INFO [train.py:903] (0/4) Epoch 10, batch 3000, loss[loss=0.2213, simple_loss=0.3006, pruned_loss=0.07094, over 19684.00 frames. ], tot_loss[loss=0.2389, simple_loss=0.3111, pruned_loss=0.08336, over 3836714.05 frames. ], batch size: 59, lr: 8.37e-03, grad_scale: 8.0 +2023-04-01 17:03:48,252 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 17:04:00,870 INFO [train.py:937] (0/4) Epoch 10, validation: loss=0.1811, simple_loss=0.2816, pruned_loss=0.04036, over 944034.00 frames. +2023-04-01 17:04:00,871 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 17:04:02,458 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6359, 1.6512, 1.5022, 1.2761, 1.1309, 1.3138, 0.2286, 0.6043], + device='cuda:0'), covar=tensor([0.0551, 0.0554, 0.0322, 0.0484, 0.1052, 0.0605, 0.0868, 0.0889], + device='cuda:0'), in_proj_covar=tensor([0.0326, 0.0327, 0.0325, 0.0344, 0.0416, 0.0340, 0.0303, 0.0321], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 17:04:04,335 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 17:04:13,260 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.74 vs. limit=2.0 +2023-04-01 17:04:18,125 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.319e+02 5.533e+02 6.661e+02 8.174e+02 1.809e+03, threshold=1.332e+03, percent-clipped=2.0 +2023-04-01 17:04:41,351 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64483.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:05:03,197 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64500.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:05:04,019 INFO [train.py:903] (0/4) Epoch 10, batch 3050, loss[loss=0.2013, simple_loss=0.2681, pruned_loss=0.06726, over 19737.00 frames. ], tot_loss[loss=0.2389, simple_loss=0.311, pruned_loss=0.08335, over 3846287.81 frames. ], batch size: 46, lr: 8.37e-03, grad_scale: 8.0 +2023-04-01 17:05:48,629 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8133, 1.5131, 1.5203, 2.1264, 1.8139, 2.0913, 2.0458, 1.9003], + device='cuda:0'), covar=tensor([0.0791, 0.0996, 0.1024, 0.0825, 0.0845, 0.0728, 0.0854, 0.0683], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0226, 0.0225, 0.0253, 0.0239, 0.0216, 0.0202, 0.0208], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 17:05:57,723 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3974, 1.8969, 1.8467, 2.6883, 2.2213, 2.5289, 2.6477, 2.4290], + device='cuda:0'), covar=tensor([0.0691, 0.0890, 0.0948, 0.0836, 0.0868, 0.0654, 0.0812, 0.0580], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0226, 0.0225, 0.0254, 0.0240, 0.0216, 0.0202, 0.0208], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 17:05:57,751 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64543.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:06:06,591 INFO [train.py:903] (0/4) Epoch 10, batch 3100, loss[loss=0.2764, simple_loss=0.342, pruned_loss=0.1054, over 18744.00 frames. ], tot_loss[loss=0.239, simple_loss=0.3111, pruned_loss=0.08342, over 3845952.05 frames. ], batch size: 74, lr: 8.37e-03, grad_scale: 8.0 +2023-04-01 17:06:22,863 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.367e+02 5.961e+02 7.216e+02 8.690e+02 2.208e+03, threshold=1.443e+03, percent-clipped=3.0 +2023-04-01 17:06:27,980 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64568.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:06:40,849 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64578.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:07:01,269 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64594.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:07:10,478 INFO [train.py:903] (0/4) Epoch 10, batch 3150, loss[loss=0.1899, simple_loss=0.2638, pruned_loss=0.05796, over 19285.00 frames. ], tot_loss[loss=0.2379, simple_loss=0.3101, pruned_loss=0.08287, over 3848473.92 frames. ], batch size: 44, lr: 8.36e-03, grad_scale: 8.0 +2023-04-01 17:07:13,030 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64603.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:07:17,450 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64607.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:07:21,738 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5961, 4.0847, 4.2512, 4.2380, 1.4986, 3.9442, 3.5080, 3.9467], + device='cuda:0'), covar=tensor([0.1268, 0.0672, 0.0511, 0.0522, 0.4757, 0.0570, 0.0545, 0.0962], + device='cuda:0'), in_proj_covar=tensor([0.0645, 0.0581, 0.0768, 0.0645, 0.0713, 0.0518, 0.0476, 0.0712], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 17:07:31,625 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-04-01 17:07:39,773 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 17:08:10,662 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64649.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:08:12,600 INFO [train.py:903] (0/4) Epoch 10, batch 3200, loss[loss=0.2534, simple_loss=0.3329, pruned_loss=0.087, over 19673.00 frames. ], tot_loss[loss=0.2373, simple_loss=0.3098, pruned_loss=0.08243, over 3853060.87 frames. ], batch size: 60, lr: 8.36e-03, grad_scale: 8.0 +2023-04-01 17:08:30,168 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.006e+02 5.483e+02 6.754e+02 8.204e+02 1.644e+03, threshold=1.351e+03, percent-clipped=2.0 +2023-04-01 17:08:32,795 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64666.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:08:43,793 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64674.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:09:16,028 INFO [train.py:903] (0/4) Epoch 10, batch 3250, loss[loss=0.1996, simple_loss=0.2743, pruned_loss=0.06249, over 19372.00 frames. ], tot_loss[loss=0.2378, simple_loss=0.3103, pruned_loss=0.08267, over 3849712.44 frames. ], batch size: 47, lr: 8.36e-03, grad_scale: 8.0 +2023-04-01 17:09:24,031 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64707.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:09:26,661 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64709.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:09:42,842 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64721.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:10:20,336 INFO [train.py:903] (0/4) Epoch 10, batch 3300, loss[loss=0.2704, simple_loss=0.3405, pruned_loss=0.1001, over 19530.00 frames. ], tot_loss[loss=0.2389, simple_loss=0.3114, pruned_loss=0.08319, over 3837856.25 frames. ], batch size: 56, lr: 8.35e-03, grad_scale: 8.0 +2023-04-01 17:10:26,800 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64756.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:10:28,776 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 17:10:35,495 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64763.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:10:37,431 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.278e+02 5.384e+02 6.605e+02 8.281e+02 2.311e+03, threshold=1.321e+03, percent-clipped=9.0 +2023-04-01 17:10:43,924 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64770.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:10:57,727 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64781.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:11:23,426 INFO [train.py:903] (0/4) Epoch 10, batch 3350, loss[loss=0.2219, simple_loss=0.2846, pruned_loss=0.07961, over 19783.00 frames. ], tot_loss[loss=0.2395, simple_loss=0.3116, pruned_loss=0.08371, over 3844464.08 frames. ], batch size: 49, lr: 8.35e-03, grad_scale: 4.0 +2023-04-01 17:11:49,545 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64822.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:11:55,098 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64827.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:12:04,623 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6602, 1.4782, 1.3916, 1.9392, 1.5556, 1.9430, 1.8900, 1.8012], + device='cuda:0'), covar=tensor([0.0788, 0.0916, 0.1008, 0.0774, 0.0840, 0.0746, 0.0823, 0.0652], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0227, 0.0225, 0.0253, 0.0239, 0.0215, 0.0201, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 17:12:24,011 INFO [train.py:903] (0/4) Epoch 10, batch 3400, loss[loss=0.2378, simple_loss=0.3151, pruned_loss=0.08026, over 18156.00 frames. ], tot_loss[loss=0.2406, simple_loss=0.3119, pruned_loss=0.08461, over 3844819.41 frames. ], batch size: 83, lr: 8.35e-03, grad_scale: 4.0 +2023-04-01 17:12:40,676 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2419, 1.3211, 1.9342, 1.5002, 2.4434, 1.9916, 2.5631, 1.1686], + device='cuda:0'), covar=tensor([0.2313, 0.3923, 0.2052, 0.1836, 0.1656, 0.2157, 0.1892, 0.3680], + device='cuda:0'), in_proj_covar=tensor([0.0480, 0.0566, 0.0583, 0.0428, 0.0587, 0.0483, 0.0645, 0.0482], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 17:12:42,252 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.893e+02 5.724e+02 7.342e+02 8.665e+02 1.913e+03, threshold=1.468e+03, percent-clipped=6.0 +2023-04-01 17:12:59,015 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9574, 1.1585, 1.3505, 1.2892, 2.4548, 0.9226, 1.6991, 2.7933], + device='cuda:0'), covar=tensor([0.0633, 0.2852, 0.2901, 0.1872, 0.0968, 0.2606, 0.1529, 0.0449], + device='cuda:0'), in_proj_covar=tensor([0.0342, 0.0330, 0.0344, 0.0316, 0.0341, 0.0329, 0.0322, 0.0343], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 17:13:27,824 INFO [train.py:903] (0/4) Epoch 10, batch 3450, loss[loss=0.2398, simple_loss=0.3193, pruned_loss=0.0802, over 19763.00 frames. ], tot_loss[loss=0.2407, simple_loss=0.3123, pruned_loss=0.08455, over 3826594.76 frames. ], batch size: 54, lr: 8.34e-03, grad_scale: 4.0 +2023-04-01 17:13:35,102 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 17:14:20,585 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=64942.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:14:30,662 INFO [train.py:903] (0/4) Epoch 10, batch 3500, loss[loss=0.2759, simple_loss=0.3342, pruned_loss=0.1088, over 18091.00 frames. ], tot_loss[loss=0.2409, simple_loss=0.3124, pruned_loss=0.0847, over 3817309.52 frames. ], batch size: 83, lr: 8.34e-03, grad_scale: 4.0 +2023-04-01 17:14:31,778 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=64951.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:14:48,653 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.495e+02 5.616e+02 6.822e+02 8.996e+02 1.764e+03, threshold=1.364e+03, percent-clipped=1.0 +2023-04-01 17:14:49,030 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=64965.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:14:59,244 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0778, 5.0723, 5.8934, 5.8740, 1.8904, 5.5885, 4.7529, 5.4690], + device='cuda:0'), covar=tensor([0.1267, 0.0663, 0.0521, 0.0469, 0.5194, 0.0405, 0.0499, 0.1100], + device='cuda:0'), in_proj_covar=tensor([0.0633, 0.0573, 0.0756, 0.0637, 0.0700, 0.0513, 0.0470, 0.0697], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 17:15:19,880 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=64990.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:15:25,269 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=64994.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:15:34,048 INFO [train.py:903] (0/4) Epoch 10, batch 3550, loss[loss=0.2443, simple_loss=0.3033, pruned_loss=0.09271, over 19742.00 frames. ], tot_loss[loss=0.2404, simple_loss=0.3121, pruned_loss=0.08436, over 3822359.35 frames. ], batch size: 46, lr: 8.34e-03, grad_scale: 4.0 +2023-04-01 17:15:44,361 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65010.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:16:34,666 INFO [train.py:903] (0/4) Epoch 10, batch 3600, loss[loss=0.2451, simple_loss=0.3326, pruned_loss=0.07878, over 19676.00 frames. ], tot_loss[loss=0.2398, simple_loss=0.3112, pruned_loss=0.08418, over 3806267.30 frames. ], batch size: 55, lr: 8.33e-03, grad_scale: 8.0 +2023-04-01 17:16:52,000 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.218e+02 5.791e+02 6.813e+02 8.568e+02 1.743e+03, threshold=1.363e+03, percent-clipped=4.0 +2023-04-01 17:16:52,163 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65065.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:16:53,551 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65066.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:17:02,326 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65073.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:17:08,111 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65078.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:17:35,956 INFO [train.py:903] (0/4) Epoch 10, batch 3650, loss[loss=0.217, simple_loss=0.2981, pruned_loss=0.06797, over 19846.00 frames. ], tot_loss[loss=0.2399, simple_loss=0.3115, pruned_loss=0.08422, over 3789383.56 frames. ], batch size: 52, lr: 8.33e-03, grad_scale: 8.0 +2023-04-01 17:17:38,604 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65103.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:17:42,739 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65107.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:17:51,899 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65114.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:18:05,484 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65125.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:18:36,262 INFO [train.py:903] (0/4) Epoch 10, batch 3700, loss[loss=0.2061, simple_loss=0.2808, pruned_loss=0.06569, over 19763.00 frames. ], tot_loss[loss=0.2411, simple_loss=0.3122, pruned_loss=0.08503, over 3791401.76 frames. ], batch size: 54, lr: 8.33e-03, grad_scale: 8.0 +2023-04-01 17:18:53,971 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.395e+02 5.629e+02 7.088e+02 8.754e+02 1.818e+03, threshold=1.418e+03, percent-clipped=5.0 +2023-04-01 17:19:02,242 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-04-01 17:19:12,177 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65180.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:19:33,192 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65198.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:19:37,055 INFO [train.py:903] (0/4) Epoch 10, batch 3750, loss[loss=0.2524, simple_loss=0.3264, pruned_loss=0.08923, over 18114.00 frames. ], tot_loss[loss=0.2429, simple_loss=0.3137, pruned_loss=0.08607, over 3792879.63 frames. ], batch size: 83, lr: 8.32e-03, grad_scale: 8.0 +2023-04-01 17:19:49,767 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-01 17:20:02,498 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65222.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:20:03,645 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65223.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:20:10,495 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65229.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:20:36,858 INFO [train.py:903] (0/4) Epoch 10, batch 3800, loss[loss=0.2409, simple_loss=0.3147, pruned_loss=0.08348, over 19647.00 frames. ], tot_loss[loss=0.2443, simple_loss=0.3151, pruned_loss=0.08674, over 3787142.28 frames. ], batch size: 55, lr: 8.32e-03, grad_scale: 8.0 +2023-04-01 17:20:54,008 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.166e+02 5.620e+02 6.620e+02 8.005e+02 1.692e+03, threshold=1.324e+03, percent-clipped=4.0 +2023-04-01 17:21:06,349 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 17:21:20,019 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.23 vs. limit=5.0 +2023-04-01 17:21:20,174 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-04-01 17:21:37,764 INFO [train.py:903] (0/4) Epoch 10, batch 3850, loss[loss=0.2007, simple_loss=0.2747, pruned_loss=0.06337, over 15143.00 frames. ], tot_loss[loss=0.244, simple_loss=0.3151, pruned_loss=0.08648, over 3793996.90 frames. ], batch size: 33, lr: 8.32e-03, grad_scale: 8.0 +2023-04-01 17:21:40,381 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65303.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:22:03,660 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65322.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:22:21,617 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65338.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:22:33,961 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65347.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:22:37,972 INFO [train.py:903] (0/4) Epoch 10, batch 3900, loss[loss=0.2139, simple_loss=0.2822, pruned_loss=0.07281, over 19749.00 frames. ], tot_loss[loss=0.2429, simple_loss=0.3142, pruned_loss=0.08576, over 3797376.85 frames. ], batch size: 46, lr: 8.31e-03, grad_scale: 8.0 +2023-04-01 17:22:55,833 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.027e+02 5.653e+02 7.180e+02 9.093e+02 1.633e+03, threshold=1.436e+03, percent-clipped=2.0 +2023-04-01 17:23:14,896 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65381.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:23:32,382 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-04-01 17:23:40,380 INFO [train.py:903] (0/4) Epoch 10, batch 3950, loss[loss=0.2106, simple_loss=0.2771, pruned_loss=0.07207, over 19073.00 frames. ], tot_loss[loss=0.2407, simple_loss=0.312, pruned_loss=0.08472, over 3796155.07 frames. ], batch size: 42, lr: 8.31e-03, grad_scale: 8.0 +2023-04-01 17:23:40,415 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 17:23:46,482 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65406.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:23:58,940 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65417.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:24:23,326 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65436.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:24:41,325 INFO [train.py:903] (0/4) Epoch 10, batch 4000, loss[loss=0.2348, simple_loss=0.3039, pruned_loss=0.0828, over 19576.00 frames. ], tot_loss[loss=0.2413, simple_loss=0.313, pruned_loss=0.08482, over 3798300.31 frames. ], batch size: 52, lr: 8.31e-03, grad_scale: 8.0 +2023-04-01 17:24:44,028 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65453.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:24:54,635 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65461.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:24:58,706 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.016e+02 5.448e+02 7.265e+02 9.508e+02 1.942e+03, threshold=1.453e+03, percent-clipped=2.0 +2023-04-01 17:25:14,537 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65478.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:25:23,061 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 17:25:23,384 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65485.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:25:41,795 INFO [train.py:903] (0/4) Epoch 10, batch 4050, loss[loss=0.2789, simple_loss=0.3441, pruned_loss=0.1068, over 18099.00 frames. ], tot_loss[loss=0.2414, simple_loss=0.313, pruned_loss=0.08485, over 3790469.53 frames. ], batch size: 83, lr: 8.30e-03, grad_scale: 8.0 +2023-04-01 17:25:45,297 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65503.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:25:53,252 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65510.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:26:19,256 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65532.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:26:40,676 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65550.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:26:41,431 INFO [train.py:903] (0/4) Epoch 10, batch 4100, loss[loss=0.2768, simple_loss=0.3472, pruned_loss=0.1032, over 18047.00 frames. ], tot_loss[loss=0.2437, simple_loss=0.3148, pruned_loss=0.08633, over 3794512.31 frames. ], batch size: 83, lr: 8.30e-03, grad_scale: 8.0 +2023-04-01 17:26:59,041 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.532e+02 5.925e+02 7.032e+02 8.463e+02 2.911e+03, threshold=1.406e+03, percent-clipped=6.0 +2023-04-01 17:27:11,838 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 17:27:42,545 INFO [train.py:903] (0/4) Epoch 10, batch 4150, loss[loss=0.1897, simple_loss=0.2642, pruned_loss=0.0576, over 19734.00 frames. ], tot_loss[loss=0.2415, simple_loss=0.3127, pruned_loss=0.08518, over 3801043.28 frames. ], batch size: 46, lr: 8.30e-03, grad_scale: 8.0 +2023-04-01 17:28:39,191 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65647.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:28:43,357 INFO [train.py:903] (0/4) Epoch 10, batch 4200, loss[loss=0.2434, simple_loss=0.3007, pruned_loss=0.09303, over 19689.00 frames. ], tot_loss[loss=0.2396, simple_loss=0.3111, pruned_loss=0.08399, over 3814854.05 frames. ], batch size: 53, lr: 8.30e-03, grad_scale: 8.0 +2023-04-01 17:28:43,405 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 17:28:59,415 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.075e+02 5.851e+02 6.725e+02 9.221e+02 2.199e+03, threshold=1.345e+03, percent-clipped=6.0 +2023-04-01 17:29:02,913 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.64 vs. limit=5.0 +2023-04-01 17:29:42,450 INFO [train.py:903] (0/4) Epoch 10, batch 4250, loss[loss=0.2419, simple_loss=0.3086, pruned_loss=0.08757, over 19623.00 frames. ], tot_loss[loss=0.2391, simple_loss=0.3107, pruned_loss=0.08377, over 3833753.76 frames. ], batch size: 50, lr: 8.29e-03, grad_scale: 8.0 +2023-04-01 17:29:54,090 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65709.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:29:55,765 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 17:30:06,600 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 17:30:22,834 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65734.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:30:43,321 INFO [train.py:903] (0/4) Epoch 10, batch 4300, loss[loss=0.2533, simple_loss=0.3332, pruned_loss=0.08664, over 19658.00 frames. ], tot_loss[loss=0.24, simple_loss=0.3117, pruned_loss=0.08412, over 3844038.70 frames. ], batch size: 55, lr: 8.29e-03, grad_scale: 8.0 +2023-04-01 17:30:56,215 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=65762.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:31:00,063 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.946e+02 5.358e+02 7.223e+02 8.854e+02 2.636e+03, threshold=1.445e+03, percent-clipped=3.0 +2023-04-01 17:31:27,986 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=65788.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:31:35,317 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 17:31:43,143 INFO [train.py:903] (0/4) Epoch 10, batch 4350, loss[loss=0.252, simple_loss=0.3233, pruned_loss=0.09037, over 19741.00 frames. ], tot_loss[loss=0.2403, simple_loss=0.3117, pruned_loss=0.08445, over 3852377.15 frames. ], batch size: 63, lr: 8.29e-03, grad_scale: 8.0 +2023-04-01 17:31:58,631 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=65813.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:32:44,687 INFO [train.py:903] (0/4) Epoch 10, batch 4400, loss[loss=0.2155, simple_loss=0.2801, pruned_loss=0.07541, over 19716.00 frames. ], tot_loss[loss=0.2395, simple_loss=0.3107, pruned_loss=0.08412, over 3856636.92 frames. ], batch size: 46, lr: 8.28e-03, grad_scale: 8.0 +2023-04-01 17:32:50,620 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=65856.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:33:00,118 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.366e+02 5.748e+02 7.554e+02 9.760e+02 1.805e+03, threshold=1.511e+03, percent-clipped=4.0 +2023-04-01 17:33:07,121 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7509, 1.7533, 1.5128, 1.3627, 1.2769, 1.4119, 0.2263, 0.6739], + device='cuda:0'), covar=tensor([0.0386, 0.0371, 0.0233, 0.0355, 0.0816, 0.0446, 0.0752, 0.0655], + device='cuda:0'), in_proj_covar=tensor([0.0326, 0.0326, 0.0325, 0.0347, 0.0418, 0.0344, 0.0303, 0.0317], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 17:33:10,962 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 17:33:16,848 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6705, 2.0910, 1.9121, 2.8325, 2.6329, 2.4177, 2.2199, 2.9493], + device='cuda:0'), covar=tensor([0.0810, 0.1736, 0.1442, 0.0930, 0.1181, 0.0416, 0.1050, 0.0465], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0352, 0.0293, 0.0240, 0.0297, 0.0244, 0.0276, 0.0237], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 17:33:19,853 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 17:33:36,071 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=65894.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:33:44,051 INFO [train.py:903] (0/4) Epoch 10, batch 4450, loss[loss=0.2403, simple_loss=0.3219, pruned_loss=0.07941, over 19411.00 frames. ], tot_loss[loss=0.2404, simple_loss=0.3119, pruned_loss=0.08443, over 3850163.06 frames. ], batch size: 66, lr: 8.28e-03, grad_scale: 8.0 +2023-04-01 17:34:45,020 INFO [train.py:903] (0/4) Epoch 10, batch 4500, loss[loss=0.2657, simple_loss=0.3317, pruned_loss=0.09986, over 13985.00 frames. ], tot_loss[loss=0.2397, simple_loss=0.3111, pruned_loss=0.08412, over 3834668.79 frames. ], batch size: 136, lr: 8.28e-03, grad_scale: 8.0 +2023-04-01 17:35:01,306 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.423e+02 5.376e+02 6.626e+02 8.325e+02 1.832e+03, threshold=1.325e+03, percent-clipped=3.0 +2023-04-01 17:35:14,782 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7596, 4.2946, 2.4056, 3.7984, 1.0562, 4.0811, 4.0326, 4.1710], + device='cuda:0'), covar=tensor([0.0612, 0.1150, 0.2105, 0.0727, 0.3997, 0.0774, 0.0805, 0.0957], + device='cuda:0'), in_proj_covar=tensor([0.0423, 0.0362, 0.0425, 0.0316, 0.0376, 0.0355, 0.0351, 0.0381], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 17:35:43,036 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2363, 1.2957, 1.4275, 1.4102, 1.8023, 1.8282, 1.7874, 0.5818], + device='cuda:0'), covar=tensor([0.2167, 0.3733, 0.2307, 0.1742, 0.1323, 0.1961, 0.1296, 0.3567], + device='cuda:0'), in_proj_covar=tensor([0.0481, 0.0561, 0.0585, 0.0430, 0.0587, 0.0486, 0.0645, 0.0484], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 17:35:44,998 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-66000.pt +2023-04-01 17:35:47,224 INFO [train.py:903] (0/4) Epoch 10, batch 4550, loss[loss=0.2519, simple_loss=0.3288, pruned_loss=0.0875, over 19337.00 frames. ], tot_loss[loss=0.2392, simple_loss=0.3112, pruned_loss=0.08363, over 3829356.53 frames. ], batch size: 66, lr: 8.27e-03, grad_scale: 8.0 +2023-04-01 17:35:56,014 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 17:35:56,360 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66009.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:36:06,973 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66018.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:36:17,574 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 17:36:36,243 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66043.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:36:45,879 INFO [train.py:903] (0/4) Epoch 10, batch 4600, loss[loss=0.3007, simple_loss=0.3611, pruned_loss=0.1201, over 19602.00 frames. ], tot_loss[loss=0.2394, simple_loss=0.311, pruned_loss=0.08393, over 3819003.65 frames. ], batch size: 57, lr: 8.27e-03, grad_scale: 8.0 +2023-04-01 17:37:00,874 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.957e+02 5.689e+02 6.858e+02 9.462e+02 1.667e+03, threshold=1.372e+03, percent-clipped=8.0 +2023-04-01 17:37:02,832 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-04-01 17:37:43,792 INFO [train.py:903] (0/4) Epoch 10, batch 4650, loss[loss=0.2088, simple_loss=0.2813, pruned_loss=0.06811, over 19391.00 frames. ], tot_loss[loss=0.2397, simple_loss=0.311, pruned_loss=0.08414, over 3826920.17 frames. ], batch size: 48, lr: 8.27e-03, grad_scale: 8.0 +2023-04-01 17:38:00,786 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 17:38:10,782 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 17:38:44,405 INFO [train.py:903] (0/4) Epoch 10, batch 4700, loss[loss=0.2563, simple_loss=0.3318, pruned_loss=0.09044, over 19617.00 frames. ], tot_loss[loss=0.2398, simple_loss=0.3114, pruned_loss=0.08414, over 3824352.92 frames. ], batch size: 57, lr: 8.26e-03, grad_scale: 8.0 +2023-04-01 17:38:57,769 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66162.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:39:00,973 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.007e+02 6.063e+02 7.892e+02 1.039e+03 2.104e+03, threshold=1.578e+03, percent-clipped=6.0 +2023-04-01 17:39:05,614 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 17:39:17,535 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66178.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:39:43,291 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66200.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:39:44,104 INFO [train.py:903] (0/4) Epoch 10, batch 4750, loss[loss=0.2657, simple_loss=0.3314, pruned_loss=0.09995, over 19479.00 frames. ], tot_loss[loss=0.2408, simple_loss=0.3121, pruned_loss=0.08475, over 3829694.80 frames. ], batch size: 64, lr: 8.26e-03, grad_scale: 8.0 +2023-04-01 17:40:43,453 INFO [train.py:903] (0/4) Epoch 10, batch 4800, loss[loss=0.23, simple_loss=0.3103, pruned_loss=0.07483, over 19432.00 frames. ], tot_loss[loss=0.2413, simple_loss=0.3124, pruned_loss=0.08513, over 3816029.84 frames. ], batch size: 70, lr: 8.26e-03, grad_scale: 8.0 +2023-04-01 17:41:01,217 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.573e+02 5.914e+02 7.080e+02 8.206e+02 1.527e+03, threshold=1.416e+03, percent-clipped=0.0 +2023-04-01 17:41:01,646 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66265.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:41:15,548 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66276.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:41:31,322 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66290.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:41:35,390 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9848, 3.6564, 2.1967, 1.5390, 3.4117, 1.4042, 1.1421, 2.1607], + device='cuda:0'), covar=tensor([0.0991, 0.0316, 0.0669, 0.0793, 0.0338, 0.1049, 0.0883, 0.0494], + device='cuda:0'), in_proj_covar=tensor([0.0288, 0.0296, 0.0325, 0.0241, 0.0235, 0.0318, 0.0287, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 17:41:44,772 INFO [train.py:903] (0/4) Epoch 10, batch 4850, loss[loss=0.1851, simple_loss=0.2639, pruned_loss=0.05312, over 19761.00 frames. ], tot_loss[loss=0.2402, simple_loss=0.3115, pruned_loss=0.0844, over 3795160.16 frames. ], batch size: 47, lr: 8.25e-03, grad_scale: 8.0 +2023-04-01 17:41:56,283 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-04-01 17:42:01,790 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66315.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:42:10,353 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 17:42:18,476 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2129, 1.8190, 1.9814, 2.5279, 1.9139, 2.5223, 2.5829, 2.3718], + device='cuda:0'), covar=tensor([0.0716, 0.0897, 0.0885, 0.0849, 0.0893, 0.0589, 0.0754, 0.0560], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0226, 0.0222, 0.0250, 0.0240, 0.0213, 0.0199, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 17:42:30,019 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 17:42:35,765 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 17:42:35,790 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 17:42:45,282 INFO [train.py:903] (0/4) Epoch 10, batch 4900, loss[loss=0.2662, simple_loss=0.3281, pruned_loss=0.1021, over 13770.00 frames. ], tot_loss[loss=0.2409, simple_loss=0.3125, pruned_loss=0.08464, over 3791815.81 frames. ], batch size: 136, lr: 8.25e-03, grad_scale: 8.0 +2023-04-01 17:42:45,297 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 17:43:01,825 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.050e+02 5.967e+02 7.269e+02 9.008e+02 2.184e+03, threshold=1.454e+03, percent-clipped=11.0 +2023-04-01 17:43:03,057 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 17:43:25,418 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1070, 1.2044, 1.6525, 1.0568, 2.4728, 3.2982, 3.0121, 3.4573], + device='cuda:0'), covar=tensor([0.1606, 0.3372, 0.2872, 0.2161, 0.0480, 0.0160, 0.0216, 0.0184], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0293, 0.0318, 0.0247, 0.0211, 0.0151, 0.0203, 0.0193], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 17:43:45,941 INFO [train.py:903] (0/4) Epoch 10, batch 4950, loss[loss=0.2474, simple_loss=0.3154, pruned_loss=0.08973, over 19541.00 frames. ], tot_loss[loss=0.2426, simple_loss=0.3139, pruned_loss=0.08563, over 3797298.65 frames. ], batch size: 54, lr: 8.25e-03, grad_scale: 8.0 +2023-04-01 17:44:03,262 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 17:44:26,211 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 17:44:44,945 INFO [train.py:903] (0/4) Epoch 10, batch 5000, loss[loss=0.2347, simple_loss=0.3034, pruned_loss=0.08296, over 19422.00 frames. ], tot_loss[loss=0.2421, simple_loss=0.3131, pruned_loss=0.08554, over 3803639.32 frames. ], batch size: 48, lr: 8.25e-03, grad_scale: 8.0 +2023-04-01 17:44:54,555 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8811, 2.1335, 2.3684, 2.6568, 2.6603, 2.4213, 2.3350, 2.8432], + device='cuda:0'), covar=tensor([0.0750, 0.1704, 0.1234, 0.0905, 0.1144, 0.0435, 0.1001, 0.0497], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0352, 0.0293, 0.0240, 0.0294, 0.0245, 0.0277, 0.0238], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 17:44:56,393 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 17:45:01,936 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.596e+02 5.419e+02 6.464e+02 8.305e+02 1.628e+03, threshold=1.293e+03, percent-clipped=3.0 +2023-04-01 17:45:06,368 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 17:45:45,174 INFO [train.py:903] (0/4) Epoch 10, batch 5050, loss[loss=0.2241, simple_loss=0.2981, pruned_loss=0.0751, over 19739.00 frames. ], tot_loss[loss=0.2422, simple_loss=0.3132, pruned_loss=0.08558, over 3810089.50 frames. ], batch size: 51, lr: 8.24e-03, grad_scale: 8.0 +2023-04-01 17:45:51,070 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66506.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:46:10,883 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66522.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:46:19,330 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 17:46:28,956 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2339, 1.4583, 1.6215, 1.8219, 2.8708, 1.3229, 2.1821, 3.1038], + device='cuda:0'), covar=tensor([0.0481, 0.2460, 0.2482, 0.1492, 0.0685, 0.2229, 0.1172, 0.0333], + device='cuda:0'), in_proj_covar=tensor([0.0345, 0.0330, 0.0348, 0.0314, 0.0339, 0.0329, 0.0318, 0.0340], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 17:46:31,331 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66539.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:46:44,531 INFO [train.py:903] (0/4) Epoch 10, batch 5100, loss[loss=0.2099, simple_loss=0.2926, pruned_loss=0.06358, over 19841.00 frames. ], tot_loss[loss=0.2431, simple_loss=0.3137, pruned_loss=0.08624, over 3792495.76 frames. ], batch size: 52, lr: 8.24e-03, grad_scale: 8.0 +2023-04-01 17:46:57,222 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 17:46:59,528 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 17:47:01,863 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.567e+02 5.864e+02 7.142e+02 9.030e+02 2.803e+03, threshold=1.428e+03, percent-clipped=6.0 +2023-04-01 17:47:04,008 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 17:47:08,937 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66571.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:47:39,596 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66596.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:47:45,738 INFO [train.py:903] (0/4) Epoch 10, batch 5150, loss[loss=0.2583, simple_loss=0.3271, pruned_loss=0.09473, over 19451.00 frames. ], tot_loss[loss=0.2419, simple_loss=0.3129, pruned_loss=0.08549, over 3783637.47 frames. ], batch size: 70, lr: 8.24e-03, grad_scale: 8.0 +2023-04-01 17:47:56,055 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2211, 2.0900, 2.0473, 2.4095, 2.2990, 1.9209, 1.9748, 2.2495], + device='cuda:0'), covar=tensor([0.0850, 0.1498, 0.1292, 0.0845, 0.1050, 0.0729, 0.1159, 0.0723], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0349, 0.0291, 0.0237, 0.0291, 0.0244, 0.0274, 0.0237], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 17:47:56,922 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 17:48:00,632 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-04-01 17:48:09,059 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66620.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:48:10,460 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66621.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:48:29,345 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66637.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:48:33,111 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 17:48:45,644 INFO [train.py:903] (0/4) Epoch 10, batch 5200, loss[loss=0.2487, simple_loss=0.323, pruned_loss=0.08721, over 18820.00 frames. ], tot_loss[loss=0.2401, simple_loss=0.3117, pruned_loss=0.08426, over 3801532.89 frames. ], batch size: 74, lr: 8.23e-03, grad_scale: 8.0 +2023-04-01 17:49:00,681 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 17:49:02,885 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.107e+02 5.688e+02 7.032e+02 8.430e+02 1.656e+03, threshold=1.406e+03, percent-clipped=2.0 +2023-04-01 17:49:44,615 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 17:49:46,899 INFO [train.py:903] (0/4) Epoch 10, batch 5250, loss[loss=0.2165, simple_loss=0.2792, pruned_loss=0.07695, over 19803.00 frames. ], tot_loss[loss=0.2416, simple_loss=0.313, pruned_loss=0.08511, over 3802017.36 frames. ], batch size: 48, lr: 8.23e-03, grad_scale: 8.0 +2023-04-01 17:50:28,104 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66735.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:50:44,976 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66749.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:50:47,100 INFO [train.py:903] (0/4) Epoch 10, batch 5300, loss[loss=0.202, simple_loss=0.2736, pruned_loss=0.06519, over 19765.00 frames. ], tot_loss[loss=0.2414, simple_loss=0.3128, pruned_loss=0.08502, over 3789788.08 frames. ], batch size: 46, lr: 8.23e-03, grad_scale: 8.0 +2023-04-01 17:51:03,216 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 17:51:04,350 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.476e+02 5.985e+02 7.858e+02 1.069e+03 1.957e+03, threshold=1.572e+03, percent-clipped=7.0 +2023-04-01 17:51:47,684 INFO [train.py:903] (0/4) Epoch 10, batch 5350, loss[loss=0.2444, simple_loss=0.3225, pruned_loss=0.08316, over 19785.00 frames. ], tot_loss[loss=0.2421, simple_loss=0.3135, pruned_loss=0.0853, over 3799848.53 frames. ], batch size: 56, lr: 8.22e-03, grad_scale: 8.0 +2023-04-01 17:52:19,369 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 17:52:46,929 INFO [train.py:903] (0/4) Epoch 10, batch 5400, loss[loss=0.2384, simple_loss=0.3079, pruned_loss=0.08445, over 19576.00 frames. ], tot_loss[loss=0.2422, simple_loss=0.3136, pruned_loss=0.08542, over 3814149.15 frames. ], batch size: 52, lr: 8.22e-03, grad_scale: 8.0 +2023-04-01 17:52:52,114 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-04-01 17:53:05,690 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.674e+02 5.633e+02 6.871e+02 8.512e+02 1.525e+03, threshold=1.374e+03, percent-clipped=0.0 +2023-04-01 17:53:18,082 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66877.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:53:26,403 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=66883.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:53:37,766 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66893.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:53:47,839 INFO [train.py:903] (0/4) Epoch 10, batch 5450, loss[loss=0.2489, simple_loss=0.322, pruned_loss=0.08789, over 19487.00 frames. ], tot_loss[loss=0.2426, simple_loss=0.3139, pruned_loss=0.08561, over 3827517.57 frames. ], batch size: 64, lr: 8.22e-03, grad_scale: 8.0 +2023-04-01 17:53:49,495 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66902.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:54:08,206 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=66918.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:54:15,781 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9896, 1.6810, 1.5653, 2.0122, 1.8730, 1.7911, 1.7190, 1.8843], + device='cuda:0'), covar=tensor([0.0875, 0.1573, 0.1309, 0.0959, 0.1194, 0.0491, 0.1074, 0.0682], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0353, 0.0291, 0.0239, 0.0295, 0.0244, 0.0278, 0.0238], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 17:54:47,623 INFO [train.py:903] (0/4) Epoch 10, batch 5500, loss[loss=0.2181, simple_loss=0.293, pruned_loss=0.07161, over 19831.00 frames. ], tot_loss[loss=0.2418, simple_loss=0.3133, pruned_loss=0.08512, over 3832639.28 frames. ], batch size: 52, lr: 8.21e-03, grad_scale: 8.0 +2023-04-01 17:55:06,146 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.718e+02 5.851e+02 7.428e+02 9.674e+02 2.066e+03, threshold=1.486e+03, percent-clipped=6.0 +2023-04-01 17:55:10,682 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 17:55:16,769 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=66975.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:55:36,746 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=66991.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:55:41,770 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.64 vs. limit=2.0 +2023-04-01 17:55:44,766 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=66998.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:55:48,023 INFO [train.py:903] (0/4) Epoch 10, batch 5550, loss[loss=0.1961, simple_loss=0.2648, pruned_loss=0.06371, over 19735.00 frames. ], tot_loss[loss=0.2405, simple_loss=0.3122, pruned_loss=0.08434, over 3828887.87 frames. ], batch size: 45, lr: 8.21e-03, grad_scale: 8.0 +2023-04-01 17:55:54,309 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 17:56:05,789 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67016.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:56:42,310 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 17:56:47,795 INFO [train.py:903] (0/4) Epoch 10, batch 5600, loss[loss=0.2566, simple_loss=0.333, pruned_loss=0.09009, over 18187.00 frames. ], tot_loss[loss=0.2409, simple_loss=0.3126, pruned_loss=0.08465, over 3824495.07 frames. ], batch size: 83, lr: 8.21e-03, grad_scale: 8.0 +2023-04-01 17:56:52,340 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.59 vs. limit=5.0 +2023-04-01 17:57:06,482 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.953e+02 5.918e+02 7.804e+02 1.015e+03 2.269e+03, threshold=1.561e+03, percent-clipped=7.0 +2023-04-01 17:57:38,369 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=67093.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 17:57:47,808 INFO [train.py:903] (0/4) Epoch 10, batch 5650, loss[loss=0.2582, simple_loss=0.3315, pruned_loss=0.09243, over 19776.00 frames. ], tot_loss[loss=0.2414, simple_loss=0.3133, pruned_loss=0.08473, over 3840106.26 frames. ], batch size: 56, lr: 8.21e-03, grad_scale: 8.0 +2023-04-01 17:57:48,111 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67101.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 17:58:33,181 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 17:58:47,425 INFO [train.py:903] (0/4) Epoch 10, batch 5700, loss[loss=0.2116, simple_loss=0.2868, pruned_loss=0.06814, over 19614.00 frames. ], tot_loss[loss=0.2427, simple_loss=0.3138, pruned_loss=0.08582, over 3830730.23 frames. ], batch size: 50, lr: 8.20e-03, grad_scale: 8.0 +2023-04-01 17:59:05,254 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.236e+02 6.145e+02 7.592e+02 1.064e+03 2.520e+03, threshold=1.518e+03, percent-clipped=7.0 +2023-04-01 17:59:06,717 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2573, 1.2828, 1.5108, 0.9746, 2.5103, 3.3396, 2.9785, 3.4704], + device='cuda:0'), covar=tensor([0.1428, 0.3225, 0.2955, 0.2145, 0.0455, 0.0157, 0.0211, 0.0196], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0290, 0.0313, 0.0245, 0.0206, 0.0150, 0.0202, 0.0193], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 17:59:47,165 INFO [train.py:903] (0/4) Epoch 10, batch 5750, loss[loss=0.2609, simple_loss=0.3354, pruned_loss=0.09323, over 19748.00 frames. ], tot_loss[loss=0.2427, simple_loss=0.314, pruned_loss=0.08572, over 3816540.56 frames. ], batch size: 63, lr: 8.20e-03, grad_scale: 8.0 +2023-04-01 17:59:48,359 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 17:59:48,838 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1318, 2.1142, 2.2546, 3.1013, 2.0725, 2.9899, 2.7638, 2.1395], + device='cuda:0'), covar=tensor([0.3557, 0.3157, 0.1374, 0.1963, 0.3658, 0.1463, 0.3006, 0.2549], + device='cuda:0'), in_proj_covar=tensor([0.0773, 0.0785, 0.0639, 0.0885, 0.0762, 0.0690, 0.0773, 0.0697], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 17:59:57,179 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67208.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 17:59:57,974 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 18:00:00,570 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9755, 4.5450, 2.6991, 3.9194, 1.0069, 4.1883, 4.2531, 4.4046], + device='cuda:0'), covar=tensor([0.0549, 0.0932, 0.1886, 0.0755, 0.4092, 0.0649, 0.0750, 0.0908], + device='cuda:0'), in_proj_covar=tensor([0.0422, 0.0358, 0.0424, 0.0311, 0.0372, 0.0352, 0.0346, 0.0377], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 18:00:02,616 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 18:00:03,008 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2861, 2.8605, 2.0499, 2.1098, 2.0278, 2.3676, 0.7260, 1.9867], + device='cuda:0'), covar=tensor([0.0407, 0.0468, 0.0502, 0.0760, 0.0754, 0.0713, 0.0967, 0.0840], + device='cuda:0'), in_proj_covar=tensor([0.0327, 0.0328, 0.0323, 0.0348, 0.0416, 0.0344, 0.0302, 0.0322], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 18:00:47,882 INFO [train.py:903] (0/4) Epoch 10, batch 5800, loss[loss=0.2201, simple_loss=0.2932, pruned_loss=0.07353, over 19840.00 frames. ], tot_loss[loss=0.2422, simple_loss=0.3137, pruned_loss=0.08539, over 3825075.70 frames. ], batch size: 52, lr: 8.20e-03, grad_scale: 8.0 +2023-04-01 18:00:51,529 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67254.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:01:06,015 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.662e+02 5.628e+02 6.923e+02 8.923e+02 2.275e+03, threshold=1.385e+03, percent-clipped=6.0 +2023-04-01 18:01:20,781 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67279.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:01:47,312 INFO [train.py:903] (0/4) Epoch 10, batch 5850, loss[loss=0.209, simple_loss=0.2789, pruned_loss=0.06955, over 19743.00 frames. ], tot_loss[loss=0.2423, simple_loss=0.3137, pruned_loss=0.0854, over 3830643.91 frames. ], batch size: 46, lr: 8.19e-03, grad_scale: 8.0 +2023-04-01 18:02:09,366 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=67319.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:02:48,435 INFO [train.py:903] (0/4) Epoch 10, batch 5900, loss[loss=0.2124, simple_loss=0.2836, pruned_loss=0.07058, over 19419.00 frames. ], tot_loss[loss=0.2406, simple_loss=0.3122, pruned_loss=0.08449, over 3835281.29 frames. ], batch size: 48, lr: 8.19e-03, grad_scale: 8.0 +2023-04-01 18:02:52,959 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 18:03:05,153 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.548e+02 5.688e+02 6.588e+02 8.588e+02 1.646e+03, threshold=1.318e+03, percent-clipped=2.0 +2023-04-01 18:03:11,570 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 18:03:26,047 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6179, 1.2807, 1.3203, 2.0427, 1.6315, 1.8567, 2.0190, 1.7835], + device='cuda:0'), covar=tensor([0.0837, 0.1029, 0.1136, 0.0843, 0.0821, 0.0694, 0.0782, 0.0646], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0226, 0.0224, 0.0251, 0.0237, 0.0214, 0.0198, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 18:03:47,172 INFO [train.py:903] (0/4) Epoch 10, batch 5950, loss[loss=0.248, simple_loss=0.314, pruned_loss=0.09096, over 19657.00 frames. ], tot_loss[loss=0.2404, simple_loss=0.3119, pruned_loss=0.08441, over 3831613.26 frames. ], batch size: 55, lr: 8.19e-03, grad_scale: 8.0 +2023-04-01 18:04:26,279 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67434.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:04:38,296 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=67445.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:04:45,815 INFO [train.py:903] (0/4) Epoch 10, batch 6000, loss[loss=0.2135, simple_loss=0.2812, pruned_loss=0.07287, over 19765.00 frames. ], tot_loss[loss=0.2405, simple_loss=0.3117, pruned_loss=0.08467, over 3820260.96 frames. ], batch size: 47, lr: 8.18e-03, grad_scale: 8.0 +2023-04-01 18:04:45,816 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 18:04:58,243 INFO [train.py:937] (0/4) Epoch 10, validation: loss=0.1798, simple_loss=0.2805, pruned_loss=0.03952, over 944034.00 frames. +2023-04-01 18:04:58,244 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 18:05:08,753 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4513, 1.6076, 2.0306, 1.7338, 3.4002, 2.8255, 3.5489, 1.6622], + device='cuda:0'), covar=tensor([0.2054, 0.3472, 0.2091, 0.1554, 0.1241, 0.1572, 0.1411, 0.3167], + device='cuda:0'), in_proj_covar=tensor([0.0476, 0.0561, 0.0582, 0.0427, 0.0580, 0.0483, 0.0638, 0.0478], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 18:05:15,104 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67464.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:05:17,959 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.621e+02 5.414e+02 6.867e+02 8.657e+02 1.897e+03, threshold=1.373e+03, percent-clipped=4.0 +2023-04-01 18:05:22,907 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0895, 1.7936, 1.6793, 2.1376, 1.8554, 1.8195, 1.7441, 2.0489], + device='cuda:0'), covar=tensor([0.0813, 0.1484, 0.1303, 0.0837, 0.1152, 0.0481, 0.1144, 0.0619], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0357, 0.0294, 0.0241, 0.0297, 0.0246, 0.0282, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 18:05:44,639 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67489.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 18:05:59,559 INFO [train.py:903] (0/4) Epoch 10, batch 6050, loss[loss=0.217, simple_loss=0.2958, pruned_loss=0.06907, over 19590.00 frames. ], tot_loss[loss=0.2395, simple_loss=0.3111, pruned_loss=0.08395, over 3832977.02 frames. ], batch size: 61, lr: 8.18e-03, grad_scale: 8.0 +2023-04-01 18:06:35,143 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67531.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:06:59,156 INFO [train.py:903] (0/4) Epoch 10, batch 6100, loss[loss=0.1872, simple_loss=0.2694, pruned_loss=0.05251, over 19853.00 frames. ], tot_loss[loss=0.2391, simple_loss=0.3109, pruned_loss=0.08365, over 3845395.03 frames. ], batch size: 52, lr: 8.18e-03, grad_scale: 4.0 +2023-04-01 18:07:10,593 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67560.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:07:19,957 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.598e+02 5.925e+02 6.739e+02 8.410e+02 2.337e+03, threshold=1.348e+03, percent-clipped=5.0 +2023-04-01 18:07:59,365 INFO [train.py:903] (0/4) Epoch 10, batch 6150, loss[loss=0.2412, simple_loss=0.3086, pruned_loss=0.08692, over 19482.00 frames. ], tot_loss[loss=0.2406, simple_loss=0.3119, pruned_loss=0.08462, over 3840247.99 frames. ], batch size: 49, lr: 8.18e-03, grad_scale: 4.0 +2023-04-01 18:08:28,208 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 18:08:59,762 INFO [train.py:903] (0/4) Epoch 10, batch 6200, loss[loss=0.2583, simple_loss=0.3342, pruned_loss=0.09122, over 19595.00 frames. ], tot_loss[loss=0.2411, simple_loss=0.3125, pruned_loss=0.08483, over 3834406.74 frames. ], batch size: 61, lr: 8.17e-03, grad_scale: 4.0 +2023-04-01 18:09:02,991 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-01 18:09:20,095 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.127e+02 5.059e+02 6.776e+02 9.553e+02 2.024e+03, threshold=1.355e+03, percent-clipped=7.0 +2023-04-01 18:09:46,790 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67690.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:09:56,453 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67698.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:09:59,569 INFO [train.py:903] (0/4) Epoch 10, batch 6250, loss[loss=0.2118, simple_loss=0.2961, pruned_loss=0.0638, over 19741.00 frames. ], tot_loss[loss=0.2406, simple_loss=0.312, pruned_loss=0.08456, over 3836152.01 frames. ], batch size: 51, lr: 8.17e-03, grad_scale: 4.0 +2023-04-01 18:10:09,268 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-01 18:10:09,897 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67710.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:10:15,754 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67715.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:10:30,842 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 18:10:59,334 INFO [train.py:903] (0/4) Epoch 10, batch 6300, loss[loss=0.2118, simple_loss=0.2823, pruned_loss=0.07069, over 19744.00 frames. ], tot_loss[loss=0.2395, simple_loss=0.3113, pruned_loss=0.08388, over 3839829.09 frames. ], batch size: 46, lr: 8.17e-03, grad_scale: 4.0 +2023-04-01 18:11:19,322 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.616e+02 5.371e+02 6.798e+02 8.698e+02 1.915e+03, threshold=1.360e+03, percent-clipped=7.0 +2023-04-01 18:11:58,679 INFO [train.py:903] (0/4) Epoch 10, batch 6350, loss[loss=0.2127, simple_loss=0.2915, pruned_loss=0.06697, over 19746.00 frames. ], tot_loss[loss=0.24, simple_loss=0.3115, pruned_loss=0.08418, over 3833161.50 frames. ], batch size: 51, lr: 8.16e-03, grad_scale: 4.0 +2023-04-01 18:12:16,854 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=67816.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:12:47,253 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=67841.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:12:58,567 INFO [train.py:903] (0/4) Epoch 10, batch 6400, loss[loss=0.2505, simple_loss=0.325, pruned_loss=0.08802, over 18778.00 frames. ], tot_loss[loss=0.241, simple_loss=0.3122, pruned_loss=0.08483, over 3828975.85 frames. ], batch size: 74, lr: 8.16e-03, grad_scale: 8.0 +2023-04-01 18:13:18,749 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.354e+02 5.283e+02 6.601e+02 9.298e+02 1.582e+03, threshold=1.320e+03, percent-clipped=4.0 +2023-04-01 18:13:26,972 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=67875.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:13:59,329 INFO [train.py:903] (0/4) Epoch 10, batch 6450, loss[loss=0.2355, simple_loss=0.3057, pruned_loss=0.0826, over 19341.00 frames. ], tot_loss[loss=0.2412, simple_loss=0.3122, pruned_loss=0.08507, over 3821701.44 frames. ], batch size: 66, lr: 8.16e-03, grad_scale: 8.0 +2023-04-01 18:14:42,673 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 18:14:59,317 INFO [train.py:903] (0/4) Epoch 10, batch 6500, loss[loss=0.1986, simple_loss=0.2718, pruned_loss=0.06274, over 19361.00 frames. ], tot_loss[loss=0.2402, simple_loss=0.3116, pruned_loss=0.08435, over 3817681.95 frames. ], batch size: 47, lr: 8.15e-03, grad_scale: 8.0 +2023-04-01 18:15:05,071 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 18:15:19,435 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.752e+02 5.758e+02 7.165e+02 9.309e+02 2.290e+03, threshold=1.433e+03, percent-clipped=7.0 +2023-04-01 18:15:47,551 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=67990.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:15:57,624 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=67999.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:15:59,435 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-68000.pt +2023-04-01 18:16:01,758 INFO [train.py:903] (0/4) Epoch 10, batch 6550, loss[loss=0.2413, simple_loss=0.3116, pruned_loss=0.08555, over 19833.00 frames. ], tot_loss[loss=0.2405, simple_loss=0.3124, pruned_loss=0.08436, over 3824903.94 frames. ], batch size: 52, lr: 8.15e-03, grad_scale: 8.0 +2023-04-01 18:16:51,299 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68042.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:17:02,075 INFO [train.py:903] (0/4) Epoch 10, batch 6600, loss[loss=0.2514, simple_loss=0.3232, pruned_loss=0.0898, over 19780.00 frames. ], tot_loss[loss=0.2406, simple_loss=0.3128, pruned_loss=0.08413, over 3822535.25 frames. ], batch size: 56, lr: 8.15e-03, grad_scale: 8.0 +2023-04-01 18:17:05,653 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68054.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:17:23,334 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.670e+02 5.913e+02 7.324e+02 8.709e+02 1.479e+03, threshold=1.465e+03, percent-clipped=1.0 +2023-04-01 18:18:02,824 INFO [train.py:903] (0/4) Epoch 10, batch 6650, loss[loss=0.2715, simple_loss=0.3345, pruned_loss=0.1042, over 19685.00 frames. ], tot_loss[loss=0.2417, simple_loss=0.3133, pruned_loss=0.08511, over 3808588.66 frames. ], batch size: 58, lr: 8.15e-03, grad_scale: 8.0 +2023-04-01 18:18:07,728 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68105.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:19:02,662 INFO [train.py:903] (0/4) Epoch 10, batch 6700, loss[loss=0.2294, simple_loss=0.287, pruned_loss=0.08588, over 19787.00 frames. ], tot_loss[loss=0.2405, simple_loss=0.3122, pruned_loss=0.08435, over 3818069.66 frames. ], batch size: 48, lr: 8.14e-03, grad_scale: 8.0 +2023-04-01 18:19:09,883 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68157.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:19:22,650 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.762e+02 5.919e+02 7.224e+02 9.287e+02 2.274e+03, threshold=1.445e+03, percent-clipped=4.0 +2023-04-01 18:19:24,149 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68169.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:20:00,032 INFO [train.py:903] (0/4) Epoch 10, batch 6750, loss[loss=0.2128, simple_loss=0.2977, pruned_loss=0.06396, over 19702.00 frames. ], tot_loss[loss=0.2402, simple_loss=0.3117, pruned_loss=0.08434, over 3822842.65 frames. ], batch size: 59, lr: 8.14e-03, grad_scale: 8.0 +2023-04-01 18:20:51,822 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68246.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:20:56,750 INFO [train.py:903] (0/4) Epoch 10, batch 6800, loss[loss=0.2616, simple_loss=0.3351, pruned_loss=0.09405, over 19313.00 frames. ], tot_loss[loss=0.2404, simple_loss=0.312, pruned_loss=0.08437, over 3820577.22 frames. ], batch size: 66, lr: 8.14e-03, grad_scale: 8.0 +2023-04-01 18:21:15,044 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.210e+02 6.150e+02 7.610e+02 9.082e+02 1.904e+03, threshold=1.522e+03, percent-clipped=4.0 +2023-04-01 18:21:18,274 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68271.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:21:25,413 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-10.pt +2023-04-01 18:21:41,028 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 18:21:42,080 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 18:21:44,655 INFO [train.py:903] (0/4) Epoch 11, batch 0, loss[loss=0.2342, simple_loss=0.3074, pruned_loss=0.08051, over 19677.00 frames. ], tot_loss[loss=0.2342, simple_loss=0.3074, pruned_loss=0.08051, over 19677.00 frames. ], batch size: 53, lr: 7.77e-03, grad_scale: 8.0 +2023-04-01 18:21:44,656 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 18:21:56,759 INFO [train.py:937] (0/4) Epoch 11, validation: loss=0.181, simple_loss=0.2818, pruned_loss=0.04012, over 944034.00 frames. +2023-04-01 18:21:56,759 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 18:22:09,351 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 18:22:57,934 INFO [train.py:903] (0/4) Epoch 11, batch 50, loss[loss=0.2649, simple_loss=0.3319, pruned_loss=0.09889, over 19532.00 frames. ], tot_loss[loss=0.2379, simple_loss=0.3096, pruned_loss=0.08311, over 877066.71 frames. ], batch size: 54, lr: 7.76e-03, grad_scale: 8.0 +2023-04-01 18:23:15,230 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68343.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:23:35,577 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 18:23:46,499 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-04-01 18:23:46,895 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.915e+02 5.747e+02 7.027e+02 9.557e+02 1.564e+03, threshold=1.405e+03, percent-clipped=1.0 +2023-04-01 18:24:00,205 INFO [train.py:903] (0/4) Epoch 11, batch 100, loss[loss=0.2341, simple_loss=0.312, pruned_loss=0.07808, over 19702.00 frames. ], tot_loss[loss=0.2342, simple_loss=0.3067, pruned_loss=0.08086, over 1536018.84 frames. ], batch size: 59, lr: 7.76e-03, grad_scale: 8.0 +2023-04-01 18:24:13,707 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 18:24:18,543 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0745, 1.2355, 1.5982, 0.8853, 2.5214, 3.0708, 2.7522, 3.2348], + device='cuda:0'), covar=tensor([0.1506, 0.3191, 0.2795, 0.2170, 0.0417, 0.0177, 0.0215, 0.0189], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0294, 0.0319, 0.0250, 0.0211, 0.0154, 0.0203, 0.0193], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 18:24:42,541 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68413.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:24:47,923 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68417.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:24:57,051 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68425.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:25:01,043 INFO [train.py:903] (0/4) Epoch 11, batch 150, loss[loss=0.3229, simple_loss=0.3663, pruned_loss=0.1397, over 13682.00 frames. ], tot_loss[loss=0.2378, simple_loss=0.3094, pruned_loss=0.08313, over 2035474.63 frames. ], batch size: 136, lr: 7.76e-03, grad_scale: 8.0 +2023-04-01 18:25:11,923 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68438.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:25:25,283 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68449.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:25:26,556 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68450.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:25:36,604 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68458.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:25:47,788 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.848e+02 5.281e+02 6.719e+02 8.986e+02 1.619e+03, threshold=1.344e+03, percent-clipped=5.0 +2023-04-01 18:26:00,691 INFO [train.py:903] (0/4) Epoch 11, batch 200, loss[loss=0.1813, simple_loss=0.258, pruned_loss=0.05231, over 19047.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.3068, pruned_loss=0.08146, over 2443258.04 frames. ], batch size: 42, lr: 7.76e-03, grad_scale: 8.0 +2023-04-01 18:26:02,031 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 18:27:03,595 INFO [train.py:903] (0/4) Epoch 11, batch 250, loss[loss=0.2386, simple_loss=0.3101, pruned_loss=0.08351, over 19737.00 frames. ], tot_loss[loss=0.2355, simple_loss=0.3078, pruned_loss=0.08161, over 2753531.97 frames. ], batch size: 63, lr: 7.75e-03, grad_scale: 8.0 +2023-04-01 18:27:46,843 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68564.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:27:51,041 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.001e+02 5.512e+02 6.613e+02 8.406e+02 1.798e+03, threshold=1.323e+03, percent-clipped=1.0 +2023-04-01 18:28:07,069 INFO [train.py:903] (0/4) Epoch 11, batch 300, loss[loss=0.2379, simple_loss=0.3179, pruned_loss=0.07892, over 19441.00 frames. ], tot_loss[loss=0.2337, simple_loss=0.3066, pruned_loss=0.08043, over 3004924.86 frames. ], batch size: 64, lr: 7.75e-03, grad_scale: 8.0 +2023-04-01 18:28:27,736 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68596.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:29:09,812 INFO [train.py:903] (0/4) Epoch 11, batch 350, loss[loss=0.2125, simple_loss=0.288, pruned_loss=0.06853, over 19568.00 frames. ], tot_loss[loss=0.2354, simple_loss=0.3085, pruned_loss=0.08117, over 3183840.95 frames. ], batch size: 52, lr: 7.75e-03, grad_scale: 8.0 +2023-04-01 18:29:16,655 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 18:29:44,107 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3811, 2.1364, 2.1982, 2.4253, 2.2042, 1.9563, 2.1204, 2.4327], + device='cuda:0'), covar=tensor([0.0681, 0.1275, 0.0971, 0.0705, 0.0951, 0.0455, 0.0892, 0.0475], + device='cuda:0'), in_proj_covar=tensor([0.0251, 0.0353, 0.0291, 0.0239, 0.0298, 0.0247, 0.0277, 0.0240], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 18:29:57,394 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.055e+02 5.409e+02 6.235e+02 7.842e+02 1.948e+03, threshold=1.247e+03, percent-clipped=7.0 +2023-04-01 18:30:09,871 INFO [train.py:903] (0/4) Epoch 11, batch 400, loss[loss=0.2203, simple_loss=0.2914, pruned_loss=0.07457, over 19276.00 frames. ], tot_loss[loss=0.2373, simple_loss=0.3099, pruned_loss=0.08232, over 3321243.23 frames. ], batch size: 44, lr: 7.74e-03, grad_scale: 8.0 +2023-04-01 18:30:40,464 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=68703.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:30:54,254 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68714.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:31:11,175 INFO [train.py:903] (0/4) Epoch 11, batch 450, loss[loss=0.2258, simple_loss=0.3047, pruned_loss=0.07347, over 19693.00 frames. ], tot_loss[loss=0.237, simple_loss=0.3095, pruned_loss=0.08225, over 3442150.22 frames. ], batch size: 60, lr: 7.74e-03, grad_scale: 8.0 +2023-04-01 18:31:25,452 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68739.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:31:49,567 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 18:31:50,636 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 18:31:51,642 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68761.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:31:59,522 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.166e+02 5.180e+02 6.491e+02 8.751e+02 1.660e+03, threshold=1.298e+03, percent-clipped=7.0 +2023-04-01 18:32:13,200 INFO [train.py:903] (0/4) Epoch 11, batch 500, loss[loss=0.2434, simple_loss=0.321, pruned_loss=0.08285, over 19472.00 frames. ], tot_loss[loss=0.2366, simple_loss=0.3095, pruned_loss=0.08181, over 3538974.82 frames. ], batch size: 64, lr: 7.74e-03, grad_scale: 8.0 +2023-04-01 18:33:04,547 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=68820.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:33:17,983 INFO [train.py:903] (0/4) Epoch 11, batch 550, loss[loss=0.1806, simple_loss=0.2565, pruned_loss=0.05232, over 19387.00 frames. ], tot_loss[loss=0.2364, simple_loss=0.3094, pruned_loss=0.08174, over 3601617.82 frames. ], batch size: 47, lr: 7.74e-03, grad_scale: 8.0 +2023-04-01 18:33:36,723 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=68845.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:34:06,685 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.095e+02 5.206e+02 6.523e+02 8.559e+02 1.532e+03, threshold=1.305e+03, percent-clipped=5.0 +2023-04-01 18:34:18,030 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=68876.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:34:21,111 INFO [train.py:903] (0/4) Epoch 11, batch 600, loss[loss=0.2737, simple_loss=0.3271, pruned_loss=0.1101, over 19497.00 frames. ], tot_loss[loss=0.238, simple_loss=0.3104, pruned_loss=0.08275, over 3642355.37 frames. ], batch size: 49, lr: 7.73e-03, grad_scale: 8.0 +2023-04-01 18:35:07,359 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 18:35:23,591 INFO [train.py:903] (0/4) Epoch 11, batch 650, loss[loss=0.2831, simple_loss=0.3482, pruned_loss=0.1091, over 19300.00 frames. ], tot_loss[loss=0.238, simple_loss=0.3106, pruned_loss=0.08274, over 3676877.04 frames. ], batch size: 66, lr: 7.73e-03, grad_scale: 4.0 +2023-04-01 18:35:36,868 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=68940.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:36:11,511 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0346, 1.9305, 2.0229, 1.7700, 4.4383, 1.1383, 2.5046, 4.7869], + device='cuda:0'), covar=tensor([0.0343, 0.2428, 0.2384, 0.1685, 0.0774, 0.2584, 0.1258, 0.0197], + device='cuda:0'), in_proj_covar=tensor([0.0353, 0.0336, 0.0351, 0.0318, 0.0346, 0.0332, 0.0330, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 18:36:14,702 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.863e+02 5.561e+02 7.157e+02 8.945e+02 2.089e+03, threshold=1.431e+03, percent-clipped=8.0 +2023-04-01 18:36:26,502 INFO [train.py:903] (0/4) Epoch 11, batch 700, loss[loss=0.2256, simple_loss=0.2988, pruned_loss=0.07622, over 19487.00 frames. ], tot_loss[loss=0.2367, simple_loss=0.3098, pruned_loss=0.0818, over 3706788.90 frames. ], batch size: 49, lr: 7.73e-03, grad_scale: 4.0 +2023-04-01 18:36:41,605 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.36 vs. limit=5.0 +2023-04-01 18:37:30,468 INFO [train.py:903] (0/4) Epoch 11, batch 750, loss[loss=0.2195, simple_loss=0.3089, pruned_loss=0.06508, over 19652.00 frames. ], tot_loss[loss=0.2375, simple_loss=0.3103, pruned_loss=0.08233, over 3730359.02 frames. ], batch size: 58, lr: 7.72e-03, grad_scale: 4.0 +2023-04-01 18:37:44,838 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3750, 2.2121, 1.8726, 1.7428, 1.4741, 1.7163, 0.5555, 1.2502], + device='cuda:0'), covar=tensor([0.0396, 0.0443, 0.0379, 0.0617, 0.0923, 0.0759, 0.0905, 0.0756], + device='cuda:0'), in_proj_covar=tensor([0.0333, 0.0329, 0.0330, 0.0352, 0.0425, 0.0349, 0.0309, 0.0324], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 18:37:52,518 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69047.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:38:01,875 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69055.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:38:09,149 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69061.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:38:20,856 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.322e+02 5.289e+02 6.423e+02 8.063e+02 1.861e+03, threshold=1.285e+03, percent-clipped=2.0 +2023-04-01 18:38:33,469 INFO [train.py:903] (0/4) Epoch 11, batch 800, loss[loss=0.2172, simple_loss=0.302, pruned_loss=0.06618, over 19307.00 frames. ], tot_loss[loss=0.238, simple_loss=0.311, pruned_loss=0.08253, over 3739259.52 frames. ], batch size: 66, lr: 7.72e-03, grad_scale: 8.0 +2023-04-01 18:38:49,975 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 18:39:34,238 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6084, 4.1606, 2.6826, 3.7216, 1.1418, 3.8965, 3.9266, 3.9913], + device='cuda:0'), covar=tensor([0.0578, 0.1034, 0.1909, 0.0759, 0.3593, 0.0808, 0.0727, 0.0946], + device='cuda:0'), in_proj_covar=tensor([0.0421, 0.0362, 0.0429, 0.0315, 0.0373, 0.0361, 0.0349, 0.0382], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 18:39:35,198 INFO [train.py:903] (0/4) Epoch 11, batch 850, loss[loss=0.1842, simple_loss=0.259, pruned_loss=0.05467, over 19329.00 frames. ], tot_loss[loss=0.2366, simple_loss=0.3098, pruned_loss=0.08169, over 3773559.54 frames. ], batch size: 44, lr: 7.72e-03, grad_scale: 8.0 +2023-04-01 18:39:39,307 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69132.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:39:41,085 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-04-01 18:40:04,976 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7958, 1.8267, 2.0037, 2.5302, 1.6301, 2.3238, 2.2853, 1.9172], + device='cuda:0'), covar=tensor([0.3388, 0.2936, 0.1486, 0.1488, 0.3096, 0.1444, 0.3338, 0.2641], + device='cuda:0'), in_proj_covar=tensor([0.0779, 0.0795, 0.0641, 0.0886, 0.0771, 0.0698, 0.0780, 0.0704], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 18:40:09,383 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3565, 3.6853, 3.9870, 4.1158, 1.5380, 3.8280, 3.3916, 3.3984], + device='cuda:0'), covar=tensor([0.1775, 0.1489, 0.0940, 0.0975, 0.6325, 0.1159, 0.1064, 0.1876], + device='cuda:0'), in_proj_covar=tensor([0.0651, 0.0589, 0.0772, 0.0660, 0.0711, 0.0531, 0.0476, 0.0716], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 18:40:12,541 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69157.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:40:18,364 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69162.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:40:26,036 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.965e+02 5.808e+02 7.359e+02 9.451e+02 2.011e+03, threshold=1.472e+03, percent-clipped=12.0 +2023-04-01 18:40:32,104 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 18:40:38,098 INFO [train.py:903] (0/4) Epoch 11, batch 900, loss[loss=0.2463, simple_loss=0.324, pruned_loss=0.0843, over 19542.00 frames. ], tot_loss[loss=0.237, simple_loss=0.31, pruned_loss=0.08204, over 3790686.06 frames. ], batch size: 56, lr: 7.72e-03, grad_scale: 8.0 +2023-04-01 18:40:39,576 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69180.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:40:47,291 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69186.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:41:28,531 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-01 18:41:42,669 INFO [train.py:903] (0/4) Epoch 11, batch 950, loss[loss=0.205, simple_loss=0.2799, pruned_loss=0.06501, over 19612.00 frames. ], tot_loss[loss=0.237, simple_loss=0.31, pruned_loss=0.08197, over 3793919.32 frames. ], batch size: 50, lr: 7.71e-03, grad_scale: 8.0 +2023-04-01 18:41:49,486 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 18:42:01,254 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69243.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:42:11,228 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0 +2023-04-01 18:42:24,823 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69263.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:42:33,741 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.446e+02 5.204e+02 6.315e+02 7.601e+02 2.315e+03, threshold=1.263e+03, percent-clipped=1.0 +2023-04-01 18:42:47,223 INFO [train.py:903] (0/4) Epoch 11, batch 1000, loss[loss=0.2165, simple_loss=0.3016, pruned_loss=0.06568, over 19521.00 frames. ], tot_loss[loss=0.2373, simple_loss=0.3106, pruned_loss=0.08205, over 3786082.25 frames. ], batch size: 54, lr: 7.71e-03, grad_scale: 8.0 +2023-04-01 18:43:01,237 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69291.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:43:27,058 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69311.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:43:43,508 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 18:43:48,898 INFO [train.py:903] (0/4) Epoch 11, batch 1050, loss[loss=0.3022, simple_loss=0.3516, pruned_loss=0.1263, over 13095.00 frames. ], tot_loss[loss=0.237, simple_loss=0.3101, pruned_loss=0.082, over 3795293.17 frames. ], batch size: 136, lr: 7.71e-03, grad_scale: 8.0 +2023-04-01 18:43:56,987 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69336.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:43:59,148 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7995, 4.3295, 2.5265, 3.8105, 1.3100, 4.1024, 4.0904, 4.1225], + device='cuda:0'), covar=tensor([0.0508, 0.0866, 0.2003, 0.0788, 0.3561, 0.0735, 0.0731, 0.1015], + device='cuda:0'), in_proj_covar=tensor([0.0423, 0.0361, 0.0428, 0.0312, 0.0375, 0.0361, 0.0350, 0.0383], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 18:43:59,236 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69338.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:43:59,354 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1218, 2.1020, 1.7117, 1.5180, 1.3729, 1.6111, 0.4847, 1.0775], + device='cuda:0'), covar=tensor([0.0669, 0.0589, 0.0475, 0.0842, 0.1215, 0.0877, 0.1084, 0.0973], + device='cuda:0'), in_proj_covar=tensor([0.0336, 0.0329, 0.0331, 0.0352, 0.0425, 0.0348, 0.0309, 0.0324], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 18:44:02,804 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8846, 1.3309, 1.0404, 1.0085, 1.1376, 0.9687, 0.9725, 1.2357], + device='cuda:0'), covar=tensor([0.0555, 0.0699, 0.1007, 0.0545, 0.0490, 0.1071, 0.0500, 0.0400], + device='cuda:0'), in_proj_covar=tensor([0.0285, 0.0301, 0.0328, 0.0246, 0.0237, 0.0318, 0.0289, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 18:44:24,559 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 18:44:38,148 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.334e+02 5.222e+02 6.475e+02 7.672e+02 1.315e+03, threshold=1.295e+03, percent-clipped=1.0 +2023-04-01 18:44:49,372 INFO [train.py:903] (0/4) Epoch 11, batch 1100, loss[loss=0.1938, simple_loss=0.2663, pruned_loss=0.06066, over 19753.00 frames. ], tot_loss[loss=0.2378, simple_loss=0.3107, pruned_loss=0.08248, over 3810337.12 frames. ], batch size: 46, lr: 7.70e-03, grad_scale: 8.0 +2023-04-01 18:44:57,866 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69386.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:45:24,080 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69405.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:45:32,705 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69412.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:45:39,566 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69418.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:45:53,092 INFO [train.py:903] (0/4) Epoch 11, batch 1150, loss[loss=0.2348, simple_loss=0.3137, pruned_loss=0.07795, over 19763.00 frames. ], tot_loss[loss=0.2371, simple_loss=0.3103, pruned_loss=0.08197, over 3829273.20 frames. ], batch size: 56, lr: 7.70e-03, grad_scale: 8.0 +2023-04-01 18:45:59,884 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-04-01 18:46:11,223 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69443.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:46:41,982 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.975e+02 5.885e+02 7.181e+02 8.495e+02 1.651e+03, threshold=1.436e+03, percent-clipped=3.0 +2023-04-01 18:46:55,768 INFO [train.py:903] (0/4) Epoch 11, batch 1200, loss[loss=0.2735, simple_loss=0.3391, pruned_loss=0.1039, over 18181.00 frames. ], tot_loss[loss=0.2395, simple_loss=0.3125, pruned_loss=0.08322, over 3827823.81 frames. ], batch size: 83, lr: 7.70e-03, grad_scale: 8.0 +2023-04-01 18:46:59,269 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5941, 1.3955, 1.3766, 2.1610, 1.5980, 2.0209, 2.1531, 1.8298], + device='cuda:0'), covar=tensor([0.0740, 0.0910, 0.0972, 0.0720, 0.0838, 0.0604, 0.0710, 0.0612], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0227, 0.0224, 0.0250, 0.0238, 0.0213, 0.0199, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 18:47:27,273 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 18:47:48,958 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69520.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:47:53,380 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69524.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:47:59,006 INFO [train.py:903] (0/4) Epoch 11, batch 1250, loss[loss=0.2233, simple_loss=0.2892, pruned_loss=0.07868, over 16944.00 frames. ], tot_loss[loss=0.2407, simple_loss=0.3134, pruned_loss=0.08401, over 3800305.82 frames. ], batch size: 37, lr: 7.70e-03, grad_scale: 8.0 +2023-04-01 18:48:00,168 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69530.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:48:49,516 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.132e+02 5.575e+02 6.899e+02 8.428e+02 1.860e+03, threshold=1.380e+03, percent-clipped=3.0 +2023-04-01 18:49:00,828 INFO [train.py:903] (0/4) Epoch 11, batch 1300, loss[loss=0.2362, simple_loss=0.3077, pruned_loss=0.08234, over 19770.00 frames. ], tot_loss[loss=0.2403, simple_loss=0.3127, pruned_loss=0.08396, over 3807082.15 frames. ], batch size: 54, lr: 7.69e-03, grad_scale: 8.0 +2023-04-01 18:49:10,187 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69587.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:49:38,567 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69607.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 18:49:39,102 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-01 18:50:04,325 INFO [train.py:903] (0/4) Epoch 11, batch 1350, loss[loss=0.2316, simple_loss=0.3166, pruned_loss=0.07325, over 19712.00 frames. ], tot_loss[loss=0.2391, simple_loss=0.3117, pruned_loss=0.08328, over 3815692.28 frames. ], batch size: 59, lr: 7.69e-03, grad_scale: 8.0 +2023-04-01 18:50:13,495 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69635.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:50:19,156 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69639.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:50:26,756 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69645.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:50:54,481 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.470e+02 5.709e+02 6.895e+02 8.476e+02 1.895e+03, threshold=1.379e+03, percent-clipped=5.0 +2023-04-01 18:51:08,228 INFO [train.py:903] (0/4) Epoch 11, batch 1400, loss[loss=0.2764, simple_loss=0.3386, pruned_loss=0.1071, over 13037.00 frames. ], tot_loss[loss=0.2395, simple_loss=0.3116, pruned_loss=0.08368, over 3809371.32 frames. ], batch size: 136, lr: 7.69e-03, grad_scale: 8.0 +2023-04-01 18:51:11,985 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69682.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:51:14,172 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0973, 1.3139, 1.5546, 1.5418, 2.6632, 1.0152, 2.0826, 2.8808], + device='cuda:0'), covar=tensor([0.0536, 0.2597, 0.2416, 0.1506, 0.0731, 0.2337, 0.1094, 0.0400], + device='cuda:0'), in_proj_covar=tensor([0.0348, 0.0334, 0.0346, 0.0314, 0.0343, 0.0330, 0.0327, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 18:51:25,008 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=69692.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:51:36,450 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69702.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:52:03,526 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69722.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 18:52:09,980 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 18:52:10,949 INFO [train.py:903] (0/4) Epoch 11, batch 1450, loss[loss=0.3296, simple_loss=0.3803, pruned_loss=0.1394, over 19490.00 frames. ], tot_loss[loss=0.2408, simple_loss=0.3128, pruned_loss=0.08446, over 3799801.73 frames. ], batch size: 64, lr: 7.69e-03, grad_scale: 8.0 +2023-04-01 18:52:12,374 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69730.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:52:26,628 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2578, 1.2054, 1.6866, 1.0850, 2.5952, 3.6047, 3.2110, 3.7391], + device='cuda:0'), covar=tensor([0.1513, 0.3540, 0.2975, 0.2120, 0.0527, 0.0130, 0.0211, 0.0185], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0297, 0.0321, 0.0250, 0.0215, 0.0154, 0.0203, 0.0198], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 18:52:36,731 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69750.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:52:44,780 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=69756.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:52:49,242 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4083, 1.3674, 1.4188, 1.5122, 2.9796, 1.0478, 2.1663, 3.2595], + device='cuda:0'), covar=tensor([0.0452, 0.2302, 0.2395, 0.1521, 0.0638, 0.2258, 0.1109, 0.0293], + device='cuda:0'), in_proj_covar=tensor([0.0346, 0.0330, 0.0344, 0.0312, 0.0340, 0.0328, 0.0326, 0.0345], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 18:53:01,499 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.562e+02 5.344e+02 6.531e+02 8.357e+02 2.062e+03, threshold=1.306e+03, percent-clipped=2.0 +2023-04-01 18:53:10,356 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69776.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:53:13,432 INFO [train.py:903] (0/4) Epoch 11, batch 1500, loss[loss=0.29, simple_loss=0.3444, pruned_loss=0.1178, over 12945.00 frames. ], tot_loss[loss=0.2418, simple_loss=0.3135, pruned_loss=0.08499, over 3804981.38 frames. ], batch size: 136, lr: 7.68e-03, grad_scale: 8.0 +2023-04-01 18:53:36,789 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69797.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:53:43,086 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69801.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:54:15,941 INFO [train.py:903] (0/4) Epoch 11, batch 1550, loss[loss=0.214, simple_loss=0.2975, pruned_loss=0.06523, over 19513.00 frames. ], tot_loss[loss=0.2401, simple_loss=0.3121, pruned_loss=0.08406, over 3822346.66 frames. ], batch size: 54, lr: 7.68e-03, grad_scale: 8.0 +2023-04-01 18:54:34,319 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-01 18:54:35,129 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5729, 1.2471, 1.1608, 1.4011, 1.1578, 1.3646, 1.0892, 1.3813], + device='cuda:0'), covar=tensor([0.0969, 0.1178, 0.1503, 0.0965, 0.1180, 0.0581, 0.1385, 0.0791], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0350, 0.0292, 0.0241, 0.0300, 0.0244, 0.0278, 0.0240], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 18:54:38,372 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69845.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:55:07,629 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.582e+02 5.814e+02 6.999e+02 8.951e+02 2.972e+03, threshold=1.400e+03, percent-clipped=5.0 +2023-04-01 18:55:09,075 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=69871.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:55:20,441 INFO [train.py:903] (0/4) Epoch 11, batch 1600, loss[loss=0.2578, simple_loss=0.3313, pruned_loss=0.09214, over 19544.00 frames. ], tot_loss[loss=0.2388, simple_loss=0.3109, pruned_loss=0.08331, over 3825416.30 frames. ], batch size: 56, lr: 7.68e-03, grad_scale: 8.0 +2023-04-01 18:55:36,919 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 18:55:41,192 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69895.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:55:46,717 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 18:55:48,135 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69901.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:56:07,298 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6816, 1.7540, 1.8542, 1.8139, 4.1603, 0.9710, 2.5556, 4.3768], + device='cuda:0'), covar=tensor([0.0316, 0.2459, 0.2407, 0.1629, 0.0669, 0.2669, 0.1255, 0.0211], + device='cuda:0'), in_proj_covar=tensor([0.0350, 0.0335, 0.0348, 0.0316, 0.0344, 0.0331, 0.0329, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 18:56:12,820 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69920.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:56:20,575 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69926.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:56:23,826 INFO [train.py:903] (0/4) Epoch 11, batch 1650, loss[loss=0.19, simple_loss=0.263, pruned_loss=0.05855, over 18603.00 frames. ], tot_loss[loss=0.2389, simple_loss=0.3108, pruned_loss=0.08355, over 3815255.18 frames. ], batch size: 41, lr: 7.67e-03, grad_scale: 8.0 +2023-04-01 18:56:33,402 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9928, 1.9486, 1.7028, 1.5508, 1.4203, 1.5642, 0.2576, 0.8861], + device='cuda:0'), covar=tensor([0.0434, 0.0409, 0.0290, 0.0426, 0.0861, 0.0541, 0.0858, 0.0714], + device='cuda:0'), in_proj_covar=tensor([0.0334, 0.0325, 0.0326, 0.0349, 0.0422, 0.0346, 0.0306, 0.0320], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 18:57:00,739 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69958.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:57:16,658 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.105e+02 5.294e+02 6.504e+02 8.314e+02 1.576e+03, threshold=1.301e+03, percent-clipped=2.0 +2023-04-01 18:57:26,314 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=69978.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 18:57:27,016 INFO [train.py:903] (0/4) Epoch 11, batch 1700, loss[loss=0.2672, simple_loss=0.3412, pruned_loss=0.09661, over 18883.00 frames. ], tot_loss[loss=0.2388, simple_loss=0.3107, pruned_loss=0.08339, over 3803005.75 frames. ], batch size: 74, lr: 7.67e-03, grad_scale: 8.0 +2023-04-01 18:57:32,055 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=69983.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:57:53,386 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-70000.pt +2023-04-01 18:57:59,049 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70003.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 18:58:03,148 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70006.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:58:09,476 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 18:58:29,317 INFO [train.py:903] (0/4) Epoch 11, batch 1750, loss[loss=0.2655, simple_loss=0.3309, pruned_loss=0.1001, over 17204.00 frames. ], tot_loss[loss=0.2396, simple_loss=0.3116, pruned_loss=0.08385, over 3795844.67 frames. ], batch size: 101, lr: 7.67e-03, grad_scale: 8.0 +2023-04-01 18:58:31,967 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70031.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:58:39,008 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=70036.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 18:59:02,476 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70053.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:59:22,247 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.188e+02 6.480e+02 8.127e+02 9.904e+02 1.581e+03, threshold=1.625e+03, percent-clipped=5.0 +2023-04-01 18:59:34,154 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70078.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 18:59:34,848 INFO [train.py:903] (0/4) Epoch 11, batch 1800, loss[loss=0.2867, simple_loss=0.3618, pruned_loss=0.1058, over 19674.00 frames. ], tot_loss[loss=0.2401, simple_loss=0.3124, pruned_loss=0.08388, over 3784349.41 frames. ], batch size: 58, lr: 7.67e-03, grad_scale: 8.0 +2023-04-01 19:00:02,644 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70101.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:00:34,759 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70126.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:00:35,503 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 19:00:35,892 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70127.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:00:37,763 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=70128.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:00:38,699 INFO [train.py:903] (0/4) Epoch 11, batch 1850, loss[loss=0.2138, simple_loss=0.2776, pruned_loss=0.07503, over 14845.00 frames. ], tot_loss[loss=0.2383, simple_loss=0.3109, pruned_loss=0.08288, over 3791802.72 frames. ], batch size: 32, lr: 7.66e-03, grad_scale: 8.0 +2023-04-01 19:01:05,157 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70151.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 19:01:06,365 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70152.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:01:13,686 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 19:01:30,923 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.636e+02 5.171e+02 6.551e+02 7.968e+02 1.780e+03, threshold=1.310e+03, percent-clipped=1.0 +2023-04-01 19:01:41,528 INFO [train.py:903] (0/4) Epoch 11, batch 1900, loss[loss=0.2606, simple_loss=0.339, pruned_loss=0.09113, over 18257.00 frames. ], tot_loss[loss=0.2386, simple_loss=0.3109, pruned_loss=0.08311, over 3798416.88 frames. ], batch size: 83, lr: 7.66e-03, grad_scale: 8.0 +2023-04-01 19:01:57,132 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 19:02:04,772 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 19:02:28,870 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 19:02:42,955 INFO [train.py:903] (0/4) Epoch 11, batch 1950, loss[loss=0.2482, simple_loss=0.3219, pruned_loss=0.0872, over 19505.00 frames. ], tot_loss[loss=0.2385, simple_loss=0.3107, pruned_loss=0.0832, over 3795215.14 frames. ], batch size: 64, lr: 7.66e-03, grad_scale: 8.0 +2023-04-01 19:03:35,375 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.353e+02 4.987e+02 6.114e+02 7.804e+02 3.131e+03, threshold=1.223e+03, percent-clipped=9.0 +2023-04-01 19:03:45,733 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=70278.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:03:46,666 INFO [train.py:903] (0/4) Epoch 11, batch 2000, loss[loss=0.2808, simple_loss=0.3434, pruned_loss=0.109, over 18865.00 frames. ], tot_loss[loss=0.2373, simple_loss=0.3099, pruned_loss=0.08233, over 3806901.94 frames. ], batch size: 74, lr: 7.66e-03, grad_scale: 8.0 +2023-04-01 19:04:04,122 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4326, 3.9023, 4.1187, 4.1143, 1.7369, 3.8706, 3.3944, 3.7677], + device='cuda:0'), covar=tensor([0.1434, 0.1038, 0.0634, 0.0637, 0.4933, 0.0837, 0.0646, 0.1196], + device='cuda:0'), in_proj_covar=tensor([0.0665, 0.0598, 0.0789, 0.0669, 0.0720, 0.0544, 0.0480, 0.0723], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 19:04:48,761 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 19:04:51,059 INFO [train.py:903] (0/4) Epoch 11, batch 2050, loss[loss=0.2197, simple_loss=0.2884, pruned_loss=0.07549, over 19015.00 frames. ], tot_loss[loss=0.2362, simple_loss=0.309, pruned_loss=0.0817, over 3799368.35 frames. ], batch size: 42, lr: 7.65e-03, grad_scale: 8.0 +2023-04-01 19:05:07,040 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 19:05:08,140 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 19:05:27,817 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 19:05:43,665 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.125e+02 5.481e+02 7.126e+02 9.531e+02 2.002e+03, threshold=1.425e+03, percent-clipped=10.0 +2023-04-01 19:05:54,412 INFO [train.py:903] (0/4) Epoch 11, batch 2100, loss[loss=0.2373, simple_loss=0.3196, pruned_loss=0.07754, over 19538.00 frames. ], tot_loss[loss=0.2375, simple_loss=0.3101, pruned_loss=0.08251, over 3801319.42 frames. ], batch size: 56, lr: 7.65e-03, grad_scale: 8.0 +2023-04-01 19:06:24,749 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 19:06:30,846 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70407.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 19:06:48,478 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 19:06:57,573 INFO [train.py:903] (0/4) Epoch 11, batch 2150, loss[loss=0.2511, simple_loss=0.3243, pruned_loss=0.08896, over 17451.00 frames. ], tot_loss[loss=0.2381, simple_loss=0.3104, pruned_loss=0.08294, over 3799387.50 frames. ], batch size: 101, lr: 7.65e-03, grad_scale: 8.0 +2023-04-01 19:07:01,541 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70432.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 19:07:45,868 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1894, 1.3523, 1.7033, 1.4104, 2.6452, 2.0860, 2.8203, 0.9745], + device='cuda:0'), covar=tensor([0.2198, 0.3576, 0.2165, 0.1752, 0.1362, 0.1886, 0.1381, 0.3638], + device='cuda:0'), in_proj_covar=tensor([0.0481, 0.0569, 0.0594, 0.0433, 0.0590, 0.0487, 0.0645, 0.0487], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 19:07:51,365 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.544e+02 5.585e+02 6.733e+02 8.759e+02 1.859e+03, threshold=1.347e+03, percent-clipped=4.0 +2023-04-01 19:07:52,797 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=70472.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:08:02,117 INFO [train.py:903] (0/4) Epoch 11, batch 2200, loss[loss=0.2338, simple_loss=0.2982, pruned_loss=0.08467, over 19751.00 frames. ], tot_loss[loss=0.236, simple_loss=0.3083, pruned_loss=0.08184, over 3812804.25 frames. ], batch size: 47, lr: 7.64e-03, grad_scale: 4.0 +2023-04-01 19:09:06,392 INFO [train.py:903] (0/4) Epoch 11, batch 2250, loss[loss=0.2286, simple_loss=0.3009, pruned_loss=0.07818, over 19761.00 frames. ], tot_loss[loss=0.2355, simple_loss=0.3077, pruned_loss=0.08163, over 3821556.01 frames. ], batch size: 54, lr: 7.64e-03, grad_scale: 4.0 +2023-04-01 19:10:00,993 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.382e+02 4.990e+02 6.839e+02 8.349e+02 1.575e+03, threshold=1.368e+03, percent-clipped=2.0 +2023-04-01 19:10:10,414 INFO [train.py:903] (0/4) Epoch 11, batch 2300, loss[loss=0.195, simple_loss=0.2728, pruned_loss=0.0586, over 19394.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.3073, pruned_loss=0.08117, over 3820700.19 frames. ], batch size: 48, lr: 7.64e-03, grad_scale: 4.0 +2023-04-01 19:10:19,765 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70587.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:10:23,155 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 19:11:05,003 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=70622.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:11:13,192 INFO [train.py:903] (0/4) Epoch 11, batch 2350, loss[loss=0.1924, simple_loss=0.2684, pruned_loss=0.05824, over 19771.00 frames. ], tot_loss[loss=0.2345, simple_loss=0.3072, pruned_loss=0.08096, over 3811411.25 frames. ], batch size: 46, lr: 7.64e-03, grad_scale: 4.0 +2023-04-01 19:11:56,886 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 19:12:06,973 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.408e+02 5.270e+02 6.524e+02 8.351e+02 2.247e+03, threshold=1.305e+03, percent-clipped=5.0 +2023-04-01 19:12:12,603 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 19:12:16,798 INFO [train.py:903] (0/4) Epoch 11, batch 2400, loss[loss=0.2515, simple_loss=0.3184, pruned_loss=0.09233, over 19739.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.3075, pruned_loss=0.0809, over 3814616.95 frames. ], batch size: 63, lr: 7.63e-03, grad_scale: 8.0 +2023-04-01 19:13:20,230 INFO [train.py:903] (0/4) Epoch 11, batch 2450, loss[loss=0.2457, simple_loss=0.3193, pruned_loss=0.08607, over 19736.00 frames. ], tot_loss[loss=0.2342, simple_loss=0.3071, pruned_loss=0.08068, over 3815846.91 frames. ], batch size: 63, lr: 7.63e-03, grad_scale: 8.0 +2023-04-01 19:13:32,391 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=70737.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:13:52,888 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3880, 1.3846, 1.7407, 1.5107, 2.5925, 2.2734, 2.7666, 1.1037], + device='cuda:0'), covar=tensor([0.2160, 0.3887, 0.2325, 0.1776, 0.1357, 0.1780, 0.1336, 0.3769], + device='cuda:0'), in_proj_covar=tensor([0.0485, 0.0574, 0.0599, 0.0437, 0.0594, 0.0490, 0.0649, 0.0491], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 19:14:09,923 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4415, 2.4430, 1.7285, 1.6958, 2.2597, 1.2803, 1.2103, 1.8931], + device='cuda:0'), covar=tensor([0.0970, 0.0567, 0.0971, 0.0668, 0.0445, 0.1142, 0.0802, 0.0442], + device='cuda:0'), in_proj_covar=tensor([0.0285, 0.0297, 0.0324, 0.0242, 0.0232, 0.0317, 0.0286, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 19:14:14,278 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.733e+02 5.717e+02 6.752e+02 8.721e+02 4.215e+03, threshold=1.350e+03, percent-clipped=8.0 +2023-04-01 19:14:25,709 INFO [train.py:903] (0/4) Epoch 11, batch 2500, loss[loss=0.1948, simple_loss=0.2689, pruned_loss=0.06032, over 19323.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.3078, pruned_loss=0.08065, over 3820274.35 frames. ], batch size: 44, lr: 7.63e-03, grad_scale: 8.0 +2023-04-01 19:15:27,558 INFO [train.py:903] (0/4) Epoch 11, batch 2550, loss[loss=0.2455, simple_loss=0.322, pruned_loss=0.08445, over 19338.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.3083, pruned_loss=0.08063, over 3832046.08 frames. ], batch size: 66, lr: 7.63e-03, grad_scale: 8.0 +2023-04-01 19:15:45,656 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70843.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:15:52,562 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7538, 3.1917, 3.2474, 3.3000, 1.2478, 3.1392, 2.7568, 2.9850], + device='cuda:0'), covar=tensor([0.1596, 0.0954, 0.0818, 0.0859, 0.5031, 0.0822, 0.0806, 0.1415], + device='cuda:0'), in_proj_covar=tensor([0.0665, 0.0597, 0.0793, 0.0676, 0.0723, 0.0544, 0.0482, 0.0733], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 19:15:58,216 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7320, 1.4808, 1.4038, 2.1340, 1.6420, 2.1842, 2.0932, 1.9058], + device='cuda:0'), covar=tensor([0.0808, 0.0932, 0.1025, 0.0859, 0.0873, 0.0600, 0.0867, 0.0663], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0225, 0.0224, 0.0250, 0.0237, 0.0212, 0.0198, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 19:16:17,501 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=70868.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:16:20,507 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.586e+02 5.428e+02 6.511e+02 7.969e+02 1.423e+03, threshold=1.302e+03, percent-clipped=4.0 +2023-04-01 19:16:26,507 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 19:16:30,145 INFO [train.py:903] (0/4) Epoch 11, batch 2600, loss[loss=0.2533, simple_loss=0.3272, pruned_loss=0.08967, over 19302.00 frames. ], tot_loss[loss=0.2353, simple_loss=0.3088, pruned_loss=0.08087, over 3830942.06 frames. ], batch size: 66, lr: 7.62e-03, grad_scale: 8.0 +2023-04-01 19:17:10,817 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.2383, 5.5236, 3.0263, 4.9437, 1.1914, 5.5522, 5.5476, 5.7256], + device='cuda:0'), covar=tensor([0.0380, 0.0859, 0.1744, 0.0627, 0.3788, 0.0492, 0.0599, 0.0752], + device='cuda:0'), in_proj_covar=tensor([0.0430, 0.0367, 0.0438, 0.0321, 0.0381, 0.0369, 0.0360, 0.0393], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 19:17:34,833 INFO [train.py:903] (0/4) Epoch 11, batch 2650, loss[loss=0.1905, simple_loss=0.2649, pruned_loss=0.05803, over 19763.00 frames. ], tot_loss[loss=0.2354, simple_loss=0.3089, pruned_loss=0.08098, over 3829612.94 frames. ], batch size: 48, lr: 7.62e-03, grad_scale: 8.0 +2023-04-01 19:17:41,049 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8832, 1.9345, 2.0737, 2.8319, 1.9028, 2.5083, 2.3836, 1.9428], + device='cuda:0'), covar=tensor([0.3451, 0.2970, 0.1495, 0.1693, 0.3293, 0.1504, 0.3306, 0.2609], + device='cuda:0'), in_proj_covar=tensor([0.0786, 0.0800, 0.0646, 0.0889, 0.0773, 0.0694, 0.0781, 0.0706], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 19:17:56,635 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 19:18:28,056 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.326e+02 5.322e+02 6.264e+02 9.085e+02 2.401e+03, threshold=1.253e+03, percent-clipped=10.0 +2023-04-01 19:18:39,372 INFO [train.py:903] (0/4) Epoch 11, batch 2700, loss[loss=0.2392, simple_loss=0.3167, pruned_loss=0.08085, over 19635.00 frames. ], tot_loss[loss=0.2373, simple_loss=0.31, pruned_loss=0.08223, over 3804402.00 frames. ], batch size: 57, lr: 7.62e-03, grad_scale: 8.0 +2023-04-01 19:18:56,213 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=70993.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:19:18,212 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71010.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:19:30,050 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=71018.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:19:43,440 INFO [train.py:903] (0/4) Epoch 11, batch 2750, loss[loss=0.1725, simple_loss=0.2474, pruned_loss=0.04882, over 19719.00 frames. ], tot_loss[loss=0.2354, simple_loss=0.3084, pruned_loss=0.08123, over 3805440.20 frames. ], batch size: 46, lr: 7.62e-03, grad_scale: 8.0 +2023-04-01 19:20:08,207 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71048.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:20:15,976 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-04-01 19:20:37,001 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.532e+02 5.301e+02 6.565e+02 7.951e+02 1.458e+03, threshold=1.313e+03, percent-clipped=6.0 +2023-04-01 19:20:45,882 INFO [train.py:903] (0/4) Epoch 11, batch 2800, loss[loss=0.2436, simple_loss=0.3203, pruned_loss=0.08338, over 19759.00 frames. ], tot_loss[loss=0.237, simple_loss=0.3095, pruned_loss=0.08225, over 3807531.24 frames. ], batch size: 63, lr: 7.61e-03, grad_scale: 8.0 +2023-04-01 19:21:51,269 INFO [train.py:903] (0/4) Epoch 11, batch 2850, loss[loss=0.1941, simple_loss=0.2703, pruned_loss=0.05894, over 19753.00 frames. ], tot_loss[loss=0.2349, simple_loss=0.3078, pruned_loss=0.08106, over 3815455.96 frames. ], batch size: 48, lr: 7.61e-03, grad_scale: 8.0 +2023-04-01 19:22:45,075 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.239e+02 5.466e+02 6.713e+02 8.640e+02 2.236e+03, threshold=1.343e+03, percent-clipped=7.0 +2023-04-01 19:22:54,319 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 19:22:55,475 INFO [train.py:903] (0/4) Epoch 11, batch 2900, loss[loss=0.276, simple_loss=0.3412, pruned_loss=0.1054, over 19454.00 frames. ], tot_loss[loss=0.2351, simple_loss=0.3078, pruned_loss=0.08119, over 3804644.79 frames. ], batch size: 64, lr: 7.61e-03, grad_scale: 8.0 +2023-04-01 19:24:00,138 INFO [train.py:903] (0/4) Epoch 11, batch 2950, loss[loss=0.2977, simple_loss=0.3549, pruned_loss=0.1202, over 19297.00 frames. ], tot_loss[loss=0.2358, simple_loss=0.3083, pruned_loss=0.08164, over 3786424.77 frames. ], batch size: 66, lr: 7.60e-03, grad_scale: 8.0 +2023-04-01 19:24:53,525 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.649e+02 5.471e+02 6.960e+02 8.532e+02 1.699e+03, threshold=1.392e+03, percent-clipped=7.0 +2023-04-01 19:25:02,835 INFO [train.py:903] (0/4) Epoch 11, batch 3000, loss[loss=0.2348, simple_loss=0.3236, pruned_loss=0.07301, over 18844.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.3084, pruned_loss=0.08098, over 3809299.87 frames. ], batch size: 74, lr: 7.60e-03, grad_scale: 8.0 +2023-04-01 19:25:02,836 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 19:25:16,083 INFO [train.py:937] (0/4) Epoch 11, validation: loss=0.1785, simple_loss=0.2793, pruned_loss=0.0389, over 944034.00 frames. +2023-04-01 19:25:16,084 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 19:25:20,543 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 19:26:20,108 INFO [train.py:903] (0/4) Epoch 11, batch 3050, loss[loss=0.2588, simple_loss=0.3296, pruned_loss=0.09402, over 18857.00 frames. ], tot_loss[loss=0.2366, simple_loss=0.3094, pruned_loss=0.08189, over 3792220.42 frames. ], batch size: 74, lr: 7.60e-03, grad_scale: 8.0 +2023-04-01 19:26:41,478 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-01 19:26:51,662 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=71354.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:27:13,585 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.278e+02 5.484e+02 7.005e+02 9.170e+02 2.820e+03, threshold=1.401e+03, percent-clipped=8.0 +2023-04-01 19:27:23,002 INFO [train.py:903] (0/4) Epoch 11, batch 3100, loss[loss=0.2289, simple_loss=0.3061, pruned_loss=0.07583, over 19523.00 frames. ], tot_loss[loss=0.2369, simple_loss=0.3093, pruned_loss=0.08222, over 3786807.37 frames. ], batch size: 54, lr: 7.60e-03, grad_scale: 8.0 +2023-04-01 19:27:36,327 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.61 vs. limit=2.0 +2023-04-01 19:27:39,073 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=71392.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:28:24,248 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6040, 1.3046, 1.4412, 1.7788, 1.4506, 1.7853, 1.7835, 1.6444], + device='cuda:0'), covar=tensor([0.0742, 0.0977, 0.0961, 0.0684, 0.0794, 0.0710, 0.0837, 0.0667], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0226, 0.0223, 0.0250, 0.0236, 0.0214, 0.0196, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 19:28:26,265 INFO [train.py:903] (0/4) Epoch 11, batch 3150, loss[loss=0.2061, simple_loss=0.2884, pruned_loss=0.0619, over 19595.00 frames. ], tot_loss[loss=0.2365, simple_loss=0.3088, pruned_loss=0.08214, over 3804644.39 frames. ], batch size: 52, lr: 7.59e-03, grad_scale: 8.0 +2023-04-01 19:28:55,931 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 19:29:17,613 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=71469.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:29:19,605 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.562e+02 5.450e+02 6.508e+02 8.826e+02 1.534e+03, threshold=1.302e+03, percent-clipped=1.0 +2023-04-01 19:29:31,367 INFO [train.py:903] (0/4) Epoch 11, batch 3200, loss[loss=0.1904, simple_loss=0.2591, pruned_loss=0.06081, over 19348.00 frames. ], tot_loss[loss=0.2337, simple_loss=0.3064, pruned_loss=0.08048, over 3814495.90 frames. ], batch size: 47, lr: 7.59e-03, grad_scale: 8.0 +2023-04-01 19:30:03,841 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7951, 1.5738, 1.4220, 1.8852, 1.6238, 1.5619, 1.5075, 1.7856], + device='cuda:0'), covar=tensor([0.0938, 0.1464, 0.1428, 0.0866, 0.1212, 0.0546, 0.1178, 0.0691], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0348, 0.0288, 0.0236, 0.0296, 0.0244, 0.0277, 0.0238], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 19:30:05,881 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=71507.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:30:32,439 INFO [train.py:903] (0/4) Epoch 11, batch 3250, loss[loss=0.2786, simple_loss=0.3496, pruned_loss=0.1038, over 18176.00 frames. ], tot_loss[loss=0.2359, simple_loss=0.3084, pruned_loss=0.08172, over 3822689.22 frames. ], batch size: 83, lr: 7.59e-03, grad_scale: 8.0 +2023-04-01 19:31:26,706 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.701e+02 5.157e+02 6.415e+02 8.641e+02 1.397e+03, threshold=1.283e+03, percent-clipped=1.0 +2023-04-01 19:31:36,061 INFO [train.py:903] (0/4) Epoch 11, batch 3300, loss[loss=0.228, simple_loss=0.3081, pruned_loss=0.07391, over 18105.00 frames. ], tot_loss[loss=0.2347, simple_loss=0.3072, pruned_loss=0.08115, over 3820258.81 frames. ], batch size: 83, lr: 7.59e-03, grad_scale: 8.0 +2023-04-01 19:31:41,739 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 19:32:11,536 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.72 vs. limit=5.0 +2023-04-01 19:32:39,941 INFO [train.py:903] (0/4) Epoch 11, batch 3350, loss[loss=0.2241, simple_loss=0.3046, pruned_loss=0.07176, over 18771.00 frames. ], tot_loss[loss=0.2356, simple_loss=0.3081, pruned_loss=0.08152, over 3815552.86 frames. ], batch size: 74, lr: 7.58e-03, grad_scale: 8.0 +2023-04-01 19:32:46,619 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.10 vs. limit=2.0 +2023-04-01 19:32:47,351 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71634.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:33:34,249 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.542e+02 5.202e+02 6.815e+02 8.173e+02 2.322e+03, threshold=1.363e+03, percent-clipped=5.0 +2023-04-01 19:33:44,450 INFO [train.py:903] (0/4) Epoch 11, batch 3400, loss[loss=0.2706, simple_loss=0.3388, pruned_loss=0.1012, over 19581.00 frames. ], tot_loss[loss=0.2358, simple_loss=0.3085, pruned_loss=0.08154, over 3815470.65 frames. ], batch size: 52, lr: 7.58e-03, grad_scale: 8.0 +2023-04-01 19:34:44,104 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71725.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:34:48,499 INFO [train.py:903] (0/4) Epoch 11, batch 3450, loss[loss=0.2267, simple_loss=0.3073, pruned_loss=0.07301, over 18107.00 frames. ], tot_loss[loss=0.2356, simple_loss=0.3085, pruned_loss=0.08132, over 3817720.32 frames. ], batch size: 83, lr: 7.58e-03, grad_scale: 8.0 +2023-04-01 19:34:52,170 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 19:35:15,992 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=71750.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:35:33,028 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=71763.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:35:42,112 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.543e+02 5.649e+02 6.835e+02 8.667e+02 1.565e+03, threshold=1.367e+03, percent-clipped=4.0 +2023-04-01 19:35:47,962 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7978, 4.3742, 2.8415, 3.9104, 1.1678, 4.0918, 4.1440, 4.2809], + device='cuda:0'), covar=tensor([0.0689, 0.1164, 0.1842, 0.0713, 0.3984, 0.0742, 0.0791, 0.1098], + device='cuda:0'), in_proj_covar=tensor([0.0424, 0.0360, 0.0431, 0.0313, 0.0373, 0.0365, 0.0350, 0.0387], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 19:35:52,286 INFO [train.py:903] (0/4) Epoch 11, batch 3500, loss[loss=0.2405, simple_loss=0.3158, pruned_loss=0.08257, over 19665.00 frames. ], tot_loss[loss=0.2361, simple_loss=0.3092, pruned_loss=0.0815, over 3826415.29 frames. ], batch size: 60, lr: 7.58e-03, grad_scale: 8.0 +2023-04-01 19:36:04,486 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=71788.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:36:30,082 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5820, 1.5792, 1.6665, 1.6923, 3.1151, 1.1097, 2.4372, 3.5867], + device='cuda:0'), covar=tensor([0.0448, 0.2298, 0.2409, 0.1541, 0.0688, 0.2428, 0.1160, 0.0239], + device='cuda:0'), in_proj_covar=tensor([0.0351, 0.0332, 0.0349, 0.0315, 0.0343, 0.0329, 0.0328, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 19:36:56,411 INFO [train.py:903] (0/4) Epoch 11, batch 3550, loss[loss=0.2272, simple_loss=0.3114, pruned_loss=0.0715, over 19481.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.3084, pruned_loss=0.08097, over 3820080.94 frames. ], batch size: 64, lr: 7.57e-03, grad_scale: 8.0 +2023-04-01 19:37:49,026 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.173e+02 5.129e+02 6.236e+02 8.069e+02 1.994e+03, threshold=1.247e+03, percent-clipped=3.0 +2023-04-01 19:37:58,903 INFO [train.py:903] (0/4) Epoch 11, batch 3600, loss[loss=0.2897, simple_loss=0.3455, pruned_loss=0.117, over 19786.00 frames. ], tot_loss[loss=0.2356, simple_loss=0.3087, pruned_loss=0.08122, over 3812609.96 frames. ], batch size: 56, lr: 7.57e-03, grad_scale: 8.0 +2023-04-01 19:39:03,005 INFO [train.py:903] (0/4) Epoch 11, batch 3650, loss[loss=0.2327, simple_loss=0.3026, pruned_loss=0.0814, over 19482.00 frames. ], tot_loss[loss=0.235, simple_loss=0.3077, pruned_loss=0.08114, over 3795828.50 frames. ], batch size: 49, lr: 7.57e-03, grad_scale: 8.0 +2023-04-01 19:39:13,487 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=71937.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:39:55,946 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.339e+02 5.422e+02 6.726e+02 7.997e+02 1.955e+03, threshold=1.345e+03, percent-clipped=5.0 +2023-04-01 19:40:05,223 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=71978.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:40:06,260 INFO [train.py:903] (0/4) Epoch 11, batch 3700, loss[loss=0.2836, simple_loss=0.3428, pruned_loss=0.1122, over 13240.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.3083, pruned_loss=0.08104, over 3797937.33 frames. ], batch size: 136, lr: 7.57e-03, grad_scale: 8.0 +2023-04-01 19:40:33,202 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-72000.pt +2023-04-01 19:41:02,232 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-04-01 19:41:11,920 INFO [train.py:903] (0/4) Epoch 11, batch 3750, loss[loss=0.2372, simple_loss=0.3183, pruned_loss=0.07804, over 19755.00 frames. ], tot_loss[loss=0.2344, simple_loss=0.3071, pruned_loss=0.08086, over 3796651.55 frames. ], batch size: 63, lr: 7.56e-03, grad_scale: 8.0 +2023-04-01 19:42:06,342 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.116e+02 5.141e+02 6.053e+02 7.322e+02 1.150e+03, threshold=1.211e+03, percent-clipped=0.0 +2023-04-01 19:42:17,090 INFO [train.py:903] (0/4) Epoch 11, batch 3800, loss[loss=0.1878, simple_loss=0.2673, pruned_loss=0.05418, over 19383.00 frames. ], tot_loss[loss=0.2349, simple_loss=0.3079, pruned_loss=0.08095, over 3791128.66 frames. ], batch size: 48, lr: 7.56e-03, grad_scale: 8.0 +2023-04-01 19:42:22,598 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-04-01 19:42:35,662 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72093.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:42:46,962 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 19:43:21,151 INFO [train.py:903] (0/4) Epoch 11, batch 3850, loss[loss=0.2437, simple_loss=0.3205, pruned_loss=0.08343, over 19318.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.3083, pruned_loss=0.08063, over 3799761.12 frames. ], batch size: 66, lr: 7.56e-03, grad_scale: 8.0 +2023-04-01 19:43:45,116 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72147.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:44:16,373 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.216e+02 5.600e+02 6.929e+02 8.545e+02 2.074e+03, threshold=1.386e+03, percent-clipped=5.0 +2023-04-01 19:44:25,945 INFO [train.py:903] (0/4) Epoch 11, batch 3900, loss[loss=0.2336, simple_loss=0.3186, pruned_loss=0.07428, over 19532.00 frames. ], tot_loss[loss=0.236, simple_loss=0.3097, pruned_loss=0.08118, over 3807134.01 frames. ], batch size: 56, lr: 7.55e-03, grad_scale: 8.0 +2023-04-01 19:45:04,300 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1768, 1.2913, 1.7556, 1.2774, 2.6285, 3.5280, 3.2199, 3.7249], + device='cuda:0'), covar=tensor([0.1501, 0.3254, 0.2814, 0.2032, 0.0469, 0.0167, 0.0203, 0.0171], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0294, 0.0321, 0.0251, 0.0213, 0.0154, 0.0206, 0.0199], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 19:45:30,660 INFO [train.py:903] (0/4) Epoch 11, batch 3950, loss[loss=0.255, simple_loss=0.3251, pruned_loss=0.09241, over 18932.00 frames. ], tot_loss[loss=0.2359, simple_loss=0.3096, pruned_loss=0.08116, over 3814366.99 frames. ], batch size: 75, lr: 7.55e-03, grad_scale: 8.0 +2023-04-01 19:45:31,930 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 19:46:23,815 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.514e+02 5.760e+02 7.487e+02 9.416e+02 2.541e+03, threshold=1.497e+03, percent-clipped=9.0 +2023-04-01 19:46:34,101 INFO [train.py:903] (0/4) Epoch 11, batch 4000, loss[loss=0.2416, simple_loss=0.3157, pruned_loss=0.08373, over 19677.00 frames. ], tot_loss[loss=0.2363, simple_loss=0.3097, pruned_loss=0.08143, over 3816384.52 frames. ], batch size: 53, lr: 7.55e-03, grad_scale: 8.0 +2023-04-01 19:46:36,413 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=72281.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:47:22,613 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 19:47:37,789 INFO [train.py:903] (0/4) Epoch 11, batch 4050, loss[loss=0.2192, simple_loss=0.3057, pruned_loss=0.06633, over 19678.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.3084, pruned_loss=0.08066, over 3830994.41 frames. ], batch size: 60, lr: 7.55e-03, grad_scale: 8.0 +2023-04-01 19:47:57,848 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72344.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:48:05,153 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=72349.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:48:33,046 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.561e+02 4.924e+02 6.616e+02 8.581e+02 2.031e+03, threshold=1.323e+03, percent-clipped=3.0 +2023-04-01 19:48:37,060 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72374.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:48:43,655 INFO [train.py:903] (0/4) Epoch 11, batch 4100, loss[loss=0.2374, simple_loss=0.3085, pruned_loss=0.08313, over 19587.00 frames. ], tot_loss[loss=0.2338, simple_loss=0.3074, pruned_loss=0.08014, over 3826809.36 frames. ], batch size: 52, lr: 7.54e-03, grad_scale: 8.0 +2023-04-01 19:49:05,224 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72396.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:49:18,922 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 19:49:48,315 INFO [train.py:903] (0/4) Epoch 11, batch 4150, loss[loss=0.2532, simple_loss=0.3249, pruned_loss=0.09081, over 18787.00 frames. ], tot_loss[loss=0.2354, simple_loss=0.3088, pruned_loss=0.081, over 3814702.02 frames. ], batch size: 74, lr: 7.54e-03, grad_scale: 8.0 +2023-04-01 19:50:01,639 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2489, 3.7831, 3.8849, 3.8780, 1.5612, 3.6676, 3.2049, 3.6386], + device='cuda:0'), covar=tensor([0.1421, 0.0723, 0.0612, 0.0670, 0.4851, 0.0623, 0.0651, 0.1109], + device='cuda:0'), in_proj_covar=tensor([0.0665, 0.0601, 0.0793, 0.0674, 0.0723, 0.0550, 0.0485, 0.0733], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 19:50:42,220 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.080e+02 4.902e+02 6.015e+02 8.099e+02 1.569e+03, threshold=1.203e+03, percent-clipped=3.0 +2023-04-01 19:50:42,485 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1241, 2.8645, 2.2081, 2.6278, 1.1522, 2.7157, 2.6926, 2.7649], + device='cuda:0'), covar=tensor([0.1156, 0.1287, 0.1796, 0.0740, 0.2730, 0.0932, 0.0890, 0.1149], + device='cuda:0'), in_proj_covar=tensor([0.0433, 0.0370, 0.0440, 0.0320, 0.0379, 0.0373, 0.0358, 0.0393], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 19:50:47,136 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8898, 4.3586, 4.6188, 4.5708, 1.7026, 4.2602, 3.7288, 4.3004], + device='cuda:0'), covar=tensor([0.1333, 0.0719, 0.0472, 0.0568, 0.4910, 0.0682, 0.0556, 0.0918], + device='cuda:0'), in_proj_covar=tensor([0.0668, 0.0602, 0.0795, 0.0676, 0.0724, 0.0552, 0.0485, 0.0734], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 19:50:51,446 INFO [train.py:903] (0/4) Epoch 11, batch 4200, loss[loss=0.2582, simple_loss=0.3285, pruned_loss=0.094, over 18083.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.3083, pruned_loss=0.08066, over 3815810.70 frames. ], batch size: 83, lr: 7.54e-03, grad_scale: 16.0 +2023-04-01 19:50:57,210 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 19:51:07,191 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=72491.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:51:56,560 INFO [train.py:903] (0/4) Epoch 11, batch 4250, loss[loss=0.2587, simple_loss=0.3336, pruned_loss=0.09186, over 19667.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.3081, pruned_loss=0.08058, over 3817058.58 frames. ], batch size: 55, lr: 7.54e-03, grad_scale: 8.0 +2023-04-01 19:52:17,354 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 19:52:23,704 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8051, 3.2310, 3.2950, 3.2959, 1.2928, 3.1358, 2.7484, 3.0454], + device='cuda:0'), covar=tensor([0.1514, 0.0879, 0.0757, 0.0860, 0.4708, 0.0854, 0.0725, 0.1232], + device='cuda:0'), in_proj_covar=tensor([0.0671, 0.0603, 0.0795, 0.0679, 0.0726, 0.0554, 0.0485, 0.0735], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 19:52:23,906 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6764, 1.7211, 1.8415, 2.0390, 1.4282, 1.8610, 2.0414, 1.7775], + device='cuda:0'), covar=tensor([0.3265, 0.2705, 0.1531, 0.1614, 0.2811, 0.1498, 0.3647, 0.2596], + device='cuda:0'), in_proj_covar=tensor([0.0792, 0.0811, 0.0651, 0.0896, 0.0781, 0.0707, 0.0790, 0.0713], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 19:52:28,268 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 19:52:51,987 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.976e+02 5.433e+02 6.415e+02 7.675e+02 1.468e+03, threshold=1.283e+03, percent-clipped=2.0 +2023-04-01 19:52:58,543 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5000, 1.5759, 2.0220, 1.6802, 3.0542, 2.5570, 3.2200, 1.6325], + device='cuda:0'), covar=tensor([0.2261, 0.3658, 0.2241, 0.1812, 0.1499, 0.1840, 0.1726, 0.3583], + device='cuda:0'), in_proj_covar=tensor([0.0490, 0.0572, 0.0596, 0.0435, 0.0589, 0.0489, 0.0645, 0.0493], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 19:53:01,653 INFO [train.py:903] (0/4) Epoch 11, batch 4300, loss[loss=0.1962, simple_loss=0.2802, pruned_loss=0.05612, over 19724.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.3082, pruned_loss=0.08112, over 3808599.88 frames. ], batch size: 51, lr: 7.53e-03, grad_scale: 8.0 +2023-04-01 19:53:26,277 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4927, 1.4237, 1.4675, 1.8412, 3.0505, 1.0571, 2.3337, 3.4247], + device='cuda:0'), covar=tensor([0.0470, 0.2467, 0.2593, 0.1433, 0.0695, 0.2437, 0.1143, 0.0273], + device='cuda:0'), in_proj_covar=tensor([0.0354, 0.0336, 0.0350, 0.0319, 0.0345, 0.0331, 0.0333, 0.0352], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 19:53:36,383 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72606.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:54:00,212 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 19:54:04,822 INFO [train.py:903] (0/4) Epoch 11, batch 4350, loss[loss=0.2728, simple_loss=0.3312, pruned_loss=0.1071, over 19536.00 frames. ], tot_loss[loss=0.2351, simple_loss=0.3077, pruned_loss=0.08128, over 3811887.28 frames. ], batch size: 54, lr: 7.53e-03, grad_scale: 8.0 +2023-04-01 19:54:33,998 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=72652.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:54:58,901 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.400e+02 5.206e+02 6.524e+02 8.372e+02 1.509e+03, threshold=1.305e+03, percent-clipped=4.0 +2023-04-01 19:55:06,320 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72677.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:55:08,340 INFO [train.py:903] (0/4) Epoch 11, batch 4400, loss[loss=0.1902, simple_loss=0.2622, pruned_loss=0.05909, over 19752.00 frames. ], tot_loss[loss=0.235, simple_loss=0.3076, pruned_loss=0.08121, over 3819753.77 frames. ], batch size: 48, lr: 7.53e-03, grad_scale: 8.0 +2023-04-01 19:55:20,139 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=72688.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:55:31,030 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8670, 1.9891, 2.0947, 2.6956, 1.8520, 2.5030, 2.4148, 1.9778], + device='cuda:0'), covar=tensor([0.3451, 0.2949, 0.1412, 0.1834, 0.3387, 0.1500, 0.3141, 0.2481], + device='cuda:0'), in_proj_covar=tensor([0.0788, 0.0806, 0.0646, 0.0890, 0.0776, 0.0705, 0.0787, 0.0706], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 19:55:37,583 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 19:55:47,758 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 19:56:12,269 INFO [train.py:903] (0/4) Epoch 11, batch 4450, loss[loss=0.268, simple_loss=0.3288, pruned_loss=0.1036, over 19778.00 frames. ], tot_loss[loss=0.2334, simple_loss=0.3065, pruned_loss=0.08014, over 3823053.91 frames. ], batch size: 54, lr: 7.53e-03, grad_scale: 8.0 +2023-04-01 19:57:06,052 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.681e+02 5.500e+02 6.922e+02 8.863e+02 1.965e+03, threshold=1.384e+03, percent-clipped=10.0 +2023-04-01 19:57:14,267 INFO [train.py:903] (0/4) Epoch 11, batch 4500, loss[loss=0.2296, simple_loss=0.3039, pruned_loss=0.07769, over 19704.00 frames. ], tot_loss[loss=0.235, simple_loss=0.3076, pruned_loss=0.08123, over 3823413.67 frames. ], batch size: 59, lr: 7.52e-03, grad_scale: 8.0 +2023-04-01 19:57:46,498 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=72803.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:58:19,529 INFO [train.py:903] (0/4) Epoch 11, batch 4550, loss[loss=0.2442, simple_loss=0.3166, pruned_loss=0.0859, over 19673.00 frames. ], tot_loss[loss=0.2351, simple_loss=0.3079, pruned_loss=0.0812, over 3807593.58 frames. ], batch size: 58, lr: 7.52e-03, grad_scale: 8.0 +2023-04-01 19:58:28,676 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 19:58:53,317 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 19:59:02,253 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=72862.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 19:59:15,579 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.061e+02 5.895e+02 7.318e+02 9.053e+02 1.617e+03, threshold=1.464e+03, percent-clipped=1.0 +2023-04-01 19:59:16,613 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-04-01 19:59:23,893 INFO [train.py:903] (0/4) Epoch 11, batch 4600, loss[loss=0.2148, simple_loss=0.296, pruned_loss=0.06679, over 19610.00 frames. ], tot_loss[loss=0.2342, simple_loss=0.307, pruned_loss=0.08067, over 3809184.05 frames. ], batch size: 57, lr: 7.52e-03, grad_scale: 8.0 +2023-04-01 19:59:34,658 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=72887.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:00:04,229 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 20:00:06,131 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72912.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:00:27,747 INFO [train.py:903] (0/4) Epoch 11, batch 4650, loss[loss=0.2772, simple_loss=0.3475, pruned_loss=0.1035, over 18180.00 frames. ], tot_loss[loss=0.2338, simple_loss=0.3067, pruned_loss=0.08051, over 3806005.96 frames. ], batch size: 83, lr: 7.52e-03, grad_scale: 8.0 +2023-04-01 20:00:46,738 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 20:00:57,807 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 20:01:22,039 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.839e+02 5.234e+02 6.593e+02 8.122e+02 1.324e+03, threshold=1.319e+03, percent-clipped=0.0 +2023-04-01 20:01:30,191 INFO [train.py:903] (0/4) Epoch 11, batch 4700, loss[loss=0.2057, simple_loss=0.2711, pruned_loss=0.07012, over 19098.00 frames. ], tot_loss[loss=0.2331, simple_loss=0.3063, pruned_loss=0.08001, over 3807708.74 frames. ], batch size: 42, lr: 7.51e-03, grad_scale: 8.0 +2023-04-01 20:01:34,977 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=72982.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 20:01:55,722 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 20:02:34,869 INFO [train.py:903] (0/4) Epoch 11, batch 4750, loss[loss=0.2983, simple_loss=0.3563, pruned_loss=0.1201, over 18048.00 frames. ], tot_loss[loss=0.2339, simple_loss=0.3068, pruned_loss=0.08049, over 3797010.72 frames. ], batch size: 83, lr: 7.51e-03, grad_scale: 8.0 +2023-04-01 20:03:11,979 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73059.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:03:28,794 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.570e+02 5.728e+02 6.958e+02 8.518e+02 2.013e+03, threshold=1.392e+03, percent-clipped=2.0 +2023-04-01 20:03:37,833 INFO [train.py:903] (0/4) Epoch 11, batch 4800, loss[loss=0.2325, simple_loss=0.3147, pruned_loss=0.07513, over 19668.00 frames. ], tot_loss[loss=0.2337, simple_loss=0.3068, pruned_loss=0.0803, over 3801556.05 frames. ], batch size: 60, lr: 7.51e-03, grad_scale: 8.0 +2023-04-01 20:03:38,476 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 20:03:44,153 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=73084.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:04:18,422 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9628, 3.6021, 2.3618, 3.2819, 0.9593, 3.4683, 3.3582, 3.4213], + device='cuda:0'), covar=tensor([0.0800, 0.1065, 0.1996, 0.0759, 0.3699, 0.0843, 0.0890, 0.1192], + device='cuda:0'), in_proj_covar=tensor([0.0430, 0.0364, 0.0433, 0.0315, 0.0376, 0.0370, 0.0356, 0.0389], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:04:26,323 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2850, 1.2962, 1.6463, 1.4400, 2.1630, 1.9188, 2.1087, 0.8478], + device='cuda:0'), covar=tensor([0.2348, 0.4224, 0.2459, 0.1974, 0.1528, 0.2239, 0.1606, 0.3993], + device='cuda:0'), in_proj_covar=tensor([0.0490, 0.0577, 0.0601, 0.0438, 0.0591, 0.0493, 0.0645, 0.0497], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 20:04:40,996 INFO [train.py:903] (0/4) Epoch 11, batch 4850, loss[loss=0.2508, simple_loss=0.3238, pruned_loss=0.08885, over 19466.00 frames. ], tot_loss[loss=0.2342, simple_loss=0.3071, pruned_loss=0.08068, over 3791715.14 frames. ], batch size: 64, lr: 7.51e-03, grad_scale: 8.0 +2023-04-01 20:05:01,952 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 20:05:22,868 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 20:05:29,925 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 20:05:29,953 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 20:05:35,783 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.053e+02 5.592e+02 6.791e+02 8.044e+02 1.982e+03, threshold=1.358e+03, percent-clipped=5.0 +2023-04-01 20:05:39,477 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 20:05:44,089 INFO [train.py:903] (0/4) Epoch 11, batch 4900, loss[loss=0.2352, simple_loss=0.3145, pruned_loss=0.07793, over 18730.00 frames. ], tot_loss[loss=0.2347, simple_loss=0.3076, pruned_loss=0.08095, over 3794059.53 frames. ], batch size: 74, lr: 7.50e-03, grad_scale: 8.0 +2023-04-01 20:05:57,050 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3349, 1.2275, 1.4333, 1.3799, 2.8463, 0.8932, 2.0516, 3.2301], + device='cuda:0'), covar=tensor([0.0488, 0.2702, 0.2715, 0.1809, 0.0769, 0.2679, 0.1338, 0.0308], + device='cuda:0'), in_proj_covar=tensor([0.0355, 0.0337, 0.0352, 0.0321, 0.0344, 0.0332, 0.0331, 0.0351], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:06:01,202 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 20:06:48,389 INFO [train.py:903] (0/4) Epoch 11, batch 4950, loss[loss=0.2058, simple_loss=0.2867, pruned_loss=0.06244, over 19667.00 frames. ], tot_loss[loss=0.2334, simple_loss=0.3067, pruned_loss=0.08007, over 3798806.81 frames. ], batch size: 53, lr: 7.50e-03, grad_scale: 8.0 +2023-04-01 20:07:00,929 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 20:07:21,265 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=73256.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:07:24,672 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 20:07:42,671 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.726e+02 5.448e+02 7.218e+02 1.051e+03 2.092e+03, threshold=1.444e+03, percent-clipped=5.0 +2023-04-01 20:07:51,005 INFO [train.py:903] (0/4) Epoch 11, batch 5000, loss[loss=0.2209, simple_loss=0.3035, pruned_loss=0.06913, over 19670.00 frames. ], tot_loss[loss=0.2333, simple_loss=0.3065, pruned_loss=0.08004, over 3808639.65 frames. ], batch size: 58, lr: 7.50e-03, grad_scale: 8.0 +2023-04-01 20:07:55,889 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 20:08:08,619 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 20:08:13,781 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3529, 2.0419, 2.4980, 2.7682, 2.1622, 2.7222, 2.7974, 2.6731], + device='cuda:0'), covar=tensor([0.0762, 0.0914, 0.0832, 0.0936, 0.0914, 0.0672, 0.0876, 0.0546], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0224, 0.0223, 0.0250, 0.0235, 0.0213, 0.0194, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 20:08:32,026 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73311.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:08:50,649 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=73326.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 20:08:54,859 INFO [train.py:903] (0/4) Epoch 11, batch 5050, loss[loss=0.2363, simple_loss=0.3226, pruned_loss=0.07497, over 19695.00 frames. ], tot_loss[loss=0.2315, simple_loss=0.3054, pruned_loss=0.07879, over 3827870.63 frames. ], batch size: 59, lr: 7.50e-03, grad_scale: 8.0 +2023-04-01 20:09:28,969 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 20:09:47,509 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=73371.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:09:48,292 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.996e+02 5.337e+02 6.487e+02 8.820e+02 1.986e+03, threshold=1.297e+03, percent-clipped=3.0 +2023-04-01 20:09:56,474 INFO [train.py:903] (0/4) Epoch 11, batch 5100, loss[loss=0.2338, simple_loss=0.3116, pruned_loss=0.07804, over 19668.00 frames. ], tot_loss[loss=0.2338, simple_loss=0.307, pruned_loss=0.08035, over 3808419.56 frames. ], batch size: 60, lr: 7.49e-03, grad_scale: 8.0 +2023-04-01 20:10:04,900 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 20:10:10,277 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 20:10:13,775 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 20:10:22,121 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3125, 3.8858, 2.5435, 3.4992, 0.8793, 3.7784, 3.6453, 3.8111], + device='cuda:0'), covar=tensor([0.0678, 0.1011, 0.1992, 0.0738, 0.4139, 0.0756, 0.0802, 0.1015], + device='cuda:0'), in_proj_covar=tensor([0.0432, 0.0363, 0.0437, 0.0317, 0.0379, 0.0373, 0.0357, 0.0390], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:11:00,241 INFO [train.py:903] (0/4) Epoch 11, batch 5150, loss[loss=0.2571, simple_loss=0.3302, pruned_loss=0.09205, over 19541.00 frames. ], tot_loss[loss=0.2329, simple_loss=0.3063, pruned_loss=0.07978, over 3808817.39 frames. ], batch size: 54, lr: 7.49e-03, grad_scale: 8.0 +2023-04-01 20:11:09,671 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 20:11:16,019 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=73441.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 20:11:31,498 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3140, 1.1407, 1.2020, 1.1539, 2.0161, 0.8602, 1.7634, 2.1782], + device='cuda:0'), covar=tensor([0.0657, 0.2308, 0.2453, 0.1478, 0.0824, 0.1998, 0.1066, 0.0512], + device='cuda:0'), in_proj_covar=tensor([0.0354, 0.0334, 0.0348, 0.0320, 0.0344, 0.0331, 0.0328, 0.0350], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:11:43,077 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 20:11:50,701 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-04-01 20:11:54,720 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.282e+02 5.910e+02 7.077e+02 9.438e+02 2.777e+03, threshold=1.415e+03, percent-clipped=4.0 +2023-04-01 20:12:04,001 INFO [train.py:903] (0/4) Epoch 11, batch 5200, loss[loss=0.2273, simple_loss=0.3103, pruned_loss=0.07217, over 19529.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.3065, pruned_loss=0.07996, over 3818899.60 frames. ], batch size: 54, lr: 7.49e-03, grad_scale: 8.0 +2023-04-01 20:12:16,643 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 20:13:00,031 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 20:13:07,094 INFO [train.py:903] (0/4) Epoch 11, batch 5250, loss[loss=0.257, simple_loss=0.3297, pruned_loss=0.09213, over 19597.00 frames. ], tot_loss[loss=0.2335, simple_loss=0.307, pruned_loss=0.07996, over 3830431.31 frames. ], batch size: 57, lr: 7.49e-03, grad_scale: 8.0 +2023-04-01 20:14:02,591 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.872e+02 5.338e+02 6.658e+02 8.757e+02 1.866e+03, threshold=1.332e+03, percent-clipped=3.0 +2023-04-01 20:14:11,102 INFO [train.py:903] (0/4) Epoch 11, batch 5300, loss[loss=0.2227, simple_loss=0.303, pruned_loss=0.07122, over 19696.00 frames. ], tot_loss[loss=0.2337, simple_loss=0.3073, pruned_loss=0.08, over 3829531.15 frames. ], batch size: 59, lr: 7.48e-03, grad_scale: 8.0 +2023-04-01 20:14:26,095 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 20:14:29,971 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3461, 2.2364, 1.8985, 1.7363, 1.5580, 1.7495, 0.5974, 1.2245], + device='cuda:0'), covar=tensor([0.0400, 0.0382, 0.0327, 0.0519, 0.0874, 0.0623, 0.0750, 0.0689], + device='cuda:0'), in_proj_covar=tensor([0.0334, 0.0326, 0.0329, 0.0348, 0.0424, 0.0349, 0.0304, 0.0318], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 20:14:43,191 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2929, 3.7916, 3.9053, 3.8932, 1.5659, 3.6903, 3.1999, 3.6318], + device='cuda:0'), covar=tensor([0.1427, 0.0759, 0.0594, 0.0667, 0.4777, 0.0659, 0.0668, 0.1089], + device='cuda:0'), in_proj_covar=tensor([0.0666, 0.0600, 0.0794, 0.0678, 0.0724, 0.0553, 0.0485, 0.0730], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 20:14:49,204 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73608.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:15:12,298 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73627.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:15:15,335 INFO [train.py:903] (0/4) Epoch 11, batch 5350, loss[loss=0.2382, simple_loss=0.2999, pruned_loss=0.08831, over 19479.00 frames. ], tot_loss[loss=0.2347, simple_loss=0.3085, pruned_loss=0.08045, over 3823666.16 frames. ], batch size: 49, lr: 7.48e-03, grad_scale: 8.0 +2023-04-01 20:15:46,113 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=73652.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:15:46,881 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 20:15:49,348 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=73655.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:16:09,903 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.174e+02 5.285e+02 6.573e+02 8.256e+02 1.333e+03, threshold=1.315e+03, percent-clipped=1.0 +2023-04-01 20:16:19,929 INFO [train.py:903] (0/4) Epoch 11, batch 5400, loss[loss=0.2879, simple_loss=0.3458, pruned_loss=0.115, over 18905.00 frames. ], tot_loss[loss=0.235, simple_loss=0.3087, pruned_loss=0.08063, over 3812441.81 frames. ], batch size: 75, lr: 7.48e-03, grad_scale: 8.0 +2023-04-01 20:16:41,368 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=73697.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 20:17:13,022 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=73722.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 20:17:20,603 INFO [train.py:903] (0/4) Epoch 11, batch 5450, loss[loss=0.2773, simple_loss=0.3423, pruned_loss=0.1062, over 19544.00 frames. ], tot_loss[loss=0.2365, simple_loss=0.3096, pruned_loss=0.08169, over 3821586.68 frames. ], batch size: 64, lr: 7.48e-03, grad_scale: 8.0 +2023-04-01 20:17:45,878 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=73750.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:18:11,991 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=73770.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:18:14,014 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.246e+02 5.707e+02 7.323e+02 9.356e+02 2.024e+03, threshold=1.465e+03, percent-clipped=8.0 +2023-04-01 20:18:23,353 INFO [train.py:903] (0/4) Epoch 11, batch 5500, loss[loss=0.1991, simple_loss=0.2855, pruned_loss=0.05635, over 19678.00 frames. ], tot_loss[loss=0.2363, simple_loss=0.3096, pruned_loss=0.08147, over 3836822.73 frames. ], batch size: 59, lr: 7.47e-03, grad_scale: 8.0 +2023-04-01 20:18:50,423 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 20:19:25,475 INFO [train.py:903] (0/4) Epoch 11, batch 5550, loss[loss=0.2097, simple_loss=0.2929, pruned_loss=0.06329, over 19614.00 frames. ], tot_loss[loss=0.2363, simple_loss=0.3095, pruned_loss=0.08158, over 3831852.34 frames. ], batch size: 57, lr: 7.47e-03, grad_scale: 8.0 +2023-04-01 20:19:35,069 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 20:20:04,808 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1156, 1.2672, 1.7328, 1.1226, 2.4868, 3.3275, 3.1161, 3.5655], + device='cuda:0'), covar=tensor([0.1525, 0.3281, 0.2870, 0.2106, 0.0457, 0.0162, 0.0209, 0.0180], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0292, 0.0321, 0.0248, 0.0212, 0.0156, 0.0205, 0.0200], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 20:20:20,656 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.349e+02 5.761e+02 6.752e+02 8.536e+02 1.570e+03, threshold=1.350e+03, percent-clipped=1.0 +2023-04-01 20:20:26,340 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 20:20:29,855 INFO [train.py:903] (0/4) Epoch 11, batch 5600, loss[loss=0.3334, simple_loss=0.3865, pruned_loss=0.1402, over 17883.00 frames. ], tot_loss[loss=0.2351, simple_loss=0.3081, pruned_loss=0.081, over 3839827.88 frames. ], batch size: 83, lr: 7.47e-03, grad_scale: 8.0 +2023-04-01 20:21:24,344 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6217, 1.2524, 1.5513, 1.4427, 3.1010, 1.0262, 2.2111, 3.5302], + device='cuda:0'), covar=tensor([0.0388, 0.2727, 0.2504, 0.1797, 0.0709, 0.2413, 0.1169, 0.0260], + device='cuda:0'), in_proj_covar=tensor([0.0353, 0.0338, 0.0349, 0.0322, 0.0347, 0.0331, 0.0332, 0.0353], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:21:34,078 INFO [train.py:903] (0/4) Epoch 11, batch 5650, loss[loss=0.226, simple_loss=0.3052, pruned_loss=0.07343, over 19525.00 frames. ], tot_loss[loss=0.2343, simple_loss=0.3075, pruned_loss=0.0806, over 3826488.54 frames. ], batch size: 54, lr: 7.47e-03, grad_scale: 8.0 +2023-04-01 20:22:02,981 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=73952.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:22:23,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 20:22:28,536 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.512e+02 5.880e+02 7.006e+02 8.632e+02 1.687e+03, threshold=1.401e+03, percent-clipped=2.0 +2023-04-01 20:22:37,807 INFO [train.py:903] (0/4) Epoch 11, batch 5700, loss[loss=0.3035, simple_loss=0.361, pruned_loss=0.123, over 13508.00 frames. ], tot_loss[loss=0.2349, simple_loss=0.3081, pruned_loss=0.08084, over 3813582.92 frames. ], batch size: 136, lr: 7.46e-03, grad_scale: 8.0 +2023-04-01 20:23:03,726 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-74000.pt +2023-04-01 20:23:13,090 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4500, 1.5698, 2.0544, 1.8337, 3.3838, 2.7189, 3.6442, 1.6856], + device='cuda:0'), covar=tensor([0.2182, 0.3696, 0.2225, 0.1560, 0.1290, 0.1698, 0.1363, 0.3287], + device='cuda:0'), in_proj_covar=tensor([0.0492, 0.0574, 0.0599, 0.0436, 0.0594, 0.0492, 0.0645, 0.0494], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 20:23:38,867 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74026.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:23:42,023 INFO [train.py:903] (0/4) Epoch 11, batch 5750, loss[loss=0.2306, simple_loss=0.3112, pruned_loss=0.07494, over 19592.00 frames. ], tot_loss[loss=0.2347, simple_loss=0.3083, pruned_loss=0.08059, over 3821377.56 frames. ], batch size: 61, lr: 7.46e-03, grad_scale: 8.0 +2023-04-01 20:23:43,286 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 20:23:51,324 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 20:23:56,774 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 20:24:11,299 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74051.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:24:30,647 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74067.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:24:36,267 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.134e+02 5.413e+02 6.547e+02 7.617e+02 1.298e+03, threshold=1.309e+03, percent-clipped=0.0 +2023-04-01 20:24:45,221 INFO [train.py:903] (0/4) Epoch 11, batch 5800, loss[loss=0.2374, simple_loss=0.3026, pruned_loss=0.08615, over 19336.00 frames. ], tot_loss[loss=0.2353, simple_loss=0.3087, pruned_loss=0.08094, over 3813397.04 frames. ], batch size: 47, lr: 7.46e-03, grad_scale: 8.0 +2023-04-01 20:24:50,554 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74082.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:25:04,170 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=74094.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:25:46,671 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2602, 1.3251, 1.2582, 1.0677, 1.1133, 1.1341, 0.0199, 0.4067], + device='cuda:0'), covar=tensor([0.0436, 0.0416, 0.0271, 0.0336, 0.0798, 0.0379, 0.0771, 0.0720], + device='cuda:0'), in_proj_covar=tensor([0.0331, 0.0325, 0.0327, 0.0346, 0.0421, 0.0344, 0.0305, 0.0318], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 20:25:48,487 INFO [train.py:903] (0/4) Epoch 11, batch 5850, loss[loss=0.1963, simple_loss=0.27, pruned_loss=0.06133, over 19791.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.3084, pruned_loss=0.08099, over 3826823.06 frames. ], batch size: 48, lr: 7.46e-03, grad_scale: 8.0 +2023-04-01 20:25:52,588 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0 +2023-04-01 20:26:04,395 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-04-01 20:26:41,999 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.778e+02 5.734e+02 7.251e+02 9.100e+02 2.751e+03, threshold=1.450e+03, percent-clipped=7.0 +2023-04-01 20:26:51,303 INFO [train.py:903] (0/4) Epoch 11, batch 5900, loss[loss=0.2281, simple_loss=0.3065, pruned_loss=0.07485, over 19751.00 frames. ], tot_loss[loss=0.2354, simple_loss=0.3082, pruned_loss=0.08131, over 3833387.96 frames. ], batch size: 54, lr: 7.45e-03, grad_scale: 8.0 +2023-04-01 20:26:53,554 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 20:27:15,038 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 20:27:21,460 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4096, 1.2391, 1.6129, 1.3024, 2.8151, 3.8336, 3.6108, 4.0352], + device='cuda:0'), covar=tensor([0.1386, 0.3318, 0.3013, 0.2028, 0.0444, 0.0140, 0.0156, 0.0156], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0295, 0.0324, 0.0250, 0.0214, 0.0158, 0.0207, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 20:27:30,970 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74209.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:27:55,157 INFO [train.py:903] (0/4) Epoch 11, batch 5950, loss[loss=0.1868, simple_loss=0.2772, pruned_loss=0.04821, over 19545.00 frames. ], tot_loss[loss=0.2339, simple_loss=0.3071, pruned_loss=0.08033, over 3823294.72 frames. ], batch size: 56, lr: 7.45e-03, grad_scale: 8.0 +2023-04-01 20:28:33,399 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74258.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:28:49,683 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.166e+02 4.883e+02 6.160e+02 7.607e+02 1.521e+03, threshold=1.232e+03, percent-clipped=1.0 +2023-04-01 20:28:59,179 INFO [train.py:903] (0/4) Epoch 11, batch 6000, loss[loss=0.187, simple_loss=0.2637, pruned_loss=0.05514, over 19732.00 frames. ], tot_loss[loss=0.234, simple_loss=0.3072, pruned_loss=0.0804, over 3805595.23 frames. ], batch size: 46, lr: 7.45e-03, grad_scale: 8.0 +2023-04-01 20:28:59,180 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 20:29:11,916 INFO [train.py:937] (0/4) Epoch 11, validation: loss=0.1778, simple_loss=0.2787, pruned_loss=0.03847, over 944034.00 frames. +2023-04-01 20:29:11,918 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 20:30:09,775 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74323.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:30:18,373 INFO [train.py:903] (0/4) Epoch 11, batch 6050, loss[loss=0.2261, simple_loss=0.3097, pruned_loss=0.07125, over 19672.00 frames. ], tot_loss[loss=0.2328, simple_loss=0.3061, pruned_loss=0.07973, over 3815810.75 frames. ], batch size: 59, lr: 7.45e-03, grad_scale: 8.0 +2023-04-01 20:30:28,818 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9175, 4.3319, 4.6687, 4.6326, 1.8660, 4.3517, 3.7187, 4.3356], + device='cuda:0'), covar=tensor([0.1446, 0.0810, 0.0496, 0.0595, 0.4831, 0.0598, 0.0615, 0.1007], + device='cuda:0'), in_proj_covar=tensor([0.0672, 0.0600, 0.0800, 0.0682, 0.0725, 0.0557, 0.0485, 0.0732], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 20:30:42,877 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74348.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:31:12,596 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.290e+02 5.870e+02 7.295e+02 8.913e+02 1.728e+03, threshold=1.459e+03, percent-clipped=8.0 +2023-04-01 20:31:21,970 INFO [train.py:903] (0/4) Epoch 11, batch 6100, loss[loss=0.2423, simple_loss=0.318, pruned_loss=0.08332, over 18299.00 frames. ], tot_loss[loss=0.2324, simple_loss=0.306, pruned_loss=0.07938, over 3822682.20 frames. ], batch size: 83, lr: 7.44e-03, grad_scale: 8.0 +2023-04-01 20:32:22,429 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=74426.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:32:25,923 INFO [train.py:903] (0/4) Epoch 11, batch 6150, loss[loss=0.2426, simple_loss=0.3229, pruned_loss=0.0812, over 19676.00 frames. ], tot_loss[loss=0.2323, simple_loss=0.3063, pruned_loss=0.07922, over 3822878.06 frames. ], batch size: 58, lr: 7.44e-03, grad_scale: 8.0 +2023-04-01 20:32:37,681 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8699, 2.0162, 2.1132, 2.7042, 1.8855, 2.5536, 2.2983, 1.9887], + device='cuda:0'), covar=tensor([0.3493, 0.2889, 0.1431, 0.1848, 0.3459, 0.1465, 0.3384, 0.2573], + device='cuda:0'), in_proj_covar=tensor([0.0790, 0.0811, 0.0644, 0.0894, 0.0779, 0.0699, 0.0783, 0.0708], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 20:32:54,718 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 20:33:12,550 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74465.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:33:20,926 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.809e+02 5.576e+02 6.382e+02 8.608e+02 2.367e+03, threshold=1.276e+03, percent-clipped=4.0 +2023-04-01 20:33:29,309 INFO [train.py:903] (0/4) Epoch 11, batch 6200, loss[loss=0.2161, simple_loss=0.3049, pruned_loss=0.06362, over 19667.00 frames. ], tot_loss[loss=0.2331, simple_loss=0.307, pruned_loss=0.07957, over 3823602.25 frames. ], batch size: 58, lr: 7.44e-03, grad_scale: 8.0 +2023-04-01 20:33:43,940 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74490.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:34:31,916 INFO [train.py:903] (0/4) Epoch 11, batch 6250, loss[loss=0.2729, simple_loss=0.344, pruned_loss=0.1009, over 19705.00 frames. ], tot_loss[loss=0.2341, simple_loss=0.3075, pruned_loss=0.08038, over 3798651.14 frames. ], batch size: 59, lr: 7.44e-03, grad_scale: 8.0 +2023-04-01 20:34:48,583 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74541.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:35:02,959 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 20:35:30,743 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.915e+02 5.334e+02 6.699e+02 8.918e+02 1.691e+03, threshold=1.340e+03, percent-clipped=6.0 +2023-04-01 20:35:38,660 INFO [train.py:903] (0/4) Epoch 11, batch 6300, loss[loss=0.2575, simple_loss=0.3116, pruned_loss=0.1017, over 19763.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.3078, pruned_loss=0.08067, over 3792859.16 frames. ], batch size: 47, lr: 7.43e-03, grad_scale: 8.0 +2023-04-01 20:36:06,782 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=74602.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:36:21,216 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74613.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:36:40,491 INFO [train.py:903] (0/4) Epoch 11, batch 6350, loss[loss=0.2164, simple_loss=0.292, pruned_loss=0.07039, over 19759.00 frames. ], tot_loss[loss=0.2342, simple_loss=0.3074, pruned_loss=0.08051, over 3805498.34 frames. ], batch size: 51, lr: 7.43e-03, grad_scale: 8.0 +2023-04-01 20:37:36,488 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.847e+02 5.622e+02 6.880e+02 9.256e+02 3.776e+03, threshold=1.376e+03, percent-clipped=5.0 +2023-04-01 20:37:44,840 INFO [train.py:903] (0/4) Epoch 11, batch 6400, loss[loss=0.2272, simple_loss=0.3101, pruned_loss=0.07216, over 19670.00 frames. ], tot_loss[loss=0.2335, simple_loss=0.3066, pruned_loss=0.08016, over 3800749.44 frames. ], batch size: 58, lr: 7.43e-03, grad_scale: 8.0 +2023-04-01 20:37:56,012 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 20:38:01,361 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9274, 1.6120, 1.5187, 1.9007, 1.7647, 1.6829, 1.5261, 1.7947], + device='cuda:0'), covar=tensor([0.0917, 0.1406, 0.1325, 0.0933, 0.1138, 0.0497, 0.1214, 0.0693], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0351, 0.0291, 0.0238, 0.0298, 0.0243, 0.0278, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:38:33,487 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=74717.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:38:46,859 INFO [train.py:903] (0/4) Epoch 11, batch 6450, loss[loss=0.2441, simple_loss=0.3186, pruned_loss=0.08477, over 19658.00 frames. ], tot_loss[loss=0.2337, simple_loss=0.3072, pruned_loss=0.08016, over 3801458.79 frames. ], batch size: 60, lr: 7.43e-03, grad_scale: 8.0 +2023-04-01 20:39:25,226 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6748, 1.4503, 1.4478, 1.7100, 1.6589, 1.5002, 1.4365, 1.6677], + device='cuda:0'), covar=tensor([0.0809, 0.1218, 0.1143, 0.0710, 0.0941, 0.0486, 0.1060, 0.0601], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0351, 0.0293, 0.0239, 0.0298, 0.0244, 0.0279, 0.0240], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:39:28,506 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 20:39:44,918 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.542e+02 5.590e+02 7.165e+02 9.175e+02 2.194e+03, threshold=1.433e+03, percent-clipped=7.0 +2023-04-01 20:39:53,337 INFO [train.py:903] (0/4) Epoch 11, batch 6500, loss[loss=0.2479, simple_loss=0.3184, pruned_loss=0.08867, over 19280.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.3083, pruned_loss=0.08063, over 3794632.89 frames. ], batch size: 66, lr: 7.42e-03, grad_scale: 8.0 +2023-04-01 20:39:54,549 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 20:40:16,064 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74797.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:40:48,466 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74822.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:40:56,365 INFO [train.py:903] (0/4) Epoch 11, batch 6550, loss[loss=0.2228, simple_loss=0.299, pruned_loss=0.07334, over 19603.00 frames. ], tot_loss[loss=0.2338, simple_loss=0.3072, pruned_loss=0.08022, over 3798366.51 frames. ], batch size: 50, lr: 7.42e-03, grad_scale: 8.0 +2023-04-01 20:40:56,799 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1896, 2.0994, 1.8171, 1.6761, 1.5366, 1.6727, 0.3788, 1.0303], + device='cuda:0'), covar=tensor([0.0483, 0.0422, 0.0310, 0.0512, 0.0893, 0.0549, 0.0884, 0.0766], + device='cuda:0'), in_proj_covar=tensor([0.0341, 0.0331, 0.0333, 0.0354, 0.0429, 0.0353, 0.0312, 0.0328], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 20:41:52,115 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.386e+02 5.036e+02 6.201e+02 8.020e+02 1.527e+03, threshold=1.240e+03, percent-clipped=1.0 +2023-04-01 20:41:59,118 INFO [train.py:903] (0/4) Epoch 11, batch 6600, loss[loss=0.1826, simple_loss=0.2573, pruned_loss=0.05389, over 19796.00 frames. ], tot_loss[loss=0.2327, simple_loss=0.3063, pruned_loss=0.07953, over 3817225.14 frames. ], batch size: 46, lr: 7.42e-03, grad_scale: 8.0 +2023-04-01 20:42:42,899 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-01 20:43:01,202 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74927.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:43:03,149 INFO [train.py:903] (0/4) Epoch 11, batch 6650, loss[loss=0.2254, simple_loss=0.3051, pruned_loss=0.07286, over 19752.00 frames. ], tot_loss[loss=0.2318, simple_loss=0.3056, pruned_loss=0.07905, over 3815154.78 frames. ], batch size: 63, lr: 7.42e-03, grad_scale: 8.0 +2023-04-01 20:43:40,234 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=74957.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:44:00,891 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.349e+02 5.506e+02 7.046e+02 9.043e+02 1.602e+03, threshold=1.409e+03, percent-clipped=2.0 +2023-04-01 20:44:01,356 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=74973.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:44:07,814 INFO [train.py:903] (0/4) Epoch 11, batch 6700, loss[loss=0.2597, simple_loss=0.3298, pruned_loss=0.09479, over 19588.00 frames. ], tot_loss[loss=0.2333, simple_loss=0.3067, pruned_loss=0.07998, over 3812536.23 frames. ], batch size: 61, lr: 7.41e-03, grad_scale: 8.0 +2023-04-01 20:44:12,414 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=74982.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:44:29,466 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6054, 1.4349, 1.4079, 1.9891, 1.6316, 1.9015, 2.0169, 1.7371], + device='cuda:0'), covar=tensor([0.0780, 0.0925, 0.0978, 0.0713, 0.0788, 0.0644, 0.0737, 0.0596], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0227, 0.0224, 0.0247, 0.0237, 0.0214, 0.0197, 0.0200], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 20:44:31,726 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=74998.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:44:46,244 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7005, 4.3231, 2.9234, 3.7408, 1.3863, 4.0669, 4.0851, 4.1567], + device='cuda:0'), covar=tensor([0.0568, 0.0921, 0.1809, 0.0770, 0.3357, 0.0709, 0.0701, 0.1055], + device='cuda:0'), in_proj_covar=tensor([0.0437, 0.0369, 0.0438, 0.0317, 0.0379, 0.0369, 0.0357, 0.0393], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:45:06,539 INFO [train.py:903] (0/4) Epoch 11, batch 6750, loss[loss=0.246, simple_loss=0.3105, pruned_loss=0.09069, over 19844.00 frames. ], tot_loss[loss=0.2338, simple_loss=0.3068, pruned_loss=0.0804, over 3808794.21 frames. ], batch size: 52, lr: 7.41e-03, grad_scale: 8.0 +2023-04-01 20:45:55,367 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75072.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:45:56,170 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.565e+02 5.649e+02 7.002e+02 8.709e+02 1.691e+03, threshold=1.400e+03, percent-clipped=2.0 +2023-04-01 20:46:04,204 INFO [train.py:903] (0/4) Epoch 11, batch 6800, loss[loss=0.2081, simple_loss=0.2848, pruned_loss=0.06566, over 19569.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.3065, pruned_loss=0.07998, over 3813694.76 frames. ], batch size: 52, lr: 7.41e-03, grad_scale: 8.0 +2023-04-01 20:46:36,713 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-11.pt +2023-04-01 20:46:52,796 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 20:46:53,242 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 20:46:56,121 INFO [train.py:903] (0/4) Epoch 12, batch 0, loss[loss=0.2534, simple_loss=0.3278, pruned_loss=0.08946, over 19773.00 frames. ], tot_loss[loss=0.2534, simple_loss=0.3278, pruned_loss=0.08946, over 19773.00 frames. ], batch size: 56, lr: 7.10e-03, grad_scale: 8.0 +2023-04-01 20:46:56,122 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 20:47:04,986 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8579, 3.4850, 2.6127, 3.3424, 0.8746, 3.2446, 3.2848, 3.4875], + device='cuda:0'), covar=tensor([0.0722, 0.0833, 0.1925, 0.0855, 0.4086, 0.1101, 0.0807, 0.0988], + device='cuda:0'), in_proj_covar=tensor([0.0430, 0.0361, 0.0432, 0.0312, 0.0374, 0.0364, 0.0352, 0.0385], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003], + device='cuda:0') +2023-04-01 20:47:08,131 INFO [train.py:937] (0/4) Epoch 12, validation: loss=0.1777, simple_loss=0.2788, pruned_loss=0.03825, over 944034.00 frames. +2023-04-01 20:47:08,132 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 20:47:20,785 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 20:48:11,523 INFO [train.py:903] (0/4) Epoch 12, batch 50, loss[loss=0.2351, simple_loss=0.3145, pruned_loss=0.07789, over 19731.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.3072, pruned_loss=0.07964, over 853262.88 frames. ], batch size: 63, lr: 7.09e-03, grad_scale: 8.0 +2023-04-01 20:48:20,830 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3016, 1.9703, 2.0751, 2.7843, 2.1061, 2.4987, 2.3448, 2.6629], + device='cuda:0'), covar=tensor([0.0684, 0.0840, 0.0832, 0.0829, 0.0899, 0.0644, 0.0856, 0.0515], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0225, 0.0222, 0.0246, 0.0235, 0.0212, 0.0195, 0.0200], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 20:48:29,484 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.226e+02 5.251e+02 6.823e+02 1.011e+03 3.055e+03, threshold=1.365e+03, percent-clipped=9.0 +2023-04-01 20:48:41,956 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 20:49:13,865 INFO [train.py:903] (0/4) Epoch 12, batch 100, loss[loss=0.2268, simple_loss=0.3105, pruned_loss=0.07154, over 19348.00 frames. ], tot_loss[loss=0.2362, simple_loss=0.3098, pruned_loss=0.08133, over 1516561.91 frames. ], batch size: 70, lr: 7.09e-03, grad_scale: 8.0 +2023-04-01 20:49:22,080 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 20:50:02,889 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75246.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:50:16,779 INFO [train.py:903] (0/4) Epoch 12, batch 150, loss[loss=0.2246, simple_loss=0.2859, pruned_loss=0.08161, over 18185.00 frames. ], tot_loss[loss=0.2357, simple_loss=0.3084, pruned_loss=0.08148, over 2023507.05 frames. ], batch size: 40, lr: 7.09e-03, grad_scale: 8.0 +2023-04-01 20:50:34,009 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=75271.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:50:36,118 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.416e+02 5.206e+02 6.373e+02 8.343e+02 1.576e+03, threshold=1.275e+03, percent-clipped=5.0 +2023-04-01 20:50:56,177 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=75288.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:51:15,331 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 20:51:18,701 INFO [train.py:903] (0/4) Epoch 12, batch 200, loss[loss=0.2299, simple_loss=0.304, pruned_loss=0.07792, over 19530.00 frames. ], tot_loss[loss=0.2351, simple_loss=0.308, pruned_loss=0.08109, over 2427130.32 frames. ], batch size: 54, lr: 7.09e-03, grad_scale: 8.0 +2023-04-01 20:51:42,606 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=75326.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:51:45,311 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75328.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:51:46,567 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9046, 2.0059, 2.1842, 2.7736, 1.8678, 2.6106, 2.4650, 1.9854], + device='cuda:0'), covar=tensor([0.3713, 0.3183, 0.1418, 0.1829, 0.3550, 0.1498, 0.3507, 0.2751], + device='cuda:0'), in_proj_covar=tensor([0.0789, 0.0808, 0.0642, 0.0891, 0.0775, 0.0703, 0.0777, 0.0703], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 20:52:16,202 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=75353.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:52:22,155 INFO [train.py:903] (0/4) Epoch 12, batch 250, loss[loss=0.1948, simple_loss=0.2768, pruned_loss=0.05644, over 19483.00 frames. ], tot_loss[loss=0.236, simple_loss=0.3087, pruned_loss=0.08163, over 2727327.64 frames. ], batch size: 49, lr: 7.08e-03, grad_scale: 8.0 +2023-04-01 20:52:41,710 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.835e+02 5.714e+02 6.835e+02 8.470e+02 1.829e+03, threshold=1.367e+03, percent-clipped=2.0 +2023-04-01 20:52:56,682 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75386.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:53:25,598 INFO [train.py:903] (0/4) Epoch 12, batch 300, loss[loss=0.2322, simple_loss=0.3139, pruned_loss=0.07526, over 19612.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.3073, pruned_loss=0.08098, over 2964912.88 frames. ], batch size: 57, lr: 7.08e-03, grad_scale: 8.0 +2023-04-01 20:53:34,062 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4114, 1.4114, 1.5294, 1.6787, 2.9635, 1.0148, 2.2004, 3.2590], + device='cuda:0'), covar=tensor([0.0414, 0.2393, 0.2449, 0.1456, 0.0671, 0.2366, 0.1161, 0.0298], + device='cuda:0'), in_proj_covar=tensor([0.0357, 0.0340, 0.0352, 0.0320, 0.0346, 0.0332, 0.0336, 0.0356], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:54:07,330 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75441.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:54:28,458 INFO [train.py:903] (0/4) Epoch 12, batch 350, loss[loss=0.3003, simple_loss=0.3553, pruned_loss=0.1227, over 13814.00 frames. ], tot_loss[loss=0.2355, simple_loss=0.3079, pruned_loss=0.08159, over 3157645.60 frames. ], batch size: 136, lr: 7.08e-03, grad_scale: 8.0 +2023-04-01 20:54:31,961 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 20:54:45,903 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.322e+02 5.463e+02 6.814e+02 8.627e+02 1.955e+03, threshold=1.363e+03, percent-clipped=3.0 +2023-04-01 20:55:24,622 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-04-01 20:55:30,815 INFO [train.py:903] (0/4) Epoch 12, batch 400, loss[loss=0.2277, simple_loss=0.2933, pruned_loss=0.08111, over 19408.00 frames. ], tot_loss[loss=0.2347, simple_loss=0.3076, pruned_loss=0.08093, over 3321172.99 frames. ], batch size: 48, lr: 7.08e-03, grad_scale: 8.0 +2023-04-01 20:56:16,856 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2614, 1.3775, 1.6648, 1.4652, 2.4146, 2.0984, 2.5006, 0.9151], + device='cuda:0'), covar=tensor([0.2147, 0.3618, 0.2066, 0.1648, 0.1330, 0.1762, 0.1338, 0.3689], + device='cuda:0'), in_proj_covar=tensor([0.0492, 0.0577, 0.0604, 0.0438, 0.0598, 0.0493, 0.0646, 0.0492], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 20:56:32,362 INFO [train.py:903] (0/4) Epoch 12, batch 450, loss[loss=0.2708, simple_loss=0.3396, pruned_loss=0.101, over 19602.00 frames. ], tot_loss[loss=0.2343, simple_loss=0.3072, pruned_loss=0.08071, over 3449664.64 frames. ], batch size: 61, lr: 7.07e-03, grad_scale: 8.0 +2023-04-01 20:56:51,536 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.480e+02 5.172e+02 6.423e+02 7.976e+02 2.291e+03, threshold=1.285e+03, percent-clipped=4.0 +2023-04-01 20:57:09,457 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 20:57:10,696 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 20:57:13,220 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=75590.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:57:17,020 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0613, 4.4031, 4.7356, 4.7140, 1.4675, 4.3887, 3.7960, 4.4006], + device='cuda:0'), covar=tensor([0.1263, 0.0788, 0.0564, 0.0554, 0.5654, 0.0682, 0.0630, 0.1071], + device='cuda:0'), in_proj_covar=tensor([0.0677, 0.0603, 0.0804, 0.0688, 0.0731, 0.0559, 0.0492, 0.0741], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 20:57:36,542 INFO [train.py:903] (0/4) Epoch 12, batch 500, loss[loss=0.2689, simple_loss=0.3383, pruned_loss=0.09974, over 19389.00 frames. ], tot_loss[loss=0.2352, simple_loss=0.3081, pruned_loss=0.08112, over 3525489.18 frames. ], batch size: 66, lr: 7.07e-03, grad_scale: 8.0 +2023-04-01 20:58:06,137 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=75632.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:58:17,921 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75642.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:58:38,555 INFO [train.py:903] (0/4) Epoch 12, batch 550, loss[loss=0.2625, simple_loss=0.322, pruned_loss=0.1015, over 17271.00 frames. ], tot_loss[loss=0.2347, simple_loss=0.3079, pruned_loss=0.0808, over 3587399.09 frames. ], batch size: 101, lr: 7.07e-03, grad_scale: 8.0 +2023-04-01 20:58:50,382 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=75667.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:58:56,742 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.239e+02 5.816e+02 6.939e+02 9.327e+02 2.224e+03, threshold=1.388e+03, percent-clipped=13.0 +2023-04-01 20:59:09,128 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-01 20:59:24,531 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9531, 1.5573, 1.7347, 1.9920, 4.3750, 1.0996, 2.4018, 4.6619], + device='cuda:0'), covar=tensor([0.0346, 0.2620, 0.2666, 0.1723, 0.0775, 0.2707, 0.1400, 0.0221], + device='cuda:0'), in_proj_covar=tensor([0.0357, 0.0340, 0.0351, 0.0319, 0.0347, 0.0330, 0.0333, 0.0354], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 20:59:26,865 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75697.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:59:38,276 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75705.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 20:59:41,416 INFO [train.py:903] (0/4) Epoch 12, batch 600, loss[loss=0.2567, simple_loss=0.3334, pruned_loss=0.08999, over 19548.00 frames. ], tot_loss[loss=0.2344, simple_loss=0.3078, pruned_loss=0.08047, over 3645522.00 frames. ], batch size: 61, lr: 7.07e-03, grad_scale: 8.0 +2023-04-01 20:59:57,482 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=75722.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:00:22,873 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 21:00:30,155 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=75747.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:00:42,500 INFO [train.py:903] (0/4) Epoch 12, batch 650, loss[loss=0.1955, simple_loss=0.2696, pruned_loss=0.0607, over 19764.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.3082, pruned_loss=0.08056, over 3688637.76 frames. ], batch size: 48, lr: 7.07e-03, grad_scale: 8.0 +2023-04-01 21:01:01,243 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.249e+02 5.224e+02 6.354e+02 7.929e+02 1.382e+03, threshold=1.271e+03, percent-clipped=0.0 +2023-04-01 21:01:22,978 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5380, 2.3891, 1.7888, 1.5348, 2.2227, 1.2770, 1.2578, 1.9652], + device='cuda:0'), covar=tensor([0.0906, 0.0598, 0.0840, 0.0673, 0.0380, 0.1000, 0.0690, 0.0392], + device='cuda:0'), in_proj_covar=tensor([0.0294, 0.0304, 0.0329, 0.0249, 0.0237, 0.0320, 0.0292, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:01:24,545 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.73 vs. limit=5.0 +2023-04-01 21:01:29,683 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4017, 1.2674, 1.7825, 1.1052, 2.5011, 3.3683, 3.0789, 3.4964], + device='cuda:0'), covar=tensor([0.1356, 0.3389, 0.2783, 0.2185, 0.0488, 0.0154, 0.0202, 0.0211], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0293, 0.0323, 0.0250, 0.0214, 0.0156, 0.0206, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 21:01:45,574 INFO [train.py:903] (0/4) Epoch 12, batch 700, loss[loss=0.2101, simple_loss=0.2914, pruned_loss=0.06442, over 19666.00 frames. ], tot_loss[loss=0.2336, simple_loss=0.3073, pruned_loss=0.07997, over 3717818.73 frames. ], batch size: 55, lr: 7.06e-03, grad_scale: 8.0 +2023-04-01 21:02:04,537 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.68 vs. limit=5.0 +2023-04-01 21:02:46,468 INFO [train.py:903] (0/4) Epoch 12, batch 750, loss[loss=0.2312, simple_loss=0.3027, pruned_loss=0.07988, over 19497.00 frames. ], tot_loss[loss=0.2335, simple_loss=0.3072, pruned_loss=0.07994, over 3745926.15 frames. ], batch size: 49, lr: 7.06e-03, grad_scale: 8.0 +2023-04-01 21:03:02,335 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1836, 2.2085, 2.3245, 3.3463, 2.2198, 3.0667, 2.7779, 2.2420], + device='cuda:0'), covar=tensor([0.3845, 0.3378, 0.1456, 0.1895, 0.3805, 0.1520, 0.3450, 0.2810], + device='cuda:0'), in_proj_covar=tensor([0.0792, 0.0813, 0.0647, 0.0893, 0.0781, 0.0704, 0.0781, 0.0711], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 21:03:05,206 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.592e+02 5.547e+02 6.814e+02 8.571e+02 2.504e+03, threshold=1.363e+03, percent-clipped=8.0 +2023-04-01 21:03:40,164 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-04-01 21:03:49,650 INFO [train.py:903] (0/4) Epoch 12, batch 800, loss[loss=0.2065, simple_loss=0.2768, pruned_loss=0.0681, over 19373.00 frames. ], tot_loss[loss=0.2336, simple_loss=0.3071, pruned_loss=0.08006, over 3758665.71 frames. ], batch size: 47, lr: 7.06e-03, grad_scale: 8.0 +2023-04-01 21:04:07,048 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 21:04:50,832 INFO [train.py:903] (0/4) Epoch 12, batch 850, loss[loss=0.2152, simple_loss=0.3014, pruned_loss=0.06448, over 19572.00 frames. ], tot_loss[loss=0.2331, simple_loss=0.307, pruned_loss=0.07958, over 3785229.65 frames. ], batch size: 61, lr: 7.06e-03, grad_scale: 8.0 +2023-04-01 21:04:54,842 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=75961.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:05:10,029 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.002e+02 5.208e+02 6.361e+02 7.722e+02 1.579e+03, threshold=1.272e+03, percent-clipped=2.0 +2023-04-01 21:05:25,762 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=75986.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:05:29,792 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5526, 1.4183, 1.8187, 1.6491, 2.9911, 4.6127, 4.5098, 4.9003], + device='cuda:0'), covar=tensor([0.1434, 0.3320, 0.3040, 0.1913, 0.0496, 0.0137, 0.0146, 0.0128], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0294, 0.0324, 0.0251, 0.0214, 0.0157, 0.0206, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 21:05:43,537 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-76000.pt +2023-04-01 21:05:46,886 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 21:05:48,501 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=76003.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:05:53,895 INFO [train.py:903] (0/4) Epoch 12, batch 900, loss[loss=0.2618, simple_loss=0.3318, pruned_loss=0.09587, over 17173.00 frames. ], tot_loss[loss=0.232, simple_loss=0.3064, pruned_loss=0.07887, over 3807209.94 frames. ], batch size: 101, lr: 7.05e-03, grad_scale: 8.0 +2023-04-01 21:06:13,717 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4571, 1.2648, 1.3068, 1.7167, 1.3945, 1.6982, 1.6925, 1.5566], + device='cuda:0'), covar=tensor([0.0792, 0.0984, 0.1028, 0.0713, 0.0880, 0.0694, 0.0875, 0.0658], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0224, 0.0222, 0.0244, 0.0235, 0.0212, 0.0195, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 21:06:20,498 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=76028.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:06:57,921 INFO [train.py:903] (0/4) Epoch 12, batch 950, loss[loss=0.247, simple_loss=0.3191, pruned_loss=0.08744, over 18208.00 frames. ], tot_loss[loss=0.2319, simple_loss=0.3058, pruned_loss=0.07899, over 3807544.93 frames. ], batch size: 83, lr: 7.05e-03, grad_scale: 8.0 +2023-04-01 21:07:02,628 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 21:07:17,511 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.890e+02 5.102e+02 6.493e+02 8.387e+02 1.985e+03, threshold=1.299e+03, percent-clipped=2.0 +2023-04-01 21:07:59,842 INFO [train.py:903] (0/4) Epoch 12, batch 1000, loss[loss=0.3108, simple_loss=0.3661, pruned_loss=0.1278, over 17526.00 frames. ], tot_loss[loss=0.2315, simple_loss=0.3055, pruned_loss=0.07878, over 3811921.83 frames. ], batch size: 101, lr: 7.05e-03, grad_scale: 8.0 +2023-04-01 21:08:42,442 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1900, 1.9459, 1.5030, 1.3139, 1.8214, 1.1288, 1.0504, 1.7174], + device='cuda:0'), covar=tensor([0.0843, 0.0698, 0.0948, 0.0620, 0.0428, 0.1035, 0.0680, 0.0374], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0303, 0.0326, 0.0247, 0.0233, 0.0314, 0.0289, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:08:54,949 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 21:09:02,680 INFO [train.py:903] (0/4) Epoch 12, batch 1050, loss[loss=0.2281, simple_loss=0.2962, pruned_loss=0.08002, over 19399.00 frames. ], tot_loss[loss=0.2321, simple_loss=0.3057, pruned_loss=0.07921, over 3803183.64 frames. ], batch size: 48, lr: 7.05e-03, grad_scale: 8.0 +2023-04-01 21:09:20,679 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.538e+02 5.178e+02 6.428e+02 8.624e+02 1.751e+03, threshold=1.286e+03, percent-clipped=4.0 +2023-04-01 21:09:35,979 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 21:10:05,080 INFO [train.py:903] (0/4) Epoch 12, batch 1100, loss[loss=0.2563, simple_loss=0.3297, pruned_loss=0.0915, over 18715.00 frames. ], tot_loss[loss=0.2318, simple_loss=0.3053, pruned_loss=0.07911, over 3816590.01 frames. ], batch size: 74, lr: 7.04e-03, grad_scale: 8.0 +2023-04-01 21:10:12,398 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9130, 2.0931, 2.3546, 2.8294, 2.7318, 2.3919, 2.2425, 2.8056], + device='cuda:0'), covar=tensor([0.0691, 0.1593, 0.1079, 0.0821, 0.1061, 0.0426, 0.1117, 0.0495], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0352, 0.0292, 0.0241, 0.0296, 0.0243, 0.0280, 0.0238], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:11:08,891 INFO [train.py:903] (0/4) Epoch 12, batch 1150, loss[loss=0.2029, simple_loss=0.2777, pruned_loss=0.064, over 19420.00 frames. ], tot_loss[loss=0.2313, simple_loss=0.3048, pruned_loss=0.0789, over 3820217.27 frames. ], batch size: 48, lr: 7.04e-03, grad_scale: 8.0 +2023-04-01 21:11:27,481 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.258e+02 5.432e+02 6.664e+02 8.568e+02 1.731e+03, threshold=1.333e+03, percent-clipped=3.0 +2023-04-01 21:11:44,823 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76287.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 21:12:10,722 INFO [train.py:903] (0/4) Epoch 12, batch 1200, loss[loss=0.199, simple_loss=0.2739, pruned_loss=0.06202, over 19478.00 frames. ], tot_loss[loss=0.2308, simple_loss=0.3047, pruned_loss=0.07847, over 3820850.07 frames. ], batch size: 49, lr: 7.04e-03, grad_scale: 8.0 +2023-04-01 21:12:15,520 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76312.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:12:45,736 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 21:12:52,921 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76341.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 21:13:14,059 INFO [train.py:903] (0/4) Epoch 12, batch 1250, loss[loss=0.1889, simple_loss=0.2585, pruned_loss=0.05961, over 18605.00 frames. ], tot_loss[loss=0.2308, simple_loss=0.3046, pruned_loss=0.07849, over 3816983.39 frames. ], batch size: 41, lr: 7.04e-03, grad_scale: 8.0 +2023-04-01 21:13:31,221 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.467e+02 5.242e+02 6.279e+02 7.669e+02 1.575e+03, threshold=1.256e+03, percent-clipped=2.0 +2023-04-01 21:13:40,368 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6834, 2.2834, 2.1602, 2.7642, 2.4569, 2.2127, 2.2205, 2.4807], + device='cuda:0'), covar=tensor([0.0759, 0.1518, 0.1226, 0.0818, 0.1132, 0.0509, 0.1083, 0.0627], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0353, 0.0292, 0.0241, 0.0298, 0.0245, 0.0280, 0.0241], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:14:15,607 INFO [train.py:903] (0/4) Epoch 12, batch 1300, loss[loss=0.2123, simple_loss=0.2803, pruned_loss=0.07216, over 19422.00 frames. ], tot_loss[loss=0.2309, simple_loss=0.3051, pruned_loss=0.07837, over 3817503.76 frames. ], batch size: 48, lr: 7.04e-03, grad_scale: 8.0 +2023-04-01 21:15:18,797 INFO [train.py:903] (0/4) Epoch 12, batch 1350, loss[loss=0.2163, simple_loss=0.2973, pruned_loss=0.0676, over 19673.00 frames. ], tot_loss[loss=0.2311, simple_loss=0.305, pruned_loss=0.07859, over 3810320.27 frames. ], batch size: 60, lr: 7.03e-03, grad_scale: 8.0 +2023-04-01 21:15:37,069 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.121e+02 5.247e+02 6.486e+02 7.910e+02 1.390e+03, threshold=1.297e+03, percent-clipped=4.0 +2023-04-01 21:16:20,438 INFO [train.py:903] (0/4) Epoch 12, batch 1400, loss[loss=0.2497, simple_loss=0.322, pruned_loss=0.08866, over 19588.00 frames. ], tot_loss[loss=0.2295, simple_loss=0.3038, pruned_loss=0.07757, over 3812835.18 frames. ], batch size: 52, lr: 7.03e-03, grad_scale: 16.0 +2023-04-01 21:16:44,277 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-04-01 21:17:23,464 INFO [train.py:903] (0/4) Epoch 12, batch 1450, loss[loss=0.2545, simple_loss=0.3273, pruned_loss=0.09081, over 19301.00 frames. ], tot_loss[loss=0.2293, simple_loss=0.304, pruned_loss=0.07726, over 3824948.15 frames. ], batch size: 66, lr: 7.03e-03, grad_scale: 16.0 +2023-04-01 21:17:25,871 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 21:17:26,560 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-01 21:17:40,980 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.252e+02 4.935e+02 6.510e+02 8.250e+02 1.774e+03, threshold=1.302e+03, percent-clipped=3.0 +2023-04-01 21:18:24,877 INFO [train.py:903] (0/4) Epoch 12, batch 1500, loss[loss=0.2035, simple_loss=0.2803, pruned_loss=0.0633, over 19735.00 frames. ], tot_loss[loss=0.2306, simple_loss=0.3047, pruned_loss=0.07822, over 3818298.44 frames. ], batch size: 51, lr: 7.03e-03, grad_scale: 8.0 +2023-04-01 21:18:53,773 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=76631.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 21:19:24,220 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=76656.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:19:27,181 INFO [train.py:903] (0/4) Epoch 12, batch 1550, loss[loss=0.2822, simple_loss=0.3427, pruned_loss=0.1108, over 19319.00 frames. ], tot_loss[loss=0.2309, simple_loss=0.3051, pruned_loss=0.07837, over 3829667.02 frames. ], batch size: 66, lr: 7.02e-03, grad_scale: 8.0 +2023-04-01 21:19:46,336 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.911e+02 5.709e+02 6.905e+02 9.210e+02 1.884e+03, threshold=1.381e+03, percent-clipped=4.0 +2023-04-01 21:20:01,202 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=76685.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 21:20:19,370 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-04-01 21:20:23,675 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0 +2023-04-01 21:20:29,928 INFO [train.py:903] (0/4) Epoch 12, batch 1600, loss[loss=0.2679, simple_loss=0.3357, pruned_loss=0.1001, over 18330.00 frames. ], tot_loss[loss=0.2302, simple_loss=0.3049, pruned_loss=0.07777, over 3833174.27 frames. ], batch size: 83, lr: 7.02e-03, grad_scale: 8.0 +2023-04-01 21:20:53,997 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 21:21:16,444 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=76746.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 21:21:23,391 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4284, 2.3843, 1.6993, 1.4587, 2.2768, 1.2131, 1.2013, 2.0147], + device='cuda:0'), covar=tensor([0.1037, 0.0567, 0.0888, 0.0751, 0.0393, 0.1111, 0.0824, 0.0420], + device='cuda:0'), in_proj_covar=tensor([0.0289, 0.0300, 0.0321, 0.0243, 0.0231, 0.0316, 0.0289, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:21:29,664 INFO [train.py:903] (0/4) Epoch 12, batch 1650, loss[loss=0.2724, simple_loss=0.3424, pruned_loss=0.1012, over 18766.00 frames. ], tot_loss[loss=0.2314, simple_loss=0.3059, pruned_loss=0.07848, over 3833398.01 frames. ], batch size: 74, lr: 7.02e-03, grad_scale: 8.0 +2023-04-01 21:21:30,358 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0 +2023-04-01 21:21:46,971 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=76771.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:21:49,944 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.672e+02 5.088e+02 6.360e+02 7.707e+02 1.579e+03, threshold=1.272e+03, percent-clipped=3.0 +2023-04-01 21:22:06,337 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7300, 1.8551, 1.6824, 2.8498, 1.8857, 2.7034, 2.0081, 1.4598], + device='cuda:0'), covar=tensor([0.4385, 0.3611, 0.2251, 0.2117, 0.3760, 0.1638, 0.4978, 0.4209], + device='cuda:0'), in_proj_covar=tensor([0.0791, 0.0816, 0.0646, 0.0889, 0.0781, 0.0706, 0.0779, 0.0709], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 21:22:22,685 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=76800.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 21:22:33,406 INFO [train.py:903] (0/4) Epoch 12, batch 1700, loss[loss=0.2386, simple_loss=0.3215, pruned_loss=0.07781, over 19539.00 frames. ], tot_loss[loss=0.2315, simple_loss=0.3059, pruned_loss=0.07855, over 3835514.26 frames. ], batch size: 56, lr: 7.02e-03, grad_scale: 8.0 +2023-04-01 21:22:59,173 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5588, 1.3533, 1.3866, 1.9154, 1.5319, 1.8386, 1.9231, 1.6608], + device='cuda:0'), covar=tensor([0.0799, 0.0946, 0.0996, 0.0783, 0.0881, 0.0712, 0.0851, 0.0651], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0225, 0.0224, 0.0246, 0.0236, 0.0213, 0.0195, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 21:23:10,291 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 21:23:16,988 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3555, 2.1412, 1.9215, 1.8267, 1.5742, 1.8178, 0.6045, 1.2396], + device='cuda:0'), covar=tensor([0.0419, 0.0454, 0.0384, 0.0601, 0.0940, 0.0710, 0.0947, 0.0772], + device='cuda:0'), in_proj_covar=tensor([0.0339, 0.0329, 0.0332, 0.0356, 0.0427, 0.0354, 0.0309, 0.0322], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 21:23:33,286 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=76857.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:23:34,183 INFO [train.py:903] (0/4) Epoch 12, batch 1750, loss[loss=0.2176, simple_loss=0.2801, pruned_loss=0.07758, over 19765.00 frames. ], tot_loss[loss=0.233, simple_loss=0.307, pruned_loss=0.07951, over 3825957.59 frames. ], batch size: 46, lr: 7.01e-03, grad_scale: 8.0 +2023-04-01 21:23:36,417 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-04-01 21:23:53,640 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.317e+02 5.587e+02 7.065e+02 8.864e+02 2.096e+03, threshold=1.413e+03, percent-clipped=6.0 +2023-04-01 21:24:37,122 INFO [train.py:903] (0/4) Epoch 12, batch 1800, loss[loss=0.2511, simple_loss=0.3221, pruned_loss=0.09002, over 19670.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.3046, pruned_loss=0.07813, over 3833385.88 frames. ], batch size: 60, lr: 7.01e-03, grad_scale: 8.0 +2023-04-01 21:24:44,446 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3050, 1.9540, 1.5350, 1.4004, 1.8359, 1.1324, 1.2942, 1.7986], + device='cuda:0'), covar=tensor([0.0775, 0.0672, 0.0888, 0.0628, 0.0421, 0.1039, 0.0566, 0.0325], + device='cuda:0'), in_proj_covar=tensor([0.0287, 0.0300, 0.0320, 0.0243, 0.0231, 0.0317, 0.0289, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:25:32,721 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 21:25:38,680 INFO [train.py:903] (0/4) Epoch 12, batch 1850, loss[loss=0.202, simple_loss=0.2784, pruned_loss=0.06285, over 19738.00 frames. ], tot_loss[loss=0.2299, simple_loss=0.3041, pruned_loss=0.07784, over 3829528.37 frames. ], batch size: 51, lr: 7.01e-03, grad_scale: 8.0 +2023-04-01 21:25:45,874 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.06 vs. limit=5.0 +2023-04-01 21:25:59,472 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.186e+02 5.640e+02 7.026e+02 8.457e+02 1.689e+03, threshold=1.405e+03, percent-clipped=1.0 +2023-04-01 21:26:12,306 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 21:26:34,345 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77002.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 21:26:41,893 INFO [train.py:903] (0/4) Epoch 12, batch 1900, loss[loss=0.22, simple_loss=0.2891, pruned_loss=0.07544, over 19719.00 frames. ], tot_loss[loss=0.2318, simple_loss=0.3059, pruned_loss=0.07888, over 3832637.94 frames. ], batch size: 51, lr: 7.01e-03, grad_scale: 8.0 +2023-04-01 21:27:00,222 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 21:27:04,920 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 21:27:06,367 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77027.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 21:27:06,384 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77027.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:27:28,514 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 21:27:36,567 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77052.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:27:42,293 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77056.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 21:27:43,966 INFO [train.py:903] (0/4) Epoch 12, batch 1950, loss[loss=0.2153, simple_loss=0.2809, pruned_loss=0.07487, over 19383.00 frames. ], tot_loss[loss=0.2327, simple_loss=0.3066, pruned_loss=0.07934, over 3836868.64 frames. ], batch size: 47, lr: 7.01e-03, grad_scale: 8.0 +2023-04-01 21:28:03,223 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.700e+02 5.468e+02 7.092e+02 9.065e+02 1.810e+03, threshold=1.418e+03, percent-clipped=4.0 +2023-04-01 21:28:11,815 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77081.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 21:28:34,064 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-04-01 21:28:45,461 INFO [train.py:903] (0/4) Epoch 12, batch 2000, loss[loss=0.2919, simple_loss=0.3433, pruned_loss=0.1203, over 13330.00 frames. ], tot_loss[loss=0.2343, simple_loss=0.3079, pruned_loss=0.08031, over 3818409.36 frames. ], batch size: 136, lr: 7.00e-03, grad_scale: 8.0 +2023-04-01 21:28:48,543 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.64 vs. limit=5.0 +2023-04-01 21:28:51,499 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77113.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:29:15,575 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5838, 1.2828, 1.2083, 1.5158, 1.1418, 1.3687, 1.2140, 1.4044], + device='cuda:0'), covar=tensor([0.0932, 0.1072, 0.1400, 0.0794, 0.1112, 0.0542, 0.1263, 0.0725], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0349, 0.0292, 0.0238, 0.0294, 0.0242, 0.0278, 0.0236], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:29:43,044 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 21:29:46,528 INFO [train.py:903] (0/4) Epoch 12, batch 2050, loss[loss=0.2049, simple_loss=0.2956, pruned_loss=0.05709, over 19627.00 frames. ], tot_loss[loss=0.2346, simple_loss=0.3079, pruned_loss=0.08068, over 3816317.13 frames. ], batch size: 57, lr: 7.00e-03, grad_scale: 8.0 +2023-04-01 21:30:02,209 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 21:30:03,448 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 21:30:06,833 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.225e+02 5.837e+02 7.308e+02 9.815e+02 2.165e+03, threshold=1.462e+03, percent-clipped=5.0 +2023-04-01 21:30:26,552 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 21:30:40,584 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77201.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:30:49,626 INFO [train.py:903] (0/4) Epoch 12, batch 2100, loss[loss=0.2634, simple_loss=0.3337, pruned_loss=0.09649, over 19672.00 frames. ], tot_loss[loss=0.2335, simple_loss=0.3067, pruned_loss=0.08012, over 3815922.64 frames. ], batch size: 58, lr: 7.00e-03, grad_scale: 4.0 +2023-04-01 21:31:10,325 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-01 21:31:12,333 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2144, 2.9459, 2.1533, 2.6913, 0.8656, 2.8560, 2.7080, 2.8318], + device='cuda:0'), covar=tensor([0.1155, 0.1364, 0.2097, 0.1044, 0.3847, 0.1087, 0.1194, 0.1441], + device='cuda:0'), in_proj_covar=tensor([0.0445, 0.0369, 0.0444, 0.0321, 0.0384, 0.0378, 0.0364, 0.0397], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:31:17,042 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77229.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:31:20,288 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 21:31:28,747 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2024, 1.1564, 1.1523, 1.3359, 1.1183, 1.3815, 1.3149, 1.2534], + device='cuda:0'), covar=tensor([0.0882, 0.0996, 0.1077, 0.0713, 0.0856, 0.0774, 0.0833, 0.0756], + device='cuda:0'), in_proj_covar=tensor([0.0214, 0.0226, 0.0226, 0.0249, 0.0236, 0.0215, 0.0197, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 21:31:40,690 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 21:31:52,760 INFO [train.py:903] (0/4) Epoch 12, batch 2150, loss[loss=0.315, simple_loss=0.3578, pruned_loss=0.1361, over 13324.00 frames. ], tot_loss[loss=0.2326, simple_loss=0.3063, pruned_loss=0.07944, over 3819315.21 frames. ], batch size: 136, lr: 7.00e-03, grad_scale: 4.0 +2023-04-01 21:32:13,144 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.202e+02 5.327e+02 7.168e+02 9.347e+02 2.125e+03, threshold=1.434e+03, percent-clipped=4.0 +2023-04-01 21:32:33,243 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77291.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:32:39,469 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77295.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:32:51,909 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.57 vs. limit=5.0 +2023-04-01 21:32:53,884 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1120, 1.1353, 1.6674, 1.1037, 2.6077, 3.5729, 3.3009, 3.7237], + device='cuda:0'), covar=tensor([0.1586, 0.3654, 0.3064, 0.2177, 0.0465, 0.0139, 0.0196, 0.0175], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0295, 0.0324, 0.0250, 0.0215, 0.0158, 0.0205, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 21:32:55,903 INFO [train.py:903] (0/4) Epoch 12, batch 2200, loss[loss=0.196, simple_loss=0.2684, pruned_loss=0.06184, over 19777.00 frames. ], tot_loss[loss=0.2321, simple_loss=0.3058, pruned_loss=0.07927, over 3823843.55 frames. ], batch size: 47, lr: 6.99e-03, grad_scale: 4.0 +2023-04-01 21:33:05,527 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77316.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:33:25,217 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-01 21:33:57,354 INFO [train.py:903] (0/4) Epoch 12, batch 2250, loss[loss=0.2327, simple_loss=0.3053, pruned_loss=0.08002, over 19604.00 frames. ], tot_loss[loss=0.2319, simple_loss=0.3055, pruned_loss=0.07914, over 3835137.35 frames. ], batch size: 50, lr: 6.99e-03, grad_scale: 4.0 +2023-04-01 21:34:08,706 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0349, 1.2780, 1.7051, 1.2469, 2.8455, 3.7168, 3.4695, 3.9189], + device='cuda:0'), covar=tensor([0.1686, 0.3299, 0.3009, 0.2128, 0.0482, 0.0140, 0.0187, 0.0174], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0294, 0.0325, 0.0250, 0.0215, 0.0158, 0.0206, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 21:34:18,146 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.729e+02 5.298e+02 6.341e+02 7.997e+02 1.542e+03, threshold=1.268e+03, percent-clipped=2.0 +2023-04-01 21:34:27,379 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4834, 4.0223, 4.1978, 4.2037, 1.4501, 3.9070, 3.4183, 3.8715], + device='cuda:0'), covar=tensor([0.1468, 0.0886, 0.0618, 0.0607, 0.5442, 0.0810, 0.0687, 0.1179], + device='cuda:0'), in_proj_covar=tensor([0.0683, 0.0613, 0.0812, 0.0690, 0.0738, 0.0565, 0.0498, 0.0749], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 21:34:35,671 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-01 21:34:58,603 INFO [train.py:903] (0/4) Epoch 12, batch 2300, loss[loss=0.2098, simple_loss=0.2806, pruned_loss=0.06948, over 19136.00 frames. ], tot_loss[loss=0.2342, simple_loss=0.3071, pruned_loss=0.08067, over 3811292.12 frames. ], batch size: 42, lr: 6.99e-03, grad_scale: 4.0 +2023-04-01 21:35:10,948 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 21:35:59,716 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77457.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:36:00,828 INFO [train.py:903] (0/4) Epoch 12, batch 2350, loss[loss=0.2286, simple_loss=0.3002, pruned_loss=0.07848, over 19587.00 frames. ], tot_loss[loss=0.2347, simple_loss=0.3075, pruned_loss=0.08096, over 3815896.66 frames. ], batch size: 52, lr: 6.99e-03, grad_scale: 4.0 +2023-04-01 21:36:22,285 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.400e+02 5.245e+02 6.457e+02 8.417e+02 4.507e+03, threshold=1.291e+03, percent-clipped=6.0 +2023-04-01 21:36:41,243 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 21:36:59,073 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 21:37:02,548 INFO [train.py:903] (0/4) Epoch 12, batch 2400, loss[loss=0.1711, simple_loss=0.2536, pruned_loss=0.0443, over 19810.00 frames. ], tot_loss[loss=0.2339, simple_loss=0.3071, pruned_loss=0.08035, over 3832654.52 frames. ], batch size: 49, lr: 6.99e-03, grad_scale: 8.0 +2023-04-01 21:37:17,451 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5489, 1.3490, 1.3198, 1.8915, 1.5776, 1.7543, 1.9561, 1.6162], + device='cuda:0'), covar=tensor([0.0815, 0.0951, 0.1049, 0.0802, 0.0844, 0.0694, 0.0747, 0.0708], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0224, 0.0224, 0.0248, 0.0234, 0.0214, 0.0196, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 21:37:36,476 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8437, 1.8521, 2.1622, 2.6770, 1.7914, 2.4892, 2.4084, 1.9587], + device='cuda:0'), covar=tensor([0.3682, 0.3271, 0.1457, 0.1639, 0.3459, 0.1435, 0.3470, 0.2733], + device='cuda:0'), in_proj_covar=tensor([0.0802, 0.0823, 0.0653, 0.0897, 0.0788, 0.0711, 0.0789, 0.0719], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 21:38:04,502 INFO [train.py:903] (0/4) Epoch 12, batch 2450, loss[loss=0.2235, simple_loss=0.3052, pruned_loss=0.07094, over 19621.00 frames. ], tot_loss[loss=0.2325, simple_loss=0.3061, pruned_loss=0.07952, over 3834325.47 frames. ], batch size: 61, lr: 6.98e-03, grad_scale: 8.0 +2023-04-01 21:38:21,436 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77572.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:38:21,508 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77572.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:38:22,405 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77573.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:38:24,483 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.532e+02 5.530e+02 6.529e+02 8.263e+02 1.639e+03, threshold=1.306e+03, percent-clipped=2.0 +2023-04-01 21:38:52,856 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77597.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:38:56,122 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6051, 4.1817, 2.5901, 3.8084, 1.1041, 3.9786, 3.9436, 4.0585], + device='cuda:0'), covar=tensor([0.0584, 0.0951, 0.2062, 0.0693, 0.3915, 0.0715, 0.0799, 0.0967], + device='cuda:0'), in_proj_covar=tensor([0.0440, 0.0365, 0.0440, 0.0319, 0.0380, 0.0373, 0.0360, 0.0392], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:39:05,893 INFO [train.py:903] (0/4) Epoch 12, batch 2500, loss[loss=0.2849, simple_loss=0.3542, pruned_loss=0.1078, over 19515.00 frames. ], tot_loss[loss=0.2332, simple_loss=0.3067, pruned_loss=0.07984, over 3846961.01 frames. ], batch size: 64, lr: 6.98e-03, grad_scale: 8.0 +2023-04-01 21:39:27,209 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4280, 1.2973, 1.2912, 1.9114, 1.5174, 1.7073, 1.9036, 1.6593], + device='cuda:0'), covar=tensor([0.0874, 0.0982, 0.1068, 0.0786, 0.0882, 0.0742, 0.0793, 0.0640], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0221, 0.0222, 0.0244, 0.0232, 0.0211, 0.0193, 0.0199], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 21:39:39,443 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77634.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:39:40,513 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77635.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:39:40,758 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6668, 1.4673, 1.5359, 2.1864, 1.6747, 2.0775, 2.1524, 2.0573], + device='cuda:0'), covar=tensor([0.0798, 0.0921, 0.0980, 0.0789, 0.0875, 0.0691, 0.0813, 0.0551], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0220, 0.0221, 0.0243, 0.0231, 0.0211, 0.0193, 0.0199], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 21:39:45,169 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77639.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:40:08,014 INFO [train.py:903] (0/4) Epoch 12, batch 2550, loss[loss=0.2786, simple_loss=0.3378, pruned_loss=0.1097, over 19716.00 frames. ], tot_loss[loss=0.2317, simple_loss=0.3057, pruned_loss=0.07891, over 3845954.14 frames. ], batch size: 63, lr: 6.98e-03, grad_scale: 8.0 +2023-04-01 21:40:30,541 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.096e+02 5.340e+02 6.709e+02 8.508e+02 1.809e+03, threshold=1.342e+03, percent-clipped=3.0 +2023-04-01 21:40:46,046 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77688.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:40:54,790 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=77695.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:41:05,573 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-01 21:41:10,773 INFO [train.py:903] (0/4) Epoch 12, batch 2600, loss[loss=0.2011, simple_loss=0.2731, pruned_loss=0.06458, over 19000.00 frames. ], tot_loss[loss=0.2318, simple_loss=0.3059, pruned_loss=0.07885, over 3836731.19 frames. ], batch size: 42, lr: 6.98e-03, grad_scale: 8.0 +2023-04-01 21:41:54,844 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-01 21:42:02,077 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77750.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:42:07,395 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=77754.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:42:11,589 INFO [train.py:903] (0/4) Epoch 12, batch 2650, loss[loss=0.2057, simple_loss=0.2756, pruned_loss=0.06787, over 19801.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.3047, pruned_loss=0.0781, over 3829279.78 frames. ], batch size: 48, lr: 6.97e-03, grad_scale: 8.0 +2023-04-01 21:42:32,181 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.366e+02 5.775e+02 6.860e+02 8.613e+02 1.817e+03, threshold=1.372e+03, percent-clipped=5.0 +2023-04-01 21:42:33,300 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-01 21:43:06,917 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9781, 3.5955, 2.4028, 3.3093, 0.7206, 3.4013, 3.4110, 3.4688], + device='cuda:0'), covar=tensor([0.0852, 0.1299, 0.2117, 0.0815, 0.4296, 0.0904, 0.0883, 0.1286], + device='cuda:0'), in_proj_covar=tensor([0.0449, 0.0374, 0.0450, 0.0326, 0.0386, 0.0379, 0.0367, 0.0400], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:43:12,307 INFO [train.py:903] (0/4) Epoch 12, batch 2700, loss[loss=0.2164, simple_loss=0.2816, pruned_loss=0.07561, over 19415.00 frames. ], tot_loss[loss=0.2306, simple_loss=0.3047, pruned_loss=0.0783, over 3837281.09 frames. ], batch size: 48, lr: 6.97e-03, grad_scale: 8.0 +2023-04-01 21:43:38,253 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77828.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:44:08,848 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77853.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:44:14,412 INFO [train.py:903] (0/4) Epoch 12, batch 2750, loss[loss=0.22, simple_loss=0.3049, pruned_loss=0.06754, over 19664.00 frames. ], tot_loss[loss=0.2309, simple_loss=0.3049, pruned_loss=0.0785, over 3827940.22 frames. ], batch size: 58, lr: 6.97e-03, grad_scale: 8.0 +2023-04-01 21:44:36,443 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.412e+02 5.785e+02 6.935e+02 8.560e+02 1.739e+03, threshold=1.387e+03, percent-clipped=4.0 +2023-04-01 21:45:15,202 INFO [train.py:903] (0/4) Epoch 12, batch 2800, loss[loss=0.1985, simple_loss=0.2663, pruned_loss=0.06538, over 19743.00 frames. ], tot_loss[loss=0.2313, simple_loss=0.3054, pruned_loss=0.07862, over 3820176.95 frames. ], batch size: 47, lr: 6.97e-03, grad_scale: 8.0 +2023-04-01 21:46:01,262 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=77944.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:46:19,703 INFO [train.py:903] (0/4) Epoch 12, batch 2850, loss[loss=0.2064, simple_loss=0.2779, pruned_loss=0.06742, over 19395.00 frames. ], tot_loss[loss=0.2306, simple_loss=0.3048, pruned_loss=0.0782, over 3815943.57 frames. ], batch size: 48, lr: 6.97e-03, grad_scale: 4.0 +2023-04-01 21:46:22,361 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6778, 1.2883, 1.5044, 1.5925, 3.2164, 0.9485, 2.1786, 3.5105], + device='cuda:0'), covar=tensor([0.0482, 0.2688, 0.2654, 0.1701, 0.0771, 0.2672, 0.1267, 0.0293], + device='cuda:0'), in_proj_covar=tensor([0.0354, 0.0333, 0.0345, 0.0315, 0.0340, 0.0330, 0.0331, 0.0351], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:46:32,713 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=77969.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:46:41,213 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.493e+02 5.663e+02 6.634e+02 8.773e+02 3.814e+03, threshold=1.327e+03, percent-clipped=6.0 +2023-04-01 21:46:43,717 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=77978.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:47:11,981 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-78000.pt +2023-04-01 21:47:20,361 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78006.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:47:22,310 INFO [train.py:903] (0/4) Epoch 12, batch 2900, loss[loss=0.1793, simple_loss=0.2587, pruned_loss=0.04991, over 19410.00 frames. ], tot_loss[loss=0.2303, simple_loss=0.3046, pruned_loss=0.07799, over 3818076.19 frames. ], batch size: 48, lr: 6.96e-03, grad_scale: 4.0 +2023-04-01 21:47:22,334 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-01 21:47:25,105 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78010.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:47:39,767 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2502, 1.3789, 1.6770, 1.4736, 2.4434, 2.1287, 2.5534, 0.9067], + device='cuda:0'), covar=tensor([0.2132, 0.3742, 0.2294, 0.1765, 0.1296, 0.1834, 0.1312, 0.3718], + device='cuda:0'), in_proj_covar=tensor([0.0489, 0.0578, 0.0608, 0.0437, 0.0594, 0.0489, 0.0644, 0.0495], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 21:47:52,652 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78031.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:47:57,902 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78035.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:48:02,313 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=78039.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:48:25,033 INFO [train.py:903] (0/4) Epoch 12, batch 2950, loss[loss=0.2411, simple_loss=0.3148, pruned_loss=0.0837, over 19672.00 frames. ], tot_loss[loss=0.2309, simple_loss=0.3052, pruned_loss=0.07828, over 3820003.79 frames. ], batch size: 53, lr: 6.96e-03, grad_scale: 4.0 +2023-04-01 21:48:48,726 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.235e+02 5.583e+02 6.866e+02 9.102e+02 1.641e+03, threshold=1.373e+03, percent-clipped=7.0 +2023-04-01 21:49:09,996 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78093.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:49:28,174 INFO [train.py:903] (0/4) Epoch 12, batch 3000, loss[loss=0.2336, simple_loss=0.3185, pruned_loss=0.07429, over 19628.00 frames. ], tot_loss[loss=0.2302, simple_loss=0.3048, pruned_loss=0.07779, over 3838199.14 frames. ], batch size: 57, lr: 6.96e-03, grad_scale: 4.0 +2023-04-01 21:49:28,175 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 21:49:40,666 INFO [train.py:937] (0/4) Epoch 12, validation: loss=0.1772, simple_loss=0.2779, pruned_loss=0.0383, over 944034.00 frames. +2023-04-01 21:49:40,667 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18224MB +2023-04-01 21:49:45,494 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-01 21:50:38,050 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78154.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:50:42,293 INFO [train.py:903] (0/4) Epoch 12, batch 3050, loss[loss=0.2393, simple_loss=0.3167, pruned_loss=0.081, over 19625.00 frames. ], tot_loss[loss=0.2301, simple_loss=0.3047, pruned_loss=0.07773, over 3842159.65 frames. ], batch size: 57, lr: 6.96e-03, grad_scale: 4.0 +2023-04-01 21:51:04,779 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.897e+02 5.238e+02 6.566e+02 8.426e+02 1.854e+03, threshold=1.313e+03, percent-clipped=6.0 +2023-04-01 21:51:20,716 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9253, 1.9663, 2.1789, 2.6371, 1.7554, 2.5518, 2.4671, 2.0804], + device='cuda:0'), covar=tensor([0.3281, 0.2882, 0.1338, 0.1591, 0.3144, 0.1359, 0.3302, 0.2531], + device='cuda:0'), in_proj_covar=tensor([0.0795, 0.0822, 0.0648, 0.0891, 0.0783, 0.0708, 0.0785, 0.0715], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 21:51:43,590 INFO [train.py:903] (0/4) Epoch 12, batch 3100, loss[loss=0.2148, simple_loss=0.3012, pruned_loss=0.06422, over 19531.00 frames. ], tot_loss[loss=0.2322, simple_loss=0.3062, pruned_loss=0.07909, over 3829418.72 frames. ], batch size: 56, lr: 6.95e-03, grad_scale: 4.0 +2023-04-01 21:52:14,923 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3817, 2.1532, 1.6232, 1.4625, 2.0152, 1.2215, 1.4350, 1.8876], + device='cuda:0'), covar=tensor([0.0928, 0.0744, 0.1128, 0.0742, 0.0480, 0.1242, 0.0592, 0.0351], + device='cuda:0'), in_proj_covar=tensor([0.0291, 0.0301, 0.0323, 0.0244, 0.0234, 0.0322, 0.0286, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 21:52:18,458 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-04-01 21:52:36,393 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78250.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:52:45,562 INFO [train.py:903] (0/4) Epoch 12, batch 3150, loss[loss=0.2465, simple_loss=0.3002, pruned_loss=0.09639, over 19359.00 frames. ], tot_loss[loss=0.2321, simple_loss=0.3062, pruned_loss=0.07895, over 3835173.60 frames. ], batch size: 44, lr: 6.95e-03, grad_scale: 4.0 +2023-04-01 21:52:53,234 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.18 vs. limit=5.0 +2023-04-01 21:53:07,630 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.745e+02 5.622e+02 7.364e+02 8.683e+02 1.879e+03, threshold=1.473e+03, percent-clipped=3.0 +2023-04-01 21:53:15,176 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-01 21:53:32,902 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0159, 1.3574, 1.9423, 1.4695, 2.9376, 4.4628, 4.4629, 4.9154], + device='cuda:0'), covar=tensor([0.1753, 0.3499, 0.2981, 0.2058, 0.0529, 0.0173, 0.0155, 0.0122], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0298, 0.0326, 0.0252, 0.0216, 0.0159, 0.0205, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 21:53:47,084 INFO [train.py:903] (0/4) Epoch 12, batch 3200, loss[loss=0.2363, simple_loss=0.3076, pruned_loss=0.08253, over 18784.00 frames. ], tot_loss[loss=0.2337, simple_loss=0.3075, pruned_loss=0.07989, over 3833620.45 frames. ], batch size: 74, lr: 6.95e-03, grad_scale: 8.0 +2023-04-01 21:54:39,708 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78349.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:54:50,776 INFO [train.py:903] (0/4) Epoch 12, batch 3250, loss[loss=0.2213, simple_loss=0.3014, pruned_loss=0.07055, over 19539.00 frames. ], tot_loss[loss=0.2322, simple_loss=0.3061, pruned_loss=0.07917, over 3830141.26 frames. ], batch size: 56, lr: 6.95e-03, grad_scale: 8.0 +2023-04-01 21:55:11,630 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78374.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:55:13,651 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.865e+02 5.227e+02 6.164e+02 7.465e+02 1.234e+03, threshold=1.233e+03, percent-clipped=0.0 +2023-04-01 21:55:54,299 INFO [train.py:903] (0/4) Epoch 12, batch 3300, loss[loss=0.2152, simple_loss=0.288, pruned_loss=0.07123, over 19740.00 frames. ], tot_loss[loss=0.2324, simple_loss=0.3062, pruned_loss=0.07935, over 3825121.59 frames. ], batch size: 51, lr: 6.95e-03, grad_scale: 8.0 +2023-04-01 21:55:57,128 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78410.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:55:57,908 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-01 21:56:26,566 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78435.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:56:55,610 INFO [train.py:903] (0/4) Epoch 12, batch 3350, loss[loss=0.2244, simple_loss=0.3045, pruned_loss=0.07216, over 18194.00 frames. ], tot_loss[loss=0.2321, simple_loss=0.3061, pruned_loss=0.07911, over 3828532.45 frames. ], batch size: 83, lr: 6.94e-03, grad_scale: 8.0 +2023-04-01 21:57:18,021 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.479e+02 5.251e+02 6.145e+02 6.896e+02 1.617e+03, threshold=1.229e+03, percent-clipped=1.0 +2023-04-01 21:57:20,569 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78478.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:57:57,211 INFO [train.py:903] (0/4) Epoch 12, batch 3400, loss[loss=0.2275, simple_loss=0.3087, pruned_loss=0.07315, over 19413.00 frames. ], tot_loss[loss=0.2312, simple_loss=0.3055, pruned_loss=0.07839, over 3824288.00 frames. ], batch size: 70, lr: 6.94e-03, grad_scale: 8.0 +2023-04-01 21:59:00,757 INFO [train.py:903] (0/4) Epoch 12, batch 3450, loss[loss=0.2128, simple_loss=0.2992, pruned_loss=0.06321, over 19760.00 frames. ], tot_loss[loss=0.2302, simple_loss=0.3049, pruned_loss=0.0777, over 3824971.00 frames. ], batch size: 54, lr: 6.94e-03, grad_scale: 8.0 +2023-04-01 21:59:06,193 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-01 21:59:21,424 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78574.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 21:59:23,278 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.785e+02 5.692e+02 7.472e+02 9.495e+02 2.057e+03, threshold=1.494e+03, percent-clipped=9.0 +2023-04-01 21:59:44,669 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=78594.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:00:03,311 INFO [train.py:903] (0/4) Epoch 12, batch 3500, loss[loss=0.2269, simple_loss=0.3064, pruned_loss=0.07371, over 19675.00 frames. ], tot_loss[loss=0.2318, simple_loss=0.3061, pruned_loss=0.07877, over 3804879.02 frames. ], batch size: 58, lr: 6.94e-03, grad_scale: 8.0 +2023-04-01 22:00:34,900 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.65 vs. limit=5.0 +2023-04-01 22:00:55,180 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9063, 1.7328, 1.5441, 2.0008, 1.6057, 1.6840, 1.5253, 1.8559], + device='cuda:0'), covar=tensor([0.0938, 0.1421, 0.1402, 0.0901, 0.1293, 0.0502, 0.1238, 0.0649], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0349, 0.0293, 0.0239, 0.0294, 0.0241, 0.0278, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:01:05,331 INFO [train.py:903] (0/4) Epoch 12, batch 3550, loss[loss=0.2165, simple_loss=0.302, pruned_loss=0.06547, over 19601.00 frames. ], tot_loss[loss=0.2319, simple_loss=0.3064, pruned_loss=0.07874, over 3796264.76 frames. ], batch size: 61, lr: 6.93e-03, grad_scale: 8.0 +2023-04-01 22:01:26,755 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.897e+02 5.737e+02 7.042e+02 1.025e+03 1.962e+03, threshold=1.408e+03, percent-clipped=6.0 +2023-04-01 22:02:07,318 INFO [train.py:903] (0/4) Epoch 12, batch 3600, loss[loss=0.273, simple_loss=0.3321, pruned_loss=0.1069, over 13452.00 frames. ], tot_loss[loss=0.2329, simple_loss=0.3071, pruned_loss=0.07939, over 3787924.54 frames. ], batch size: 137, lr: 6.93e-03, grad_scale: 8.0 +2023-04-01 22:02:07,771 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6871, 1.6582, 1.5739, 1.3090, 1.2299, 1.4259, 0.2168, 0.6414], + device='cuda:0'), covar=tensor([0.0467, 0.0483, 0.0286, 0.0454, 0.0920, 0.0487, 0.0895, 0.0787], + device='cuda:0'), in_proj_covar=tensor([0.0342, 0.0336, 0.0334, 0.0360, 0.0432, 0.0359, 0.0317, 0.0328], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 22:02:08,900 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78709.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:03:09,180 INFO [train.py:903] (0/4) Epoch 12, batch 3650, loss[loss=0.2472, simple_loss=0.3209, pruned_loss=0.08673, over 19243.00 frames. ], tot_loss[loss=0.2326, simple_loss=0.3065, pruned_loss=0.07941, over 3785547.92 frames. ], batch size: 66, lr: 6.93e-03, grad_scale: 8.0 +2023-04-01 22:03:33,805 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.478e+02 5.318e+02 6.562e+02 8.219e+02 2.478e+03, threshold=1.312e+03, percent-clipped=4.0 +2023-04-01 22:03:37,877 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2708, 1.3076, 1.5571, 1.4874, 2.1962, 1.9411, 2.2745, 0.7072], + device='cuda:0'), covar=tensor([0.2308, 0.3930, 0.2475, 0.1778, 0.1434, 0.2041, 0.1373, 0.3876], + device='cuda:0'), in_proj_covar=tensor([0.0488, 0.0577, 0.0606, 0.0437, 0.0592, 0.0488, 0.0643, 0.0492], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 22:04:14,201 INFO [train.py:903] (0/4) Epoch 12, batch 3700, loss[loss=0.2531, simple_loss=0.3272, pruned_loss=0.08952, over 18720.00 frames. ], tot_loss[loss=0.2323, simple_loss=0.3065, pruned_loss=0.07902, over 3793218.10 frames. ], batch size: 74, lr: 6.93e-03, grad_scale: 8.0 +2023-04-01 22:04:31,377 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=78822.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:05:15,862 INFO [train.py:903] (0/4) Epoch 12, batch 3750, loss[loss=0.2173, simple_loss=0.2958, pruned_loss=0.06944, over 19837.00 frames. ], tot_loss[loss=0.2316, simple_loss=0.3057, pruned_loss=0.07872, over 3817007.10 frames. ], batch size: 52, lr: 6.93e-03, grad_scale: 8.0 +2023-04-01 22:05:18,441 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3202, 3.8926, 2.6399, 3.5022, 0.8965, 3.6726, 3.6360, 3.8286], + device='cuda:0'), covar=tensor([0.0704, 0.1074, 0.2058, 0.0798, 0.4185, 0.0831, 0.0880, 0.1085], + device='cuda:0'), in_proj_covar=tensor([0.0444, 0.0369, 0.0445, 0.0321, 0.0381, 0.0376, 0.0364, 0.0399], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:05:37,732 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.608e+02 5.233e+02 6.179e+02 8.242e+02 1.500e+03, threshold=1.236e+03, percent-clipped=2.0 +2023-04-01 22:06:16,424 INFO [train.py:903] (0/4) Epoch 12, batch 3800, loss[loss=0.2418, simple_loss=0.3125, pruned_loss=0.08562, over 19847.00 frames. ], tot_loss[loss=0.2306, simple_loss=0.3046, pruned_loss=0.07826, over 3832904.71 frames. ], batch size: 52, lr: 6.92e-03, grad_scale: 8.0 +2023-04-01 22:06:29,201 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=78918.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:06:53,690 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-01 22:06:54,058 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=78937.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:07:17,949 INFO [train.py:903] (0/4) Epoch 12, batch 3850, loss[loss=0.2453, simple_loss=0.3223, pruned_loss=0.08412, over 19786.00 frames. ], tot_loss[loss=0.2308, simple_loss=0.305, pruned_loss=0.07825, over 3832037.70 frames. ], batch size: 56, lr: 6.92e-03, grad_scale: 8.0 +2023-04-01 22:07:19,947 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-01 22:07:27,650 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-04-01 22:07:28,312 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=78965.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:07:40,738 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.425e+02 5.982e+02 6.977e+02 9.373e+02 2.137e+03, threshold=1.395e+03, percent-clipped=8.0 +2023-04-01 22:07:49,874 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=78983.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:07:58,091 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=78990.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:08:21,045 INFO [train.py:903] (0/4) Epoch 12, batch 3900, loss[loss=0.2533, simple_loss=0.3227, pruned_loss=0.09196, over 19291.00 frames. ], tot_loss[loss=0.2304, simple_loss=0.3049, pruned_loss=0.07797, over 3831341.63 frames. ], batch size: 66, lr: 6.92e-03, grad_scale: 8.0 +2023-04-01 22:08:51,031 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79033.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:09:22,295 INFO [train.py:903] (0/4) Epoch 12, batch 3950, loss[loss=0.2406, simple_loss=0.3179, pruned_loss=0.08161, over 19659.00 frames. ], tot_loss[loss=0.2308, simple_loss=0.305, pruned_loss=0.07825, over 3828807.52 frames. ], batch size: 58, lr: 6.92e-03, grad_scale: 8.0 +2023-04-01 22:09:29,071 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-01 22:09:40,906 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5397, 2.3966, 1.6548, 1.4920, 2.1490, 1.1867, 1.4226, 1.9006], + device='cuda:0'), covar=tensor([0.0970, 0.0614, 0.1065, 0.0739, 0.0498, 0.1227, 0.0766, 0.0452], + device='cuda:0'), in_proj_covar=tensor([0.0291, 0.0300, 0.0326, 0.0244, 0.0235, 0.0318, 0.0287, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:09:43,636 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.615e+02 6.132e+02 6.993e+02 8.356e+02 2.478e+03, threshold=1.399e+03, percent-clipped=5.0 +2023-04-01 22:10:04,420 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.52 vs. limit=5.0 +2023-04-01 22:10:18,509 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4554, 1.3022, 1.3221, 1.8517, 1.5832, 1.6784, 1.7806, 1.5435], + device='cuda:0'), covar=tensor([0.0873, 0.0985, 0.1094, 0.0684, 0.0790, 0.0726, 0.0780, 0.0710], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0225, 0.0223, 0.0247, 0.0236, 0.0211, 0.0195, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 22:10:22,895 INFO [train.py:903] (0/4) Epoch 12, batch 4000, loss[loss=0.2497, simple_loss=0.3335, pruned_loss=0.08293, over 19532.00 frames. ], tot_loss[loss=0.2312, simple_loss=0.3055, pruned_loss=0.07847, over 3840746.55 frames. ], batch size: 64, lr: 6.91e-03, grad_scale: 8.0 +2023-04-01 22:11:13,119 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-01 22:11:24,601 INFO [train.py:903] (0/4) Epoch 12, batch 4050, loss[loss=0.2017, simple_loss=0.2834, pruned_loss=0.05999, over 19601.00 frames. ], tot_loss[loss=0.2304, simple_loss=0.305, pruned_loss=0.07793, over 3845793.41 frames. ], batch size: 52, lr: 6.91e-03, grad_scale: 8.0 +2023-04-01 22:11:47,121 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.257e+02 5.067e+02 6.190e+02 7.758e+02 2.001e+03, threshold=1.238e+03, percent-clipped=2.0 +2023-04-01 22:12:07,801 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79193.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:12:26,660 INFO [train.py:903] (0/4) Epoch 12, batch 4100, loss[loss=0.2481, simple_loss=0.3202, pruned_loss=0.08803, over 19773.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.3048, pruned_loss=0.07807, over 3835907.70 frames. ], batch size: 56, lr: 6.91e-03, grad_scale: 8.0 +2023-04-01 22:12:39,286 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79218.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:13:03,848 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-01 22:13:27,006 INFO [train.py:903] (0/4) Epoch 12, batch 4150, loss[loss=0.1758, simple_loss=0.2493, pruned_loss=0.0511, over 19749.00 frames. ], tot_loss[loss=0.2298, simple_loss=0.3042, pruned_loss=0.07774, over 3833463.70 frames. ], batch size: 47, lr: 6.91e-03, grad_scale: 8.0 +2023-04-01 22:13:45,778 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9341, 1.9920, 2.1749, 2.8840, 1.9423, 2.5956, 2.4927, 1.9981], + device='cuda:0'), covar=tensor([0.3675, 0.3335, 0.1503, 0.1700, 0.3497, 0.1537, 0.3525, 0.2771], + device='cuda:0'), in_proj_covar=tensor([0.0801, 0.0826, 0.0654, 0.0897, 0.0788, 0.0714, 0.0793, 0.0715], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 22:13:49,455 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.496e+02 5.552e+02 6.928e+02 9.304e+02 2.111e+03, threshold=1.386e+03, percent-clipped=6.0 +2023-04-01 22:14:06,919 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79289.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:14:23,416 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79303.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 22:14:28,774 INFO [train.py:903] (0/4) Epoch 12, batch 4200, loss[loss=0.1948, simple_loss=0.2724, pruned_loss=0.05861, over 16070.00 frames. ], tot_loss[loss=0.2309, simple_loss=0.305, pruned_loss=0.07838, over 3821801.54 frames. ], batch size: 35, lr: 6.91e-03, grad_scale: 8.0 +2023-04-01 22:14:33,322 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-01 22:14:35,749 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79314.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:14:53,761 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79327.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:15:22,560 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79351.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:15:31,218 INFO [train.py:903] (0/4) Epoch 12, batch 4250, loss[loss=0.2752, simple_loss=0.3509, pruned_loss=0.09975, over 17532.00 frames. ], tot_loss[loss=0.231, simple_loss=0.3053, pruned_loss=0.07832, over 3822920.82 frames. ], batch size: 101, lr: 6.90e-03, grad_scale: 8.0 +2023-04-01 22:15:43,239 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-01 22:15:52,361 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.517e+02 4.925e+02 6.557e+02 8.003e+02 1.515e+03, threshold=1.311e+03, percent-clipped=3.0 +2023-04-01 22:15:54,751 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-01 22:16:00,677 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79382.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:16:13,150 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79393.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 22:16:33,044 INFO [train.py:903] (0/4) Epoch 12, batch 4300, loss[loss=0.2011, simple_loss=0.2808, pruned_loss=0.06064, over 19752.00 frames. ], tot_loss[loss=0.2313, simple_loss=0.3055, pruned_loss=0.07859, over 3837519.65 frames. ], batch size: 51, lr: 6.90e-03, grad_scale: 8.0 +2023-04-01 22:17:13,283 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79442.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:17:24,354 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-01 22:17:29,506 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2234, 1.9625, 1.8456, 2.0837, 1.9468, 1.8826, 1.9009, 2.0997], + device='cuda:0'), covar=tensor([0.0758, 0.1249, 0.1198, 0.0922, 0.1137, 0.0444, 0.1066, 0.0567], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0352, 0.0292, 0.0240, 0.0297, 0.0242, 0.0281, 0.0240], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:17:32,597 INFO [train.py:903] (0/4) Epoch 12, batch 4350, loss[loss=0.2489, simple_loss=0.3199, pruned_loss=0.08898, over 18236.00 frames. ], tot_loss[loss=0.2323, simple_loss=0.3067, pruned_loss=0.079, over 3834055.75 frames. ], batch size: 83, lr: 6.90e-03, grad_scale: 8.0 +2023-04-01 22:17:52,327 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5075, 3.6694, 4.0068, 3.9896, 2.2018, 3.7467, 3.4179, 3.7201], + device='cuda:0'), covar=tensor([0.1151, 0.2850, 0.0551, 0.0628, 0.4101, 0.1028, 0.0558, 0.0970], + device='cuda:0'), in_proj_covar=tensor([0.0682, 0.0618, 0.0812, 0.0691, 0.0740, 0.0567, 0.0496, 0.0749], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 22:17:54,382 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.719e+02 5.544e+02 6.866e+02 8.754e+02 2.036e+03, threshold=1.373e+03, percent-clipped=4.0 +2023-04-01 22:18:34,725 INFO [train.py:903] (0/4) Epoch 12, batch 4400, loss[loss=0.2381, simple_loss=0.3168, pruned_loss=0.07971, over 18269.00 frames. ], tot_loss[loss=0.2342, simple_loss=0.3081, pruned_loss=0.08016, over 3807590.93 frames. ], batch size: 83, lr: 6.90e-03, grad_scale: 8.0 +2023-04-01 22:18:58,949 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-01 22:19:07,280 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-01 22:19:38,007 INFO [train.py:903] (0/4) Epoch 12, batch 4450, loss[loss=0.221, simple_loss=0.3015, pruned_loss=0.07026, over 19674.00 frames. ], tot_loss[loss=0.2336, simple_loss=0.3075, pruned_loss=0.07983, over 3796525.94 frames. ], batch size: 53, lr: 6.90e-03, grad_scale: 8.0 +2023-04-01 22:20:00,025 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.129e+02 5.261e+02 6.909e+02 8.531e+02 1.990e+03, threshold=1.382e+03, percent-clipped=5.0 +2023-04-01 22:20:00,429 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1177, 1.2987, 1.7811, 0.8795, 2.4301, 3.0805, 2.7743, 3.2594], + device='cuda:0'), covar=tensor([0.1553, 0.3438, 0.2862, 0.2334, 0.0484, 0.0188, 0.0251, 0.0231], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0300, 0.0329, 0.0252, 0.0219, 0.0161, 0.0207, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 22:20:33,583 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79603.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:20:41,077 INFO [train.py:903] (0/4) Epoch 12, batch 4500, loss[loss=0.2107, simple_loss=0.2828, pruned_loss=0.06926, over 19776.00 frames. ], tot_loss[loss=0.231, simple_loss=0.3052, pruned_loss=0.07838, over 3821235.12 frames. ], batch size: 48, lr: 6.89e-03, grad_scale: 8.0 +2023-04-01 22:21:29,749 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79647.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 22:21:43,346 INFO [train.py:903] (0/4) Epoch 12, batch 4550, loss[loss=0.2141, simple_loss=0.292, pruned_loss=0.06813, over 19753.00 frames. ], tot_loss[loss=0.2307, simple_loss=0.3049, pruned_loss=0.07827, over 3826864.07 frames. ], batch size: 54, lr: 6.89e-03, grad_scale: 8.0 +2023-04-01 22:21:52,482 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-01 22:22:01,888 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9779, 1.5655, 1.4261, 1.6283, 1.5073, 1.4118, 1.2428, 1.6787], + device='cuda:0'), covar=tensor([0.0856, 0.1187, 0.1458, 0.0977, 0.1243, 0.0709, 0.1588, 0.0759], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0353, 0.0293, 0.0241, 0.0298, 0.0243, 0.0283, 0.0241], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:22:04,157 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79675.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:22:04,981 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.248e+02 5.458e+02 6.277e+02 7.572e+02 1.495e+03, threshold=1.255e+03, percent-clipped=2.0 +2023-04-01 22:22:08,805 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-01 22:22:15,964 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-01 22:22:23,291 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3863, 1.4649, 1.7097, 1.5829, 2.6889, 2.2649, 2.8463, 1.0486], + device='cuda:0'), covar=tensor([0.2085, 0.3500, 0.2127, 0.1637, 0.1296, 0.1745, 0.1255, 0.3613], + device='cuda:0'), in_proj_covar=tensor([0.0491, 0.0578, 0.0608, 0.0438, 0.0593, 0.0493, 0.0643, 0.0494], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 22:22:28,829 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9292, 1.5694, 1.4335, 1.7389, 1.6273, 1.5898, 1.4085, 1.7525], + device='cuda:0'), covar=tensor([0.0863, 0.1197, 0.1389, 0.0965, 0.1052, 0.0501, 0.1321, 0.0673], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0353, 0.0292, 0.0240, 0.0297, 0.0242, 0.0281, 0.0241], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:22:29,747 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79695.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:22:33,576 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=79698.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:22:44,778 INFO [train.py:903] (0/4) Epoch 12, batch 4600, loss[loss=0.2313, simple_loss=0.3156, pruned_loss=0.07346, over 19659.00 frames. ], tot_loss[loss=0.2313, simple_loss=0.3054, pruned_loss=0.07861, over 3816125.91 frames. ], batch size: 55, lr: 6.89e-03, grad_scale: 8.0 +2023-04-01 22:23:04,274 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=79723.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:23:07,617 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79726.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:23:18,661 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1405, 1.9893, 1.8297, 1.6435, 1.3751, 1.6361, 0.3423, 0.9610], + device='cuda:0'), covar=tensor([0.0393, 0.0436, 0.0330, 0.0567, 0.0947, 0.0680, 0.0958, 0.0786], + device='cuda:0'), in_proj_covar=tensor([0.0340, 0.0330, 0.0330, 0.0355, 0.0430, 0.0357, 0.0310, 0.0325], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 22:23:21,764 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79737.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 22:23:46,648 INFO [train.py:903] (0/4) Epoch 12, batch 4650, loss[loss=0.2361, simple_loss=0.3009, pruned_loss=0.08562, over 19806.00 frames. ], tot_loss[loss=0.2321, simple_loss=0.306, pruned_loss=0.07908, over 3820760.54 frames. ], batch size: 48, lr: 6.89e-03, grad_scale: 8.0 +2023-04-01 22:23:51,764 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79762.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 22:24:06,322 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-01 22:24:09,613 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.145e+02 5.614e+02 7.010e+02 8.934e+02 1.991e+03, threshold=1.402e+03, percent-clipped=7.0 +2023-04-01 22:24:14,656 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=79780.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:24:15,648 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-01 22:24:37,945 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-01 22:24:48,915 INFO [train.py:903] (0/4) Epoch 12, batch 4700, loss[loss=0.2274, simple_loss=0.3041, pruned_loss=0.07539, over 19785.00 frames. ], tot_loss[loss=0.2317, simple_loss=0.3054, pruned_loss=0.079, over 3824784.28 frames. ], batch size: 56, lr: 6.88e-03, grad_scale: 8.0 +2023-04-01 22:24:52,486 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79810.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:25:12,482 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-01 22:25:30,356 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79841.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:25:45,609 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=79852.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 22:25:52,118 INFO [train.py:903] (0/4) Epoch 12, batch 4750, loss[loss=0.2282, simple_loss=0.3011, pruned_loss=0.07766, over 18283.00 frames. ], tot_loss[loss=0.2316, simple_loss=0.3056, pruned_loss=0.07879, over 3819946.27 frames. ], batch size: 84, lr: 6.88e-03, grad_scale: 8.0 +2023-04-01 22:26:14,209 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.848e+02 5.173e+02 6.366e+02 7.623e+02 1.625e+03, threshold=1.273e+03, percent-clipped=2.0 +2023-04-01 22:26:54,256 INFO [train.py:903] (0/4) Epoch 12, batch 4800, loss[loss=0.2123, simple_loss=0.2979, pruned_loss=0.06338, over 19531.00 frames. ], tot_loss[loss=0.2299, simple_loss=0.3039, pruned_loss=0.07795, over 3819359.02 frames. ], batch size: 64, lr: 6.88e-03, grad_scale: 8.0 +2023-04-01 22:27:42,673 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=79947.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:27:56,277 INFO [train.py:903] (0/4) Epoch 12, batch 4850, loss[loss=0.2194, simple_loss=0.2997, pruned_loss=0.06952, over 19497.00 frames. ], tot_loss[loss=0.2303, simple_loss=0.3042, pruned_loss=0.07816, over 3810593.53 frames. ], batch size: 64, lr: 6.88e-03, grad_scale: 16.0 +2023-04-01 22:28:19,179 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.933e+02 5.432e+02 6.685e+02 9.186e+02 1.976e+03, threshold=1.337e+03, percent-clipped=11.0 +2023-04-01 22:28:23,590 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-01 22:28:43,826 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-01 22:28:48,541 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-80000.pt +2023-04-01 22:28:49,943 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-01 22:28:51,011 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-01 22:28:57,919 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1060, 1.8270, 1.3947, 1.1268, 1.5644, 1.0737, 1.1182, 1.6515], + device='cuda:0'), covar=tensor([0.0766, 0.0721, 0.1059, 0.0761, 0.0532, 0.1204, 0.0643, 0.0323], + device='cuda:0'), in_proj_covar=tensor([0.0291, 0.0298, 0.0324, 0.0244, 0.0237, 0.0315, 0.0287, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:28:59,976 INFO [train.py:903] (0/4) Epoch 12, batch 4900, loss[loss=0.2087, simple_loss=0.2868, pruned_loss=0.06534, over 19653.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.3031, pruned_loss=0.07734, over 3812688.06 frames. ], batch size: 53, lr: 6.88e-03, grad_scale: 8.0 +2023-04-01 22:29:02,310 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-01 22:29:13,755 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80018.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 22:29:14,564 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80019.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:29:21,258 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-01 22:29:43,698 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80043.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 22:30:03,749 INFO [train.py:903] (0/4) Epoch 12, batch 4950, loss[loss=0.2759, simple_loss=0.3398, pruned_loss=0.106, over 19347.00 frames. ], tot_loss[loss=0.2297, simple_loss=0.3039, pruned_loss=0.07778, over 3814812.71 frames. ], batch size: 66, lr: 6.87e-03, grad_scale: 8.0 +2023-04-01 22:30:08,680 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80062.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:30:13,639 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80066.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:30:20,457 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-01 22:30:26,823 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.980e+02 5.263e+02 6.788e+02 8.958e+02 2.034e+03, threshold=1.358e+03, percent-clipped=5.0 +2023-04-01 22:30:44,554 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80091.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:30:45,319 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-01 22:30:52,457 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80097.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:31:04,653 INFO [train.py:903] (0/4) Epoch 12, batch 5000, loss[loss=0.2109, simple_loss=0.2962, pruned_loss=0.06282, over 19645.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.3045, pruned_loss=0.07823, over 3818982.13 frames. ], batch size: 60, lr: 6.87e-03, grad_scale: 8.0 +2023-04-01 22:31:05,093 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80108.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 22:31:13,747 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-01 22:31:22,084 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80122.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:31:24,052 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80124.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:31:25,187 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-01 22:31:37,099 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80133.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 22:31:38,193 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80134.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:31:54,278 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0596, 1.6184, 1.6932, 2.0235, 1.8522, 1.7731, 1.6704, 1.8913], + device='cuda:0'), covar=tensor([0.0865, 0.1517, 0.1393, 0.0895, 0.1207, 0.0498, 0.1198, 0.0689], + device='cuda:0'), in_proj_covar=tensor([0.0254, 0.0351, 0.0294, 0.0240, 0.0298, 0.0243, 0.0281, 0.0241], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:32:06,837 INFO [train.py:903] (0/4) Epoch 12, batch 5050, loss[loss=0.2399, simple_loss=0.3173, pruned_loss=0.0812, over 19652.00 frames. ], tot_loss[loss=0.2317, simple_loss=0.3057, pruned_loss=0.07889, over 3826245.31 frames. ], batch size: 58, lr: 6.87e-03, grad_scale: 8.0 +2023-04-01 22:32:30,894 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.181e+02 5.736e+02 7.301e+02 9.465e+02 2.500e+03, threshold=1.460e+03, percent-clipped=5.0 +2023-04-01 22:32:33,496 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.7013, 0.9540, 1.2048, 0.5569, 1.4686, 1.7113, 1.4984, 1.8017], + device='cuda:0'), covar=tensor([0.1223, 0.2513, 0.2214, 0.2054, 0.0825, 0.0365, 0.0299, 0.0302], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0299, 0.0332, 0.0253, 0.0218, 0.0162, 0.0207, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 22:32:41,351 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-01 22:33:08,740 INFO [train.py:903] (0/4) Epoch 12, batch 5100, loss[loss=0.2173, simple_loss=0.3013, pruned_loss=0.06666, over 18089.00 frames. ], tot_loss[loss=0.2315, simple_loss=0.3054, pruned_loss=0.07884, over 3823286.76 frames. ], batch size: 83, lr: 6.87e-03, grad_scale: 8.0 +2023-04-01 22:33:21,044 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-01 22:33:23,204 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-01 22:33:26,508 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-01 22:33:47,569 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80239.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:33:52,237 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9878, 1.9012, 1.8314, 1.6774, 1.6080, 1.7310, 0.9451, 1.3159], + device='cuda:0'), covar=tensor([0.0379, 0.0461, 0.0267, 0.0401, 0.0641, 0.0572, 0.0756, 0.0595], + device='cuda:0'), in_proj_covar=tensor([0.0341, 0.0330, 0.0331, 0.0356, 0.0429, 0.0358, 0.0311, 0.0325], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 22:34:11,774 INFO [train.py:903] (0/4) Epoch 12, batch 5150, loss[loss=0.3211, simple_loss=0.3712, pruned_loss=0.1355, over 18730.00 frames. ], tot_loss[loss=0.231, simple_loss=0.305, pruned_loss=0.0785, over 3824207.72 frames. ], batch size: 74, lr: 6.87e-03, grad_scale: 8.0 +2023-04-01 22:34:23,660 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-01 22:34:34,583 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.377e+02 5.205e+02 6.062e+02 7.794e+02 1.645e+03, threshold=1.212e+03, percent-clipped=2.0 +2023-04-01 22:34:59,508 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 22:35:05,763 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80301.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:35:13,712 INFO [train.py:903] (0/4) Epoch 12, batch 5200, loss[loss=0.2052, simple_loss=0.2738, pruned_loss=0.06828, over 19730.00 frames. ], tot_loss[loss=0.2304, simple_loss=0.3042, pruned_loss=0.07824, over 3825743.65 frames. ], batch size: 46, lr: 6.86e-03, grad_scale: 8.0 +2023-04-01 22:35:26,505 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80318.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:35:27,311 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-01 22:35:58,809 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80343.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:36:00,979 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8181, 4.1875, 4.5159, 4.5163, 1.8060, 4.2109, 3.6911, 4.1870], + device='cuda:0'), covar=tensor([0.1589, 0.0974, 0.0629, 0.0637, 0.5655, 0.0782, 0.0619, 0.1195], + device='cuda:0'), in_proj_covar=tensor([0.0682, 0.0614, 0.0811, 0.0690, 0.0730, 0.0561, 0.0490, 0.0735], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 22:36:13,580 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-01 22:36:16,622 INFO [train.py:903] (0/4) Epoch 12, batch 5250, loss[loss=0.249, simple_loss=0.3269, pruned_loss=0.0855, over 19406.00 frames. ], tot_loss[loss=0.2303, simple_loss=0.3045, pruned_loss=0.07803, over 3824053.63 frames. ], batch size: 70, lr: 6.86e-03, grad_scale: 8.0 +2023-04-01 22:36:40,819 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0 +2023-04-01 22:36:41,380 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.664e+02 5.603e+02 6.465e+02 8.351e+02 1.434e+03, threshold=1.293e+03, percent-clipped=3.0 +2023-04-01 22:36:56,989 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80390.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:37:18,481 INFO [train.py:903] (0/4) Epoch 12, batch 5300, loss[loss=0.2106, simple_loss=0.2955, pruned_loss=0.06288, over 19532.00 frames. ], tot_loss[loss=0.2312, simple_loss=0.3053, pruned_loss=0.07853, over 3830598.84 frames. ], batch size: 54, lr: 6.86e-03, grad_scale: 8.0 +2023-04-01 22:37:22,932 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80411.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:37:26,625 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-04-01 22:37:28,731 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80415.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:37:42,145 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-01 22:38:23,298 INFO [train.py:903] (0/4) Epoch 12, batch 5350, loss[loss=0.2003, simple_loss=0.2804, pruned_loss=0.06005, over 19832.00 frames. ], tot_loss[loss=0.2306, simple_loss=0.3048, pruned_loss=0.07815, over 3832011.77 frames. ], batch size: 52, lr: 6.86e-03, grad_scale: 8.0 +2023-04-01 22:38:44,529 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.162e+02 5.085e+02 6.728e+02 8.660e+02 2.071e+03, threshold=1.346e+03, percent-clipped=4.0 +2023-04-01 22:38:50,333 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80481.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:38:59,002 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-01 22:38:59,483 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2294, 2.2720, 2.5364, 3.2675, 2.2725, 3.2461, 2.7102, 2.2531], + device='cuda:0'), covar=tensor([0.3661, 0.3301, 0.1431, 0.1917, 0.3672, 0.1451, 0.3436, 0.2709], + device='cuda:0'), in_proj_covar=tensor([0.0805, 0.0826, 0.0656, 0.0894, 0.0790, 0.0716, 0.0790, 0.0718], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 22:39:08,254 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=80495.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:39:23,541 INFO [train.py:903] (0/4) Epoch 12, batch 5400, loss[loss=0.2322, simple_loss=0.3116, pruned_loss=0.07637, over 19675.00 frames. ], tot_loss[loss=0.2293, simple_loss=0.3038, pruned_loss=0.07736, over 3842680.37 frames. ], batch size: 53, lr: 6.85e-03, grad_scale: 8.0 +2023-04-01 22:39:28,316 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80512.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:39:38,446 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=80520.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:40:24,799 INFO [train.py:903] (0/4) Epoch 12, batch 5450, loss[loss=0.1769, simple_loss=0.2544, pruned_loss=0.04964, over 19741.00 frames. ], tot_loss[loss=0.2295, simple_loss=0.3037, pruned_loss=0.07759, over 3841002.23 frames. ], batch size: 46, lr: 6.85e-03, grad_scale: 8.0 +2023-04-01 22:40:35,127 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1468, 2.0790, 1.8415, 1.7020, 1.5018, 1.6870, 0.5339, 1.1262], + device='cuda:0'), covar=tensor([0.0414, 0.0475, 0.0353, 0.0552, 0.0922, 0.0732, 0.0982, 0.0830], + device='cuda:0'), in_proj_covar=tensor([0.0343, 0.0330, 0.0333, 0.0356, 0.0428, 0.0358, 0.0312, 0.0326], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 22:40:49,072 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.130e+02 5.165e+02 6.319e+02 8.444e+02 1.726e+03, threshold=1.264e+03, percent-clipped=5.0 +2023-04-01 22:41:26,455 INFO [train.py:903] (0/4) Epoch 12, batch 5500, loss[loss=0.279, simple_loss=0.3457, pruned_loss=0.1062, over 19304.00 frames. ], tot_loss[loss=0.2304, simple_loss=0.3046, pruned_loss=0.07812, over 3837573.61 frames. ], batch size: 66, lr: 6.85e-03, grad_scale: 8.0 +2023-04-01 22:41:53,718 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-01 22:42:12,355 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80645.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:42:29,082 INFO [train.py:903] (0/4) Epoch 12, batch 5550, loss[loss=0.2113, simple_loss=0.2931, pruned_loss=0.06475, over 19525.00 frames. ], tot_loss[loss=0.2312, simple_loss=0.3053, pruned_loss=0.07856, over 3833065.37 frames. ], batch size: 54, lr: 6.85e-03, grad_scale: 8.0 +2023-04-01 22:42:38,034 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-01 22:42:51,877 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.910e+02 5.291e+02 6.725e+02 8.423e+02 1.958e+03, threshold=1.345e+03, percent-clipped=4.0 +2023-04-01 22:42:53,857 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-01 22:43:28,266 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-01 22:43:31,866 INFO [train.py:903] (0/4) Epoch 12, batch 5600, loss[loss=0.2455, simple_loss=0.3222, pruned_loss=0.08441, over 19304.00 frames. ], tot_loss[loss=0.2301, simple_loss=0.3049, pruned_loss=0.07763, over 3836538.59 frames. ], batch size: 66, lr: 6.85e-03, grad_scale: 8.0 +2023-04-01 22:44:10,246 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-04-01 22:44:29,082 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80754.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:44:30,092 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80755.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:44:33,519 INFO [train.py:903] (0/4) Epoch 12, batch 5650, loss[loss=0.1987, simple_loss=0.2791, pruned_loss=0.0591, over 19476.00 frames. ], tot_loss[loss=0.2296, simple_loss=0.3042, pruned_loss=0.07745, over 3834868.73 frames. ], batch size: 49, lr: 6.84e-03, grad_scale: 8.0 +2023-04-01 22:44:36,179 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80760.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:44:57,730 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.312e+02 5.237e+02 6.303e+02 7.862e+02 2.175e+03, threshold=1.261e+03, percent-clipped=3.0 +2023-04-01 22:45:15,165 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=80791.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:45:18,720 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3672, 1.3358, 1.7734, 1.2846, 2.4457, 3.1315, 2.9388, 3.3256], + device='cuda:0'), covar=tensor([0.1481, 0.3345, 0.2927, 0.2215, 0.0785, 0.0273, 0.0231, 0.0263], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0299, 0.0330, 0.0252, 0.0219, 0.0161, 0.0207, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 22:45:20,711 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-01 22:45:35,185 INFO [train.py:903] (0/4) Epoch 12, batch 5700, loss[loss=0.2522, simple_loss=0.3178, pruned_loss=0.09326, over 19524.00 frames. ], tot_loss[loss=0.2301, simple_loss=0.3046, pruned_loss=0.07781, over 3833432.02 frames. ], batch size: 54, lr: 6.84e-03, grad_scale: 8.0 +2023-04-01 22:45:57,696 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80825.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:46:21,578 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7924, 1.3660, 1.4633, 1.6183, 3.2944, 1.1198, 2.1577, 3.6579], + device='cuda:0'), covar=tensor([0.0415, 0.2552, 0.2716, 0.1795, 0.0754, 0.2513, 0.1416, 0.0281], + device='cuda:0'), in_proj_covar=tensor([0.0361, 0.0341, 0.0353, 0.0320, 0.0347, 0.0332, 0.0336, 0.0361], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:46:35,800 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=80856.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:46:38,760 INFO [train.py:903] (0/4) Epoch 12, batch 5750, loss[loss=0.2657, simple_loss=0.3364, pruned_loss=0.09745, over 19343.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.304, pruned_loss=0.07685, over 3843521.30 frames. ], batch size: 70, lr: 6.84e-03, grad_scale: 8.0 +2023-04-01 22:46:39,996 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-01 22:46:44,887 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4345, 2.3626, 1.7584, 1.4704, 2.1030, 1.2686, 1.2821, 1.8377], + device='cuda:0'), covar=tensor([0.0973, 0.0646, 0.0978, 0.0733, 0.0460, 0.1211, 0.0716, 0.0460], + device='cuda:0'), in_proj_covar=tensor([0.0289, 0.0302, 0.0323, 0.0244, 0.0239, 0.0318, 0.0284, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:46:47,879 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-01 22:46:52,463 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-01 22:46:52,781 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80870.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:47:00,405 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.450e+02 5.415e+02 6.686e+02 8.336e+02 1.819e+03, threshold=1.337e+03, percent-clipped=1.0 +2023-04-01 22:47:13,129 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4454, 1.5039, 1.8627, 1.6455, 2.8100, 2.3712, 2.9665, 1.2036], + device='cuda:0'), covar=tensor([0.2126, 0.3686, 0.2247, 0.1678, 0.1322, 0.1732, 0.1321, 0.3535], + device='cuda:0'), in_proj_covar=tensor([0.0497, 0.0587, 0.0617, 0.0440, 0.0599, 0.0501, 0.0651, 0.0500], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 22:47:40,322 INFO [train.py:903] (0/4) Epoch 12, batch 5800, loss[loss=0.2864, simple_loss=0.3445, pruned_loss=0.1141, over 13893.00 frames. ], tot_loss[loss=0.2303, simple_loss=0.3051, pruned_loss=0.07774, over 3822587.48 frames. ], batch size: 139, lr: 6.84e-03, grad_scale: 8.0 +2023-04-01 22:48:16,645 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-04-01 22:48:21,624 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80940.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:48:34,430 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8812, 1.8817, 2.2239, 1.9863, 3.0389, 2.4900, 3.0736, 2.0685], + device='cuda:0'), covar=tensor([0.1936, 0.3293, 0.2127, 0.1630, 0.1238, 0.1843, 0.1279, 0.3006], + device='cuda:0'), in_proj_covar=tensor([0.0496, 0.0586, 0.0615, 0.0440, 0.0597, 0.0500, 0.0649, 0.0499], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 22:48:41,885 INFO [train.py:903] (0/4) Epoch 12, batch 5850, loss[loss=0.2048, simple_loss=0.2911, pruned_loss=0.05921, over 19756.00 frames. ], tot_loss[loss=0.2314, simple_loss=0.306, pruned_loss=0.07843, over 3827241.77 frames. ], batch size: 54, lr: 6.84e-03, grad_scale: 8.0 +2023-04-01 22:48:57,974 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=80971.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:49:06,079 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.947e+02 5.387e+02 6.409e+02 7.183e+02 1.679e+03, threshold=1.282e+03, percent-clipped=1.0 +2023-04-01 22:49:10,724 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.2677, 3.8708, 2.6359, 3.5370, 1.0581, 3.5970, 3.6282, 3.6992], + device='cuda:0'), covar=tensor([0.0777, 0.1106, 0.1920, 0.0764, 0.3958, 0.0881, 0.0870, 0.1217], + device='cuda:0'), in_proj_covar=tensor([0.0443, 0.0372, 0.0442, 0.0320, 0.0379, 0.0375, 0.0366, 0.0397], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:49:18,793 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5674, 2.9564, 3.1195, 3.1737, 1.2071, 2.9248, 2.6681, 2.6372], + device='cuda:0'), covar=tensor([0.2895, 0.1748, 0.1496, 0.1808, 0.7214, 0.1894, 0.1318, 0.2804], + device='cuda:0'), in_proj_covar=tensor([0.0687, 0.0617, 0.0814, 0.0700, 0.0738, 0.0567, 0.0500, 0.0750], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 22:49:43,638 INFO [train.py:903] (0/4) Epoch 12, batch 5900, loss[loss=0.2426, simple_loss=0.3258, pruned_loss=0.07976, over 19676.00 frames. ], tot_loss[loss=0.2325, simple_loss=0.3067, pruned_loss=0.07913, over 3823556.51 frames. ], batch size: 60, lr: 6.83e-03, grad_scale: 8.0 +2023-04-01 22:49:47,131 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-01 22:49:55,116 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4506, 2.4140, 1.6828, 1.5466, 2.1857, 1.2518, 1.2269, 1.8489], + device='cuda:0'), covar=tensor([0.1055, 0.0616, 0.0987, 0.0690, 0.0440, 0.1162, 0.0746, 0.0472], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0307, 0.0328, 0.0249, 0.0242, 0.0323, 0.0287, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:49:55,137 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81016.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:50:09,727 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-01 22:50:25,032 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81041.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:50:35,662 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0714, 1.6862, 1.6077, 1.9781, 1.7084, 1.7415, 1.6170, 1.7901], + device='cuda:0'), covar=tensor([0.0802, 0.1259, 0.1345, 0.0892, 0.1202, 0.0476, 0.1183, 0.0719], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0358, 0.0298, 0.0242, 0.0304, 0.0246, 0.0284, 0.0246], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:50:47,169 INFO [train.py:903] (0/4) Epoch 12, batch 5950, loss[loss=0.2379, simple_loss=0.3136, pruned_loss=0.08108, over 19765.00 frames. ], tot_loss[loss=0.2319, simple_loss=0.3064, pruned_loss=0.07873, over 3828132.51 frames. ], batch size: 63, lr: 6.83e-03, grad_scale: 8.0 +2023-04-01 22:51:10,057 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.575e+02 5.377e+02 6.760e+02 8.757e+02 1.989e+03, threshold=1.352e+03, percent-clipped=8.0 +2023-04-01 22:51:36,957 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81098.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:51:49,665 INFO [train.py:903] (0/4) Epoch 12, batch 6000, loss[loss=0.2187, simple_loss=0.3026, pruned_loss=0.06742, over 19530.00 frames. ], tot_loss[loss=0.2314, simple_loss=0.3059, pruned_loss=0.07845, over 3825622.15 frames. ], batch size: 56, lr: 6.83e-03, grad_scale: 8.0 +2023-04-01 22:51:49,666 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 22:52:03,357 INFO [train.py:937] (0/4) Epoch 12, validation: loss=0.1765, simple_loss=0.2774, pruned_loss=0.03779, over 944034.00 frames. +2023-04-01 22:52:03,358 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18321MB +2023-04-01 22:52:25,557 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81126.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:52:34,805 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81134.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:52:35,837 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81135.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:52:57,877 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81151.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:53:05,457 INFO [train.py:903] (0/4) Epoch 12, batch 6050, loss[loss=0.2124, simple_loss=0.2882, pruned_loss=0.06828, over 19469.00 frames. ], tot_loss[loss=0.2298, simple_loss=0.3039, pruned_loss=0.07789, over 3815011.63 frames. ], batch size: 49, lr: 6.83e-03, grad_scale: 8.0 +2023-04-01 22:53:27,708 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.001e+02 5.041e+02 6.677e+02 8.260e+02 1.738e+03, threshold=1.335e+03, percent-clipped=2.0 +2023-04-01 22:53:52,518 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81196.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:53:57,464 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-04-01 22:54:05,930 INFO [train.py:903] (0/4) Epoch 12, batch 6100, loss[loss=0.2049, simple_loss=0.287, pruned_loss=0.06144, over 19590.00 frames. ], tot_loss[loss=0.2301, simple_loss=0.3044, pruned_loss=0.07795, over 3815192.68 frames. ], batch size: 52, lr: 6.83e-03, grad_scale: 8.0 +2023-04-01 22:54:11,795 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81213.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:54:20,933 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81221.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:54:28,710 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81227.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:54:57,826 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81250.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:55:00,301 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81252.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:55:03,881 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9114, 1.3443, 1.0557, 0.9638, 1.1506, 0.9450, 0.9808, 1.2725], + device='cuda:0'), covar=tensor([0.0524, 0.0640, 0.0945, 0.0548, 0.0501, 0.1103, 0.0456, 0.0379], + device='cuda:0'), in_proj_covar=tensor([0.0290, 0.0303, 0.0325, 0.0247, 0.0240, 0.0320, 0.0282, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:55:06,768 INFO [train.py:903] (0/4) Epoch 12, batch 6150, loss[loss=0.2182, simple_loss=0.2884, pruned_loss=0.07403, over 19604.00 frames. ], tot_loss[loss=0.2295, simple_loss=0.3039, pruned_loss=0.0775, over 3826798.63 frames. ], batch size: 50, lr: 6.82e-03, grad_scale: 8.0 +2023-04-01 22:55:31,812 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-04-01 22:55:33,032 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.393e+02 5.279e+02 6.402e+02 8.020e+02 2.167e+03, threshold=1.280e+03, percent-clipped=2.0 +2023-04-01 22:55:33,435 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4968, 1.0453, 1.1466, 1.3467, 1.0618, 1.3190, 1.1881, 1.3017], + device='cuda:0'), covar=tensor([0.0950, 0.1226, 0.1386, 0.0874, 0.1079, 0.0540, 0.1185, 0.0750], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0352, 0.0293, 0.0240, 0.0296, 0.0242, 0.0280, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:55:39,180 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-01 22:56:12,035 INFO [train.py:903] (0/4) Epoch 12, batch 6200, loss[loss=0.2807, simple_loss=0.3408, pruned_loss=0.1103, over 13324.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.3032, pruned_loss=0.07681, over 3822437.72 frames. ], batch size: 137, lr: 6.82e-03, grad_scale: 8.0 +2023-04-01 22:56:22,821 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81317.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:56:25,114 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81319.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:56:32,149 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9905, 3.5925, 2.3598, 3.2569, 0.8622, 3.3862, 3.4062, 3.4998], + device='cuda:0'), covar=tensor([0.0794, 0.1248, 0.2170, 0.0854, 0.3961, 0.0891, 0.0907, 0.1081], + device='cuda:0'), in_proj_covar=tensor([0.0443, 0.0373, 0.0442, 0.0320, 0.0380, 0.0376, 0.0365, 0.0396], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:56:57,108 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3559, 2.0675, 1.6180, 1.4194, 1.9229, 1.2746, 1.3465, 1.7999], + device='cuda:0'), covar=tensor([0.0970, 0.0655, 0.0879, 0.0681, 0.0467, 0.1098, 0.0634, 0.0430], + device='cuda:0'), in_proj_covar=tensor([0.0287, 0.0301, 0.0323, 0.0246, 0.0239, 0.0318, 0.0282, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:57:11,771 INFO [train.py:903] (0/4) Epoch 12, batch 6250, loss[loss=0.219, simple_loss=0.2958, pruned_loss=0.07116, over 18148.00 frames. ], tot_loss[loss=0.2283, simple_loss=0.3031, pruned_loss=0.07675, over 3819107.33 frames. ], batch size: 83, lr: 6.82e-03, grad_scale: 8.0 +2023-04-01 22:57:33,904 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.420e+02 5.051e+02 6.163e+02 7.297e+02 1.401e+03, threshold=1.233e+03, percent-clipped=3.0 +2023-04-01 22:57:44,021 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-01 22:58:13,175 INFO [train.py:903] (0/4) Epoch 12, batch 6300, loss[loss=0.2494, simple_loss=0.3229, pruned_loss=0.08792, over 19541.00 frames. ], tot_loss[loss=0.2277, simple_loss=0.3024, pruned_loss=0.07653, over 3818131.90 frames. ], batch size: 54, lr: 6.82e-03, grad_scale: 8.0 +2023-04-01 22:58:48,844 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8350, 1.2785, 1.4251, 1.5515, 3.2931, 1.0635, 2.2715, 3.7784], + device='cuda:0'), covar=tensor([0.0398, 0.2685, 0.2891, 0.1799, 0.0773, 0.2531, 0.1378, 0.0239], + device='cuda:0'), in_proj_covar=tensor([0.0358, 0.0339, 0.0351, 0.0318, 0.0345, 0.0327, 0.0335, 0.0358], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 22:59:14,521 INFO [train.py:903] (0/4) Epoch 12, batch 6350, loss[loss=0.2464, simple_loss=0.3226, pruned_loss=0.0851, over 19525.00 frames. ], tot_loss[loss=0.228, simple_loss=0.3028, pruned_loss=0.07662, over 3830178.32 frames. ], batch size: 56, lr: 6.81e-03, grad_scale: 8.0 +2023-04-01 22:59:28,523 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81469.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:59:39,342 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.788e+02 5.412e+02 6.997e+02 8.497e+02 1.750e+03, threshold=1.399e+03, percent-clipped=2.0 +2023-04-01 22:59:39,898 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8185, 1.9363, 2.0583, 2.4461, 1.7274, 2.3469, 2.3168, 1.9753], + device='cuda:0'), covar=tensor([0.3373, 0.2747, 0.1432, 0.1626, 0.2977, 0.1423, 0.3390, 0.2542], + device='cuda:0'), in_proj_covar=tensor([0.0803, 0.0826, 0.0650, 0.0889, 0.0788, 0.0715, 0.0786, 0.0717], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 22:59:40,742 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81478.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 22:59:59,507 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81494.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:00:14,198 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81506.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:00:16,025 INFO [train.py:903] (0/4) Epoch 12, batch 6400, loss[loss=0.2327, simple_loss=0.3088, pruned_loss=0.07836, over 19531.00 frames. ], tot_loss[loss=0.2291, simple_loss=0.3036, pruned_loss=0.07725, over 3832781.48 frames. ], batch size: 56, lr: 6.81e-03, grad_scale: 8.0 +2023-04-01 23:00:45,261 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81531.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:01:19,141 INFO [train.py:903] (0/4) Epoch 12, batch 6450, loss[loss=0.208, simple_loss=0.2903, pruned_loss=0.06291, over 19671.00 frames. ], tot_loss[loss=0.2295, simple_loss=0.3041, pruned_loss=0.0774, over 3835658.94 frames. ], batch size: 53, lr: 6.81e-03, grad_scale: 8.0 +2023-04-01 23:01:41,365 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.161e+02 5.839e+02 6.972e+02 8.326e+02 2.886e+03, threshold=1.394e+03, percent-clipped=3.0 +2023-04-01 23:02:03,327 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81593.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:02:06,515 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-01 23:02:20,371 INFO [train.py:903] (0/4) Epoch 12, batch 6500, loss[loss=0.2117, simple_loss=0.2909, pruned_loss=0.06626, over 19689.00 frames. ], tot_loss[loss=0.229, simple_loss=0.3035, pruned_loss=0.07724, over 3819171.77 frames. ], batch size: 59, lr: 6.81e-03, grad_scale: 8.0 +2023-04-01 23:02:27,383 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-01 23:03:22,584 INFO [train.py:903] (0/4) Epoch 12, batch 6550, loss[loss=0.2442, simple_loss=0.3198, pruned_loss=0.08425, over 19457.00 frames. ], tot_loss[loss=0.2282, simple_loss=0.3028, pruned_loss=0.07686, over 3819912.31 frames. ], batch size: 64, lr: 6.81e-03, grad_scale: 8.0 +2023-04-01 23:03:26,197 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81661.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:03:28,570 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=81663.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:03:47,197 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.342e+02 5.338e+02 6.617e+02 7.892e+02 1.534e+03, threshold=1.323e+03, percent-clipped=1.0 +2023-04-01 23:04:00,918 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=81688.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:04:08,904 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9192, 3.5588, 2.5030, 3.1744, 0.8754, 3.4148, 3.2755, 3.4797], + device='cuda:0'), covar=tensor([0.0797, 0.1085, 0.1906, 0.0874, 0.3806, 0.0834, 0.0942, 0.1072], + device='cuda:0'), in_proj_covar=tensor([0.0441, 0.0372, 0.0440, 0.0317, 0.0378, 0.0372, 0.0366, 0.0397], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:04:24,219 INFO [train.py:903] (0/4) Epoch 12, batch 6600, loss[loss=0.2163, simple_loss=0.2964, pruned_loss=0.06812, over 19611.00 frames. ], tot_loss[loss=0.2293, simple_loss=0.3037, pruned_loss=0.07749, over 3814554.26 frames. ], batch size: 61, lr: 6.80e-03, grad_scale: 8.0 +2023-04-01 23:04:56,138 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-04-01 23:05:14,743 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1689, 1.2551, 1.8350, 1.4618, 3.1604, 4.5656, 4.4749, 4.9769], + device='cuda:0'), covar=tensor([0.1655, 0.3714, 0.3177, 0.2131, 0.0492, 0.0179, 0.0163, 0.0132], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0299, 0.0328, 0.0252, 0.0220, 0.0163, 0.0206, 0.0208], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 23:05:26,102 INFO [train.py:903] (0/4) Epoch 12, batch 6650, loss[loss=0.2136, simple_loss=0.3014, pruned_loss=0.06291, over 19617.00 frames. ], tot_loss[loss=0.2293, simple_loss=0.3039, pruned_loss=0.07734, over 3816318.43 frames. ], batch size: 57, lr: 6.80e-03, grad_scale: 8.0 +2023-04-01 23:05:46,833 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81776.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:05:47,591 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.542e+02 5.145e+02 6.461e+02 8.134e+02 1.737e+03, threshold=1.292e+03, percent-clipped=3.0 +2023-04-01 23:05:49,127 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=81778.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:06:26,739 INFO [train.py:903] (0/4) Epoch 12, batch 6700, loss[loss=0.201, simple_loss=0.2681, pruned_loss=0.06695, over 19763.00 frames. ], tot_loss[loss=0.2299, simple_loss=0.3046, pruned_loss=0.0776, over 3821710.58 frames. ], batch size: 46, lr: 6.80e-03, grad_scale: 8.0 +2023-04-01 23:07:16,949 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=81849.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:07:26,538 INFO [train.py:903] (0/4) Epoch 12, batch 6750, loss[loss=0.27, simple_loss=0.3333, pruned_loss=0.1034, over 18173.00 frames. ], tot_loss[loss=0.2313, simple_loss=0.3053, pruned_loss=0.07861, over 3813604.32 frames. ], batch size: 83, lr: 6.80e-03, grad_scale: 4.0 +2023-04-01 23:07:37,174 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3395, 2.0827, 1.6233, 1.3439, 1.8892, 1.2217, 1.3579, 1.8572], + device='cuda:0'), covar=tensor([0.0800, 0.0660, 0.1025, 0.0718, 0.0490, 0.1191, 0.0551, 0.0359], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0306, 0.0325, 0.0246, 0.0240, 0.0325, 0.0288, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:07:45,296 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=81874.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:07:48,725 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7831, 1.4820, 1.4214, 1.7794, 1.5469, 1.6078, 1.4195, 1.6600], + device='cuda:0'), covar=tensor([0.0953, 0.1274, 0.1406, 0.0853, 0.1075, 0.0490, 0.1229, 0.0713], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0349, 0.0293, 0.0237, 0.0293, 0.0240, 0.0278, 0.0240], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:07:49,449 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.155e+02 6.276e+02 7.333e+02 1.082e+03 2.540e+03, threshold=1.467e+03, percent-clipped=11.0 +2023-04-01 23:08:23,231 INFO [train.py:903] (0/4) Epoch 12, batch 6800, loss[loss=0.2101, simple_loss=0.2877, pruned_loss=0.0662, over 19674.00 frames. ], tot_loss[loss=0.2303, simple_loss=0.3048, pruned_loss=0.07789, over 3820689.19 frames. ], batch size: 53, lr: 6.80e-03, grad_scale: 8.0 +2023-04-01 23:08:53,196 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-12.pt +2023-04-01 23:09:08,550 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-01 23:09:09,583 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-01 23:09:12,677 INFO [train.py:903] (0/4) Epoch 13, batch 0, loss[loss=0.2348, simple_loss=0.3162, pruned_loss=0.07675, over 19662.00 frames. ], tot_loss[loss=0.2348, simple_loss=0.3162, pruned_loss=0.07675, over 19662.00 frames. ], batch size: 55, lr: 6.53e-03, grad_scale: 8.0 +2023-04-01 23:09:12,678 INFO [train.py:928] (0/4) Computing validation loss +2023-04-01 23:09:23,576 INFO [train.py:937] (0/4) Epoch 13, validation: loss=0.176, simple_loss=0.2772, pruned_loss=0.03738, over 944034.00 frames. +2023-04-01 23:09:23,577 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18321MB +2023-04-01 23:09:35,400 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-01 23:10:14,566 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.117e+02 5.222e+02 6.740e+02 8.452e+02 3.268e+03, threshold=1.348e+03, percent-clipped=4.0 +2023-04-01 23:10:23,827 INFO [train.py:903] (0/4) Epoch 13, batch 50, loss[loss=0.2159, simple_loss=0.2924, pruned_loss=0.06971, over 19610.00 frames. ], tot_loss[loss=0.2298, simple_loss=0.3027, pruned_loss=0.07846, over 867891.07 frames. ], batch size: 61, lr: 6.53e-03, grad_scale: 8.0 +2023-04-01 23:10:39,647 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-82000.pt +2023-04-01 23:10:46,542 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82005.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:10:59,200 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-01 23:11:20,791 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82032.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:11:20,993 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82032.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:11:23,307 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82034.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:11:25,080 INFO [train.py:903] (0/4) Epoch 13, batch 100, loss[loss=0.2489, simple_loss=0.3178, pruned_loss=0.08996, over 19762.00 frames. ], tot_loss[loss=0.2317, simple_loss=0.3048, pruned_loss=0.07926, over 1519491.93 frames. ], batch size: 54, lr: 6.53e-03, grad_scale: 8.0 +2023-04-01 23:11:36,612 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-01 23:11:52,558 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82057.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:11:55,595 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82059.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:12:11,458 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82073.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:12:16,818 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.549e+02 4.805e+02 6.218e+02 7.513e+02 1.266e+03, threshold=1.244e+03, percent-clipped=0.0 +2023-04-01 23:12:25,782 INFO [train.py:903] (0/4) Epoch 13, batch 150, loss[loss=0.2024, simple_loss=0.279, pruned_loss=0.06285, over 19591.00 frames. ], tot_loss[loss=0.2283, simple_loss=0.3022, pruned_loss=0.07723, over 2037519.95 frames. ], batch size: 52, lr: 6.52e-03, grad_scale: 8.0 +2023-04-01 23:13:10,404 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0353, 2.7833, 2.0849, 2.5320, 0.9061, 2.6522, 2.6186, 2.6557], + device='cuda:0'), covar=tensor([0.1269, 0.1407, 0.1977, 0.0902, 0.3426, 0.1073, 0.1063, 0.1435], + device='cuda:0'), in_proj_covar=tensor([0.0443, 0.0370, 0.0444, 0.0316, 0.0380, 0.0374, 0.0367, 0.0399], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:13:23,620 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-01 23:13:24,765 INFO [train.py:903] (0/4) Epoch 13, batch 200, loss[loss=0.219, simple_loss=0.2985, pruned_loss=0.06969, over 19533.00 frames. ], tot_loss[loss=0.2294, simple_loss=0.3038, pruned_loss=0.07756, over 2440191.41 frames. ], batch size: 54, lr: 6.52e-03, grad_scale: 8.0 +2023-04-01 23:13:40,467 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82147.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:13:42,569 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5445, 2.2519, 2.1296, 2.7920, 2.6166, 2.1981, 2.0700, 2.6630], + device='cuda:0'), covar=tensor([0.0830, 0.1560, 0.1299, 0.0879, 0.1089, 0.0467, 0.1183, 0.0559], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0348, 0.0292, 0.0238, 0.0292, 0.0240, 0.0278, 0.0240], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:14:04,568 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9353, 1.9233, 1.7181, 1.4594, 1.3642, 1.4949, 0.3541, 0.7024], + device='cuda:0'), covar=tensor([0.0634, 0.0578, 0.0374, 0.0641, 0.1247, 0.0730, 0.1068, 0.1097], + device='cuda:0'), in_proj_covar=tensor([0.0341, 0.0332, 0.0333, 0.0359, 0.0431, 0.0354, 0.0314, 0.0326], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 23:14:14,395 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.039e+02 5.002e+02 5.972e+02 7.403e+02 2.257e+03, threshold=1.194e+03, percent-clipped=4.0 +2023-04-01 23:14:26,897 INFO [train.py:903] (0/4) Epoch 13, batch 250, loss[loss=0.1883, simple_loss=0.2592, pruned_loss=0.05865, over 19312.00 frames. ], tot_loss[loss=0.2294, simple_loss=0.3036, pruned_loss=0.07756, over 2750535.92 frames. ], batch size: 44, lr: 6.52e-03, grad_scale: 8.0 +2023-04-01 23:15:25,416 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-04-01 23:15:26,892 INFO [train.py:903] (0/4) Epoch 13, batch 300, loss[loss=0.2081, simple_loss=0.2792, pruned_loss=0.06851, over 18235.00 frames. ], tot_loss[loss=0.2286, simple_loss=0.3029, pruned_loss=0.07712, over 2985816.75 frames. ], batch size: 40, lr: 6.52e-03, grad_scale: 8.0 +2023-04-01 23:16:18,766 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.276e+02 5.753e+02 6.815e+02 9.164e+02 1.837e+03, threshold=1.363e+03, percent-clipped=5.0 +2023-04-01 23:16:28,129 INFO [train.py:903] (0/4) Epoch 13, batch 350, loss[loss=0.2417, simple_loss=0.3126, pruned_loss=0.08538, over 19587.00 frames. ], tot_loss[loss=0.2282, simple_loss=0.3024, pruned_loss=0.07696, over 3167534.45 frames. ], batch size: 61, lr: 6.52e-03, grad_scale: 8.0 +2023-04-01 23:16:30,469 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-01 23:17:28,525 INFO [train.py:903] (0/4) Epoch 13, batch 400, loss[loss=0.2161, simple_loss=0.3025, pruned_loss=0.06489, over 19614.00 frames. ], tot_loss[loss=0.2279, simple_loss=0.302, pruned_loss=0.07684, over 3313157.18 frames. ], batch size: 57, lr: 6.51e-03, grad_scale: 8.0 +2023-04-01 23:17:47,110 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82349.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:18:04,816 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82364.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:18:10,378 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0030, 3.6263, 2.5667, 3.2409, 0.9858, 3.4812, 3.4114, 3.5108], + device='cuda:0'), covar=tensor([0.0788, 0.1147, 0.1830, 0.0862, 0.3614, 0.0812, 0.0884, 0.1066], + device='cuda:0'), in_proj_covar=tensor([0.0440, 0.0370, 0.0442, 0.0318, 0.0378, 0.0371, 0.0365, 0.0398], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:18:21,965 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.254e+02 5.325e+02 6.166e+02 7.720e+02 2.046e+03, threshold=1.233e+03, percent-clipped=4.0 +2023-04-01 23:18:28,196 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6344, 1.3164, 1.3261, 2.1675, 1.8358, 1.9927, 2.1367, 1.9285], + device='cuda:0'), covar=tensor([0.0860, 0.1064, 0.1133, 0.0829, 0.0841, 0.0754, 0.0845, 0.0633], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0226, 0.0226, 0.0245, 0.0234, 0.0211, 0.0193, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 23:18:31,252 INFO [train.py:903] (0/4) Epoch 13, batch 450, loss[loss=0.2072, simple_loss=0.276, pruned_loss=0.06922, over 19742.00 frames. ], tot_loss[loss=0.2272, simple_loss=0.3017, pruned_loss=0.07633, over 3432375.08 frames. ], batch size: 46, lr: 6.51e-03, grad_scale: 8.0 +2023-04-01 23:18:53,559 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82403.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:19:04,667 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-01 23:19:04,702 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-01 23:19:09,580 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82417.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:19:23,716 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82428.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:19:25,766 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8764, 1.2916, 1.4024, 1.6326, 3.3935, 1.0616, 2.5156, 3.7394], + device='cuda:0'), covar=tensor([0.0414, 0.2663, 0.2807, 0.1795, 0.0714, 0.2467, 0.1090, 0.0273], + device='cuda:0'), in_proj_covar=tensor([0.0362, 0.0344, 0.0356, 0.0324, 0.0350, 0.0331, 0.0340, 0.0365], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:19:33,820 INFO [train.py:903] (0/4) Epoch 13, batch 500, loss[loss=0.2465, simple_loss=0.3205, pruned_loss=0.08624, over 19672.00 frames. ], tot_loss[loss=0.2256, simple_loss=0.3005, pruned_loss=0.07538, over 3540059.22 frames. ], batch size: 55, lr: 6.51e-03, grad_scale: 8.0 +2023-04-01 23:20:06,459 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82464.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:20:27,379 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.531e+02 5.153e+02 6.569e+02 8.401e+02 1.477e+03, threshold=1.314e+03, percent-clipped=3.0 +2023-04-01 23:20:35,266 INFO [train.py:903] (0/4) Epoch 13, batch 550, loss[loss=0.1951, simple_loss=0.2667, pruned_loss=0.06178, over 19752.00 frames. ], tot_loss[loss=0.2269, simple_loss=0.3019, pruned_loss=0.07598, over 3606078.18 frames. ], batch size: 45, lr: 6.51e-03, grad_scale: 8.0 +2023-04-01 23:21:05,879 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82511.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 23:21:30,821 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82532.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:21:35,045 INFO [train.py:903] (0/4) Epoch 13, batch 600, loss[loss=0.228, simple_loss=0.311, pruned_loss=0.07252, over 19549.00 frames. ], tot_loss[loss=0.2272, simple_loss=0.3023, pruned_loss=0.07605, over 3664532.25 frames. ], batch size: 56, lr: 6.51e-03, grad_scale: 8.0 +2023-04-01 23:22:17,360 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-01 23:22:20,658 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6444, 4.1860, 2.8553, 3.6959, 0.8830, 4.0430, 4.0151, 4.0827], + device='cuda:0'), covar=tensor([0.0571, 0.1026, 0.1746, 0.0782, 0.3980, 0.0691, 0.0800, 0.0857], + device='cuda:0'), in_proj_covar=tensor([0.0442, 0.0372, 0.0442, 0.0319, 0.0381, 0.0374, 0.0367, 0.0401], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:22:28,772 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.675e+02 5.313e+02 6.751e+02 8.249e+02 1.619e+03, threshold=1.350e+03, percent-clipped=3.0 +2023-04-01 23:22:36,729 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.74 vs. limit=5.0 +2023-04-01 23:22:36,936 INFO [train.py:903] (0/4) Epoch 13, batch 650, loss[loss=0.2297, simple_loss=0.3094, pruned_loss=0.07501, over 19736.00 frames. ], tot_loss[loss=0.2286, simple_loss=0.3036, pruned_loss=0.07681, over 3706810.11 frames. ], batch size: 63, lr: 6.50e-03, grad_scale: 8.0 +2023-04-01 23:22:38,442 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82587.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:23:40,973 INFO [train.py:903] (0/4) Epoch 13, batch 700, loss[loss=0.2698, simple_loss=0.3381, pruned_loss=0.1008, over 19591.00 frames. ], tot_loss[loss=0.2267, simple_loss=0.3016, pruned_loss=0.07592, over 3733671.29 frames. ], batch size: 57, lr: 6.50e-03, grad_scale: 8.0 +2023-04-01 23:24:21,261 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3224, 2.1452, 1.6181, 1.3913, 2.0046, 1.2251, 1.2135, 1.8269], + device='cuda:0'), covar=tensor([0.0877, 0.0638, 0.0943, 0.0716, 0.0406, 0.1086, 0.0662, 0.0396], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0307, 0.0329, 0.0248, 0.0242, 0.0323, 0.0289, 0.0267], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:24:36,357 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.507e+02 5.358e+02 6.603e+02 8.553e+02 2.977e+03, threshold=1.321e+03, percent-clipped=4.0 +2023-04-01 23:24:44,575 INFO [train.py:903] (0/4) Epoch 13, batch 750, loss[loss=0.2709, simple_loss=0.3364, pruned_loss=0.1027, over 18297.00 frames. ], tot_loss[loss=0.2277, simple_loss=0.3024, pruned_loss=0.0765, over 3743761.53 frames. ], batch size: 84, lr: 6.50e-03, grad_scale: 8.0 +2023-04-01 23:25:10,154 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82708.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:25:28,756 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82720.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:25:33,509 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6981, 1.4886, 1.4976, 1.4285, 3.2612, 1.0439, 2.3385, 3.7044], + device='cuda:0'), covar=tensor([0.0424, 0.2473, 0.2588, 0.1873, 0.0670, 0.2503, 0.1236, 0.0222], + device='cuda:0'), in_proj_covar=tensor([0.0363, 0.0344, 0.0356, 0.0324, 0.0353, 0.0332, 0.0342, 0.0366], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:25:47,341 INFO [train.py:903] (0/4) Epoch 13, batch 800, loss[loss=0.2725, simple_loss=0.3432, pruned_loss=0.1009, over 19706.00 frames. ], tot_loss[loss=0.2285, simple_loss=0.3031, pruned_loss=0.07694, over 3766919.09 frames. ], batch size: 59, lr: 6.50e-03, grad_scale: 8.0 +2023-04-01 23:25:58,387 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82745.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:26:01,436 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-01 23:26:20,720 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4338, 2.2662, 1.6668, 1.4441, 2.0757, 1.2657, 1.2746, 1.8984], + device='cuda:0'), covar=tensor([0.0965, 0.0630, 0.0993, 0.0754, 0.0442, 0.1137, 0.0742, 0.0449], + device='cuda:0'), in_proj_covar=tensor([0.0294, 0.0307, 0.0327, 0.0247, 0.0240, 0.0321, 0.0287, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:26:39,178 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1005, 1.3228, 1.4788, 1.4250, 2.6954, 1.1003, 1.9982, 2.9933], + device='cuda:0'), covar=tensor([0.0491, 0.2525, 0.2531, 0.1642, 0.0759, 0.2162, 0.1115, 0.0331], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0346, 0.0358, 0.0326, 0.0354, 0.0333, 0.0344, 0.0367], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:26:41,539 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7409, 4.1734, 4.4270, 4.4398, 1.7215, 4.1788, 3.6520, 4.0880], + device='cuda:0'), covar=tensor([0.1377, 0.0750, 0.0577, 0.0587, 0.5075, 0.0634, 0.0579, 0.1051], + device='cuda:0'), in_proj_covar=tensor([0.0689, 0.0613, 0.0815, 0.0698, 0.0737, 0.0569, 0.0495, 0.0745], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-01 23:26:42,349 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.821e+02 5.442e+02 6.436e+02 7.821e+02 1.140e+03, threshold=1.287e+03, percent-clipped=0.0 +2023-04-01 23:26:50,570 INFO [train.py:903] (0/4) Epoch 13, batch 850, loss[loss=0.2212, simple_loss=0.2899, pruned_loss=0.07627, over 19305.00 frames. ], tot_loss[loss=0.23, simple_loss=0.3045, pruned_loss=0.07776, over 3782679.06 frames. ], batch size: 44, lr: 6.50e-03, grad_scale: 8.0 +2023-04-01 23:26:53,185 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=82788.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:27:26,991 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=82813.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:27:38,332 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82823.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:27:44,763 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-01 23:27:52,714 INFO [train.py:903] (0/4) Epoch 13, batch 900, loss[loss=0.2931, simple_loss=0.3549, pruned_loss=0.1156, over 18919.00 frames. ], tot_loss[loss=0.2301, simple_loss=0.3045, pruned_loss=0.07785, over 3789420.96 frames. ], batch size: 74, lr: 6.49e-03, grad_scale: 8.0 +2023-04-01 23:28:19,289 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82855.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 23:28:47,653 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.460e+02 5.906e+02 6.958e+02 9.103e+02 2.196e+03, threshold=1.392e+03, percent-clipped=5.0 +2023-04-01 23:28:59,852 INFO [train.py:903] (0/4) Epoch 13, batch 950, loss[loss=0.2444, simple_loss=0.3265, pruned_loss=0.08112, over 18738.00 frames. ], tot_loss[loss=0.2313, simple_loss=0.3058, pruned_loss=0.07845, over 3780032.44 frames. ], batch size: 74, lr: 6.49e-03, grad_scale: 8.0 +2023-04-01 23:29:04,322 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-01 23:29:55,917 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=82931.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:30:01,532 INFO [train.py:903] (0/4) Epoch 13, batch 1000, loss[loss=0.2584, simple_loss=0.3344, pruned_loss=0.09121, over 18825.00 frames. ], tot_loss[loss=0.23, simple_loss=0.3046, pruned_loss=0.07768, over 3780933.19 frames. ], batch size: 74, lr: 6.49e-03, grad_scale: 8.0 +2023-04-01 23:30:14,230 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.10 vs. limit=2.0 +2023-04-01 23:30:44,807 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=82970.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 23:30:48,284 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5408, 2.2233, 2.2257, 2.8567, 2.5495, 2.3730, 2.2938, 2.7728], + device='cuda:0'), covar=tensor([0.0840, 0.1649, 0.1277, 0.0904, 0.1181, 0.0432, 0.1086, 0.0571], + device='cuda:0'), in_proj_covar=tensor([0.0255, 0.0350, 0.0296, 0.0238, 0.0296, 0.0241, 0.0280, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:30:48,368 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4818, 1.5385, 2.0748, 1.7130, 3.3296, 2.6442, 3.5900, 1.5980], + device='cuda:0'), covar=tensor([0.2172, 0.3806, 0.2384, 0.1689, 0.1315, 0.1735, 0.1434, 0.3563], + device='cuda:0'), in_proj_covar=tensor([0.0491, 0.0582, 0.0613, 0.0441, 0.0594, 0.0496, 0.0643, 0.0495], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 23:30:51,730 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5479, 1.0979, 1.3561, 1.1177, 2.2018, 0.9118, 2.0998, 2.4018], + device='cuda:0'), covar=tensor([0.0674, 0.2748, 0.2673, 0.1681, 0.0904, 0.2056, 0.0866, 0.0503], + device='cuda:0'), in_proj_covar=tensor([0.0366, 0.0346, 0.0358, 0.0326, 0.0355, 0.0335, 0.0342, 0.0368], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:30:52,621 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-01 23:30:54,574 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.083e+02 5.161e+02 6.395e+02 8.326e+02 2.115e+03, threshold=1.279e+03, percent-clipped=2.0 +2023-04-01 23:30:57,042 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=82981.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:31:02,716 INFO [train.py:903] (0/4) Epoch 13, batch 1050, loss[loss=0.2238, simple_loss=0.2875, pruned_loss=0.08008, over 19764.00 frames. ], tot_loss[loss=0.2309, simple_loss=0.3053, pruned_loss=0.07831, over 3789460.67 frames. ], batch size: 48, lr: 6.49e-03, grad_scale: 8.0 +2023-04-01 23:31:27,833 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1605, 2.0737, 1.7946, 1.7172, 1.5136, 1.6977, 0.3746, 1.0239], + device='cuda:0'), covar=tensor([0.0450, 0.0441, 0.0370, 0.0529, 0.1007, 0.0604, 0.1018, 0.0827], + device='cuda:0'), in_proj_covar=tensor([0.0341, 0.0333, 0.0333, 0.0360, 0.0429, 0.0356, 0.0316, 0.0325], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 23:31:34,341 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-01 23:32:04,664 INFO [train.py:903] (0/4) Epoch 13, batch 1100, loss[loss=0.2225, simple_loss=0.2942, pruned_loss=0.07542, over 19487.00 frames. ], tot_loss[loss=0.2297, simple_loss=0.3046, pruned_loss=0.07746, over 3810299.60 frames. ], batch size: 49, lr: 6.49e-03, grad_scale: 8.0 +2023-04-01 23:32:19,489 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83046.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:32:30,388 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1822, 1.3862, 1.6078, 1.2290, 2.7093, 3.4944, 3.2732, 3.7493], + device='cuda:0'), covar=tensor([0.1596, 0.3296, 0.3136, 0.2191, 0.0528, 0.0188, 0.0202, 0.0212], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0300, 0.0329, 0.0253, 0.0221, 0.0163, 0.0207, 0.0212], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 23:32:57,833 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.212e+02 5.014e+02 6.117e+02 7.878e+02 1.226e+03, threshold=1.223e+03, percent-clipped=0.0 +2023-04-01 23:32:58,253 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83079.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:33:08,489 INFO [train.py:903] (0/4) Epoch 13, batch 1150, loss[loss=0.2243, simple_loss=0.2917, pruned_loss=0.07843, over 19097.00 frames. ], tot_loss[loss=0.2299, simple_loss=0.3047, pruned_loss=0.07753, over 3809121.16 frames. ], batch size: 42, lr: 6.48e-03, grad_scale: 8.0 +2023-04-01 23:33:30,516 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83104.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:33:40,674 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1941, 1.8145, 1.4187, 1.2048, 1.6091, 1.0786, 1.1796, 1.6600], + device='cuda:0'), covar=tensor([0.0681, 0.0718, 0.1010, 0.0654, 0.0460, 0.1210, 0.0542, 0.0368], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0311, 0.0331, 0.0248, 0.0241, 0.0327, 0.0291, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:34:10,939 INFO [train.py:903] (0/4) Epoch 13, batch 1200, loss[loss=0.1868, simple_loss=0.266, pruned_loss=0.05386, over 19596.00 frames. ], tot_loss[loss=0.2287, simple_loss=0.3037, pruned_loss=0.07689, over 3819729.20 frames. ], batch size: 52, lr: 6.48e-03, grad_scale: 8.0 +2023-04-01 23:34:40,325 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-01 23:35:06,465 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.490e+02 6.101e+02 7.395e+02 1.032e+03 1.939e+03, threshold=1.479e+03, percent-clipped=13.0 +2023-04-01 23:35:07,336 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.82 vs. limit=5.0 +2023-04-01 23:35:12,376 INFO [train.py:903] (0/4) Epoch 13, batch 1250, loss[loss=0.2338, simple_loss=0.304, pruned_loss=0.08186, over 19141.00 frames. ], tot_loss[loss=0.2286, simple_loss=0.3035, pruned_loss=0.07685, over 3810176.15 frames. ], batch size: 42, lr: 6.48e-03, grad_scale: 4.0 +2023-04-01 23:35:48,373 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83214.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:36:02,476 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83226.0, num_to_drop=1, layers_to_drop={1} +2023-04-01 23:36:13,263 INFO [train.py:903] (0/4) Epoch 13, batch 1300, loss[loss=0.2379, simple_loss=0.314, pruned_loss=0.08086, over 19544.00 frames. ], tot_loss[loss=0.2276, simple_loss=0.3027, pruned_loss=0.07625, over 3808658.92 frames. ], batch size: 54, lr: 6.48e-03, grad_scale: 4.0 +2023-04-01 23:36:33,713 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83251.0, num_to_drop=1, layers_to_drop={0} +2023-04-01 23:37:05,206 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5357, 2.2524, 2.2244, 2.8357, 1.9379, 2.9456, 2.7417, 2.7432], + device='cuda:0'), covar=tensor([0.0628, 0.0728, 0.0810, 0.0789, 0.0918, 0.0567, 0.0801, 0.0526], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0222, 0.0222, 0.0243, 0.0233, 0.0210, 0.0193, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 23:37:08,171 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.019e+02 5.042e+02 6.195e+02 7.677e+02 1.204e+03, threshold=1.239e+03, percent-clipped=0.0 +2023-04-01 23:37:17,151 INFO [train.py:903] (0/4) Epoch 13, batch 1350, loss[loss=0.2328, simple_loss=0.2916, pruned_loss=0.08695, over 19733.00 frames. ], tot_loss[loss=0.2285, simple_loss=0.3033, pruned_loss=0.07681, over 3811265.08 frames. ], batch size: 46, lr: 6.48e-03, grad_scale: 4.0 +2023-04-01 23:37:37,997 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83302.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:37:38,117 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83302.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:37:39,612 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.09 vs. limit=5.0 +2023-04-01 23:37:42,732 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4333, 1.3442, 1.2948, 1.6931, 1.3690, 1.7152, 1.7636, 1.5769], + device='cuda:0'), covar=tensor([0.0928, 0.1019, 0.1120, 0.0804, 0.0865, 0.0747, 0.0816, 0.0699], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0222, 0.0222, 0.0243, 0.0232, 0.0210, 0.0193, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-01 23:38:06,338 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=83325.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:38:09,935 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83327.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:38:21,070 INFO [train.py:903] (0/4) Epoch 13, batch 1400, loss[loss=0.2348, simple_loss=0.3106, pruned_loss=0.07943, over 19669.00 frames. ], tot_loss[loss=0.2286, simple_loss=0.3036, pruned_loss=0.07681, over 3823593.96 frames. ], batch size: 55, lr: 6.47e-03, grad_scale: 4.0 +2023-04-01 23:38:38,125 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-04-01 23:39:16,728 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.257e+02 5.533e+02 6.494e+02 7.603e+02 1.656e+03, threshold=1.299e+03, percent-clipped=2.0 +2023-04-01 23:39:20,314 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-01 23:39:22,625 INFO [train.py:903] (0/4) Epoch 13, batch 1450, loss[loss=0.2077, simple_loss=0.2768, pruned_loss=0.0693, over 19395.00 frames. ], tot_loss[loss=0.2309, simple_loss=0.3056, pruned_loss=0.0781, over 3815622.28 frames. ], batch size: 47, lr: 6.47e-03, grad_scale: 4.0 +2023-04-01 23:40:24,405 INFO [train.py:903] (0/4) Epoch 13, batch 1500, loss[loss=0.216, simple_loss=0.3063, pruned_loss=0.06288, over 19699.00 frames. ], tot_loss[loss=0.2296, simple_loss=0.3047, pruned_loss=0.07727, over 3810488.89 frames. ], batch size: 59, lr: 6.47e-03, grad_scale: 4.0 +2023-04-01 23:40:29,216 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83440.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:40:44,776 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-04-01 23:41:19,826 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.552e+02 5.496e+02 6.437e+02 7.955e+02 2.023e+03, threshold=1.287e+03, percent-clipped=5.0 +2023-04-01 23:41:26,536 INFO [train.py:903] (0/4) Epoch 13, batch 1550, loss[loss=0.2024, simple_loss=0.2856, pruned_loss=0.05956, over 19534.00 frames. ], tot_loss[loss=0.2287, simple_loss=0.3038, pruned_loss=0.07678, over 3814978.64 frames. ], batch size: 56, lr: 6.47e-03, grad_scale: 4.0 +2023-04-01 23:42:30,076 INFO [train.py:903] (0/4) Epoch 13, batch 1600, loss[loss=0.2403, simple_loss=0.3171, pruned_loss=0.08173, over 19550.00 frames. ], tot_loss[loss=0.229, simple_loss=0.3041, pruned_loss=0.07691, over 3811584.62 frames. ], batch size: 56, lr: 6.47e-03, grad_scale: 8.0 +2023-04-01 23:42:53,317 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-01 23:42:55,756 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=83558.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:43:25,328 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.823e+02 5.445e+02 6.386e+02 7.908e+02 1.256e+03, threshold=1.277e+03, percent-clipped=0.0 +2023-04-01 23:43:31,085 INFO [train.py:903] (0/4) Epoch 13, batch 1650, loss[loss=0.2136, simple_loss=0.281, pruned_loss=0.07312, over 19422.00 frames. ], tot_loss[loss=0.2293, simple_loss=0.3041, pruned_loss=0.0772, over 3831732.40 frames. ], batch size: 48, lr: 6.47e-03, grad_scale: 8.0 +2023-04-01 23:43:44,217 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3004, 1.5204, 1.9368, 1.6513, 3.1420, 2.5889, 3.4084, 1.5968], + device='cuda:0'), covar=tensor([0.2298, 0.3833, 0.2341, 0.1682, 0.1390, 0.1783, 0.1546, 0.3411], + device='cuda:0'), in_proj_covar=tensor([0.0495, 0.0584, 0.0618, 0.0440, 0.0598, 0.0495, 0.0649, 0.0501], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 23:44:33,584 INFO [train.py:903] (0/4) Epoch 13, batch 1700, loss[loss=0.1925, simple_loss=0.2651, pruned_loss=0.05994, over 18642.00 frames. ], tot_loss[loss=0.2291, simple_loss=0.3036, pruned_loss=0.07726, over 3837474.46 frames. ], batch size: 41, lr: 6.46e-03, grad_scale: 8.0 +2023-04-01 23:44:45,676 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=83646.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:45:15,476 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-01 23:45:19,397 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83673.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:45:27,804 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.234e+02 5.386e+02 6.785e+02 9.131e+02 1.645e+03, threshold=1.357e+03, percent-clipped=5.0 +2023-04-01 23:45:33,657 INFO [train.py:903] (0/4) Epoch 13, batch 1750, loss[loss=0.1999, simple_loss=0.2698, pruned_loss=0.06494, over 19731.00 frames. ], tot_loss[loss=0.2273, simple_loss=0.3019, pruned_loss=0.07639, over 3838854.03 frames. ], batch size: 45, lr: 6.46e-03, grad_scale: 8.0 +2023-04-01 23:45:48,921 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83696.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:46:19,378 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83721.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:46:37,929 INFO [train.py:903] (0/4) Epoch 13, batch 1800, loss[loss=0.2578, simple_loss=0.3269, pruned_loss=0.09432, over 19667.00 frames. ], tot_loss[loss=0.228, simple_loss=0.3029, pruned_loss=0.07659, over 3835038.94 frames. ], batch size: 60, lr: 6.46e-03, grad_scale: 8.0 +2023-04-01 23:47:01,559 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8444, 0.8968, 0.8269, 0.7217, 0.7188, 0.7661, 0.0808, 0.2782], + device='cuda:0'), covar=tensor([0.0365, 0.0377, 0.0242, 0.0324, 0.0692, 0.0347, 0.0774, 0.0618], + device='cuda:0'), in_proj_covar=tensor([0.0340, 0.0334, 0.0333, 0.0358, 0.0426, 0.0356, 0.0316, 0.0324], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 23:47:08,530 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=83761.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:47:29,808 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83778.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:47:32,958 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.146e+02 5.542e+02 6.939e+02 8.095e+02 2.139e+03, threshold=1.388e+03, percent-clipped=3.0 +2023-04-01 23:47:35,255 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-01 23:47:39,805 INFO [train.py:903] (0/4) Epoch 13, batch 1850, loss[loss=0.2325, simple_loss=0.3136, pruned_loss=0.07575, over 19672.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.3035, pruned_loss=0.07714, over 3822749.91 frames. ], batch size: 59, lr: 6.46e-03, grad_scale: 8.0 +2023-04-01 23:47:42,529 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2490, 1.3954, 1.7775, 1.3920, 2.7050, 3.3599, 3.1715, 3.5111], + device='cuda:0'), covar=tensor([0.1545, 0.3336, 0.2984, 0.2175, 0.0596, 0.0231, 0.0210, 0.0231], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0299, 0.0327, 0.0250, 0.0218, 0.0161, 0.0206, 0.0210], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 23:48:11,504 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-01 23:48:17,624 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0763, 1.3517, 1.8372, 1.3822, 3.0124, 4.6718, 4.5492, 5.0195], + device='cuda:0'), covar=tensor([0.1657, 0.3440, 0.3070, 0.2073, 0.0533, 0.0156, 0.0164, 0.0121], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0299, 0.0328, 0.0252, 0.0219, 0.0162, 0.0206, 0.0211], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-01 23:48:38,904 INFO [train.py:903] (0/4) Epoch 13, batch 1900, loss[loss=0.241, simple_loss=0.3058, pruned_loss=0.08808, over 19722.00 frames. ], tot_loss[loss=0.229, simple_loss=0.3037, pruned_loss=0.07712, over 3830325.84 frames. ], batch size: 51, lr: 6.46e-03, grad_scale: 8.0 +2023-04-01 23:48:55,303 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-01 23:49:00,751 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-01 23:49:23,812 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-01 23:49:32,948 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.547e+02 5.453e+02 6.640e+02 7.751e+02 1.927e+03, threshold=1.328e+03, percent-clipped=4.0 +2023-04-01 23:49:38,629 INFO [train.py:903] (0/4) Epoch 13, batch 1950, loss[loss=0.2422, simple_loss=0.3119, pruned_loss=0.08624, over 19832.00 frames. ], tot_loss[loss=0.2275, simple_loss=0.3025, pruned_loss=0.07625, over 3834771.46 frames. ], batch size: 52, lr: 6.45e-03, grad_scale: 8.0 +2023-04-01 23:50:31,098 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=83929.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:50:40,980 INFO [train.py:903] (0/4) Epoch 13, batch 2000, loss[loss=0.2148, simple_loss=0.2983, pruned_loss=0.06564, over 19427.00 frames. ], tot_loss[loss=0.229, simple_loss=0.3039, pruned_loss=0.0771, over 3831024.15 frames. ], batch size: 66, lr: 6.45e-03, grad_scale: 8.0 +2023-04-01 23:51:02,440 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=83954.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:51:04,038 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.43 vs. limit=5.0 +2023-04-01 23:51:17,802 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=83967.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:51:36,104 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.295e+02 5.067e+02 6.527e+02 8.467e+02 1.955e+03, threshold=1.305e+03, percent-clipped=7.0 +2023-04-01 23:51:38,314 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-01 23:51:42,654 INFO [train.py:903] (0/4) Epoch 13, batch 2050, loss[loss=0.2176, simple_loss=0.3006, pruned_loss=0.06732, over 19666.00 frames. ], tot_loss[loss=0.2287, simple_loss=0.3032, pruned_loss=0.07708, over 3820085.20 frames. ], batch size: 58, lr: 6.45e-03, grad_scale: 8.0 +2023-04-01 23:51:56,810 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-01 23:51:57,787 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-01 23:51:59,003 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-84000.pt +2023-04-01 23:52:03,629 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84003.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:52:21,314 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-01 23:52:22,833 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84017.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:52:44,638 INFO [train.py:903] (0/4) Epoch 13, batch 2100, loss[loss=0.2384, simple_loss=0.3155, pruned_loss=0.08069, over 19536.00 frames. ], tot_loss[loss=0.2287, simple_loss=0.3028, pruned_loss=0.0773, over 3823306.12 frames. ], batch size: 64, lr: 6.45e-03, grad_scale: 8.0 +2023-04-01 23:52:52,266 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84042.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:52:58,248 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7930, 2.3009, 2.3094, 2.8211, 2.7431, 2.4015, 2.2366, 2.8781], + device='cuda:0'), covar=tensor([0.0735, 0.1519, 0.1254, 0.0952, 0.1152, 0.0439, 0.1061, 0.0511], + device='cuda:0'), in_proj_covar=tensor([0.0252, 0.0345, 0.0292, 0.0237, 0.0293, 0.0242, 0.0278, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-01 23:53:14,853 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-01 23:53:28,758 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-01 23:53:36,214 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-01 23:53:39,575 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.864e+02 5.545e+02 6.946e+02 9.457e+02 3.064e+03, threshold=1.389e+03, percent-clipped=12.0 +2023-04-01 23:53:45,272 INFO [train.py:903] (0/4) Epoch 13, batch 2150, loss[loss=0.252, simple_loss=0.3222, pruned_loss=0.09086, over 19582.00 frames. ], tot_loss[loss=0.229, simple_loss=0.3034, pruned_loss=0.07731, over 3827549.70 frames. ], batch size: 52, lr: 6.45e-03, grad_scale: 8.0 +2023-04-01 23:54:23,367 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84115.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:54:31,335 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84122.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:54:49,422 INFO [train.py:903] (0/4) Epoch 13, batch 2200, loss[loss=0.2783, simple_loss=0.3503, pruned_loss=0.1031, over 19790.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.3032, pruned_loss=0.07677, over 3832373.44 frames. ], batch size: 56, lr: 6.44e-03, grad_scale: 8.0 +2023-04-01 23:55:44,487 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.166e+02 5.936e+02 7.647e+02 9.699e+02 2.302e+03, threshold=1.529e+03, percent-clipped=8.0 +2023-04-01 23:55:50,238 INFO [train.py:903] (0/4) Epoch 13, batch 2250, loss[loss=0.2149, simple_loss=0.3045, pruned_loss=0.06269, over 19671.00 frames. ], tot_loss[loss=0.229, simple_loss=0.3036, pruned_loss=0.07721, over 3827620.97 frames. ], batch size: 58, lr: 6.44e-03, grad_scale: 8.0 +2023-04-01 23:56:09,019 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4062, 1.5983, 2.1676, 1.6853, 2.9254, 2.3086, 2.9633, 1.5773], + device='cuda:0'), covar=tensor([0.2375, 0.4019, 0.2244, 0.1897, 0.1612, 0.2200, 0.2012, 0.3786], + device='cuda:0'), in_proj_covar=tensor([0.0497, 0.0589, 0.0624, 0.0444, 0.0603, 0.0499, 0.0651, 0.0505], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-01 23:56:18,362 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.59 vs. limit=2.0 +2023-04-01 23:56:28,831 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84217.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:56:51,972 INFO [train.py:903] (0/4) Epoch 13, batch 2300, loss[loss=0.196, simple_loss=0.2697, pruned_loss=0.06111, over 19383.00 frames. ], tot_loss[loss=0.2285, simple_loss=0.3034, pruned_loss=0.07681, over 3828838.17 frames. ], batch size: 47, lr: 6.44e-03, grad_scale: 8.0 +2023-04-01 23:56:53,470 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84237.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:57:05,845 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-01 23:57:15,052 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84254.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:57:47,143 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.862e+02 5.192e+02 6.483e+02 8.696e+02 2.103e+03, threshold=1.297e+03, percent-clipped=4.0 +2023-04-01 23:57:52,877 INFO [train.py:903] (0/4) Epoch 13, batch 2350, loss[loss=0.2143, simple_loss=0.2967, pruned_loss=0.0659, over 18158.00 frames. ], tot_loss[loss=0.2275, simple_loss=0.3025, pruned_loss=0.07619, over 3838668.69 frames. ], batch size: 83, lr: 6.44e-03, grad_scale: 8.0 +2023-04-01 23:57:53,516 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-04-01 23:58:25,978 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84311.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:58:37,176 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-01 23:58:54,356 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-01 23:58:57,891 INFO [train.py:903] (0/4) Epoch 13, batch 2400, loss[loss=0.256, simple_loss=0.3235, pruned_loss=0.09423, over 19522.00 frames. ], tot_loss[loss=0.2278, simple_loss=0.3028, pruned_loss=0.07643, over 3838913.62 frames. ], batch size: 54, lr: 6.44e-03, grad_scale: 8.0 +2023-04-01 23:59:11,464 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84347.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:59:20,647 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84354.0, num_to_drop=0, layers_to_drop=set() +2023-04-01 23:59:54,315 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.264e+02 5.138e+02 6.932e+02 8.383e+02 1.660e+03, threshold=1.386e+03, percent-clipped=4.0 +2023-04-01 23:59:59,973 INFO [train.py:903] (0/4) Epoch 13, batch 2450, loss[loss=0.2487, simple_loss=0.3254, pruned_loss=0.08607, over 19849.00 frames. ], tot_loss[loss=0.2286, simple_loss=0.3034, pruned_loss=0.07688, over 3850609.87 frames. ], batch size: 52, lr: 6.43e-03, grad_scale: 8.0 +2023-04-02 00:00:16,771 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.14 vs. limit=5.0 +2023-04-02 00:00:51,567 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84426.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:01:03,733 INFO [train.py:903] (0/4) Epoch 13, batch 2500, loss[loss=0.2507, simple_loss=0.3329, pruned_loss=0.08423, over 19672.00 frames. ], tot_loss[loss=0.229, simple_loss=0.3037, pruned_loss=0.07716, over 3835422.39 frames. ], batch size: 58, lr: 6.43e-03, grad_scale: 8.0 +2023-04-02 00:01:33,211 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84459.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:01:38,000 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84462.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:02:00,516 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.656e+02 5.326e+02 7.100e+02 9.098e+02 1.657e+03, threshold=1.420e+03, percent-clipped=3.0 +2023-04-02 00:02:06,422 INFO [train.py:903] (0/4) Epoch 13, batch 2550, loss[loss=0.2501, simple_loss=0.3115, pruned_loss=0.09432, over 19608.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.3033, pruned_loss=0.07722, over 3834797.27 frames. ], batch size: 50, lr: 6.43e-03, grad_scale: 8.0 +2023-04-02 00:02:16,039 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84493.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:02:48,818 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4300, 2.0228, 2.0504, 2.6776, 2.2861, 2.1556, 2.0075, 2.4202], + device='cuda:0'), covar=tensor([0.0869, 0.1673, 0.1318, 0.0834, 0.1290, 0.0470, 0.1163, 0.0639], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0346, 0.0292, 0.0236, 0.0293, 0.0240, 0.0278, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:02:48,828 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84518.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:03:04,532 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 00:03:10,388 INFO [train.py:903] (0/4) Epoch 13, batch 2600, loss[loss=0.2479, simple_loss=0.3257, pruned_loss=0.08502, over 19601.00 frames. ], tot_loss[loss=0.2286, simple_loss=0.3032, pruned_loss=0.07703, over 3826004.37 frames. ], batch size: 52, lr: 6.43e-03, grad_scale: 8.0 +2023-04-02 00:03:21,083 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84543.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:03:42,738 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84561.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:03:46,342 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0802, 1.6506, 1.6606, 1.8888, 1.7112, 1.8028, 1.6304, 1.8790], + device='cuda:0'), covar=tensor([0.0837, 0.1360, 0.1261, 0.0835, 0.1202, 0.0484, 0.1163, 0.0654], + device='cuda:0'), in_proj_covar=tensor([0.0253, 0.0346, 0.0291, 0.0237, 0.0293, 0.0240, 0.0278, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:03:59,322 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84574.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:04:09,373 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.575e+02 5.097e+02 6.359e+02 8.045e+02 2.004e+03, threshold=1.272e+03, percent-clipped=4.0 +2023-04-02 00:04:15,154 INFO [train.py:903] (0/4) Epoch 13, batch 2650, loss[loss=0.2051, simple_loss=0.2818, pruned_loss=0.0642, over 19591.00 frames. ], tot_loss[loss=0.2287, simple_loss=0.3034, pruned_loss=0.07698, over 3818798.85 frames. ], batch size: 52, lr: 6.43e-03, grad_scale: 8.0 +2023-04-02 00:04:30,349 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84598.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:04:34,941 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 00:04:35,152 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4117, 1.0270, 1.3317, 1.5837, 2.7224, 1.0772, 2.2881, 3.2653], + device='cuda:0'), covar=tensor([0.0639, 0.3502, 0.3132, 0.1978, 0.1217, 0.2774, 0.1323, 0.0436], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0344, 0.0356, 0.0324, 0.0346, 0.0332, 0.0342, 0.0367], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:05:17,487 INFO [train.py:903] (0/4) Epoch 13, batch 2700, loss[loss=0.1823, simple_loss=0.2694, pruned_loss=0.04759, over 19676.00 frames. ], tot_loss[loss=0.2303, simple_loss=0.3051, pruned_loss=0.07772, over 3825106.03 frames. ], batch size: 53, lr: 6.43e-03, grad_scale: 8.0 +2023-04-02 00:05:36,217 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84651.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:06:08,620 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84676.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:06:14,005 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.791e+02 5.394e+02 6.395e+02 8.456e+02 1.799e+03, threshold=1.279e+03, percent-clipped=4.0 +2023-04-02 00:06:16,790 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84682.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:06:21,022 INFO [train.py:903] (0/4) Epoch 13, batch 2750, loss[loss=0.197, simple_loss=0.2696, pruned_loss=0.06221, over 19282.00 frames. ], tot_loss[loss=0.2305, simple_loss=0.3045, pruned_loss=0.07822, over 3835739.35 frames. ], batch size: 44, lr: 6.42e-03, grad_scale: 8.0 +2023-04-02 00:06:37,479 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84698.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:06:37,744 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4508, 2.1609, 1.5704, 1.5254, 1.9990, 1.1980, 1.3859, 1.8354], + device='cuda:0'), covar=tensor([0.0814, 0.0690, 0.1053, 0.0688, 0.0534, 0.1187, 0.0601, 0.0454], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0305, 0.0327, 0.0252, 0.0240, 0.0320, 0.0289, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:06:49,112 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84707.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:06:56,188 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84713.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:07:01,951 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84718.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:07:24,547 INFO [train.py:903] (0/4) Epoch 13, batch 2800, loss[loss=0.1711, simple_loss=0.2501, pruned_loss=0.046, over 17791.00 frames. ], tot_loss[loss=0.2291, simple_loss=0.3036, pruned_loss=0.07731, over 3838524.25 frames. ], batch size: 39, lr: 6.42e-03, grad_scale: 8.0 +2023-04-02 00:07:34,100 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84743.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:07:52,563 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3247, 2.9709, 2.1493, 2.7788, 0.7657, 2.9066, 2.8943, 2.9827], + device='cuda:0'), covar=tensor([0.1050, 0.1431, 0.2201, 0.0972, 0.3967, 0.1052, 0.1029, 0.1345], + device='cuda:0'), in_proj_covar=tensor([0.0443, 0.0368, 0.0443, 0.0319, 0.0380, 0.0374, 0.0368, 0.0400], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:08:22,665 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.267e+02 5.423e+02 6.884e+02 8.957e+02 1.568e+03, threshold=1.377e+03, percent-clipped=4.0 +2023-04-02 00:08:29,876 INFO [train.py:903] (0/4) Epoch 13, batch 2850, loss[loss=0.2596, simple_loss=0.3305, pruned_loss=0.09438, over 19535.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.303, pruned_loss=0.07684, over 3842118.60 frames. ], batch size: 54, lr: 6.42e-03, grad_scale: 8.0 +2023-04-02 00:08:31,518 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2290, 2.0960, 1.8508, 1.6448, 1.5841, 1.6989, 0.4407, 1.0131], + device='cuda:0'), covar=tensor([0.0418, 0.0437, 0.0324, 0.0577, 0.0972, 0.0610, 0.1003, 0.0814], + device='cuda:0'), in_proj_covar=tensor([0.0340, 0.0333, 0.0334, 0.0357, 0.0429, 0.0356, 0.0315, 0.0323], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 00:09:04,276 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=84813.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:09:08,027 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7456, 1.4711, 1.5029, 2.2257, 1.8627, 1.9363, 2.0957, 1.8436], + device='cuda:0'), covar=tensor([0.0815, 0.1009, 0.1089, 0.0771, 0.0823, 0.0786, 0.0878, 0.0644], + device='cuda:0'), in_proj_covar=tensor([0.0207, 0.0223, 0.0223, 0.0243, 0.0230, 0.0209, 0.0193, 0.0200], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 00:09:11,184 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84818.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:09:26,597 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84830.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:09:29,145 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9564, 1.9571, 2.2481, 2.6299, 1.8184, 2.4693, 2.4514, 2.0711], + device='cuda:0'), covar=tensor([0.3619, 0.3315, 0.1478, 0.2004, 0.3635, 0.1659, 0.3572, 0.2701], + device='cuda:0'), in_proj_covar=tensor([0.0814, 0.0844, 0.0663, 0.0899, 0.0796, 0.0725, 0.0800, 0.0723], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 00:09:33,381 INFO [train.py:903] (0/4) Epoch 13, batch 2900, loss[loss=0.2376, simple_loss=0.3108, pruned_loss=0.08219, over 19666.00 frames. ], tot_loss[loss=0.227, simple_loss=0.3019, pruned_loss=0.07604, over 3855075.38 frames. ], batch size: 60, lr: 6.42e-03, grad_scale: 8.0 +2023-04-02 00:09:33,417 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 00:09:58,713 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84855.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:10:31,984 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.833e+02 5.191e+02 6.710e+02 8.572e+02 2.238e+03, threshold=1.342e+03, percent-clipped=4.0 +2023-04-02 00:10:38,036 INFO [train.py:903] (0/4) Epoch 13, batch 2950, loss[loss=0.1984, simple_loss=0.2814, pruned_loss=0.05767, over 19599.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.3009, pruned_loss=0.07543, over 3842524.91 frames. ], batch size: 52, lr: 6.42e-03, grad_scale: 8.0 +2023-04-02 00:10:40,474 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84887.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:11:39,621 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84932.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:11:45,156 INFO [train.py:903] (0/4) Epoch 13, batch 3000, loss[loss=0.2488, simple_loss=0.3225, pruned_loss=0.08757, over 19664.00 frames. ], tot_loss[loss=0.2281, simple_loss=0.3027, pruned_loss=0.07669, over 3821721.23 frames. ], batch size: 60, lr: 6.41e-03, grad_scale: 8.0 +2023-04-02 00:11:45,157 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 00:12:00,840 INFO [train.py:937] (0/4) Epoch 13, validation: loss=0.1754, simple_loss=0.276, pruned_loss=0.03742, over 944034.00 frames. +2023-04-02 00:12:00,842 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18321MB +2023-04-02 00:12:05,792 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 00:12:29,830 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84957.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:12:43,877 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=84969.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:12:44,017 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=84969.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:12:59,475 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.728e+02 4.901e+02 6.331e+02 8.447e+02 1.208e+03, threshold=1.266e+03, percent-clipped=0.0 +2023-04-02 00:13:05,457 INFO [train.py:903] (0/4) Epoch 13, batch 3050, loss[loss=0.2439, simple_loss=0.3199, pruned_loss=0.08397, over 19650.00 frames. ], tot_loss[loss=0.2278, simple_loss=0.3024, pruned_loss=0.07661, over 3817326.53 frames. ], batch size: 58, lr: 6.41e-03, grad_scale: 8.0 +2023-04-02 00:13:17,125 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=84994.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:13:18,145 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=84995.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:13:26,443 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85002.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:14:09,068 INFO [train.py:903] (0/4) Epoch 13, batch 3100, loss[loss=0.2525, simple_loss=0.3239, pruned_loss=0.09059, over 19751.00 frames. ], tot_loss[loss=0.23, simple_loss=0.3039, pruned_loss=0.07804, over 3790062.79 frames. ], batch size: 63, lr: 6.41e-03, grad_scale: 8.0 +2023-04-02 00:14:19,296 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-04-02 00:14:50,680 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85069.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:15:05,349 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.206e+02 5.118e+02 6.727e+02 8.323e+02 1.616e+03, threshold=1.345e+03, percent-clipped=6.0 +2023-04-02 00:15:11,218 INFO [train.py:903] (0/4) Epoch 13, batch 3150, loss[loss=0.2416, simple_loss=0.3308, pruned_loss=0.0762, over 19665.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.303, pruned_loss=0.07686, over 3808863.24 frames. ], batch size: 55, lr: 6.41e-03, grad_scale: 8.0 +2023-04-02 00:15:20,631 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85094.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:15:37,411 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 00:15:41,902 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85110.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:16:00,130 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85125.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:16:09,356 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85133.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:16:12,580 INFO [train.py:903] (0/4) Epoch 13, batch 3200, loss[loss=0.2279, simple_loss=0.3179, pruned_loss=0.06893, over 19660.00 frames. ], tot_loss[loss=0.2285, simple_loss=0.303, pruned_loss=0.07696, over 3812018.68 frames. ], batch size: 60, lr: 6.41e-03, grad_scale: 8.0 +2023-04-02 00:16:46,231 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85162.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:17:10,847 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.817e+02 5.513e+02 6.647e+02 8.266e+02 1.326e+03, threshold=1.329e+03, percent-clipped=0.0 +2023-04-02 00:17:16,618 INFO [train.py:903] (0/4) Epoch 13, batch 3250, loss[loss=0.2108, simple_loss=0.2924, pruned_loss=0.0646, over 19778.00 frames. ], tot_loss[loss=0.2274, simple_loss=0.3021, pruned_loss=0.07632, over 3816109.11 frames. ], batch size: 56, lr: 6.40e-03, grad_scale: 8.0 +2023-04-02 00:17:54,816 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.96 vs. limit=5.0 +2023-04-02 00:18:20,664 INFO [train.py:903] (0/4) Epoch 13, batch 3300, loss[loss=0.2322, simple_loss=0.3188, pruned_loss=0.07274, over 19576.00 frames. ], tot_loss[loss=0.2278, simple_loss=0.3025, pruned_loss=0.07656, over 3819885.71 frames. ], batch size: 61, lr: 6.40e-03, grad_scale: 8.0 +2023-04-02 00:18:21,898 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 00:18:47,636 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85258.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:18:57,446 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.2255, 5.5937, 3.0659, 4.9219, 1.2921, 5.5300, 5.5214, 5.6939], + device='cuda:0'), covar=tensor([0.0411, 0.0953, 0.1821, 0.0637, 0.3884, 0.0582, 0.0655, 0.0823], + device='cuda:0'), in_proj_covar=tensor([0.0447, 0.0373, 0.0448, 0.0322, 0.0387, 0.0380, 0.0371, 0.0403], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:19:12,393 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85277.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:19:17,832 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.152e+02 5.269e+02 6.447e+02 8.165e+02 1.494e+03, threshold=1.289e+03, percent-clipped=3.0 +2023-04-02 00:19:19,488 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85283.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:19:22,466 INFO [train.py:903] (0/4) Epoch 13, batch 3350, loss[loss=0.2493, simple_loss=0.3245, pruned_loss=0.08705, over 19445.00 frames. ], tot_loss[loss=0.2282, simple_loss=0.3027, pruned_loss=0.07683, over 3819077.21 frames. ], batch size: 64, lr: 6.40e-03, grad_scale: 8.0 +2023-04-02 00:19:26,586 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 00:19:57,319 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85313.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:20:24,674 INFO [train.py:903] (0/4) Epoch 13, batch 3400, loss[loss=0.1883, simple_loss=0.2596, pruned_loss=0.05851, over 19707.00 frames. ], tot_loss[loss=0.2288, simple_loss=0.3031, pruned_loss=0.07722, over 3802058.51 frames. ], batch size: 45, lr: 6.40e-03, grad_scale: 8.0 +2023-04-02 00:20:42,480 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9935, 2.6407, 1.8640, 1.9537, 2.4474, 1.7770, 1.6131, 2.0168], + device='cuda:0'), covar=tensor([0.0812, 0.0627, 0.0732, 0.0587, 0.0388, 0.0854, 0.0599, 0.0508], + device='cuda:0'), in_proj_covar=tensor([0.0290, 0.0301, 0.0323, 0.0247, 0.0236, 0.0317, 0.0283, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:20:42,554 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3263, 2.3713, 2.5840, 3.1799, 2.4755, 3.1115, 2.7677, 2.3370], + device='cuda:0'), covar=tensor([0.3256, 0.2943, 0.1300, 0.1623, 0.3168, 0.1306, 0.2866, 0.2327], + device='cuda:0'), in_proj_covar=tensor([0.0806, 0.0835, 0.0658, 0.0890, 0.0793, 0.0718, 0.0794, 0.0718], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 00:21:03,052 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85366.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:21:22,545 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.171e+02 5.371e+02 6.826e+02 8.500e+02 2.504e+03, threshold=1.365e+03, percent-clipped=9.0 +2023-04-02 00:21:27,161 INFO [train.py:903] (0/4) Epoch 13, batch 3450, loss[loss=0.2003, simple_loss=0.2745, pruned_loss=0.06306, over 19576.00 frames. ], tot_loss[loss=0.228, simple_loss=0.3026, pruned_loss=0.07671, over 3813506.46 frames. ], batch size: 52, lr: 6.40e-03, grad_scale: 8.0 +2023-04-02 00:21:30,635 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 00:21:34,246 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85391.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:21:47,629 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6056, 1.4217, 1.4071, 1.9178, 1.6422, 1.9968, 1.9633, 1.7690], + device='cuda:0'), covar=tensor([0.0819, 0.1032, 0.1088, 0.0946, 0.0874, 0.0699, 0.0902, 0.0689], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0225, 0.0225, 0.0245, 0.0231, 0.0212, 0.0193, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 00:22:19,348 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85428.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:22:28,803 INFO [train.py:903] (0/4) Epoch 13, batch 3500, loss[loss=0.181, simple_loss=0.2516, pruned_loss=0.05515, over 19323.00 frames. ], tot_loss[loss=0.2287, simple_loss=0.3032, pruned_loss=0.07709, over 3811262.28 frames. ], batch size: 44, lr: 6.40e-03, grad_scale: 8.0 +2023-04-02 00:22:42,072 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85446.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:23:10,405 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85469.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:23:19,016 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-02 00:23:20,658 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85477.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:23:27,192 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.349e+02 5.339e+02 6.675e+02 8.042e+02 1.644e+03, threshold=1.335e+03, percent-clipped=3.0 +2023-04-02 00:23:31,543 INFO [train.py:903] (0/4) Epoch 13, batch 3550, loss[loss=0.2077, simple_loss=0.2717, pruned_loss=0.07181, over 18611.00 frames. ], tot_loss[loss=0.2283, simple_loss=0.3028, pruned_loss=0.07686, over 3807555.15 frames. ], batch size: 41, lr: 6.39e-03, grad_scale: 8.0 +2023-04-02 00:23:50,596 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85501.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:24:31,217 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85533.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:24:34,224 INFO [train.py:903] (0/4) Epoch 13, batch 3600, loss[loss=0.2119, simple_loss=0.2774, pruned_loss=0.07324, over 19779.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.3027, pruned_loss=0.07702, over 3799411.16 frames. ], batch size: 46, lr: 6.39e-03, grad_scale: 8.0 +2023-04-02 00:24:50,355 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85548.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:25:03,579 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85558.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:25:33,198 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.573e+02 5.969e+02 7.367e+02 9.197e+02 1.857e+03, threshold=1.473e+03, percent-clipped=2.0 +2023-04-02 00:25:36,788 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85584.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:25:38,625 INFO [train.py:903] (0/4) Epoch 13, batch 3650, loss[loss=0.2781, simple_loss=0.3439, pruned_loss=0.1061, over 19282.00 frames. ], tot_loss[loss=0.2295, simple_loss=0.304, pruned_loss=0.07753, over 3795042.43 frames. ], batch size: 66, lr: 6.39e-03, grad_scale: 8.0 +2023-04-02 00:25:46,137 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85592.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:26:41,857 INFO [train.py:903] (0/4) Epoch 13, batch 3700, loss[loss=0.2273, simple_loss=0.2929, pruned_loss=0.08083, over 18719.00 frames. ], tot_loss[loss=0.2299, simple_loss=0.3042, pruned_loss=0.07777, over 3799840.96 frames. ], batch size: 41, lr: 6.39e-03, grad_scale: 8.0 +2023-04-02 00:26:51,220 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85643.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:27:41,799 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.698e+02 5.337e+02 6.385e+02 1.025e+03 1.989e+03, threshold=1.277e+03, percent-clipped=4.0 +2023-04-02 00:27:44,649 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85684.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:27:46,553 INFO [train.py:903] (0/4) Epoch 13, batch 3750, loss[loss=0.2853, simple_loss=0.3538, pruned_loss=0.1083, over 19613.00 frames. ], tot_loss[loss=0.2293, simple_loss=0.3037, pruned_loss=0.07744, over 3789295.14 frames. ], batch size: 57, lr: 6.39e-03, grad_scale: 8.0 +2023-04-02 00:28:14,925 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85709.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:28:48,148 INFO [train.py:903] (0/4) Epoch 13, batch 3800, loss[loss=0.1791, simple_loss=0.2531, pruned_loss=0.0525, over 19730.00 frames. ], tot_loss[loss=0.2296, simple_loss=0.304, pruned_loss=0.07762, over 3792592.13 frames. ], batch size: 46, lr: 6.38e-03, grad_scale: 8.0 +2023-04-02 00:29:13,288 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0578, 5.3942, 2.8611, 4.6920, 1.1403, 5.4908, 5.3758, 5.5367], + device='cuda:0'), covar=tensor([0.0403, 0.0802, 0.1897, 0.0664, 0.3850, 0.0501, 0.0650, 0.0779], + device='cuda:0'), in_proj_covar=tensor([0.0446, 0.0375, 0.0449, 0.0323, 0.0388, 0.0383, 0.0373, 0.0406], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:29:20,841 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 00:29:44,888 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.310e+02 5.587e+02 7.205e+02 9.194e+02 2.888e+03, threshold=1.441e+03, percent-clipped=8.0 +2023-04-02 00:29:50,626 INFO [train.py:903] (0/4) Epoch 13, batch 3850, loss[loss=0.2512, simple_loss=0.3263, pruned_loss=0.08805, over 19141.00 frames. ], tot_loss[loss=0.2311, simple_loss=0.3052, pruned_loss=0.07847, over 3803268.62 frames. ], batch size: 69, lr: 6.38e-03, grad_scale: 8.0 +2023-04-02 00:29:56,219 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85790.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:29:56,549 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8411, 1.9098, 2.1698, 2.6307, 1.8537, 2.5061, 2.3288, 2.0159], + device='cuda:0'), covar=tensor([0.3724, 0.3247, 0.1495, 0.1802, 0.3431, 0.1542, 0.3746, 0.2854], + device='cuda:0'), in_proj_covar=tensor([0.0812, 0.0846, 0.0661, 0.0899, 0.0797, 0.0723, 0.0800, 0.0722], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 00:30:53,504 INFO [train.py:903] (0/4) Epoch 13, batch 3900, loss[loss=0.2221, simple_loss=0.3097, pruned_loss=0.06722, over 19600.00 frames. ], tot_loss[loss=0.2294, simple_loss=0.304, pruned_loss=0.07742, over 3802409.76 frames. ], batch size: 57, lr: 6.38e-03, grad_scale: 8.0 +2023-04-02 00:31:00,094 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85840.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:31:06,801 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85845.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:31:10,512 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=85848.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:31:30,218 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85865.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:31:40,433 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=85873.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:31:51,444 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.894e+02 5.450e+02 6.738e+02 8.252e+02 2.044e+03, threshold=1.348e+03, percent-clipped=4.0 +2023-04-02 00:31:57,035 INFO [train.py:903] (0/4) Epoch 13, batch 3950, loss[loss=0.2066, simple_loss=0.269, pruned_loss=0.07206, over 19769.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.3034, pruned_loss=0.07719, over 3817703.59 frames. ], batch size: 47, lr: 6.38e-03, grad_scale: 8.0 +2023-04-02 00:32:00,536 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 00:32:04,154 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85892.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:32:05,530 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85893.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:32:20,776 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85905.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:32:51,825 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2917, 3.7136, 3.8656, 3.8726, 1.5367, 3.6495, 3.2437, 3.5538], + device='cuda:0'), covar=tensor([0.1494, 0.1076, 0.0633, 0.0689, 0.5119, 0.0900, 0.0628, 0.1086], + device='cuda:0'), in_proj_covar=tensor([0.0695, 0.0619, 0.0825, 0.0702, 0.0747, 0.0574, 0.0496, 0.0760], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 00:32:59,797 INFO [train.py:903] (0/4) Epoch 13, batch 4000, loss[loss=0.2513, simple_loss=0.3281, pruned_loss=0.08724, over 19302.00 frames. ], tot_loss[loss=0.2286, simple_loss=0.3034, pruned_loss=0.07693, over 3805946.38 frames. ], batch size: 66, lr: 6.38e-03, grad_scale: 8.0 +2023-04-02 00:33:14,761 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85948.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 00:33:30,751 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=85960.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:33:46,158 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 00:33:48,581 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=85975.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:33:56,481 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.510e+02 5.280e+02 6.258e+02 8.022e+02 1.377e+03, threshold=1.252e+03, percent-clipped=2.0 +2023-04-02 00:34:02,168 INFO [train.py:903] (0/4) Epoch 13, batch 4050, loss[loss=0.2024, simple_loss=0.2776, pruned_loss=0.06366, over 19712.00 frames. ], tot_loss[loss=0.2294, simple_loss=0.3039, pruned_loss=0.07746, over 3810610.28 frames. ], batch size: 46, lr: 6.37e-03, grad_scale: 8.0 +2023-04-02 00:34:03,493 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=85987.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:34:20,259 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-86000.pt +2023-04-02 00:34:30,787 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86007.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:34:46,102 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86020.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:34:46,294 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1720, 2.2301, 2.3569, 3.1481, 2.2567, 3.0923, 2.6117, 2.2178], + device='cuda:0'), covar=tensor([0.3831, 0.3535, 0.1589, 0.2044, 0.3960, 0.1622, 0.3723, 0.2804], + device='cuda:0'), in_proj_covar=tensor([0.0813, 0.0845, 0.0660, 0.0899, 0.0798, 0.0725, 0.0800, 0.0723], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 00:35:05,744 INFO [train.py:903] (0/4) Epoch 13, batch 4100, loss[loss=0.2624, simple_loss=0.3307, pruned_loss=0.09707, over 17695.00 frames. ], tot_loss[loss=0.2294, simple_loss=0.3038, pruned_loss=0.07746, over 3810106.25 frames. ], batch size: 101, lr: 6.37e-03, grad_scale: 8.0 +2023-04-02 00:35:07,489 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3550, 1.8941, 1.9893, 2.9135, 2.0987, 2.6979, 2.6398, 2.3155], + device='cuda:0'), covar=tensor([0.0699, 0.0909, 0.0947, 0.0795, 0.0857, 0.0684, 0.0879, 0.0632], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0224, 0.0225, 0.0243, 0.0232, 0.0210, 0.0191, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 00:35:41,511 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 00:35:48,384 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86070.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:36:04,014 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.212e+02 4.943e+02 6.094e+02 7.890e+02 1.986e+03, threshold=1.219e+03, percent-clipped=5.0 +2023-04-02 00:36:08,782 INFO [train.py:903] (0/4) Epoch 13, batch 4150, loss[loss=0.1848, simple_loss=0.2605, pruned_loss=0.05452, over 19761.00 frames. ], tot_loss[loss=0.2285, simple_loss=0.3029, pruned_loss=0.07701, over 3810949.78 frames. ], batch size: 47, lr: 6.37e-03, grad_scale: 8.0 +2023-04-02 00:36:11,029 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86087.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:36:24,547 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-04-02 00:36:28,536 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86102.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:37:11,325 INFO [train.py:903] (0/4) Epoch 13, batch 4200, loss[loss=0.1699, simple_loss=0.2459, pruned_loss=0.04691, over 19283.00 frames. ], tot_loss[loss=0.2282, simple_loss=0.3029, pruned_loss=0.07675, over 3818195.16 frames. ], batch size: 44, lr: 6.37e-03, grad_scale: 8.0 +2023-04-02 00:37:14,988 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 00:37:41,978 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86161.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:38:08,160 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.547e+02 5.506e+02 6.651e+02 9.015e+02 1.898e+03, threshold=1.330e+03, percent-clipped=8.0 +2023-04-02 00:38:12,740 INFO [train.py:903] (0/4) Epoch 13, batch 4250, loss[loss=0.2119, simple_loss=0.2749, pruned_loss=0.0745, over 19722.00 frames. ], tot_loss[loss=0.228, simple_loss=0.3028, pruned_loss=0.07663, over 3835291.54 frames. ], batch size: 46, lr: 6.37e-03, grad_scale: 8.0 +2023-04-02 00:38:13,158 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86186.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:38:31,135 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 00:38:43,583 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 00:38:52,336 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86216.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:39:15,638 INFO [train.py:903] (0/4) Epoch 13, batch 4300, loss[loss=0.2453, simple_loss=0.3225, pruned_loss=0.08404, over 19448.00 frames. ], tot_loss[loss=0.2268, simple_loss=0.3019, pruned_loss=0.07585, over 3846501.61 frames. ], batch size: 70, lr: 6.37e-03, grad_scale: 8.0 +2023-04-02 00:39:17,000 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86237.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:39:17,306 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5303, 2.3870, 1.8257, 1.5802, 2.1876, 1.4059, 1.4341, 1.9261], + device='cuda:0'), covar=tensor([0.0914, 0.0623, 0.0882, 0.0708, 0.0438, 0.1061, 0.0579, 0.0406], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0304, 0.0327, 0.0251, 0.0237, 0.0320, 0.0287, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:39:22,955 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86241.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:39:50,451 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86263.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:39:51,561 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86264.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:40:13,250 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.647e+02 5.244e+02 6.314e+02 8.271e+02 2.210e+03, threshold=1.263e+03, percent-clipped=7.0 +2023-04-02 00:40:15,192 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 00:40:19,673 INFO [train.py:903] (0/4) Epoch 13, batch 4350, loss[loss=0.2489, simple_loss=0.3139, pruned_loss=0.09191, over 19597.00 frames. ], tot_loss[loss=0.2271, simple_loss=0.3018, pruned_loss=0.07615, over 3841875.85 frames. ], batch size: 52, lr: 6.36e-03, grad_scale: 8.0 +2023-04-02 00:40:22,529 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86288.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:40:27,935 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86292.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 00:41:00,086 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86319.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:41:22,565 INFO [train.py:903] (0/4) Epoch 13, batch 4400, loss[loss=0.2364, simple_loss=0.3098, pruned_loss=0.0815, over 19668.00 frames. ], tot_loss[loss=0.2267, simple_loss=0.3014, pruned_loss=0.07599, over 3827521.83 frames. ], batch size: 58, lr: 6.36e-03, grad_scale: 8.0 +2023-04-02 00:41:42,034 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86352.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:41:49,162 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86358.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:41:50,097 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 00:41:58,578 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86364.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:42:00,813 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 00:42:20,371 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.626e+02 5.303e+02 6.614e+02 7.903e+02 1.976e+03, threshold=1.323e+03, percent-clipped=7.0 +2023-04-02 00:42:21,972 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86383.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:42:25,157 INFO [train.py:903] (0/4) Epoch 13, batch 4450, loss[loss=0.2282, simple_loss=0.3083, pruned_loss=0.07401, over 19610.00 frames. ], tot_loss[loss=0.2257, simple_loss=0.3004, pruned_loss=0.07555, over 3813697.72 frames. ], batch size: 57, lr: 6.36e-03, grad_scale: 8.0 +2023-04-02 00:42:52,178 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86407.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 00:43:00,988 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86414.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:43:21,599 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86431.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:43:25,407 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86434.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:43:27,352 INFO [train.py:903] (0/4) Epoch 13, batch 4500, loss[loss=0.236, simple_loss=0.3121, pruned_loss=0.07995, over 19533.00 frames. ], tot_loss[loss=0.2257, simple_loss=0.3007, pruned_loss=0.07538, over 3832986.42 frames. ], batch size: 54, lr: 6.36e-03, grad_scale: 8.0 +2023-04-02 00:43:52,035 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7295, 1.7304, 1.5966, 1.3806, 1.3267, 1.4072, 0.2030, 0.6842], + device='cuda:0'), covar=tensor([0.0452, 0.0449, 0.0281, 0.0437, 0.0859, 0.0525, 0.0906, 0.0811], + device='cuda:0'), in_proj_covar=tensor([0.0343, 0.0337, 0.0334, 0.0360, 0.0433, 0.0360, 0.0316, 0.0327], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 00:43:55,654 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86458.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:44:01,502 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86463.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:44:21,173 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86479.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:44:24,468 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.347e+02 5.430e+02 6.459e+02 7.870e+02 1.871e+03, threshold=1.292e+03, percent-clipped=3.0 +2023-04-02 00:44:29,963 INFO [train.py:903] (0/4) Epoch 13, batch 4550, loss[loss=0.2449, simple_loss=0.3243, pruned_loss=0.08279, over 18730.00 frames. ], tot_loss[loss=0.2262, simple_loss=0.3011, pruned_loss=0.07567, over 3823493.31 frames. ], batch size: 74, lr: 6.36e-03, grad_scale: 8.0 +2023-04-02 00:44:39,219 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 00:45:02,103 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 00:45:07,127 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3837, 4.0351, 2.5080, 3.5624, 1.1139, 3.7484, 3.7553, 3.8750], + device='cuda:0'), covar=tensor([0.0663, 0.1025, 0.2086, 0.0829, 0.3765, 0.0813, 0.0826, 0.1026], + device='cuda:0'), in_proj_covar=tensor([0.0448, 0.0374, 0.0450, 0.0323, 0.0388, 0.0382, 0.0373, 0.0408], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:45:16,171 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.55 vs. limit=2.0 +2023-04-02 00:45:24,953 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86529.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:45:33,594 INFO [train.py:903] (0/4) Epoch 13, batch 4600, loss[loss=0.1915, simple_loss=0.257, pruned_loss=0.06305, over 19762.00 frames. ], tot_loss[loss=0.2269, simple_loss=0.3016, pruned_loss=0.07605, over 3821812.54 frames. ], batch size: 47, lr: 6.35e-03, grad_scale: 8.0 +2023-04-02 00:45:45,502 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86546.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:45:50,023 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6362, 1.4019, 1.2757, 2.0909, 1.5866, 2.0387, 2.0466, 1.7170], + device='cuda:0'), covar=tensor([0.0808, 0.0993, 0.1136, 0.0899, 0.0942, 0.0747, 0.0894, 0.0714], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0225, 0.0224, 0.0245, 0.0231, 0.0211, 0.0192, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 00:46:31,573 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.083e+02 5.452e+02 6.634e+02 8.020e+02 1.693e+03, threshold=1.327e+03, percent-clipped=1.0 +2023-04-02 00:46:35,925 INFO [train.py:903] (0/4) Epoch 13, batch 4650, loss[loss=0.2745, simple_loss=0.3324, pruned_loss=0.1082, over 13321.00 frames. ], tot_loss[loss=0.2263, simple_loss=0.3016, pruned_loss=0.07547, over 3826592.35 frames. ], batch size: 135, lr: 6.35e-03, grad_scale: 8.0 +2023-04-02 00:46:51,339 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86598.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:46:52,259 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 00:46:57,295 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-02 00:47:03,875 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 00:47:03,989 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86608.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:47:04,257 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86608.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:47:35,012 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86633.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:47:35,322 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-02 00:47:38,173 INFO [train.py:903] (0/4) Epoch 13, batch 4700, loss[loss=0.2364, simple_loss=0.3133, pruned_loss=0.07975, over 19581.00 frames. ], tot_loss[loss=0.2265, simple_loss=0.3017, pruned_loss=0.07561, over 3823247.97 frames. ], batch size: 67, lr: 6.35e-03, grad_scale: 8.0 +2023-04-02 00:47:53,756 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-02 00:48:04,331 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 00:48:06,008 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1761, 2.0418, 1.8462, 1.6757, 1.4956, 1.6667, 0.3837, 1.0830], + device='cuda:0'), covar=tensor([0.0441, 0.0454, 0.0350, 0.0575, 0.0986, 0.0601, 0.0976, 0.0754], + device='cuda:0'), in_proj_covar=tensor([0.0341, 0.0335, 0.0332, 0.0357, 0.0432, 0.0357, 0.0314, 0.0325], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 00:48:13,273 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86663.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 00:48:35,960 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.257e+02 5.083e+02 5.953e+02 7.054e+02 1.649e+03, threshold=1.191e+03, percent-clipped=1.0 +2023-04-02 00:48:42,060 INFO [train.py:903] (0/4) Epoch 13, batch 4750, loss[loss=0.2413, simple_loss=0.3173, pruned_loss=0.08268, over 18846.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.3015, pruned_loss=0.07515, over 3820128.23 frames. ], batch size: 74, lr: 6.35e-03, grad_scale: 8.0 +2023-04-02 00:48:45,709 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86688.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 00:48:48,026 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86690.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:48:58,159 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7045, 1.3314, 1.4021, 1.4951, 3.2379, 1.0906, 2.2105, 3.5898], + device='cuda:0'), covar=tensor([0.0459, 0.2677, 0.2922, 0.1848, 0.0730, 0.2512, 0.1378, 0.0278], + device='cuda:0'), in_proj_covar=tensor([0.0365, 0.0345, 0.0356, 0.0325, 0.0352, 0.0332, 0.0346, 0.0365], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:49:07,677 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-02 00:49:10,036 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-04-02 00:49:15,251 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.10 vs. limit=2.0 +2023-04-02 00:49:18,489 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86715.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:49:27,240 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86723.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:49:31,060 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-04-02 00:49:43,045 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86735.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:49:43,760 INFO [train.py:903] (0/4) Epoch 13, batch 4800, loss[loss=0.2911, simple_loss=0.3502, pruned_loss=0.116, over 19273.00 frames. ], tot_loss[loss=0.2273, simple_loss=0.3023, pruned_loss=0.07609, over 3809447.56 frames. ], batch size: 66, lr: 6.35e-03, grad_scale: 8.0 +2023-04-02 00:50:12,879 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86760.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:50:41,009 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.218e+02 5.270e+02 6.552e+02 8.488e+02 1.715e+03, threshold=1.310e+03, percent-clipped=7.0 +2023-04-02 00:50:45,087 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86785.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:50:45,826 INFO [train.py:903] (0/4) Epoch 13, batch 4850, loss[loss=0.2183, simple_loss=0.2812, pruned_loss=0.07771, over 19742.00 frames. ], tot_loss[loss=0.2273, simple_loss=0.3022, pruned_loss=0.07618, over 3791319.37 frames. ], batch size: 46, lr: 6.35e-03, grad_scale: 8.0 +2023-04-02 00:51:01,584 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.3096, 5.6701, 3.1402, 4.8318, 1.1260, 5.6133, 5.6861, 5.7313], + device='cuda:0'), covar=tensor([0.0339, 0.0835, 0.1650, 0.0644, 0.4046, 0.0507, 0.0556, 0.0835], + device='cuda:0'), in_proj_covar=tensor([0.0441, 0.0369, 0.0444, 0.0321, 0.0382, 0.0379, 0.0368, 0.0406], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 00:51:06,432 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86802.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:51:06,703 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86802.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:51:09,592 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 00:51:14,191 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86807.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:51:18,076 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86810.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:51:31,690 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 00:51:37,264 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 00:51:37,288 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 00:51:38,810 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=86827.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:51:45,325 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86832.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:51:47,403 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 00:51:49,797 INFO [train.py:903] (0/4) Epoch 13, batch 4900, loss[loss=0.2295, simple_loss=0.3101, pruned_loss=0.07448, over 19779.00 frames. ], tot_loss[loss=0.2263, simple_loss=0.301, pruned_loss=0.0758, over 3794398.64 frames. ], batch size: 56, lr: 6.34e-03, grad_scale: 8.0 +2023-04-02 00:52:07,083 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 00:52:11,972 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-04-02 00:52:46,044 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.083e+02 5.393e+02 6.845e+02 8.725e+02 1.892e+03, threshold=1.369e+03, percent-clipped=5.0 +2023-04-02 00:52:50,737 INFO [train.py:903] (0/4) Epoch 13, batch 4950, loss[loss=0.2177, simple_loss=0.2878, pruned_loss=0.07379, over 19772.00 frames. ], tot_loss[loss=0.2264, simple_loss=0.3014, pruned_loss=0.07571, over 3813040.33 frames. ], batch size: 47, lr: 6.34e-03, grad_scale: 8.0 +2023-04-02 00:53:05,765 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=86896.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:53:08,078 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 00:53:31,755 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86917.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:53:32,645 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 00:53:37,639 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=86922.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:53:44,587 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0595, 5.0757, 5.9163, 5.8683, 1.9539, 5.4859, 4.7379, 5.5294], + device='cuda:0'), covar=tensor([0.1425, 0.0639, 0.0457, 0.0483, 0.5387, 0.0520, 0.0544, 0.0952], + device='cuda:0'), in_proj_covar=tensor([0.0710, 0.0626, 0.0837, 0.0713, 0.0753, 0.0584, 0.0506, 0.0770], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 00:53:56,223 INFO [train.py:903] (0/4) Epoch 13, batch 5000, loss[loss=0.2296, simple_loss=0.292, pruned_loss=0.08357, over 16487.00 frames. ], tot_loss[loss=0.2258, simple_loss=0.3013, pruned_loss=0.07519, over 3818270.54 frames. ], batch size: 36, lr: 6.34e-03, grad_scale: 8.0 +2023-04-02 00:54:03,375 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 00:54:03,510 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=86942.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:54:14,004 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 00:54:49,307 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=86979.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:54:52,320 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.998e+02 5.650e+02 6.679e+02 8.615e+02 1.943e+03, threshold=1.336e+03, percent-clipped=4.0 +2023-04-02 00:54:56,911 INFO [train.py:903] (0/4) Epoch 13, batch 5050, loss[loss=0.2125, simple_loss=0.2805, pruned_loss=0.07219, over 19356.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.3006, pruned_loss=0.07461, over 3831686.26 frames. ], batch size: 47, lr: 6.34e-03, grad_scale: 8.0 +2023-04-02 00:55:19,920 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87004.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:55:32,409 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 00:55:41,731 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87020.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 00:56:01,009 INFO [train.py:903] (0/4) Epoch 13, batch 5100, loss[loss=0.2399, simple_loss=0.3191, pruned_loss=0.08033, over 18333.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.3007, pruned_loss=0.07478, over 3817979.23 frames. ], batch size: 83, lr: 6.34e-03, grad_scale: 8.0 +2023-04-02 00:56:07,911 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 00:56:11,374 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 00:56:18,115 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 00:56:28,896 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87057.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:56:58,517 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.198e+02 5.089e+02 6.519e+02 8.713e+02 1.677e+03, threshold=1.304e+03, percent-clipped=2.0 +2023-04-02 00:57:03,281 INFO [train.py:903] (0/4) Epoch 13, batch 5150, loss[loss=0.2109, simple_loss=0.2798, pruned_loss=0.07105, over 19379.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.3013, pruned_loss=0.07525, over 3817493.73 frames. ], batch size: 47, lr: 6.33e-03, grad_scale: 8.0 +2023-04-02 00:57:15,161 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 00:57:48,511 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 00:58:07,742 INFO [train.py:903] (0/4) Epoch 13, batch 5200, loss[loss=0.2327, simple_loss=0.2993, pruned_loss=0.08306, over 19578.00 frames. ], tot_loss[loss=0.2265, simple_loss=0.3016, pruned_loss=0.07574, over 3818902.12 frames. ], batch size: 52, lr: 6.33e-03, grad_scale: 8.0 +2023-04-02 00:58:18,415 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 00:58:54,661 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87173.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:58:57,947 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87176.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:59:01,501 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87178.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:59:04,669 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 00:59:05,685 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.198e+02 5.751e+02 6.794e+02 9.141e+02 2.147e+03, threshold=1.359e+03, percent-clipped=10.0 +2023-04-02 00:59:10,441 INFO [train.py:903] (0/4) Epoch 13, batch 5250, loss[loss=0.2939, simple_loss=0.3478, pruned_loss=0.12, over 12760.00 frames. ], tot_loss[loss=0.2263, simple_loss=0.3017, pruned_loss=0.0755, over 3813124.99 frames. ], batch size: 135, lr: 6.33e-03, grad_scale: 16.0 +2023-04-02 00:59:25,790 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87198.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:59:31,679 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87203.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 00:59:55,611 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 01:00:13,789 INFO [train.py:903] (0/4) Epoch 13, batch 5300, loss[loss=0.2352, simple_loss=0.3043, pruned_loss=0.08301, over 19745.00 frames. ], tot_loss[loss=0.226, simple_loss=0.3013, pruned_loss=0.07536, over 3824573.23 frames. ], batch size: 51, lr: 6.33e-03, grad_scale: 16.0 +2023-04-02 01:00:18,347 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87240.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:00:27,869 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87248.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:00:28,783 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 01:00:51,912 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 01:01:04,391 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2257, 1.2774, 1.7536, 1.2633, 2.8389, 3.7173, 3.5044, 3.8804], + device='cuda:0'), covar=tensor([0.1598, 0.3581, 0.3165, 0.2197, 0.0495, 0.0183, 0.0193, 0.0208], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0302, 0.0332, 0.0252, 0.0223, 0.0164, 0.0209, 0.0216], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 01:01:10,966 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.140e+02 5.034e+02 6.226e+02 7.840e+02 1.929e+03, threshold=1.245e+03, percent-clipped=3.0 +2023-04-02 01:01:15,805 INFO [train.py:903] (0/4) Epoch 13, batch 5350, loss[loss=0.2039, simple_loss=0.2921, pruned_loss=0.05785, over 19528.00 frames. ], tot_loss[loss=0.226, simple_loss=0.3015, pruned_loss=0.07528, over 3830204.34 frames. ], batch size: 54, lr: 6.33e-03, grad_scale: 16.0 +2023-04-02 01:01:24,090 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87291.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:01:49,518 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 01:01:51,120 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87313.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:02:21,295 INFO [train.py:903] (0/4) Epoch 13, batch 5400, loss[loss=0.2779, simple_loss=0.3411, pruned_loss=0.1074, over 19666.00 frames. ], tot_loss[loss=0.2264, simple_loss=0.3014, pruned_loss=0.07566, over 3837596.51 frames. ], batch size: 60, lr: 6.33e-03, grad_scale: 8.0 +2023-04-02 01:02:24,062 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87338.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:02:44,267 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87355.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:02:55,485 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87364.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 01:03:00,737 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.92 vs. limit=5.0 +2023-04-02 01:03:13,269 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5551, 4.1689, 2.4865, 3.6970, 0.8833, 3.9556, 3.9690, 4.0027], + device='cuda:0'), covar=tensor([0.0642, 0.0935, 0.2144, 0.0749, 0.4076, 0.0713, 0.0767, 0.1048], + device='cuda:0'), in_proj_covar=tensor([0.0448, 0.0373, 0.0451, 0.0323, 0.0385, 0.0384, 0.0373, 0.0410], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:03:16,663 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87380.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:03:19,988 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.794e+02 4.952e+02 6.479e+02 7.900e+02 1.578e+03, threshold=1.296e+03, percent-clipped=3.0 +2023-04-02 01:03:23,439 INFO [train.py:903] (0/4) Epoch 13, batch 5450, loss[loss=0.2105, simple_loss=0.2984, pruned_loss=0.06136, over 19534.00 frames. ], tot_loss[loss=0.2255, simple_loss=0.3008, pruned_loss=0.07514, over 3847394.11 frames. ], batch size: 54, lr: 6.32e-03, grad_scale: 8.0 +2023-04-02 01:04:10,192 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-04-02 01:04:25,745 INFO [train.py:903] (0/4) Epoch 13, batch 5500, loss[loss=0.1824, simple_loss=0.2591, pruned_loss=0.05286, over 19325.00 frames. ], tot_loss[loss=0.2264, simple_loss=0.3018, pruned_loss=0.07547, over 3844247.43 frames. ], batch size: 44, lr: 6.32e-03, grad_scale: 8.0 +2023-04-02 01:04:47,836 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 01:05:21,394 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87479.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:05:21,469 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87479.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 01:05:25,843 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.829e+02 5.041e+02 6.443e+02 8.590e+02 2.643e+03, threshold=1.289e+03, percent-clipped=7.0 +2023-04-02 01:05:29,326 INFO [train.py:903] (0/4) Epoch 13, batch 5550, loss[loss=0.2006, simple_loss=0.2727, pruned_loss=0.06423, over 19757.00 frames. ], tot_loss[loss=0.2271, simple_loss=0.3025, pruned_loss=0.07587, over 3847562.76 frames. ], batch size: 46, lr: 6.32e-03, grad_scale: 8.0 +2023-04-02 01:05:31,735 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8456, 4.2718, 4.5456, 4.5255, 1.7484, 4.2466, 3.6385, 4.2012], + device='cuda:0'), covar=tensor([0.1428, 0.0872, 0.0558, 0.0624, 0.5445, 0.0813, 0.0682, 0.1090], + device='cuda:0'), in_proj_covar=tensor([0.0709, 0.0630, 0.0837, 0.0716, 0.0752, 0.0582, 0.0504, 0.0773], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 01:05:33,835 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 01:05:58,112 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4437, 0.9336, 1.1110, 1.4564, 2.7300, 1.0518, 2.1287, 3.2692], + device='cuda:0'), covar=tensor([0.0658, 0.3934, 0.3828, 0.2302, 0.1358, 0.3053, 0.1579, 0.0477], + device='cuda:0'), in_proj_covar=tensor([0.0366, 0.0345, 0.0356, 0.0324, 0.0352, 0.0333, 0.0343, 0.0366], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:06:12,062 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8193, 1.4198, 1.6205, 1.5082, 3.3987, 1.0269, 2.3162, 3.7926], + device='cuda:0'), covar=tensor([0.0407, 0.2513, 0.2586, 0.1867, 0.0658, 0.2527, 0.1312, 0.0209], + device='cuda:0'), in_proj_covar=tensor([0.0365, 0.0345, 0.0355, 0.0324, 0.0352, 0.0332, 0.0343, 0.0365], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:06:16,481 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87523.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:06:23,053 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 01:06:33,781 INFO [train.py:903] (0/4) Epoch 13, batch 5600, loss[loss=0.2417, simple_loss=0.3082, pruned_loss=0.08755, over 19344.00 frames. ], tot_loss[loss=0.2271, simple_loss=0.3025, pruned_loss=0.0758, over 3834455.03 frames. ], batch size: 66, lr: 6.32e-03, grad_scale: 8.0 +2023-04-02 01:06:47,615 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87547.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:06:50,187 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.54 vs. limit=2.0 +2023-04-02 01:07:17,509 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87572.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:07:32,977 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.526e+02 5.023e+02 6.222e+02 8.051e+02 1.328e+03, threshold=1.244e+03, percent-clipped=2.0 +2023-04-02 01:07:36,350 INFO [train.py:903] (0/4) Epoch 13, batch 5650, loss[loss=0.1984, simple_loss=0.2714, pruned_loss=0.06276, over 19360.00 frames. ], tot_loss[loss=0.2273, simple_loss=0.3024, pruned_loss=0.07608, over 3839276.79 frames. ], batch size: 47, lr: 6.32e-03, grad_scale: 8.0 +2023-04-02 01:07:38,473 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.45 vs. limit=2.0 +2023-04-02 01:07:39,053 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8353, 1.2089, 0.9736, 0.9208, 1.0584, 0.8923, 0.8982, 1.1702], + device='cuda:0'), covar=tensor([0.0581, 0.0735, 0.0930, 0.0586, 0.0492, 0.1053, 0.0483, 0.0404], + device='cuda:0'), in_proj_covar=tensor([0.0292, 0.0305, 0.0330, 0.0251, 0.0241, 0.0324, 0.0292, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:07:43,440 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87592.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:07:56,265 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87602.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:08:01,807 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87607.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:08:06,352 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87611.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:08:23,588 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=87623.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:08:24,450 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 01:08:38,517 INFO [train.py:903] (0/4) Epoch 13, batch 5700, loss[loss=0.2256, simple_loss=0.3027, pruned_loss=0.07424, over 19869.00 frames. ], tot_loss[loss=0.2279, simple_loss=0.3031, pruned_loss=0.07635, over 3838031.51 frames. ], batch size: 52, lr: 6.31e-03, grad_scale: 8.0 +2023-04-02 01:08:38,942 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87636.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:08:53,309 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-04-02 01:09:36,750 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.413e+02 4.997e+02 6.690e+02 8.841e+02 1.921e+03, threshold=1.338e+03, percent-clipped=6.0 +2023-04-02 01:09:40,130 INFO [train.py:903] (0/4) Epoch 13, batch 5750, loss[loss=0.2536, simple_loss=0.3261, pruned_loss=0.09049, over 19595.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.3029, pruned_loss=0.07698, over 3834319.21 frames. ], batch size: 57, lr: 6.31e-03, grad_scale: 8.0 +2023-04-02 01:09:41,288 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 01:09:49,517 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 01:09:55,218 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 01:10:07,744 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87707.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:10:27,823 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87724.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:10:37,291 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4593, 2.2562, 1.6949, 1.4550, 2.1523, 1.1797, 1.2888, 1.9755], + device='cuda:0'), covar=tensor([0.0967, 0.0767, 0.0978, 0.0754, 0.0453, 0.1237, 0.0702, 0.0405], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0307, 0.0332, 0.0252, 0.0241, 0.0326, 0.0294, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:10:41,787 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87735.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 01:10:42,502 INFO [train.py:903] (0/4) Epoch 13, batch 5800, loss[loss=0.2788, simple_loss=0.3525, pruned_loss=0.1026, over 18102.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.3035, pruned_loss=0.07718, over 3818917.78 frames. ], batch size: 83, lr: 6.31e-03, grad_scale: 8.0 +2023-04-02 01:10:46,844 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-04-02 01:11:07,375 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.76 vs. limit=5.0 +2023-04-02 01:11:13,107 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87760.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 01:11:27,850 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.3209, 5.2084, 6.0980, 6.0760, 1.7597, 5.7875, 4.9489, 5.7577], + device='cuda:0'), covar=tensor([0.1405, 0.0594, 0.0463, 0.0469, 0.5638, 0.0432, 0.0490, 0.0928], + device='cuda:0'), in_proj_covar=tensor([0.0703, 0.0625, 0.0836, 0.0714, 0.0745, 0.0581, 0.0502, 0.0765], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 01:11:43,250 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.294e+02 5.041e+02 6.242e+02 7.928e+02 1.581e+03, threshold=1.248e+03, percent-clipped=4.0 +2023-04-02 01:11:46,962 INFO [train.py:903] (0/4) Epoch 13, batch 5850, loss[loss=0.2151, simple_loss=0.2832, pruned_loss=0.07344, over 19853.00 frames. ], tot_loss[loss=0.227, simple_loss=0.3016, pruned_loss=0.07624, over 3819086.69 frames. ], batch size: 52, lr: 6.31e-03, grad_scale: 8.0 +2023-04-02 01:12:18,351 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1843, 1.1713, 1.1337, 1.2794, 1.0927, 1.3202, 1.3631, 1.2002], + device='cuda:0'), covar=tensor([0.0834, 0.0874, 0.1011, 0.0673, 0.0792, 0.0745, 0.0764, 0.0716], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0223, 0.0224, 0.0242, 0.0229, 0.0211, 0.0191, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 01:12:33,158 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87823.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:12:47,993 INFO [train.py:903] (0/4) Epoch 13, batch 5900, loss[loss=0.2261, simple_loss=0.2924, pruned_loss=0.07986, over 19684.00 frames. ], tot_loss[loss=0.2285, simple_loss=0.3028, pruned_loss=0.07714, over 3813996.89 frames. ], batch size: 53, lr: 6.31e-03, grad_scale: 8.0 +2023-04-02 01:12:49,020 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 01:12:51,584 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87839.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:13:03,839 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4151, 1.2615, 1.0899, 1.2940, 1.0645, 1.1616, 1.0718, 1.2440], + device='cuda:0'), covar=tensor([0.0885, 0.0926, 0.1341, 0.0814, 0.0997, 0.0597, 0.1241, 0.0794], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0353, 0.0296, 0.0239, 0.0295, 0.0243, 0.0284, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:13:10,522 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 01:13:27,506 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87867.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:13:46,999 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.002e+02 5.447e+02 6.662e+02 8.353e+02 2.279e+03, threshold=1.332e+03, percent-clipped=6.0 +2023-04-02 01:13:50,406 INFO [train.py:903] (0/4) Epoch 13, batch 5950, loss[loss=0.2869, simple_loss=0.3461, pruned_loss=0.1138, over 19679.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.3027, pruned_loss=0.07704, over 3809876.59 frames. ], batch size: 60, lr: 6.31e-03, grad_scale: 8.0 +2023-04-02 01:14:30,753 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7056, 1.5078, 1.4579, 2.1680, 1.7481, 2.0095, 2.0234, 1.8076], + device='cuda:0'), covar=tensor([0.0782, 0.0900, 0.1015, 0.0725, 0.0782, 0.0668, 0.0817, 0.0628], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0223, 0.0224, 0.0241, 0.0227, 0.0210, 0.0191, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 01:14:53,067 INFO [train.py:903] (0/4) Epoch 13, batch 6000, loss[loss=0.216, simple_loss=0.2979, pruned_loss=0.06703, over 19522.00 frames. ], tot_loss[loss=0.228, simple_loss=0.3023, pruned_loss=0.07687, over 3807785.91 frames. ], batch size: 54, lr: 6.30e-03, grad_scale: 8.0 +2023-04-02 01:14:53,068 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 01:15:06,409 INFO [train.py:937] (0/4) Epoch 13, validation: loss=0.175, simple_loss=0.2755, pruned_loss=0.03726, over 944034.00 frames. +2023-04-02 01:15:06,409 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18321MB +2023-04-02 01:15:09,180 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87938.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:15:19,417 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87946.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:15:25,217 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87951.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:15:35,018 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8376, 1.7041, 1.6015, 1.9072, 1.7309, 1.6331, 1.6109, 1.8101], + device='cuda:0'), covar=tensor([0.0836, 0.1325, 0.1263, 0.0853, 0.1132, 0.0515, 0.1180, 0.0650], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0352, 0.0296, 0.0240, 0.0296, 0.0243, 0.0285, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:15:41,077 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=87963.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:15:45,484 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=87967.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:16:04,016 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=87982.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:16:04,762 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.636e+02 5.142e+02 6.649e+02 8.424e+02 2.027e+03, threshold=1.330e+03, percent-clipped=9.0 +2023-04-02 01:16:08,327 INFO [train.py:903] (0/4) Epoch 13, batch 6050, loss[loss=0.2244, simple_loss=0.3017, pruned_loss=0.0736, over 19400.00 frames. ], tot_loss[loss=0.228, simple_loss=0.3023, pruned_loss=0.07686, over 3803340.25 frames. ], batch size: 48, lr: 6.30e-03, grad_scale: 8.0 +2023-04-02 01:16:12,361 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=87988.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:16:26,863 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-88000.pt +2023-04-02 01:17:12,303 INFO [train.py:903] (0/4) Epoch 13, batch 6100, loss[loss=0.2561, simple_loss=0.3254, pruned_loss=0.09341, over 19776.00 frames. ], tot_loss[loss=0.227, simple_loss=0.3012, pruned_loss=0.07636, over 3819336.85 frames. ], batch size: 54, lr: 6.30e-03, grad_scale: 8.0 +2023-04-02 01:17:42,722 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88061.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:17:48,664 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88066.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:17:55,244 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2232, 1.3042, 1.5078, 1.3933, 2.1603, 1.9648, 2.2007, 0.6854], + device='cuda:0'), covar=tensor([0.2207, 0.3755, 0.2339, 0.1827, 0.1386, 0.1928, 0.1317, 0.3987], + device='cuda:0'), in_proj_covar=tensor([0.0499, 0.0589, 0.0629, 0.0447, 0.0600, 0.0496, 0.0647, 0.0507], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 01:18:09,952 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88082.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:18:10,750 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.507e+02 5.311e+02 6.775e+02 8.712e+02 1.953e+03, threshold=1.355e+03, percent-clipped=3.0 +2023-04-02 01:18:14,178 INFO [train.py:903] (0/4) Epoch 13, batch 6150, loss[loss=0.27, simple_loss=0.335, pruned_loss=0.1026, over 18741.00 frames. ], tot_loss[loss=0.2292, simple_loss=0.3033, pruned_loss=0.07756, over 3805766.02 frames. ], batch size: 74, lr: 6.30e-03, grad_scale: 8.0 +2023-04-02 01:18:24,908 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88095.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:18:38,384 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 01:18:54,515 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=88118.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:18:56,799 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88120.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:19:15,204 INFO [train.py:903] (0/4) Epoch 13, batch 6200, loss[loss=0.3075, simple_loss=0.3538, pruned_loss=0.1307, over 13212.00 frames. ], tot_loss[loss=0.2307, simple_loss=0.3043, pruned_loss=0.07855, over 3785571.70 frames. ], batch size: 136, lr: 6.30e-03, grad_scale: 8.0 +2023-04-02 01:19:29,261 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9429, 4.3541, 4.6461, 4.6195, 1.7381, 4.2798, 3.7290, 4.3148], + device='cuda:0'), covar=tensor([0.1418, 0.0673, 0.0576, 0.0578, 0.5339, 0.0717, 0.0612, 0.1112], + device='cuda:0'), in_proj_covar=tensor([0.0696, 0.0622, 0.0826, 0.0705, 0.0739, 0.0574, 0.0496, 0.0758], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 01:19:32,520 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2948, 3.7761, 3.9033, 3.9039, 1.5674, 3.6868, 3.2135, 3.6043], + device='cuda:0'), covar=tensor([0.1470, 0.0806, 0.0669, 0.0668, 0.5172, 0.0847, 0.0672, 0.1187], + device='cuda:0'), in_proj_covar=tensor([0.0696, 0.0622, 0.0825, 0.0705, 0.0739, 0.0574, 0.0495, 0.0757], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 01:19:49,133 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-04-02 01:20:13,334 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.409e+02 5.304e+02 6.722e+02 8.627e+02 2.252e+03, threshold=1.344e+03, percent-clipped=3.0 +2023-04-02 01:20:16,890 INFO [train.py:903] (0/4) Epoch 13, batch 6250, loss[loss=0.2022, simple_loss=0.2836, pruned_loss=0.06041, over 19748.00 frames. ], tot_loss[loss=0.2309, simple_loss=0.3044, pruned_loss=0.07873, over 3771122.95 frames. ], batch size: 51, lr: 6.29e-03, grad_scale: 8.0 +2023-04-02 01:20:29,078 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88194.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:20:44,745 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 01:20:58,225 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88219.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:21:20,740 INFO [train.py:903] (0/4) Epoch 13, batch 6300, loss[loss=0.2021, simple_loss=0.2716, pruned_loss=0.06629, over 19389.00 frames. ], tot_loss[loss=0.2304, simple_loss=0.3044, pruned_loss=0.0782, over 3794436.76 frames. ], batch size: 47, lr: 6.29e-03, grad_scale: 8.0 +2023-04-02 01:21:23,572 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88238.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:21:32,847 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.10 vs. limit=5.0 +2023-04-02 01:21:52,933 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88263.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:22:18,337 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.601e+02 5.052e+02 6.199e+02 7.712e+02 1.989e+03, threshold=1.240e+03, percent-clipped=5.0 +2023-04-02 01:22:21,909 INFO [train.py:903] (0/4) Epoch 13, batch 6350, loss[loss=0.2402, simple_loss=0.307, pruned_loss=0.08674, over 19470.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.3033, pruned_loss=0.07727, over 3815806.04 frames. ], batch size: 49, lr: 6.29e-03, grad_scale: 8.0 +2023-04-02 01:22:32,660 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6239, 4.1742, 2.6257, 3.7717, 1.1780, 3.8756, 3.9865, 4.0648], + device='cuda:0'), covar=tensor([0.0573, 0.1032, 0.2095, 0.0753, 0.3907, 0.0814, 0.0815, 0.1115], + device='cuda:0'), in_proj_covar=tensor([0.0444, 0.0374, 0.0451, 0.0321, 0.0386, 0.0380, 0.0373, 0.0408], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:22:50,064 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=88309.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 01:23:02,580 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88317.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:23:08,247 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88322.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:23:24,604 INFO [train.py:903] (0/4) Epoch 13, batch 6400, loss[loss=0.2107, simple_loss=0.2888, pruned_loss=0.06632, over 19583.00 frames. ], tot_loss[loss=0.2283, simple_loss=0.303, pruned_loss=0.07677, over 3814692.58 frames. ], batch size: 52, lr: 6.29e-03, grad_scale: 8.0 +2023-04-02 01:23:27,262 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88338.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:23:31,723 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88342.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:23:38,151 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88347.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:23:58,464 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88363.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:23:58,527 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3549, 1.4061, 1.6382, 1.5578, 2.4370, 2.1243, 2.5146, 0.8577], + device='cuda:0'), covar=tensor([0.2116, 0.3757, 0.2231, 0.1701, 0.1364, 0.1877, 0.1244, 0.3867], + device='cuda:0'), in_proj_covar=tensor([0.0499, 0.0592, 0.0629, 0.0448, 0.0603, 0.0499, 0.0645, 0.0505], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 01:24:16,128 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8951, 1.9925, 2.2846, 2.7205, 1.8594, 2.6217, 2.4216, 2.0523], + device='cuda:0'), covar=tensor([0.3969, 0.3450, 0.1571, 0.1973, 0.3666, 0.1600, 0.3851, 0.2900], + device='cuda:0'), in_proj_covar=tensor([0.0817, 0.0848, 0.0662, 0.0901, 0.0798, 0.0733, 0.0800, 0.0726], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 01:24:18,740 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-04-02 01:24:22,198 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.590e+02 5.639e+02 6.967e+02 8.699e+02 1.659e+03, threshold=1.393e+03, percent-clipped=7.0 +2023-04-02 01:24:25,785 INFO [train.py:903] (0/4) Epoch 13, batch 6450, loss[loss=0.2262, simple_loss=0.3177, pruned_loss=0.06732, over 19678.00 frames. ], tot_loss[loss=0.2283, simple_loss=0.3029, pruned_loss=0.07679, over 3806368.89 frames. ], batch size: 60, lr: 6.29e-03, grad_scale: 8.0 +2023-04-02 01:25:05,097 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 01:25:18,090 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7646, 1.8677, 2.0821, 2.3488, 1.6019, 2.2601, 2.2448, 1.9010], + device='cuda:0'), covar=tensor([0.3915, 0.3343, 0.1753, 0.1933, 0.3601, 0.1717, 0.4135, 0.3024], + device='cuda:0'), in_proj_covar=tensor([0.0813, 0.0845, 0.0661, 0.0898, 0.0796, 0.0729, 0.0797, 0.0722], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 01:25:28,933 INFO [train.py:903] (0/4) Epoch 13, batch 6500, loss[loss=0.1844, simple_loss=0.2577, pruned_loss=0.05562, over 19135.00 frames. ], tot_loss[loss=0.2287, simple_loss=0.3037, pruned_loss=0.07686, over 3810957.76 frames. ], batch size: 42, lr: 6.29e-03, grad_scale: 8.0 +2023-04-02 01:25:30,993 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 01:26:00,474 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88462.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:26:26,788 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.125e+02 5.190e+02 6.145e+02 7.751e+02 1.614e+03, threshold=1.229e+03, percent-clipped=4.0 +2023-04-02 01:26:30,119 INFO [train.py:903] (0/4) Epoch 13, batch 6550, loss[loss=0.2651, simple_loss=0.3176, pruned_loss=0.1063, over 19744.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.3029, pruned_loss=0.07689, over 3806854.49 frames. ], batch size: 47, lr: 6.28e-03, grad_scale: 8.0 +2023-04-02 01:27:11,705 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1623, 1.8123, 1.4868, 1.2058, 1.6513, 1.1885, 1.2074, 1.6742], + device='cuda:0'), covar=tensor([0.0725, 0.0768, 0.0982, 0.0697, 0.0454, 0.1102, 0.0532, 0.0345], + device='cuda:0'), in_proj_covar=tensor([0.0292, 0.0304, 0.0326, 0.0249, 0.0239, 0.0321, 0.0293, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:27:31,540 INFO [train.py:903] (0/4) Epoch 13, batch 6600, loss[loss=0.2416, simple_loss=0.3191, pruned_loss=0.08203, over 19472.00 frames. ], tot_loss[loss=0.2284, simple_loss=0.3035, pruned_loss=0.07671, over 3808707.94 frames. ], batch size: 64, lr: 6.28e-03, grad_scale: 8.0 +2023-04-02 01:28:21,373 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88577.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:28:28,438 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.648e+02 5.175e+02 6.632e+02 9.239e+02 1.861e+03, threshold=1.326e+03, percent-clipped=7.0 +2023-04-02 01:28:29,320 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 01:28:32,111 INFO [train.py:903] (0/4) Epoch 13, batch 6650, loss[loss=0.1962, simple_loss=0.2689, pruned_loss=0.06174, over 19329.00 frames. ], tot_loss[loss=0.2289, simple_loss=0.3035, pruned_loss=0.07721, over 3804656.39 frames. ], batch size: 44, lr: 6.28e-03, grad_scale: 8.0 +2023-04-02 01:28:42,994 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5322, 2.3783, 1.7777, 1.3435, 2.2209, 1.2342, 1.3241, 2.0148], + device='cuda:0'), covar=tensor([0.1072, 0.0632, 0.0964, 0.0824, 0.0483, 0.1219, 0.0750, 0.0432], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0307, 0.0330, 0.0252, 0.0240, 0.0326, 0.0297, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:29:33,913 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6075, 1.6666, 1.9604, 1.7632, 3.1283, 2.5074, 3.2919, 1.8149], + device='cuda:0'), covar=tensor([0.2184, 0.3922, 0.2374, 0.1736, 0.1371, 0.1878, 0.1401, 0.3456], + device='cuda:0'), in_proj_covar=tensor([0.0502, 0.0592, 0.0629, 0.0447, 0.0604, 0.0497, 0.0648, 0.0507], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 01:29:34,644 INFO [train.py:903] (0/4) Epoch 13, batch 6700, loss[loss=0.2063, simple_loss=0.2918, pruned_loss=0.06041, over 17457.00 frames. ], tot_loss[loss=0.2288, simple_loss=0.3036, pruned_loss=0.07698, over 3805277.44 frames. ], batch size: 102, lr: 6.28e-03, grad_scale: 8.0 +2023-04-02 01:29:55,527 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=88653.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 01:29:55,691 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5721, 1.4431, 1.3786, 1.8862, 1.5208, 2.0153, 1.9459, 1.8019], + device='cuda:0'), covar=tensor([0.0820, 0.0916, 0.1030, 0.0844, 0.0846, 0.0650, 0.0800, 0.0617], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0223, 0.0222, 0.0242, 0.0227, 0.0209, 0.0192, 0.0200], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 01:30:29,681 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.623e+02 5.222e+02 6.682e+02 8.627e+02 1.913e+03, threshold=1.336e+03, percent-clipped=6.0 +2023-04-02 01:30:33,139 INFO [train.py:903] (0/4) Epoch 13, batch 6750, loss[loss=0.1821, simple_loss=0.2656, pruned_loss=0.04932, over 19488.00 frames. ], tot_loss[loss=0.2285, simple_loss=0.3032, pruned_loss=0.07691, over 3804172.50 frames. ], batch size: 49, lr: 6.28e-03, grad_scale: 8.0 +2023-04-02 01:31:31,219 INFO [train.py:903] (0/4) Epoch 13, batch 6800, loss[loss=0.2368, simple_loss=0.3195, pruned_loss=0.077, over 19342.00 frames. ], tot_loss[loss=0.228, simple_loss=0.303, pruned_loss=0.07655, over 3798708.81 frames. ], batch size: 70, lr: 6.28e-03, grad_scale: 8.0 +2023-04-02 01:32:01,944 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-13.pt +2023-04-02 01:32:18,413 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 01:32:18,907 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 01:32:21,845 INFO [train.py:903] (0/4) Epoch 14, batch 0, loss[loss=0.3079, simple_loss=0.3756, pruned_loss=0.1201, over 19586.00 frames. ], tot_loss[loss=0.3079, simple_loss=0.3756, pruned_loss=0.1201, over 19586.00 frames. ], batch size: 61, lr: 6.05e-03, grad_scale: 8.0 +2023-04-02 01:32:21,846 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 01:32:33,656 INFO [train.py:937] (0/4) Epoch 14, validation: loss=0.1763, simple_loss=0.2772, pruned_loss=0.03773, over 944034.00 frames. +2023-04-02 01:32:33,660 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 01:32:41,835 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=88768.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 01:32:49,786 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 01:32:59,173 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.227e+02 4.998e+02 6.377e+02 7.989e+02 1.719e+03, threshold=1.275e+03, percent-clipped=2.0 +2023-04-02 01:33:10,437 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-04-02 01:33:40,052 INFO [train.py:903] (0/4) Epoch 14, batch 50, loss[loss=0.1951, simple_loss=0.2799, pruned_loss=0.05516, over 19683.00 frames. ], tot_loss[loss=0.221, simple_loss=0.298, pruned_loss=0.072, over 862967.96 frames. ], batch size: 53, lr: 6.04e-03, grad_scale: 8.0 +2023-04-02 01:34:01,734 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=88833.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:34:15,373 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 01:34:34,180 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=88858.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:34:40,591 INFO [train.py:903] (0/4) Epoch 14, batch 100, loss[loss=0.2077, simple_loss=0.2884, pruned_loss=0.0635, over 19708.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.3007, pruned_loss=0.07393, over 1524846.09 frames. ], batch size: 53, lr: 6.04e-03, grad_scale: 8.0 +2023-04-02 01:34:51,071 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 01:35:02,787 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.971e+02 5.271e+02 6.621e+02 8.700e+02 2.391e+03, threshold=1.324e+03, percent-clipped=10.0 +2023-04-02 01:35:41,065 INFO [train.py:903] (0/4) Epoch 14, batch 150, loss[loss=0.1663, simple_loss=0.2469, pruned_loss=0.04285, over 19412.00 frames. ], tot_loss[loss=0.2232, simple_loss=0.2994, pruned_loss=0.07353, over 2049630.45 frames. ], batch size: 48, lr: 6.04e-03, grad_scale: 8.0 +2023-04-02 01:35:53,873 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2632, 1.1529, 1.2411, 1.3456, 1.0275, 1.3309, 1.3268, 1.2576], + device='cuda:0'), covar=tensor([0.0850, 0.1056, 0.1071, 0.0691, 0.0836, 0.0853, 0.0856, 0.0775], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0223, 0.0225, 0.0244, 0.0228, 0.0210, 0.0193, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 01:36:00,295 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.45 vs. limit=2.0 +2023-04-02 01:36:29,581 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0143, 1.4144, 1.8775, 1.5582, 2.9721, 4.3099, 4.2703, 4.7918], + device='cuda:0'), covar=tensor([0.1748, 0.3460, 0.3114, 0.2010, 0.0546, 0.0186, 0.0182, 0.0144], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0301, 0.0331, 0.0251, 0.0220, 0.0165, 0.0207, 0.0217], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 01:36:38,764 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 01:36:39,942 INFO [train.py:903] (0/4) Epoch 14, batch 200, loss[loss=0.2337, simple_loss=0.3156, pruned_loss=0.07593, over 19805.00 frames. ], tot_loss[loss=0.224, simple_loss=0.2999, pruned_loss=0.0741, over 2442609.13 frames. ], batch size: 56, lr: 6.04e-03, grad_scale: 8.0 +2023-04-02 01:37:03,948 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.747e+02 5.170e+02 6.545e+02 8.695e+02 1.666e+03, threshold=1.309e+03, percent-clipped=4.0 +2023-04-02 01:37:41,107 INFO [train.py:903] (0/4) Epoch 14, batch 250, loss[loss=0.1846, simple_loss=0.2765, pruned_loss=0.04639, over 19667.00 frames. ], tot_loss[loss=0.2246, simple_loss=0.3003, pruned_loss=0.07441, over 2737992.99 frames. ], batch size: 55, lr: 6.04e-03, grad_scale: 8.0 +2023-04-02 01:37:53,626 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89024.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 01:38:17,611 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89045.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:38:22,447 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89049.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 01:38:43,490 INFO [train.py:903] (0/4) Epoch 14, batch 300, loss[loss=0.2156, simple_loss=0.3074, pruned_loss=0.0619, over 19598.00 frames. ], tot_loss[loss=0.2239, simple_loss=0.2999, pruned_loss=0.07391, over 2981558.45 frames. ], batch size: 57, lr: 6.04e-03, grad_scale: 8.0 +2023-04-02 01:38:52,772 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7500, 4.2492, 4.4321, 4.4143, 1.5042, 4.1380, 3.6056, 4.1103], + device='cuda:0'), covar=tensor([0.1427, 0.0729, 0.0576, 0.0615, 0.5519, 0.0746, 0.0649, 0.1095], + device='cuda:0'), in_proj_covar=tensor([0.0707, 0.0630, 0.0836, 0.0715, 0.0753, 0.0583, 0.0504, 0.0764], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 01:39:05,434 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.258e+02 4.830e+02 5.923e+02 7.544e+02 2.111e+03, threshold=1.185e+03, percent-clipped=1.0 +2023-04-02 01:39:45,147 INFO [train.py:903] (0/4) Epoch 14, batch 350, loss[loss=0.292, simple_loss=0.3477, pruned_loss=0.1182, over 13378.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.3004, pruned_loss=0.07389, over 3169929.50 frames. ], batch size: 136, lr: 6.03e-03, grad_scale: 8.0 +2023-04-02 01:39:47,483 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 01:40:29,897 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2057, 2.2215, 2.4204, 3.1960, 2.1847, 3.0162, 2.6659, 2.1767], + device='cuda:0'), covar=tensor([0.3892, 0.3621, 0.1579, 0.2122, 0.3990, 0.1713, 0.3685, 0.2832], + device='cuda:0'), in_proj_covar=tensor([0.0824, 0.0859, 0.0666, 0.0908, 0.0807, 0.0738, 0.0811, 0.0728], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 01:40:46,839 INFO [train.py:903] (0/4) Epoch 14, batch 400, loss[loss=0.2608, simple_loss=0.3274, pruned_loss=0.09709, over 19670.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.301, pruned_loss=0.07439, over 3313624.44 frames. ], batch size: 60, lr: 6.03e-03, grad_scale: 8.0 +2023-04-02 01:41:11,936 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.999e+02 4.893e+02 5.942e+02 7.582e+02 1.529e+03, threshold=1.188e+03, percent-clipped=4.0 +2023-04-02 01:41:16,770 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89187.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:41:36,314 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3857, 2.1291, 2.0699, 2.4972, 2.2833, 2.1417, 2.0233, 2.5806], + device='cuda:0'), covar=tensor([0.0872, 0.1550, 0.1324, 0.0914, 0.1261, 0.0473, 0.1147, 0.0583], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0353, 0.0297, 0.0240, 0.0298, 0.0243, 0.0285, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:41:47,623 INFO [train.py:903] (0/4) Epoch 14, batch 450, loss[loss=0.2242, simple_loss=0.3024, pruned_loss=0.07295, over 19478.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.3007, pruned_loss=0.07375, over 3431116.92 frames. ], batch size: 49, lr: 6.03e-03, grad_scale: 8.0 +2023-04-02 01:42:19,958 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 01:42:20,928 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 01:42:51,726 INFO [train.py:903] (0/4) Epoch 14, batch 500, loss[loss=0.236, simple_loss=0.3085, pruned_loss=0.08172, over 19494.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.3012, pruned_loss=0.07454, over 3525901.39 frames. ], batch size: 49, lr: 6.03e-03, grad_scale: 8.0 +2023-04-02 01:43:04,783 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7282, 1.5057, 1.3585, 1.6122, 1.4770, 1.3617, 1.3336, 1.5693], + device='cuda:0'), covar=tensor([0.1042, 0.1327, 0.1560, 0.1001, 0.1223, 0.0729, 0.1479, 0.0834], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0350, 0.0295, 0.0239, 0.0295, 0.0241, 0.0285, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:43:13,298 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.040e+02 5.086e+02 6.501e+02 8.394e+02 1.936e+03, threshold=1.300e+03, percent-clipped=6.0 +2023-04-02 01:43:51,231 INFO [train.py:903] (0/4) Epoch 14, batch 550, loss[loss=0.2364, simple_loss=0.3128, pruned_loss=0.08005, over 19662.00 frames. ], tot_loss[loss=0.2246, simple_loss=0.3001, pruned_loss=0.07455, over 3599090.50 frames. ], batch size: 60, lr: 6.03e-03, grad_scale: 8.0 +2023-04-02 01:44:05,920 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-02 01:44:50,891 INFO [train.py:903] (0/4) Epoch 14, batch 600, loss[loss=0.2613, simple_loss=0.3329, pruned_loss=0.09488, over 19701.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2993, pruned_loss=0.07412, over 3639423.30 frames. ], batch size: 59, lr: 6.03e-03, grad_scale: 8.0 +2023-04-02 01:45:14,768 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.733e+02 5.437e+02 6.463e+02 8.394e+02 1.645e+03, threshold=1.293e+03, percent-clipped=4.0 +2023-04-02 01:45:23,617 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=89389.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:45:27,402 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89392.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:45:32,883 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 01:45:52,142 INFO [train.py:903] (0/4) Epoch 14, batch 650, loss[loss=0.2201, simple_loss=0.2995, pruned_loss=0.07035, over 19726.00 frames. ], tot_loss[loss=0.224, simple_loss=0.2998, pruned_loss=0.07407, over 3695225.61 frames. ], batch size: 51, lr: 6.02e-03, grad_scale: 8.0 +2023-04-02 01:45:54,681 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.9534, 5.2895, 2.7640, 4.8207, 0.9903, 5.3211, 5.3211, 5.4952], + device='cuda:0'), covar=tensor([0.0426, 0.0929, 0.2140, 0.0674, 0.4413, 0.0576, 0.0703, 0.0782], + device='cuda:0'), in_proj_covar=tensor([0.0452, 0.0379, 0.0456, 0.0332, 0.0394, 0.0386, 0.0383, 0.0413], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:46:06,882 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-04-02 01:46:29,617 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89443.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:46:45,791 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89457.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 01:46:53,908 INFO [train.py:903] (0/4) Epoch 14, batch 700, loss[loss=0.2305, simple_loss=0.3082, pruned_loss=0.07638, over 19751.00 frames. ], tot_loss[loss=0.2235, simple_loss=0.2993, pruned_loss=0.07384, over 3733615.72 frames. ], batch size: 54, lr: 6.02e-03, grad_scale: 8.0 +2023-04-02 01:47:21,078 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.092e+02 5.132e+02 6.716e+02 8.076e+02 1.293e+03, threshold=1.343e+03, percent-clipped=1.0 +2023-04-02 01:47:44,491 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89504.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:47:59,821 INFO [train.py:903] (0/4) Epoch 14, batch 750, loss[loss=0.2692, simple_loss=0.3288, pruned_loss=0.1048, over 18285.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2988, pruned_loss=0.07365, over 3754554.21 frames. ], batch size: 83, lr: 6.02e-03, grad_scale: 8.0 +2023-04-02 01:48:14,781 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89527.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:48:19,166 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=89531.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:49:01,550 INFO [train.py:903] (0/4) Epoch 14, batch 800, loss[loss=0.2775, simple_loss=0.349, pruned_loss=0.103, over 19330.00 frames. ], tot_loss[loss=0.2239, simple_loss=0.2994, pruned_loss=0.07417, over 3770767.65 frames. ], batch size: 70, lr: 6.02e-03, grad_scale: 8.0 +2023-04-02 01:49:16,468 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 01:49:24,433 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.547e+02 5.343e+02 6.523e+02 8.164e+02 1.688e+03, threshold=1.305e+03, percent-clipped=4.0 +2023-04-02 01:49:50,617 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5913, 1.6880, 1.8252, 2.0130, 1.3669, 1.8520, 1.9285, 1.6917], + device='cuda:0'), covar=tensor([0.3580, 0.2980, 0.1614, 0.1848, 0.3286, 0.1628, 0.3953, 0.2815], + device='cuda:0'), in_proj_covar=tensor([0.0813, 0.0851, 0.0662, 0.0896, 0.0798, 0.0730, 0.0803, 0.0724], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 01:50:01,793 INFO [train.py:903] (0/4) Epoch 14, batch 850, loss[loss=0.2494, simple_loss=0.3261, pruned_loss=0.08638, over 18847.00 frames. ], tot_loss[loss=0.2261, simple_loss=0.3015, pruned_loss=0.07537, over 3777591.70 frames. ], batch size: 74, lr: 6.02e-03, grad_scale: 8.0 +2023-04-02 01:50:42,876 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89646.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:50:44,471 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-04-02 01:50:52,347 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1651, 1.2613, 1.6715, 1.0317, 2.5632, 3.3268, 3.1056, 3.5606], + device='cuda:0'), covar=tensor([0.1598, 0.3550, 0.3175, 0.2262, 0.0502, 0.0191, 0.0226, 0.0244], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0302, 0.0333, 0.0253, 0.0221, 0.0166, 0.0206, 0.0216], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 01:50:55,564 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 01:51:03,553 INFO [train.py:903] (0/4) Epoch 14, batch 900, loss[loss=0.2087, simple_loss=0.2802, pruned_loss=0.06859, over 19482.00 frames. ], tot_loss[loss=0.2251, simple_loss=0.3006, pruned_loss=0.07481, over 3795246.23 frames. ], batch size: 49, lr: 6.02e-03, grad_scale: 8.0 +2023-04-02 01:51:29,667 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.240e+02 4.983e+02 6.131e+02 7.467e+02 1.791e+03, threshold=1.226e+03, percent-clipped=3.0 +2023-04-02 01:51:46,402 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89698.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:52:07,440 INFO [train.py:903] (0/4) Epoch 14, batch 950, loss[loss=0.2902, simple_loss=0.3646, pruned_loss=0.1079, over 19608.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.3006, pruned_loss=0.0745, over 3812977.18 frames. ], batch size: 57, lr: 6.01e-03, grad_scale: 8.0 +2023-04-02 01:52:07,503 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 01:52:33,950 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=89736.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:52:47,338 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0 +2023-04-02 01:53:06,405 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89760.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:53:10,804 INFO [train.py:903] (0/4) Epoch 14, batch 1000, loss[loss=0.2385, simple_loss=0.3175, pruned_loss=0.07974, over 19337.00 frames. ], tot_loss[loss=0.2266, simple_loss=0.3022, pruned_loss=0.07546, over 3799486.05 frames. ], batch size: 66, lr: 6.01e-03, grad_scale: 8.0 +2023-04-02 01:53:19,417 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6601, 1.5451, 1.5147, 2.0744, 1.6992, 1.9301, 2.0819, 1.7873], + device='cuda:0'), covar=tensor([0.0728, 0.0836, 0.0962, 0.0744, 0.0831, 0.0687, 0.0746, 0.0613], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0222, 0.0223, 0.0242, 0.0229, 0.0209, 0.0192, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 01:53:34,646 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.523e+02 5.300e+02 6.720e+02 8.420e+02 1.635e+03, threshold=1.344e+03, percent-clipped=5.0 +2023-04-02 01:53:36,373 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89785.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:53:38,545 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=89787.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:53:59,141 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=89801.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 01:54:02,274 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 01:54:13,787 INFO [train.py:903] (0/4) Epoch 14, batch 1050, loss[loss=0.2614, simple_loss=0.3269, pruned_loss=0.09795, over 17573.00 frames. ], tot_loss[loss=0.227, simple_loss=0.3026, pruned_loss=0.07575, over 3801561.66 frames. ], batch size: 101, lr: 6.01e-03, grad_scale: 8.0 +2023-04-02 01:54:21,209 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6783, 3.3454, 2.5037, 3.0358, 1.4623, 3.2034, 3.1465, 3.2540], + device='cuda:0'), covar=tensor([0.0908, 0.1069, 0.2012, 0.0852, 0.2986, 0.0872, 0.0912, 0.1257], + device='cuda:0'), in_proj_covar=tensor([0.0446, 0.0372, 0.0447, 0.0325, 0.0390, 0.0381, 0.0373, 0.0408], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 01:54:46,491 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 01:55:02,247 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89851.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:55:16,496 INFO [train.py:903] (0/4) Epoch 14, batch 1100, loss[loss=0.2184, simple_loss=0.2802, pruned_loss=0.07829, over 19028.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.3014, pruned_loss=0.07525, over 3806638.55 frames. ], batch size: 42, lr: 6.01e-03, grad_scale: 8.0 +2023-04-02 01:55:24,734 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=89871.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:55:43,015 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=89883.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:55:43,842 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.219e+02 5.450e+02 6.738e+02 9.006e+02 2.173e+03, threshold=1.348e+03, percent-clipped=6.0 +2023-04-02 01:56:05,591 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89902.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:56:05,661 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=89902.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:56:19,138 INFO [train.py:903] (0/4) Epoch 14, batch 1150, loss[loss=0.2008, simple_loss=0.282, pruned_loss=0.05976, over 19461.00 frames. ], tot_loss[loss=0.2264, simple_loss=0.3018, pruned_loss=0.07555, over 3813092.95 frames. ], batch size: 49, lr: 6.01e-03, grad_scale: 8.0 +2023-04-02 01:56:22,861 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89916.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 01:56:39,102 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=89927.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:57:25,203 INFO [train.py:903] (0/4) Epoch 14, batch 1200, loss[loss=0.2267, simple_loss=0.3035, pruned_loss=0.07495, over 19747.00 frames. ], tot_loss[loss=0.2275, simple_loss=0.3028, pruned_loss=0.07614, over 3801976.56 frames. ], batch size: 51, lr: 6.01e-03, grad_scale: 8.0 +2023-04-02 01:57:49,396 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.071e+02 4.889e+02 5.798e+02 7.146e+02 1.027e+03, threshold=1.160e+03, percent-clipped=0.0 +2023-04-02 01:57:51,958 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=89986.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:57:55,100 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 01:58:10,200 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-90000.pt +2023-04-02 01:58:28,857 INFO [train.py:903] (0/4) Epoch 14, batch 1250, loss[loss=0.1912, simple_loss=0.2757, pruned_loss=0.05331, over 19768.00 frames. ], tot_loss[loss=0.2274, simple_loss=0.3027, pruned_loss=0.07603, over 3810441.57 frames. ], batch size: 54, lr: 6.00e-03, grad_scale: 8.0 +2023-04-02 01:59:04,891 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90042.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 01:59:31,268 INFO [train.py:903] (0/4) Epoch 14, batch 1300, loss[loss=0.2363, simple_loss=0.3125, pruned_loss=0.08008, over 19656.00 frames. ], tot_loss[loss=0.2273, simple_loss=0.3024, pruned_loss=0.07613, over 3809877.77 frames. ], batch size: 58, lr: 6.00e-03, grad_scale: 8.0 +2023-04-02 01:59:36,504 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7719, 1.7479, 1.5446, 1.3163, 1.3465, 1.4327, 0.1913, 0.6475], + device='cuda:0'), covar=tensor([0.0499, 0.0491, 0.0309, 0.0486, 0.1000, 0.0569, 0.0909, 0.0818], + device='cuda:0'), in_proj_covar=tensor([0.0344, 0.0337, 0.0335, 0.0364, 0.0438, 0.0361, 0.0319, 0.0328], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 01:59:57,908 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.456e+02 5.410e+02 6.769e+02 7.813e+02 2.183e+03, threshold=1.354e+03, percent-clipped=2.0 +2023-04-02 02:00:02,793 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0763, 1.7652, 1.3553, 1.0965, 1.6152, 1.0058, 1.1714, 1.6383], + device='cuda:0'), covar=tensor([0.0716, 0.0676, 0.0976, 0.0679, 0.0417, 0.1162, 0.0567, 0.0374], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0307, 0.0327, 0.0250, 0.0238, 0.0326, 0.0295, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:00:25,735 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90107.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:00:33,598 INFO [train.py:903] (0/4) Epoch 14, batch 1350, loss[loss=0.2246, simple_loss=0.3056, pruned_loss=0.07178, over 19514.00 frames. ], tot_loss[loss=0.2269, simple_loss=0.3021, pruned_loss=0.07584, over 3813974.03 frames. ], batch size: 56, lr: 6.00e-03, grad_scale: 8.0 +2023-04-02 02:00:58,138 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90132.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:01:16,473 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90148.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:01:24,049 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-02 02:01:28,058 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90157.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:01:29,272 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90158.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:01:35,841 INFO [train.py:903] (0/4) Epoch 14, batch 1400, loss[loss=0.187, simple_loss=0.2609, pruned_loss=0.05649, over 19383.00 frames. ], tot_loss[loss=0.2267, simple_loss=0.3023, pruned_loss=0.07561, over 3817195.20 frames. ], batch size: 47, lr: 6.00e-03, grad_scale: 8.0 +2023-04-02 02:01:46,540 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90172.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 02:01:59,331 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90183.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:02:00,110 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.442e+02 5.450e+02 7.234e+02 1.006e+03 2.886e+03, threshold=1.447e+03, percent-clipped=11.0 +2023-04-02 02:02:16,740 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90197.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 02:02:34,949 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 02:02:38,346 INFO [train.py:903] (0/4) Epoch 14, batch 1450, loss[loss=0.242, simple_loss=0.3054, pruned_loss=0.08927, over 19747.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.3006, pruned_loss=0.07485, over 3814818.41 frames. ], batch size: 51, lr: 6.00e-03, grad_scale: 8.0 +2023-04-02 02:02:53,333 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90227.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:03:12,031 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90242.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:03:22,173 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90250.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:03:38,190 INFO [train.py:903] (0/4) Epoch 14, batch 1500, loss[loss=0.2038, simple_loss=0.2721, pruned_loss=0.06775, over 19754.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.2993, pruned_loss=0.07446, over 3830032.41 frames. ], batch size: 47, lr: 6.00e-03, grad_scale: 8.0 +2023-04-02 02:03:42,138 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90267.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:04:03,122 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.947e+02 5.072e+02 6.110e+02 7.921e+02 2.000e+03, threshold=1.222e+03, percent-clipped=2.0 +2023-04-02 02:04:09,139 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2515, 3.7439, 3.8678, 3.8572, 1.5821, 3.6171, 3.1516, 3.5910], + device='cuda:0'), covar=tensor([0.1405, 0.0932, 0.0582, 0.0672, 0.4868, 0.0818, 0.0648, 0.1052], + device='cuda:0'), in_proj_covar=tensor([0.0700, 0.0624, 0.0827, 0.0710, 0.0746, 0.0576, 0.0499, 0.0764], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 02:04:23,047 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0701, 3.5935, 1.9827, 2.1810, 3.2088, 1.7839, 1.2730, 2.0680], + device='cuda:0'), covar=tensor([0.1274, 0.0433, 0.1008, 0.0782, 0.0469, 0.1154, 0.1046, 0.0720], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0308, 0.0327, 0.0251, 0.0238, 0.0327, 0.0298, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:04:39,097 INFO [train.py:903] (0/4) Epoch 14, batch 1550, loss[loss=0.237, simple_loss=0.3172, pruned_loss=0.0784, over 19478.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.3002, pruned_loss=0.07455, over 3845936.63 frames. ], batch size: 64, lr: 5.99e-03, grad_scale: 8.0 +2023-04-02 02:05:16,362 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90342.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:05:42,760 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9058, 2.5861, 1.7429, 1.8197, 2.3749, 1.7384, 1.5032, 1.9958], + device='cuda:0'), covar=tensor([0.0917, 0.0580, 0.0755, 0.0647, 0.0432, 0.0863, 0.0732, 0.0543], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0308, 0.0327, 0.0250, 0.0239, 0.0327, 0.0297, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:05:44,735 INFO [train.py:903] (0/4) Epoch 14, batch 1600, loss[loss=0.2621, simple_loss=0.3322, pruned_loss=0.09605, over 19560.00 frames. ], tot_loss[loss=0.2253, simple_loss=0.3007, pruned_loss=0.07494, over 3819950.79 frames. ], batch size: 61, lr: 5.99e-03, grad_scale: 8.0 +2023-04-02 02:05:51,752 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90369.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:06:02,642 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 02:06:08,372 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.055e+02 5.288e+02 6.387e+02 7.705e+02 1.564e+03, threshold=1.277e+03, percent-clipped=2.0 +2023-04-02 02:06:46,534 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90413.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:06:47,284 INFO [train.py:903] (0/4) Epoch 14, batch 1650, loss[loss=0.2005, simple_loss=0.2651, pruned_loss=0.06796, over 19743.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.3001, pruned_loss=0.07475, over 3813896.82 frames. ], batch size: 46, lr: 5.99e-03, grad_scale: 8.0 +2023-04-02 02:07:14,984 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90438.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:07:47,914 INFO [train.py:903] (0/4) Epoch 14, batch 1700, loss[loss=0.1908, simple_loss=0.2702, pruned_loss=0.0557, over 19456.00 frames. ], tot_loss[loss=0.2254, simple_loss=0.3009, pruned_loss=0.07492, over 3826540.70 frames. ], batch size: 49, lr: 5.99e-03, grad_scale: 8.0 +2023-04-02 02:07:50,645 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3057, 2.3902, 2.4597, 3.1632, 2.2497, 2.9272, 2.6711, 2.4034], + device='cuda:0'), covar=tensor([0.3597, 0.3246, 0.1568, 0.2079, 0.3660, 0.1754, 0.3795, 0.2714], + device='cuda:0'), in_proj_covar=tensor([0.0817, 0.0852, 0.0665, 0.0893, 0.0797, 0.0737, 0.0800, 0.0729], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 02:08:03,304 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90477.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:08:13,896 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.443e+02 5.354e+02 6.441e+02 8.114e+02 1.316e+03, threshold=1.288e+03, percent-clipped=1.0 +2023-04-02 02:08:19,178 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-02 02:08:23,204 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90492.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:08:25,331 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 02:08:49,500 INFO [train.py:903] (0/4) Epoch 14, batch 1750, loss[loss=0.2426, simple_loss=0.3246, pruned_loss=0.08028, over 19353.00 frames. ], tot_loss[loss=0.2244, simple_loss=0.3005, pruned_loss=0.07418, over 3842857.81 frames. ], batch size: 66, lr: 5.99e-03, grad_scale: 8.0 +2023-04-02 02:08:49,818 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90514.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:09:54,005 INFO [train.py:903] (0/4) Epoch 14, batch 1800, loss[loss=0.2687, simple_loss=0.3438, pruned_loss=0.09684, over 19480.00 frames. ], tot_loss[loss=0.224, simple_loss=0.3002, pruned_loss=0.07385, over 3842069.97 frames. ], batch size: 64, lr: 5.99e-03, grad_scale: 8.0 +2023-04-02 02:10:17,889 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.96 vs. limit=5.0 +2023-04-02 02:10:18,178 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.749e+02 5.076e+02 6.157e+02 8.007e+02 1.318e+03, threshold=1.231e+03, percent-clipped=1.0 +2023-04-02 02:10:29,854 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90594.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:10:34,930 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90598.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:10:40,545 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-04-02 02:10:47,953 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90607.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:10:48,799 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 02:10:55,664 INFO [train.py:903] (0/4) Epoch 14, batch 1850, loss[loss=0.2694, simple_loss=0.3372, pruned_loss=0.1008, over 19715.00 frames. ], tot_loss[loss=0.2239, simple_loss=0.2997, pruned_loss=0.07406, over 3831678.68 frames. ], batch size: 63, lr: 5.98e-03, grad_scale: 8.0 +2023-04-02 02:11:06,609 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-02 02:11:07,313 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90623.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:11:24,438 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 02:11:30,931 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0103, 3.6155, 2.5646, 3.2585, 0.7879, 3.4912, 3.4070, 3.4908], + device='cuda:0'), covar=tensor([0.0723, 0.1070, 0.1830, 0.0892, 0.4062, 0.0816, 0.0921, 0.1071], + device='cuda:0'), in_proj_covar=tensor([0.0448, 0.0374, 0.0451, 0.0326, 0.0391, 0.0384, 0.0379, 0.0412], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:11:38,451 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-04-02 02:11:58,527 INFO [train.py:903] (0/4) Epoch 14, batch 1900, loss[loss=0.1928, simple_loss=0.2777, pruned_loss=0.05394, over 19479.00 frames. ], tot_loss[loss=0.2232, simple_loss=0.2993, pruned_loss=0.07352, over 3830669.93 frames. ], batch size: 49, lr: 5.98e-03, grad_scale: 8.0 +2023-04-02 02:12:07,987 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5638, 4.0833, 4.2414, 4.2354, 1.4699, 3.9533, 3.4817, 3.9636], + device='cuda:0'), covar=tensor([0.1529, 0.0754, 0.0580, 0.0646, 0.5451, 0.0807, 0.0632, 0.1086], + device='cuda:0'), in_proj_covar=tensor([0.0713, 0.0635, 0.0844, 0.0730, 0.0758, 0.0590, 0.0508, 0.0778], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 02:12:12,350 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 02:12:14,023 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5662, 2.4030, 1.8004, 1.6535, 2.2167, 1.4272, 1.4259, 1.9812], + device='cuda:0'), covar=tensor([0.0918, 0.0651, 0.0897, 0.0671, 0.0453, 0.1142, 0.0680, 0.0435], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0309, 0.0329, 0.0252, 0.0241, 0.0328, 0.0297, 0.0269], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:12:15,968 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 02:12:18,349 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8903, 4.3292, 4.5949, 4.5928, 1.6845, 4.3075, 3.7942, 4.2683], + device='cuda:0'), covar=tensor([0.1476, 0.0814, 0.0513, 0.0580, 0.5280, 0.0712, 0.0596, 0.1040], + device='cuda:0'), in_proj_covar=tensor([0.0713, 0.0634, 0.0842, 0.0729, 0.0756, 0.0590, 0.0507, 0.0777], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 02:12:24,250 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.000e+02 5.065e+02 6.349e+02 7.558e+02 1.663e+03, threshold=1.270e+03, percent-clipped=1.0 +2023-04-02 02:12:29,237 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2090, 3.7922, 3.9036, 3.8738, 1.4369, 3.6750, 3.2590, 3.6029], + device='cuda:0'), covar=tensor([0.1612, 0.0768, 0.0642, 0.0693, 0.5423, 0.0837, 0.0655, 0.1204], + device='cuda:0'), in_proj_covar=tensor([0.0713, 0.0633, 0.0842, 0.0729, 0.0756, 0.0590, 0.0507, 0.0778], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 02:12:43,823 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 02:12:48,705 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6183, 1.3868, 1.4742, 2.0485, 1.5943, 1.8752, 2.0397, 1.7362], + device='cuda:0'), covar=tensor([0.0866, 0.1010, 0.1070, 0.0870, 0.0886, 0.0773, 0.0838, 0.0724], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0222, 0.0223, 0.0243, 0.0230, 0.0209, 0.0192, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 02:12:54,521 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90709.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:12:59,211 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90713.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:13:00,230 INFO [train.py:903] (0/4) Epoch 14, batch 1950, loss[loss=0.2289, simple_loss=0.3067, pruned_loss=0.07554, over 19695.00 frames. ], tot_loss[loss=0.2244, simple_loss=0.3006, pruned_loss=0.07413, over 3833466.48 frames. ], batch size: 53, lr: 5.98e-03, grad_scale: 8.0 +2023-04-02 02:13:40,096 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2282, 1.3273, 1.2223, 1.0139, 1.0735, 1.0976, 0.0442, 0.3927], + device='cuda:0'), covar=tensor([0.0541, 0.0510, 0.0324, 0.0424, 0.1036, 0.0472, 0.0958, 0.0873], + device='cuda:0'), in_proj_covar=tensor([0.0339, 0.0336, 0.0331, 0.0359, 0.0434, 0.0358, 0.0315, 0.0324], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 02:14:03,108 INFO [train.py:903] (0/4) Epoch 14, batch 2000, loss[loss=0.2678, simple_loss=0.3334, pruned_loss=0.1011, over 19527.00 frames. ], tot_loss[loss=0.2251, simple_loss=0.3011, pruned_loss=0.07448, over 3843564.75 frames. ], batch size: 56, lr: 5.98e-03, grad_scale: 8.0 +2023-04-02 02:14:22,438 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4311, 1.3093, 1.4728, 1.3926, 2.9869, 0.8947, 2.3099, 3.3464], + device='cuda:0'), covar=tensor([0.0458, 0.2644, 0.2637, 0.1782, 0.0712, 0.2546, 0.1065, 0.0279], + device='cuda:0'), in_proj_covar=tensor([0.0370, 0.0347, 0.0360, 0.0329, 0.0351, 0.0335, 0.0346, 0.0368], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:14:27,795 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.086e+02 5.207e+02 5.999e+02 7.179e+02 1.269e+03, threshold=1.200e+03, percent-clipped=0.0 +2023-04-02 02:14:59,995 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 02:15:05,781 INFO [train.py:903] (0/4) Epoch 14, batch 2050, loss[loss=0.2161, simple_loss=0.2891, pruned_loss=0.07149, over 19470.00 frames. ], tot_loss[loss=0.2239, simple_loss=0.3001, pruned_loss=0.07379, over 3838793.39 frames. ], batch size: 49, lr: 5.98e-03, grad_scale: 8.0 +2023-04-02 02:15:14,225 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90821.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:15:18,967 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 02:15:20,147 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 02:15:23,959 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90828.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:15:42,039 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 02:16:01,413 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=90858.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:16:07,355 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90863.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:16:08,093 INFO [train.py:903] (0/4) Epoch 14, batch 2100, loss[loss=0.2059, simple_loss=0.2855, pruned_loss=0.06317, over 19616.00 frames. ], tot_loss[loss=0.2242, simple_loss=0.3005, pruned_loss=0.07389, over 3836437.58 frames. ], batch size: 50, lr: 5.98e-03, grad_scale: 8.0 +2023-04-02 02:16:12,820 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=90867.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:16:32,691 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.435e+02 5.132e+02 6.276e+02 7.348e+02 1.970e+03, threshold=1.255e+03, percent-clipped=2.0 +2023-04-02 02:16:37,024 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 02:16:38,554 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90888.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:16:46,370 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0071, 1.7843, 1.7160, 2.1725, 1.9183, 1.7678, 1.8505, 2.0152], + device='cuda:0'), covar=tensor([0.0890, 0.1455, 0.1250, 0.0836, 0.1098, 0.0476, 0.1054, 0.0651], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0354, 0.0296, 0.0240, 0.0295, 0.0240, 0.0288, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:16:59,492 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 02:17:09,967 INFO [train.py:903] (0/4) Epoch 14, batch 2150, loss[loss=0.1861, simple_loss=0.2636, pruned_loss=0.05427, over 19402.00 frames. ], tot_loss[loss=0.224, simple_loss=0.3001, pruned_loss=0.07398, over 3830356.73 frames. ], batch size: 48, lr: 5.97e-03, grad_scale: 8.0 +2023-04-02 02:17:38,816 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90936.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:17:53,046 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-04-02 02:18:10,987 INFO [train.py:903] (0/4) Epoch 14, batch 2200, loss[loss=0.1822, simple_loss=0.2554, pruned_loss=0.05445, over 19818.00 frames. ], tot_loss[loss=0.2245, simple_loss=0.3001, pruned_loss=0.07445, over 3832226.16 frames. ], batch size: 48, lr: 5.97e-03, grad_scale: 8.0 +2023-04-02 02:18:12,455 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=90965.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:18:18,949 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3576, 1.4395, 2.0042, 1.7259, 3.2251, 4.8208, 4.7226, 5.1210], + device='cuda:0'), covar=tensor([0.1569, 0.3390, 0.2898, 0.1864, 0.0469, 0.0172, 0.0138, 0.0152], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0302, 0.0332, 0.0254, 0.0225, 0.0166, 0.0206, 0.0218], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 02:18:23,309 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=90973.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:18:35,015 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.72 vs. limit=5.0 +2023-04-02 02:18:35,325 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.217e+02 5.132e+02 5.971e+02 7.684e+02 1.888e+03, threshold=1.194e+03, percent-clipped=2.0 +2023-04-02 02:18:43,488 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=90990.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:19:13,355 INFO [train.py:903] (0/4) Epoch 14, batch 2250, loss[loss=0.2102, simple_loss=0.2796, pruned_loss=0.07043, over 19426.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.2999, pruned_loss=0.07429, over 3820776.29 frames. ], batch size: 48, lr: 5.97e-03, grad_scale: 8.0 +2023-04-02 02:20:14,199 INFO [train.py:903] (0/4) Epoch 14, batch 2300, loss[loss=0.2386, simple_loss=0.3173, pruned_loss=0.07991, over 19674.00 frames. ], tot_loss[loss=0.2255, simple_loss=0.3006, pruned_loss=0.07519, over 3793024.99 frames. ], batch size: 55, lr: 5.97e-03, grad_scale: 8.0 +2023-04-02 02:20:26,833 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 02:20:38,186 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.297e+02 5.417e+02 6.829e+02 8.730e+02 1.535e+03, threshold=1.366e+03, percent-clipped=12.0 +2023-04-02 02:20:38,633 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91084.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:20:47,630 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9101, 1.7886, 1.7519, 2.0441, 1.9398, 1.8036, 1.7122, 2.0066], + device='cuda:0'), covar=tensor([0.0964, 0.1468, 0.1288, 0.0945, 0.1123, 0.0477, 0.1154, 0.0605], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0350, 0.0294, 0.0239, 0.0293, 0.0239, 0.0285, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:21:11,049 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91109.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:21:16,451 INFO [train.py:903] (0/4) Epoch 14, batch 2350, loss[loss=0.2181, simple_loss=0.2823, pruned_loss=0.07697, over 19379.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.3003, pruned_loss=0.07462, over 3808145.49 frames. ], batch size: 47, lr: 5.97e-03, grad_scale: 8.0 +2023-04-02 02:21:58,620 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 02:22:13,121 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 02:22:14,983 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.84 vs. limit=5.0 +2023-04-02 02:22:17,784 INFO [train.py:903] (0/4) Epoch 14, batch 2400, loss[loss=0.1968, simple_loss=0.2776, pruned_loss=0.05802, over 19495.00 frames. ], tot_loss[loss=0.2254, simple_loss=0.301, pruned_loss=0.07485, over 3811367.40 frames. ], batch size: 49, lr: 5.97e-03, grad_scale: 8.0 +2023-04-02 02:22:42,189 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.550e+02 5.466e+02 6.599e+02 8.174e+02 1.804e+03, threshold=1.320e+03, percent-clipped=5.0 +2023-04-02 02:22:52,555 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9854, 4.3805, 4.6672, 4.6356, 1.6227, 4.3117, 3.7914, 4.3414], + device='cuda:0'), covar=tensor([0.1261, 0.0770, 0.0505, 0.0507, 0.5313, 0.0630, 0.0588, 0.1022], + device='cuda:0'), in_proj_covar=tensor([0.0708, 0.0636, 0.0839, 0.0724, 0.0756, 0.0586, 0.0504, 0.0777], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 02:22:52,716 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91192.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:23:15,504 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=91211.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:23:19,601 INFO [train.py:903] (0/4) Epoch 14, batch 2450, loss[loss=0.2323, simple_loss=0.3126, pruned_loss=0.07599, over 19609.00 frames. ], tot_loss[loss=0.2265, simple_loss=0.3019, pruned_loss=0.07554, over 3808037.54 frames. ], batch size: 57, lr: 5.96e-03, grad_scale: 8.0 +2023-04-02 02:23:23,082 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91217.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:23:37,774 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91229.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:24:08,229 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91254.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:24:19,340 INFO [train.py:903] (0/4) Epoch 14, batch 2500, loss[loss=0.2072, simple_loss=0.2795, pruned_loss=0.06749, over 19380.00 frames. ], tot_loss[loss=0.2262, simple_loss=0.3014, pruned_loss=0.07544, over 3811320.53 frames. ], batch size: 48, lr: 5.96e-03, grad_scale: 8.0 +2023-04-02 02:24:32,865 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2949, 1.2763, 1.2457, 1.3807, 1.0779, 1.3441, 1.3987, 1.3575], + device='cuda:0'), covar=tensor([0.0875, 0.0908, 0.1050, 0.0640, 0.0790, 0.0798, 0.0783, 0.0731], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0223, 0.0224, 0.0241, 0.0228, 0.0209, 0.0191, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 02:24:35,619 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.54 vs. limit=2.0 +2023-04-02 02:24:39,580 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7292, 3.4084, 2.6578, 3.0472, 1.4089, 3.2000, 3.2156, 3.2798], + device='cuda:0'), covar=tensor([0.0980, 0.1247, 0.1852, 0.0961, 0.3252, 0.0976, 0.0940, 0.1252], + device='cuda:0'), in_proj_covar=tensor([0.0452, 0.0380, 0.0454, 0.0330, 0.0392, 0.0386, 0.0378, 0.0413], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:24:42,799 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.074e+02 5.867e+02 6.968e+02 8.618e+02 1.617e+03, threshold=1.394e+03, percent-clipped=2.0 +2023-04-02 02:24:45,406 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9380, 1.2221, 1.6607, 0.8170, 2.3003, 3.0233, 2.7196, 3.2157], + device='cuda:0'), covar=tensor([0.1683, 0.3626, 0.3134, 0.2466, 0.0549, 0.0193, 0.0246, 0.0236], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0301, 0.0330, 0.0253, 0.0223, 0.0166, 0.0206, 0.0217], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 02:24:58,179 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2323, 1.9476, 1.9183, 2.3908, 2.4360, 1.8970, 1.9449, 2.3468], + device='cuda:0'), covar=tensor([0.1112, 0.1934, 0.1787, 0.1115, 0.1373, 0.0899, 0.1599, 0.0859], + device='cuda:0'), in_proj_covar=tensor([0.0257, 0.0352, 0.0294, 0.0240, 0.0294, 0.0240, 0.0288, 0.0241], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:25:06,895 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91303.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:25:19,881 INFO [train.py:903] (0/4) Epoch 14, batch 2550, loss[loss=0.208, simple_loss=0.2908, pruned_loss=0.06256, over 19527.00 frames. ], tot_loss[loss=0.2255, simple_loss=0.3011, pruned_loss=0.07492, over 3805006.78 frames. ], batch size: 54, lr: 5.96e-03, grad_scale: 8.0 +2023-04-02 02:25:33,187 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91326.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:26:12,380 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 02:26:19,133 INFO [train.py:903] (0/4) Epoch 14, batch 2600, loss[loss=0.1831, simple_loss=0.2582, pruned_loss=0.05394, over 18641.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.3003, pruned_loss=0.07454, over 3816631.25 frames. ], batch size: 41, lr: 5.96e-03, grad_scale: 8.0 +2023-04-02 02:26:44,916 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.041e+02 5.557e+02 7.203e+02 9.059e+02 1.995e+03, threshold=1.441e+03, percent-clipped=7.0 +2023-04-02 02:27:21,520 INFO [train.py:903] (0/4) Epoch 14, batch 2650, loss[loss=0.246, simple_loss=0.3147, pruned_loss=0.08863, over 19695.00 frames. ], tot_loss[loss=0.2259, simple_loss=0.3013, pruned_loss=0.07524, over 3819512.51 frames. ], batch size: 58, lr: 5.96e-03, grad_scale: 8.0 +2023-04-02 02:27:41,950 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 02:28:11,767 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5925, 1.2725, 1.4510, 1.3036, 2.2209, 0.9887, 2.0156, 2.4284], + device='cuda:0'), covar=tensor([0.0636, 0.2385, 0.2552, 0.1463, 0.0820, 0.1953, 0.0974, 0.0447], + device='cuda:0'), in_proj_covar=tensor([0.0371, 0.0346, 0.0363, 0.0327, 0.0355, 0.0338, 0.0347, 0.0369], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:28:18,977 INFO [train.py:903] (0/4) Epoch 14, batch 2700, loss[loss=0.2281, simple_loss=0.3123, pruned_loss=0.07193, over 19589.00 frames. ], tot_loss[loss=0.2265, simple_loss=0.3018, pruned_loss=0.07562, over 3831164.68 frames. ], batch size: 61, lr: 5.96e-03, grad_scale: 8.0 +2023-04-02 02:28:43,685 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.192e+02 5.145e+02 6.524e+02 8.606e+02 2.089e+03, threshold=1.305e+03, percent-clipped=4.0 +2023-04-02 02:29:13,059 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.32 vs. limit=5.0 +2023-04-02 02:29:20,022 INFO [train.py:903] (0/4) Epoch 14, batch 2750, loss[loss=0.1859, simple_loss=0.2617, pruned_loss=0.05503, over 19419.00 frames. ], tot_loss[loss=0.2254, simple_loss=0.3006, pruned_loss=0.07507, over 3815159.89 frames. ], batch size: 48, lr: 5.95e-03, grad_scale: 8.0 +2023-04-02 02:29:44,615 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-04-02 02:30:14,297 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91560.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 02:30:18,370 INFO [train.py:903] (0/4) Epoch 14, batch 2800, loss[loss=0.2347, simple_loss=0.311, pruned_loss=0.07917, over 17237.00 frames. ], tot_loss[loss=0.2262, simple_loss=0.3015, pruned_loss=0.07547, over 3804989.50 frames. ], batch size: 101, lr: 5.95e-03, grad_scale: 8.0 +2023-04-02 02:30:41,195 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=91582.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:30:44,142 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.583e+02 4.991e+02 6.176e+02 8.428e+02 2.269e+03, threshold=1.235e+03, percent-clipped=6.0 +2023-04-02 02:31:09,979 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=91607.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:31:19,256 INFO [train.py:903] (0/4) Epoch 14, batch 2850, loss[loss=0.2415, simple_loss=0.3225, pruned_loss=0.08023, over 17544.00 frames. ], tot_loss[loss=0.2256, simple_loss=0.3008, pruned_loss=0.0752, over 3783197.48 frames. ], batch size: 101, lr: 5.95e-03, grad_scale: 8.0 +2023-04-02 02:31:40,067 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.13 vs. limit=5.0 +2023-04-02 02:31:49,776 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-02 02:31:58,526 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=91647.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:32:20,012 INFO [train.py:903] (0/4) Epoch 14, batch 2900, loss[loss=0.2482, simple_loss=0.3289, pruned_loss=0.08381, over 19496.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.3004, pruned_loss=0.07505, over 3790924.00 frames. ], batch size: 64, lr: 5.95e-03, grad_scale: 8.0 +2023-04-02 02:32:20,835 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 02:32:44,081 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.659e+02 5.126e+02 6.310e+02 7.919e+02 2.445e+03, threshold=1.262e+03, percent-clipped=4.0 +2023-04-02 02:33:05,190 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5032, 1.6466, 2.1109, 1.7626, 3.3468, 2.8503, 3.7104, 1.6954], + device='cuda:0'), covar=tensor([0.2228, 0.3669, 0.2376, 0.1676, 0.1237, 0.1685, 0.1263, 0.3405], + device='cuda:0'), in_proj_covar=tensor([0.0500, 0.0587, 0.0633, 0.0446, 0.0597, 0.0500, 0.0644, 0.0504], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 02:33:19,572 INFO [train.py:903] (0/4) Epoch 14, batch 2950, loss[loss=0.2406, simple_loss=0.3171, pruned_loss=0.08204, over 19769.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.3001, pruned_loss=0.07469, over 3797081.32 frames. ], batch size: 56, lr: 5.95e-03, grad_scale: 8.0 +2023-04-02 02:34:06,808 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.30 vs. limit=5.0 +2023-04-02 02:34:17,831 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=91762.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:34:19,623 INFO [train.py:903] (0/4) Epoch 14, batch 3000, loss[loss=0.2181, simple_loss=0.2988, pruned_loss=0.0687, over 19668.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.3006, pruned_loss=0.0749, over 3801705.93 frames. ], batch size: 58, lr: 5.95e-03, grad_scale: 8.0 +2023-04-02 02:34:19,623 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 02:34:36,638 INFO [train.py:937] (0/4) Epoch 14, validation: loss=0.1742, simple_loss=0.2751, pruned_loss=0.03671, over 944034.00 frames. +2023-04-02 02:34:36,639 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 02:34:42,026 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 02:35:02,710 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.309e+02 5.261e+02 6.370e+02 8.625e+02 1.479e+03, threshold=1.274e+03, percent-clipped=4.0 +2023-04-02 02:35:37,796 INFO [train.py:903] (0/4) Epoch 14, batch 3050, loss[loss=0.1869, simple_loss=0.2649, pruned_loss=0.05446, over 19360.00 frames. ], tot_loss[loss=0.2252, simple_loss=0.3007, pruned_loss=0.07487, over 3807684.76 frames. ], batch size: 47, lr: 5.94e-03, grad_scale: 8.0 +2023-04-02 02:36:37,008 INFO [train.py:903] (0/4) Epoch 14, batch 3100, loss[loss=0.2578, simple_loss=0.3252, pruned_loss=0.09519, over 17310.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.3004, pruned_loss=0.07463, over 3819358.64 frames. ], batch size: 101, lr: 5.94e-03, grad_scale: 8.0 +2023-04-02 02:37:02,384 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.205e+02 5.361e+02 6.622e+02 8.860e+02 2.580e+03, threshold=1.324e+03, percent-clipped=11.0 +2023-04-02 02:37:05,897 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2849, 3.8154, 2.2166, 2.4567, 3.4355, 2.0793, 1.6662, 2.2211], + device='cuda:0'), covar=tensor([0.1109, 0.0464, 0.0905, 0.0659, 0.0396, 0.1011, 0.0861, 0.0634], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0306, 0.0325, 0.0249, 0.0238, 0.0325, 0.0293, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:37:24,385 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=91904.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 02:37:31,769 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91909.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:37:37,887 INFO [train.py:903] (0/4) Epoch 14, batch 3150, loss[loss=0.2057, simple_loss=0.2757, pruned_loss=0.06786, over 18261.00 frames. ], tot_loss[loss=0.225, simple_loss=0.3001, pruned_loss=0.07496, over 3819096.43 frames. ], batch size: 40, lr: 5.94e-03, grad_scale: 8.0 +2023-04-02 02:38:04,271 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 02:38:10,518 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3346, 1.3698, 1.7389, 1.5140, 2.5451, 2.2213, 2.7339, 0.9971], + device='cuda:0'), covar=tensor([0.2442, 0.4179, 0.2627, 0.1916, 0.1622, 0.2098, 0.1535, 0.4283], + device='cuda:0'), in_proj_covar=tensor([0.0505, 0.0591, 0.0637, 0.0451, 0.0602, 0.0504, 0.0651, 0.0507], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 02:38:37,363 INFO [train.py:903] (0/4) Epoch 14, batch 3200, loss[loss=0.2423, simple_loss=0.3228, pruned_loss=0.08087, over 19409.00 frames. ], tot_loss[loss=0.2253, simple_loss=0.3007, pruned_loss=0.07493, over 3814853.59 frames. ], batch size: 70, lr: 5.94e-03, grad_scale: 8.0 +2023-04-02 02:38:48,663 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91973.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:39:02,848 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.458e+02 5.150e+02 6.206e+02 7.874e+02 1.849e+03, threshold=1.241e+03, percent-clipped=5.0 +2023-04-02 02:39:20,951 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=91999.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 02:39:21,911 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-92000.pt +2023-04-02 02:39:39,710 INFO [train.py:903] (0/4) Epoch 14, batch 3250, loss[loss=0.2195, simple_loss=0.3049, pruned_loss=0.06707, over 19606.00 frames. ], tot_loss[loss=0.2253, simple_loss=0.3008, pruned_loss=0.07495, over 3812260.34 frames. ], batch size: 57, lr: 5.94e-03, grad_scale: 8.0 +2023-04-02 02:39:44,780 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92018.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:39:45,909 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92019.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 02:40:15,651 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92043.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:40:39,291 INFO [train.py:903] (0/4) Epoch 14, batch 3300, loss[loss=0.2612, simple_loss=0.3344, pruned_loss=0.094, over 19566.00 frames. ], tot_loss[loss=0.2258, simple_loss=0.3009, pruned_loss=0.07535, over 3804948.43 frames. ], batch size: 61, lr: 5.94e-03, grad_scale: 8.0 +2023-04-02 02:40:44,821 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 02:40:52,239 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-02 02:41:05,000 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.031e+02 4.974e+02 6.176e+02 7.406e+02 2.018e+03, threshold=1.235e+03, percent-clipped=5.0 +2023-04-02 02:41:41,642 INFO [train.py:903] (0/4) Epoch 14, batch 3350, loss[loss=0.2371, simple_loss=0.314, pruned_loss=0.08006, over 18647.00 frames. ], tot_loss[loss=0.225, simple_loss=0.3004, pruned_loss=0.07478, over 3819512.20 frames. ], batch size: 74, lr: 5.94e-03, grad_scale: 8.0 +2023-04-02 02:42:40,612 INFO [train.py:903] (0/4) Epoch 14, batch 3400, loss[loss=0.1869, simple_loss=0.2658, pruned_loss=0.05397, over 19780.00 frames. ], tot_loss[loss=0.2251, simple_loss=0.3008, pruned_loss=0.0747, over 3821029.61 frames. ], batch size: 48, lr: 5.93e-03, grad_scale: 8.0 +2023-04-02 02:42:46,824 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0982, 5.1499, 5.9691, 5.9241, 1.9885, 5.5289, 4.7515, 5.5258], + device='cuda:0'), covar=tensor([0.1509, 0.0847, 0.0458, 0.0546, 0.5463, 0.0567, 0.0597, 0.1118], + device='cuda:0'), in_proj_covar=tensor([0.0717, 0.0647, 0.0857, 0.0738, 0.0763, 0.0594, 0.0517, 0.0782], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 02:43:05,849 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.972e+02 4.934e+02 6.017e+02 7.496e+02 1.650e+03, threshold=1.203e+03, percent-clipped=3.0 +2023-04-02 02:43:42,226 INFO [train.py:903] (0/4) Epoch 14, batch 3450, loss[loss=0.2364, simple_loss=0.3136, pruned_loss=0.0796, over 19610.00 frames. ], tot_loss[loss=0.225, simple_loss=0.301, pruned_loss=0.07452, over 3822893.12 frames. ], batch size: 57, lr: 5.93e-03, grad_scale: 8.0 +2023-04-02 02:43:44,674 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 02:44:14,161 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-02 02:44:28,719 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=92253.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:44:40,673 INFO [train.py:903] (0/4) Epoch 14, batch 3500, loss[loss=0.1941, simple_loss=0.2876, pruned_loss=0.05028, over 19610.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.3005, pruned_loss=0.07445, over 3823831.01 frames. ], batch size: 57, lr: 5.93e-03, grad_scale: 8.0 +2023-04-02 02:44:42,973 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4011, 2.1490, 1.5861, 1.4183, 2.0160, 1.1941, 1.2226, 1.8512], + device='cuda:0'), covar=tensor([0.0977, 0.0709, 0.0947, 0.0779, 0.0483, 0.1164, 0.0758, 0.0444], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0311, 0.0331, 0.0254, 0.0241, 0.0331, 0.0299, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:44:43,936 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8617, 4.5024, 3.2320, 3.9641, 2.0432, 4.1977, 4.2813, 4.3396], + device='cuda:0'), covar=tensor([0.0553, 0.0976, 0.1728, 0.0754, 0.2813, 0.0676, 0.0785, 0.1045], + device='cuda:0'), in_proj_covar=tensor([0.0449, 0.0377, 0.0451, 0.0327, 0.0390, 0.0389, 0.0376, 0.0413], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:44:45,988 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1695, 1.8488, 1.9404, 2.6897, 2.0700, 2.5187, 2.6748, 2.3536], + device='cuda:0'), covar=tensor([0.0686, 0.0832, 0.0891, 0.0819, 0.0817, 0.0622, 0.0747, 0.0544], + device='cuda:0'), in_proj_covar=tensor([0.0207, 0.0221, 0.0221, 0.0240, 0.0225, 0.0206, 0.0189, 0.0199], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 02:44:54,880 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92275.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 02:45:05,729 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.029e+02 5.239e+02 6.377e+02 7.871e+02 1.606e+03, threshold=1.275e+03, percent-clipped=5.0 +2023-04-02 02:45:24,223 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92300.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 02:45:34,190 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3993, 2.1945, 1.6625, 1.5792, 2.0018, 1.3010, 1.3011, 1.8546], + device='cuda:0'), covar=tensor([0.1034, 0.0690, 0.1000, 0.0736, 0.0529, 0.1168, 0.0677, 0.0439], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0312, 0.0332, 0.0254, 0.0242, 0.0332, 0.0300, 0.0271], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:45:41,102 INFO [train.py:903] (0/4) Epoch 14, batch 3550, loss[loss=0.2021, simple_loss=0.2849, pruned_loss=0.05968, over 19465.00 frames. ], tot_loss[loss=0.2258, simple_loss=0.3012, pruned_loss=0.07515, over 3804004.91 frames. ], batch size: 64, lr: 5.93e-03, grad_scale: 8.0 +2023-04-02 02:45:44,515 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=92317.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:46:15,218 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=92343.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 02:46:39,901 INFO [train.py:903] (0/4) Epoch 14, batch 3600, loss[loss=0.2053, simple_loss=0.2797, pruned_loss=0.06546, over 19625.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.2998, pruned_loss=0.07438, over 3814391.06 frames. ], batch size: 50, lr: 5.93e-03, grad_scale: 8.0 +2023-04-02 02:46:44,983 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92368.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:46:55,930 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9479, 3.5806, 2.2212, 3.2747, 0.8638, 3.4175, 3.3607, 3.5078], + device='cuda:0'), covar=tensor([0.0839, 0.1167, 0.2298, 0.0812, 0.3984, 0.0854, 0.0834, 0.1072], + device='cuda:0'), in_proj_covar=tensor([0.0449, 0.0378, 0.0453, 0.0329, 0.0392, 0.0391, 0.0378, 0.0413], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:47:04,638 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.193e+02 5.209e+02 6.321e+02 7.842e+02 1.520e+03, threshold=1.264e+03, percent-clipped=2.0 +2023-04-02 02:47:40,755 INFO [train.py:903] (0/4) Epoch 14, batch 3650, loss[loss=0.204, simple_loss=0.2899, pruned_loss=0.05908, over 19788.00 frames. ], tot_loss[loss=0.2244, simple_loss=0.3, pruned_loss=0.07436, over 3817444.75 frames. ], batch size: 56, lr: 5.93e-03, grad_scale: 8.0 +2023-04-02 02:48:02,178 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92432.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:48:34,072 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=92458.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 02:48:40,400 INFO [train.py:903] (0/4) Epoch 14, batch 3700, loss[loss=0.2048, simple_loss=0.2914, pruned_loss=0.05912, over 19648.00 frames. ], tot_loss[loss=0.2239, simple_loss=0.3, pruned_loss=0.0739, over 3822252.85 frames. ], batch size: 58, lr: 5.92e-03, grad_scale: 8.0 +2023-04-02 02:48:51,192 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-02 02:49:05,824 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.433e+02 4.888e+02 6.023e+02 8.004e+02 1.682e+03, threshold=1.205e+03, percent-clipped=3.0 +2023-04-02 02:49:10,927 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1281, 1.9793, 2.0678, 2.4095, 2.0555, 2.2603, 2.2067, 2.1861], + device='cuda:0'), covar=tensor([0.0615, 0.0685, 0.0724, 0.0732, 0.0756, 0.0580, 0.0842, 0.0537], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0222, 0.0222, 0.0242, 0.0226, 0.0207, 0.0191, 0.0200], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 02:49:11,217 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.10 vs. limit=2.0 +2023-04-02 02:49:13,132 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9183, 1.6471, 1.8730, 1.8826, 4.4259, 1.0523, 2.4862, 4.8821], + device='cuda:0'), covar=tensor([0.0360, 0.2576, 0.2585, 0.1692, 0.0732, 0.2666, 0.1370, 0.0149], + device='cuda:0'), in_proj_covar=tensor([0.0370, 0.0345, 0.0362, 0.0326, 0.0356, 0.0336, 0.0345, 0.0369], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:49:41,647 INFO [train.py:903] (0/4) Epoch 14, batch 3750, loss[loss=0.2154, simple_loss=0.294, pruned_loss=0.06837, over 19677.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2989, pruned_loss=0.07358, over 3833791.51 frames. ], batch size: 58, lr: 5.92e-03, grad_scale: 8.0 +2023-04-02 02:50:42,055 INFO [train.py:903] (0/4) Epoch 14, batch 3800, loss[loss=0.2336, simple_loss=0.3088, pruned_loss=0.0792, over 19730.00 frames. ], tot_loss[loss=0.2237, simple_loss=0.2997, pruned_loss=0.07384, over 3826590.63 frames. ], batch size: 63, lr: 5.92e-03, grad_scale: 8.0 +2023-04-02 02:51:06,368 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.816e+02 4.992e+02 6.384e+02 8.353e+02 1.667e+03, threshold=1.277e+03, percent-clipped=5.0 +2023-04-02 02:51:10,997 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 02:51:40,635 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-04-02 02:51:42,072 INFO [train.py:903] (0/4) Epoch 14, batch 3850, loss[loss=0.2367, simple_loss=0.3138, pruned_loss=0.0798, over 19663.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2991, pruned_loss=0.07339, over 3823295.85 frames. ], batch size: 58, lr: 5.92e-03, grad_scale: 8.0 +2023-04-02 02:51:54,735 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92624.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:52:19,012 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92643.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:52:26,123 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92649.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:52:43,617 INFO [train.py:903] (0/4) Epoch 14, batch 3900, loss[loss=0.2345, simple_loss=0.3145, pruned_loss=0.07727, over 19649.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2992, pruned_loss=0.0731, over 3829479.79 frames. ], batch size: 58, lr: 5.92e-03, grad_scale: 4.0 +2023-04-02 02:53:10,396 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.910e+02 5.705e+02 7.277e+02 9.203e+02 2.913e+03, threshold=1.455e+03, percent-clipped=9.0 +2023-04-02 02:53:13,153 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92688.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:53:43,639 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92713.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:53:44,354 INFO [train.py:903] (0/4) Epoch 14, batch 3950, loss[loss=0.2466, simple_loss=0.3147, pruned_loss=0.08928, over 19790.00 frames. ], tot_loss[loss=0.2228, simple_loss=0.299, pruned_loss=0.07327, over 3829514.84 frames. ], batch size: 56, lr: 5.92e-03, grad_scale: 4.0 +2023-04-02 02:53:44,792 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=92714.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 02:53:48,548 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 02:54:12,965 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92738.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:54:14,267 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=92739.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 02:54:45,458 INFO [train.py:903] (0/4) Epoch 14, batch 4000, loss[loss=0.2387, simple_loss=0.3188, pruned_loss=0.07934, over 18088.00 frames. ], tot_loss[loss=0.2229, simple_loss=0.2994, pruned_loss=0.07324, over 3827917.33 frames. ], batch size: 83, lr: 5.91e-03, grad_scale: 8.0 +2023-04-02 02:55:11,213 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.263e+02 4.983e+02 6.436e+02 8.255e+02 1.908e+03, threshold=1.287e+03, percent-clipped=2.0 +2023-04-02 02:55:33,053 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 02:55:45,554 INFO [train.py:903] (0/4) Epoch 14, batch 4050, loss[loss=0.1939, simple_loss=0.2895, pruned_loss=0.04915, over 19660.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2993, pruned_loss=0.07335, over 3837556.21 frames. ], batch size: 58, lr: 5.91e-03, grad_scale: 4.0 +2023-04-02 02:55:52,824 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7897, 1.5352, 1.5525, 1.7725, 1.6082, 1.5542, 1.4798, 1.7514], + device='cuda:0'), covar=tensor([0.0780, 0.1226, 0.1069, 0.0846, 0.1057, 0.0465, 0.1102, 0.0570], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0349, 0.0294, 0.0241, 0.0297, 0.0243, 0.0287, 0.0241], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 02:56:28,992 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92849.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:56:45,589 INFO [train.py:903] (0/4) Epoch 14, batch 4100, loss[loss=0.2168, simple_loss=0.2985, pruned_loss=0.06759, over 19443.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2994, pruned_loss=0.07347, over 3833888.44 frames. ], batch size: 64, lr: 5.91e-03, grad_scale: 4.0 +2023-04-02 02:57:12,584 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-04-02 02:57:13,887 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.303e+02 5.414e+02 6.604e+02 8.302e+02 1.654e+03, threshold=1.321e+03, percent-clipped=7.0 +2023-04-02 02:57:21,830 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 02:57:45,523 INFO [train.py:903] (0/4) Epoch 14, batch 4150, loss[loss=0.2098, simple_loss=0.2787, pruned_loss=0.07043, over 19618.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2991, pruned_loss=0.0735, over 3815125.98 frames. ], batch size: 50, lr: 5.91e-03, grad_scale: 4.0 +2023-04-02 02:57:50,222 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92917.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:58:20,028 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3002, 1.3262, 1.5630, 1.5025, 2.2493, 2.0812, 2.2907, 0.7689], + device='cuda:0'), covar=tensor([0.2259, 0.4011, 0.2407, 0.1803, 0.1432, 0.1999, 0.1328, 0.4128], + device='cuda:0'), in_proj_covar=tensor([0.0499, 0.0588, 0.0633, 0.0446, 0.0599, 0.0500, 0.0645, 0.0503], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 02:58:28,832 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92949.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:58:48,455 INFO [train.py:903] (0/4) Epoch 14, batch 4200, loss[loss=0.2021, simple_loss=0.2818, pruned_loss=0.06121, over 19782.00 frames. ], tot_loss[loss=0.2225, simple_loss=0.2987, pruned_loss=0.07313, over 3820913.07 frames. ], batch size: 49, lr: 5.91e-03, grad_scale: 4.0 +2023-04-02 02:58:51,720 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 02:59:15,443 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.424e+02 5.036e+02 6.286e+02 7.807e+02 1.684e+03, threshold=1.257e+03, percent-clipped=4.0 +2023-04-02 02:59:15,610 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=92987.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:59:22,676 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=92993.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 02:59:48,128 INFO [train.py:903] (0/4) Epoch 14, batch 4250, loss[loss=0.2149, simple_loss=0.2941, pruned_loss=0.06786, over 19083.00 frames. ], tot_loss[loss=0.2242, simple_loss=0.3007, pruned_loss=0.07388, over 3832774.29 frames. ], batch size: 69, lr: 5.91e-03, grad_scale: 4.0 +2023-04-02 03:00:03,996 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 03:00:05,139 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93028.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:00:15,041 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 03:00:37,776 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93054.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:00:47,386 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.66 vs. limit=5.0 +2023-04-02 03:00:48,957 INFO [train.py:903] (0/4) Epoch 14, batch 4300, loss[loss=0.2402, simple_loss=0.3186, pruned_loss=0.08091, over 19554.00 frames. ], tot_loss[loss=0.2246, simple_loss=0.3008, pruned_loss=0.07419, over 3827365.77 frames. ], batch size: 56, lr: 5.90e-03, grad_scale: 4.0 +2023-04-02 03:01:12,678 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93082.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:01:18,339 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.279e+02 5.152e+02 6.308e+02 8.497e+02 2.668e+03, threshold=1.262e+03, percent-clipped=7.0 +2023-04-02 03:01:18,720 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7758, 1.4482, 1.3429, 1.6664, 1.5121, 1.3771, 1.2256, 1.6203], + device='cuda:0'), covar=tensor([0.1021, 0.1331, 0.1559, 0.1043, 0.1199, 0.0740, 0.1541, 0.0809], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0349, 0.0294, 0.0241, 0.0295, 0.0244, 0.0287, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:01:36,136 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93102.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:01:42,511 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 03:01:50,403 INFO [train.py:903] (0/4) Epoch 14, batch 4350, loss[loss=0.1832, simple_loss=0.2669, pruned_loss=0.04974, over 19589.00 frames. ], tot_loss[loss=0.2233, simple_loss=0.2996, pruned_loss=0.07347, over 3828717.84 frames. ], batch size: 52, lr: 5.90e-03, grad_scale: 4.0 +2023-04-02 03:01:53,885 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4784, 2.1786, 1.5953, 1.5571, 2.0570, 1.3256, 1.2588, 1.8057], + device='cuda:0'), covar=tensor([0.0966, 0.0705, 0.1005, 0.0693, 0.0510, 0.1145, 0.0728, 0.0486], + device='cuda:0'), in_proj_covar=tensor([0.0296, 0.0306, 0.0326, 0.0249, 0.0239, 0.0327, 0.0292, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:02:14,533 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93133.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:02:52,476 INFO [train.py:903] (0/4) Epoch 14, batch 4400, loss[loss=0.2205, simple_loss=0.2794, pruned_loss=0.0808, over 19763.00 frames. ], tot_loss[loss=0.2224, simple_loss=0.2987, pruned_loss=0.07308, over 3833600.68 frames. ], batch size: 47, lr: 5.90e-03, grad_scale: 8.0 +2023-04-02 03:03:15,285 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 03:03:18,717 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.086e+02 5.134e+02 6.154e+02 7.916e+02 2.681e+03, threshold=1.231e+03, percent-clipped=4.0 +2023-04-02 03:03:25,257 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 03:03:26,649 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93193.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:03:31,844 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93197.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:03:52,930 INFO [train.py:903] (0/4) Epoch 14, batch 4450, loss[loss=0.2035, simple_loss=0.2772, pruned_loss=0.06485, over 19748.00 frames. ], tot_loss[loss=0.2218, simple_loss=0.2981, pruned_loss=0.07274, over 3841019.88 frames. ], batch size: 51, lr: 5.90e-03, grad_scale: 8.0 +2023-04-02 03:03:54,412 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0121, 1.6759, 1.5882, 2.0745, 1.7623, 1.6788, 1.5935, 1.8666], + device='cuda:0'), covar=tensor([0.0950, 0.1496, 0.1435, 0.1039, 0.1324, 0.0535, 0.1280, 0.0739], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0349, 0.0296, 0.0242, 0.0297, 0.0244, 0.0288, 0.0243], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:04:49,431 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93261.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:04:52,743 INFO [train.py:903] (0/4) Epoch 14, batch 4500, loss[loss=0.2444, simple_loss=0.319, pruned_loss=0.08494, over 19278.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2988, pruned_loss=0.07335, over 3827187.48 frames. ], batch size: 66, lr: 5.90e-03, grad_scale: 8.0 +2023-04-02 03:05:21,647 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.340e+02 5.314e+02 6.344e+02 7.896e+02 1.749e+03, threshold=1.269e+03, percent-clipped=5.0 +2023-04-02 03:05:28,481 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93293.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:05:45,652 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93308.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:05:52,290 INFO [train.py:903] (0/4) Epoch 14, batch 4550, loss[loss=0.2023, simple_loss=0.2725, pruned_loss=0.06605, over 18643.00 frames. ], tot_loss[loss=0.2221, simple_loss=0.2985, pruned_loss=0.07289, over 3838576.08 frames. ], batch size: 41, lr: 5.90e-03, grad_scale: 8.0 +2023-04-02 03:06:06,152 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 03:06:21,893 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93337.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:06:27,652 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 03:06:46,899 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93358.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:06:55,968 INFO [train.py:903] (0/4) Epoch 14, batch 4600, loss[loss=0.2057, simple_loss=0.2788, pruned_loss=0.06632, over 19386.00 frames. ], tot_loss[loss=0.2217, simple_loss=0.2977, pruned_loss=0.07288, over 3837436.37 frames. ], batch size: 48, lr: 5.90e-03, grad_scale: 8.0 +2023-04-02 03:07:02,072 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9713, 1.8827, 1.6025, 1.5117, 1.3413, 1.4936, 0.3374, 0.8891], + device='cuda:0'), covar=tensor([0.0493, 0.0542, 0.0416, 0.0606, 0.1130, 0.0764, 0.1029, 0.0870], + device='cuda:0'), in_proj_covar=tensor([0.0342, 0.0339, 0.0333, 0.0363, 0.0436, 0.0363, 0.0316, 0.0327], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 03:07:05,014 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93372.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:07:09,509 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93376.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:07:13,249 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3396, 1.3276, 1.5445, 1.5384, 2.2406, 2.0740, 2.3000, 0.7916], + device='cuda:0'), covar=tensor([0.2208, 0.3894, 0.2422, 0.1749, 0.1408, 0.1877, 0.1359, 0.3852], + device='cuda:0'), in_proj_covar=tensor([0.0498, 0.0587, 0.0633, 0.0448, 0.0599, 0.0499, 0.0643, 0.0504], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 03:07:17,806 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93383.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:07:22,006 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.600e+02 5.339e+02 6.388e+02 8.227e+02 2.509e+03, threshold=1.278e+03, percent-clipped=2.0 +2023-04-02 03:07:35,566 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93398.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:07:49,408 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93408.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:07:56,015 INFO [train.py:903] (0/4) Epoch 14, batch 4650, loss[loss=0.2142, simple_loss=0.2902, pruned_loss=0.06913, over 19596.00 frames. ], tot_loss[loss=0.2235, simple_loss=0.2993, pruned_loss=0.07382, over 3823991.53 frames. ], batch size: 52, lr: 5.89e-03, grad_scale: 8.0 +2023-04-02 03:07:56,405 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9106, 1.8250, 1.6756, 2.1407, 1.9563, 1.7212, 1.7380, 1.8783], + device='cuda:0'), covar=tensor([0.1074, 0.1594, 0.1390, 0.0996, 0.1255, 0.0561, 0.1235, 0.0746], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0350, 0.0297, 0.0245, 0.0298, 0.0245, 0.0289, 0.0244], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:08:12,040 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 03:08:23,193 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 03:08:43,061 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93452.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:08:44,250 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93453.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:08:45,374 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9791, 1.9690, 1.8499, 1.7648, 1.6373, 1.8483, 0.9680, 1.3957], + device='cuda:0'), covar=tensor([0.0369, 0.0471, 0.0286, 0.0435, 0.0663, 0.0584, 0.0746, 0.0653], + device='cuda:0'), in_proj_covar=tensor([0.0341, 0.0337, 0.0332, 0.0362, 0.0435, 0.0362, 0.0314, 0.0326], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 03:08:56,301 INFO [train.py:903] (0/4) Epoch 14, batch 4700, loss[loss=0.2336, simple_loss=0.3104, pruned_loss=0.07842, over 19603.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2989, pruned_loss=0.07358, over 3827753.27 frames. ], batch size: 57, lr: 5.89e-03, grad_scale: 8.0 +2023-04-02 03:09:11,849 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=93477.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:09:13,361 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93478.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:09:20,329 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 03:09:25,717 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.143e+02 5.199e+02 6.337e+02 7.857e+02 1.524e+03, threshold=1.267e+03, percent-clipped=2.0 +2023-04-02 03:09:26,087 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93487.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:09:55,133 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93513.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:09:55,932 INFO [train.py:903] (0/4) Epoch 14, batch 4750, loss[loss=0.2191, simple_loss=0.2985, pruned_loss=0.0699, over 19609.00 frames. ], tot_loss[loss=0.2225, simple_loss=0.2984, pruned_loss=0.07333, over 3822216.75 frames. ], batch size: 61, lr: 5.89e-03, grad_scale: 8.0 +2023-04-02 03:10:55,740 INFO [train.py:903] (0/4) Epoch 14, batch 4800, loss[loss=0.2647, simple_loss=0.338, pruned_loss=0.09567, over 19688.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.3006, pruned_loss=0.07463, over 3827898.00 frames. ], batch size: 59, lr: 5.89e-03, grad_scale: 8.0 +2023-04-02 03:10:56,108 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93564.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:11:22,943 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.531e+02 5.541e+02 6.642e+02 8.296e+02 2.320e+03, threshold=1.328e+03, percent-clipped=4.0 +2023-04-02 03:11:23,356 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7076, 1.7244, 1.5489, 1.3862, 1.3480, 1.4525, 0.2050, 0.6565], + device='cuda:0'), covar=tensor([0.0483, 0.0491, 0.0316, 0.0515, 0.1027, 0.0563, 0.0955, 0.0875], + device='cuda:0'), in_proj_covar=tensor([0.0343, 0.0338, 0.0334, 0.0363, 0.0438, 0.0363, 0.0314, 0.0327], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 03:11:25,690 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93589.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:11:29,129 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=93592.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:11:57,049 INFO [train.py:903] (0/4) Epoch 14, batch 4850, loss[loss=0.2236, simple_loss=0.2984, pruned_loss=0.0744, over 19567.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.3, pruned_loss=0.07424, over 3828397.20 frames. ], batch size: 52, lr: 5.89e-03, grad_scale: 8.0 +2023-04-02 03:12:17,832 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93632.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:12:23,247 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 03:12:42,785 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 03:12:47,279 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 03:12:48,586 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 03:12:48,922 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93657.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:12:56,532 INFO [train.py:903] (0/4) Epoch 14, batch 4900, loss[loss=0.2312, simple_loss=0.3093, pruned_loss=0.07652, over 19541.00 frames. ], tot_loss[loss=0.225, simple_loss=0.3008, pruned_loss=0.07454, over 3813627.49 frames. ], batch size: 56, lr: 5.89e-03, grad_scale: 8.0 +2023-04-02 03:12:56,971 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93664.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:12:57,699 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 03:13:18,094 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 03:13:22,700 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.1991, 3.8707, 2.8734, 3.4428, 1.0734, 3.6443, 3.5788, 3.7558], + device='cuda:0'), covar=tensor([0.0811, 0.0993, 0.1638, 0.0806, 0.3791, 0.0869, 0.0924, 0.1244], + device='cuda:0'), in_proj_covar=tensor([0.0449, 0.0377, 0.0451, 0.0326, 0.0389, 0.0387, 0.0377, 0.0409], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:13:25,555 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.007e+02 4.842e+02 5.818e+02 7.123e+02 1.786e+03, threshold=1.164e+03, percent-clipped=2.0 +2023-04-02 03:13:28,157 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93689.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:13:49,653 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93708.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:13:55,992 INFO [train.py:903] (0/4) Epoch 14, batch 4950, loss[loss=0.2253, simple_loss=0.3019, pruned_loss=0.07435, over 19618.00 frames. ], tot_loss[loss=0.2238, simple_loss=0.2999, pruned_loss=0.07388, over 3813452.00 frames. ], batch size: 50, lr: 5.88e-03, grad_scale: 8.0 +2023-04-02 03:14:16,294 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 03:14:21,093 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93733.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:14:32,430 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93743.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:14:36,699 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 03:14:58,214 INFO [train.py:903] (0/4) Epoch 14, batch 5000, loss[loss=0.2595, simple_loss=0.3223, pruned_loss=0.09834, over 19605.00 frames. ], tot_loss[loss=0.2242, simple_loss=0.3, pruned_loss=0.07422, over 3805651.82 frames. ], batch size: 50, lr: 5.88e-03, grad_scale: 8.0 +2023-04-02 03:15:04,133 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93768.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:15:05,345 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93769.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:15:07,263 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 03:15:14,498 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1260, 1.8779, 1.6991, 2.1658, 2.0872, 1.7859, 1.7032, 2.0385], + device='cuda:0'), covar=tensor([0.0913, 0.1451, 0.1359, 0.0899, 0.1187, 0.0518, 0.1249, 0.0617], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0350, 0.0296, 0.0243, 0.0298, 0.0245, 0.0290, 0.0243], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:15:17,362 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 03:15:25,226 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 4.057e+02 5.347e+02 6.962e+02 9.103e+02 2.417e+03, threshold=1.392e+03, percent-clipped=9.0 +2023-04-02 03:15:26,625 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93788.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:15:33,765 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93794.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:15:59,315 INFO [train.py:903] (0/4) Epoch 14, batch 5050, loss[loss=0.1929, simple_loss=0.2706, pruned_loss=0.05763, over 19748.00 frames. ], tot_loss[loss=0.2246, simple_loss=0.3002, pruned_loss=0.07456, over 3810399.32 frames. ], batch size: 47, lr: 5.88e-03, grad_scale: 8.0 +2023-04-02 03:16:35,021 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 03:16:40,853 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=93848.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:16:59,044 INFO [train.py:903] (0/4) Epoch 14, batch 5100, loss[loss=0.2125, simple_loss=0.2873, pruned_loss=0.06888, over 19453.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.2998, pruned_loss=0.07425, over 3814076.62 frames. ], batch size: 49, lr: 5.88e-03, grad_scale: 8.0 +2023-04-02 03:17:09,068 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=93873.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:17:09,890 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 03:17:13,173 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 03:17:16,721 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 03:17:26,258 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.008e+02 4.904e+02 5.934e+02 7.645e+02 1.361e+03, threshold=1.187e+03, percent-clipped=0.0 +2023-04-02 03:17:47,167 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0160, 5.4465, 2.8160, 4.7458, 1.2535, 5.5563, 5.4145, 5.5702], + device='cuda:0'), covar=tensor([0.0411, 0.0768, 0.1989, 0.0618, 0.3598, 0.0500, 0.0713, 0.0898], + device='cuda:0'), in_proj_covar=tensor([0.0452, 0.0380, 0.0454, 0.0329, 0.0392, 0.0389, 0.0378, 0.0411], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:17:56,876 INFO [train.py:903] (0/4) Epoch 14, batch 5150, loss[loss=0.2057, simple_loss=0.2799, pruned_loss=0.06576, over 19733.00 frames. ], tot_loss[loss=0.2253, simple_loss=0.3009, pruned_loss=0.07487, over 3808979.78 frames. ], batch size: 51, lr: 5.88e-03, grad_scale: 8.0 +2023-04-02 03:18:09,225 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 03:18:43,202 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 03:18:58,008 INFO [train.py:903] (0/4) Epoch 14, batch 5200, loss[loss=0.258, simple_loss=0.3112, pruned_loss=0.1024, over 19759.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.3, pruned_loss=0.07405, over 3816453.41 frames. ], batch size: 45, lr: 5.88e-03, grad_scale: 8.0 +2023-04-02 03:19:13,837 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 03:19:16,649 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2118, 1.2167, 1.3989, 1.3786, 1.6560, 1.7257, 1.7043, 0.4462], + device='cuda:0'), covar=tensor([0.2477, 0.4149, 0.2536, 0.1992, 0.1666, 0.2254, 0.1399, 0.4450], + device='cuda:0'), in_proj_covar=tensor([0.0501, 0.0587, 0.0634, 0.0448, 0.0600, 0.0500, 0.0645, 0.0505], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 03:19:25,471 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.052e+02 5.268e+02 6.485e+02 8.631e+02 2.638e+03, threshold=1.297e+03, percent-clipped=7.0 +2023-04-02 03:19:39,250 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=93999.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:19:40,158 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-94000.pt +2023-04-02 03:19:57,306 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 03:19:59,391 INFO [train.py:903] (0/4) Epoch 14, batch 5250, loss[loss=0.2703, simple_loss=0.3497, pruned_loss=0.09539, over 19678.00 frames. ], tot_loss[loss=0.2237, simple_loss=0.2995, pruned_loss=0.07395, over 3824600.80 frames. ], batch size: 60, lr: 5.88e-03, grad_scale: 8.0 +2023-04-02 03:20:05,577 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6143, 1.6253, 1.7791, 1.8275, 4.1444, 1.1960, 2.4028, 4.4560], + device='cuda:0'), covar=tensor([0.0382, 0.2613, 0.2695, 0.1736, 0.0721, 0.2611, 0.1522, 0.0190], + device='cuda:0'), in_proj_covar=tensor([0.0372, 0.0345, 0.0364, 0.0324, 0.0351, 0.0334, 0.0343, 0.0369], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:20:43,656 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9043, 1.9915, 2.2342, 2.7117, 1.9654, 2.6185, 2.4057, 2.0139], + device='cuda:0'), covar=tensor([0.3784, 0.3300, 0.1570, 0.1979, 0.3682, 0.1653, 0.3845, 0.2833], + device='cuda:0'), in_proj_covar=tensor([0.0821, 0.0858, 0.0666, 0.0895, 0.0808, 0.0742, 0.0803, 0.0730], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 03:20:47,424 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.88 vs. limit=2.0 +2023-04-02 03:20:59,247 INFO [train.py:903] (0/4) Epoch 14, batch 5300, loss[loss=0.1938, simple_loss=0.2727, pruned_loss=0.05748, over 19787.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.3001, pruned_loss=0.07471, over 3822247.05 frames. ], batch size: 48, lr: 5.87e-03, grad_scale: 8.0 +2023-04-02 03:21:16,464 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 03:21:27,961 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.389e+02 5.368e+02 7.020e+02 9.283e+02 2.840e+03, threshold=1.404e+03, percent-clipped=4.0 +2023-04-02 03:21:45,478 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-02 03:21:59,007 INFO [train.py:903] (0/4) Epoch 14, batch 5350, loss[loss=0.2276, simple_loss=0.3069, pruned_loss=0.07419, over 19670.00 frames. ], tot_loss[loss=0.2251, simple_loss=0.3005, pruned_loss=0.0748, over 3826860.45 frames. ], batch size: 58, lr: 5.87e-03, grad_scale: 8.0 +2023-04-02 03:22:23,137 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=94132.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:22:34,755 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 03:23:01,402 INFO [train.py:903] (0/4) Epoch 14, batch 5400, loss[loss=0.2224, simple_loss=0.3058, pruned_loss=0.06954, over 19526.00 frames. ], tot_loss[loss=0.2248, simple_loss=0.3001, pruned_loss=0.07476, over 3810654.07 frames. ], batch size: 56, lr: 5.87e-03, grad_scale: 8.0 +2023-04-02 03:23:29,188 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.364e+02 5.537e+02 7.248e+02 8.700e+02 2.021e+03, threshold=1.450e+03, percent-clipped=3.0 +2023-04-02 03:24:03,211 INFO [train.py:903] (0/4) Epoch 14, batch 5450, loss[loss=0.1897, simple_loss=0.2596, pruned_loss=0.05991, over 19721.00 frames. ], tot_loss[loss=0.2251, simple_loss=0.3005, pruned_loss=0.0748, over 3818189.89 frames. ], batch size: 45, lr: 5.87e-03, grad_scale: 8.0 +2023-04-02 03:24:33,951 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94241.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 03:24:43,529 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94247.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:25:02,862 INFO [train.py:903] (0/4) Epoch 14, batch 5500, loss[loss=0.2961, simple_loss=0.352, pruned_loss=0.1201, over 13533.00 frames. ], tot_loss[loss=0.2245, simple_loss=0.3003, pruned_loss=0.07436, over 3812937.77 frames. ], batch size: 136, lr: 5.87e-03, grad_scale: 8.0 +2023-04-02 03:25:24,845 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 03:25:30,870 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.158e+02 4.805e+02 5.794e+02 7.462e+02 1.465e+03, threshold=1.159e+03, percent-clipped=1.0 +2023-04-02 03:25:44,162 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4065, 2.2585, 1.7174, 1.5775, 2.1053, 1.4263, 1.3215, 1.9047], + device='cuda:0'), covar=tensor([0.0897, 0.0685, 0.0962, 0.0731, 0.0447, 0.1048, 0.0650, 0.0389], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0304, 0.0327, 0.0249, 0.0238, 0.0325, 0.0292, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:25:51,157 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.56 vs. limit=2.0 +2023-04-02 03:26:01,459 INFO [train.py:903] (0/4) Epoch 14, batch 5550, loss[loss=0.2114, simple_loss=0.2992, pruned_loss=0.06174, over 19666.00 frames. ], tot_loss[loss=0.2244, simple_loss=0.2998, pruned_loss=0.07448, over 3824280.53 frames. ], batch size: 59, lr: 5.87e-03, grad_scale: 8.0 +2023-04-02 03:26:08,356 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 03:26:37,722 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=94343.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:26:57,886 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 03:27:01,434 INFO [train.py:903] (0/4) Epoch 14, batch 5600, loss[loss=0.1858, simple_loss=0.2678, pruned_loss=0.0519, over 19623.00 frames. ], tot_loss[loss=0.2234, simple_loss=0.299, pruned_loss=0.07387, over 3826696.45 frames. ], batch size: 50, lr: 5.86e-03, grad_scale: 8.0 +2023-04-02 03:27:04,764 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=94366.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:27:30,034 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.083e+02 5.188e+02 6.005e+02 7.911e+02 1.925e+03, threshold=1.201e+03, percent-clipped=8.0 +2023-04-02 03:28:03,376 INFO [train.py:903] (0/4) Epoch 14, batch 5650, loss[loss=0.2549, simple_loss=0.3185, pruned_loss=0.09567, over 13608.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2987, pruned_loss=0.07363, over 3829031.40 frames. ], batch size: 136, lr: 5.86e-03, grad_scale: 8.0 +2023-04-02 03:28:40,691 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.55 vs. limit=2.0 +2023-04-02 03:28:48,908 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6008, 1.5545, 1.5852, 1.9244, 3.1856, 1.3791, 2.2052, 3.6474], + device='cuda:0'), covar=tensor([0.0496, 0.2397, 0.2536, 0.1396, 0.0669, 0.2098, 0.1282, 0.0224], + device='cuda:0'), in_proj_covar=tensor([0.0373, 0.0346, 0.0365, 0.0325, 0.0352, 0.0335, 0.0345, 0.0369], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:28:49,721 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 03:28:55,926 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94458.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:29:02,022 INFO [train.py:903] (0/4) Epoch 14, batch 5700, loss[loss=0.2675, simple_loss=0.3259, pruned_loss=0.1046, over 19606.00 frames. ], tot_loss[loss=0.2245, simple_loss=0.2999, pruned_loss=0.07453, over 3822812.24 frames. ], batch size: 50, lr: 5.86e-03, grad_scale: 8.0 +2023-04-02 03:29:10,686 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4361, 1.4711, 1.7937, 1.6732, 2.7808, 2.4045, 2.8530, 1.2043], + device='cuda:0'), covar=tensor([0.2302, 0.3923, 0.2470, 0.1798, 0.1410, 0.1866, 0.1458, 0.3931], + device='cuda:0'), in_proj_covar=tensor([0.0505, 0.0592, 0.0642, 0.0451, 0.0604, 0.0508, 0.0651, 0.0510], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 03:29:23,216 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-02 03:29:29,829 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.310e+02 4.949e+02 6.008e+02 7.817e+02 2.884e+03, threshold=1.202e+03, percent-clipped=11.0 +2023-04-02 03:29:50,170 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=94503.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:29:58,116 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8556, 1.4321, 1.5525, 1.7080, 3.3930, 1.0389, 2.2583, 3.7103], + device='cuda:0'), covar=tensor([0.0440, 0.2683, 0.2715, 0.1657, 0.0632, 0.2712, 0.1421, 0.0265], + device='cuda:0'), in_proj_covar=tensor([0.0375, 0.0347, 0.0367, 0.0326, 0.0354, 0.0337, 0.0346, 0.0370], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:30:02,384 INFO [train.py:903] (0/4) Epoch 14, batch 5750, loss[loss=0.2786, simple_loss=0.3424, pruned_loss=0.1074, over 19697.00 frames. ], tot_loss[loss=0.2237, simple_loss=0.2994, pruned_loss=0.07405, over 3821417.72 frames. ], batch size: 63, lr: 5.86e-03, grad_scale: 8.0 +2023-04-02 03:30:04,719 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 03:30:11,529 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 03:30:17,749 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 03:30:21,289 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=94528.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:30:32,322 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6850, 1.4820, 1.3899, 2.1028, 1.6476, 2.0096, 2.0090, 1.7486], + device='cuda:0'), covar=tensor([0.0841, 0.0936, 0.1101, 0.0811, 0.0868, 0.0705, 0.0898, 0.0714], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0220, 0.0223, 0.0241, 0.0226, 0.0208, 0.0191, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 03:30:45,585 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.10 vs. limit=2.0 +2023-04-02 03:31:04,686 INFO [train.py:903] (0/4) Epoch 14, batch 5800, loss[loss=0.2222, simple_loss=0.2942, pruned_loss=0.07512, over 19734.00 frames. ], tot_loss[loss=0.2236, simple_loss=0.2994, pruned_loss=0.07389, over 3821075.68 frames. ], batch size: 51, lr: 5.86e-03, grad_scale: 8.0 +2023-04-02 03:31:30,513 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=94585.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 03:31:32,533 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.175e+02 5.430e+02 7.155e+02 9.192e+02 1.752e+03, threshold=1.431e+03, percent-clipped=10.0 +2023-04-02 03:32:06,961 INFO [train.py:903] (0/4) Epoch 14, batch 5850, loss[loss=0.218, simple_loss=0.2996, pruned_loss=0.06822, over 19671.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2985, pruned_loss=0.0734, over 3822767.47 frames. ], batch size: 58, lr: 5.86e-03, grad_scale: 8.0 +2023-04-02 03:33:06,741 INFO [train.py:903] (0/4) Epoch 14, batch 5900, loss[loss=0.2069, simple_loss=0.2762, pruned_loss=0.06875, over 19715.00 frames. ], tot_loss[loss=0.2236, simple_loss=0.2991, pruned_loss=0.07409, over 3823926.44 frames. ], batch size: 45, lr: 5.85e-03, grad_scale: 8.0 +2023-04-02 03:33:07,937 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 03:33:27,827 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 03:33:33,162 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.177e+02 5.122e+02 5.971e+02 8.409e+02 2.018e+03, threshold=1.194e+03, percent-clipped=4.0 +2023-04-02 03:33:50,045 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94700.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 03:34:01,307 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=94710.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:34:05,815 INFO [train.py:903] (0/4) Epoch 14, batch 5950, loss[loss=0.2037, simple_loss=0.2785, pruned_loss=0.06447, over 19064.00 frames. ], tot_loss[loss=0.2239, simple_loss=0.2994, pruned_loss=0.07422, over 3814622.52 frames. ], batch size: 42, lr: 5.85e-03, grad_scale: 8.0 +2023-04-02 03:34:06,212 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=94714.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:34:37,244 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=94739.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:34:38,743 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.33 vs. limit=5.0 +2023-04-02 03:35:04,630 INFO [train.py:903] (0/4) Epoch 14, batch 6000, loss[loss=0.2457, simple_loss=0.3145, pruned_loss=0.08847, over 19539.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.3, pruned_loss=0.07491, over 3817494.12 frames. ], batch size: 56, lr: 5.85e-03, grad_scale: 8.0 +2023-04-02 03:35:04,631 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 03:35:17,169 INFO [train.py:937] (0/4) Epoch 14, validation: loss=0.1744, simple_loss=0.2748, pruned_loss=0.03705, over 944034.00 frames. +2023-04-02 03:35:17,170 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 03:35:25,778 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8670, 1.7419, 1.5050, 1.9485, 1.7277, 1.6438, 1.4508, 1.8426], + device='cuda:0'), covar=tensor([0.0960, 0.1268, 0.1408, 0.0889, 0.1175, 0.0512, 0.1342, 0.0701], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0354, 0.0299, 0.0245, 0.0300, 0.0245, 0.0291, 0.0246], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:35:47,198 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.285e+02 5.018e+02 6.191e+02 7.483e+02 1.325e+03, threshold=1.238e+03, percent-clipped=2.0 +2023-04-02 03:36:17,852 INFO [train.py:903] (0/4) Epoch 14, batch 6050, loss[loss=0.2233, simple_loss=0.2956, pruned_loss=0.07547, over 19844.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.2998, pruned_loss=0.07497, over 3817840.41 frames. ], batch size: 52, lr: 5.85e-03, grad_scale: 8.0 +2023-04-02 03:36:33,144 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=94825.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:36:46,534 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1146, 2.7872, 2.2015, 2.1410, 2.0092, 2.3951, 0.8192, 1.9583], + device='cuda:0'), covar=tensor([0.0563, 0.0513, 0.0591, 0.0921, 0.1000, 0.0950, 0.1187, 0.0989], + device='cuda:0'), in_proj_covar=tensor([0.0345, 0.0340, 0.0335, 0.0366, 0.0436, 0.0364, 0.0316, 0.0326], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 03:37:20,902 INFO [train.py:903] (0/4) Epoch 14, batch 6100, loss[loss=0.2561, simple_loss=0.3309, pruned_loss=0.09063, over 17441.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.2999, pruned_loss=0.07476, over 3820256.63 frames. ], batch size: 101, lr: 5.85e-03, grad_scale: 8.0 +2023-04-02 03:37:48,987 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.227e+02 5.260e+02 6.294e+02 8.137e+02 1.551e+03, threshold=1.259e+03, percent-clipped=3.0 +2023-04-02 03:38:10,927 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5823, 1.4102, 1.4354, 1.8388, 1.4377, 1.9780, 1.9081, 1.6184], + device='cuda:0'), covar=tensor([0.0805, 0.0933, 0.1037, 0.0825, 0.0900, 0.0666, 0.0796, 0.0698], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0220, 0.0223, 0.0243, 0.0227, 0.0210, 0.0193, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 03:38:21,674 INFO [train.py:903] (0/4) Epoch 14, batch 6150, loss[loss=0.2467, simple_loss=0.3153, pruned_loss=0.08904, over 19741.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2986, pruned_loss=0.07379, over 3825766.30 frames. ], batch size: 51, lr: 5.85e-03, grad_scale: 8.0 +2023-04-02 03:38:48,803 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 03:39:13,126 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=94956.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 03:39:21,882 INFO [train.py:903] (0/4) Epoch 14, batch 6200, loss[loss=0.2134, simple_loss=0.2968, pruned_loss=0.06504, over 18215.00 frames. ], tot_loss[loss=0.2223, simple_loss=0.2983, pruned_loss=0.07313, over 3832347.72 frames. ], batch size: 83, lr: 5.85e-03, grad_scale: 8.0 +2023-04-02 03:39:43,629 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=94981.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 03:39:48,242 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 03:39:51,883 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.111e+02 5.470e+02 6.385e+02 8.085e+02 2.296e+03, threshold=1.277e+03, percent-clipped=5.0 +2023-04-02 03:40:22,496 INFO [train.py:903] (0/4) Epoch 14, batch 6250, loss[loss=0.2317, simple_loss=0.3102, pruned_loss=0.07662, over 19622.00 frames. ], tot_loss[loss=0.2218, simple_loss=0.2979, pruned_loss=0.07283, over 3838337.19 frames. ], batch size: 57, lr: 5.84e-03, grad_scale: 8.0 +2023-04-02 03:40:55,024 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 03:41:24,163 INFO [train.py:903] (0/4) Epoch 14, batch 6300, loss[loss=0.2205, simple_loss=0.3061, pruned_loss=0.06746, over 19594.00 frames. ], tot_loss[loss=0.2222, simple_loss=0.2982, pruned_loss=0.07309, over 3840110.04 frames. ], batch size: 61, lr: 5.84e-03, grad_scale: 8.0 +2023-04-02 03:41:44,520 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95081.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:41:51,896 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.537e+02 5.238e+02 6.215e+02 7.195e+02 1.642e+03, threshold=1.243e+03, percent-clipped=4.0 +2023-04-02 03:42:15,045 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95106.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:42:24,132 INFO [train.py:903] (0/4) Epoch 14, batch 6350, loss[loss=0.1859, simple_loss=0.2646, pruned_loss=0.05359, over 19383.00 frames. ], tot_loss[loss=0.2235, simple_loss=0.2993, pruned_loss=0.0738, over 3831124.59 frames. ], batch size: 47, lr: 5.84e-03, grad_scale: 8.0 +2023-04-02 03:42:34,667 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95123.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 03:43:23,517 INFO [train.py:903] (0/4) Epoch 14, batch 6400, loss[loss=0.2501, simple_loss=0.3236, pruned_loss=0.08829, over 19508.00 frames. ], tot_loss[loss=0.2245, simple_loss=0.3002, pruned_loss=0.07438, over 3825714.57 frames. ], batch size: 64, lr: 5.84e-03, grad_scale: 8.0 +2023-04-02 03:43:40,120 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0538, 3.5848, 1.9175, 2.1471, 3.1004, 1.6497, 1.3221, 2.2005], + device='cuda:0'), covar=tensor([0.1296, 0.0472, 0.1041, 0.0774, 0.0480, 0.1168, 0.0964, 0.0684], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0305, 0.0328, 0.0248, 0.0240, 0.0327, 0.0291, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:43:52,823 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.332e+02 5.689e+02 7.116e+02 8.755e+02 2.889e+03, threshold=1.423e+03, percent-clipped=3.0 +2023-04-02 03:44:23,629 INFO [train.py:903] (0/4) Epoch 14, batch 6450, loss[loss=0.2434, simple_loss=0.3174, pruned_loss=0.08465, over 18822.00 frames. ], tot_loss[loss=0.2247, simple_loss=0.3001, pruned_loss=0.07469, over 3821926.55 frames. ], batch size: 74, lr: 5.84e-03, grad_scale: 8.0 +2023-04-02 03:45:09,467 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 03:45:25,845 INFO [train.py:903] (0/4) Epoch 14, batch 6500, loss[loss=0.2331, simple_loss=0.3152, pruned_loss=0.07551, over 19694.00 frames. ], tot_loss[loss=0.2253, simple_loss=0.3008, pruned_loss=0.07487, over 3816825.66 frames. ], batch size: 60, lr: 5.84e-03, grad_scale: 8.0 +2023-04-02 03:45:32,331 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 03:45:32,718 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8325, 1.9081, 2.0878, 2.3588, 1.6499, 2.2665, 2.2544, 1.9986], + device='cuda:0'), covar=tensor([0.3513, 0.3025, 0.1631, 0.1867, 0.3337, 0.1618, 0.3883, 0.2819], + device='cuda:0'), in_proj_covar=tensor([0.0829, 0.0867, 0.0669, 0.0906, 0.0816, 0.0753, 0.0814, 0.0739], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 03:45:43,516 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95278.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:45:52,731 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95286.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:45:54,574 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.481e+02 5.245e+02 6.559e+02 8.783e+02 2.152e+03, threshold=1.312e+03, percent-clipped=6.0 +2023-04-02 03:46:27,867 INFO [train.py:903] (0/4) Epoch 14, batch 6550, loss[loss=0.2059, simple_loss=0.2943, pruned_loss=0.05877, over 19619.00 frames. ], tot_loss[loss=0.2242, simple_loss=0.3001, pruned_loss=0.07411, over 3815258.26 frames. ], batch size: 57, lr: 5.84e-03, grad_scale: 8.0 +2023-04-02 03:47:20,347 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95357.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:47:28,129 INFO [train.py:903] (0/4) Epoch 14, batch 6600, loss[loss=0.2182, simple_loss=0.2954, pruned_loss=0.07044, over 19510.00 frames. ], tot_loss[loss=0.2225, simple_loss=0.2988, pruned_loss=0.07315, over 3829128.23 frames. ], batch size: 54, lr: 5.83e-03, grad_scale: 8.0 +2023-04-02 03:47:57,394 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.401e+02 5.166e+02 6.061e+02 7.266e+02 1.890e+03, threshold=1.212e+03, percent-clipped=2.0 +2023-04-02 03:48:28,428 INFO [train.py:903] (0/4) Epoch 14, batch 6650, loss[loss=0.302, simple_loss=0.3691, pruned_loss=0.1174, over 19303.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.299, pruned_loss=0.07364, over 3819235.33 frames. ], batch size: 66, lr: 5.83e-03, grad_scale: 8.0 +2023-04-02 03:48:35,365 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-02 03:49:29,376 INFO [train.py:903] (0/4) Epoch 14, batch 6700, loss[loss=0.2339, simple_loss=0.3032, pruned_loss=0.08224, over 13459.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.2999, pruned_loss=0.07436, over 3813873.21 frames. ], batch size: 137, lr: 5.83e-03, grad_scale: 8.0 +2023-04-02 03:49:33,795 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=95467.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 03:49:57,461 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.175e+02 5.267e+02 5.901e+02 8.158e+02 1.902e+03, threshold=1.180e+03, percent-clipped=6.0 +2023-04-02 03:50:01,573 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 03:50:25,738 INFO [train.py:903] (0/4) Epoch 14, batch 6750, loss[loss=0.2292, simple_loss=0.3001, pruned_loss=0.07913, over 19662.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.3002, pruned_loss=0.07474, over 3812863.96 frames. ], batch size: 53, lr: 5.83e-03, grad_scale: 8.0 +2023-04-02 03:51:01,630 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2273, 1.3941, 1.8285, 0.8966, 2.3578, 3.0743, 2.7827, 3.2938], + device='cuda:0'), covar=tensor([0.1430, 0.3255, 0.2610, 0.2276, 0.0501, 0.0192, 0.0238, 0.0215], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0303, 0.0329, 0.0253, 0.0223, 0.0167, 0.0207, 0.0219], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 03:51:21,227 INFO [train.py:903] (0/4) Epoch 14, batch 6800, loss[loss=0.2523, simple_loss=0.3331, pruned_loss=0.08573, over 19479.00 frames. ], tot_loss[loss=0.2261, simple_loss=0.3015, pruned_loss=0.07533, over 3813738.92 frames. ], batch size: 64, lr: 5.83e-03, grad_scale: 8.0 +2023-04-02 03:51:41,546 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95582.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 03:51:46,734 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.931e+02 5.198e+02 6.166e+02 8.008e+02 1.508e+03, threshold=1.233e+03, percent-clipped=6.0 +2023-04-02 03:51:51,245 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-14.pt +2023-04-02 03:52:07,410 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 03:52:07,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 03:52:10,209 INFO [train.py:903] (0/4) Epoch 15, batch 0, loss[loss=0.2446, simple_loss=0.3207, pruned_loss=0.08429, over 18759.00 frames. ], tot_loss[loss=0.2446, simple_loss=0.3207, pruned_loss=0.08429, over 18759.00 frames. ], batch size: 74, lr: 5.63e-03, grad_scale: 8.0 +2023-04-02 03:52:10,209 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 03:52:21,738 INFO [train.py:937] (0/4) Epoch 15, validation: loss=0.1744, simple_loss=0.2751, pruned_loss=0.03681, over 944034.00 frames. +2023-04-02 03:52:21,739 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 03:52:26,536 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2992, 1.4728, 1.8186, 1.5999, 2.6766, 2.1905, 2.8090, 1.1449], + device='cuda:0'), covar=tensor([0.2478, 0.4112, 0.2543, 0.1885, 0.1589, 0.2074, 0.1577, 0.4190], + device='cuda:0'), in_proj_covar=tensor([0.0503, 0.0592, 0.0642, 0.0452, 0.0602, 0.0507, 0.0650, 0.0510], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 03:52:33,141 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 03:52:58,933 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=95622.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:53:08,213 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=95630.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:53:14,610 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95635.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:53:22,142 INFO [train.py:903] (0/4) Epoch 15, batch 50, loss[loss=0.2149, simple_loss=0.2829, pruned_loss=0.07347, over 19393.00 frames. ], tot_loss[loss=0.2264, simple_loss=0.3024, pruned_loss=0.07516, over 870129.00 frames. ], batch size: 48, lr: 5.63e-03, grad_scale: 8.0 +2023-04-02 03:53:57,894 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5110, 1.2467, 1.4801, 1.2451, 2.1240, 0.9864, 2.0693, 2.4352], + device='cuda:0'), covar=tensor([0.0698, 0.2564, 0.2461, 0.1569, 0.0953, 0.1975, 0.0904, 0.0447], + device='cuda:0'), in_proj_covar=tensor([0.0374, 0.0347, 0.0367, 0.0327, 0.0357, 0.0336, 0.0344, 0.0371], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:53:58,793 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 03:54:07,437 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-04-02 03:54:20,259 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.439e+02 5.472e+02 6.461e+02 8.212e+02 1.912e+03, threshold=1.292e+03, percent-clipped=7.0 +2023-04-02 03:54:26,852 INFO [train.py:903] (0/4) Epoch 15, batch 100, loss[loss=0.1869, simple_loss=0.2577, pruned_loss=0.05802, over 19748.00 frames. ], tot_loss[loss=0.2223, simple_loss=0.2982, pruned_loss=0.07316, over 1518597.88 frames. ], batch size: 46, lr: 5.62e-03, grad_scale: 8.0 +2023-04-02 03:54:37,471 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 03:54:37,590 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=95701.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:55:22,830 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95737.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:55:28,253 INFO [train.py:903] (0/4) Epoch 15, batch 150, loss[loss=0.2096, simple_loss=0.2926, pruned_loss=0.06325, over 19543.00 frames. ], tot_loss[loss=0.2218, simple_loss=0.2979, pruned_loss=0.07288, over 2037877.21 frames. ], batch size: 56, lr: 5.62e-03, grad_scale: 8.0 +2023-04-02 03:55:32,135 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95745.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:56:23,974 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.923e+02 5.351e+02 6.322e+02 7.452e+02 1.833e+03, threshold=1.264e+03, percent-clipped=1.0 +2023-04-02 03:56:27,309 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 03:56:28,492 INFO [train.py:903] (0/4) Epoch 15, batch 200, loss[loss=0.2102, simple_loss=0.2976, pruned_loss=0.06142, over 19419.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2967, pruned_loss=0.07196, over 2435627.11 frames. ], batch size: 70, lr: 5.62e-03, grad_scale: 8.0 +2023-04-02 03:56:45,596 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4992, 2.1761, 2.1989, 2.7019, 2.4901, 2.3229, 1.9936, 2.7218], + device='cuda:0'), covar=tensor([0.0903, 0.1746, 0.1373, 0.0958, 0.1260, 0.0454, 0.1367, 0.0592], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0357, 0.0301, 0.0247, 0.0301, 0.0247, 0.0291, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:56:51,096 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0830, 1.2283, 1.4440, 1.4092, 2.6872, 1.0793, 2.1271, 3.0111], + device='cuda:0'), covar=tensor([0.0571, 0.2708, 0.2700, 0.1692, 0.0773, 0.2264, 0.1072, 0.0333], + device='cuda:0'), in_proj_covar=tensor([0.0374, 0.0346, 0.0369, 0.0328, 0.0354, 0.0337, 0.0344, 0.0372], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 03:56:59,713 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=95816.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:57:15,624 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95830.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 03:57:24,872 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95838.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 03:57:29,640 INFO [train.py:903] (0/4) Epoch 15, batch 250, loss[loss=0.2215, simple_loss=0.2854, pruned_loss=0.07883, over 19328.00 frames. ], tot_loss[loss=0.2207, simple_loss=0.297, pruned_loss=0.07223, over 2742416.84 frames. ], batch size: 44, lr: 5.62e-03, grad_scale: 8.0 +2023-04-02 03:57:56,091 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=95863.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 03:57:58,213 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95865.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 03:58:24,514 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.450e+02 5.277e+02 6.948e+02 9.039e+02 3.101e+03, threshold=1.390e+03, percent-clipped=9.0 +2023-04-02 03:58:28,338 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2829, 1.3045, 1.5391, 1.4400, 2.2830, 2.0007, 2.2881, 0.8136], + device='cuda:0'), covar=tensor([0.2264, 0.3903, 0.2373, 0.1809, 0.1340, 0.1963, 0.1328, 0.3960], + device='cuda:0'), in_proj_covar=tensor([0.0498, 0.0589, 0.0636, 0.0449, 0.0599, 0.0503, 0.0644, 0.0507], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 03:58:30,109 INFO [train.py:903] (0/4) Epoch 15, batch 300, loss[loss=0.2352, simple_loss=0.3178, pruned_loss=0.07628, over 19663.00 frames. ], tot_loss[loss=0.2229, simple_loss=0.2987, pruned_loss=0.07357, over 2975016.84 frames. ], batch size: 58, lr: 5.62e-03, grad_scale: 8.0 +2023-04-02 03:59:32,840 INFO [train.py:903] (0/4) Epoch 15, batch 350, loss[loss=0.2313, simple_loss=0.3101, pruned_loss=0.07624, over 19466.00 frames. ], tot_loss[loss=0.2222, simple_loss=0.2977, pruned_loss=0.07331, over 3164679.19 frames. ], batch size: 64, lr: 5.62e-03, grad_scale: 8.0 +2023-04-02 03:59:33,871 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 03:59:57,911 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5483, 4.0332, 4.2027, 4.1817, 1.7394, 3.9249, 3.4467, 3.9176], + device='cuda:0'), covar=tensor([0.1473, 0.1036, 0.0574, 0.0659, 0.5245, 0.0911, 0.0637, 0.1076], + device='cuda:0'), in_proj_covar=tensor([0.0719, 0.0646, 0.0856, 0.0740, 0.0762, 0.0600, 0.0516, 0.0787], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 04:00:17,404 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=95979.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:00:28,195 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.384e+02 5.018e+02 5.906e+02 6.897e+02 1.495e+03, threshold=1.181e+03, percent-clipped=1.0 +2023-04-02 04:00:32,777 INFO [train.py:903] (0/4) Epoch 15, batch 400, loss[loss=0.2399, simple_loss=0.3188, pruned_loss=0.0805, over 19669.00 frames. ], tot_loss[loss=0.224, simple_loss=0.2993, pruned_loss=0.07433, over 3311753.18 frames. ], batch size: 55, lr: 5.62e-03, grad_scale: 8.0 +2023-04-02 04:00:34,350 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=95993.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:00:39,871 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=95998.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:00:42,036 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-96000.pt +2023-04-02 04:00:44,427 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96001.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:01:04,731 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96018.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:01:16,006 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96026.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:01:33,842 INFO [train.py:903] (0/4) Epoch 15, batch 450, loss[loss=0.2326, simple_loss=0.3037, pruned_loss=0.08078, over 13338.00 frames. ], tot_loss[loss=0.2229, simple_loss=0.2985, pruned_loss=0.07364, over 3429887.04 frames. ], batch size: 135, lr: 5.61e-03, grad_scale: 8.0 +2023-04-02 04:02:07,802 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 04:02:07,849 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 04:02:12,861 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96072.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:02:31,244 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.196e+02 4.837e+02 5.995e+02 7.453e+02 1.580e+03, threshold=1.199e+03, percent-clipped=6.0 +2023-04-02 04:02:36,671 INFO [train.py:903] (0/4) Epoch 15, batch 500, loss[loss=0.233, simple_loss=0.311, pruned_loss=0.07753, over 19527.00 frames. ], tot_loss[loss=0.2218, simple_loss=0.2975, pruned_loss=0.0731, over 3522286.65 frames. ], batch size: 54, lr: 5.61e-03, grad_scale: 8.0 +2023-04-02 04:02:39,300 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96094.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:02:43,774 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96097.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:02:44,798 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96098.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:03:38,908 INFO [train.py:903] (0/4) Epoch 15, batch 550, loss[loss=0.277, simple_loss=0.3369, pruned_loss=0.1085, over 19362.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.2997, pruned_loss=0.07442, over 3585489.37 frames. ], batch size: 66, lr: 5.61e-03, grad_scale: 8.0 +2023-04-02 04:04:12,240 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-02 04:04:18,069 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96174.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 04:04:29,236 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96183.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:04:31,635 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-04-02 04:04:35,629 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.160e+02 5.342e+02 6.491e+02 8.104e+02 1.503e+03, threshold=1.298e+03, percent-clipped=3.0 +2023-04-02 04:04:40,051 INFO [train.py:903] (0/4) Epoch 15, batch 600, loss[loss=0.216, simple_loss=0.302, pruned_loss=0.065, over 19770.00 frames. ], tot_loss[loss=0.2243, simple_loss=0.2997, pruned_loss=0.07441, over 3648266.23 frames. ], batch size: 56, lr: 5.61e-03, grad_scale: 8.0 +2023-04-02 04:04:53,090 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5698, 1.1827, 1.4089, 1.2424, 2.2273, 0.9961, 1.9777, 2.4832], + device='cuda:0'), covar=tensor([0.0621, 0.2633, 0.2629, 0.1556, 0.0868, 0.1925, 0.0964, 0.0440], + device='cuda:0'), in_proj_covar=tensor([0.0371, 0.0346, 0.0366, 0.0328, 0.0352, 0.0335, 0.0343, 0.0373], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:05:00,801 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96209.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:05:20,405 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 04:05:43,304 INFO [train.py:903] (0/4) Epoch 15, batch 650, loss[loss=0.2331, simple_loss=0.309, pruned_loss=0.07861, over 19538.00 frames. ], tot_loss[loss=0.2244, simple_loss=0.3004, pruned_loss=0.07421, over 3684893.00 frames. ], batch size: 64, lr: 5.61e-03, grad_scale: 8.0 +2023-04-02 04:06:41,549 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.029e+02 5.105e+02 6.385e+02 8.770e+02 1.706e+03, threshold=1.277e+03, percent-clipped=3.0 +2023-04-02 04:06:42,955 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96289.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 04:06:46,038 INFO [train.py:903] (0/4) Epoch 15, batch 700, loss[loss=0.218, simple_loss=0.3016, pruned_loss=0.06722, over 19385.00 frames. ], tot_loss[loss=0.2223, simple_loss=0.2988, pruned_loss=0.07295, over 3716475.55 frames. ], batch size: 70, lr: 5.61e-03, grad_scale: 8.0 +2023-04-02 04:07:26,437 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96324.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:07:47,381 INFO [train.py:903] (0/4) Epoch 15, batch 750, loss[loss=0.2127, simple_loss=0.2968, pruned_loss=0.0643, over 19728.00 frames. ], tot_loss[loss=0.2221, simple_loss=0.2987, pruned_loss=0.07281, over 3747181.99 frames. ], batch size: 63, lr: 5.61e-03, grad_scale: 8.0 +2023-04-02 04:07:47,558 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96342.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:07:47,709 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96342.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:07:57,775 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96350.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:08:28,492 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96375.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:08:44,429 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.337e+02 5.276e+02 6.207e+02 7.536e+02 1.572e+03, threshold=1.241e+03, percent-clipped=2.0 +2023-04-02 04:08:47,393 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.66 vs. limit=5.0 +2023-04-02 04:08:49,803 INFO [train.py:903] (0/4) Epoch 15, batch 800, loss[loss=0.1976, simple_loss=0.2825, pruned_loss=0.05633, over 19668.00 frames. ], tot_loss[loss=0.2224, simple_loss=0.2986, pruned_loss=0.07308, over 3773235.63 frames. ], batch size: 58, lr: 5.60e-03, grad_scale: 8.0 +2023-04-02 04:08:53,730 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7236, 1.4573, 1.5021, 2.1029, 1.5644, 1.9228, 1.9422, 1.7241], + device='cuda:0'), covar=tensor([0.0808, 0.0945, 0.1027, 0.0811, 0.0910, 0.0787, 0.0886, 0.0694], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0223, 0.0224, 0.0243, 0.0228, 0.0210, 0.0191, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 04:09:04,712 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 04:09:09,582 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96408.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:09:17,553 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8277, 2.0808, 2.3612, 2.2968, 3.1437, 3.6114, 3.5534, 3.8803], + device='cuda:0'), covar=tensor([0.1251, 0.2575, 0.2321, 0.1618, 0.0874, 0.0332, 0.0164, 0.0204], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0303, 0.0330, 0.0253, 0.0225, 0.0167, 0.0206, 0.0220], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 04:09:50,620 INFO [train.py:903] (0/4) Epoch 15, batch 850, loss[loss=0.201, simple_loss=0.2773, pruned_loss=0.06237, over 19411.00 frames. ], tot_loss[loss=0.2228, simple_loss=0.2987, pruned_loss=0.0734, over 3772566.21 frames. ], batch size: 48, lr: 5.60e-03, grad_scale: 8.0 +2023-04-02 04:09:51,930 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96442.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:10:09,368 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96457.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:10:41,144 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 04:10:47,719 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.001e+02 5.238e+02 6.465e+02 7.879e+02 1.664e+03, threshold=1.293e+03, percent-clipped=4.0 +2023-04-02 04:10:52,491 INFO [train.py:903] (0/4) Epoch 15, batch 900, loss[loss=0.1989, simple_loss=0.2716, pruned_loss=0.06306, over 19378.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2985, pruned_loss=0.07347, over 3768924.67 frames. ], batch size: 47, lr: 5.60e-03, grad_scale: 8.0 +2023-04-02 04:11:36,464 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96527.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:11:55,050 INFO [train.py:903] (0/4) Epoch 15, batch 950, loss[loss=0.1934, simple_loss=0.2762, pruned_loss=0.05526, over 19671.00 frames. ], tot_loss[loss=0.2219, simple_loss=0.2982, pruned_loss=0.07283, over 3794066.54 frames. ], batch size: 53, lr: 5.60e-03, grad_scale: 8.0 +2023-04-02 04:11:56,235 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 04:11:59,021 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96545.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 04:12:06,745 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7613, 1.8594, 2.0159, 2.3497, 1.7261, 2.2887, 2.1792, 1.9026], + device='cuda:0'), covar=tensor([0.3593, 0.3128, 0.1682, 0.1817, 0.3261, 0.1637, 0.3805, 0.2892], + device='cuda:0'), in_proj_covar=tensor([0.0826, 0.0864, 0.0668, 0.0898, 0.0814, 0.0748, 0.0802, 0.0731], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 04:12:14,112 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96557.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:12:30,398 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96570.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 04:12:41,763 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96580.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:12:52,447 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.674e+02 5.225e+02 6.143e+02 7.839e+02 1.754e+03, threshold=1.229e+03, percent-clipped=1.0 +2023-04-02 04:12:57,222 INFO [train.py:903] (0/4) Epoch 15, batch 1000, loss[loss=0.1824, simple_loss=0.2574, pruned_loss=0.05369, over 19302.00 frames. ], tot_loss[loss=0.2226, simple_loss=0.2988, pruned_loss=0.07321, over 3798047.91 frames. ], batch size: 44, lr: 5.60e-03, grad_scale: 8.0 +2023-04-02 04:13:13,515 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96605.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:13:51,325 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 04:13:59,278 INFO [train.py:903] (0/4) Epoch 15, batch 1050, loss[loss=0.2042, simple_loss=0.2667, pruned_loss=0.07087, over 19720.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2989, pruned_loss=0.07326, over 3792788.99 frames. ], batch size: 45, lr: 5.60e-03, grad_scale: 8.0 +2023-04-02 04:13:59,656 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96642.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:14:31,326 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 04:14:35,039 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4025, 1.6601, 2.0687, 1.6730, 3.1371, 2.5975, 3.2115, 1.5550], + device='cuda:0'), covar=tensor([0.2318, 0.3778, 0.2304, 0.1789, 0.1514, 0.1938, 0.1758, 0.3678], + device='cuda:0'), in_proj_covar=tensor([0.0501, 0.0589, 0.0640, 0.0449, 0.0601, 0.0505, 0.0643, 0.0507], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 04:14:53,737 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96686.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:14:57,023 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.831e+02 5.170e+02 6.505e+02 8.376e+02 1.590e+03, threshold=1.301e+03, percent-clipped=4.0 +2023-04-02 04:15:01,350 INFO [train.py:903] (0/4) Epoch 15, batch 1100, loss[loss=0.2553, simple_loss=0.3298, pruned_loss=0.09033, over 19536.00 frames. ], tot_loss[loss=0.2235, simple_loss=0.2995, pruned_loss=0.07374, over 3799356.71 frames. ], batch size: 56, lr: 5.60e-03, grad_scale: 4.0 +2023-04-02 04:15:28,113 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96713.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:15:58,779 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96738.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:16:02,837 INFO [train.py:903] (0/4) Epoch 15, batch 1150, loss[loss=0.1981, simple_loss=0.2699, pruned_loss=0.06314, over 19733.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2979, pruned_loss=0.07258, over 3806470.34 frames. ], batch size: 47, lr: 5.59e-03, grad_scale: 4.0 +2023-04-02 04:16:15,357 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=96752.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:17:01,609 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.369e+02 5.122e+02 6.460e+02 8.216e+02 1.619e+03, threshold=1.292e+03, percent-clipped=5.0 +2023-04-02 04:17:06,203 INFO [train.py:903] (0/4) Epoch 15, batch 1200, loss[loss=0.1733, simple_loss=0.2552, pruned_loss=0.04567, over 19037.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.2972, pruned_loss=0.07216, over 3818694.13 frames. ], batch size: 42, lr: 5.59e-03, grad_scale: 8.0 +2023-04-02 04:17:08,719 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=96794.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:17:17,132 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96801.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:17:32,624 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96813.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:17:39,079 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 04:18:03,059 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96838.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:18:07,992 INFO [train.py:903] (0/4) Epoch 15, batch 1250, loss[loss=0.2659, simple_loss=0.3318, pruned_loss=0.1, over 18202.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2991, pruned_loss=0.07346, over 3803793.32 frames. ], batch size: 83, lr: 5.59e-03, grad_scale: 8.0 +2023-04-02 04:18:38,268 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=96867.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:19:05,686 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.585e+02 5.442e+02 6.897e+02 8.528e+02 1.967e+03, threshold=1.379e+03, percent-clipped=7.0 +2023-04-02 04:19:09,069 INFO [train.py:903] (0/4) Epoch 15, batch 1300, loss[loss=0.2713, simple_loss=0.3318, pruned_loss=0.1054, over 13866.00 frames. ], tot_loss[loss=0.224, simple_loss=0.2999, pruned_loss=0.07405, over 3814556.29 frames. ], batch size: 135, lr: 5.59e-03, grad_scale: 8.0 +2023-04-02 04:19:17,636 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=96898.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:19:48,600 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=96923.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:20:12,478 INFO [train.py:903] (0/4) Epoch 15, batch 1350, loss[loss=0.1866, simple_loss=0.2562, pruned_loss=0.05849, over 16083.00 frames. ], tot_loss[loss=0.2235, simple_loss=0.2995, pruned_loss=0.07378, over 3809959.65 frames. ], batch size: 35, lr: 5.59e-03, grad_scale: 8.0 +2023-04-02 04:21:11,449 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.829e+02 4.520e+02 5.515e+02 7.272e+02 1.782e+03, threshold=1.103e+03, percent-clipped=1.0 +2023-04-02 04:21:15,868 INFO [train.py:903] (0/4) Epoch 15, batch 1400, loss[loss=0.2209, simple_loss=0.3072, pruned_loss=0.0673, over 19620.00 frames. ], tot_loss[loss=0.2233, simple_loss=0.2992, pruned_loss=0.0737, over 3805330.98 frames. ], batch size: 57, lr: 5.59e-03, grad_scale: 8.0 +2023-04-02 04:21:49,669 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97018.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:22:15,030 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.59 vs. limit=2.0 +2023-04-02 04:22:19,077 INFO [train.py:903] (0/4) Epoch 15, batch 1450, loss[loss=0.2306, simple_loss=0.315, pruned_loss=0.07305, over 19660.00 frames. ], tot_loss[loss=0.2222, simple_loss=0.2984, pruned_loss=0.07302, over 3815927.72 frames. ], batch size: 58, lr: 5.59e-03, grad_scale: 8.0 +2023-04-02 04:22:20,281 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 04:22:38,930 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97057.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:23:09,820 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97082.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:23:11,121 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2765, 2.2769, 2.4209, 3.1816, 2.1883, 3.0580, 2.8367, 2.2983], + device='cuda:0'), covar=tensor([0.3923, 0.3746, 0.1590, 0.2155, 0.4181, 0.1786, 0.3699, 0.2870], + device='cuda:0'), in_proj_covar=tensor([0.0833, 0.0871, 0.0673, 0.0905, 0.0817, 0.0750, 0.0808, 0.0737], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 04:23:18,581 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.200e+02 5.198e+02 6.417e+02 9.294e+02 1.968e+03, threshold=1.283e+03, percent-clipped=11.0 +2023-04-02 04:23:21,917 INFO [train.py:903] (0/4) Epoch 15, batch 1500, loss[loss=0.1912, simple_loss=0.28, pruned_loss=0.05119, over 19756.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2987, pruned_loss=0.0733, over 3797749.84 frames. ], batch size: 54, lr: 5.58e-03, grad_scale: 8.0 +2023-04-02 04:24:01,252 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97123.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:24:20,464 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97138.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:24:24,794 INFO [train.py:903] (0/4) Epoch 15, batch 1550, loss[loss=0.2203, simple_loss=0.3054, pruned_loss=0.0676, over 19773.00 frames. ], tot_loss[loss=0.2221, simple_loss=0.2986, pruned_loss=0.07283, over 3807985.96 frames. ], batch size: 56, lr: 5.58e-03, grad_scale: 8.0 +2023-04-02 04:24:25,253 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6889, 1.8147, 1.9411, 2.0671, 1.5249, 1.9635, 2.0916, 1.8783], + device='cuda:0'), covar=tensor([0.3437, 0.2774, 0.1566, 0.1700, 0.3125, 0.1592, 0.3812, 0.2654], + device='cuda:0'), in_proj_covar=tensor([0.0835, 0.0871, 0.0674, 0.0903, 0.0817, 0.0752, 0.0806, 0.0738], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 04:24:31,871 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97148.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:25:22,554 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.982e+02 5.223e+02 6.490e+02 8.468e+02 1.572e+03, threshold=1.298e+03, percent-clipped=7.0 +2023-04-02 04:25:26,805 INFO [train.py:903] (0/4) Epoch 15, batch 1600, loss[loss=0.2119, simple_loss=0.2981, pruned_loss=0.06289, over 19292.00 frames. ], tot_loss[loss=0.2217, simple_loss=0.2983, pruned_loss=0.07259, over 3809541.93 frames. ], batch size: 66, lr: 5.58e-03, grad_scale: 8.0 +2023-04-02 04:25:51,578 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 04:26:20,094 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-02 04:26:28,830 INFO [train.py:903] (0/4) Epoch 15, batch 1650, loss[loss=0.2881, simple_loss=0.3507, pruned_loss=0.1127, over 17417.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2976, pruned_loss=0.0727, over 3814984.05 frames. ], batch size: 101, lr: 5.58e-03, grad_scale: 8.0 +2023-04-02 04:26:42,629 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97253.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:27:03,865 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5833, 1.2321, 1.1857, 1.4433, 1.1800, 1.3584, 1.2093, 1.4073], + device='cuda:0'), covar=tensor([0.1008, 0.1251, 0.1481, 0.0990, 0.1130, 0.0579, 0.1407, 0.0794], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0349, 0.0296, 0.0242, 0.0293, 0.0244, 0.0287, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:27:14,835 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97278.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:27:27,488 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.574e+02 5.168e+02 6.563e+02 8.938e+02 1.305e+03, threshold=1.313e+03, percent-clipped=1.0 +2023-04-02 04:27:30,913 INFO [train.py:903] (0/4) Epoch 15, batch 1700, loss[loss=0.2198, simple_loss=0.2956, pruned_loss=0.07202, over 19653.00 frames. ], tot_loss[loss=0.2217, simple_loss=0.2978, pruned_loss=0.07281, over 3824184.68 frames. ], batch size: 55, lr: 5.58e-03, grad_scale: 8.0 +2023-04-02 04:27:49,503 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-04-02 04:28:02,153 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3307, 3.0211, 2.2981, 2.7558, 0.7418, 2.9958, 2.8420, 2.9382], + device='cuda:0'), covar=tensor([0.1040, 0.1399, 0.1954, 0.1046, 0.3894, 0.0989, 0.0991, 0.1269], + device='cuda:0'), in_proj_covar=tensor([0.0457, 0.0384, 0.0458, 0.0326, 0.0389, 0.0391, 0.0382, 0.0414], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:28:11,284 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 04:28:32,699 INFO [train.py:903] (0/4) Epoch 15, batch 1750, loss[loss=0.2543, simple_loss=0.3184, pruned_loss=0.09509, over 13430.00 frames. ], tot_loss[loss=0.2211, simple_loss=0.2969, pruned_loss=0.07268, over 3828338.48 frames. ], batch size: 137, lr: 5.58e-03, grad_scale: 8.0 +2023-04-02 04:28:55,667 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97360.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:28:57,837 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97362.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:29:01,593 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97365.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:29:30,674 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.259e+02 5.096e+02 6.393e+02 8.037e+02 1.428e+03, threshold=1.279e+03, percent-clipped=5.0 +2023-04-02 04:29:33,929 INFO [train.py:903] (0/4) Epoch 15, batch 1800, loss[loss=0.2233, simple_loss=0.3065, pruned_loss=0.07007, over 19594.00 frames. ], tot_loss[loss=0.2214, simple_loss=0.2971, pruned_loss=0.07279, over 3821725.49 frames. ], batch size: 61, lr: 5.58e-03, grad_scale: 8.0 +2023-04-02 04:29:34,223 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97392.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:30:03,912 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1991, 1.2867, 1.2249, 1.0699, 1.0862, 1.1360, 0.0949, 0.4135], + device='cuda:0'), covar=tensor([0.0581, 0.0560, 0.0356, 0.0467, 0.1144, 0.0478, 0.1010, 0.0888], + device='cuda:0'), in_proj_covar=tensor([0.0342, 0.0340, 0.0337, 0.0366, 0.0438, 0.0365, 0.0319, 0.0325], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 04:30:32,249 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 04:30:36,842 INFO [train.py:903] (0/4) Epoch 15, batch 1850, loss[loss=0.1935, simple_loss=0.2718, pruned_loss=0.05759, over 19769.00 frames. ], tot_loss[loss=0.2212, simple_loss=0.2972, pruned_loss=0.07257, over 3830471.95 frames. ], batch size: 48, lr: 5.57e-03, grad_scale: 8.0 +2023-04-02 04:31:10,814 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 04:31:21,325 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97477.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:31:25,820 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8743, 4.2864, 4.5922, 4.6001, 1.4729, 4.2919, 3.7083, 4.2720], + device='cuda:0'), covar=tensor([0.1547, 0.0803, 0.0624, 0.0591, 0.5899, 0.0879, 0.0641, 0.1101], + device='cuda:0'), in_proj_covar=tensor([0.0728, 0.0655, 0.0865, 0.0744, 0.0773, 0.0609, 0.0523, 0.0797], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 04:31:35,889 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.266e+02 4.535e+02 5.674e+02 7.772e+02 1.771e+03, threshold=1.135e+03, percent-clipped=3.0 +2023-04-02 04:31:39,245 INFO [train.py:903] (0/4) Epoch 15, batch 1900, loss[loss=0.2343, simple_loss=0.3123, pruned_loss=0.07817, over 19283.00 frames. ], tot_loss[loss=0.2218, simple_loss=0.2978, pruned_loss=0.07287, over 3825304.81 frames. ], batch size: 66, lr: 5.57e-03, grad_scale: 8.0 +2023-04-02 04:31:47,722 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8343, 3.2389, 3.2841, 3.3210, 1.2283, 3.1597, 2.7135, 3.0386], + device='cuda:0'), covar=tensor([0.1483, 0.0956, 0.0803, 0.0816, 0.5039, 0.0927, 0.0833, 0.1274], + device='cuda:0'), in_proj_covar=tensor([0.0725, 0.0654, 0.0863, 0.0743, 0.0771, 0.0607, 0.0522, 0.0794], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 04:31:58,018 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 04:32:00,969 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97509.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:32:02,870 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 04:32:28,310 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 04:32:32,290 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97534.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:32:40,694 INFO [train.py:903] (0/4) Epoch 15, batch 1950, loss[loss=0.1983, simple_loss=0.2817, pruned_loss=0.05745, over 19464.00 frames. ], tot_loss[loss=0.2222, simple_loss=0.2984, pruned_loss=0.07299, over 3819847.15 frames. ], batch size: 49, lr: 5.57e-03, grad_scale: 8.0 +2023-04-02 04:33:19,888 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97573.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:33:39,667 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.621e+02 4.980e+02 6.456e+02 8.636e+02 2.349e+03, threshold=1.291e+03, percent-clipped=8.0 +2023-04-02 04:33:43,308 INFO [train.py:903] (0/4) Epoch 15, batch 2000, loss[loss=0.2308, simple_loss=0.3114, pruned_loss=0.07512, over 19602.00 frames. ], tot_loss[loss=0.2225, simple_loss=0.2988, pruned_loss=0.07306, over 3820674.90 frames. ], batch size: 61, lr: 5.57e-03, grad_scale: 8.0 +2023-04-02 04:34:06,549 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-04-02 04:34:20,870 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97622.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:34:35,245 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-04-02 04:34:42,316 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 04:34:47,029 INFO [train.py:903] (0/4) Epoch 15, batch 2050, loss[loss=0.2006, simple_loss=0.2848, pruned_loss=0.0582, over 19743.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2984, pruned_loss=0.07231, over 3830437.59 frames. ], batch size: 51, lr: 5.57e-03, grad_scale: 8.0 +2023-04-02 04:35:01,879 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 04:35:03,038 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 04:35:23,967 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 04:35:47,019 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.214e+02 4.998e+02 6.604e+02 7.928e+02 1.987e+03, threshold=1.321e+03, percent-clipped=5.0 +2023-04-02 04:35:50,611 INFO [train.py:903] (0/4) Epoch 15, batch 2100, loss[loss=0.1996, simple_loss=0.2905, pruned_loss=0.05431, over 19524.00 frames. ], tot_loss[loss=0.2214, simple_loss=0.2982, pruned_loss=0.0723, over 3839203.28 frames. ], batch size: 56, lr: 5.57e-03, grad_scale: 8.0 +2023-04-02 04:35:50,941 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97692.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 04:35:56,743 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5092, 4.0572, 4.2145, 4.2097, 1.5667, 3.9613, 3.4928, 3.9172], + device='cuda:0'), covar=tensor([0.1551, 0.0919, 0.0669, 0.0658, 0.5692, 0.0898, 0.0666, 0.1134], + device='cuda:0'), in_proj_covar=tensor([0.0732, 0.0658, 0.0867, 0.0747, 0.0777, 0.0612, 0.0524, 0.0798], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 04:36:04,922 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97704.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:36:11,599 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97709.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:36:20,730 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 04:36:41,964 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97733.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:36:43,973 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 04:36:45,265 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97736.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:36:46,642 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97737.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:36:49,876 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9941, 1.1976, 1.4714, 0.6233, 2.0155, 2.4190, 2.1069, 2.5794], + device='cuda:0'), covar=tensor([0.1522, 0.3658, 0.3200, 0.2547, 0.0585, 0.0252, 0.0350, 0.0309], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0305, 0.0331, 0.0253, 0.0225, 0.0169, 0.0208, 0.0221], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 04:36:51,864 INFO [train.py:903] (0/4) Epoch 15, batch 2150, loss[loss=0.2426, simple_loss=0.3178, pruned_loss=0.0837, over 18708.00 frames. ], tot_loss[loss=0.2222, simple_loss=0.299, pruned_loss=0.07268, over 3837709.26 frames. ], batch size: 74, lr: 5.57e-03, grad_scale: 8.0 +2023-04-02 04:37:12,246 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=97758.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:37:49,742 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.357e+02 4.862e+02 6.186e+02 7.185e+02 1.323e+03, threshold=1.237e+03, percent-clipped=1.0 +2023-04-02 04:37:54,005 INFO [train.py:903] (0/4) Epoch 15, batch 2200, loss[loss=0.1834, simple_loss=0.2578, pruned_loss=0.05444, over 19410.00 frames. ], tot_loss[loss=0.2216, simple_loss=0.2985, pruned_loss=0.07238, over 3830079.59 frames. ], batch size: 48, lr: 5.56e-03, grad_scale: 8.0 +2023-04-02 04:37:55,569 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1725, 2.0196, 1.7914, 1.6548, 1.5723, 1.7545, 0.3485, 0.9856], + device='cuda:0'), covar=tensor([0.0465, 0.0486, 0.0383, 0.0610, 0.0901, 0.0650, 0.1004, 0.0829], + device='cuda:0'), in_proj_covar=tensor([0.0342, 0.0340, 0.0336, 0.0365, 0.0438, 0.0364, 0.0318, 0.0324], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 04:38:28,065 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97819.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:38:33,738 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97824.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:38:57,679 INFO [train.py:903] (0/4) Epoch 15, batch 2250, loss[loss=0.2591, simple_loss=0.3268, pruned_loss=0.09567, over 19547.00 frames. ], tot_loss[loss=0.2221, simple_loss=0.2985, pruned_loss=0.0728, over 3819224.82 frames. ], batch size: 54, lr: 5.56e-03, grad_scale: 8.0 +2023-04-02 04:39:09,291 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=97851.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:39:56,866 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.382e+02 4.977e+02 6.292e+02 8.077e+02 1.831e+03, threshold=1.258e+03, percent-clipped=5.0 +2023-04-02 04:40:00,323 INFO [train.py:903] (0/4) Epoch 15, batch 2300, loss[loss=0.2263, simple_loss=0.3107, pruned_loss=0.07093, over 19523.00 frames. ], tot_loss[loss=0.2214, simple_loss=0.2984, pruned_loss=0.07218, over 3819778.68 frames. ], batch size: 56, lr: 5.56e-03, grad_scale: 8.0 +2023-04-02 04:40:12,674 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 04:40:16,283 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=97906.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:40:22,276 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2754, 1.1792, 1.2512, 1.3599, 1.0976, 1.3958, 1.3062, 1.3282], + device='cuda:0'), covar=tensor([0.0922, 0.1034, 0.1081, 0.0697, 0.0864, 0.0829, 0.0896, 0.0765], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0223, 0.0242, 0.0228, 0.0208, 0.0191, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 04:40:29,875 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=97917.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:40:54,965 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2982, 3.0001, 2.3015, 2.6871, 0.7555, 2.9460, 2.8298, 2.9427], + device='cuda:0'), covar=tensor([0.1100, 0.1352, 0.2013, 0.1023, 0.3990, 0.1000, 0.1081, 0.1320], + device='cuda:0'), in_proj_covar=tensor([0.0462, 0.0381, 0.0461, 0.0325, 0.0392, 0.0393, 0.0382, 0.0417], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:41:01,540 INFO [train.py:903] (0/4) Epoch 15, batch 2350, loss[loss=0.2273, simple_loss=0.305, pruned_loss=0.07481, over 19701.00 frames. ], tot_loss[loss=0.2214, simple_loss=0.2981, pruned_loss=0.07237, over 3821746.10 frames. ], batch size: 59, lr: 5.56e-03, grad_scale: 8.0 +2023-04-02 04:41:44,926 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 04:41:58,283 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.032e+02 5.416e+02 6.244e+02 7.823e+02 1.587e+03, threshold=1.249e+03, percent-clipped=4.0 +2023-04-02 04:41:58,399 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 04:42:02,746 INFO [train.py:903] (0/4) Epoch 15, batch 2400, loss[loss=0.2079, simple_loss=0.2906, pruned_loss=0.06263, over 19525.00 frames. ], tot_loss[loss=0.2219, simple_loss=0.2986, pruned_loss=0.07261, over 3815768.23 frames. ], batch size: 56, lr: 5.56e-03, grad_scale: 8.0 +2023-04-02 04:42:04,439 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=97993.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:42:12,201 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-98000.pt +2023-04-02 04:42:29,190 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9341, 4.4998, 2.6315, 4.0222, 1.1370, 4.3459, 4.3044, 4.3887], + device='cuda:0'), covar=tensor([0.0542, 0.0901, 0.1995, 0.0695, 0.3682, 0.0690, 0.0763, 0.1129], + device='cuda:0'), in_proj_covar=tensor([0.0459, 0.0379, 0.0457, 0.0323, 0.0388, 0.0391, 0.0379, 0.0415], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:42:36,216 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98018.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:42:53,753 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98032.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:42:58,104 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98036.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 04:43:05,605 INFO [train.py:903] (0/4) Epoch 15, batch 2450, loss[loss=0.2704, simple_loss=0.3344, pruned_loss=0.1032, over 19575.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2982, pruned_loss=0.07245, over 3823533.02 frames. ], batch size: 61, lr: 5.56e-03, grad_scale: 8.0 +2023-04-02 04:43:32,178 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1654, 3.7222, 2.2367, 2.3179, 3.3310, 2.1518, 1.2979, 2.2266], + device='cuda:0'), covar=tensor([0.1257, 0.0529, 0.0955, 0.0780, 0.0550, 0.1027, 0.1006, 0.0634], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0306, 0.0322, 0.0248, 0.0239, 0.0327, 0.0290, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:43:47,599 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98075.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:43:54,239 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98080.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:44:05,272 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.133e+02 5.305e+02 6.204e+02 8.091e+02 1.870e+03, threshold=1.241e+03, percent-clipped=5.0 +2023-04-02 04:44:09,743 INFO [train.py:903] (0/4) Epoch 15, batch 2500, loss[loss=0.1784, simple_loss=0.2635, pruned_loss=0.04663, over 19841.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2972, pruned_loss=0.07173, over 3821430.01 frames. ], batch size: 52, lr: 5.56e-03, grad_scale: 8.0 +2023-04-02 04:44:19,636 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98100.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:44:25,566 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98105.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:44:27,995 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98107.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:44:29,491 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.09 vs. limit=2.0 +2023-04-02 04:44:51,253 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98124.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:45:00,624 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98132.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:45:12,189 INFO [train.py:903] (0/4) Epoch 15, batch 2550, loss[loss=0.2241, simple_loss=0.3114, pruned_loss=0.06836, over 19790.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2968, pruned_loss=0.07136, over 3836116.53 frames. ], batch size: 56, lr: 5.55e-03, grad_scale: 8.0 +2023-04-02 04:45:23,230 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98151.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 04:45:46,897 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3708, 1.4682, 1.7768, 1.6239, 2.6572, 2.2534, 2.7467, 1.0771], + device='cuda:0'), covar=tensor([0.2341, 0.4008, 0.2541, 0.1797, 0.1425, 0.2076, 0.1477, 0.4047], + device='cuda:0'), in_proj_covar=tensor([0.0504, 0.0593, 0.0646, 0.0450, 0.0602, 0.0507, 0.0646, 0.0507], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 04:46:07,105 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 04:46:10,537 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.981e+02 5.247e+02 6.339e+02 8.638e+02 2.352e+03, threshold=1.268e+03, percent-clipped=5.0 +2023-04-02 04:46:14,051 INFO [train.py:903] (0/4) Epoch 15, batch 2600, loss[loss=0.2218, simple_loss=0.3094, pruned_loss=0.0671, over 19684.00 frames. ], tot_loss[loss=0.2204, simple_loss=0.2977, pruned_loss=0.07157, over 3839020.12 frames. ], batch size: 59, lr: 5.55e-03, grad_scale: 8.0 +2023-04-02 04:46:24,898 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0398, 4.9894, 5.8436, 5.7801, 1.8350, 5.4558, 4.7000, 5.4586], + device='cuda:0'), covar=tensor([0.1535, 0.0830, 0.0523, 0.0582, 0.5868, 0.0622, 0.0578, 0.1122], + device='cuda:0'), in_proj_covar=tensor([0.0729, 0.0660, 0.0867, 0.0746, 0.0777, 0.0614, 0.0525, 0.0799], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 04:47:18,038 INFO [train.py:903] (0/4) Epoch 15, batch 2650, loss[loss=0.2011, simple_loss=0.2669, pruned_loss=0.06763, over 19726.00 frames. ], tot_loss[loss=0.2206, simple_loss=0.2973, pruned_loss=0.07194, over 3828585.11 frames. ], batch size: 46, lr: 5.55e-03, grad_scale: 8.0 +2023-04-02 04:47:28,824 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98250.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:47:39,902 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 04:48:17,337 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98288.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:48:18,022 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.302e+02 4.990e+02 6.118e+02 7.575e+02 1.335e+03, threshold=1.224e+03, percent-clipped=1.0 +2023-04-02 04:48:21,604 INFO [train.py:903] (0/4) Epoch 15, batch 2700, loss[loss=0.2388, simple_loss=0.3207, pruned_loss=0.07844, over 19752.00 frames. ], tot_loss[loss=0.2199, simple_loss=0.297, pruned_loss=0.07135, over 3840104.13 frames. ], batch size: 63, lr: 5.55e-03, grad_scale: 8.0 +2023-04-02 04:48:47,470 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98313.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:49:24,166 INFO [train.py:903] (0/4) Epoch 15, batch 2750, loss[loss=0.2767, simple_loss=0.3348, pruned_loss=0.1093, over 13597.00 frames. ], tot_loss[loss=0.2212, simple_loss=0.2978, pruned_loss=0.07233, over 3818647.64 frames. ], batch size: 136, lr: 5.55e-03, grad_scale: 8.0 +2023-04-02 04:49:45,380 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0230, 1.7969, 1.5599, 2.1812, 1.8056, 1.7642, 1.6145, 1.9592], + device='cuda:0'), covar=tensor([0.0957, 0.1490, 0.1531, 0.0944, 0.1397, 0.0551, 0.1298, 0.0714], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0348, 0.0295, 0.0243, 0.0293, 0.0244, 0.0288, 0.0246], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:49:54,711 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98365.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:50:02,026 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1202, 2.7963, 1.9095, 2.0689, 1.8499, 2.3044, 0.7635, 1.9826], + device='cuda:0'), covar=tensor([0.0554, 0.0529, 0.0687, 0.0973, 0.1095, 0.1028, 0.1306, 0.0963], + device='cuda:0'), in_proj_covar=tensor([0.0343, 0.0342, 0.0340, 0.0368, 0.0443, 0.0369, 0.0320, 0.0328], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 04:50:23,807 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.213e+02 5.173e+02 6.107e+02 7.991e+02 1.431e+03, threshold=1.221e+03, percent-clipped=3.0 +2023-04-02 04:50:27,276 INFO [train.py:903] (0/4) Epoch 15, batch 2800, loss[loss=0.2248, simple_loss=0.3096, pruned_loss=0.07003, over 18134.00 frames. ], tot_loss[loss=0.2211, simple_loss=0.2977, pruned_loss=0.07223, over 3831047.32 frames. ], batch size: 83, lr: 5.55e-03, grad_scale: 8.0 +2023-04-02 04:50:28,753 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3511, 3.0751, 2.3107, 2.7996, 0.6313, 2.9530, 2.8919, 2.9666], + device='cuda:0'), covar=tensor([0.1028, 0.1341, 0.1908, 0.1015, 0.3904, 0.1058, 0.1012, 0.1210], + device='cuda:0'), in_proj_covar=tensor([0.0467, 0.0386, 0.0463, 0.0329, 0.0392, 0.0397, 0.0387, 0.0421], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:50:47,397 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.9803, 5.2994, 3.0065, 4.6758, 1.3450, 5.3200, 5.2723, 5.4938], + device='cuda:0'), covar=tensor([0.0409, 0.0939, 0.1900, 0.0712, 0.3595, 0.0578, 0.0703, 0.0878], + device='cuda:0'), in_proj_covar=tensor([0.0467, 0.0386, 0.0463, 0.0329, 0.0392, 0.0397, 0.0387, 0.0421], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:50:48,826 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98407.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 04:51:13,923 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98428.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:51:18,584 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98432.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 04:51:30,822 INFO [train.py:903] (0/4) Epoch 15, batch 2850, loss[loss=0.2145, simple_loss=0.2955, pruned_loss=0.0668, over 19775.00 frames. ], tot_loss[loss=0.2212, simple_loss=0.2981, pruned_loss=0.07219, over 3836631.07 frames. ], batch size: 56, lr: 5.55e-03, grad_scale: 8.0 +2023-04-02 04:52:03,911 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98468.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:52:30,661 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.993e+02 5.164e+02 6.295e+02 8.927e+02 2.262e+03, threshold=1.259e+03, percent-clipped=4.0 +2023-04-02 04:52:34,177 INFO [train.py:903] (0/4) Epoch 15, batch 2900, loss[loss=0.2096, simple_loss=0.2975, pruned_loss=0.06088, over 19283.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2967, pruned_loss=0.0714, over 3850880.21 frames. ], batch size: 66, lr: 5.54e-03, grad_scale: 8.0 +2023-04-02 04:52:35,474 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 04:52:53,173 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0273, 1.8748, 1.5347, 2.0583, 1.7842, 1.6920, 1.6214, 1.8831], + device='cuda:0'), covar=tensor([0.0966, 0.1351, 0.1568, 0.0966, 0.1371, 0.0587, 0.1334, 0.0815], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0352, 0.0297, 0.0245, 0.0295, 0.0246, 0.0290, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 04:53:36,713 INFO [train.py:903] (0/4) Epoch 15, batch 2950, loss[loss=0.1895, simple_loss=0.2602, pruned_loss=0.05935, over 19403.00 frames. ], tot_loss[loss=0.2207, simple_loss=0.2978, pruned_loss=0.07179, over 3853997.87 frames. ], batch size: 47, lr: 5.54e-03, grad_scale: 8.0 +2023-04-02 04:54:00,567 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98561.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:54:29,226 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98583.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:54:35,383 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.243e+02 5.043e+02 6.244e+02 8.249e+02 2.456e+03, threshold=1.249e+03, percent-clipped=2.0 +2023-04-02 04:54:38,824 INFO [train.py:903] (0/4) Epoch 15, batch 3000, loss[loss=0.2223, simple_loss=0.3065, pruned_loss=0.06907, over 19707.00 frames. ], tot_loss[loss=0.2204, simple_loss=0.2972, pruned_loss=0.07177, over 3849636.37 frames. ], batch size: 59, lr: 5.54e-03, grad_scale: 8.0 +2023-04-02 04:54:38,825 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 04:54:51,337 INFO [train.py:937] (0/4) Epoch 15, validation: loss=0.1735, simple_loss=0.2738, pruned_loss=0.0366, over 944034.00 frames. +2023-04-02 04:54:51,338 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 04:54:51,791 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1328, 2.1578, 2.3673, 2.9157, 2.1816, 2.7316, 2.5177, 2.1737], + device='cuda:0'), covar=tensor([0.3621, 0.3171, 0.1447, 0.1850, 0.3373, 0.1551, 0.3522, 0.2603], + device='cuda:0'), in_proj_covar=tensor([0.0828, 0.0873, 0.0669, 0.0899, 0.0813, 0.0746, 0.0806, 0.0735], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 04:54:53,552 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 04:55:28,720 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98621.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:55:52,976 INFO [train.py:903] (0/4) Epoch 15, batch 3050, loss[loss=0.2183, simple_loss=0.2898, pruned_loss=0.07345, over 19768.00 frames. ], tot_loss[loss=0.2209, simple_loss=0.2973, pruned_loss=0.07228, over 3851162.48 frames. ], batch size: 47, lr: 5.54e-03, grad_scale: 8.0 +2023-04-02 04:55:57,865 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98646.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:56:51,965 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.865e+02 5.607e+02 7.054e+02 9.171e+02 2.046e+03, threshold=1.411e+03, percent-clipped=6.0 +2023-04-02 04:56:54,317 INFO [train.py:903] (0/4) Epoch 15, batch 3100, loss[loss=0.1964, simple_loss=0.2789, pruned_loss=0.057, over 19765.00 frames. ], tot_loss[loss=0.2206, simple_loss=0.2968, pruned_loss=0.07217, over 3844818.93 frames. ], batch size: 46, lr: 5.54e-03, grad_scale: 8.0 +2023-04-02 04:57:58,294 INFO [train.py:903] (0/4) Epoch 15, batch 3150, loss[loss=0.2198, simple_loss=0.3043, pruned_loss=0.06772, over 19459.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2967, pruned_loss=0.0722, over 3848153.02 frames. ], batch size: 64, lr: 5.54e-03, grad_scale: 8.0 +2023-04-02 04:58:26,344 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 04:58:34,636 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98772.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 04:58:58,674 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.173e+02 5.220e+02 6.366e+02 8.908e+02 1.802e+03, threshold=1.273e+03, percent-clipped=4.0 +2023-04-02 04:59:01,102 INFO [train.py:903] (0/4) Epoch 15, batch 3200, loss[loss=0.1936, simple_loss=0.2655, pruned_loss=0.06088, over 19747.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2964, pruned_loss=0.07206, over 3838727.68 frames. ], batch size: 46, lr: 5.54e-03, grad_scale: 8.0 +2023-04-02 04:59:37,257 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0013, 1.4921, 1.6222, 1.4930, 2.8677, 4.4942, 4.3509, 4.8188], + device='cuda:0'), covar=tensor([0.1799, 0.3454, 0.3327, 0.2069, 0.0621, 0.0187, 0.0160, 0.0157], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0303, 0.0330, 0.0251, 0.0223, 0.0168, 0.0206, 0.0222], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 04:59:59,409 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=98839.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:00:02,403 INFO [train.py:903] (0/4) Epoch 15, batch 3250, loss[loss=0.1932, simple_loss=0.2697, pruned_loss=0.05832, over 19751.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2964, pruned_loss=0.07184, over 3830172.35 frames. ], batch size: 47, lr: 5.53e-03, grad_scale: 8.0 +2023-04-02 05:00:24,181 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9396, 4.4903, 2.7553, 3.8216, 0.8671, 4.3533, 4.2514, 4.3202], + device='cuda:0'), covar=tensor([0.0537, 0.0938, 0.1853, 0.0776, 0.4115, 0.0615, 0.0800, 0.0851], + device='cuda:0'), in_proj_covar=tensor([0.0458, 0.0380, 0.0453, 0.0322, 0.0387, 0.0392, 0.0380, 0.0416], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:00:29,958 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=98864.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:00:56,820 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=98887.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:00:59,973 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.355e+02 4.885e+02 6.090e+02 7.322e+02 1.846e+03, threshold=1.218e+03, percent-clipped=3.0 +2023-04-02 05:01:02,397 INFO [train.py:903] (0/4) Epoch 15, batch 3300, loss[loss=0.2915, simple_loss=0.3518, pruned_loss=0.1156, over 14001.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2975, pruned_loss=0.07272, over 3811689.19 frames. ], batch size: 138, lr: 5.53e-03, grad_scale: 8.0 +2023-04-02 05:01:08,217 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 05:01:20,963 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=98905.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:01:25,246 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98908.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:01:30,597 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98913.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:02:07,327 INFO [train.py:903] (0/4) Epoch 15, batch 3350, loss[loss=0.2065, simple_loss=0.2792, pruned_loss=0.06691, over 19766.00 frames. ], tot_loss[loss=0.2201, simple_loss=0.2965, pruned_loss=0.07183, over 3824186.99 frames. ], batch size: 48, lr: 5.53e-03, grad_scale: 8.0 +2023-04-02 05:02:26,973 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9346, 1.1571, 1.5434, 0.5561, 2.0796, 2.4186, 2.0622, 2.5642], + device='cuda:0'), covar=tensor([0.1551, 0.3710, 0.3137, 0.2619, 0.0571, 0.0265, 0.0372, 0.0320], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0306, 0.0333, 0.0254, 0.0225, 0.0169, 0.0208, 0.0224], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:02:38,129 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5606, 4.1684, 2.6208, 3.6571, 0.8265, 4.0639, 3.9229, 4.0381], + device='cuda:0'), covar=tensor([0.0658, 0.0969, 0.1990, 0.0828, 0.4283, 0.0692, 0.0838, 0.1040], + device='cuda:0'), in_proj_covar=tensor([0.0461, 0.0381, 0.0455, 0.0325, 0.0391, 0.0394, 0.0383, 0.0418], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:03:06,864 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=98989.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:03:07,573 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.074e+02 5.628e+02 7.317e+02 9.748e+02 2.071e+03, threshold=1.463e+03, percent-clipped=8.0 +2023-04-02 05:03:09,827 INFO [train.py:903] (0/4) Epoch 15, batch 3400, loss[loss=0.2216, simple_loss=0.3022, pruned_loss=0.07052, over 19687.00 frames. ], tot_loss[loss=0.2212, simple_loss=0.2975, pruned_loss=0.07248, over 3817800.84 frames. ], batch size: 59, lr: 5.53e-03, grad_scale: 8.0 +2023-04-02 05:03:11,396 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1749, 1.1192, 1.1392, 1.2433, 0.9922, 1.3473, 1.3114, 1.2218], + device='cuda:0'), covar=tensor([0.0883, 0.0988, 0.1061, 0.0740, 0.0893, 0.0771, 0.0837, 0.0749], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0222, 0.0222, 0.0241, 0.0228, 0.0208, 0.0189, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 05:03:44,041 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99020.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:03:48,621 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3396, 1.4151, 1.8979, 1.7088, 3.0602, 4.8025, 4.6954, 5.0532], + device='cuda:0'), covar=tensor([0.1524, 0.3477, 0.3000, 0.1899, 0.0492, 0.0137, 0.0143, 0.0135], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0305, 0.0331, 0.0253, 0.0225, 0.0168, 0.0207, 0.0223], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:04:10,820 INFO [train.py:903] (0/4) Epoch 15, batch 3450, loss[loss=0.2691, simple_loss=0.3227, pruned_loss=0.1078, over 19726.00 frames. ], tot_loss[loss=0.2219, simple_loss=0.2981, pruned_loss=0.07285, over 3824977.79 frames. ], batch size: 51, lr: 5.53e-03, grad_scale: 4.0 +2023-04-02 05:04:14,067 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 05:04:45,815 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99069.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:05:11,145 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.718e+02 4.946e+02 5.886e+02 7.201e+02 1.354e+03, threshold=1.177e+03, percent-clipped=0.0 +2023-04-02 05:05:12,326 INFO [train.py:903] (0/4) Epoch 15, batch 3500, loss[loss=0.237, simple_loss=0.3174, pruned_loss=0.07825, over 18118.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2988, pruned_loss=0.0736, over 3819511.55 frames. ], batch size: 83, lr: 5.53e-03, grad_scale: 4.0 +2023-04-02 05:06:15,628 INFO [train.py:903] (0/4) Epoch 15, batch 3550, loss[loss=0.2287, simple_loss=0.3075, pruned_loss=0.07498, over 19312.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2993, pruned_loss=0.07345, over 3817595.45 frames. ], batch size: 66, lr: 5.53e-03, grad_scale: 4.0 +2023-04-02 05:06:18,511 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99143.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:06:48,295 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99168.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:07:18,044 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.289e+02 4.734e+02 6.110e+02 7.809e+02 1.736e+03, threshold=1.222e+03, percent-clipped=8.0 +2023-04-02 05:07:19,078 INFO [train.py:903] (0/4) Epoch 15, batch 3600, loss[loss=0.2693, simple_loss=0.3321, pruned_loss=0.1032, over 19504.00 frames. ], tot_loss[loss=0.2225, simple_loss=0.2988, pruned_loss=0.07313, over 3819143.06 frames. ], batch size: 64, lr: 5.52e-03, grad_scale: 8.0 +2023-04-02 05:07:57,243 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.95 vs. limit=5.0 +2023-04-02 05:08:20,326 INFO [train.py:903] (0/4) Epoch 15, batch 3650, loss[loss=0.1907, simple_loss=0.2662, pruned_loss=0.05755, over 19578.00 frames. ], tot_loss[loss=0.2236, simple_loss=0.2997, pruned_loss=0.07375, over 3821401.08 frames. ], batch size: 52, lr: 5.52e-03, grad_scale: 8.0 +2023-04-02 05:08:26,417 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99247.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:08:32,039 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99252.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:08:37,707 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99257.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:09:03,827 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99276.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:09:15,205 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99286.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:09:20,750 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.641e+02 5.562e+02 6.527e+02 8.007e+02 1.277e+03, threshold=1.305e+03, percent-clipped=3.0 +2023-04-02 05:09:21,900 INFO [train.py:903] (0/4) Epoch 15, batch 3700, loss[loss=0.1944, simple_loss=0.2855, pruned_loss=0.05167, over 19669.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2991, pruned_loss=0.07349, over 3822328.26 frames. ], batch size: 55, lr: 5.52e-03, grad_scale: 8.0 +2023-04-02 05:09:32,862 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99301.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:10:13,725 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99333.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:10:14,277 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-04-02 05:10:24,819 INFO [train.py:903] (0/4) Epoch 15, batch 3750, loss[loss=0.2223, simple_loss=0.2961, pruned_loss=0.07423, over 19610.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2991, pruned_loss=0.0735, over 3814763.40 frames. ], batch size: 50, lr: 5.52e-03, grad_scale: 8.0 +2023-04-02 05:10:25,119 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99342.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:10:56,172 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99367.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:11:01,969 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99372.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:11:26,382 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.684e+02 5.640e+02 6.861e+02 8.808e+02 1.909e+03, threshold=1.372e+03, percent-clipped=3.0 +2023-04-02 05:11:28,553 INFO [train.py:903] (0/4) Epoch 15, batch 3800, loss[loss=0.1963, simple_loss=0.2717, pruned_loss=0.06046, over 19330.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.2999, pruned_loss=0.07417, over 3802730.73 frames. ], batch size: 44, lr: 5.52e-03, grad_scale: 8.0 +2023-04-02 05:11:53,186 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99413.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:11:58,878 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 05:12:26,312 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99438.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:12:30,515 INFO [train.py:903] (0/4) Epoch 15, batch 3850, loss[loss=0.2198, simple_loss=0.2879, pruned_loss=0.0759, over 19475.00 frames. ], tot_loss[loss=0.2241, simple_loss=0.3, pruned_loss=0.07408, over 3798954.86 frames. ], batch size: 49, lr: 5.52e-03, grad_scale: 4.0 +2023-04-02 05:12:37,721 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99448.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:13:25,880 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99486.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:13:32,482 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.342e+02 4.923e+02 6.054e+02 7.410e+02 1.518e+03, threshold=1.211e+03, percent-clipped=1.0 +2023-04-02 05:13:32,500 INFO [train.py:903] (0/4) Epoch 15, batch 3900, loss[loss=0.2059, simple_loss=0.2898, pruned_loss=0.06101, over 19661.00 frames. ], tot_loss[loss=0.2225, simple_loss=0.2987, pruned_loss=0.07316, over 3805130.34 frames. ], batch size: 55, lr: 5.52e-03, grad_scale: 4.0 +2023-04-02 05:14:18,211 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99528.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:14:33,486 INFO [train.py:903] (0/4) Epoch 15, batch 3950, loss[loss=0.1888, simple_loss=0.2621, pruned_loss=0.05775, over 19748.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2988, pruned_loss=0.07326, over 3812112.29 frames. ], batch size: 47, lr: 5.52e-03, grad_scale: 4.0 +2023-04-02 05:14:41,181 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 05:15:00,513 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.86 vs. limit=5.0 +2023-04-02 05:15:02,362 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2843, 1.3869, 1.7069, 1.4885, 2.3706, 1.9839, 2.4512, 1.0961], + device='cuda:0'), covar=tensor([0.2293, 0.3909, 0.2233, 0.1814, 0.1449, 0.2071, 0.1405, 0.3846], + device='cuda:0'), in_proj_covar=tensor([0.0506, 0.0601, 0.0649, 0.0453, 0.0606, 0.0506, 0.0648, 0.0512], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:15:24,187 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8897, 2.6125, 2.5310, 3.0460, 2.7920, 2.5316, 2.2477, 2.8837], + device='cuda:0'), covar=tensor([0.0813, 0.1453, 0.1200, 0.0929, 0.1190, 0.0439, 0.1198, 0.0544], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0355, 0.0298, 0.0245, 0.0299, 0.0249, 0.0294, 0.0249], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:15:28,356 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99586.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:15:35,970 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99591.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:15:36,879 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.680e+02 5.329e+02 6.331e+02 8.679e+02 1.427e+03, threshold=1.266e+03, percent-clipped=6.0 +2023-04-02 05:15:36,897 INFO [train.py:903] (0/4) Epoch 15, batch 4000, loss[loss=0.2293, simple_loss=0.2923, pruned_loss=0.08311, over 19764.00 frames. ], tot_loss[loss=0.2235, simple_loss=0.2994, pruned_loss=0.07384, over 3817331.82 frames. ], batch size: 47, lr: 5.51e-03, grad_scale: 8.0 +2023-04-02 05:15:54,646 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99607.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:16:14,189 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99623.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:16:20,937 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99628.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:16:23,035 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 05:16:23,148 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99630.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:16:36,049 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8749, 4.1750, 4.5177, 4.5112, 1.9641, 4.2137, 3.7901, 4.1840], + device='cuda:0'), covar=tensor([0.1418, 0.1395, 0.0502, 0.0562, 0.5123, 0.0774, 0.0601, 0.1029], + device='cuda:0'), in_proj_covar=tensor([0.0723, 0.0658, 0.0857, 0.0735, 0.0769, 0.0607, 0.0518, 0.0793], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 05:16:38,161 INFO [train.py:903] (0/4) Epoch 15, batch 4050, loss[loss=0.2305, simple_loss=0.3092, pruned_loss=0.07595, over 19729.00 frames. ], tot_loss[loss=0.2227, simple_loss=0.2991, pruned_loss=0.07314, over 3810254.69 frames. ], batch size: 63, lr: 5.51e-03, grad_scale: 4.0 +2023-04-02 05:16:45,318 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99648.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:16:50,833 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99653.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:17:32,701 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99686.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:17:39,407 INFO [train.py:903] (0/4) Epoch 15, batch 4100, loss[loss=0.2239, simple_loss=0.2868, pruned_loss=0.08047, over 16848.00 frames. ], tot_loss[loss=0.222, simple_loss=0.2983, pruned_loss=0.07285, over 3811232.29 frames. ], batch size: 37, lr: 5.51e-03, grad_scale: 4.0 +2023-04-02 05:17:40,554 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.802e+02 5.566e+02 7.429e+02 9.161e+02 2.166e+03, threshold=1.486e+03, percent-clipped=8.0 +2023-04-02 05:17:46,974 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=99698.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:17:54,963 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99704.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:17:58,060 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99706.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:18:14,937 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 05:18:27,022 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99729.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:18:43,272 INFO [train.py:903] (0/4) Epoch 15, batch 4150, loss[loss=0.2291, simple_loss=0.3126, pruned_loss=0.07285, over 18513.00 frames. ], tot_loss[loss=0.222, simple_loss=0.2985, pruned_loss=0.0727, over 3799587.75 frames. ], batch size: 84, lr: 5.51e-03, grad_scale: 4.0 +2023-04-02 05:18:47,045 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99745.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:19:31,728 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99782.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:19:35,422 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99784.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:19:45,183 INFO [train.py:903] (0/4) Epoch 15, batch 4200, loss[loss=0.2981, simple_loss=0.3445, pruned_loss=0.1259, over 13192.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.298, pruned_loss=0.07249, over 3801532.56 frames. ], batch size: 137, lr: 5.51e-03, grad_scale: 4.0 +2023-04-02 05:19:47,442 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.476e+02 5.549e+02 6.772e+02 8.720e+02 1.402e+03, threshold=1.354e+03, percent-clipped=0.0 +2023-04-02 05:19:50,948 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 05:19:57,176 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99801.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:20:06,466 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99809.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:20:29,572 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3212, 1.3162, 1.5558, 1.4859, 2.2408, 2.0009, 2.2106, 0.7788], + device='cuda:0'), covar=tensor([0.2311, 0.4123, 0.2417, 0.1837, 0.1444, 0.2036, 0.1432, 0.4185], + device='cuda:0'), in_proj_covar=tensor([0.0504, 0.0596, 0.0646, 0.0452, 0.0603, 0.0504, 0.0645, 0.0511], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:20:31,690 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99830.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:20:47,368 INFO [train.py:903] (0/4) Epoch 15, batch 4250, loss[loss=0.2429, simple_loss=0.3191, pruned_loss=0.08334, over 19654.00 frames. ], tot_loss[loss=0.2216, simple_loss=0.2984, pruned_loss=0.07243, over 3815734.89 frames. ], batch size: 55, lr: 5.51e-03, grad_scale: 4.0 +2023-04-02 05:21:03,978 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 05:21:15,094 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 05:21:30,020 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8682, 1.9092, 2.1581, 2.5679, 1.7515, 2.4046, 2.2262, 1.9780], + device='cuda:0'), covar=tensor([0.3951, 0.3515, 0.1763, 0.1944, 0.3741, 0.1756, 0.4540, 0.3124], + device='cuda:0'), in_proj_covar=tensor([0.0831, 0.0876, 0.0670, 0.0903, 0.0817, 0.0751, 0.0807, 0.0739], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 05:21:37,938 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5086, 1.4342, 1.4088, 1.8231, 1.5631, 1.6720, 1.7862, 1.6344], + device='cuda:0'), covar=tensor([0.0834, 0.0920, 0.1009, 0.0640, 0.0722, 0.0767, 0.0820, 0.0661], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0220, 0.0219, 0.0239, 0.0226, 0.0207, 0.0188, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 05:21:38,410 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-02 05:21:47,822 INFO [train.py:903] (0/4) Epoch 15, batch 4300, loss[loss=0.2263, simple_loss=0.3117, pruned_loss=0.07047, over 19516.00 frames. ], tot_loss[loss=0.2213, simple_loss=0.2978, pruned_loss=0.07235, over 3811833.38 frames. ], batch size: 54, lr: 5.51e-03, grad_scale: 4.0 +2023-04-02 05:21:48,962 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.369e+02 5.517e+02 6.494e+02 8.260e+02 1.741e+03, threshold=1.299e+03, percent-clipped=4.0 +2023-04-02 05:21:53,839 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99897.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:22:35,378 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99930.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:22:41,691 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 05:22:48,543 INFO [train.py:903] (0/4) Epoch 15, batch 4350, loss[loss=0.18, simple_loss=0.2545, pruned_loss=0.05271, over 18729.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2981, pruned_loss=0.07243, over 3824035.37 frames. ], batch size: 41, lr: 5.50e-03, grad_scale: 4.0 +2023-04-02 05:22:52,427 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=99945.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:23:00,233 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=99951.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:23:12,148 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3587, 3.1047, 2.0658, 2.8008, 0.9478, 2.9740, 2.8671, 2.9798], + device='cuda:0'), covar=tensor([0.0975, 0.1159, 0.2083, 0.0917, 0.3445, 0.0999, 0.1033, 0.1203], + device='cuda:0'), in_proj_covar=tensor([0.0467, 0.0386, 0.0462, 0.0328, 0.0396, 0.0396, 0.0389, 0.0425], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:23:16,023 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=99962.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:23:45,416 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=99987.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:23:51,932 INFO [train.py:903] (0/4) Epoch 15, batch 4400, loss[loss=0.1798, simple_loss=0.2586, pruned_loss=0.05046, over 19741.00 frames. ], tot_loss[loss=0.2202, simple_loss=0.2968, pruned_loss=0.07179, over 3834907.48 frames. ], batch size: 47, lr: 5.50e-03, grad_scale: 8.0 +2023-04-02 05:23:53,158 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.223e+02 4.844e+02 5.787e+02 7.470e+02 1.170e+03, threshold=1.157e+03, percent-clipped=0.0 +2023-04-02 05:24:04,174 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-100000.pt +2023-04-02 05:24:06,755 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100001.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:24:23,740 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 05:24:29,805 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6896, 4.0884, 4.3714, 4.3929, 1.9218, 4.0688, 3.6458, 4.0710], + device='cuda:0'), covar=tensor([0.1472, 0.1211, 0.0543, 0.0580, 0.4995, 0.0793, 0.0613, 0.1032], + device='cuda:0'), in_proj_covar=tensor([0.0732, 0.0665, 0.0870, 0.0744, 0.0775, 0.0613, 0.0524, 0.0802], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 05:24:31,862 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 05:24:34,572 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100025.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 05:24:35,747 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100026.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:24:57,064 INFO [train.py:903] (0/4) Epoch 15, batch 4450, loss[loss=0.1984, simple_loss=0.2752, pruned_loss=0.06078, over 19478.00 frames. ], tot_loss[loss=0.2189, simple_loss=0.2954, pruned_loss=0.07121, over 3831512.85 frames. ], batch size: 49, lr: 5.50e-03, grad_scale: 8.0 +2023-04-02 05:24:57,236 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100042.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:25:00,894 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100045.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:25:14,927 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100057.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:25:25,306 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100066.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:25:47,664 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100082.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:25:59,167 INFO [train.py:903] (0/4) Epoch 15, batch 4500, loss[loss=0.2471, simple_loss=0.3232, pruned_loss=0.08555, over 19663.00 frames. ], tot_loss[loss=0.2201, simple_loss=0.2965, pruned_loss=0.0718, over 3820871.90 frames. ], batch size: 60, lr: 5.50e-03, grad_scale: 8.0 +2023-04-02 05:26:00,181 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.254e+02 5.166e+02 6.659e+02 8.670e+02 1.796e+03, threshold=1.332e+03, percent-clipped=6.0 +2023-04-02 05:26:59,885 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-04-02 05:27:01,361 INFO [train.py:903] (0/4) Epoch 15, batch 4550, loss[loss=0.239, simple_loss=0.3322, pruned_loss=0.07293, over 19321.00 frames. ], tot_loss[loss=0.2194, simple_loss=0.2963, pruned_loss=0.07122, over 3829051.96 frames. ], batch size: 66, lr: 5.50e-03, grad_scale: 8.0 +2023-04-02 05:27:10,395 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 05:27:16,310 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100153.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:27:21,847 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100157.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:27:38,124 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 05:27:47,888 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100178.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:28:04,927 INFO [train.py:903] (0/4) Epoch 15, batch 4600, loss[loss=0.2324, simple_loss=0.3115, pruned_loss=0.07666, over 19141.00 frames. ], tot_loss[loss=0.2195, simple_loss=0.2963, pruned_loss=0.0713, over 3833173.35 frames. ], batch size: 69, lr: 5.50e-03, grad_scale: 8.0 +2023-04-02 05:28:06,060 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.047e+02 4.707e+02 5.654e+02 7.541e+02 1.184e+03, threshold=1.131e+03, percent-clipped=0.0 +2023-04-02 05:28:19,023 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100201.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 05:28:26,342 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1030, 1.2052, 1.4761, 1.3742, 2.7058, 1.0967, 2.0404, 3.0018], + device='cuda:0'), covar=tensor([0.0548, 0.2750, 0.2796, 0.1794, 0.0751, 0.2348, 0.1170, 0.0332], + device='cuda:0'), in_proj_covar=tensor([0.0376, 0.0347, 0.0367, 0.0329, 0.0356, 0.0337, 0.0347, 0.0370], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:28:36,836 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4682, 1.4922, 1.8889, 1.7278, 3.2947, 2.7522, 3.5754, 1.7090], + device='cuda:0'), covar=tensor([0.2416, 0.4218, 0.2628, 0.1938, 0.1369, 0.1826, 0.1391, 0.3595], + device='cuda:0'), in_proj_covar=tensor([0.0504, 0.0596, 0.0648, 0.0452, 0.0603, 0.0505, 0.0641, 0.0511], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:28:47,106 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100226.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:29:08,987 INFO [train.py:903] (0/4) Epoch 15, batch 4650, loss[loss=0.2037, simple_loss=0.2829, pruned_loss=0.06222, over 19470.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2964, pruned_loss=0.07135, over 3830338.15 frames. ], batch size: 49, lr: 5.50e-03, grad_scale: 8.0 +2023-04-02 05:29:20,287 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-04-02 05:29:25,498 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 05:29:25,774 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100256.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:29:35,870 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 05:30:11,078 INFO [train.py:903] (0/4) Epoch 15, batch 4700, loss[loss=0.2207, simple_loss=0.2963, pruned_loss=0.07253, over 19777.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2975, pruned_loss=0.07174, over 3818492.57 frames. ], batch size: 54, lr: 5.49e-03, grad_scale: 8.0 +2023-04-02 05:30:12,228 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.373e+02 5.601e+02 6.328e+02 8.331e+02 3.311e+03, threshold=1.266e+03, percent-clipped=13.0 +2023-04-02 05:30:22,181 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100301.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:30:33,248 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 05:30:50,889 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100322.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:30:55,446 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100326.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:31:13,905 INFO [train.py:903] (0/4) Epoch 15, batch 4750, loss[loss=0.2246, simple_loss=0.3024, pruned_loss=0.07346, over 19617.00 frames. ], tot_loss[loss=0.2204, simple_loss=0.2978, pruned_loss=0.07156, over 3824760.48 frames. ], batch size: 61, lr: 5.49e-03, grad_scale: 8.0 +2023-04-02 05:31:21,199 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100347.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:31:49,672 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100369.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:32:16,842 INFO [train.py:903] (0/4) Epoch 15, batch 4800, loss[loss=0.243, simple_loss=0.3228, pruned_loss=0.08166, over 19539.00 frames. ], tot_loss[loss=0.2199, simple_loss=0.2972, pruned_loss=0.07133, over 3835207.81 frames. ], batch size: 56, lr: 5.49e-03, grad_scale: 8.0 +2023-04-02 05:32:18,029 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.926e+02 4.713e+02 6.387e+02 8.455e+02 2.006e+03, threshold=1.277e+03, percent-clipped=3.0 +2023-04-02 05:32:43,225 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2884, 2.0155, 2.1042, 2.8580, 2.1647, 2.6773, 2.7621, 2.5030], + device='cuda:0'), covar=tensor([0.0699, 0.0850, 0.0907, 0.0784, 0.0784, 0.0675, 0.0726, 0.0556], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0221, 0.0220, 0.0239, 0.0226, 0.0208, 0.0187, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 05:32:44,441 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100413.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:32:52,646 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100420.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:33:17,175 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100438.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:33:21,571 INFO [train.py:903] (0/4) Epoch 15, batch 4850, loss[loss=0.2344, simple_loss=0.3114, pruned_loss=0.07875, over 19538.00 frames. ], tot_loss[loss=0.2192, simple_loss=0.2961, pruned_loss=0.07115, over 3826257.01 frames. ], batch size: 56, lr: 5.49e-03, grad_scale: 8.0 +2023-04-02 05:33:48,721 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 05:34:10,983 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 05:34:14,843 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100484.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:34:16,742 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 05:34:17,778 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 05:34:24,795 INFO [train.py:903] (0/4) Epoch 15, batch 4900, loss[loss=0.2449, simple_loss=0.3152, pruned_loss=0.08731, over 12910.00 frames. ], tot_loss[loss=0.2206, simple_loss=0.2972, pruned_loss=0.07205, over 3802596.07 frames. ], batch size: 135, lr: 5.49e-03, grad_scale: 8.0 +2023-04-02 05:34:25,929 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.390e+02 5.042e+02 5.846e+02 7.925e+02 1.600e+03, threshold=1.169e+03, percent-clipped=3.0 +2023-04-02 05:34:25,996 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 05:34:46,442 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 05:35:27,066 INFO [train.py:903] (0/4) Epoch 15, batch 4950, loss[loss=0.2018, simple_loss=0.279, pruned_loss=0.06225, over 19482.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2973, pruned_loss=0.07167, over 3790818.76 frames. ], batch size: 49, lr: 5.49e-03, grad_scale: 8.0 +2023-04-02 05:35:45,778 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 05:36:04,528 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-02 05:36:05,452 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4897, 2.2791, 1.6757, 1.5945, 2.1440, 1.3852, 1.3189, 1.8958], + device='cuda:0'), covar=tensor([0.1126, 0.0775, 0.0973, 0.0742, 0.0469, 0.1146, 0.0767, 0.0434], + device='cuda:0'), in_proj_covar=tensor([0.0292, 0.0305, 0.0323, 0.0250, 0.0237, 0.0325, 0.0292, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:36:09,666 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 05:36:29,672 INFO [train.py:903] (0/4) Epoch 15, batch 5000, loss[loss=0.214, simple_loss=0.2794, pruned_loss=0.07427, over 19800.00 frames. ], tot_loss[loss=0.221, simple_loss=0.2981, pruned_loss=0.07194, over 3810538.23 frames. ], batch size: 47, lr: 5.49e-03, grad_scale: 4.0 +2023-04-02 05:36:31,853 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.571e+02 5.199e+02 6.623e+02 8.418e+02 1.165e+03, threshold=1.325e+03, percent-clipped=0.0 +2023-04-02 05:36:40,682 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 05:36:40,823 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100600.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:36:52,300 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 05:36:54,960 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7175, 1.7379, 1.5755, 1.3530, 1.3052, 1.3862, 0.2301, 0.6193], + device='cuda:0'), covar=tensor([0.0550, 0.0521, 0.0345, 0.0544, 0.1075, 0.0611, 0.0994, 0.0885], + device='cuda:0'), in_proj_covar=tensor([0.0348, 0.0342, 0.0341, 0.0370, 0.0444, 0.0371, 0.0321, 0.0327], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:37:25,100 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8407, 1.1826, 1.5468, 0.5400, 2.0227, 2.4353, 2.0896, 2.5723], + device='cuda:0'), covar=tensor([0.1618, 0.3621, 0.3159, 0.2612, 0.0574, 0.0257, 0.0368, 0.0337], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0304, 0.0334, 0.0252, 0.0226, 0.0170, 0.0208, 0.0224], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:37:33,612 INFO [train.py:903] (0/4) Epoch 15, batch 5050, loss[loss=0.2216, simple_loss=0.3062, pruned_loss=0.06848, over 19107.00 frames. ], tot_loss[loss=0.2213, simple_loss=0.2981, pruned_loss=0.07226, over 3805593.75 frames. ], batch size: 69, lr: 5.49e-03, grad_scale: 4.0 +2023-04-02 05:38:09,678 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 05:38:37,496 INFO [train.py:903] (0/4) Epoch 15, batch 5100, loss[loss=0.2227, simple_loss=0.3061, pruned_loss=0.06967, over 19493.00 frames. ], tot_loss[loss=0.2213, simple_loss=0.2981, pruned_loss=0.07231, over 3808980.87 frames. ], batch size: 64, lr: 5.48e-03, grad_scale: 4.0 +2023-04-02 05:38:39,893 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.188e+02 5.134e+02 6.124e+02 7.830e+02 1.941e+03, threshold=1.225e+03, percent-clipped=4.0 +2023-04-02 05:38:49,382 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 05:38:51,816 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 05:38:57,362 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 05:39:05,812 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100715.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:39:37,087 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100740.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:39:38,888 INFO [train.py:903] (0/4) Epoch 15, batch 5150, loss[loss=0.2363, simple_loss=0.3131, pruned_loss=0.07978, over 19730.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2962, pruned_loss=0.07166, over 3810891.82 frames. ], batch size: 51, lr: 5.48e-03, grad_scale: 4.0 +2023-04-02 05:39:47,030 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=100748.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:39:51,337 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 05:40:07,666 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=100764.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:40:09,039 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100765.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 05:40:20,529 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5461, 1.4723, 1.4506, 1.8278, 1.4258, 1.8217, 1.7492, 1.6609], + device='cuda:0'), covar=tensor([0.0799, 0.0909, 0.0991, 0.0680, 0.0773, 0.0699, 0.0790, 0.0657], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0221, 0.0221, 0.0241, 0.0227, 0.0209, 0.0189, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 05:40:28,224 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 05:40:38,849 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1275, 3.5466, 2.0315, 2.2791, 3.1251, 1.8460, 1.4962, 2.1125], + device='cuda:0'), covar=tensor([0.1387, 0.0519, 0.1043, 0.0745, 0.0476, 0.1173, 0.0956, 0.0696], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0305, 0.0325, 0.0251, 0.0237, 0.0326, 0.0294, 0.0267], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:40:41,950 INFO [train.py:903] (0/4) Epoch 15, batch 5200, loss[loss=0.2746, simple_loss=0.3283, pruned_loss=0.1105, over 13488.00 frames. ], tot_loss[loss=0.2209, simple_loss=0.2975, pruned_loss=0.07216, over 3809290.56 frames. ], batch size: 136, lr: 5.48e-03, grad_scale: 8.0 +2023-04-02 05:40:44,524 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.350e+02 5.325e+02 6.515e+02 8.501e+02 1.618e+03, threshold=1.303e+03, percent-clipped=5.0 +2023-04-02 05:40:58,648 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 05:41:43,060 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 05:41:47,607 INFO [train.py:903] (0/4) Epoch 15, batch 5250, loss[loss=0.2412, simple_loss=0.315, pruned_loss=0.08373, over 18391.00 frames. ], tot_loss[loss=0.2209, simple_loss=0.2976, pruned_loss=0.07209, over 3816791.00 frames. ], batch size: 84, lr: 5.48e-03, grad_scale: 8.0 +2023-04-02 05:42:08,359 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4524, 2.2604, 2.0428, 1.8540, 1.7620, 1.8747, 0.6810, 1.1971], + device='cuda:0'), covar=tensor([0.0419, 0.0431, 0.0353, 0.0610, 0.0889, 0.0683, 0.0993, 0.0816], + device='cuda:0'), in_proj_covar=tensor([0.0346, 0.0340, 0.0338, 0.0368, 0.0438, 0.0367, 0.0320, 0.0326], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:42:33,416 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=100879.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:42:50,409 INFO [train.py:903] (0/4) Epoch 15, batch 5300, loss[loss=0.227, simple_loss=0.3082, pruned_loss=0.07294, over 18783.00 frames. ], tot_loss[loss=0.2223, simple_loss=0.2988, pruned_loss=0.0729, over 3814399.77 frames. ], batch size: 74, lr: 5.48e-03, grad_scale: 8.0 +2023-04-02 05:42:52,727 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.288e+02 5.131e+02 6.120e+02 8.353e+02 1.768e+03, threshold=1.224e+03, percent-clipped=4.0 +2023-04-02 05:43:08,007 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 05:43:52,931 INFO [train.py:903] (0/4) Epoch 15, batch 5350, loss[loss=0.242, simple_loss=0.3161, pruned_loss=0.084, over 19515.00 frames. ], tot_loss[loss=0.2222, simple_loss=0.2986, pruned_loss=0.07292, over 3818752.63 frames. ], batch size: 54, lr: 5.48e-03, grad_scale: 8.0 +2023-04-02 05:44:29,363 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 05:44:30,867 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=100971.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:44:56,006 INFO [train.py:903] (0/4) Epoch 15, batch 5400, loss[loss=0.2189, simple_loss=0.2907, pruned_loss=0.07362, over 19461.00 frames. ], tot_loss[loss=0.2229, simple_loss=0.299, pruned_loss=0.07337, over 3802933.53 frames. ], batch size: 49, lr: 5.48e-03, grad_scale: 8.0 +2023-04-02 05:44:58,250 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.408e+02 5.355e+02 6.832e+02 8.412e+02 2.240e+03, threshold=1.366e+03, percent-clipped=7.0 +2023-04-02 05:45:00,855 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=100996.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:45:55,909 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2246, 2.0705, 1.8662, 1.6760, 1.5751, 1.6916, 0.4741, 1.1169], + device='cuda:0'), covar=tensor([0.0518, 0.0528, 0.0417, 0.0649, 0.1040, 0.0747, 0.1172, 0.0853], + device='cuda:0'), in_proj_covar=tensor([0.0346, 0.0342, 0.0338, 0.0367, 0.0441, 0.0369, 0.0321, 0.0329], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:46:00,097 INFO [train.py:903] (0/4) Epoch 15, batch 5450, loss[loss=0.1993, simple_loss=0.2872, pruned_loss=0.05564, over 19666.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2989, pruned_loss=0.07363, over 3803359.09 frames. ], batch size: 55, lr: 5.47e-03, grad_scale: 8.0 +2023-04-02 05:47:04,590 INFO [train.py:903] (0/4) Epoch 15, batch 5500, loss[loss=0.1848, simple_loss=0.2649, pruned_loss=0.0523, over 19770.00 frames. ], tot_loss[loss=0.2233, simple_loss=0.2994, pruned_loss=0.07355, over 3809415.48 frames. ], batch size: 48, lr: 5.47e-03, grad_scale: 8.0 +2023-04-02 05:47:04,778 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=101092.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:47:06,825 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.049e+02 5.715e+02 6.860e+02 8.623e+02 1.531e+03, threshold=1.372e+03, percent-clipped=1.0 +2023-04-02 05:47:27,323 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 05:47:45,322 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=101125.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:47:58,078 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=101135.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:48:05,680 INFO [train.py:903] (0/4) Epoch 15, batch 5550, loss[loss=0.2718, simple_loss=0.3449, pruned_loss=0.09939, over 19663.00 frames. ], tot_loss[loss=0.2232, simple_loss=0.2994, pruned_loss=0.07353, over 3818385.68 frames. ], batch size: 59, lr: 5.47e-03, grad_scale: 8.0 +2023-04-02 05:48:12,815 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 05:48:28,830 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101160.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:48:38,080 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.40 vs. limit=5.0 +2023-04-02 05:49:04,234 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 05:49:08,923 INFO [train.py:903] (0/4) Epoch 15, batch 5600, loss[loss=0.2555, simple_loss=0.3289, pruned_loss=0.09108, over 19589.00 frames. ], tot_loss[loss=0.2228, simple_loss=0.2991, pruned_loss=0.07323, over 3822518.71 frames. ], batch size: 61, lr: 5.47e-03, grad_scale: 8.0 +2023-04-02 05:49:11,017 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.323e+02 5.272e+02 6.555e+02 8.017e+02 1.573e+03, threshold=1.311e+03, percent-clipped=2.0 +2023-04-02 05:49:28,229 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4353, 1.2220, 1.4547, 1.4729, 2.9133, 1.0772, 2.1863, 3.4406], + device='cuda:0'), covar=tensor([0.0647, 0.3266, 0.3150, 0.2090, 0.0954, 0.2796, 0.1603, 0.0329], + device='cuda:0'), in_proj_covar=tensor([0.0382, 0.0350, 0.0368, 0.0334, 0.0359, 0.0339, 0.0352, 0.0371], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:49:28,275 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=101207.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:49:52,607 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9867, 1.0245, 1.3232, 1.2631, 2.3366, 0.9637, 2.0870, 2.7696], + device='cuda:0'), covar=tensor([0.0767, 0.3693, 0.3343, 0.2199, 0.1283, 0.2766, 0.1305, 0.0521], + device='cuda:0'), in_proj_covar=tensor([0.0381, 0.0350, 0.0368, 0.0333, 0.0358, 0.0339, 0.0351, 0.0371], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:50:11,404 INFO [train.py:903] (0/4) Epoch 15, batch 5650, loss[loss=0.2102, simple_loss=0.2859, pruned_loss=0.06724, over 19680.00 frames. ], tot_loss[loss=0.2224, simple_loss=0.2988, pruned_loss=0.07303, over 3822794.07 frames. ], batch size: 53, lr: 5.47e-03, grad_scale: 8.0 +2023-04-02 05:50:59,686 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 05:51:01,872 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8014, 1.5426, 1.9457, 1.6424, 4.2339, 1.1890, 2.3697, 4.6531], + device='cuda:0'), covar=tensor([0.0403, 0.2834, 0.2567, 0.1987, 0.0776, 0.2573, 0.1441, 0.0183], + device='cuda:0'), in_proj_covar=tensor([0.0381, 0.0352, 0.0369, 0.0334, 0.0360, 0.0340, 0.0352, 0.0372], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:51:14,090 INFO [train.py:903] (0/4) Epoch 15, batch 5700, loss[loss=0.2143, simple_loss=0.2924, pruned_loss=0.06813, over 19330.00 frames. ], tot_loss[loss=0.2232, simple_loss=0.2993, pruned_loss=0.07351, over 3821046.48 frames. ], batch size: 66, lr: 5.47e-03, grad_scale: 8.0 +2023-04-02 05:51:17,702 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.786e+02 5.651e+02 6.610e+02 8.419e+02 1.957e+03, threshold=1.322e+03, percent-clipped=7.0 +2023-04-02 05:52:17,923 INFO [train.py:903] (0/4) Epoch 15, batch 5750, loss[loss=0.2157, simple_loss=0.2953, pruned_loss=0.06802, over 19426.00 frames. ], tot_loss[loss=0.223, simple_loss=0.2994, pruned_loss=0.07326, over 3820282.89 frames. ], batch size: 70, lr: 5.47e-03, grad_scale: 8.0 +2023-04-02 05:52:20,199 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 05:52:28,370 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 05:52:33,016 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 05:53:21,260 INFO [train.py:903] (0/4) Epoch 15, batch 5800, loss[loss=0.2049, simple_loss=0.2776, pruned_loss=0.06609, over 19617.00 frames. ], tot_loss[loss=0.2228, simple_loss=0.2993, pruned_loss=0.07319, over 3819552.83 frames. ], batch size: 50, lr: 5.46e-03, grad_scale: 8.0 +2023-04-02 05:53:23,496 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.554e+02 5.132e+02 6.075e+02 7.663e+02 2.298e+03, threshold=1.215e+03, percent-clipped=3.0 +2023-04-02 05:54:09,484 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0401, 1.8604, 1.7593, 2.1725, 1.9615, 1.8264, 1.7881, 2.0523], + device='cuda:0'), covar=tensor([0.0930, 0.1497, 0.1371, 0.0980, 0.1234, 0.0510, 0.1227, 0.0662], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0351, 0.0295, 0.0244, 0.0297, 0.0243, 0.0291, 0.0245], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:54:24,297 INFO [train.py:903] (0/4) Epoch 15, batch 5850, loss[loss=0.2226, simple_loss=0.3007, pruned_loss=0.07225, over 19719.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2992, pruned_loss=0.07347, over 3819937.62 frames. ], batch size: 63, lr: 5.46e-03, grad_scale: 8.0 +2023-04-02 05:54:52,512 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=101463.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:54:58,854 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=101469.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:55:15,851 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-04-02 05:55:24,183 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101488.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 05:55:28,447 INFO [train.py:903] (0/4) Epoch 15, batch 5900, loss[loss=0.1896, simple_loss=0.2619, pruned_loss=0.05869, over 19767.00 frames. ], tot_loss[loss=0.2212, simple_loss=0.2974, pruned_loss=0.07254, over 3829454.46 frames. ], batch size: 48, lr: 5.46e-03, grad_scale: 8.0 +2023-04-02 05:55:30,756 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.093e+02 5.163e+02 6.557e+02 7.983e+02 1.612e+03, threshold=1.311e+03, percent-clipped=3.0 +2023-04-02 05:55:30,817 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 05:55:51,844 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 05:56:03,645 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0446, 3.6890, 2.4137, 3.3245, 0.8897, 3.5540, 3.4906, 3.5553], + device='cuda:0'), covar=tensor([0.0751, 0.1120, 0.2005, 0.0843, 0.3726, 0.0738, 0.0868, 0.1072], + device='cuda:0'), in_proj_covar=tensor([0.0462, 0.0382, 0.0458, 0.0327, 0.0390, 0.0393, 0.0385, 0.0420], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:56:30,791 INFO [train.py:903] (0/4) Epoch 15, batch 5950, loss[loss=0.2215, simple_loss=0.2918, pruned_loss=0.0756, over 19476.00 frames. ], tot_loss[loss=0.2222, simple_loss=0.298, pruned_loss=0.07323, over 3813939.71 frames. ], batch size: 49, lr: 5.46e-03, grad_scale: 8.0 +2023-04-02 05:56:31,089 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6413, 1.5503, 1.9454, 2.0115, 4.1324, 1.3336, 2.4332, 4.3677], + device='cuda:0'), covar=tensor([0.0406, 0.2843, 0.2480, 0.1703, 0.0749, 0.2466, 0.1493, 0.0233], + device='cuda:0'), in_proj_covar=tensor([0.0380, 0.0350, 0.0368, 0.0333, 0.0360, 0.0339, 0.0352, 0.0372], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:57:17,244 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9423, 1.1052, 1.4824, 0.5342, 2.0654, 2.4497, 2.1402, 2.6036], + device='cuda:0'), covar=tensor([0.1590, 0.3942, 0.3369, 0.2606, 0.0601, 0.0263, 0.0367, 0.0334], + device='cuda:0'), in_proj_covar=tensor([0.0259, 0.0305, 0.0334, 0.0252, 0.0225, 0.0169, 0.0208, 0.0223], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:57:24,037 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=101584.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 05:57:34,464 INFO [train.py:903] (0/4) Epoch 15, batch 6000, loss[loss=0.2403, simple_loss=0.319, pruned_loss=0.08083, over 19624.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2968, pruned_loss=0.0721, over 3828354.83 frames. ], batch size: 57, lr: 5.46e-03, grad_scale: 8.0 +2023-04-02 05:57:34,465 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 05:57:47,190 INFO [train.py:937] (0/4) Epoch 15, validation: loss=0.1729, simple_loss=0.2735, pruned_loss=0.0362, over 944034.00 frames. +2023-04-02 05:57:47,192 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 05:57:49,633 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.269e+02 5.208e+02 6.128e+02 8.316e+02 1.573e+03, threshold=1.226e+03, percent-clipped=3.0 +2023-04-02 05:57:54,415 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0854, 1.8344, 1.7109, 2.0836, 1.8252, 1.8449, 1.6635, 2.0652], + device='cuda:0'), covar=tensor([0.0898, 0.1346, 0.1367, 0.0963, 0.1245, 0.0482, 0.1302, 0.0655], + device='cuda:0'), in_proj_covar=tensor([0.0256, 0.0348, 0.0293, 0.0243, 0.0294, 0.0242, 0.0287, 0.0244], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 05:58:30,053 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4564, 1.6241, 2.0459, 1.8235, 3.1469, 2.7391, 3.5909, 1.6353], + device='cuda:0'), covar=tensor([0.2215, 0.3776, 0.2416, 0.1630, 0.1483, 0.1804, 0.1440, 0.3669], + device='cuda:0'), in_proj_covar=tensor([0.0505, 0.0604, 0.0650, 0.0454, 0.0605, 0.0509, 0.0648, 0.0513], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 05:58:49,961 INFO [train.py:903] (0/4) Epoch 15, batch 6050, loss[loss=0.2152, simple_loss=0.2986, pruned_loss=0.06585, over 19539.00 frames. ], tot_loss[loss=0.2219, simple_loss=0.298, pruned_loss=0.07289, over 3812663.73 frames. ], batch size: 56, lr: 5.46e-03, grad_scale: 8.0 +2023-04-02 05:59:52,013 INFO [train.py:903] (0/4) Epoch 15, batch 6100, loss[loss=0.2186, simple_loss=0.3073, pruned_loss=0.06496, over 19681.00 frames. ], tot_loss[loss=0.2226, simple_loss=0.2989, pruned_loss=0.07316, over 3815329.79 frames. ], batch size: 58, lr: 5.46e-03, grad_scale: 8.0 +2023-04-02 05:59:55,081 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.053e+02 5.047e+02 6.326e+02 7.867e+02 1.574e+03, threshold=1.265e+03, percent-clipped=9.0 +2023-04-02 06:00:05,572 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=101702.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:00:56,523 INFO [train.py:903] (0/4) Epoch 15, batch 6150, loss[loss=0.2318, simple_loss=0.3138, pruned_loss=0.07488, over 19698.00 frames. ], tot_loss[loss=0.2231, simple_loss=0.2993, pruned_loss=0.07347, over 3814913.68 frames. ], batch size: 59, lr: 5.46e-03, grad_scale: 8.0 +2023-04-02 06:01:23,863 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 06:01:59,359 INFO [train.py:903] (0/4) Epoch 15, batch 6200, loss[loss=0.2335, simple_loss=0.3259, pruned_loss=0.07058, over 19665.00 frames. ], tot_loss[loss=0.2225, simple_loss=0.2989, pruned_loss=0.07306, over 3819162.22 frames. ], batch size: 60, lr: 5.45e-03, grad_scale: 8.0 +2023-04-02 06:02:01,550 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.367e+02 4.959e+02 6.303e+02 7.991e+02 1.526e+03, threshold=1.261e+03, percent-clipped=1.0 +2023-04-02 06:02:07,378 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8877, 4.9789, 5.7054, 5.7013, 1.9908, 5.3766, 4.5598, 5.3455], + device='cuda:0'), covar=tensor([0.1441, 0.0883, 0.0525, 0.0487, 0.5475, 0.0656, 0.0566, 0.1025], + device='cuda:0'), in_proj_covar=tensor([0.0737, 0.0672, 0.0880, 0.0754, 0.0785, 0.0625, 0.0528, 0.0813], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 06:02:59,272 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=101840.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:03:02,066 INFO [train.py:903] (0/4) Epoch 15, batch 6250, loss[loss=0.2143, simple_loss=0.2851, pruned_loss=0.07179, over 19386.00 frames. ], tot_loss[loss=0.2213, simple_loss=0.2984, pruned_loss=0.07214, over 3828921.37 frames. ], batch size: 47, lr: 5.45e-03, grad_scale: 8.0 +2023-04-02 06:03:19,659 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-04-02 06:03:30,484 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 06:03:30,844 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=101865.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:03:44,783 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3432, 2.3589, 2.1321, 2.6854, 2.5700, 2.1505, 2.1113, 2.6689], + device='cuda:0'), covar=tensor([0.0932, 0.1556, 0.1322, 0.1011, 0.1208, 0.0510, 0.1194, 0.0595], + device='cuda:0'), in_proj_covar=tensor([0.0258, 0.0350, 0.0296, 0.0244, 0.0297, 0.0244, 0.0290, 0.0246], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:03:53,398 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.44 vs. limit=5.0 +2023-04-02 06:04:04,515 INFO [train.py:903] (0/4) Epoch 15, batch 6300, loss[loss=0.2013, simple_loss=0.2765, pruned_loss=0.06308, over 19732.00 frames. ], tot_loss[loss=0.2211, simple_loss=0.298, pruned_loss=0.07213, over 3817193.32 frames. ], batch size: 51, lr: 5.45e-03, grad_scale: 4.0 +2023-04-02 06:04:07,998 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.483e+02 5.666e+02 6.616e+02 7.934e+02 1.912e+03, threshold=1.323e+03, percent-clipped=2.0 +2023-04-02 06:05:01,681 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5574, 2.1865, 2.2470, 2.6625, 2.4731, 2.3424, 2.2263, 2.8977], + device='cuda:0'), covar=tensor([0.0882, 0.1687, 0.1285, 0.0999, 0.1317, 0.0488, 0.1157, 0.0518], + device='cuda:0'), in_proj_covar=tensor([0.0260, 0.0351, 0.0296, 0.0245, 0.0298, 0.0245, 0.0291, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:05:03,699 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=101938.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 06:05:08,249 INFO [train.py:903] (0/4) Epoch 15, batch 6350, loss[loss=0.1876, simple_loss=0.2533, pruned_loss=0.06097, over 19753.00 frames. ], tot_loss[loss=0.221, simple_loss=0.2978, pruned_loss=0.07208, over 3827005.83 frames. ], batch size: 46, lr: 5.45e-03, grad_scale: 2.0 +2023-04-02 06:06:11,825 INFO [train.py:903] (0/4) Epoch 15, batch 6400, loss[loss=0.2121, simple_loss=0.2943, pruned_loss=0.06499, over 19741.00 frames. ], tot_loss[loss=0.2201, simple_loss=0.2972, pruned_loss=0.07155, over 3834872.64 frames. ], batch size: 51, lr: 5.45e-03, grad_scale: 4.0 +2023-04-02 06:06:16,593 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.303e+02 4.874e+02 5.936e+02 7.490e+02 2.019e+03, threshold=1.187e+03, percent-clipped=4.0 +2023-04-02 06:06:22,637 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-102000.pt +2023-04-02 06:06:30,739 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=102005.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:07:10,213 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3563, 3.9114, 2.4766, 3.4231, 0.9466, 3.7638, 3.6656, 3.8010], + device='cuda:0'), covar=tensor([0.0677, 0.1106, 0.2157, 0.0955, 0.4059, 0.0875, 0.1058, 0.1298], + device='cuda:0'), in_proj_covar=tensor([0.0467, 0.0383, 0.0459, 0.0330, 0.0391, 0.0395, 0.0387, 0.0426], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:07:17,097 INFO [train.py:903] (0/4) Epoch 15, batch 6450, loss[loss=0.1953, simple_loss=0.2808, pruned_loss=0.0549, over 19762.00 frames. ], tot_loss[loss=0.2202, simple_loss=0.2974, pruned_loss=0.07151, over 3836380.34 frames. ], batch size: 54, lr: 5.45e-03, grad_scale: 4.0 +2023-04-02 06:07:23,142 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=102046.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:07:28,324 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5183, 3.1639, 2.6030, 2.4947, 2.6218, 2.7818, 0.9446, 2.3091], + device='cuda:0'), covar=tensor([0.0549, 0.0466, 0.0536, 0.0888, 0.0767, 0.0905, 0.1216, 0.0885], + device='cuda:0'), in_proj_covar=tensor([0.0349, 0.0340, 0.0338, 0.0370, 0.0441, 0.0370, 0.0322, 0.0331], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:08:03,626 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 06:08:20,708 INFO [train.py:903] (0/4) Epoch 15, batch 6500, loss[loss=0.1771, simple_loss=0.2574, pruned_loss=0.04835, over 19733.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2971, pruned_loss=0.07121, over 3837073.43 frames. ], batch size: 46, lr: 5.45e-03, grad_scale: 4.0 +2023-04-02 06:08:25,528 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.034e+02 4.860e+02 6.100e+02 7.866e+02 2.286e+03, threshold=1.220e+03, percent-clipped=9.0 +2023-04-02 06:08:26,634 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 06:09:23,526 INFO [train.py:903] (0/4) Epoch 15, batch 6550, loss[loss=0.1925, simple_loss=0.2761, pruned_loss=0.05445, over 19736.00 frames. ], tot_loss[loss=0.2192, simple_loss=0.2965, pruned_loss=0.07096, over 3820223.51 frames. ], batch size: 51, lr: 5.44e-03, grad_scale: 4.0 +2023-04-02 06:09:47,181 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=102161.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:10:26,473 INFO [train.py:903] (0/4) Epoch 15, batch 6600, loss[loss=0.2742, simple_loss=0.3355, pruned_loss=0.1064, over 18084.00 frames. ], tot_loss[loss=0.2204, simple_loss=0.2976, pruned_loss=0.07156, over 3829832.53 frames. ], batch size: 83, lr: 5.44e-03, grad_scale: 4.0 +2023-04-02 06:10:31,178 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.095e+02 4.937e+02 6.611e+02 8.175e+02 1.787e+03, threshold=1.322e+03, percent-clipped=6.0 +2023-04-02 06:11:01,868 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1435, 1.3557, 1.8506, 1.6078, 3.0211, 4.4130, 4.2703, 4.8370], + device='cuda:0'), covar=tensor([0.1714, 0.3732, 0.3236, 0.2110, 0.0568, 0.0195, 0.0175, 0.0139], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0308, 0.0336, 0.0256, 0.0227, 0.0171, 0.0210, 0.0225], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:11:29,859 INFO [train.py:903] (0/4) Epoch 15, batch 6650, loss[loss=0.1947, simple_loss=0.2678, pruned_loss=0.06084, over 19753.00 frames. ], tot_loss[loss=0.2202, simple_loss=0.2974, pruned_loss=0.07151, over 3828536.13 frames. ], batch size: 46, lr: 5.44e-03, grad_scale: 4.0 +2023-04-02 06:11:36,978 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2766, 1.4115, 1.8629, 1.5227, 2.7211, 2.2274, 2.7822, 1.2048], + device='cuda:0'), covar=tensor([0.2437, 0.3957, 0.2306, 0.1873, 0.1375, 0.1970, 0.1476, 0.3949], + device='cuda:0'), in_proj_covar=tensor([0.0505, 0.0602, 0.0654, 0.0455, 0.0606, 0.0512, 0.0651, 0.0514], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:12:20,476 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=102282.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 06:12:33,728 INFO [train.py:903] (0/4) Epoch 15, batch 6700, loss[loss=0.184, simple_loss=0.2569, pruned_loss=0.05557, over 19726.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2967, pruned_loss=0.0713, over 3828985.05 frames. ], batch size: 45, lr: 5.44e-03, grad_scale: 4.0 +2023-04-02 06:12:38,442 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.247e+02 5.015e+02 6.423e+02 7.841e+02 1.581e+03, threshold=1.285e+03, percent-clipped=1.0 +2023-04-02 06:13:32,251 INFO [train.py:903] (0/4) Epoch 15, batch 6750, loss[loss=0.2206, simple_loss=0.296, pruned_loss=0.07258, over 19692.00 frames. ], tot_loss[loss=0.2193, simple_loss=0.2961, pruned_loss=0.07129, over 3827126.61 frames. ], batch size: 53, lr: 5.44e-03, grad_scale: 4.0 +2023-04-02 06:13:40,325 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=102349.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:14:30,400 INFO [train.py:903] (0/4) Epoch 15, batch 6800, loss[loss=0.1801, simple_loss=0.2567, pruned_loss=0.05181, over 19771.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2949, pruned_loss=0.07054, over 3838821.27 frames. ], batch size: 47, lr: 5.44e-03, grad_scale: 8.0 +2023-04-02 06:14:35,327 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.506e+02 4.944e+02 6.022e+02 7.665e+02 3.022e+03, threshold=1.204e+03, percent-clipped=5.0 +2023-04-02 06:14:36,978 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=102397.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 06:14:57,420 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=102417.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:15:00,313 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-15.pt +2023-04-02 06:15:15,669 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 06:15:16,141 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 06:15:19,105 INFO [train.py:903] (0/4) Epoch 16, batch 0, loss[loss=0.2627, simple_loss=0.3192, pruned_loss=0.1031, over 19483.00 frames. ], tot_loss[loss=0.2627, simple_loss=0.3192, pruned_loss=0.1031, over 19483.00 frames. ], batch size: 64, lr: 5.26e-03, grad_scale: 8.0 +2023-04-02 06:15:19,106 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 06:15:29,716 INFO [train.py:937] (0/4) Epoch 16, validation: loss=0.1737, simple_loss=0.2745, pruned_loss=0.03646, over 944034.00 frames. +2023-04-02 06:15:29,717 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 06:15:45,601 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 06:15:58,413 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=102442.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:16:01,078 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.57 vs. limit=5.0 +2023-04-02 06:16:24,722 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=102464.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:16:33,055 INFO [train.py:903] (0/4) Epoch 16, batch 50, loss[loss=0.2097, simple_loss=0.2975, pruned_loss=0.06088, over 19615.00 frames. ], tot_loss[loss=0.2169, simple_loss=0.2941, pruned_loss=0.06989, over 854416.40 frames. ], batch size: 57, lr: 5.26e-03, grad_scale: 8.0 +2023-04-02 06:17:04,309 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.709e+02 4.865e+02 6.426e+02 8.395e+02 1.744e+03, threshold=1.285e+03, percent-clipped=5.0 +2023-04-02 06:17:08,787 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 06:17:33,677 INFO [train.py:903] (0/4) Epoch 16, batch 100, loss[loss=0.1814, simple_loss=0.2544, pruned_loss=0.05422, over 19730.00 frames. ], tot_loss[loss=0.2214, simple_loss=0.2976, pruned_loss=0.0726, over 1514132.03 frames. ], batch size: 45, lr: 5.26e-03, grad_scale: 8.0 +2023-04-02 06:17:47,718 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 06:18:34,758 INFO [train.py:903] (0/4) Epoch 16, batch 150, loss[loss=0.1903, simple_loss=0.2667, pruned_loss=0.05698, over 19322.00 frames. ], tot_loss[loss=0.2209, simple_loss=0.2971, pruned_loss=0.07234, over 2026322.43 frames. ], batch size: 44, lr: 5.26e-03, grad_scale: 8.0 +2023-04-02 06:18:46,806 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8539, 1.9092, 2.1862, 2.4815, 1.7670, 2.3681, 2.2932, 2.0291], + device='cuda:0'), covar=tensor([0.3884, 0.3487, 0.1699, 0.1831, 0.3558, 0.1708, 0.4286, 0.3007], + device='cuda:0'), in_proj_covar=tensor([0.0835, 0.0884, 0.0679, 0.0905, 0.0824, 0.0758, 0.0813, 0.0744], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 06:18:54,616 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4574, 1.6666, 2.1340, 1.7755, 3.0496, 2.4679, 3.2842, 1.5714], + device='cuda:0'), covar=tensor([0.2658, 0.4317, 0.2758, 0.2061, 0.1859, 0.2348, 0.1873, 0.4242], + device='cuda:0'), in_proj_covar=tensor([0.0507, 0.0602, 0.0655, 0.0457, 0.0606, 0.0514, 0.0654, 0.0517], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 06:19:07,432 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.312e+02 5.578e+02 6.638e+02 8.298e+02 1.665e+03, threshold=1.328e+03, percent-clipped=4.0 +2023-04-02 06:19:36,764 INFO [train.py:903] (0/4) Epoch 16, batch 200, loss[loss=0.1941, simple_loss=0.2697, pruned_loss=0.05928, over 19768.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2959, pruned_loss=0.07173, over 2409174.45 frames. ], batch size: 47, lr: 5.26e-03, grad_scale: 8.0 +2023-04-02 06:19:38,833 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 06:20:18,470 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=102653.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 06:20:37,454 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5262, 1.3004, 1.3895, 2.0099, 1.6505, 1.8503, 1.9831, 1.6855], + device='cuda:0'), covar=tensor([0.0903, 0.1131, 0.1076, 0.0872, 0.0885, 0.0819, 0.0829, 0.0743], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0222, 0.0223, 0.0243, 0.0226, 0.0210, 0.0190, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 06:20:38,296 INFO [train.py:903] (0/4) Epoch 16, batch 250, loss[loss=0.2358, simple_loss=0.3096, pruned_loss=0.08099, over 19746.00 frames. ], tot_loss[loss=0.2206, simple_loss=0.2969, pruned_loss=0.07217, over 2727076.29 frames. ], batch size: 63, lr: 5.26e-03, grad_scale: 8.0 +2023-04-02 06:20:50,334 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=102678.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 06:21:12,172 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.486e+02 5.293e+02 5.976e+02 7.347e+02 1.638e+03, threshold=1.195e+03, percent-clipped=3.0 +2023-04-02 06:21:43,451 INFO [train.py:903] (0/4) Epoch 16, batch 300, loss[loss=0.1965, simple_loss=0.2701, pruned_loss=0.06146, over 19315.00 frames. ], tot_loss[loss=0.2182, simple_loss=0.2949, pruned_loss=0.07077, over 2976671.27 frames. ], batch size: 44, lr: 5.26e-03, grad_scale: 8.0 +2023-04-02 06:21:43,868 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=102720.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:22:06,896 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9435, 1.3265, 1.0446, 0.9298, 1.1699, 0.9339, 0.9061, 1.2331], + device='cuda:0'), covar=tensor([0.0539, 0.0723, 0.1104, 0.0688, 0.0542, 0.1234, 0.0580, 0.0452], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0310, 0.0330, 0.0255, 0.0243, 0.0329, 0.0295, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:22:13,583 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=102745.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:22:44,073 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0167, 1.2670, 1.6420, 1.1987, 2.6295, 3.4364, 3.1720, 3.6013], + device='cuda:0'), covar=tensor([0.1828, 0.3772, 0.3398, 0.2349, 0.0567, 0.0184, 0.0230, 0.0230], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0305, 0.0334, 0.0255, 0.0228, 0.0170, 0.0210, 0.0226], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:22:44,946 INFO [train.py:903] (0/4) Epoch 16, batch 350, loss[loss=0.2373, simple_loss=0.3148, pruned_loss=0.07988, over 18744.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2951, pruned_loss=0.07075, over 3163043.82 frames. ], batch size: 74, lr: 5.25e-03, grad_scale: 8.0 +2023-04-02 06:22:50,853 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 06:23:16,122 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.122e+02 5.254e+02 6.186e+02 7.503e+02 2.205e+03, threshold=1.237e+03, percent-clipped=4.0 +2023-04-02 06:23:36,188 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2318, 1.3073, 1.2400, 1.0303, 1.0848, 1.1205, 0.0322, 0.3657], + device='cuda:0'), covar=tensor([0.0548, 0.0524, 0.0325, 0.0459, 0.1076, 0.0469, 0.1045, 0.0924], + device='cuda:0'), in_proj_covar=tensor([0.0350, 0.0342, 0.0338, 0.0371, 0.0442, 0.0371, 0.0323, 0.0333], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:23:47,561 INFO [train.py:903] (0/4) Epoch 16, batch 400, loss[loss=0.208, simple_loss=0.2901, pruned_loss=0.06298, over 19684.00 frames. ], tot_loss[loss=0.2187, simple_loss=0.296, pruned_loss=0.07072, over 3321927.23 frames. ], batch size: 53, lr: 5.25e-03, grad_scale: 8.0 +2023-04-02 06:24:49,390 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=102869.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:24:50,277 INFO [train.py:903] (0/4) Epoch 16, batch 450, loss[loss=0.2295, simple_loss=0.3095, pruned_loss=0.07476, over 19587.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2956, pruned_loss=0.07064, over 3446895.95 frames. ], batch size: 61, lr: 5.25e-03, grad_scale: 8.0 +2023-04-02 06:25:22,280 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.385e+02 4.837e+02 5.955e+02 8.003e+02 1.401e+03, threshold=1.191e+03, percent-clipped=4.0 +2023-04-02 06:25:24,661 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 06:25:25,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 06:25:33,021 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=102905.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:25:51,996 INFO [train.py:903] (0/4) Epoch 16, batch 500, loss[loss=0.2288, simple_loss=0.3134, pruned_loss=0.07211, over 19390.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2966, pruned_loss=0.07081, over 3512612.31 frames. ], batch size: 70, lr: 5.25e-03, grad_scale: 8.0 +2023-04-02 06:26:37,556 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9963, 2.0821, 2.2943, 2.7454, 1.9430, 2.6897, 2.4553, 2.1050], + device='cuda:0'), covar=tensor([0.4002, 0.3478, 0.1717, 0.2022, 0.3666, 0.1737, 0.4101, 0.3045], + device='cuda:0'), in_proj_covar=tensor([0.0837, 0.0886, 0.0679, 0.0906, 0.0824, 0.0758, 0.0814, 0.0743], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 06:26:54,172 INFO [train.py:903] (0/4) Epoch 16, batch 550, loss[loss=0.183, simple_loss=0.2564, pruned_loss=0.05479, over 19750.00 frames. ], tot_loss[loss=0.2187, simple_loss=0.2959, pruned_loss=0.0708, over 3590808.32 frames. ], batch size: 45, lr: 5.25e-03, grad_scale: 8.0 +2023-04-02 06:27:00,546 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.39 vs. limit=5.0 +2023-04-02 06:27:18,352 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2514, 1.5170, 1.9746, 1.7324, 3.0222, 4.5705, 4.4415, 4.9281], + device='cuda:0'), covar=tensor([0.1666, 0.3486, 0.3104, 0.2040, 0.0554, 0.0182, 0.0154, 0.0154], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0306, 0.0335, 0.0256, 0.0228, 0.0171, 0.0211, 0.0227], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:27:24,957 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.789e+02 5.227e+02 6.443e+02 7.702e+02 1.436e+03, threshold=1.289e+03, percent-clipped=3.0 +2023-04-02 06:27:54,425 INFO [train.py:903] (0/4) Epoch 16, batch 600, loss[loss=0.2246, simple_loss=0.306, pruned_loss=0.07163, over 19698.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2956, pruned_loss=0.07065, over 3629282.73 frames. ], batch size: 59, lr: 5.25e-03, grad_scale: 8.0 +2023-04-02 06:28:37,057 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 06:28:55,608 INFO [train.py:903] (0/4) Epoch 16, batch 650, loss[loss=0.1943, simple_loss=0.2724, pruned_loss=0.05813, over 19476.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2959, pruned_loss=0.07056, over 3680375.30 frames. ], batch size: 49, lr: 5.25e-03, grad_scale: 8.0 +2023-04-02 06:29:28,781 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.210e+02 4.715e+02 6.008e+02 7.942e+02 1.451e+03, threshold=1.202e+03, percent-clipped=1.0 +2023-04-02 06:29:53,874 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103116.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:29:58,298 INFO [train.py:903] (0/4) Epoch 16, batch 700, loss[loss=0.2525, simple_loss=0.3235, pruned_loss=0.09071, over 19681.00 frames. ], tot_loss[loss=0.2189, simple_loss=0.2962, pruned_loss=0.07081, over 3719234.68 frames. ], batch size: 59, lr: 5.25e-03, grad_scale: 8.0 +2023-04-02 06:30:34,540 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.51 vs. limit=2.0 +2023-04-02 06:30:35,103 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103150.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:31:00,294 INFO [train.py:903] (0/4) Epoch 16, batch 750, loss[loss=0.2352, simple_loss=0.3025, pruned_loss=0.08392, over 19621.00 frames. ], tot_loss[loss=0.2202, simple_loss=0.2972, pruned_loss=0.07158, over 3752559.90 frames. ], batch size: 50, lr: 5.24e-03, grad_scale: 8.0 +2023-04-02 06:31:33,687 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.142e+02 5.236e+02 6.262e+02 8.063e+02 1.865e+03, threshold=1.252e+03, percent-clipped=5.0 +2023-04-02 06:31:55,084 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=103213.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:32:03,181 INFO [train.py:903] (0/4) Epoch 16, batch 800, loss[loss=0.2728, simple_loss=0.3355, pruned_loss=0.105, over 13200.00 frames. ], tot_loss[loss=0.2209, simple_loss=0.298, pruned_loss=0.07195, over 3750638.39 frames. ], batch size: 135, lr: 5.24e-03, grad_scale: 8.0 +2023-04-02 06:32:18,116 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 06:32:26,451 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9642, 1.3493, 1.1203, 0.9637, 1.2669, 0.9444, 0.9859, 1.2694], + device='cuda:0'), covar=tensor([0.0661, 0.0740, 0.0785, 0.0667, 0.0429, 0.0962, 0.0525, 0.0367], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0309, 0.0330, 0.0254, 0.0242, 0.0330, 0.0292, 0.0267], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:32:38,486 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=103249.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:32:52,092 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7378, 4.1968, 4.4071, 4.3897, 1.6954, 4.1592, 3.5707, 4.1000], + device='cuda:0'), covar=tensor([0.1532, 0.0844, 0.0656, 0.0644, 0.5669, 0.0704, 0.0653, 0.1157], + device='cuda:0'), in_proj_covar=tensor([0.0730, 0.0666, 0.0869, 0.0746, 0.0772, 0.0615, 0.0519, 0.0802], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 06:33:04,745 INFO [train.py:903] (0/4) Epoch 16, batch 850, loss[loss=0.2337, simple_loss=0.3114, pruned_loss=0.07799, over 19336.00 frames. ], tot_loss[loss=0.2212, simple_loss=0.2981, pruned_loss=0.07215, over 3758463.14 frames. ], batch size: 66, lr: 5.24e-03, grad_scale: 8.0 +2023-04-02 06:33:38,413 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.826e+02 4.865e+02 6.263e+02 7.829e+02 1.710e+03, threshold=1.253e+03, percent-clipped=2.0 +2023-04-02 06:33:40,906 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9786, 5.0776, 5.8122, 5.7484, 2.0578, 5.4479, 4.6906, 5.4398], + device='cuda:0'), covar=tensor([0.1507, 0.0765, 0.0534, 0.0607, 0.5514, 0.0584, 0.0535, 0.1131], + device='cuda:0'), in_proj_covar=tensor([0.0730, 0.0665, 0.0870, 0.0747, 0.0770, 0.0616, 0.0520, 0.0802], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 06:33:44,516 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5808, 1.7096, 2.0882, 1.8358, 3.2877, 3.0448, 3.5530, 1.5763], + device='cuda:0'), covar=tensor([0.2254, 0.3968, 0.2439, 0.1782, 0.1472, 0.1640, 0.1622, 0.3847], + device='cuda:0'), in_proj_covar=tensor([0.0505, 0.0600, 0.0654, 0.0455, 0.0609, 0.0512, 0.0650, 0.0517], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:33:57,929 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 06:34:06,717 INFO [train.py:903] (0/4) Epoch 16, batch 900, loss[loss=0.1916, simple_loss=0.2657, pruned_loss=0.05878, over 19747.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.2979, pruned_loss=0.07182, over 3779991.32 frames. ], batch size: 45, lr: 5.24e-03, grad_scale: 8.0 +2023-04-02 06:34:17,372 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103328.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:35:01,509 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103364.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:35:06,308 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2661, 2.8560, 2.2226, 2.1036, 2.0927, 2.3844, 1.1343, 2.1006], + device='cuda:0'), covar=tensor([0.0544, 0.0457, 0.0609, 0.0890, 0.0876, 0.0844, 0.1073, 0.0838], + device='cuda:0'), in_proj_covar=tensor([0.0346, 0.0338, 0.0336, 0.0364, 0.0437, 0.0367, 0.0320, 0.0327], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:35:08,184 INFO [train.py:903] (0/4) Epoch 16, batch 950, loss[loss=0.2586, simple_loss=0.3291, pruned_loss=0.09404, over 19377.00 frames. ], tot_loss[loss=0.2194, simple_loss=0.2968, pruned_loss=0.07097, over 3799222.58 frames. ], batch size: 70, lr: 5.24e-03, grad_scale: 8.0 +2023-04-02 06:35:13,576 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 06:35:19,098 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0220, 3.6481, 2.3892, 3.2895, 1.0643, 3.4706, 3.4102, 3.4848], + device='cuda:0'), covar=tensor([0.0761, 0.1053, 0.2027, 0.0831, 0.3649, 0.0863, 0.0932, 0.1218], + device='cuda:0'), in_proj_covar=tensor([0.0470, 0.0384, 0.0460, 0.0329, 0.0391, 0.0398, 0.0392, 0.0427], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:35:27,482 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103384.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 06:35:40,826 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.146e+02 5.367e+02 6.234e+02 7.823e+02 2.113e+03, threshold=1.247e+03, percent-clipped=3.0 +2023-04-02 06:36:12,131 INFO [train.py:903] (0/4) Epoch 16, batch 1000, loss[loss=0.2306, simple_loss=0.3223, pruned_loss=0.06945, over 17452.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2978, pruned_loss=0.0716, over 3791396.42 frames. ], batch size: 101, lr: 5.24e-03, grad_scale: 8.0 +2023-04-02 06:37:03,100 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=103460.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:37:07,730 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 06:37:14,587 INFO [train.py:903] (0/4) Epoch 16, batch 1050, loss[loss=0.2153, simple_loss=0.2986, pruned_loss=0.06601, over 19688.00 frames. ], tot_loss[loss=0.2211, simple_loss=0.298, pruned_loss=0.07207, over 3804571.27 frames. ], batch size: 59, lr: 5.24e-03, grad_scale: 8.0 +2023-04-02 06:37:43,688 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=103494.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:37:45,894 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.718e+02 5.497e+02 6.450e+02 8.583e+02 2.663e+03, threshold=1.290e+03, percent-clipped=6.0 +2023-04-02 06:37:49,185 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 06:38:15,957 INFO [train.py:903] (0/4) Epoch 16, batch 1100, loss[loss=0.228, simple_loss=0.3245, pruned_loss=0.06573, over 19611.00 frames. ], tot_loss[loss=0.2206, simple_loss=0.2973, pruned_loss=0.07193, over 3815561.39 frames. ], batch size: 57, lr: 5.24e-03, grad_scale: 8.0 +2023-04-02 06:38:21,830 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3419, 3.8450, 3.9337, 3.9200, 1.5921, 3.7430, 3.2440, 3.6506], + device='cuda:0'), covar=tensor([0.1490, 0.0807, 0.0618, 0.0673, 0.5269, 0.0784, 0.0688, 0.1122], + device='cuda:0'), in_proj_covar=tensor([0.0741, 0.0676, 0.0879, 0.0754, 0.0780, 0.0623, 0.0524, 0.0810], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 06:39:18,053 INFO [train.py:903] (0/4) Epoch 16, batch 1150, loss[loss=0.2315, simple_loss=0.308, pruned_loss=0.07745, over 19536.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2963, pruned_loss=0.07158, over 3813346.68 frames. ], batch size: 56, lr: 5.23e-03, grad_scale: 8.0 +2023-04-02 06:39:24,703 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103575.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:39:37,371 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103584.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:39:50,620 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.809e+02 5.106e+02 6.190e+02 8.719e+02 1.567e+03, threshold=1.238e+03, percent-clipped=4.0 +2023-04-02 06:39:58,866 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5094, 1.5271, 1.8512, 1.6999, 2.6482, 2.2581, 2.7202, 1.2871], + device='cuda:0'), covar=tensor([0.2096, 0.3746, 0.2274, 0.1716, 0.1341, 0.1903, 0.1337, 0.3806], + device='cuda:0'), in_proj_covar=tensor([0.0501, 0.0599, 0.0653, 0.0453, 0.0604, 0.0509, 0.0647, 0.0514], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:40:06,771 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103609.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:40:06,810 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103609.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:40:21,156 INFO [train.py:903] (0/4) Epoch 16, batch 1200, loss[loss=0.2542, simple_loss=0.3181, pruned_loss=0.09509, over 13139.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2952, pruned_loss=0.0707, over 3807898.78 frames. ], batch size: 136, lr: 5.23e-03, grad_scale: 8.0 +2023-04-02 06:40:21,605 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103620.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:40:52,908 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103645.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:40:54,982 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 06:41:24,586 INFO [train.py:903] (0/4) Epoch 16, batch 1250, loss[loss=0.2584, simple_loss=0.3217, pruned_loss=0.09759, over 13636.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2972, pruned_loss=0.07169, over 3799620.21 frames. ], batch size: 138, lr: 5.23e-03, grad_scale: 8.0 +2023-04-02 06:41:56,591 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.406e+02 4.955e+02 6.144e+02 7.729e+02 1.641e+03, threshold=1.229e+03, percent-clipped=4.0 +2023-04-02 06:42:20,612 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9028, 1.1085, 1.5178, 0.6170, 2.0385, 2.3867, 2.1217, 2.5064], + device='cuda:0'), covar=tensor([0.1613, 0.3710, 0.3167, 0.2585, 0.0591, 0.0319, 0.0352, 0.0366], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0307, 0.0336, 0.0257, 0.0228, 0.0172, 0.0211, 0.0227], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 06:42:24,683 INFO [train.py:903] (0/4) Epoch 16, batch 1300, loss[loss=0.2746, simple_loss=0.3313, pruned_loss=0.109, over 19767.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2964, pruned_loss=0.07152, over 3798569.08 frames. ], batch size: 54, lr: 5.23e-03, grad_scale: 8.0 +2023-04-02 06:42:35,155 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=103728.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 06:42:53,029 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6935, 1.8485, 2.0974, 2.4257, 1.6823, 2.2058, 2.2202, 1.9870], + device='cuda:0'), covar=tensor([0.3985, 0.3416, 0.1639, 0.1837, 0.3645, 0.1746, 0.4110, 0.2939], + device='cuda:0'), in_proj_covar=tensor([0.0838, 0.0885, 0.0682, 0.0910, 0.0826, 0.0763, 0.0811, 0.0744], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 06:43:12,106 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3538, 3.9556, 2.5316, 3.4366, 0.7948, 3.8781, 3.7213, 3.8180], + device='cuda:0'), covar=tensor([0.0678, 0.1025, 0.2015, 0.0909, 0.4047, 0.0667, 0.0867, 0.1024], + device='cuda:0'), in_proj_covar=tensor([0.0473, 0.0386, 0.0462, 0.0334, 0.0395, 0.0400, 0.0395, 0.0431], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:43:12,195 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103758.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:43:26,061 INFO [train.py:903] (0/4) Epoch 16, batch 1350, loss[loss=0.2549, simple_loss=0.3243, pruned_loss=0.09274, over 19723.00 frames. ], tot_loss[loss=0.2201, simple_loss=0.2963, pruned_loss=0.07197, over 3807097.81 frames. ], batch size: 63, lr: 5.23e-03, grad_scale: 8.0 +2023-04-02 06:43:43,812 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103783.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:43:47,409 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.13 vs. limit=5.0 +2023-04-02 06:43:59,438 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.194e+02 5.214e+02 6.633e+02 9.392e+02 2.118e+03, threshold=1.327e+03, percent-clipped=8.0 +2023-04-02 06:44:30,117 INFO [train.py:903] (0/4) Epoch 16, batch 1400, loss[loss=0.2136, simple_loss=0.297, pruned_loss=0.06515, over 19296.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2963, pruned_loss=0.07145, over 3797372.48 frames. ], batch size: 66, lr: 5.23e-03, grad_scale: 8.0 +2023-04-02 06:44:43,931 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103831.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:44:57,958 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=103843.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 06:45:14,423 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103856.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:45:20,224 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5257, 1.3935, 1.1770, 1.5237, 1.2922, 1.3210, 1.2292, 1.4107], + device='cuda:0'), covar=tensor([0.1019, 0.1009, 0.1436, 0.0779, 0.1021, 0.0612, 0.1273, 0.0774], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0354, 0.0298, 0.0246, 0.0301, 0.0247, 0.0293, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:45:26,689 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=103865.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:45:32,154 INFO [train.py:903] (0/4) Epoch 16, batch 1450, loss[loss=0.1946, simple_loss=0.2721, pruned_loss=0.05852, over 19741.00 frames. ], tot_loss[loss=0.221, simple_loss=0.2974, pruned_loss=0.07233, over 3804238.76 frames. ], batch size: 51, lr: 5.23e-03, grad_scale: 8.0 +2023-04-02 06:45:32,180 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 06:45:56,430 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=103890.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:46:00,966 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=103894.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:46:03,877 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.072e+02 4.634e+02 5.962e+02 7.073e+02 1.523e+03, threshold=1.192e+03, percent-clipped=2.0 +2023-04-02 06:46:33,192 INFO [train.py:903] (0/4) Epoch 16, batch 1500, loss[loss=0.2362, simple_loss=0.2976, pruned_loss=0.08742, over 19073.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.296, pruned_loss=0.07154, over 3814316.09 frames. ], batch size: 42, lr: 5.23e-03, grad_scale: 16.0 +2023-04-02 06:47:06,840 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3164, 1.2558, 1.3024, 1.3012, 1.0782, 1.4028, 1.4662, 1.3258], + device='cuda:0'), covar=tensor([0.0867, 0.0959, 0.1026, 0.0705, 0.0788, 0.0803, 0.0759, 0.0744], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0222, 0.0224, 0.0244, 0.0226, 0.0210, 0.0192, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 06:47:35,228 INFO [train.py:903] (0/4) Epoch 16, batch 1550, loss[loss=0.229, simple_loss=0.3085, pruned_loss=0.07477, over 19350.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.296, pruned_loss=0.07164, over 3810132.07 frames. ], batch size: 70, lr: 5.22e-03, grad_scale: 8.0 +2023-04-02 06:47:58,591 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8176, 1.8382, 1.5034, 2.1182, 1.8391, 1.6176, 1.6701, 1.8991], + device='cuda:0'), covar=tensor([0.1089, 0.1480, 0.1507, 0.0908, 0.1232, 0.0605, 0.1333, 0.0725], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0355, 0.0299, 0.0247, 0.0302, 0.0247, 0.0295, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:48:09,262 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.399e+02 5.500e+02 6.727e+02 8.636e+02 1.580e+03, threshold=1.345e+03, percent-clipped=8.0 +2023-04-02 06:48:12,943 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-104000.pt +2023-04-02 06:48:24,281 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104008.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:48:32,603 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104015.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:48:39,342 INFO [train.py:903] (0/4) Epoch 16, batch 1600, loss[loss=0.2279, simple_loss=0.3085, pruned_loss=0.07362, over 19326.00 frames. ], tot_loss[loss=0.2187, simple_loss=0.2955, pruned_loss=0.07097, over 3808033.59 frames. ], batch size: 66, lr: 5.22e-03, grad_scale: 8.0 +2023-04-02 06:48:51,274 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8072, 3.4530, 1.9222, 2.0365, 3.0139, 1.6614, 1.1966, 2.1474], + device='cuda:0'), covar=tensor([0.1504, 0.0553, 0.1054, 0.0836, 0.0524, 0.1244, 0.1122, 0.0714], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0310, 0.0330, 0.0255, 0.0244, 0.0333, 0.0296, 0.0269], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:49:05,515 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 06:49:27,357 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8594, 1.9799, 2.1721, 2.4829, 1.8405, 2.4088, 2.2411, 2.0183], + device='cuda:0'), covar=tensor([0.3766, 0.3131, 0.1710, 0.2037, 0.3383, 0.1727, 0.4091, 0.2875], + device='cuda:0'), in_proj_covar=tensor([0.0840, 0.0886, 0.0683, 0.0912, 0.0825, 0.0762, 0.0815, 0.0746], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 06:49:38,461 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=104067.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:49:41,608 INFO [train.py:903] (0/4) Epoch 16, batch 1650, loss[loss=0.2134, simple_loss=0.2955, pruned_loss=0.06563, over 19566.00 frames. ], tot_loss[loss=0.2195, simple_loss=0.2963, pruned_loss=0.07137, over 3809538.58 frames. ], batch size: 52, lr: 5.22e-03, grad_scale: 8.0 +2023-04-02 06:50:14,808 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.025e+02 5.150e+02 6.179e+02 7.587e+02 1.568e+03, threshold=1.236e+03, percent-clipped=4.0 +2023-04-02 06:50:18,554 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104099.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 06:50:21,710 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104102.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:50:37,162 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9330, 1.5121, 1.7531, 1.6602, 4.4573, 1.1747, 2.4610, 4.7510], + device='cuda:0'), covar=tensor([0.0414, 0.2856, 0.2709, 0.1942, 0.0752, 0.2595, 0.1461, 0.0182], + device='cuda:0'), in_proj_covar=tensor([0.0382, 0.0351, 0.0370, 0.0335, 0.0358, 0.0341, 0.0352, 0.0376], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:50:43,583 INFO [train.py:903] (0/4) Epoch 16, batch 1700, loss[loss=0.2528, simple_loss=0.3302, pruned_loss=0.08773, over 17237.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2967, pruned_loss=0.07136, over 3821117.50 frames. ], batch size: 101, lr: 5.22e-03, grad_scale: 8.0 +2023-04-02 06:50:48,592 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=104124.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 06:50:51,804 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104127.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:51:25,772 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 06:51:29,472 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1126, 2.0118, 1.8218, 2.2359, 2.0577, 1.9046, 1.7118, 2.1428], + device='cuda:0'), covar=tensor([0.0940, 0.1470, 0.1244, 0.0978, 0.1174, 0.0477, 0.1294, 0.0629], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0356, 0.0300, 0.0249, 0.0303, 0.0249, 0.0295, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:51:44,972 INFO [train.py:903] (0/4) Epoch 16, batch 1750, loss[loss=0.2305, simple_loss=0.3027, pruned_loss=0.07915, over 19574.00 frames. ], tot_loss[loss=0.2209, simple_loss=0.2975, pruned_loss=0.07209, over 3821281.85 frames. ], batch size: 52, lr: 5.22e-03, grad_scale: 8.0 +2023-04-02 06:52:19,029 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.135e+02 4.884e+02 5.867e+02 6.930e+02 2.034e+03, threshold=1.173e+03, percent-clipped=2.0 +2023-04-02 06:52:44,500 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104217.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:52:48,726 INFO [train.py:903] (0/4) Epoch 16, batch 1800, loss[loss=0.2601, simple_loss=0.3298, pruned_loss=0.0952, over 19374.00 frames. ], tot_loss[loss=0.2206, simple_loss=0.2977, pruned_loss=0.07177, over 3828001.18 frames. ], batch size: 70, lr: 5.22e-03, grad_scale: 8.0 +2023-04-02 06:53:09,912 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104238.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:53:11,578 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-02 06:53:14,810 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104242.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:53:46,185 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 06:53:49,825 INFO [train.py:903] (0/4) Epoch 16, batch 1850, loss[loss=0.2254, simple_loss=0.314, pruned_loss=0.06841, over 19525.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2973, pruned_loss=0.07163, over 3832215.07 frames. ], batch size: 54, lr: 5.22e-03, grad_scale: 8.0 +2023-04-02 06:54:21,785 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 06:54:22,923 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.793e+02 5.248e+02 6.749e+02 7.716e+02 1.558e+03, threshold=1.350e+03, percent-clipped=5.0 +2023-04-02 06:54:26,609 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-02 06:54:46,589 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5849, 1.3540, 1.5448, 1.5560, 3.1454, 1.1327, 2.2693, 3.5441], + device='cuda:0'), covar=tensor([0.0444, 0.2629, 0.2726, 0.1735, 0.0711, 0.2406, 0.1249, 0.0245], + device='cuda:0'), in_proj_covar=tensor([0.0381, 0.0349, 0.0367, 0.0334, 0.0357, 0.0339, 0.0351, 0.0374], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 06:54:51,925 INFO [train.py:903] (0/4) Epoch 16, batch 1900, loss[loss=0.266, simple_loss=0.3369, pruned_loss=0.09756, over 19743.00 frames. ], tot_loss[loss=0.2211, simple_loss=0.2977, pruned_loss=0.0722, over 3799303.68 frames. ], batch size: 63, lr: 5.22e-03, grad_scale: 4.0 +2023-04-02 06:55:09,077 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 06:55:15,645 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 06:55:31,877 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104352.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:55:34,121 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104353.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:55:39,870 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 06:55:41,199 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104359.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:55:53,979 INFO [train.py:903] (0/4) Epoch 16, batch 1950, loss[loss=0.2119, simple_loss=0.2975, pruned_loss=0.06311, over 19510.00 frames. ], tot_loss[loss=0.2221, simple_loss=0.2986, pruned_loss=0.07277, over 3814057.83 frames. ], batch size: 64, lr: 5.21e-03, grad_scale: 4.0 +2023-04-02 06:56:30,730 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.975e+02 4.688e+02 6.377e+02 8.120e+02 1.703e+03, threshold=1.275e+03, percent-clipped=4.0 +2023-04-02 06:56:46,822 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=104411.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:56:58,261 INFO [train.py:903] (0/4) Epoch 16, batch 2000, loss[loss=0.1959, simple_loss=0.2772, pruned_loss=0.05724, over 19850.00 frames. ], tot_loss[loss=0.2207, simple_loss=0.2975, pruned_loss=0.07195, over 3812564.01 frames. ], batch size: 52, lr: 5.21e-03, grad_scale: 8.0 +2023-04-02 06:57:57,340 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 06:57:57,610 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104467.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:58:00,766 INFO [train.py:903] (0/4) Epoch 16, batch 2050, loss[loss=0.169, simple_loss=0.2396, pruned_loss=0.04918, over 19309.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2967, pruned_loss=0.07149, over 3814309.35 frames. ], batch size: 44, lr: 5.21e-03, grad_scale: 8.0 +2023-04-02 06:58:04,661 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104473.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:58:05,701 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104474.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:58:09,152 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-02 06:58:14,089 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 06:58:15,353 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 06:58:35,725 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.098e+02 4.885e+02 6.080e+02 8.555e+02 1.693e+03, threshold=1.216e+03, percent-clipped=6.0 +2023-04-02 06:58:36,103 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=104498.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:58:36,135 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104498.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:58:39,795 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-04-02 06:58:40,126 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 06:59:02,733 INFO [train.py:903] (0/4) Epoch 16, batch 2100, loss[loss=0.2278, simple_loss=0.3061, pruned_loss=0.07481, over 19689.00 frames. ], tot_loss[loss=0.2201, simple_loss=0.2968, pruned_loss=0.07168, over 3830462.05 frames. ], batch size: 59, lr: 5.21e-03, grad_scale: 8.0 +2023-04-02 06:59:06,389 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=104523.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:59:09,923 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=104526.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 06:59:33,807 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 06:59:55,524 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 06:59:56,919 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8639, 4.9221, 5.6510, 5.6293, 2.1763, 5.2768, 4.5043, 5.2558], + device='cuda:0'), covar=tensor([0.1401, 0.0906, 0.0531, 0.0565, 0.5164, 0.0653, 0.0550, 0.1105], + device='cuda:0'), in_proj_covar=tensor([0.0726, 0.0667, 0.0868, 0.0748, 0.0771, 0.0618, 0.0515, 0.0799], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 07:00:04,689 INFO [train.py:903] (0/4) Epoch 16, batch 2150, loss[loss=0.2962, simple_loss=0.3549, pruned_loss=0.1187, over 19668.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2963, pruned_loss=0.07161, over 3831950.53 frames. ], batch size: 58, lr: 5.21e-03, grad_scale: 8.0 +2023-04-02 07:00:39,764 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.712e+02 5.547e+02 6.907e+02 8.298e+02 2.194e+03, threshold=1.381e+03, percent-clipped=3.0 +2023-04-02 07:00:49,284 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0356, 1.7450, 1.8716, 1.7209, 4.5030, 1.0514, 2.4452, 4.8729], + device='cuda:0'), covar=tensor([0.0371, 0.2602, 0.2618, 0.1862, 0.0723, 0.2609, 0.1408, 0.0177], + device='cuda:0'), in_proj_covar=tensor([0.0384, 0.0351, 0.0369, 0.0335, 0.0360, 0.0340, 0.0354, 0.0376], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:00:53,773 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104609.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:01:08,291 INFO [train.py:903] (0/4) Epoch 16, batch 2200, loss[loss=0.1872, simple_loss=0.2583, pruned_loss=0.05804, over 19773.00 frames. ], tot_loss[loss=0.2193, simple_loss=0.296, pruned_loss=0.07136, over 3839945.23 frames. ], batch size: 48, lr: 5.21e-03, grad_scale: 8.0 +2023-04-02 07:01:26,328 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=104634.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:01:41,695 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6286, 2.3884, 1.7809, 1.4614, 2.1690, 1.2320, 1.3512, 1.9268], + device='cuda:0'), covar=tensor([0.0990, 0.0726, 0.1053, 0.0835, 0.0561, 0.1335, 0.0786, 0.0467], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0309, 0.0328, 0.0255, 0.0244, 0.0332, 0.0291, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:02:12,257 INFO [train.py:903] (0/4) Epoch 16, batch 2250, loss[loss=0.2229, simple_loss=0.2864, pruned_loss=0.07974, over 19768.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2959, pruned_loss=0.07121, over 3828892.43 frames. ], batch size: 48, lr: 5.21e-03, grad_scale: 8.0 +2023-04-02 07:02:46,756 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.071e+02 4.921e+02 5.898e+02 6.952e+02 1.646e+03, threshold=1.180e+03, percent-clipped=1.0 +2023-04-02 07:03:15,187 INFO [train.py:903] (0/4) Epoch 16, batch 2300, loss[loss=0.1828, simple_loss=0.2615, pruned_loss=0.0521, over 19050.00 frames. ], tot_loss[loss=0.2181, simple_loss=0.2952, pruned_loss=0.07047, over 3826525.43 frames. ], batch size: 42, lr: 5.21e-03, grad_scale: 8.0 +2023-04-02 07:03:19,214 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104723.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:03:27,409 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104730.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:03:29,350 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 07:03:51,587 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=104748.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:03:59,908 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=104755.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:04:17,946 INFO [train.py:903] (0/4) Epoch 16, batch 2350, loss[loss=0.208, simple_loss=0.2881, pruned_loss=0.06392, over 19444.00 frames. ], tot_loss[loss=0.217, simple_loss=0.294, pruned_loss=0.06996, over 3822352.31 frames. ], batch size: 64, lr: 5.20e-03, grad_scale: 8.0 +2023-04-02 07:04:34,887 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=104782.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:04:53,849 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.557e+02 5.048e+02 6.738e+02 8.844e+02 1.580e+03, threshold=1.348e+03, percent-clipped=5.0 +2023-04-02 07:05:00,895 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 07:05:05,754 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=104807.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:05:18,375 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 07:05:22,501 INFO [train.py:903] (0/4) Epoch 16, batch 2400, loss[loss=0.2244, simple_loss=0.301, pruned_loss=0.0739, over 19661.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2946, pruned_loss=0.07005, over 3831701.61 frames. ], batch size: 53, lr: 5.20e-03, grad_scale: 8.0 +2023-04-02 07:06:20,843 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-04-02 07:06:24,571 INFO [train.py:903] (0/4) Epoch 16, batch 2450, loss[loss=0.2413, simple_loss=0.3203, pruned_loss=0.08115, over 19677.00 frames. ], tot_loss[loss=0.2187, simple_loss=0.2957, pruned_loss=0.07086, over 3817765.89 frames. ], batch size: 60, lr: 5.20e-03, grad_scale: 8.0 +2023-04-02 07:07:00,038 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.524e+02 5.877e+02 7.599e+02 9.302e+02 2.010e+03, threshold=1.520e+03, percent-clipped=8.0 +2023-04-02 07:07:13,328 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5768, 1.1972, 1.4015, 1.3504, 2.2110, 0.9950, 2.0277, 2.4261], + device='cuda:0'), covar=tensor([0.0684, 0.2729, 0.2693, 0.1524, 0.0908, 0.2004, 0.0992, 0.0461], + device='cuda:0'), in_proj_covar=tensor([0.0385, 0.0354, 0.0371, 0.0336, 0.0361, 0.0342, 0.0358, 0.0379], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:07:27,189 INFO [train.py:903] (0/4) Epoch 16, batch 2500, loss[loss=0.2226, simple_loss=0.3044, pruned_loss=0.0704, over 19765.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2957, pruned_loss=0.07121, over 3813602.01 frames. ], batch size: 63, lr: 5.20e-03, grad_scale: 8.0 +2023-04-02 07:07:30,681 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9496, 3.4818, 1.8224, 2.1693, 3.0881, 1.6887, 1.3592, 2.2042], + device='cuda:0'), covar=tensor([0.1447, 0.0593, 0.1080, 0.0751, 0.0488, 0.1158, 0.1078, 0.0641], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0314, 0.0330, 0.0257, 0.0244, 0.0333, 0.0295, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:07:39,237 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.53 vs. limit=2.0 +2023-04-02 07:07:48,630 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-04-02 07:08:26,459 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8374, 4.4381, 2.7121, 3.8653, 0.9972, 4.1897, 4.2100, 4.2580], + device='cuda:0'), covar=tensor([0.0590, 0.0907, 0.1836, 0.0770, 0.3923, 0.0656, 0.0779, 0.1092], + device='cuda:0'), in_proj_covar=tensor([0.0469, 0.0384, 0.0459, 0.0329, 0.0390, 0.0396, 0.0391, 0.0424], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:08:29,683 INFO [train.py:903] (0/4) Epoch 16, batch 2550, loss[loss=0.1822, simple_loss=0.2592, pruned_loss=0.05265, over 19403.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2967, pruned_loss=0.07147, over 3821907.04 frames. ], batch size: 48, lr: 5.20e-03, grad_scale: 8.0 +2023-04-02 07:09:05,174 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.330e+02 5.239e+02 6.384e+02 8.143e+02 1.987e+03, threshold=1.277e+03, percent-clipped=1.0 +2023-04-02 07:09:24,257 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-02 07:09:24,637 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 07:09:32,157 INFO [train.py:903] (0/4) Epoch 16, batch 2600, loss[loss=0.258, simple_loss=0.3369, pruned_loss=0.08956, over 19449.00 frames. ], tot_loss[loss=0.2199, simple_loss=0.2967, pruned_loss=0.0715, over 3820403.19 frames. ], batch size: 70, lr: 5.20e-03, grad_scale: 8.0 +2023-04-02 07:10:03,892 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3690, 2.1927, 2.1435, 2.6145, 2.4244, 2.1019, 2.0387, 2.4032], + device='cuda:0'), covar=tensor([0.0890, 0.1494, 0.1209, 0.0833, 0.1168, 0.0510, 0.1186, 0.0611], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0352, 0.0298, 0.0247, 0.0300, 0.0249, 0.0294, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:10:21,330 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4508, 2.3501, 1.7136, 1.5516, 2.1898, 1.3201, 1.2981, 1.8603], + device='cuda:0'), covar=tensor([0.1146, 0.0689, 0.0963, 0.0826, 0.0466, 0.1169, 0.0777, 0.0515], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0313, 0.0329, 0.0258, 0.0243, 0.0333, 0.0294, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:10:35,304 INFO [train.py:903] (0/4) Epoch 16, batch 2650, loss[loss=0.2381, simple_loss=0.3187, pruned_loss=0.07874, over 18754.00 frames. ], tot_loss[loss=0.2188, simple_loss=0.296, pruned_loss=0.07081, over 3813833.91 frames. ], batch size: 74, lr: 5.20e-03, grad_scale: 8.0 +2023-04-02 07:10:54,808 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 07:11:09,896 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.780e+02 5.317e+02 6.316e+02 8.001e+02 1.365e+03, threshold=1.263e+03, percent-clipped=2.0 +2023-04-02 07:11:36,912 INFO [train.py:903] (0/4) Epoch 16, batch 2700, loss[loss=0.2171, simple_loss=0.303, pruned_loss=0.06566, over 19363.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2974, pruned_loss=0.07102, over 3824997.50 frames. ], batch size: 66, lr: 5.20e-03, grad_scale: 8.0 +2023-04-02 07:12:39,773 INFO [train.py:903] (0/4) Epoch 16, batch 2750, loss[loss=0.2289, simple_loss=0.3063, pruned_loss=0.07578, over 19674.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2974, pruned_loss=0.07125, over 3815802.17 frames. ], batch size: 60, lr: 5.19e-03, grad_scale: 8.0 +2023-04-02 07:13:15,057 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.123e+02 5.170e+02 6.445e+02 8.543e+02 2.677e+03, threshold=1.289e+03, percent-clipped=7.0 +2023-04-02 07:13:41,785 INFO [train.py:903] (0/4) Epoch 16, batch 2800, loss[loss=0.2438, simple_loss=0.3199, pruned_loss=0.08385, over 19684.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.2978, pruned_loss=0.07184, over 3818854.25 frames. ], batch size: 60, lr: 5.19e-03, grad_scale: 8.0 +2023-04-02 07:14:44,325 INFO [train.py:903] (0/4) Epoch 16, batch 2850, loss[loss=0.2318, simple_loss=0.3125, pruned_loss=0.07552, over 18900.00 frames. ], tot_loss[loss=0.2215, simple_loss=0.2985, pruned_loss=0.07222, over 3827769.11 frames. ], batch size: 74, lr: 5.19e-03, grad_scale: 8.0 +2023-04-02 07:15:18,937 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.908e+02 5.066e+02 6.361e+02 8.222e+02 2.548e+03, threshold=1.272e+03, percent-clipped=4.0 +2023-04-02 07:15:46,243 INFO [train.py:903] (0/4) Epoch 16, batch 2900, loss[loss=0.2342, simple_loss=0.3033, pruned_loss=0.08252, over 19855.00 frames. ], tot_loss[loss=0.2221, simple_loss=0.2989, pruned_loss=0.07262, over 3832273.08 frames. ], batch size: 52, lr: 5.19e-03, grad_scale: 8.0 +2023-04-02 07:15:46,285 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 07:16:48,798 INFO [train.py:903] (0/4) Epoch 16, batch 2950, loss[loss=0.2113, simple_loss=0.2846, pruned_loss=0.06902, over 19329.00 frames. ], tot_loss[loss=0.2212, simple_loss=0.2983, pruned_loss=0.07201, over 3825114.18 frames. ], batch size: 44, lr: 5.19e-03, grad_scale: 8.0 +2023-04-02 07:17:23,864 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.147e+02 4.947e+02 6.195e+02 7.724e+02 2.015e+03, threshold=1.239e+03, percent-clipped=3.0 +2023-04-02 07:17:24,283 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8832, 1.6090, 1.5041, 1.9252, 1.6872, 1.6832, 1.5079, 1.8122], + device='cuda:0'), covar=tensor([0.1025, 0.1479, 0.1489, 0.0958, 0.1244, 0.0533, 0.1384, 0.0706], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0353, 0.0299, 0.0247, 0.0299, 0.0250, 0.0295, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:17:33,595 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=105406.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:17:50,878 INFO [train.py:903] (0/4) Epoch 16, batch 3000, loss[loss=0.2073, simple_loss=0.2922, pruned_loss=0.06123, over 19803.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.2983, pruned_loss=0.07164, over 3843935.61 frames. ], batch size: 56, lr: 5.19e-03, grad_scale: 8.0 +2023-04-02 07:17:50,878 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 07:18:04,146 INFO [train.py:937] (0/4) Epoch 16, validation: loss=0.1725, simple_loss=0.273, pruned_loss=0.03604, over 944034.00 frames. +2023-04-02 07:18:04,147 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 07:18:07,769 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 07:19:07,110 INFO [train.py:903] (0/4) Epoch 16, batch 3050, loss[loss=0.1806, simple_loss=0.26, pruned_loss=0.0506, over 19793.00 frames. ], tot_loss[loss=0.2189, simple_loss=0.2967, pruned_loss=0.07059, over 3852427.51 frames. ], batch size: 48, lr: 5.19e-03, grad_scale: 8.0 +2023-04-02 07:19:41,628 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.577e+02 4.913e+02 6.190e+02 8.953e+02 2.496e+03, threshold=1.238e+03, percent-clipped=7.0 +2023-04-02 07:19:49,924 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=105504.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:20:11,581 INFO [train.py:903] (0/4) Epoch 16, batch 3100, loss[loss=0.2182, simple_loss=0.3045, pruned_loss=0.06594, over 19357.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2953, pruned_loss=0.06997, over 3851247.72 frames. ], batch size: 70, lr: 5.19e-03, grad_scale: 8.0 +2023-04-02 07:20:11,860 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=105520.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:20:40,008 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1063, 1.2614, 1.7145, 1.3518, 2.8155, 3.7209, 3.4926, 3.9794], + device='cuda:0'), covar=tensor([0.1752, 0.3797, 0.3313, 0.2272, 0.0550, 0.0175, 0.0200, 0.0207], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0307, 0.0337, 0.0255, 0.0230, 0.0174, 0.0210, 0.0230], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 07:21:13,220 INFO [train.py:903] (0/4) Epoch 16, batch 3150, loss[loss=0.2709, simple_loss=0.3433, pruned_loss=0.0992, over 17310.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2955, pruned_loss=0.06985, over 3854842.82 frames. ], batch size: 101, lr: 5.18e-03, grad_scale: 8.0 +2023-04-02 07:21:41,321 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 07:21:46,677 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.213e+02 5.014e+02 6.069e+02 7.545e+02 2.509e+03, threshold=1.214e+03, percent-clipped=2.0 +2023-04-02 07:22:12,912 INFO [train.py:903] (0/4) Epoch 16, batch 3200, loss[loss=0.2014, simple_loss=0.2855, pruned_loss=0.05867, over 19856.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2966, pruned_loss=0.07079, over 3854799.78 frames. ], batch size: 52, lr: 5.18e-03, grad_scale: 8.0 +2023-04-02 07:23:15,478 INFO [train.py:903] (0/4) Epoch 16, batch 3250, loss[loss=0.2127, simple_loss=0.2877, pruned_loss=0.06885, over 19186.00 frames. ], tot_loss[loss=0.2184, simple_loss=0.2959, pruned_loss=0.07041, over 3846979.28 frames. ], batch size: 69, lr: 5.18e-03, grad_scale: 8.0 +2023-04-02 07:23:50,142 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.698e+02 5.141e+02 5.939e+02 7.859e+02 1.653e+03, threshold=1.188e+03, percent-clipped=2.0 +2023-04-02 07:24:19,063 INFO [train.py:903] (0/4) Epoch 16, batch 3300, loss[loss=0.2555, simple_loss=0.3236, pruned_loss=0.09373, over 19579.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2956, pruned_loss=0.07052, over 3830157.63 frames. ], batch size: 52, lr: 5.18e-03, grad_scale: 8.0 +2023-04-02 07:24:19,297 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9801, 4.5434, 2.6634, 4.0258, 0.8638, 4.4542, 4.3827, 4.4983], + device='cuda:0'), covar=tensor([0.0518, 0.0885, 0.1804, 0.0630, 0.3760, 0.0516, 0.0704, 0.0884], + device='cuda:0'), in_proj_covar=tensor([0.0471, 0.0387, 0.0463, 0.0328, 0.0393, 0.0398, 0.0394, 0.0428], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:24:22,584 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 07:24:48,441 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4237, 2.1224, 1.8967, 2.7599, 2.1386, 2.4802, 2.4888, 2.4333], + device='cuda:0'), covar=tensor([0.0708, 0.0876, 0.0942, 0.0848, 0.0875, 0.0732, 0.0868, 0.0602], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0220, 0.0221, 0.0240, 0.0224, 0.0207, 0.0189, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 07:24:55,795 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=105750.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:25:21,449 INFO [train.py:903] (0/4) Epoch 16, batch 3350, loss[loss=0.1629, simple_loss=0.2379, pruned_loss=0.0439, over 19757.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2942, pruned_loss=0.06945, over 3837702.40 frames. ], batch size: 46, lr: 5.18e-03, grad_scale: 8.0 +2023-04-02 07:25:31,893 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-04-02 07:25:57,717 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.211e+02 5.066e+02 6.242e+02 8.095e+02 2.652e+03, threshold=1.248e+03, percent-clipped=5.0 +2023-04-02 07:26:12,306 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=105810.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:26:20,486 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=105817.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:26:23,649 INFO [train.py:903] (0/4) Epoch 16, batch 3400, loss[loss=0.2306, simple_loss=0.3089, pruned_loss=0.0761, over 19048.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2952, pruned_loss=0.07002, over 3826045.66 frames. ], batch size: 69, lr: 5.18e-03, grad_scale: 8.0 +2023-04-02 07:26:29,258 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.70 vs. limit=2.0 +2023-04-02 07:26:59,994 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=105848.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:27:18,390 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=105864.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:27:19,628 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=105865.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:27:26,111 INFO [train.py:903] (0/4) Epoch 16, batch 3450, loss[loss=0.1983, simple_loss=0.2703, pruned_loss=0.06314, over 16047.00 frames. ], tot_loss[loss=0.2174, simple_loss=0.2951, pruned_loss=0.06984, over 3828401.04 frames. ], batch size: 35, lr: 5.18e-03, grad_scale: 8.0 +2023-04-02 07:27:29,346 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 07:28:00,500 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.376e+02 5.355e+02 6.157e+02 7.690e+02 1.854e+03, threshold=1.231e+03, percent-clipped=4.0 +2023-04-02 07:28:29,093 INFO [train.py:903] (0/4) Epoch 16, batch 3500, loss[loss=0.2412, simple_loss=0.3178, pruned_loss=0.08227, over 19783.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2952, pruned_loss=0.07003, over 3831045.95 frames. ], batch size: 54, lr: 5.18e-03, grad_scale: 8.0 +2023-04-02 07:29:23,569 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=105963.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:29:31,025 INFO [train.py:903] (0/4) Epoch 16, batch 3550, loss[loss=0.2453, simple_loss=0.3175, pruned_loss=0.08651, over 19660.00 frames. ], tot_loss[loss=0.2186, simple_loss=0.2957, pruned_loss=0.07075, over 3841198.83 frames. ], batch size: 60, lr: 5.17e-03, grad_scale: 8.0 +2023-04-02 07:29:42,155 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=105979.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:30:06,596 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.318e+02 5.018e+02 6.219e+02 7.272e+02 1.453e+03, threshold=1.244e+03, percent-clipped=3.0 +2023-04-02 07:30:10,061 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-106000.pt +2023-04-02 07:30:34,290 INFO [train.py:903] (0/4) Epoch 16, batch 3600, loss[loss=0.2295, simple_loss=0.3089, pruned_loss=0.07506, over 17484.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2952, pruned_loss=0.07043, over 3826293.61 frames. ], batch size: 101, lr: 5.17e-03, grad_scale: 8.0 +2023-04-02 07:30:46,588 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.53 vs. limit=5.0 +2023-04-02 07:30:47,285 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1132, 1.1730, 1.2875, 1.2402, 1.5348, 1.5788, 1.5082, 0.5494], + device='cuda:0'), covar=tensor([0.2129, 0.3614, 0.2277, 0.1717, 0.1355, 0.1995, 0.1150, 0.3915], + device='cuda:0'), in_proj_covar=tensor([0.0507, 0.0608, 0.0659, 0.0461, 0.0604, 0.0510, 0.0648, 0.0520], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 07:31:37,380 INFO [train.py:903] (0/4) Epoch 16, batch 3650, loss[loss=0.175, simple_loss=0.2526, pruned_loss=0.04871, over 19375.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2954, pruned_loss=0.07062, over 3814615.15 frames. ], batch size: 47, lr: 5.17e-03, grad_scale: 8.0 +2023-04-02 07:32:09,751 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106095.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:32:13,044 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.400e+02 5.371e+02 6.801e+02 8.277e+02 1.518e+03, threshold=1.360e+03, percent-clipped=5.0 +2023-04-02 07:32:30,556 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9209, 2.0281, 2.1604, 2.5424, 1.8853, 2.4550, 2.2111, 2.0155], + device='cuda:0'), covar=tensor([0.3965, 0.3364, 0.1760, 0.2189, 0.3801, 0.1862, 0.4343, 0.2955], + device='cuda:0'), in_proj_covar=tensor([0.0839, 0.0889, 0.0679, 0.0905, 0.0825, 0.0759, 0.0809, 0.0743], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 07:32:42,528 INFO [train.py:903] (0/4) Epoch 16, batch 3700, loss[loss=0.1865, simple_loss=0.2672, pruned_loss=0.05296, over 19398.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.294, pruned_loss=0.06955, over 3829476.63 frames. ], batch size: 48, lr: 5.17e-03, grad_scale: 8.0 +2023-04-02 07:32:44,087 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106121.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:32:52,258 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7301, 1.8279, 2.0589, 2.3833, 1.7138, 2.2535, 2.1733, 1.9172], + device='cuda:0'), covar=tensor([0.3879, 0.3504, 0.1663, 0.1902, 0.3529, 0.1758, 0.4060, 0.3062], + device='cuda:0'), in_proj_covar=tensor([0.0838, 0.0887, 0.0678, 0.0903, 0.0822, 0.0756, 0.0808, 0.0742], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 07:33:13,534 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106146.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:33:23,569 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=106154.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:33:34,045 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=106161.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:33:44,481 INFO [train.py:903] (0/4) Epoch 16, batch 3750, loss[loss=0.2449, simple_loss=0.3173, pruned_loss=0.08623, over 18199.00 frames. ], tot_loss[loss=0.2155, simple_loss=0.293, pruned_loss=0.06901, over 3834852.21 frames. ], batch size: 83, lr: 5.17e-03, grad_scale: 8.0 +2023-04-02 07:34:00,238 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.57 vs. limit=2.0 +2023-04-02 07:34:15,822 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8232, 1.3969, 1.5767, 1.7195, 3.3920, 1.2045, 2.3159, 3.7824], + device='cuda:0'), covar=tensor([0.0448, 0.2705, 0.2675, 0.1699, 0.0655, 0.2463, 0.1380, 0.0231], + device='cuda:0'), in_proj_covar=tensor([0.0388, 0.0352, 0.0371, 0.0337, 0.0359, 0.0340, 0.0354, 0.0377], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 07:34:19,089 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.576e+02 4.705e+02 5.517e+02 6.973e+02 1.532e+03, threshold=1.103e+03, percent-clipped=3.0 +2023-04-02 07:34:45,448 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106219.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:34:46,185 INFO [train.py:903] (0/4) Epoch 16, batch 3800, loss[loss=0.2301, simple_loss=0.2918, pruned_loss=0.08419, over 19398.00 frames. ], tot_loss[loss=0.2159, simple_loss=0.293, pruned_loss=0.06945, over 3819978.86 frames. ], batch size: 48, lr: 5.17e-03, grad_scale: 8.0 +2023-04-02 07:35:06,437 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106235.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:35:18,156 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 07:35:18,458 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106244.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:35:36,738 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106260.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:35:48,039 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106269.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:35:48,820 INFO [train.py:903] (0/4) Epoch 16, batch 3850, loss[loss=0.2133, simple_loss=0.296, pruned_loss=0.06532, over 19652.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2935, pruned_loss=0.06985, over 3826535.65 frames. ], batch size: 53, lr: 5.17e-03, grad_scale: 8.0 +2023-04-02 07:35:56,804 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106276.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:36:09,573 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106286.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:36:23,153 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.040e+02 5.267e+02 6.364e+02 7.734e+02 1.610e+03, threshold=1.273e+03, percent-clipped=5.0 +2023-04-02 07:36:50,364 INFO [train.py:903] (0/4) Epoch 16, batch 3900, loss[loss=0.2383, simple_loss=0.3086, pruned_loss=0.08399, over 18223.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2941, pruned_loss=0.07024, over 3832529.63 frames. ], batch size: 84, lr: 5.17e-03, grad_scale: 16.0 +2023-04-02 07:37:32,141 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106354.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:37:53,350 INFO [train.py:903] (0/4) Epoch 16, batch 3950, loss[loss=0.2343, simple_loss=0.3009, pruned_loss=0.08383, over 19745.00 frames. ], tot_loss[loss=0.2184, simple_loss=0.2954, pruned_loss=0.07071, over 3823600.95 frames. ], batch size: 51, lr: 5.16e-03, grad_scale: 8.0 +2023-04-02 07:37:58,036 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 07:38:28,357 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.667e+02 4.931e+02 6.019e+02 7.839e+02 1.392e+03, threshold=1.204e+03, percent-clipped=3.0 +2023-04-02 07:38:54,928 INFO [train.py:903] (0/4) Epoch 16, batch 4000, loss[loss=0.2268, simple_loss=0.3064, pruned_loss=0.07366, over 19521.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2954, pruned_loss=0.07027, over 3837686.86 frames. ], batch size: 54, lr: 5.16e-03, grad_scale: 8.0 +2023-04-02 07:39:03,764 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.61 vs. limit=2.0 +2023-04-02 07:39:19,315 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=106439.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:39:23,114 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2108, 1.2750, 1.2050, 1.0312, 1.0927, 1.0814, 0.0683, 0.3879], + device='cuda:0'), covar=tensor([0.0590, 0.0565, 0.0345, 0.0485, 0.1070, 0.0530, 0.1082, 0.0926], + device='cuda:0'), in_proj_covar=tensor([0.0352, 0.0345, 0.0343, 0.0371, 0.0445, 0.0375, 0.0323, 0.0330], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 07:39:45,487 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 07:39:56,761 INFO [train.py:903] (0/4) Epoch 16, batch 4050, loss[loss=0.2266, simple_loss=0.3041, pruned_loss=0.07456, over 19313.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2951, pruned_loss=0.07008, over 3831423.02 frames. ], batch size: 66, lr: 5.16e-03, grad_scale: 8.0 +2023-04-02 07:39:59,360 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=106472.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:40:22,946 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5002, 3.0925, 2.2678, 2.4691, 2.4231, 2.6717, 1.0718, 2.2908], + device='cuda:0'), covar=tensor([0.0567, 0.0533, 0.0636, 0.0954, 0.0862, 0.0994, 0.1148, 0.0847], + device='cuda:0'), in_proj_covar=tensor([0.0353, 0.0346, 0.0345, 0.0372, 0.0447, 0.0378, 0.0325, 0.0332], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 07:40:34,186 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.509e+02 4.912e+02 6.033e+02 8.007e+02 1.551e+03, threshold=1.207e+03, percent-clipped=4.0 +2023-04-02 07:40:59,901 INFO [train.py:903] (0/4) Epoch 16, batch 4100, loss[loss=0.2068, simple_loss=0.2931, pruned_loss=0.06018, over 19549.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.294, pruned_loss=0.06968, over 3822406.58 frames. ], batch size: 56, lr: 5.16e-03, grad_scale: 8.0 +2023-04-02 07:41:07,082 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106525.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:41:15,865 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106532.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:41:34,835 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 07:41:36,425 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106550.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:41:41,938 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106554.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:41:45,517 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106557.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:42:02,445 INFO [train.py:903] (0/4) Epoch 16, batch 4150, loss[loss=0.221, simple_loss=0.3066, pruned_loss=0.06768, over 19559.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2945, pruned_loss=0.07037, over 3811201.16 frames. ], batch size: 61, lr: 5.16e-03, grad_scale: 8.0 +2023-04-02 07:42:20,385 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4011, 1.4850, 1.6198, 1.6245, 2.1990, 2.1766, 2.3081, 0.8313], + device='cuda:0'), covar=tensor([0.2115, 0.3822, 0.2462, 0.1648, 0.1438, 0.1846, 0.1242, 0.4067], + device='cuda:0'), in_proj_covar=tensor([0.0506, 0.0606, 0.0658, 0.0458, 0.0602, 0.0506, 0.0647, 0.0517], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 07:42:35,027 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1261, 2.0003, 1.7686, 1.6374, 1.5116, 1.6463, 0.4548, 1.0460], + device='cuda:0'), covar=tensor([0.0567, 0.0570, 0.0414, 0.0731, 0.0994, 0.0914, 0.1091, 0.0915], + device='cuda:0'), in_proj_covar=tensor([0.0352, 0.0346, 0.0343, 0.0371, 0.0446, 0.0375, 0.0324, 0.0331], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 07:42:36,771 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.179e+02 4.926e+02 6.073e+02 7.216e+02 1.422e+03, threshold=1.215e+03, percent-clipped=1.0 +2023-04-02 07:43:03,952 INFO [train.py:903] (0/4) Epoch 16, batch 4200, loss[loss=0.2037, simple_loss=0.2922, pruned_loss=0.05763, over 18057.00 frames. ], tot_loss[loss=0.2188, simple_loss=0.2959, pruned_loss=0.07083, over 3823072.97 frames. ], batch size: 83, lr: 5.16e-03, grad_scale: 8.0 +2023-04-02 07:43:11,042 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 07:43:15,841 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=106630.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:44:05,135 INFO [train.py:903] (0/4) Epoch 16, batch 4250, loss[loss=0.241, simple_loss=0.3086, pruned_loss=0.08672, over 19569.00 frames. ], tot_loss[loss=0.219, simple_loss=0.2961, pruned_loss=0.07091, over 3812144.41 frames. ], batch size: 61, lr: 5.16e-03, grad_scale: 8.0 +2023-04-02 07:44:20,446 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 07:44:32,720 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 07:44:41,212 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=106698.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:44:42,276 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.502e+02 4.922e+02 6.114e+02 7.451e+02 1.808e+03, threshold=1.223e+03, percent-clipped=2.0 +2023-04-02 07:45:08,519 INFO [train.py:903] (0/4) Epoch 16, batch 4300, loss[loss=0.2414, simple_loss=0.3217, pruned_loss=0.0806, over 19742.00 frames. ], tot_loss[loss=0.2195, simple_loss=0.2964, pruned_loss=0.07136, over 3822928.00 frames. ], batch size: 63, lr: 5.16e-03, grad_scale: 8.0 +2023-04-02 07:45:23,513 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=5.05 vs. limit=5.0 +2023-04-02 07:45:38,967 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106745.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:45:58,070 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 07:46:11,647 INFO [train.py:903] (0/4) Epoch 16, batch 4350, loss[loss=0.2404, simple_loss=0.3128, pruned_loss=0.08403, over 19559.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2966, pruned_loss=0.07136, over 3817206.32 frames. ], batch size: 61, lr: 5.15e-03, grad_scale: 8.0 +2023-04-02 07:46:46,924 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.450e+02 5.102e+02 6.153e+02 8.101e+02 2.041e+03, threshold=1.231e+03, percent-clipped=8.0 +2023-04-02 07:46:48,701 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 07:47:03,099 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=106810.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:47:04,401 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-04-02 07:47:06,320 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106813.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:47:09,593 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=106816.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:47:14,208 INFO [train.py:903] (0/4) Epoch 16, batch 4400, loss[loss=0.2277, simple_loss=0.305, pruned_loss=0.07521, over 18810.00 frames. ], tot_loss[loss=0.2187, simple_loss=0.2957, pruned_loss=0.07086, over 3816584.97 frames. ], batch size: 74, lr: 5.15e-03, grad_scale: 8.0 +2023-04-02 07:47:33,138 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=106835.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:47:36,416 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 07:47:46,711 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 07:48:17,289 INFO [train.py:903] (0/4) Epoch 16, batch 4450, loss[loss=0.217, simple_loss=0.2972, pruned_loss=0.06837, over 19788.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2951, pruned_loss=0.07044, over 3807801.21 frames. ], batch size: 56, lr: 5.15e-03, grad_scale: 8.0 +2023-04-02 07:48:53,920 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.305e+02 4.966e+02 6.259e+02 8.420e+02 1.632e+03, threshold=1.252e+03, percent-clipped=6.0 +2023-04-02 07:49:18,991 INFO [train.py:903] (0/4) Epoch 16, batch 4500, loss[loss=0.2342, simple_loss=0.3131, pruned_loss=0.07769, over 17744.00 frames. ], tot_loss[loss=0.2181, simple_loss=0.2955, pruned_loss=0.07036, over 3817490.33 frames. ], batch size: 101, lr: 5.15e-03, grad_scale: 8.0 +2023-04-02 07:49:34,353 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=106931.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:50:23,572 INFO [train.py:903] (0/4) Epoch 16, batch 4550, loss[loss=0.1757, simple_loss=0.2534, pruned_loss=0.04904, over 19610.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2959, pruned_loss=0.07058, over 3812749.80 frames. ], batch size: 50, lr: 5.15e-03, grad_scale: 8.0 +2023-04-02 07:50:31,659 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 07:50:44,431 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.50 vs. limit=5.0 +2023-04-02 07:50:54,389 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 07:50:59,933 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.264e+02 4.883e+02 5.860e+02 7.136e+02 1.225e+03, threshold=1.172e+03, percent-clipped=0.0 +2023-04-02 07:51:02,615 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107001.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:51:27,843 INFO [train.py:903] (0/4) Epoch 16, batch 4600, loss[loss=0.1978, simple_loss=0.2702, pruned_loss=0.06275, over 19181.00 frames. ], tot_loss[loss=0.2194, simple_loss=0.2965, pruned_loss=0.07116, over 3797758.74 frames. ], batch size: 42, lr: 5.15e-03, grad_scale: 8.0 +2023-04-02 07:51:34,890 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107026.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:52:29,051 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107069.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:52:29,793 INFO [train.py:903] (0/4) Epoch 16, batch 4650, loss[loss=0.1982, simple_loss=0.2839, pruned_loss=0.05621, over 19460.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2968, pruned_loss=0.07127, over 3812725.76 frames. ], batch size: 64, lr: 5.15e-03, grad_scale: 8.0 +2023-04-02 07:52:47,200 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 07:52:59,800 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 07:53:01,348 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107094.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:53:07,461 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.220e+02 5.459e+02 6.581e+02 8.910e+02 1.601e+03, threshold=1.316e+03, percent-clipped=6.0 +2023-04-02 07:53:31,767 INFO [train.py:903] (0/4) Epoch 16, batch 4700, loss[loss=0.2172, simple_loss=0.2986, pruned_loss=0.06791, over 19683.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2971, pruned_loss=0.07128, over 3817967.25 frames. ], batch size: 53, lr: 5.15e-03, grad_scale: 8.0 +2023-04-02 07:53:55,910 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 07:54:12,255 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-04-02 07:54:20,564 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=107158.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 07:54:36,806 INFO [train.py:903] (0/4) Epoch 16, batch 4750, loss[loss=0.1645, simple_loss=0.2455, pruned_loss=0.04178, over 19400.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.296, pruned_loss=0.07047, over 3824791.11 frames. ], batch size: 48, lr: 5.15e-03, grad_scale: 8.0 +2023-04-02 07:54:37,106 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=107170.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:54:58,932 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107187.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:55:11,905 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.171e+02 5.545e+02 6.621e+02 8.650e+02 1.971e+03, threshold=1.324e+03, percent-clipped=6.0 +2023-04-02 07:55:29,423 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107212.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 07:55:40,210 INFO [train.py:903] (0/4) Epoch 16, batch 4800, loss[loss=0.2101, simple_loss=0.2875, pruned_loss=0.06633, over 19859.00 frames. ], tot_loss[loss=0.2186, simple_loss=0.2959, pruned_loss=0.07067, over 3832317.57 frames. ], batch size: 52, lr: 5.14e-03, grad_scale: 8.0 +2023-04-02 07:55:50,298 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-04-02 07:56:41,878 INFO [train.py:903] (0/4) Epoch 16, batch 4850, loss[loss=0.2207, simple_loss=0.3014, pruned_loss=0.07003, over 19521.00 frames. ], tot_loss[loss=0.2184, simple_loss=0.2959, pruned_loss=0.07049, over 3837455.76 frames. ], batch size: 64, lr: 5.14e-03, grad_scale: 4.0 +2023-04-02 07:57:07,071 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 07:57:19,837 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.329e+02 4.930e+02 6.295e+02 8.478e+02 1.665e+03, threshold=1.259e+03, percent-clipped=1.0 +2023-04-02 07:57:27,481 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 07:57:32,764 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 07:57:32,791 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 07:57:43,096 INFO [train.py:903] (0/4) Epoch 16, batch 4900, loss[loss=0.2204, simple_loss=0.3049, pruned_loss=0.06799, over 18227.00 frames. ], tot_loss[loss=0.219, simple_loss=0.2963, pruned_loss=0.07085, over 3833290.84 frames. ], batch size: 84, lr: 5.14e-03, grad_scale: 4.0 +2023-04-02 07:57:43,121 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 07:57:43,929 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.04 vs. limit=5.0 +2023-04-02 07:58:04,172 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 07:58:46,428 INFO [train.py:903] (0/4) Epoch 16, batch 4950, loss[loss=0.2056, simple_loss=0.2916, pruned_loss=0.05975, over 19587.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2969, pruned_loss=0.07113, over 3822728.35 frames. ], batch size: 52, lr: 5.14e-03, grad_scale: 4.0 +2023-04-02 07:59:04,266 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 07:59:22,737 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.794e+02 5.680e+02 6.678e+02 8.404e+02 2.020e+03, threshold=1.336e+03, percent-clipped=4.0 +2023-04-02 07:59:27,634 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 07:59:48,956 INFO [train.py:903] (0/4) Epoch 16, batch 5000, loss[loss=0.2673, simple_loss=0.3376, pruned_loss=0.09855, over 18055.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2963, pruned_loss=0.07102, over 3811615.79 frames. ], batch size: 83, lr: 5.14e-03, grad_scale: 4.0 +2023-04-02 07:59:58,942 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 08:00:09,000 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 08:00:50,457 INFO [train.py:903] (0/4) Epoch 16, batch 5050, loss[loss=0.1705, simple_loss=0.2484, pruned_loss=0.04628, over 19724.00 frames. ], tot_loss[loss=0.2194, simple_loss=0.2967, pruned_loss=0.07104, over 3817786.19 frames. ], batch size: 46, lr: 5.14e-03, grad_scale: 4.0 +2023-04-02 08:01:27,874 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.607e+02 5.470e+02 6.454e+02 8.047e+02 2.188e+03, threshold=1.291e+03, percent-clipped=2.0 +2023-04-02 08:01:27,916 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 08:01:30,381 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=107502.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 08:01:44,961 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=107514.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:01:51,915 INFO [train.py:903] (0/4) Epoch 16, batch 5100, loss[loss=0.2283, simple_loss=0.3084, pruned_loss=0.07412, over 19569.00 frames. ], tot_loss[loss=0.2202, simple_loss=0.2974, pruned_loss=0.07153, over 3824371.57 frames. ], batch size: 61, lr: 5.14e-03, grad_scale: 4.0 +2023-04-02 08:02:02,147 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=107528.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:02:04,287 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 08:02:08,815 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 08:02:13,300 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 08:02:54,727 INFO [train.py:903] (0/4) Epoch 16, batch 5150, loss[loss=0.1993, simple_loss=0.276, pruned_loss=0.06131, over 19845.00 frames. ], tot_loss[loss=0.2194, simple_loss=0.2968, pruned_loss=0.071, over 3835501.27 frames. ], batch size: 52, lr: 5.14e-03, grad_scale: 4.0 +2023-04-02 08:03:08,977 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 08:03:31,799 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.461e+02 4.950e+02 6.087e+02 7.766e+02 1.818e+03, threshold=1.217e+03, percent-clipped=6.0 +2023-04-02 08:03:43,257 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 08:03:54,420 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=107617.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 08:03:57,185 INFO [train.py:903] (0/4) Epoch 16, batch 5200, loss[loss=0.2191, simple_loss=0.2952, pruned_loss=0.0715, over 19667.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2959, pruned_loss=0.07059, over 3835479.69 frames. ], batch size: 58, lr: 5.13e-03, grad_scale: 8.0 +2023-04-02 08:04:08,672 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=107629.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:04:09,526 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 08:04:55,314 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 08:04:59,579 INFO [train.py:903] (0/4) Epoch 16, batch 5250, loss[loss=0.2268, simple_loss=0.3074, pruned_loss=0.07312, over 18758.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2963, pruned_loss=0.0709, over 3838401.47 frames. ], batch size: 74, lr: 5.13e-03, grad_scale: 8.0 +2023-04-02 08:05:07,710 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=107677.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:05:36,443 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.729e+02 5.464e+02 6.488e+02 8.647e+02 1.622e+03, threshold=1.298e+03, percent-clipped=8.0 +2023-04-02 08:05:42,433 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7243, 1.5948, 1.4825, 2.1457, 1.6361, 2.0963, 2.1551, 1.8488], + device='cuda:0'), covar=tensor([0.0755, 0.0858, 0.1000, 0.0778, 0.0839, 0.0676, 0.0766, 0.0642], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0221, 0.0222, 0.0245, 0.0225, 0.0207, 0.0189, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 08:06:00,481 INFO [train.py:903] (0/4) Epoch 16, batch 5300, loss[loss=0.2196, simple_loss=0.2972, pruned_loss=0.07107, over 19736.00 frames. ], tot_loss[loss=0.2172, simple_loss=0.295, pruned_loss=0.06971, over 3844923.99 frames. ], batch size: 51, lr: 5.13e-03, grad_scale: 8.0 +2023-04-02 08:06:19,308 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 08:06:28,019 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.2177, 5.5647, 3.1420, 4.8790, 1.1928, 5.6775, 5.5959, 5.8096], + device='cuda:0'), covar=tensor([0.0359, 0.0826, 0.1826, 0.0642, 0.3932, 0.0482, 0.0649, 0.0707], + device='cuda:0'), in_proj_covar=tensor([0.0470, 0.0385, 0.0463, 0.0328, 0.0391, 0.0399, 0.0398, 0.0427], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:06:28,280 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4120, 1.6092, 2.0081, 1.6919, 3.4319, 2.7020, 3.6135, 1.8085], + device='cuda:0'), covar=tensor([0.2456, 0.4185, 0.2628, 0.1889, 0.1362, 0.1924, 0.1450, 0.3592], + device='cuda:0'), in_proj_covar=tensor([0.0507, 0.0609, 0.0659, 0.0461, 0.0607, 0.0508, 0.0649, 0.0520], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 08:06:45,108 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=107755.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:07:03,282 INFO [train.py:903] (0/4) Epoch 16, batch 5350, loss[loss=0.2379, simple_loss=0.3196, pruned_loss=0.07813, over 19759.00 frames. ], tot_loss[loss=0.2181, simple_loss=0.2962, pruned_loss=0.06997, over 3850104.02 frames. ], batch size: 63, lr: 5.13e-03, grad_scale: 8.0 +2023-04-02 08:07:37,239 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 08:07:39,703 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3106, 3.0266, 2.1469, 2.7867, 0.7900, 2.9627, 2.9112, 2.9798], + device='cuda:0'), covar=tensor([0.1166, 0.1346, 0.2007, 0.0894, 0.3639, 0.0983, 0.1066, 0.1314], + device='cuda:0'), in_proj_covar=tensor([0.0470, 0.0384, 0.0463, 0.0328, 0.0391, 0.0399, 0.0399, 0.0428], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:07:40,608 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.374e+02 4.506e+02 5.884e+02 6.805e+02 1.610e+03, threshold=1.177e+03, percent-clipped=2.0 +2023-04-02 08:08:06,498 INFO [train.py:903] (0/4) Epoch 16, batch 5400, loss[loss=0.2332, simple_loss=0.3098, pruned_loss=0.07832, over 19624.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2958, pruned_loss=0.07015, over 3851657.43 frames. ], batch size: 57, lr: 5.13e-03, grad_scale: 8.0 +2023-04-02 08:09:08,379 INFO [train.py:903] (0/4) Epoch 16, batch 5450, loss[loss=0.2096, simple_loss=0.2953, pruned_loss=0.06193, over 19667.00 frames. ], tot_loss[loss=0.2181, simple_loss=0.2958, pruned_loss=0.07021, over 3852099.18 frames. ], batch size: 60, lr: 5.13e-03, grad_scale: 8.0 +2023-04-02 08:09:10,565 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=107872.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:09:11,922 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107873.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 08:09:26,836 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=107885.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:09:44,600 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107898.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 08:09:46,448 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.026e+02 5.228e+02 6.592e+02 8.752e+02 1.860e+03, threshold=1.318e+03, percent-clipped=11.0 +2023-04-02 08:09:58,327 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=107910.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:10:10,018 INFO [train.py:903] (0/4) Epoch 16, batch 5500, loss[loss=0.2239, simple_loss=0.2933, pruned_loss=0.07727, over 19414.00 frames. ], tot_loss[loss=0.2206, simple_loss=0.2978, pruned_loss=0.07168, over 3837741.20 frames. ], batch size: 48, lr: 5.13e-03, grad_scale: 8.0 +2023-04-02 08:10:12,869 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2381, 2.2064, 2.5465, 3.0808, 2.1059, 2.8861, 2.6470, 2.2087], + device='cuda:0'), covar=tensor([0.4212, 0.4210, 0.1716, 0.2487, 0.4624, 0.2104, 0.4399, 0.3334], + device='cuda:0'), in_proj_covar=tensor([0.0846, 0.0892, 0.0682, 0.0905, 0.0826, 0.0763, 0.0812, 0.0746], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 08:10:27,622 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8232, 3.2793, 3.3513, 3.3450, 1.3992, 3.2084, 2.8321, 3.1256], + device='cuda:0'), covar=tensor([0.1576, 0.0901, 0.0734, 0.0828, 0.4934, 0.0808, 0.0757, 0.1197], + device='cuda:0'), in_proj_covar=tensor([0.0736, 0.0678, 0.0879, 0.0761, 0.0784, 0.0630, 0.0524, 0.0817], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 08:10:34,949 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 08:10:43,660 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0445, 1.7200, 1.6782, 2.0196, 1.8357, 1.7772, 1.5981, 1.9384], + device='cuda:0'), covar=tensor([0.1018, 0.1603, 0.1560, 0.0997, 0.1314, 0.0533, 0.1380, 0.0743], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0354, 0.0302, 0.0244, 0.0299, 0.0248, 0.0292, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:11:00,097 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-02 08:11:06,868 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6581, 1.2228, 1.2879, 1.5174, 1.1456, 1.4367, 1.2182, 1.4784], + device='cuda:0'), covar=tensor([0.1052, 0.1229, 0.1490, 0.0929, 0.1234, 0.0561, 0.1463, 0.0763], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0353, 0.0301, 0.0244, 0.0298, 0.0248, 0.0291, 0.0246], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:11:13,110 INFO [train.py:903] (0/4) Epoch 16, batch 5550, loss[loss=0.2394, simple_loss=0.3084, pruned_loss=0.08522, over 19013.00 frames. ], tot_loss[loss=0.2201, simple_loss=0.2974, pruned_loss=0.07139, over 3822747.77 frames. ], batch size: 75, lr: 5.13e-03, grad_scale: 8.0 +2023-04-02 08:11:18,631 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 08:11:34,470 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=107987.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:11:50,037 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.970e+02 4.933e+02 6.287e+02 7.608e+02 2.106e+03, threshold=1.257e+03, percent-clipped=3.0 +2023-04-02 08:11:50,201 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-108000.pt +2023-04-02 08:12:10,685 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 08:12:17,775 INFO [train.py:903] (0/4) Epoch 16, batch 5600, loss[loss=0.1971, simple_loss=0.2862, pruned_loss=0.05401, over 19654.00 frames. ], tot_loss[loss=0.2202, simple_loss=0.2975, pruned_loss=0.07148, over 3828580.98 frames. ], batch size: 55, lr: 5.13e-03, grad_scale: 8.0 +2023-04-02 08:12:19,189 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=108021.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:12:48,000 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9383, 4.3611, 4.6737, 4.6434, 1.6129, 4.3721, 3.7444, 4.3342], + device='cuda:0'), covar=tensor([0.1504, 0.0734, 0.0569, 0.0647, 0.5795, 0.0683, 0.0676, 0.1126], + device='cuda:0'), in_proj_covar=tensor([0.0741, 0.0680, 0.0883, 0.0763, 0.0789, 0.0634, 0.0527, 0.0821], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 08:12:49,340 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3062, 1.4950, 1.8051, 1.5835, 2.9347, 2.2545, 3.0338, 1.2792], + device='cuda:0'), covar=tensor([0.2383, 0.3979, 0.2461, 0.1803, 0.1297, 0.1966, 0.1331, 0.3979], + device='cuda:0'), in_proj_covar=tensor([0.0510, 0.0608, 0.0661, 0.0460, 0.0606, 0.0509, 0.0647, 0.0519], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 08:13:17,742 INFO [train.py:903] (0/4) Epoch 16, batch 5650, loss[loss=0.2649, simple_loss=0.3282, pruned_loss=0.1008, over 19704.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2975, pruned_loss=0.07122, over 3823524.52 frames. ], batch size: 63, lr: 5.12e-03, grad_scale: 8.0 +2023-04-02 08:13:34,293 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3134, 3.0058, 2.0903, 2.6853, 0.8461, 2.9286, 2.8393, 2.9412], + device='cuda:0'), covar=tensor([0.1108, 0.1425, 0.2315, 0.1084, 0.4084, 0.1084, 0.1215, 0.1414], + device='cuda:0'), in_proj_covar=tensor([0.0470, 0.0387, 0.0464, 0.0329, 0.0394, 0.0402, 0.0401, 0.0429], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:13:55,624 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=108099.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:13:56,645 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.333e+02 5.688e+02 6.420e+02 8.034e+02 1.662e+03, threshold=1.284e+03, percent-clipped=4.0 +2023-04-02 08:14:03,619 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 08:14:21,000 INFO [train.py:903] (0/4) Epoch 16, batch 5700, loss[loss=0.2329, simple_loss=0.315, pruned_loss=0.07536, over 18761.00 frames. ], tot_loss[loss=0.2205, simple_loss=0.2975, pruned_loss=0.07169, over 3811830.47 frames. ], batch size: 74, lr: 5.12e-03, grad_scale: 8.0 +2023-04-02 08:14:41,782 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=108136.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:15:22,677 INFO [train.py:903] (0/4) Epoch 16, batch 5750, loss[loss=0.1849, simple_loss=0.2637, pruned_loss=0.05306, over 19772.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2974, pruned_loss=0.07132, over 3823066.70 frames. ], batch size: 46, lr: 5.12e-03, grad_scale: 8.0 +2023-04-02 08:15:22,708 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 08:15:33,017 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 08:15:36,710 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 08:15:39,369 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1328, 1.8038, 1.4351, 1.1137, 1.6065, 1.0774, 1.1433, 1.6486], + device='cuda:0'), covar=tensor([0.0728, 0.0756, 0.0944, 0.0737, 0.0473, 0.1254, 0.0571, 0.0341], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0308, 0.0328, 0.0254, 0.0240, 0.0329, 0.0290, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:16:00,581 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.754e+02 4.860e+02 6.382e+02 7.922e+02 1.732e+03, threshold=1.276e+03, percent-clipped=4.0 +2023-04-02 08:16:18,279 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=108214.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:16:26,740 INFO [train.py:903] (0/4) Epoch 16, batch 5800, loss[loss=0.2339, simple_loss=0.3061, pruned_loss=0.08083, over 19484.00 frames. ], tot_loss[loss=0.2202, simple_loss=0.2975, pruned_loss=0.07147, over 3806715.58 frames. ], batch size: 64, lr: 5.12e-03, grad_scale: 8.0 +2023-04-02 08:16:54,058 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=108243.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:17:25,828 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=108268.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:17:27,716 INFO [train.py:903] (0/4) Epoch 16, batch 5850, loss[loss=0.2493, simple_loss=0.3232, pruned_loss=0.08769, over 19561.00 frames. ], tot_loss[loss=0.2206, simple_loss=0.2977, pruned_loss=0.07175, over 3802411.57 frames. ], batch size: 61, lr: 5.12e-03, grad_scale: 8.0 +2023-04-02 08:18:05,292 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.738e+02 5.463e+02 6.888e+02 8.802e+02 1.964e+03, threshold=1.378e+03, percent-clipped=5.0 +2023-04-02 08:18:28,340 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 08:18:29,520 INFO [train.py:903] (0/4) Epoch 16, batch 5900, loss[loss=0.1961, simple_loss=0.2904, pruned_loss=0.05092, over 19794.00 frames. ], tot_loss[loss=0.2208, simple_loss=0.298, pruned_loss=0.07182, over 3803372.65 frames. ], batch size: 56, lr: 5.12e-03, grad_scale: 8.0 +2023-04-02 08:18:52,011 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 08:18:59,398 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=108343.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:19:30,982 INFO [train.py:903] (0/4) Epoch 16, batch 5950, loss[loss=0.1807, simple_loss=0.2645, pruned_loss=0.04848, over 19836.00 frames. ], tot_loss[loss=0.2203, simple_loss=0.2975, pruned_loss=0.0715, over 3800457.20 frames. ], batch size: 52, lr: 5.12e-03, grad_scale: 8.0 +2023-04-02 08:19:59,876 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=108392.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:20:09,618 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.801e+02 5.248e+02 6.439e+02 7.965e+02 2.252e+03, threshold=1.288e+03, percent-clipped=3.0 +2023-04-02 08:20:30,472 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=108417.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:20:35,475 INFO [train.py:903] (0/4) Epoch 16, batch 6000, loss[loss=0.2849, simple_loss=0.3467, pruned_loss=0.1115, over 18244.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2969, pruned_loss=0.07111, over 3808621.28 frames. ], batch size: 83, lr: 5.12e-03, grad_scale: 8.0 +2023-04-02 08:20:35,476 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 08:20:47,900 INFO [train.py:937] (0/4) Epoch 16, validation: loss=0.1716, simple_loss=0.2723, pruned_loss=0.03545, over 944034.00 frames. +2023-04-02 08:20:47,901 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 08:21:06,653 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9616, 3.4129, 2.0316, 2.0337, 3.0808, 1.7109, 1.5127, 2.1988], + device='cuda:0'), covar=tensor([0.1368, 0.0587, 0.1027, 0.0827, 0.0502, 0.1250, 0.0914, 0.0669], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0308, 0.0329, 0.0254, 0.0241, 0.0330, 0.0290, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:21:11,132 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-02 08:21:25,601 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=108450.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:21:33,880 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.51 vs. limit=2.0 +2023-04-02 08:21:51,714 INFO [train.py:903] (0/4) Epoch 16, batch 6050, loss[loss=0.1913, simple_loss=0.2798, pruned_loss=0.05144, over 19650.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2971, pruned_loss=0.07145, over 3820022.29 frames. ], batch size: 55, lr: 5.11e-03, grad_scale: 4.0 +2023-04-02 08:21:52,152 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=108470.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:22:02,504 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8611, 1.3479, 1.1010, 0.9809, 1.1938, 0.9928, 0.9307, 1.2181], + device='cuda:0'), covar=tensor([0.0576, 0.0864, 0.1033, 0.0701, 0.0525, 0.1218, 0.0571, 0.0439], + device='cuda:0'), in_proj_covar=tensor([0.0296, 0.0307, 0.0330, 0.0254, 0.0242, 0.0329, 0.0290, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:22:18,245 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-04-02 08:22:22,121 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=108495.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:22:28,586 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.284e+02 5.272e+02 6.402e+02 8.353e+02 1.883e+03, threshold=1.280e+03, percent-clipped=4.0 +2023-04-02 08:22:53,807 INFO [train.py:903] (0/4) Epoch 16, batch 6100, loss[loss=0.1908, simple_loss=0.2628, pruned_loss=0.05941, over 19112.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2969, pruned_loss=0.07134, over 3821739.68 frames. ], batch size: 42, lr: 5.11e-03, grad_scale: 4.0 +2023-04-02 08:23:56,091 INFO [train.py:903] (0/4) Epoch 16, batch 6150, loss[loss=0.2856, simple_loss=0.3522, pruned_loss=0.1095, over 19596.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2967, pruned_loss=0.07125, over 3831552.01 frames. ], batch size: 61, lr: 5.11e-03, grad_scale: 4.0 +2023-04-02 08:24:26,267 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 08:24:28,829 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=108595.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:24:35,758 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.970e+02 4.845e+02 6.012e+02 7.583e+02 1.796e+03, threshold=1.202e+03, percent-clipped=3.0 +2023-04-02 08:24:58,796 INFO [train.py:903] (0/4) Epoch 16, batch 6200, loss[loss=0.2641, simple_loss=0.3419, pruned_loss=0.09313, over 18112.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2965, pruned_loss=0.07135, over 3824633.46 frames. ], batch size: 83, lr: 5.11e-03, grad_scale: 4.0 +2023-04-02 08:26:02,198 INFO [train.py:903] (0/4) Epoch 16, batch 6250, loss[loss=0.2235, simple_loss=0.302, pruned_loss=0.07249, over 19553.00 frames. ], tot_loss[loss=0.2199, simple_loss=0.2968, pruned_loss=0.07147, over 3823552.95 frames. ], batch size: 56, lr: 5.11e-03, grad_scale: 4.0 +2023-04-02 08:26:22,913 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=108687.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:26:27,145 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-02 08:26:29,276 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2310, 2.2393, 2.5236, 3.1126, 2.2212, 2.8977, 2.6796, 2.2936], + device='cuda:0'), covar=tensor([0.3930, 0.3868, 0.1630, 0.2294, 0.4221, 0.1936, 0.3792, 0.2898], + device='cuda:0'), in_proj_covar=tensor([0.0847, 0.0891, 0.0679, 0.0906, 0.0826, 0.0764, 0.0813, 0.0745], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 08:26:34,479 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 08:26:40,079 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.877e+02 4.966e+02 6.026e+02 7.805e+02 1.726e+03, threshold=1.205e+03, percent-clipped=3.0 +2023-04-02 08:27:04,690 INFO [train.py:903] (0/4) Epoch 16, batch 6300, loss[loss=0.1846, simple_loss=0.2619, pruned_loss=0.05369, over 19770.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.297, pruned_loss=0.07131, over 3838833.29 frames. ], batch size: 47, lr: 5.11e-03, grad_scale: 4.0 +2023-04-02 08:27:07,178 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5455, 4.0387, 4.2044, 4.2161, 1.6768, 3.9671, 3.3974, 3.8850], + device='cuda:0'), covar=tensor([0.1573, 0.0933, 0.0649, 0.0687, 0.5335, 0.0911, 0.0670, 0.1231], + device='cuda:0'), in_proj_covar=tensor([0.0741, 0.0684, 0.0886, 0.0768, 0.0789, 0.0638, 0.0532, 0.0820], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 08:27:10,280 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9232, 4.4772, 2.6906, 4.0092, 1.0296, 4.2951, 4.2291, 4.3539], + device='cuda:0'), covar=tensor([0.0543, 0.0971, 0.1995, 0.0768, 0.3947, 0.0711, 0.0856, 0.0993], + device='cuda:0'), in_proj_covar=tensor([0.0466, 0.0381, 0.0461, 0.0329, 0.0388, 0.0398, 0.0397, 0.0424], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:28:06,338 INFO [train.py:903] (0/4) Epoch 16, batch 6350, loss[loss=0.2072, simple_loss=0.2817, pruned_loss=0.06638, over 19755.00 frames. ], tot_loss[loss=0.2198, simple_loss=0.2972, pruned_loss=0.07125, over 3846188.77 frames. ], batch size: 45, lr: 5.11e-03, grad_scale: 4.0 +2023-04-02 08:28:38,916 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=108794.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:28:47,231 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.196e+02 5.125e+02 6.267e+02 8.166e+02 1.468e+03, threshold=1.253e+03, percent-clipped=8.0 +2023-04-02 08:28:48,865 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=108802.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:28:54,181 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.76 vs. limit=2.0 +2023-04-02 08:28:56,939 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=108809.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:29:01,718 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0456, 3.2831, 1.9570, 1.9699, 2.8488, 1.6712, 1.4150, 2.2221], + device='cuda:0'), covar=tensor([0.1336, 0.0549, 0.1094, 0.0840, 0.0569, 0.1132, 0.0976, 0.0616], + device='cuda:0'), in_proj_covar=tensor([0.0295, 0.0306, 0.0330, 0.0254, 0.0242, 0.0329, 0.0289, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:29:09,787 INFO [train.py:903] (0/4) Epoch 16, batch 6400, loss[loss=0.1962, simple_loss=0.2604, pruned_loss=0.06599, over 19181.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2954, pruned_loss=0.07027, over 3848249.56 frames. ], batch size: 42, lr: 5.11e-03, grad_scale: 8.0 +2023-04-02 08:29:23,218 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.65 vs. limit=2.0 +2023-04-02 08:30:12,289 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8440, 1.9067, 2.0008, 1.8015, 4.3306, 1.1213, 2.5848, 4.7499], + device='cuda:0'), covar=tensor([0.0474, 0.2382, 0.2440, 0.1782, 0.0807, 0.2598, 0.1339, 0.0196], + device='cuda:0'), in_proj_covar=tensor([0.0384, 0.0352, 0.0373, 0.0332, 0.0358, 0.0339, 0.0358, 0.0375], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:30:14,208 INFO [train.py:903] (0/4) Epoch 16, batch 6450, loss[loss=0.2132, simple_loss=0.2856, pruned_loss=0.0704, over 19612.00 frames. ], tot_loss[loss=0.2188, simple_loss=0.2956, pruned_loss=0.07103, over 3820762.94 frames. ], batch size: 50, lr: 5.11e-03, grad_scale: 8.0 +2023-04-02 08:30:52,173 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.276e+02 5.109e+02 6.250e+02 7.655e+02 1.750e+03, threshold=1.250e+03, percent-clipped=4.0 +2023-04-02 08:31:00,312 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 08:31:04,028 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=108909.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:31:18,889 INFO [train.py:903] (0/4) Epoch 16, batch 6500, loss[loss=0.2105, simple_loss=0.2893, pruned_loss=0.06585, over 19667.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2942, pruned_loss=0.07016, over 3817259.21 frames. ], batch size: 53, lr: 5.10e-03, grad_scale: 8.0 +2023-04-02 08:31:24,480 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 08:31:40,947 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=108939.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:32:20,063 INFO [train.py:903] (0/4) Epoch 16, batch 6550, loss[loss=0.2215, simple_loss=0.3035, pruned_loss=0.06973, over 19596.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.2948, pruned_loss=0.07053, over 3812619.83 frames. ], batch size: 61, lr: 5.10e-03, grad_scale: 8.0 +2023-04-02 08:32:53,527 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5992, 1.7068, 1.9350, 2.0923, 1.4761, 1.8952, 2.0010, 1.8237], + device='cuda:0'), covar=tensor([0.3987, 0.3304, 0.1689, 0.2017, 0.3463, 0.1887, 0.4335, 0.2970], + device='cuda:0'), in_proj_covar=tensor([0.0849, 0.0892, 0.0682, 0.0905, 0.0829, 0.0767, 0.0815, 0.0746], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 08:32:58,929 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.314e+02 4.624e+02 5.967e+02 6.821e+02 1.232e+03, threshold=1.193e+03, percent-clipped=0.0 +2023-04-02 08:33:21,265 INFO [train.py:903] (0/4) Epoch 16, batch 6600, loss[loss=0.2551, simple_loss=0.3312, pruned_loss=0.08955, over 19780.00 frames. ], tot_loss[loss=0.2185, simple_loss=0.2956, pruned_loss=0.07072, over 3812890.46 frames. ], batch size: 56, lr: 5.10e-03, grad_scale: 8.0 +2023-04-02 08:33:35,340 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-04-02 08:33:52,840 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8268, 1.9570, 2.1545, 2.5346, 1.7687, 2.3857, 2.2907, 1.9400], + device='cuda:0'), covar=tensor([0.4068, 0.3572, 0.1761, 0.2240, 0.3940, 0.1958, 0.4055, 0.3145], + device='cuda:0'), in_proj_covar=tensor([0.0847, 0.0892, 0.0682, 0.0905, 0.0828, 0.0766, 0.0815, 0.0747], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 08:34:03,956 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109054.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:34:05,054 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109055.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 08:34:08,530 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109058.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:34:22,850 INFO [train.py:903] (0/4) Epoch 16, batch 6650, loss[loss=0.2169, simple_loss=0.3012, pruned_loss=0.06633, over 19597.00 frames. ], tot_loss[loss=0.2192, simple_loss=0.2961, pruned_loss=0.07111, over 3829507.71 frames. ], batch size: 61, lr: 5.10e-03, grad_scale: 8.0 +2023-04-02 08:34:41,470 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109083.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:35:01,853 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.091e+02 5.420e+02 6.464e+02 8.289e+02 2.034e+03, threshold=1.293e+03, percent-clipped=5.0 +2023-04-02 08:35:27,577 INFO [train.py:903] (0/4) Epoch 16, batch 6700, loss[loss=0.23, simple_loss=0.3057, pruned_loss=0.07716, over 19596.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2956, pruned_loss=0.07051, over 3832745.70 frames. ], batch size: 52, lr: 5.10e-03, grad_scale: 8.0 +2023-04-02 08:35:32,578 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6204, 1.4916, 1.4613, 2.1030, 1.5989, 2.0385, 2.0563, 1.7492], + device='cuda:0'), covar=tensor([0.0844, 0.0950, 0.1101, 0.0819, 0.0923, 0.0680, 0.0829, 0.0682], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0220, 0.0224, 0.0243, 0.0225, 0.0206, 0.0188, 0.0200], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 08:36:06,060 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=109153.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:36:19,905 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109165.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:36:25,879 INFO [train.py:903] (0/4) Epoch 16, batch 6750, loss[loss=0.2799, simple_loss=0.3492, pruned_loss=0.1053, over 19680.00 frames. ], tot_loss[loss=0.2169, simple_loss=0.2944, pruned_loss=0.06976, over 3829335.73 frames. ], batch size: 60, lr: 5.10e-03, grad_scale: 8.0 +2023-04-02 08:36:50,437 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109190.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:37:03,548 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.450e+02 4.897e+02 5.778e+02 7.062e+02 1.289e+03, threshold=1.156e+03, percent-clipped=0.0 +2023-04-02 08:37:15,684 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109212.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:37:24,185 INFO [train.py:903] (0/4) Epoch 16, batch 6800, loss[loss=0.193, simple_loss=0.2846, pruned_loss=0.05073, over 19539.00 frames. ], tot_loss[loss=0.218, simple_loss=0.295, pruned_loss=0.07051, over 3813663.07 frames. ], batch size: 56, lr: 5.10e-03, grad_scale: 8.0 +2023-04-02 08:37:45,699 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-02 08:37:53,863 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-16.pt +2023-04-02 08:38:10,021 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 08:38:11,131 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 08:38:13,448 INFO [train.py:903] (0/4) Epoch 17, batch 0, loss[loss=0.2226, simple_loss=0.3032, pruned_loss=0.07098, over 19683.00 frames. ], tot_loss[loss=0.2226, simple_loss=0.3032, pruned_loss=0.07098, over 19683.00 frames. ], batch size: 58, lr: 4.94e-03, grad_scale: 8.0 +2023-04-02 08:38:13,449 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 08:38:24,215 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0017, 1.0347, 1.1997, 1.1405, 1.6043, 1.4863, 1.6585, 0.6652], + device='cuda:0'), covar=tensor([0.1690, 0.2947, 0.1872, 0.1356, 0.1073, 0.1528, 0.0995, 0.3086], + device='cuda:0'), in_proj_covar=tensor([0.0509, 0.0612, 0.0664, 0.0460, 0.0606, 0.0512, 0.0650, 0.0520], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 08:38:26,006 INFO [train.py:937] (0/4) Epoch 17, validation: loss=0.1721, simple_loss=0.2728, pruned_loss=0.03571, over 944034.00 frames. +2023-04-02 08:38:26,007 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 08:38:39,459 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 08:38:51,276 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109268.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:39:29,166 INFO [train.py:903] (0/4) Epoch 17, batch 50, loss[loss=0.2161, simple_loss=0.2988, pruned_loss=0.0667, over 19127.00 frames. ], tot_loss[loss=0.2194, simple_loss=0.2965, pruned_loss=0.07113, over 868811.72 frames. ], batch size: 69, lr: 4.94e-03, grad_scale: 8.0 +2023-04-02 08:39:32,717 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.530e+02 5.319e+02 6.338e+02 7.961e+02 1.981e+03, threshold=1.268e+03, percent-clipped=4.0 +2023-04-02 08:39:40,225 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4152, 1.5022, 1.7746, 1.5921, 2.5454, 2.1632, 2.6507, 1.2032], + device='cuda:0'), covar=tensor([0.2331, 0.4053, 0.2456, 0.1852, 0.1462, 0.2085, 0.1430, 0.4027], + device='cuda:0'), in_proj_covar=tensor([0.0508, 0.0611, 0.0663, 0.0461, 0.0604, 0.0511, 0.0651, 0.0521], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 08:39:43,536 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109310.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:39:45,693 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109312.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:39:59,843 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 08:40:13,464 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109335.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:40:14,438 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109336.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:40:28,820 INFO [train.py:903] (0/4) Epoch 17, batch 100, loss[loss=0.2158, simple_loss=0.2907, pruned_loss=0.07047, over 19677.00 frames. ], tot_loss[loss=0.2159, simple_loss=0.2939, pruned_loss=0.06899, over 1543504.69 frames. ], batch size: 53, lr: 4.94e-03, grad_scale: 8.0 +2023-04-02 08:40:36,778 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 08:41:16,045 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6652, 1.5461, 1.5646, 2.1234, 1.7217, 1.8865, 2.0039, 1.8341], + device='cuda:0'), covar=tensor([0.0863, 0.1016, 0.1097, 0.0852, 0.0907, 0.0840, 0.0969, 0.0698], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0222, 0.0225, 0.0245, 0.0226, 0.0207, 0.0190, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 08:41:29,305 INFO [train.py:903] (0/4) Epoch 17, batch 150, loss[loss=0.2406, simple_loss=0.3127, pruned_loss=0.08431, over 19619.00 frames. ], tot_loss[loss=0.2193, simple_loss=0.2961, pruned_loss=0.07127, over 2036275.29 frames. ], batch size: 57, lr: 4.94e-03, grad_scale: 8.0 +2023-04-02 08:41:30,561 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=109399.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 08:41:32,675 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.150e+02 5.062e+02 6.078e+02 8.331e+02 1.364e+03, threshold=1.216e+03, percent-clipped=3.0 +2023-04-02 08:41:46,300 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3015, 1.5314, 1.7164, 1.5994, 2.9472, 1.2280, 2.3042, 3.2503], + device='cuda:0'), covar=tensor([0.0495, 0.2489, 0.2453, 0.1710, 0.0624, 0.2335, 0.1334, 0.0332], + device='cuda:0'), in_proj_covar=tensor([0.0384, 0.0350, 0.0373, 0.0333, 0.0359, 0.0339, 0.0359, 0.0377], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:42:22,336 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 08:42:29,336 INFO [train.py:903] (0/4) Epoch 17, batch 200, loss[loss=0.1852, simple_loss=0.2687, pruned_loss=0.05086, over 19846.00 frames. ], tot_loss[loss=0.2192, simple_loss=0.2959, pruned_loss=0.0713, over 2418455.44 frames. ], batch size: 52, lr: 4.94e-03, grad_scale: 8.0 +2023-04-02 08:43:32,822 INFO [train.py:903] (0/4) Epoch 17, batch 250, loss[loss=0.1988, simple_loss=0.2833, pruned_loss=0.05711, over 19667.00 frames. ], tot_loss[loss=0.2186, simple_loss=0.2953, pruned_loss=0.07091, over 2740051.96 frames. ], batch size: 55, lr: 4.94e-03, grad_scale: 8.0 +2023-04-02 08:43:36,237 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.403e+02 5.064e+02 6.047e+02 7.271e+02 1.663e+03, threshold=1.209e+03, percent-clipped=2.0 +2023-04-02 08:43:46,913 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.54 vs. limit=5.0 +2023-04-02 08:43:52,410 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109514.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 08:44:04,228 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109524.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:44:16,829 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3290, 1.3617, 1.5246, 1.4628, 2.1921, 1.9733, 2.2809, 0.8873], + device='cuda:0'), covar=tensor([0.2279, 0.3976, 0.2481, 0.1939, 0.1435, 0.2025, 0.1318, 0.4098], + device='cuda:0'), in_proj_covar=tensor([0.0504, 0.0608, 0.0660, 0.0459, 0.0603, 0.0509, 0.0647, 0.0517], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 08:44:35,508 INFO [train.py:903] (0/4) Epoch 17, batch 300, loss[loss=0.2028, simple_loss=0.2875, pruned_loss=0.05903, over 19601.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.2948, pruned_loss=0.07048, over 2990383.26 frames. ], batch size: 57, lr: 4.94e-03, grad_scale: 8.0 +2023-04-02 08:44:37,141 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109549.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:44:45,067 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=109556.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:44:56,598 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7664, 1.6040, 1.5934, 2.3400, 1.8172, 2.0772, 2.2054, 1.8657], + device='cuda:0'), covar=tensor([0.0803, 0.0959, 0.1038, 0.0778, 0.0822, 0.0735, 0.0822, 0.0685], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0220, 0.0223, 0.0244, 0.0226, 0.0207, 0.0189, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 08:45:36,075 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109597.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:45:36,918 INFO [train.py:903] (0/4) Epoch 17, batch 350, loss[loss=0.2228, simple_loss=0.3057, pruned_loss=0.06993, over 19655.00 frames. ], tot_loss[loss=0.2186, simple_loss=0.2959, pruned_loss=0.0706, over 3187685.55 frames. ], batch size: 58, lr: 4.93e-03, grad_scale: 8.0 +2023-04-02 08:45:38,103 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 08:45:40,565 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.272e+02 4.783e+02 5.858e+02 7.548e+02 1.929e+03, threshold=1.172e+03, percent-clipped=3.0 +2023-04-02 08:46:38,846 INFO [train.py:903] (0/4) Epoch 17, batch 400, loss[loss=0.2367, simple_loss=0.3146, pruned_loss=0.07937, over 19325.00 frames. ], tot_loss[loss=0.2193, simple_loss=0.2964, pruned_loss=0.07109, over 3327161.15 frames. ], batch size: 66, lr: 4.93e-03, grad_scale: 8.0 +2023-04-02 08:46:49,030 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=109656.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:47:09,941 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109671.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:47:16,744 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=109677.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:47:17,198 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.51 vs. limit=2.0 +2023-04-02 08:47:20,119 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=109680.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:47:24,872 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1543, 3.5590, 3.7244, 3.7555, 1.6342, 3.5135, 3.0978, 3.4475], + device='cuda:0'), covar=tensor([0.1529, 0.1708, 0.0748, 0.0791, 0.5497, 0.1168, 0.0740, 0.1286], + device='cuda:0'), in_proj_covar=tensor([0.0732, 0.0676, 0.0877, 0.0761, 0.0785, 0.0625, 0.0526, 0.0807], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 08:47:40,321 INFO [train.py:903] (0/4) Epoch 17, batch 450, loss[loss=0.2112, simple_loss=0.2955, pruned_loss=0.06349, over 19373.00 frames. ], tot_loss[loss=0.2184, simple_loss=0.2956, pruned_loss=0.07064, over 3446908.27 frames. ], batch size: 70, lr: 4.93e-03, grad_scale: 8.0 +2023-04-02 08:47:44,685 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.768e+02 5.202e+02 6.539e+02 8.058e+02 1.631e+03, threshold=1.308e+03, percent-clipped=6.0 +2023-04-02 08:48:12,290 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 08:48:12,739 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7164, 1.8174, 1.9901, 2.3655, 1.6877, 2.1920, 2.0909, 1.8437], + device='cuda:0'), covar=tensor([0.3949, 0.3618, 0.1879, 0.2144, 0.3744, 0.2022, 0.4339, 0.3329], + device='cuda:0'), in_proj_covar=tensor([0.0851, 0.0896, 0.0681, 0.0905, 0.0832, 0.0768, 0.0815, 0.0748], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 08:48:13,421 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 08:48:44,999 INFO [train.py:903] (0/4) Epoch 17, batch 500, loss[loss=0.2024, simple_loss=0.2877, pruned_loss=0.05852, over 19527.00 frames. ], tot_loss[loss=0.2196, simple_loss=0.2967, pruned_loss=0.07129, over 3544289.99 frames. ], batch size: 54, lr: 4.93e-03, grad_scale: 8.0 +2023-04-02 08:48:48,151 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-04-02 08:48:51,183 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0199, 1.3018, 1.7893, 1.2135, 2.6676, 3.5899, 3.3138, 3.8276], + device='cuda:0'), covar=tensor([0.1760, 0.3635, 0.3064, 0.2395, 0.0530, 0.0179, 0.0197, 0.0218], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0309, 0.0337, 0.0257, 0.0231, 0.0175, 0.0208, 0.0233], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 08:49:01,157 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-04-02 08:49:11,992 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109770.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 08:49:12,970 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109771.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:49:33,451 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-02 08:49:44,274 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=109795.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:49:44,317 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109795.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 08:49:47,261 INFO [train.py:903] (0/4) Epoch 17, batch 550, loss[loss=0.2707, simple_loss=0.3402, pruned_loss=0.1006, over 19661.00 frames. ], tot_loss[loss=0.2193, simple_loss=0.2967, pruned_loss=0.07099, over 3610648.47 frames. ], batch size: 55, lr: 4.93e-03, grad_scale: 8.0 +2023-04-02 08:49:50,684 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.960e+02 5.345e+02 7.483e+02 9.617e+02 2.288e+03, threshold=1.497e+03, percent-clipped=10.0 +2023-04-02 08:50:48,355 INFO [train.py:903] (0/4) Epoch 17, batch 600, loss[loss=0.2153, simple_loss=0.2953, pruned_loss=0.06763, over 19658.00 frames. ], tot_loss[loss=0.2186, simple_loss=0.2959, pruned_loss=0.07071, over 3667260.13 frames. ], batch size: 60, lr: 4.93e-03, grad_scale: 8.0 +2023-04-02 08:51:27,270 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 08:51:49,456 INFO [train.py:903] (0/4) Epoch 17, batch 650, loss[loss=0.207, simple_loss=0.2757, pruned_loss=0.0691, over 19400.00 frames. ], tot_loss[loss=0.2187, simple_loss=0.296, pruned_loss=0.07075, over 3696192.27 frames. ], batch size: 48, lr: 4.93e-03, grad_scale: 8.0 +2023-04-02 08:51:53,012 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.996e+02 5.642e+02 6.698e+02 8.791e+02 1.815e+03, threshold=1.340e+03, percent-clipped=2.0 +2023-04-02 08:52:21,837 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1502, 1.3361, 1.6057, 1.4178, 2.7755, 0.9999, 2.1555, 3.0031], + device='cuda:0'), covar=tensor([0.0530, 0.2534, 0.2486, 0.1634, 0.0698, 0.2260, 0.1093, 0.0329], + device='cuda:0'), in_proj_covar=tensor([0.0386, 0.0351, 0.0371, 0.0333, 0.0359, 0.0340, 0.0356, 0.0378], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 08:52:27,876 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=109927.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:52:43,910 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=109941.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:52:51,842 INFO [train.py:903] (0/4) Epoch 17, batch 700, loss[loss=0.2466, simple_loss=0.3192, pruned_loss=0.08695, over 19694.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2957, pruned_loss=0.07048, over 3722349.47 frames. ], batch size: 59, lr: 4.93e-03, grad_scale: 8.0 +2023-04-02 08:52:58,938 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=109952.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:53:06,798 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8101, 1.9312, 2.1466, 1.9820, 3.0952, 2.6747, 3.3527, 1.9408], + device='cuda:0'), covar=tensor([0.1986, 0.3439, 0.2287, 0.1621, 0.1414, 0.1790, 0.1440, 0.3308], + device='cuda:0'), in_proj_covar=tensor([0.0506, 0.0611, 0.0663, 0.0458, 0.0603, 0.0510, 0.0647, 0.0516], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 08:53:57,830 INFO [train.py:903] (0/4) Epoch 17, batch 750, loss[loss=0.1747, simple_loss=0.2486, pruned_loss=0.05037, over 19746.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2942, pruned_loss=0.06951, over 3730685.73 frames. ], batch size: 45, lr: 4.93e-03, grad_scale: 8.0 +2023-04-02 08:54:00,304 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-110000.pt +2023-04-02 08:54:02,553 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.920e+02 4.697e+02 5.712e+02 7.217e+02 1.165e+03, threshold=1.142e+03, percent-clipped=0.0 +2023-04-02 08:54:25,895 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=110021.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:54:33,285 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110027.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:54:44,175 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110035.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:54:44,311 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1941, 2.1002, 1.9072, 1.7701, 1.6132, 1.7700, 0.7650, 1.1815], + device='cuda:0'), covar=tensor([0.0538, 0.0495, 0.0384, 0.0595, 0.0861, 0.0700, 0.0943, 0.0805], + device='cuda:0'), in_proj_covar=tensor([0.0351, 0.0349, 0.0347, 0.0371, 0.0448, 0.0378, 0.0326, 0.0333], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 08:55:00,572 INFO [train.py:903] (0/4) Epoch 17, batch 800, loss[loss=0.1918, simple_loss=0.2646, pruned_loss=0.05954, over 19371.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.294, pruned_loss=0.0692, over 3753487.45 frames. ], batch size: 47, lr: 4.92e-03, grad_scale: 8.0 +2023-04-02 08:55:04,524 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110051.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:55:05,675 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110052.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:55:10,549 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110056.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:55:14,915 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 08:55:17,657 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110062.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:55:35,595 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110076.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:55:54,362 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7779, 1.8768, 2.0297, 2.4076, 1.7066, 2.2522, 2.1502, 1.8876], + device='cuda:0'), covar=tensor([0.3878, 0.3422, 0.1757, 0.2039, 0.3578, 0.1847, 0.4434, 0.3152], + device='cuda:0'), in_proj_covar=tensor([0.0848, 0.0897, 0.0680, 0.0903, 0.0830, 0.0764, 0.0811, 0.0746], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 08:56:03,238 INFO [train.py:903] (0/4) Epoch 17, batch 850, loss[loss=0.2147, simple_loss=0.2957, pruned_loss=0.0669, over 19672.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.2947, pruned_loss=0.06932, over 3778529.37 frames. ], batch size: 60, lr: 4.92e-03, grad_scale: 8.0 +2023-04-02 08:56:06,202 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.347e+02 5.213e+02 6.563e+02 8.363e+02 2.159e+03, threshold=1.313e+03, percent-clipped=10.0 +2023-04-02 08:56:42,144 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110128.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:56:47,683 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110133.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:56:51,222 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110136.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 08:56:56,476 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 08:57:04,363 INFO [train.py:903] (0/4) Epoch 17, batch 900, loss[loss=0.1891, simple_loss=0.2648, pruned_loss=0.05674, over 19482.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2954, pruned_loss=0.06962, over 3793634.55 frames. ], batch size: 49, lr: 4.92e-03, grad_scale: 8.0 +2023-04-02 08:58:05,929 INFO [train.py:903] (0/4) Epoch 17, batch 950, loss[loss=0.2386, simple_loss=0.3039, pruned_loss=0.08667, over 19632.00 frames. ], tot_loss[loss=0.2169, simple_loss=0.2949, pruned_loss=0.06946, over 3799242.74 frames. ], batch size: 50, lr: 4.92e-03, grad_scale: 8.0 +2023-04-02 08:58:08,516 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 08:58:09,598 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.667e+02 4.867e+02 6.255e+02 8.229e+02 2.250e+03, threshold=1.251e+03, percent-clipped=5.0 +2023-04-02 08:59:08,986 INFO [train.py:903] (0/4) Epoch 17, batch 1000, loss[loss=0.2049, simple_loss=0.2763, pruned_loss=0.06675, over 19761.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.2958, pruned_loss=0.07003, over 3806091.14 frames. ], batch size: 47, lr: 4.92e-03, grad_scale: 4.0 +2023-04-02 08:59:44,521 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5550, 2.0338, 2.0952, 2.7427, 2.1635, 2.6303, 2.5515, 2.4989], + device='cuda:0'), covar=tensor([0.0606, 0.0836, 0.0865, 0.0859, 0.0837, 0.0659, 0.0789, 0.0563], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0220, 0.0223, 0.0244, 0.0226, 0.0207, 0.0187, 0.0200], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 09:00:05,679 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 09:00:11,636 INFO [train.py:903] (0/4) Epoch 17, batch 1050, loss[loss=0.1876, simple_loss=0.2674, pruned_loss=0.05391, over 19727.00 frames. ], tot_loss[loss=0.2172, simple_loss=0.295, pruned_loss=0.06967, over 3811977.65 frames. ], batch size: 51, lr: 4.92e-03, grad_scale: 4.0 +2023-04-02 09:00:15,422 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110301.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:00:16,233 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.124e+02 4.918e+02 6.228e+02 8.050e+02 1.872e+03, threshold=1.246e+03, percent-clipped=2.0 +2023-04-02 09:00:28,167 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110312.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:00:40,504 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110321.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:00:44,970 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 09:01:00,140 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110336.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:01:01,449 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110337.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:01:13,840 INFO [train.py:903] (0/4) Epoch 17, batch 1100, loss[loss=0.1895, simple_loss=0.2629, pruned_loss=0.0581, over 19764.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2943, pruned_loss=0.06906, over 3817501.53 frames. ], batch size: 47, lr: 4.92e-03, grad_scale: 4.0 +2023-04-02 09:01:36,065 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3780, 1.1001, 1.4583, 1.4409, 2.9223, 1.0862, 2.1728, 3.2862], + device='cuda:0'), covar=tensor([0.0488, 0.2914, 0.2856, 0.1774, 0.0743, 0.2435, 0.1224, 0.0297], + device='cuda:0'), in_proj_covar=tensor([0.0389, 0.0355, 0.0375, 0.0335, 0.0360, 0.0342, 0.0357, 0.0381], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:01:54,278 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=110379.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:02:09,863 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110392.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:02:16,588 INFO [train.py:903] (0/4) Epoch 17, batch 1150, loss[loss=0.1918, simple_loss=0.27, pruned_loss=0.0568, over 19720.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2931, pruned_loss=0.068, over 3823146.06 frames. ], batch size: 51, lr: 4.92e-03, grad_scale: 4.0 +2023-04-02 09:02:21,341 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.021e+02 4.729e+02 5.705e+02 7.310e+02 1.426e+03, threshold=1.141e+03, percent-clipped=3.0 +2023-04-02 09:02:28,343 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=110406.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:02:43,175 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110417.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:03:20,437 INFO [train.py:903] (0/4) Epoch 17, batch 1200, loss[loss=0.2766, simple_loss=0.3371, pruned_loss=0.108, over 19654.00 frames. ], tot_loss[loss=0.216, simple_loss=0.2943, pruned_loss=0.06882, over 3819455.65 frames. ], batch size: 60, lr: 4.92e-03, grad_scale: 8.0 +2023-04-02 09:03:22,075 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.13 vs. limit=5.0 +2023-04-02 09:03:49,893 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=110472.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:03:53,237 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 09:03:55,648 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=110477.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:04:19,551 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110494.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:04:23,816 INFO [train.py:903] (0/4) Epoch 17, batch 1250, loss[loss=0.2503, simple_loss=0.3102, pruned_loss=0.09518, over 13652.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2945, pruned_loss=0.06928, over 3805377.77 frames. ], batch size: 136, lr: 4.91e-03, grad_scale: 8.0 +2023-04-02 09:04:28,266 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.549e+02 5.334e+02 6.614e+02 7.905e+02 1.343e+03, threshold=1.323e+03, percent-clipped=1.0 +2023-04-02 09:04:48,466 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9904, 3.5627, 2.4376, 3.1983, 0.7663, 3.4868, 3.4295, 3.5159], + device='cuda:0'), covar=tensor([0.0884, 0.1355, 0.2182, 0.1015, 0.4318, 0.0854, 0.1032, 0.1373], + device='cuda:0'), in_proj_covar=tensor([0.0471, 0.0385, 0.0465, 0.0331, 0.0389, 0.0402, 0.0400, 0.0429], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:04:50,935 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110521.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:05:25,870 INFO [train.py:903] (0/4) Epoch 17, batch 1300, loss[loss=0.1757, simple_loss=0.2587, pruned_loss=0.04633, over 19756.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2944, pruned_loss=0.06946, over 3799929.44 frames. ], batch size: 51, lr: 4.91e-03, grad_scale: 8.0 +2023-04-02 09:06:02,879 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.56 vs. limit=5.0 +2023-04-02 09:06:14,189 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110587.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:06:20,254 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110592.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:06:26,785 INFO [train.py:903] (0/4) Epoch 17, batch 1350, loss[loss=0.2702, simple_loss=0.3394, pruned_loss=0.1005, over 17279.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2945, pruned_loss=0.06986, over 3795477.63 frames. ], batch size: 101, lr: 4.91e-03, grad_scale: 8.0 +2023-04-02 09:06:31,260 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.362e+02 4.458e+02 6.078e+02 8.226e+02 1.667e+03, threshold=1.216e+03, percent-clipped=3.0 +2023-04-02 09:06:34,173 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2939, 1.4113, 1.8357, 1.5196, 2.7150, 2.1135, 2.8194, 1.2744], + device='cuda:0'), covar=tensor([0.2684, 0.4559, 0.2786, 0.2129, 0.1600, 0.2358, 0.1598, 0.4390], + device='cuda:0'), in_proj_covar=tensor([0.0515, 0.0621, 0.0675, 0.0467, 0.0612, 0.0518, 0.0658, 0.0525], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 09:07:07,131 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110629.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:07:26,706 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=110645.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:07:28,045 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0903, 1.1823, 1.5978, 1.1131, 2.7500, 3.6288, 3.3992, 3.8348], + device='cuda:0'), covar=tensor([0.1727, 0.3913, 0.3453, 0.2496, 0.0572, 0.0180, 0.0201, 0.0253], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0308, 0.0337, 0.0258, 0.0232, 0.0175, 0.0209, 0.0234], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 09:07:29,984 INFO [train.py:903] (0/4) Epoch 17, batch 1400, loss[loss=0.2298, simple_loss=0.304, pruned_loss=0.07781, over 19675.00 frames. ], tot_loss[loss=0.2172, simple_loss=0.2947, pruned_loss=0.06988, over 3794907.86 frames. ], batch size: 60, lr: 4.91e-03, grad_scale: 4.0 +2023-04-02 09:07:51,399 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=110665.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:08:08,767 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=110680.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:08:32,849 INFO [train.py:903] (0/4) Epoch 17, batch 1450, loss[loss=0.2518, simple_loss=0.3216, pruned_loss=0.09102, over 19542.00 frames. ], tot_loss[loss=0.2188, simple_loss=0.2963, pruned_loss=0.07065, over 3806640.97 frames. ], batch size: 56, lr: 4.91e-03, grad_scale: 4.0 +2023-04-02 09:08:32,915 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 09:08:38,606 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.085e+02 5.047e+02 5.721e+02 7.117e+02 1.861e+03, threshold=1.144e+03, percent-clipped=3.0 +2023-04-02 09:09:35,390 INFO [train.py:903] (0/4) Epoch 17, batch 1500, loss[loss=0.2117, simple_loss=0.2898, pruned_loss=0.06678, over 19584.00 frames. ], tot_loss[loss=0.2177, simple_loss=0.2953, pruned_loss=0.07004, over 3795666.44 frames. ], batch size: 52, lr: 4.91e-03, grad_scale: 4.0 +2023-04-02 09:09:38,219 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110750.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:09:42,846 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110754.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:09:50,101 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110760.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:10:11,713 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110775.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:10:14,276 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110777.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:10:16,434 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2445, 3.7680, 3.8899, 3.8971, 1.4748, 3.6763, 3.1948, 3.6106], + device='cuda:0'), covar=tensor([0.1601, 0.0984, 0.0662, 0.0732, 0.5642, 0.0966, 0.0731, 0.1172], + device='cuda:0'), in_proj_covar=tensor([0.0742, 0.0686, 0.0885, 0.0771, 0.0792, 0.0638, 0.0534, 0.0818], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 09:10:17,719 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110780.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:10:27,706 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-04-02 09:10:35,425 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=110795.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:10:38,203 INFO [train.py:903] (0/4) Epoch 17, batch 1550, loss[loss=0.2258, simple_loss=0.3079, pruned_loss=0.07185, over 19623.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.294, pruned_loss=0.06882, over 3807415.26 frames. ], batch size: 57, lr: 4.91e-03, grad_scale: 4.0 +2023-04-02 09:10:43,464 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110802.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:10:44,892 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.880e+02 4.421e+02 5.256e+02 6.667e+02 1.625e+03, threshold=1.051e+03, percent-clipped=2.0 +2023-04-02 09:11:34,882 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110843.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:11:42,178 INFO [train.py:903] (0/4) Epoch 17, batch 1600, loss[loss=0.2005, simple_loss=0.2855, pruned_loss=0.05777, over 19696.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2944, pruned_loss=0.06927, over 3811404.83 frames. ], batch size: 59, lr: 4.91e-03, grad_scale: 8.0 +2023-04-02 09:11:42,572 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=110848.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:12:05,493 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 09:12:05,868 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6118, 1.3921, 1.3371, 1.9293, 1.5444, 1.8880, 1.8570, 1.6631], + device='cuda:0'), covar=tensor([0.0786, 0.0961, 0.1071, 0.0744, 0.0784, 0.0666, 0.0784, 0.0669], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0220, 0.0222, 0.0244, 0.0226, 0.0208, 0.0188, 0.0201], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 09:12:07,163 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110868.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:12:12,585 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=110873.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:12:29,552 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110886.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:12:45,458 INFO [train.py:903] (0/4) Epoch 17, batch 1650, loss[loss=0.184, simple_loss=0.2532, pruned_loss=0.05737, over 19376.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2939, pruned_loss=0.06927, over 3815958.71 frames. ], batch size: 47, lr: 4.91e-03, grad_scale: 8.0 +2023-04-02 09:12:51,300 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.530e+02 5.600e+02 6.791e+02 9.379e+02 3.114e+03, threshold=1.358e+03, percent-clipped=15.0 +2023-04-02 09:13:47,402 INFO [train.py:903] (0/4) Epoch 17, batch 1700, loss[loss=0.1875, simple_loss=0.2715, pruned_loss=0.05169, over 19469.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2941, pruned_loss=0.06904, over 3829301.47 frames. ], batch size: 49, lr: 4.90e-03, grad_scale: 8.0 +2023-04-02 09:14:19,106 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=110973.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:14:24,090 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=110976.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:14:29,659 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 09:14:49,199 INFO [train.py:903] (0/4) Epoch 17, batch 1750, loss[loss=0.2302, simple_loss=0.3088, pruned_loss=0.07578, over 18794.00 frames. ], tot_loss[loss=0.217, simple_loss=0.2945, pruned_loss=0.06976, over 3819548.96 frames. ], batch size: 74, lr: 4.90e-03, grad_scale: 8.0 +2023-04-02 09:14:55,345 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.686e+02 4.870e+02 5.792e+02 7.151e+02 1.845e+03, threshold=1.158e+03, percent-clipped=1.0 +2023-04-02 09:15:14,921 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111016.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:15:37,652 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111036.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:15:43,361 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111041.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:15:51,879 INFO [train.py:903] (0/4) Epoch 17, batch 1800, loss[loss=0.2227, simple_loss=0.3114, pruned_loss=0.06699, over 19673.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2942, pruned_loss=0.06932, over 3826117.54 frames. ], batch size: 58, lr: 4.90e-03, grad_scale: 8.0 +2023-04-02 09:15:56,905 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111051.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:16:10,622 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111061.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:16:28,040 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111076.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:16:44,373 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111088.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:16:51,937 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 09:16:57,623 INFO [train.py:903] (0/4) Epoch 17, batch 1850, loss[loss=0.2385, simple_loss=0.3115, pruned_loss=0.08277, over 19348.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.294, pruned_loss=0.06916, over 3829657.42 frames. ], batch size: 70, lr: 4.90e-03, grad_scale: 8.0 +2023-04-02 09:16:57,833 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=111098.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:17:03,773 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.487e+02 5.241e+02 6.549e+02 7.790e+02 1.838e+03, threshold=1.310e+03, percent-clipped=3.0 +2023-04-02 09:17:29,438 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 09:18:00,981 INFO [train.py:903] (0/4) Epoch 17, batch 1900, loss[loss=0.2405, simple_loss=0.322, pruned_loss=0.07951, over 19749.00 frames. ], tot_loss[loss=0.216, simple_loss=0.294, pruned_loss=0.069, over 3831458.51 frames. ], batch size: 63, lr: 4.90e-03, grad_scale: 8.0 +2023-04-02 09:18:01,482 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5048, 1.6136, 1.8592, 1.7165, 2.6889, 2.3044, 2.7793, 1.2796], + device='cuda:0'), covar=tensor([0.2259, 0.4018, 0.2548, 0.1866, 0.1459, 0.2007, 0.1395, 0.4014], + device='cuda:0'), in_proj_covar=tensor([0.0508, 0.0615, 0.0671, 0.0463, 0.0609, 0.0515, 0.0653, 0.0523], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 09:18:17,324 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 09:18:24,356 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 09:18:34,689 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6092, 2.3575, 1.7164, 1.5523, 2.2128, 1.2843, 1.5041, 2.0214], + device='cuda:0'), covar=tensor([0.1049, 0.0695, 0.0940, 0.0775, 0.0429, 0.1176, 0.0684, 0.0429], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0310, 0.0331, 0.0257, 0.0242, 0.0329, 0.0292, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:18:41,270 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-04-02 09:18:49,751 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 09:19:02,952 INFO [train.py:903] (0/4) Epoch 17, batch 1950, loss[loss=0.194, simple_loss=0.2722, pruned_loss=0.05786, over 19860.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2936, pruned_loss=0.06875, over 3836446.57 frames. ], batch size: 52, lr: 4.90e-03, grad_scale: 8.0 +2023-04-02 09:19:08,706 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.803e+02 5.118e+02 6.155e+02 7.161e+02 1.490e+03, threshold=1.231e+03, percent-clipped=2.0 +2023-04-02 09:19:19,582 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5562, 1.1287, 1.3250, 1.1346, 2.2157, 0.9281, 1.9596, 2.3928], + device='cuda:0'), covar=tensor([0.0718, 0.2760, 0.2851, 0.1718, 0.0851, 0.2173, 0.1075, 0.0502], + device='cuda:0'), in_proj_covar=tensor([0.0389, 0.0352, 0.0374, 0.0335, 0.0361, 0.0342, 0.0358, 0.0382], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:19:23,914 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111213.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:19:33,807 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0667, 3.5058, 2.0445, 2.0051, 3.0855, 1.7444, 1.5311, 2.2266], + device='cuda:0'), covar=tensor([0.1439, 0.0712, 0.0972, 0.0880, 0.0535, 0.1196, 0.0977, 0.0671], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0310, 0.0332, 0.0257, 0.0243, 0.0331, 0.0293, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:19:44,075 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=111230.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:19:54,145 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-04-02 09:19:59,242 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=111243.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:20:05,432 INFO [train.py:903] (0/4) Epoch 17, batch 2000, loss[loss=0.2194, simple_loss=0.3007, pruned_loss=0.06907, over 19616.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.293, pruned_loss=0.06856, over 3838028.23 frames. ], batch size: 57, lr: 4.90e-03, grad_scale: 8.0 +2023-04-02 09:21:04,231 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 09:21:07,735 INFO [train.py:903] (0/4) Epoch 17, batch 2050, loss[loss=0.1976, simple_loss=0.2885, pruned_loss=0.05333, over 19522.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2943, pruned_loss=0.06931, over 3811622.10 frames. ], batch size: 54, lr: 4.90e-03, grad_scale: 8.0 +2023-04-02 09:21:14,880 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.549e+02 5.413e+02 6.604e+02 7.706e+02 1.674e+03, threshold=1.321e+03, percent-clipped=5.0 +2023-04-02 09:21:22,823 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 09:21:24,037 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 09:21:29,059 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8897, 1.6377, 1.5238, 1.9451, 1.6976, 1.6276, 1.5426, 1.7664], + device='cuda:0'), covar=tensor([0.1005, 0.1406, 0.1451, 0.0908, 0.1214, 0.0559, 0.1316, 0.0747], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0356, 0.0302, 0.0245, 0.0299, 0.0249, 0.0297, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:21:34,575 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=111320.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:21:43,892 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 09:21:46,188 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7793, 1.8266, 2.0226, 2.3127, 1.8019, 2.2264, 2.0953, 1.9143], + device='cuda:0'), covar=tensor([0.3317, 0.2828, 0.1364, 0.1584, 0.2880, 0.1374, 0.3293, 0.2421], + device='cuda:0'), in_proj_covar=tensor([0.0854, 0.0905, 0.0683, 0.0910, 0.0834, 0.0769, 0.0817, 0.0752], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 09:22:05,010 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111344.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:22:06,090 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111345.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:22:09,334 INFO [train.py:903] (0/4) Epoch 17, batch 2100, loss[loss=0.2162, simple_loss=0.2994, pruned_loss=0.06654, over 19656.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.2953, pruned_loss=0.06994, over 3789149.94 frames. ], batch size: 55, lr: 4.90e-03, grad_scale: 8.0 +2023-04-02 09:22:34,028 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111369.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:22:38,049 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 09:23:01,371 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 09:23:10,554 INFO [train.py:903] (0/4) Epoch 17, batch 2150, loss[loss=0.1923, simple_loss=0.2707, pruned_loss=0.05695, over 19832.00 frames. ], tot_loss[loss=0.2172, simple_loss=0.2952, pruned_loss=0.06965, over 3795391.28 frames. ], batch size: 52, lr: 4.89e-03, grad_scale: 8.0 +2023-04-02 09:23:16,487 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.230e+02 4.943e+02 5.975e+02 7.359e+02 1.553e+03, threshold=1.195e+03, percent-clipped=3.0 +2023-04-02 09:23:25,735 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5301, 1.0189, 1.2506, 1.2235, 2.2184, 0.9428, 1.8978, 2.4750], + device='cuda:0'), covar=tensor([0.0487, 0.2357, 0.2388, 0.1476, 0.0679, 0.1915, 0.0831, 0.0319], + device='cuda:0'), in_proj_covar=tensor([0.0391, 0.0354, 0.0376, 0.0337, 0.0363, 0.0343, 0.0359, 0.0383], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:23:44,967 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0232, 1.2335, 1.7427, 1.1957, 2.6409, 3.5229, 3.2252, 3.7371], + device='cuda:0'), covar=tensor([0.1733, 0.3776, 0.3195, 0.2378, 0.0535, 0.0175, 0.0229, 0.0219], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0307, 0.0337, 0.0256, 0.0230, 0.0175, 0.0209, 0.0232], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 09:23:56,923 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111435.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:24:11,378 INFO [train.py:903] (0/4) Epoch 17, batch 2200, loss[loss=0.2207, simple_loss=0.3015, pruned_loss=0.06999, over 19698.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2952, pruned_loss=0.06947, over 3803445.94 frames. ], batch size: 59, lr: 4.89e-03, grad_scale: 8.0 +2023-04-02 09:24:38,963 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111469.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:24:48,168 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9871, 2.5435, 2.5086, 2.8389, 2.6793, 2.5310, 2.3536, 2.9941], + device='cuda:0'), covar=tensor([0.0735, 0.1477, 0.1212, 0.1032, 0.1229, 0.0451, 0.1172, 0.0537], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0357, 0.0303, 0.0246, 0.0300, 0.0250, 0.0298, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:24:58,472 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=111486.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:25:09,792 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111494.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:25:15,255 INFO [train.py:903] (0/4) Epoch 17, batch 2250, loss[loss=0.2091, simple_loss=0.2927, pruned_loss=0.06271, over 19531.00 frames. ], tot_loss[loss=0.2168, simple_loss=0.2949, pruned_loss=0.06932, over 3810868.93 frames. ], batch size: 56, lr: 4.89e-03, grad_scale: 8.0 +2023-04-02 09:25:22,029 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.234e+02 5.257e+02 6.827e+02 8.687e+02 2.303e+03, threshold=1.365e+03, percent-clipped=8.0 +2023-04-02 09:25:29,370 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9101, 2.6167, 2.4842, 2.8817, 2.6490, 2.5525, 2.4094, 2.9525], + device='cuda:0'), covar=tensor([0.0811, 0.1460, 0.1311, 0.0999, 0.1299, 0.0461, 0.1174, 0.0554], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0357, 0.0303, 0.0246, 0.0300, 0.0250, 0.0298, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:26:16,855 INFO [train.py:903] (0/4) Epoch 17, batch 2300, loss[loss=0.1948, simple_loss=0.274, pruned_loss=0.05778, over 19844.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2934, pruned_loss=0.0689, over 3814519.36 frames. ], batch size: 52, lr: 4.89e-03, grad_scale: 8.0 +2023-04-02 09:26:27,094 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 09:27:05,554 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=111587.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:27:18,432 INFO [train.py:903] (0/4) Epoch 17, batch 2350, loss[loss=0.2273, simple_loss=0.3051, pruned_loss=0.07477, over 18257.00 frames. ], tot_loss[loss=0.2168, simple_loss=0.2943, pruned_loss=0.0696, over 3812540.40 frames. ], batch size: 83, lr: 4.89e-03, grad_scale: 8.0 +2023-04-02 09:27:22,400 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111601.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:27:24,236 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.974e+02 5.149e+02 5.946e+02 7.803e+02 1.982e+03, threshold=1.189e+03, percent-clipped=4.0 +2023-04-02 09:27:54,454 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111626.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:27:58,784 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 09:28:14,862 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 09:28:19,177 INFO [train.py:903] (0/4) Epoch 17, batch 2400, loss[loss=0.2054, simple_loss=0.2904, pruned_loss=0.0602, over 18299.00 frames. ], tot_loss[loss=0.2175, simple_loss=0.2949, pruned_loss=0.07004, over 3819496.12 frames. ], batch size: 84, lr: 4.89e-03, grad_scale: 8.0 +2023-04-02 09:29:15,420 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111691.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:29:23,213 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.59 vs. limit=5.0 +2023-04-02 09:29:24,901 INFO [train.py:903] (0/4) Epoch 17, batch 2450, loss[loss=0.1872, simple_loss=0.2711, pruned_loss=0.05162, over 19616.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2951, pruned_loss=0.07042, over 3798809.31 frames. ], batch size: 50, lr: 4.89e-03, grad_scale: 4.0 +2023-04-02 09:29:29,897 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111702.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:29:32,592 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.040e+02 4.976e+02 6.292e+02 8.323e+02 1.636e+03, threshold=1.258e+03, percent-clipped=0.0 +2023-04-02 09:29:39,350 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-04-02 09:29:47,054 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111716.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:30:27,985 INFO [train.py:903] (0/4) Epoch 17, batch 2500, loss[loss=0.2139, simple_loss=0.2933, pruned_loss=0.06724, over 19407.00 frames. ], tot_loss[loss=0.2177, simple_loss=0.2951, pruned_loss=0.07018, over 3811942.12 frames. ], batch size: 70, lr: 4.89e-03, grad_scale: 4.0 +2023-04-02 09:31:31,105 INFO [train.py:903] (0/4) Epoch 17, batch 2550, loss[loss=0.2107, simple_loss=0.2782, pruned_loss=0.0716, over 19770.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.294, pruned_loss=0.06961, over 3822559.09 frames. ], batch size: 46, lr: 4.89e-03, grad_scale: 4.0 +2023-04-02 09:31:38,617 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.039e+02 5.268e+02 6.331e+02 7.722e+02 1.710e+03, threshold=1.266e+03, percent-clipped=3.0 +2023-04-02 09:31:44,767 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=111809.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 09:32:13,241 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=111830.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:32:28,446 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 09:32:34,315 INFO [train.py:903] (0/4) Epoch 17, batch 2600, loss[loss=0.1865, simple_loss=0.2583, pruned_loss=0.05732, over 18577.00 frames. ], tot_loss[loss=0.2174, simple_loss=0.2945, pruned_loss=0.07016, over 3821075.96 frames. ], batch size: 41, lr: 4.88e-03, grad_scale: 4.0 +2023-04-02 09:33:38,520 INFO [train.py:903] (0/4) Epoch 17, batch 2650, loss[loss=0.2224, simple_loss=0.2911, pruned_loss=0.07687, over 19786.00 frames. ], tot_loss[loss=0.2169, simple_loss=0.2943, pruned_loss=0.06973, over 3822276.85 frames. ], batch size: 47, lr: 4.88e-03, grad_scale: 4.0 +2023-04-02 09:33:46,341 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.387e+02 5.132e+02 6.430e+02 7.842e+02 1.964e+03, threshold=1.286e+03, percent-clipped=4.0 +2023-04-02 09:33:58,637 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 09:34:38,351 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=111945.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:34:41,295 INFO [train.py:903] (0/4) Epoch 17, batch 2700, loss[loss=0.1743, simple_loss=0.2608, pruned_loss=0.04387, over 19617.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2941, pruned_loss=0.06941, over 3830293.56 frames. ], batch size: 50, lr: 4.88e-03, grad_scale: 4.0 +2023-04-02 09:34:54,247 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=111958.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:35:26,064 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=111983.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:35:43,875 INFO [train.py:903] (0/4) Epoch 17, batch 2750, loss[loss=0.1699, simple_loss=0.2485, pruned_loss=0.0456, over 18744.00 frames. ], tot_loss[loss=0.217, simple_loss=0.2947, pruned_loss=0.06965, over 3830898.23 frames. ], batch size: 41, lr: 4.88e-03, grad_scale: 4.0 +2023-04-02 09:35:46,295 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-112000.pt +2023-04-02 09:35:52,117 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.760e+02 5.468e+02 6.814e+02 8.726e+02 1.544e+03, threshold=1.363e+03, percent-clipped=3.0 +2023-04-02 09:36:45,031 INFO [train.py:903] (0/4) Epoch 17, batch 2800, loss[loss=0.2328, simple_loss=0.3114, pruned_loss=0.07709, over 19391.00 frames. ], tot_loss[loss=0.2182, simple_loss=0.2956, pruned_loss=0.07046, over 3814731.61 frames. ], batch size: 70, lr: 4.88e-03, grad_scale: 8.0 +2023-04-02 09:37:29,646 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112084.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:37:48,060 INFO [train.py:903] (0/4) Epoch 17, batch 2850, loss[loss=0.2006, simple_loss=0.2725, pruned_loss=0.06433, over 19729.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2953, pruned_loss=0.07062, over 3798887.83 frames. ], batch size: 46, lr: 4.88e-03, grad_scale: 8.0 +2023-04-02 09:37:54,816 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.003e+02 5.207e+02 6.661e+02 8.674e+02 1.797e+03, threshold=1.332e+03, percent-clipped=6.0 +2023-04-02 09:37:59,863 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6556, 1.7054, 1.6044, 1.3761, 1.3411, 1.4120, 0.2403, 0.6772], + device='cuda:0'), covar=tensor([0.0574, 0.0575, 0.0366, 0.0579, 0.1164, 0.0731, 0.1170, 0.0984], + device='cuda:0'), in_proj_covar=tensor([0.0350, 0.0349, 0.0347, 0.0373, 0.0449, 0.0379, 0.0323, 0.0334], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 09:38:46,046 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 09:38:49,622 INFO [train.py:903] (0/4) Epoch 17, batch 2900, loss[loss=0.202, simple_loss=0.2845, pruned_loss=0.05975, over 19518.00 frames. ], tot_loss[loss=0.2186, simple_loss=0.2959, pruned_loss=0.07064, over 3800216.77 frames. ], batch size: 54, lr: 4.88e-03, grad_scale: 8.0 +2023-04-02 09:38:56,283 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=112153.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 09:39:47,788 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-02 09:39:51,807 INFO [train.py:903] (0/4) Epoch 17, batch 2950, loss[loss=0.1988, simple_loss=0.2819, pruned_loss=0.05783, over 19669.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2972, pruned_loss=0.07144, over 3802525.14 frames. ], batch size: 55, lr: 4.88e-03, grad_scale: 8.0 +2023-04-02 09:39:55,960 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=112201.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:39:58,767 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.039e+02 4.879e+02 6.137e+02 7.850e+02 1.399e+03, threshold=1.227e+03, percent-clipped=1.0 +2023-04-02 09:40:07,402 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.1501, 5.1050, 5.9404, 5.9385, 2.0544, 5.5944, 4.8538, 5.4886], + device='cuda:0'), covar=tensor([0.1412, 0.0868, 0.0519, 0.0524, 0.5386, 0.0556, 0.0551, 0.1085], + device='cuda:0'), in_proj_covar=tensor([0.0737, 0.0677, 0.0880, 0.0766, 0.0785, 0.0636, 0.0527, 0.0809], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 09:40:27,815 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=112226.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:40:54,420 INFO [train.py:903] (0/4) Epoch 17, batch 3000, loss[loss=0.1987, simple_loss=0.2848, pruned_loss=0.05628, over 19655.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.2956, pruned_loss=0.07011, over 3815429.35 frames. ], batch size: 55, lr: 4.88e-03, grad_scale: 8.0 +2023-04-02 09:40:54,421 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 09:41:09,007 INFO [train.py:937] (0/4) Epoch 17, validation: loss=0.1717, simple_loss=0.272, pruned_loss=0.03576, over 944034.00 frames. +2023-04-02 09:41:09,008 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 09:41:13,731 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 09:41:25,581 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0809, 1.0738, 1.6143, 1.3426, 2.5591, 3.5762, 3.3758, 3.9556], + device='cuda:0'), covar=tensor([0.1890, 0.5035, 0.4230, 0.2384, 0.0701, 0.0219, 0.0255, 0.0266], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0307, 0.0336, 0.0256, 0.0229, 0.0175, 0.0208, 0.0234], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 09:41:33,205 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=112268.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 09:41:46,300 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-02 09:42:09,756 INFO [train.py:903] (0/4) Epoch 17, batch 3050, loss[loss=0.204, simple_loss=0.278, pruned_loss=0.06506, over 19746.00 frames. ], tot_loss[loss=0.217, simple_loss=0.295, pruned_loss=0.06952, over 3834156.21 frames. ], batch size: 47, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:42:16,489 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.397e+02 5.197e+02 6.217e+02 9.038e+02 1.667e+03, threshold=1.243e+03, percent-clipped=8.0 +2023-04-02 09:43:10,131 INFO [train.py:903] (0/4) Epoch 17, batch 3100, loss[loss=0.1747, simple_loss=0.2521, pruned_loss=0.04862, over 19341.00 frames. ], tot_loss[loss=0.2191, simple_loss=0.2967, pruned_loss=0.07073, over 3825198.36 frames. ], batch size: 44, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:43:14,676 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0028, 1.0345, 1.4904, 1.6372, 2.5297, 4.3622, 4.2630, 4.9364], + device='cuda:0'), covar=tensor([0.2137, 0.5516, 0.4798, 0.2430, 0.0848, 0.0242, 0.0208, 0.0168], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0307, 0.0336, 0.0256, 0.0229, 0.0175, 0.0208, 0.0234], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 09:44:14,542 INFO [train.py:903] (0/4) Epoch 17, batch 3150, loss[loss=0.1969, simple_loss=0.2733, pruned_loss=0.06024, over 19731.00 frames. ], tot_loss[loss=0.2186, simple_loss=0.2961, pruned_loss=0.07056, over 3836755.09 frames. ], batch size: 46, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:44:21,812 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.470e+02 5.016e+02 6.190e+02 7.660e+02 1.883e+03, threshold=1.238e+03, percent-clipped=9.0 +2023-04-02 09:44:25,539 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112407.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:44:42,679 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 09:44:51,275 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=112428.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:45:17,099 INFO [train.py:903] (0/4) Epoch 17, batch 3200, loss[loss=0.2205, simple_loss=0.2833, pruned_loss=0.07885, over 19764.00 frames. ], tot_loss[loss=0.2187, simple_loss=0.296, pruned_loss=0.07069, over 3829347.97 frames. ], batch size: 45, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:46:19,039 INFO [train.py:903] (0/4) Epoch 17, batch 3250, loss[loss=0.1726, simple_loss=0.2482, pruned_loss=0.04847, over 19771.00 frames. ], tot_loss[loss=0.2197, simple_loss=0.2968, pruned_loss=0.07127, over 3819631.24 frames. ], batch size: 47, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:46:26,172 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.294e+02 4.958e+02 6.274e+02 7.840e+02 2.025e+03, threshold=1.255e+03, percent-clipped=2.0 +2023-04-02 09:46:51,212 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=112524.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 09:47:13,515 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=112543.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:47:18,737 INFO [train.py:903] (0/4) Epoch 17, batch 3300, loss[loss=0.2266, simple_loss=0.2905, pruned_loss=0.08136, over 19422.00 frames. ], tot_loss[loss=0.22, simple_loss=0.2972, pruned_loss=0.07139, over 3822180.14 frames. ], batch size: 48, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:47:20,334 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=112549.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 09:47:25,352 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 09:48:00,404 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112580.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:48:20,009 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112595.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:48:24,042 INFO [train.py:903] (0/4) Epoch 17, batch 3350, loss[loss=0.2071, simple_loss=0.2849, pruned_loss=0.06463, over 19582.00 frames. ], tot_loss[loss=0.2187, simple_loss=0.2962, pruned_loss=0.07059, over 3832088.37 frames. ], batch size: 52, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:48:31,317 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.087e+02 5.436e+02 6.846e+02 8.612e+02 1.565e+03, threshold=1.369e+03, percent-clipped=5.0 +2023-04-02 09:49:09,900 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112635.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:49:19,914 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8861, 4.4373, 2.5531, 3.9542, 0.9794, 4.2963, 4.2540, 4.3845], + device='cuda:0'), covar=tensor([0.0582, 0.1055, 0.2157, 0.0774, 0.4051, 0.0721, 0.0850, 0.1109], + device='cuda:0'), in_proj_covar=tensor([0.0473, 0.0389, 0.0472, 0.0331, 0.0390, 0.0407, 0.0402, 0.0431], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:49:24,124 INFO [train.py:903] (0/4) Epoch 17, batch 3400, loss[loss=0.1796, simple_loss=0.2675, pruned_loss=0.04586, over 19600.00 frames. ], tot_loss[loss=0.2183, simple_loss=0.2958, pruned_loss=0.07037, over 3840834.46 frames. ], batch size: 52, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:49:57,932 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.44 vs. limit=5.0 +2023-04-02 09:50:25,917 INFO [train.py:903] (0/4) Epoch 17, batch 3450, loss[loss=0.245, simple_loss=0.3254, pruned_loss=0.08228, over 19649.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.2954, pruned_loss=0.07017, over 3822463.14 frames. ], batch size: 58, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:50:28,244 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 09:50:33,005 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.425e+02 4.932e+02 6.092e+02 9.481e+02 2.200e+03, threshold=1.218e+03, percent-clipped=6.0 +2023-04-02 09:51:03,686 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1408, 3.6918, 2.1722, 1.9521, 3.3367, 1.8526, 1.3788, 2.2889], + device='cuda:0'), covar=tensor([0.1338, 0.0472, 0.0935, 0.0878, 0.0556, 0.1177, 0.1014, 0.0627], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0310, 0.0331, 0.0257, 0.0244, 0.0332, 0.0294, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:51:27,357 INFO [train.py:903] (0/4) Epoch 17, batch 3500, loss[loss=0.2004, simple_loss=0.2699, pruned_loss=0.06543, over 18676.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2955, pruned_loss=0.07028, over 3833877.75 frames. ], batch size: 41, lr: 4.87e-03, grad_scale: 8.0 +2023-04-02 09:51:32,006 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=112751.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:51:49,934 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5105, 1.7281, 2.0569, 1.7609, 3.2007, 2.6487, 3.5081, 1.6563], + device='cuda:0'), covar=tensor([0.2310, 0.3946, 0.2534, 0.1794, 0.1448, 0.1942, 0.1531, 0.3771], + device='cuda:0'), in_proj_covar=tensor([0.0508, 0.0610, 0.0667, 0.0463, 0.0605, 0.0514, 0.0646, 0.0521], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 09:52:31,111 INFO [train.py:903] (0/4) Epoch 17, batch 3550, loss[loss=0.2127, simple_loss=0.2853, pruned_loss=0.07002, over 19581.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2949, pruned_loss=0.06983, over 3836143.88 frames. ], batch size: 52, lr: 4.86e-03, grad_scale: 8.0 +2023-04-02 09:52:32,817 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=112799.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:52:38,378 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.345e+02 4.759e+02 5.980e+02 7.566e+02 1.638e+03, threshold=1.196e+03, percent-clipped=2.0 +2023-04-02 09:53:03,288 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=112824.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:53:09,420 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-04-02 09:53:33,310 INFO [train.py:903] (0/4) Epoch 17, batch 3600, loss[loss=0.1942, simple_loss=0.2732, pruned_loss=0.05756, over 19778.00 frames. ], tot_loss[loss=0.2174, simple_loss=0.2951, pruned_loss=0.06988, over 3824703.48 frames. ], batch size: 54, lr: 4.86e-03, grad_scale: 8.0 +2023-04-02 09:53:55,557 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=112866.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:53:56,512 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0091, 3.6039, 2.5331, 3.2199, 0.7525, 3.5615, 3.4479, 3.5462], + device='cuda:0'), covar=tensor([0.0756, 0.1072, 0.1954, 0.0954, 0.4179, 0.0857, 0.0973, 0.1216], + device='cuda:0'), in_proj_covar=tensor([0.0475, 0.0388, 0.0470, 0.0332, 0.0389, 0.0406, 0.0402, 0.0431], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:53:57,977 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8506, 1.3452, 1.0762, 0.9653, 1.1486, 1.0323, 0.8839, 1.2273], + device='cuda:0'), covar=tensor([0.0650, 0.0746, 0.1111, 0.0688, 0.0564, 0.1297, 0.0643, 0.0438], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0312, 0.0333, 0.0258, 0.0245, 0.0334, 0.0296, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:54:35,707 INFO [train.py:903] (0/4) Epoch 17, batch 3650, loss[loss=0.2949, simple_loss=0.3545, pruned_loss=0.1177, over 13541.00 frames. ], tot_loss[loss=0.2177, simple_loss=0.2952, pruned_loss=0.07015, over 3807695.43 frames. ], batch size: 136, lr: 4.86e-03, grad_scale: 8.0 +2023-04-02 09:54:43,570 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.951e+02 4.960e+02 5.826e+02 7.647e+02 1.614e+03, threshold=1.165e+03, percent-clipped=2.0 +2023-04-02 09:55:09,846 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=112924.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:55:28,600 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=112939.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:55:28,743 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=112939.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:55:38,877 INFO [train.py:903] (0/4) Epoch 17, batch 3700, loss[loss=0.208, simple_loss=0.2735, pruned_loss=0.07126, over 19050.00 frames. ], tot_loss[loss=0.2179, simple_loss=0.2949, pruned_loss=0.07049, over 3814341.44 frames. ], batch size: 42, lr: 4.86e-03, grad_scale: 8.0 +2023-04-02 09:56:17,465 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=112979.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:56:42,237 INFO [train.py:903] (0/4) Epoch 17, batch 3750, loss[loss=0.1809, simple_loss=0.2533, pruned_loss=0.05422, over 18668.00 frames. ], tot_loss[loss=0.2169, simple_loss=0.2942, pruned_loss=0.06974, over 3828600.56 frames. ], batch size: 41, lr: 4.86e-03, grad_scale: 8.0 +2023-04-02 09:56:49,245 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.286e+02 4.723e+02 6.001e+02 7.947e+02 1.345e+03, threshold=1.200e+03, percent-clipped=4.0 +2023-04-02 09:57:32,661 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113039.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:57:42,422 INFO [train.py:903] (0/4) Epoch 17, batch 3800, loss[loss=0.2069, simple_loss=0.2931, pruned_loss=0.06033, over 19557.00 frames. ], tot_loss[loss=0.2172, simple_loss=0.2945, pruned_loss=0.06995, over 3824728.44 frames. ], batch size: 56, lr: 4.86e-03, grad_scale: 8.0 +2023-04-02 09:57:49,559 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113054.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:58:14,153 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 09:58:39,380 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113094.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:58:43,325 INFO [train.py:903] (0/4) Epoch 17, batch 3850, loss[loss=0.2293, simple_loss=0.3107, pruned_loss=0.07398, over 19609.00 frames. ], tot_loss[loss=0.2168, simple_loss=0.2943, pruned_loss=0.06966, over 3830155.79 frames. ], batch size: 57, lr: 4.86e-03, grad_scale: 8.0 +2023-04-02 09:58:51,555 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.627e+02 5.316e+02 6.326e+02 9.097e+02 1.552e+03, threshold=1.265e+03, percent-clipped=8.0 +2023-04-02 09:59:08,793 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3522, 2.9962, 2.3170, 2.7378, 0.9620, 2.9651, 2.9100, 3.0025], + device='cuda:0'), covar=tensor([0.1043, 0.1426, 0.1954, 0.1114, 0.3658, 0.1096, 0.1108, 0.1378], + device='cuda:0'), in_proj_covar=tensor([0.0475, 0.0387, 0.0470, 0.0331, 0.0389, 0.0408, 0.0403, 0.0431], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 09:59:14,496 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=113122.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:59:28,704 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6526, 1.4799, 1.4799, 2.3246, 1.8824, 1.8711, 1.9704, 1.7792], + device='cuda:0'), covar=tensor([0.0841, 0.0956, 0.1057, 0.0694, 0.0745, 0.0788, 0.0848, 0.0710], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0221, 0.0222, 0.0242, 0.0226, 0.0208, 0.0188, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 09:59:35,638 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2118, 2.2250, 2.3723, 3.0306, 2.1381, 2.7653, 2.5476, 2.1952], + device='cuda:0'), covar=tensor([0.4135, 0.4020, 0.1884, 0.2477, 0.4495, 0.2170, 0.4259, 0.3308], + device='cuda:0'), in_proj_covar=tensor([0.0857, 0.0908, 0.0688, 0.0913, 0.0838, 0.0774, 0.0817, 0.0754], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 09:59:44,563 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=113147.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 09:59:45,352 INFO [train.py:903] (0/4) Epoch 17, batch 3900, loss[loss=0.2067, simple_loss=0.2837, pruned_loss=0.06483, over 19486.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.294, pruned_loss=0.06928, over 3824233.21 frames. ], batch size: 49, lr: 4.86e-03, grad_scale: 8.0 +2023-04-02 10:00:17,085 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-04-02 10:00:33,600 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.92 vs. limit=5.0 +2023-04-02 10:00:48,752 INFO [train.py:903] (0/4) Epoch 17, batch 3950, loss[loss=0.2055, simple_loss=0.2891, pruned_loss=0.06097, over 19782.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2939, pruned_loss=0.06926, over 3816429.52 frames. ], batch size: 56, lr: 4.86e-03, grad_scale: 8.0 +2023-04-02 10:00:56,114 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 10:00:57,243 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.297e+02 4.545e+02 5.288e+02 6.585e+02 1.560e+03, threshold=1.058e+03, percent-clipped=1.0 +2023-04-02 10:01:05,685 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0742, 1.7503, 1.9629, 2.0196, 4.5394, 1.2724, 2.7585, 4.9252], + device='cuda:0'), covar=tensor([0.0409, 0.2561, 0.2602, 0.1721, 0.0720, 0.2475, 0.1208, 0.0171], + device='cuda:0'), in_proj_covar=tensor([0.0389, 0.0352, 0.0375, 0.0333, 0.0360, 0.0342, 0.0358, 0.0381], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:01:51,454 INFO [train.py:903] (0/4) Epoch 17, batch 4000, loss[loss=0.1861, simple_loss=0.2573, pruned_loss=0.05747, over 19733.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2937, pruned_loss=0.06939, over 3802370.33 frames. ], batch size: 46, lr: 4.85e-03, grad_scale: 8.0 +2023-04-02 10:01:56,547 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=113252.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:02:00,264 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2267, 2.2621, 2.4452, 2.9086, 2.2052, 2.7602, 2.6000, 2.2625], + device='cuda:0'), covar=tensor([0.3822, 0.3349, 0.1639, 0.2145, 0.3621, 0.1817, 0.4015, 0.2980], + device='cuda:0'), in_proj_covar=tensor([0.0860, 0.0910, 0.0691, 0.0916, 0.0839, 0.0775, 0.0818, 0.0757], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 10:02:35,212 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=113283.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:02:39,806 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 10:02:49,612 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=113295.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:02:52,763 INFO [train.py:903] (0/4) Epoch 17, batch 4050, loss[loss=0.2301, simple_loss=0.305, pruned_loss=0.07758, over 19797.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.2943, pruned_loss=0.06915, over 3809977.31 frames. ], batch size: 56, lr: 4.85e-03, grad_scale: 8.0 +2023-04-02 10:03:00,896 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.601e+02 4.703e+02 5.716e+02 7.594e+02 1.568e+03, threshold=1.143e+03, percent-clipped=5.0 +2023-04-02 10:03:08,190 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=113310.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:03:21,566 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=113320.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:03:38,666 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=113335.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:03:54,659 INFO [train.py:903] (0/4) Epoch 17, batch 4100, loss[loss=0.1969, simple_loss=0.269, pruned_loss=0.06244, over 19795.00 frames. ], tot_loss[loss=0.2184, simple_loss=0.2958, pruned_loss=0.07054, over 3785296.24 frames. ], batch size: 48, lr: 4.85e-03, grad_scale: 8.0 +2023-04-02 10:03:57,614 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=113350.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:04:28,652 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=113375.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:04:31,489 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 10:04:56,296 INFO [train.py:903] (0/4) Epoch 17, batch 4150, loss[loss=0.2075, simple_loss=0.2926, pruned_loss=0.0612, over 19598.00 frames. ], tot_loss[loss=0.2176, simple_loss=0.295, pruned_loss=0.07009, over 3794290.39 frames. ], batch size: 61, lr: 4.85e-03, grad_scale: 8.0 +2023-04-02 10:04:56,639 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113398.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:05:03,828 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.156e+02 5.375e+02 6.520e+02 8.152e+02 2.133e+03, threshold=1.304e+03, percent-clipped=6.0 +2023-04-02 10:05:57,675 INFO [train.py:903] (0/4) Epoch 17, batch 4200, loss[loss=0.2502, simple_loss=0.3338, pruned_loss=0.08334, over 19670.00 frames. ], tot_loss[loss=0.2175, simple_loss=0.2949, pruned_loss=0.07005, over 3794476.24 frames. ], batch size: 59, lr: 4.85e-03, grad_scale: 8.0 +2023-04-02 10:06:02,360 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 10:06:15,150 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0977, 1.7719, 2.0710, 1.7226, 4.6324, 1.0944, 2.5663, 4.9945], + device='cuda:0'), covar=tensor([0.0417, 0.2560, 0.2393, 0.1797, 0.0660, 0.2576, 0.1345, 0.0152], + device='cuda:0'), in_proj_covar=tensor([0.0389, 0.0352, 0.0373, 0.0333, 0.0361, 0.0342, 0.0359, 0.0379], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:06:59,519 INFO [train.py:903] (0/4) Epoch 17, batch 4250, loss[loss=0.1818, simple_loss=0.2605, pruned_loss=0.05151, over 19400.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2932, pruned_loss=0.06921, over 3794690.01 frames. ], batch size: 47, lr: 4.85e-03, grad_scale: 8.0 +2023-04-02 10:07:06,456 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.137e+02 4.808e+02 5.898e+02 7.585e+02 1.571e+03, threshold=1.180e+03, percent-clipped=5.0 +2023-04-02 10:07:13,469 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 10:07:24,947 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 10:08:01,305 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7157, 1.7075, 1.6208, 1.3907, 1.3078, 1.4156, 0.2595, 0.6602], + device='cuda:0'), covar=tensor([0.0540, 0.0529, 0.0328, 0.0556, 0.1071, 0.0578, 0.1021, 0.0944], + device='cuda:0'), in_proj_covar=tensor([0.0347, 0.0346, 0.0345, 0.0374, 0.0445, 0.0376, 0.0323, 0.0331], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 10:08:02,055 INFO [train.py:903] (0/4) Epoch 17, batch 4300, loss[loss=0.2168, simple_loss=0.297, pruned_loss=0.06828, over 19670.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2936, pruned_loss=0.06936, over 3802992.45 frames. ], batch size: 55, lr: 4.85e-03, grad_scale: 8.0 +2023-04-02 10:08:55,437 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 10:09:02,226 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=113596.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:09:04,353 INFO [train.py:903] (0/4) Epoch 17, batch 4350, loss[loss=0.2247, simple_loss=0.2936, pruned_loss=0.07792, over 19753.00 frames. ], tot_loss[loss=0.2147, simple_loss=0.2926, pruned_loss=0.0684, over 3819820.64 frames. ], batch size: 51, lr: 4.85e-03, grad_scale: 8.0 +2023-04-02 10:09:12,444 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.419e+02 4.847e+02 6.118e+02 7.738e+02 1.753e+03, threshold=1.224e+03, percent-clipped=4.0 +2023-04-02 10:09:45,306 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 10:09:49,669 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=113635.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:10:07,375 INFO [train.py:903] (0/4) Epoch 17, batch 4400, loss[loss=0.1955, simple_loss=0.2823, pruned_loss=0.05433, over 19788.00 frames. ], tot_loss[loss=0.2146, simple_loss=0.2927, pruned_loss=0.06821, over 3830357.66 frames. ], batch size: 56, lr: 4.85e-03, grad_scale: 8.0 +2023-04-02 10:10:14,892 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=113654.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:10:26,277 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=113664.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:10:33,149 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 10:10:43,126 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 10:10:45,268 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=113679.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:11:07,787 INFO [train.py:903] (0/4) Epoch 17, batch 4450, loss[loss=0.2181, simple_loss=0.2899, pruned_loss=0.07312, over 19600.00 frames. ], tot_loss[loss=0.2154, simple_loss=0.2929, pruned_loss=0.06897, over 3831348.41 frames. ], batch size: 52, lr: 4.84e-03, grad_scale: 16.0 +2023-04-02 10:11:14,458 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.974e+02 5.100e+02 6.811e+02 8.906e+02 1.680e+03, threshold=1.362e+03, percent-clipped=7.0 +2023-04-02 10:11:22,970 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=113711.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:11:27,803 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.73 vs. limit=2.0 +2023-04-02 10:11:51,806 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-04-02 10:12:07,717 INFO [train.py:903] (0/4) Epoch 17, batch 4500, loss[loss=0.2192, simple_loss=0.3046, pruned_loss=0.06687, over 19617.00 frames. ], tot_loss[loss=0.2168, simple_loss=0.2935, pruned_loss=0.07005, over 3816237.57 frames. ], batch size: 57, lr: 4.84e-03, grad_scale: 8.0 +2023-04-02 10:12:41,931 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1172, 2.0356, 1.9028, 1.7346, 1.5201, 1.7126, 0.5504, 1.0966], + device='cuda:0'), covar=tensor([0.0528, 0.0531, 0.0332, 0.0635, 0.1030, 0.0712, 0.1127, 0.0869], + device='cuda:0'), in_proj_covar=tensor([0.0346, 0.0344, 0.0344, 0.0372, 0.0444, 0.0375, 0.0322, 0.0329], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 10:12:59,656 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-04-02 10:13:06,569 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-02 10:13:09,291 INFO [train.py:903] (0/4) Epoch 17, batch 4550, loss[loss=0.1832, simple_loss=0.2619, pruned_loss=0.05223, over 19475.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2946, pruned_loss=0.07001, over 3818250.65 frames. ], batch size: 49, lr: 4.84e-03, grad_scale: 4.0 +2023-04-02 10:13:19,360 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 10:13:20,448 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.619e+02 5.090e+02 6.214e+02 7.749e+02 1.433e+03, threshold=1.243e+03, percent-clipped=2.0 +2023-04-02 10:13:42,454 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 10:14:12,738 INFO [train.py:903] (0/4) Epoch 17, batch 4600, loss[loss=0.2176, simple_loss=0.3054, pruned_loss=0.06486, over 19625.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2942, pruned_loss=0.06952, over 3824422.20 frames. ], batch size: 57, lr: 4.84e-03, grad_scale: 4.0 +2023-04-02 10:15:14,385 INFO [train.py:903] (0/4) Epoch 17, batch 4650, loss[loss=0.2099, simple_loss=0.2978, pruned_loss=0.061, over 19680.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2941, pruned_loss=0.06943, over 3836965.04 frames. ], batch size: 53, lr: 4.84e-03, grad_scale: 4.0 +2023-04-02 10:15:23,609 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.367e+02 5.276e+02 6.482e+02 7.907e+02 1.823e+03, threshold=1.296e+03, percent-clipped=2.0 +2023-04-02 10:15:31,994 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 10:15:44,634 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 10:16:16,816 INFO [train.py:903] (0/4) Epoch 17, batch 4700, loss[loss=0.1911, simple_loss=0.2654, pruned_loss=0.0584, over 18657.00 frames. ], tot_loss[loss=0.2155, simple_loss=0.293, pruned_loss=0.06894, over 3848036.53 frames. ], batch size: 41, lr: 4.84e-03, grad_scale: 4.0 +2023-04-02 10:16:42,988 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=113967.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:16:43,756 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 10:16:56,533 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=113979.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:17:12,176 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=113992.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:17:19,855 INFO [train.py:903] (0/4) Epoch 17, batch 4750, loss[loss=0.2131, simple_loss=0.302, pruned_loss=0.06205, over 19645.00 frames. ], tot_loss[loss=0.2153, simple_loss=0.2931, pruned_loss=0.06873, over 3820583.76 frames. ], batch size: 58, lr: 4.84e-03, grad_scale: 4.0 +2023-04-02 10:17:22,233 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-114000.pt +2023-04-02 10:17:32,729 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.044e+02 4.836e+02 6.122e+02 7.624e+02 1.576e+03, threshold=1.224e+03, percent-clipped=1.0 +2023-04-02 10:17:35,054 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=114008.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:17:47,627 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5627, 4.1231, 2.6292, 3.6136, 0.8967, 3.9629, 3.9107, 3.9852], + device='cuda:0'), covar=tensor([0.0652, 0.1045, 0.1958, 0.0890, 0.4175, 0.0779, 0.0942, 0.1061], + device='cuda:0'), in_proj_covar=tensor([0.0476, 0.0389, 0.0476, 0.0335, 0.0396, 0.0410, 0.0408, 0.0435], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:18:08,355 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7377, 1.2476, 1.4353, 1.4020, 3.3033, 1.0453, 2.3359, 3.6367], + device='cuda:0'), covar=tensor([0.0468, 0.2816, 0.2896, 0.1968, 0.0749, 0.2660, 0.1307, 0.0282], + device='cuda:0'), in_proj_covar=tensor([0.0388, 0.0352, 0.0373, 0.0334, 0.0361, 0.0342, 0.0359, 0.0378], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:18:24,455 INFO [train.py:903] (0/4) Epoch 17, batch 4800, loss[loss=0.2261, simple_loss=0.2988, pruned_loss=0.07671, over 19669.00 frames. ], tot_loss[loss=0.2157, simple_loss=0.2936, pruned_loss=0.06896, over 3815669.29 frames. ], batch size: 55, lr: 4.84e-03, grad_scale: 8.0 +2023-04-02 10:19:22,101 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=114094.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:19:25,638 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2898, 3.8170, 3.9139, 3.8952, 1.5831, 3.7029, 3.2429, 3.6628], + device='cuda:0'), covar=tensor([0.1621, 0.0923, 0.0643, 0.0777, 0.5493, 0.1000, 0.0719, 0.1139], + device='cuda:0'), in_proj_covar=tensor([0.0739, 0.0691, 0.0887, 0.0772, 0.0791, 0.0643, 0.0536, 0.0814], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 10:19:26,563 INFO [train.py:903] (0/4) Epoch 17, batch 4850, loss[loss=0.2054, simple_loss=0.2892, pruned_loss=0.06074, over 19769.00 frames. ], tot_loss[loss=0.2153, simple_loss=0.2932, pruned_loss=0.06869, over 3820149.87 frames. ], batch size: 56, lr: 4.84e-03, grad_scale: 8.0 +2023-04-02 10:19:35,426 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.185e+02 5.130e+02 6.675e+02 8.728e+02 1.864e+03, threshold=1.335e+03, percent-clipped=11.0 +2023-04-02 10:19:52,816 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 10:19:57,417 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=114123.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:20:08,907 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8020, 3.2868, 3.3164, 3.3312, 1.3984, 3.1903, 2.7854, 3.0622], + device='cuda:0'), covar=tensor([0.1694, 0.0928, 0.0798, 0.0852, 0.5155, 0.0976, 0.0782, 0.1289], + device='cuda:0'), in_proj_covar=tensor([0.0742, 0.0694, 0.0892, 0.0777, 0.0798, 0.0646, 0.0539, 0.0819], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 10:20:14,653 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 10:20:19,126 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 10:20:20,346 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 10:20:25,205 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=114145.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:20:28,429 INFO [train.py:903] (0/4) Epoch 17, batch 4900, loss[loss=0.2411, simple_loss=0.3305, pruned_loss=0.07587, over 18170.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2935, pruned_loss=0.06846, over 3831797.40 frames. ], batch size: 83, lr: 4.84e-03, grad_scale: 8.0 +2023-04-02 10:20:28,493 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 10:20:45,331 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9027, 1.9437, 2.1503, 2.4818, 1.7846, 2.3452, 2.2122, 1.9359], + device='cuda:0'), covar=tensor([0.4137, 0.3732, 0.1884, 0.2389, 0.3990, 0.2033, 0.4551, 0.3316], + device='cuda:0'), in_proj_covar=tensor([0.0856, 0.0905, 0.0687, 0.0913, 0.0836, 0.0773, 0.0817, 0.0755], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 10:20:48,144 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 10:21:29,558 INFO [train.py:903] (0/4) Epoch 17, batch 4950, loss[loss=0.1898, simple_loss=0.2628, pruned_loss=0.05835, over 19786.00 frames. ], tot_loss[loss=0.2153, simple_loss=0.2935, pruned_loss=0.0685, over 3829775.09 frames. ], batch size: 48, lr: 4.83e-03, grad_scale: 8.0 +2023-04-02 10:21:41,951 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.237e+02 5.073e+02 6.090e+02 7.599e+02 1.461e+03, threshold=1.218e+03, percent-clipped=1.0 +2023-04-02 10:21:48,501 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 10:22:09,556 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 10:22:31,813 INFO [train.py:903] (0/4) Epoch 17, batch 5000, loss[loss=0.1849, simple_loss=0.2633, pruned_loss=0.05326, over 19806.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.2944, pruned_loss=0.06948, over 3816044.94 frames. ], batch size: 49, lr: 4.83e-03, grad_scale: 8.0 +2023-04-02 10:22:39,599 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 10:22:50,103 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 10:23:32,991 INFO [train.py:903] (0/4) Epoch 17, batch 5050, loss[loss=0.214, simple_loss=0.3003, pruned_loss=0.06389, over 19609.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2941, pruned_loss=0.06947, over 3814735.37 frames. ], batch size: 57, lr: 4.83e-03, grad_scale: 8.0 +2023-04-02 10:23:42,364 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.900e+02 5.068e+02 6.244e+02 7.899e+02 1.430e+03, threshold=1.249e+03, percent-clipped=5.0 +2023-04-02 10:24:10,992 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 10:24:34,964 INFO [train.py:903] (0/4) Epoch 17, batch 5100, loss[loss=0.2153, simple_loss=0.3007, pruned_loss=0.06495, over 19604.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2944, pruned_loss=0.06938, over 3816937.87 frames. ], batch size: 61, lr: 4.83e-03, grad_scale: 8.0 +2023-04-02 10:24:37,659 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=114350.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:24:44,365 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 10:24:46,816 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 10:24:52,437 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 10:25:10,350 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=114375.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:25:14,739 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=114379.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:25:19,724 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1412, 3.7050, 2.1916, 2.1562, 3.3495, 1.9030, 1.4076, 2.2912], + device='cuda:0'), covar=tensor([0.1383, 0.0490, 0.0977, 0.0882, 0.0514, 0.1207, 0.1048, 0.0656], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0309, 0.0329, 0.0256, 0.0245, 0.0328, 0.0292, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:25:36,282 INFO [train.py:903] (0/4) Epoch 17, batch 5150, loss[loss=0.1964, simple_loss=0.2867, pruned_loss=0.05304, over 19777.00 frames. ], tot_loss[loss=0.2166, simple_loss=0.2947, pruned_loss=0.06928, over 3815302.65 frames. ], batch size: 56, lr: 4.83e-03, grad_scale: 8.0 +2023-04-02 10:25:46,417 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=114404.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:25:49,639 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.419e+02 5.288e+02 6.652e+02 7.839e+02 1.735e+03, threshold=1.330e+03, percent-clipped=3.0 +2023-04-02 10:25:50,827 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 10:26:24,429 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 10:26:41,812 INFO [train.py:903] (0/4) Epoch 17, batch 5200, loss[loss=0.2453, simple_loss=0.3198, pruned_loss=0.08536, over 19687.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2936, pruned_loss=0.06882, over 3834787.03 frames. ], batch size: 59, lr: 4.83e-03, grad_scale: 8.0 +2023-04-02 10:26:47,161 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 10:26:54,845 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 10:27:33,107 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=114489.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:27:35,795 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6922, 1.6888, 1.6227, 1.3825, 1.3429, 1.3879, 0.2822, 0.7011], + device='cuda:0'), covar=tensor([0.0594, 0.0583, 0.0345, 0.0591, 0.1054, 0.0684, 0.1136, 0.0984], + device='cuda:0'), in_proj_covar=tensor([0.0349, 0.0346, 0.0345, 0.0374, 0.0448, 0.0379, 0.0325, 0.0333], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 10:27:37,829 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 10:27:43,619 INFO [train.py:903] (0/4) Epoch 17, batch 5250, loss[loss=0.1985, simple_loss=0.2807, pruned_loss=0.05811, over 19700.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2942, pruned_loss=0.06869, over 3841059.82 frames. ], batch size: 53, lr: 4.83e-03, grad_scale: 8.0 +2023-04-02 10:27:53,063 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.560e+02 4.802e+02 5.852e+02 7.465e+02 1.395e+03, threshold=1.170e+03, percent-clipped=1.0 +2023-04-02 10:28:03,704 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7808, 2.6171, 2.2653, 2.6773, 2.6198, 2.2945, 2.1786, 2.8107], + device='cuda:0'), covar=tensor([0.0818, 0.1334, 0.1218, 0.0964, 0.1147, 0.0452, 0.1158, 0.0476], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0352, 0.0299, 0.0243, 0.0297, 0.0246, 0.0293, 0.0245], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:28:26,797 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0 +2023-04-02 10:28:44,586 INFO [train.py:903] (0/4) Epoch 17, batch 5300, loss[loss=0.1947, simple_loss=0.2628, pruned_loss=0.06327, over 19388.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2934, pruned_loss=0.06844, over 3840984.33 frames. ], batch size: 47, lr: 4.83e-03, grad_scale: 8.0 +2023-04-02 10:28:59,045 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 10:29:44,122 INFO [train.py:903] (0/4) Epoch 17, batch 5350, loss[loss=0.2079, simple_loss=0.2918, pruned_loss=0.06206, over 19598.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2939, pruned_loss=0.06924, over 3836146.97 frames. ], batch size: 57, lr: 4.83e-03, grad_scale: 8.0 +2023-04-02 10:29:51,245 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=114604.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:29:54,892 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.137e+02 5.171e+02 6.688e+02 9.089e+02 2.274e+03, threshold=1.338e+03, percent-clipped=9.0 +2023-04-02 10:30:19,030 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 10:30:46,233 INFO [train.py:903] (0/4) Epoch 17, batch 5400, loss[loss=0.1876, simple_loss=0.2691, pruned_loss=0.05308, over 19395.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.2946, pruned_loss=0.06933, over 3827736.99 frames. ], batch size: 48, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:31:00,989 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8667, 4.4174, 2.8250, 3.7616, 1.0749, 4.3522, 4.2180, 4.3818], + device='cuda:0'), covar=tensor([0.0514, 0.0931, 0.1936, 0.0862, 0.3867, 0.0620, 0.0857, 0.1019], + device='cuda:0'), in_proj_covar=tensor([0.0474, 0.0390, 0.0474, 0.0336, 0.0394, 0.0410, 0.0408, 0.0435], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:31:32,773 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-02 10:31:47,161 INFO [train.py:903] (0/4) Epoch 17, batch 5450, loss[loss=0.2227, simple_loss=0.2876, pruned_loss=0.07888, over 19754.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2941, pruned_loss=0.06912, over 3828696.26 frames. ], batch size: 46, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:31:56,194 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.018e+02 4.503e+02 5.761e+02 7.243e+02 1.420e+03, threshold=1.152e+03, percent-clipped=1.0 +2023-04-02 10:32:31,001 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=114734.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:32:47,133 INFO [train.py:903] (0/4) Epoch 17, batch 5500, loss[loss=0.2028, simple_loss=0.2868, pruned_loss=0.05942, over 19776.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2951, pruned_loss=0.06954, over 3812889.49 frames. ], batch size: 56, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:32:48,638 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8170, 1.5927, 1.6406, 1.6191, 3.3578, 1.2052, 2.5318, 3.7915], + device='cuda:0'), covar=tensor([0.0441, 0.2411, 0.2518, 0.1740, 0.0671, 0.2376, 0.1102, 0.0215], + device='cuda:0'), in_proj_covar=tensor([0.0389, 0.0353, 0.0374, 0.0336, 0.0364, 0.0343, 0.0360, 0.0381], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:33:11,651 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 10:33:42,362 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-04-02 10:33:46,748 INFO [train.py:903] (0/4) Epoch 17, batch 5550, loss[loss=0.2083, simple_loss=0.2887, pruned_loss=0.06391, over 19687.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2944, pruned_loss=0.06891, over 3819474.37 frames. ], batch size: 53, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:33:54,779 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 10:33:55,923 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.429e+02 4.950e+02 6.230e+02 7.289e+02 1.704e+03, threshold=1.246e+03, percent-clipped=5.0 +2023-04-02 10:34:41,617 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 10:34:48,920 INFO [train.py:903] (0/4) Epoch 17, batch 5600, loss[loss=0.1744, simple_loss=0.2504, pruned_loss=0.04915, over 19100.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2945, pruned_loss=0.06912, over 3815638.47 frames. ], batch size: 42, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:35:03,675 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=114860.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:35:32,407 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=114885.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:35:39,355 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=114889.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:35:50,182 INFO [train.py:903] (0/4) Epoch 17, batch 5650, loss[loss=0.1649, simple_loss=0.2418, pruned_loss=0.04403, over 15179.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2936, pruned_loss=0.069, over 3798682.62 frames. ], batch size: 33, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:35:59,362 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.231e+02 5.160e+02 6.498e+02 8.575e+02 1.504e+03, threshold=1.300e+03, percent-clipped=5.0 +2023-04-02 10:36:36,133 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 10:36:51,136 INFO [train.py:903] (0/4) Epoch 17, batch 5700, loss[loss=0.2504, simple_loss=0.3203, pruned_loss=0.09025, over 19322.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2951, pruned_loss=0.06979, over 3799393.88 frames. ], batch size: 66, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:37:43,544 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8170, 1.9268, 2.1281, 2.4432, 1.7665, 2.3499, 2.1643, 1.9575], + device='cuda:0'), covar=tensor([0.4175, 0.3766, 0.1824, 0.2330, 0.3994, 0.1990, 0.4613, 0.3215], + device='cuda:0'), in_proj_covar=tensor([0.0860, 0.0912, 0.0690, 0.0919, 0.0838, 0.0777, 0.0815, 0.0761], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 10:37:44,798 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7724, 1.7776, 1.6101, 1.3794, 1.4889, 1.4409, 0.2033, 0.6495], + device='cuda:0'), covar=tensor([0.0490, 0.0539, 0.0352, 0.0549, 0.0972, 0.0646, 0.1027, 0.0951], + device='cuda:0'), in_proj_covar=tensor([0.0348, 0.0346, 0.0344, 0.0375, 0.0447, 0.0380, 0.0325, 0.0333], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 10:37:50,252 INFO [train.py:903] (0/4) Epoch 17, batch 5750, loss[loss=0.1902, simple_loss=0.2764, pruned_loss=0.05197, over 19529.00 frames. ], tot_loss[loss=0.2168, simple_loss=0.2947, pruned_loss=0.06951, over 3801833.85 frames. ], batch size: 56, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:37:50,264 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 10:37:57,221 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 10:37:59,478 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.823e+02 5.221e+02 6.429e+02 7.572e+02 1.818e+03, threshold=1.286e+03, percent-clipped=4.0 +2023-04-02 10:38:04,571 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 10:38:50,579 INFO [train.py:903] (0/4) Epoch 17, batch 5800, loss[loss=0.2313, simple_loss=0.3119, pruned_loss=0.07537, over 19676.00 frames. ], tot_loss[loss=0.2171, simple_loss=0.2948, pruned_loss=0.06974, over 3809052.10 frames. ], batch size: 59, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:39:27,224 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=115078.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:39:52,202 INFO [train.py:903] (0/4) Epoch 17, batch 5850, loss[loss=0.195, simple_loss=0.2811, pruned_loss=0.05446, over 19668.00 frames. ], tot_loss[loss=0.2167, simple_loss=0.2945, pruned_loss=0.06941, over 3810967.64 frames. ], batch size: 58, lr: 4.82e-03, grad_scale: 8.0 +2023-04-02 10:39:59,333 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=115104.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:40:01,415 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.460e+02 4.663e+02 6.050e+02 7.882e+02 1.454e+03, threshold=1.210e+03, percent-clipped=2.0 +2023-04-02 10:40:51,602 INFO [train.py:903] (0/4) Epoch 17, batch 5900, loss[loss=0.215, simple_loss=0.2934, pruned_loss=0.06826, over 17412.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.2941, pruned_loss=0.06923, over 3817385.77 frames. ], batch size: 101, lr: 4.81e-03, grad_scale: 8.0 +2023-04-02 10:40:55,183 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 10:41:03,358 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3346, 3.8617, 2.6239, 3.4902, 1.1999, 3.8491, 3.7447, 3.9261], + device='cuda:0'), covar=tensor([0.0684, 0.1155, 0.1889, 0.0813, 0.3542, 0.0689, 0.0828, 0.0969], + device='cuda:0'), in_proj_covar=tensor([0.0470, 0.0387, 0.0469, 0.0334, 0.0388, 0.0406, 0.0403, 0.0431], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:41:13,994 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 10:41:45,856 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=115193.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:41:51,143 INFO [train.py:903] (0/4) Epoch 17, batch 5950, loss[loss=0.2971, simple_loss=0.3534, pruned_loss=0.1204, over 13440.00 frames. ], tot_loss[loss=0.2169, simple_loss=0.2945, pruned_loss=0.06968, over 3805169.85 frames. ], batch size: 136, lr: 4.81e-03, grad_scale: 8.0 +2023-04-02 10:42:00,462 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.984e+02 4.938e+02 6.318e+02 8.201e+02 2.090e+03, threshold=1.264e+03, percent-clipped=7.0 +2023-04-02 10:42:35,074 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=115233.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:42:51,731 INFO [train.py:903] (0/4) Epoch 17, batch 6000, loss[loss=0.2208, simple_loss=0.286, pruned_loss=0.07781, over 19756.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2939, pruned_loss=0.06913, over 3798466.29 frames. ], batch size: 47, lr: 4.81e-03, grad_scale: 8.0 +2023-04-02 10:42:51,732 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 10:43:04,252 INFO [train.py:937] (0/4) Epoch 17, validation: loss=0.1707, simple_loss=0.2712, pruned_loss=0.03505, over 944034.00 frames. +2023-04-02 10:43:04,253 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 10:44:04,142 INFO [train.py:903] (0/4) Epoch 17, batch 6050, loss[loss=0.1839, simple_loss=0.2516, pruned_loss=0.05809, over 19086.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2937, pruned_loss=0.06899, over 3792500.58 frames. ], batch size: 42, lr: 4.81e-03, grad_scale: 8.0 +2023-04-02 10:44:15,952 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.299e+02 5.156e+02 6.136e+02 7.598e+02 1.906e+03, threshold=1.227e+03, percent-clipped=4.0 +2023-04-02 10:45:06,507 INFO [train.py:903] (0/4) Epoch 17, batch 6100, loss[loss=0.2128, simple_loss=0.3007, pruned_loss=0.06242, over 19544.00 frames. ], tot_loss[loss=0.2147, simple_loss=0.2925, pruned_loss=0.06839, over 3804999.87 frames. ], batch size: 56, lr: 4.81e-03, grad_scale: 8.0 +2023-04-02 10:45:06,872 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=115348.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:45:27,762 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-04-02 10:45:36,308 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9398, 1.6653, 1.5175, 1.8931, 1.5407, 1.5982, 1.4643, 1.8360], + device='cuda:0'), covar=tensor([0.1022, 0.1420, 0.1546, 0.1011, 0.1409, 0.0600, 0.1412, 0.0737], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0353, 0.0302, 0.0245, 0.0297, 0.0247, 0.0295, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:45:54,951 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=115388.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:45:56,132 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=115389.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:46:06,535 INFO [train.py:903] (0/4) Epoch 17, batch 6150, loss[loss=0.1923, simple_loss=0.2853, pruned_loss=0.04971, over 19546.00 frames. ], tot_loss[loss=0.2153, simple_loss=0.2934, pruned_loss=0.06864, over 3807719.46 frames. ], batch size: 54, lr: 4.81e-03, grad_scale: 8.0 +2023-04-02 10:46:06,952 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5909, 2.3142, 1.7965, 1.4475, 2.1165, 1.3927, 1.4257, 1.9405], + device='cuda:0'), covar=tensor([0.0990, 0.0736, 0.0995, 0.0861, 0.0553, 0.1262, 0.0721, 0.0420], + device='cuda:0'), in_proj_covar=tensor([0.0293, 0.0307, 0.0327, 0.0255, 0.0243, 0.0324, 0.0288, 0.0267], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:46:15,597 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.636e+02 5.209e+02 6.440e+02 8.380e+02 1.538e+03, threshold=1.288e+03, percent-clipped=5.0 +2023-04-02 10:46:33,777 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 10:46:35,170 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8102, 3.2669, 3.3241, 3.3183, 1.3059, 3.2228, 2.8214, 3.0970], + device='cuda:0'), covar=tensor([0.1707, 0.1002, 0.0851, 0.0940, 0.5426, 0.0904, 0.0770, 0.1256], + device='cuda:0'), in_proj_covar=tensor([0.0745, 0.0695, 0.0896, 0.0779, 0.0799, 0.0648, 0.0537, 0.0822], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 10:47:07,331 INFO [train.py:903] (0/4) Epoch 17, batch 6200, loss[loss=0.1975, simple_loss=0.2669, pruned_loss=0.06406, over 19336.00 frames. ], tot_loss[loss=0.2154, simple_loss=0.2932, pruned_loss=0.0688, over 3813124.27 frames. ], batch size: 47, lr: 4.81e-03, grad_scale: 8.0 +2023-04-02 10:47:07,474 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=115448.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:47:08,887 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=115449.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:47:39,842 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=115474.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:48:07,509 INFO [train.py:903] (0/4) Epoch 17, batch 6250, loss[loss=0.1791, simple_loss=0.2714, pruned_loss=0.04345, over 19769.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2931, pruned_loss=0.06859, over 3795209.99 frames. ], batch size: 56, lr: 4.81e-03, grad_scale: 8.0 +2023-04-02 10:48:16,568 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.081e+02 4.690e+02 5.769e+02 7.890e+02 2.007e+03, threshold=1.154e+03, percent-clipped=3.0 +2023-04-02 10:48:37,588 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 10:49:09,163 INFO [train.py:903] (0/4) Epoch 17, batch 6300, loss[loss=0.2107, simple_loss=0.2937, pruned_loss=0.06385, over 19654.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2942, pruned_loss=0.06903, over 3805059.69 frames. ], batch size: 60, lr: 4.81e-03, grad_scale: 8.0 +2023-04-02 10:49:27,668 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=115563.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:50:12,480 INFO [train.py:903] (0/4) Epoch 17, batch 6350, loss[loss=0.2157, simple_loss=0.2881, pruned_loss=0.07161, over 19595.00 frames. ], tot_loss[loss=0.2147, simple_loss=0.293, pruned_loss=0.0682, over 3798972.76 frames. ], batch size: 50, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:50:19,286 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-02 10:50:19,959 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=115604.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:50:21,925 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.064e+02 4.833e+02 6.077e+02 8.091e+02 1.466e+03, threshold=1.215e+03, percent-clipped=5.0 +2023-04-02 10:50:50,636 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=115629.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:51:13,881 INFO [train.py:903] (0/4) Epoch 17, batch 6400, loss[loss=0.2532, simple_loss=0.3315, pruned_loss=0.08745, over 19489.00 frames. ], tot_loss[loss=0.2148, simple_loss=0.2936, pruned_loss=0.06802, over 3810483.86 frames. ], batch size: 64, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:52:15,082 INFO [train.py:903] (0/4) Epoch 17, batch 6450, loss[loss=0.2222, simple_loss=0.3008, pruned_loss=0.07185, over 19616.00 frames. ], tot_loss[loss=0.2148, simple_loss=0.2939, pruned_loss=0.06782, over 3827811.88 frames. ], batch size: 50, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:52:25,105 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.211e+02 4.829e+02 5.862e+02 7.962e+02 1.327e+03, threshold=1.172e+03, percent-clipped=3.0 +2023-04-02 10:52:41,080 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1270, 1.2638, 1.6679, 1.0069, 2.3655, 3.0994, 2.7752, 3.2560], + device='cuda:0'), covar=tensor([0.1597, 0.3709, 0.3209, 0.2501, 0.0586, 0.0191, 0.0266, 0.0293], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0313, 0.0341, 0.0260, 0.0235, 0.0179, 0.0211, 0.0237], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 10:52:58,106 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=115732.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:52:59,222 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=115733.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:53:01,407 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 10:53:16,380 INFO [train.py:903] (0/4) Epoch 17, batch 6500, loss[loss=0.1986, simple_loss=0.2753, pruned_loss=0.06094, over 19738.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.2926, pruned_loss=0.06758, over 3834846.66 frames. ], batch size: 51, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:53:24,002 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 10:54:18,483 INFO [train.py:903] (0/4) Epoch 17, batch 6550, loss[loss=0.2677, simple_loss=0.3336, pruned_loss=0.1009, over 19649.00 frames. ], tot_loss[loss=0.2144, simple_loss=0.2927, pruned_loss=0.06801, over 3819907.20 frames. ], batch size: 55, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:54:28,764 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.723e+02 5.150e+02 6.522e+02 8.804e+02 2.234e+03, threshold=1.304e+03, percent-clipped=7.0 +2023-04-02 10:54:43,121 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=115819.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:55:15,220 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=115844.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:55:18,694 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=115847.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:55:19,547 INFO [train.py:903] (0/4) Epoch 17, batch 6600, loss[loss=0.1899, simple_loss=0.2625, pruned_loss=0.05863, over 19762.00 frames. ], tot_loss[loss=0.2132, simple_loss=0.2917, pruned_loss=0.06737, over 3828956.91 frames. ], batch size: 47, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:55:19,934 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=115848.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:56:10,716 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7572, 1.5376, 1.5952, 2.0995, 1.6749, 2.0163, 2.1556, 1.8369], + device='cuda:0'), covar=tensor([0.0809, 0.0928, 0.1023, 0.0830, 0.0864, 0.0736, 0.0776, 0.0679], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0220, 0.0224, 0.0244, 0.0226, 0.0211, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 10:56:19,801 INFO [train.py:903] (0/4) Epoch 17, batch 6650, loss[loss=0.1685, simple_loss=0.2478, pruned_loss=0.0446, over 19366.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2926, pruned_loss=0.06816, over 3837768.15 frames. ], batch size: 47, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:56:27,599 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-04-02 10:56:30,889 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.015e+02 4.833e+02 5.946e+02 8.225e+02 1.682e+03, threshold=1.189e+03, percent-clipped=7.0 +2023-04-02 10:56:34,712 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=115910.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 10:56:43,242 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1993, 1.8253, 1.5147, 1.2438, 1.6555, 1.2069, 1.1800, 1.5998], + device='cuda:0'), covar=tensor([0.0781, 0.0792, 0.0965, 0.0773, 0.0469, 0.1180, 0.0619, 0.0390], + device='cuda:0'), in_proj_covar=tensor([0.0289, 0.0304, 0.0323, 0.0253, 0.0240, 0.0322, 0.0284, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:57:21,959 INFO [train.py:903] (0/4) Epoch 17, batch 6700, loss[loss=0.1956, simple_loss=0.2641, pruned_loss=0.06349, over 19726.00 frames. ], tot_loss[loss=0.2146, simple_loss=0.2925, pruned_loss=0.06837, over 3826107.57 frames. ], batch size: 47, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:58:07,683 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9240, 1.7628, 1.5165, 1.8756, 1.7981, 1.4787, 1.4541, 1.7504], + device='cuda:0'), covar=tensor([0.1112, 0.1680, 0.1727, 0.1088, 0.1462, 0.0804, 0.1631, 0.0904], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0354, 0.0304, 0.0246, 0.0298, 0.0246, 0.0294, 0.0250], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:58:20,387 INFO [train.py:903] (0/4) Epoch 17, batch 6750, loss[loss=0.2684, simple_loss=0.3316, pruned_loss=0.1027, over 13280.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2938, pruned_loss=0.06917, over 3823312.34 frames. ], batch size: 136, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:58:22,771 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-116000.pt +2023-04-02 10:58:31,536 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.004e+02 5.304e+02 6.320e+02 7.514e+02 1.971e+03, threshold=1.264e+03, percent-clipped=7.0 +2023-04-02 10:58:44,259 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7976, 1.4673, 1.4232, 1.7840, 1.3782, 1.5903, 1.4594, 1.6879], + device='cuda:0'), covar=tensor([0.1056, 0.1380, 0.1448, 0.0954, 0.1294, 0.0552, 0.1306, 0.0732], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0351, 0.0301, 0.0244, 0.0295, 0.0244, 0.0292, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 10:58:59,769 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116032.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 10:59:17,212 INFO [train.py:903] (0/4) Epoch 17, batch 6800, loss[loss=0.2163, simple_loss=0.302, pruned_loss=0.06531, over 19537.00 frames. ], tot_loss[loss=0.218, simple_loss=0.2953, pruned_loss=0.07035, over 3805766.14 frames. ], batch size: 56, lr: 4.80e-03, grad_scale: 8.0 +2023-04-02 10:59:42,987 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4805, 1.4653, 1.7027, 1.5880, 2.3586, 2.0078, 2.3521, 1.4624], + device='cuda:0'), covar=tensor([0.1863, 0.3202, 0.2016, 0.1522, 0.1218, 0.1671, 0.1158, 0.3491], + device='cuda:0'), in_proj_covar=tensor([0.0511, 0.0610, 0.0668, 0.0460, 0.0609, 0.0512, 0.0647, 0.0521], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 10:59:48,083 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-17.pt +2023-04-02 11:00:03,644 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 11:00:04,106 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 11:00:07,138 INFO [train.py:903] (0/4) Epoch 18, batch 0, loss[loss=0.2006, simple_loss=0.2889, pruned_loss=0.05619, over 19535.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2889, pruned_loss=0.05619, over 19535.00 frames. ], batch size: 54, lr: 4.66e-03, grad_scale: 8.0 +2023-04-02 11:00:07,139 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 11:00:18,782 INFO [train.py:937] (0/4) Epoch 18, validation: loss=0.1712, simple_loss=0.2722, pruned_loss=0.03505, over 944034.00 frames. +2023-04-02 11:00:18,782 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 11:00:32,358 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 11:00:34,739 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.6935, 5.1999, 3.2400, 4.5145, 1.3241, 5.0838, 5.1045, 5.2039], + device='cuda:0'), covar=tensor([0.0361, 0.0687, 0.1645, 0.0628, 0.3787, 0.0519, 0.0718, 0.1009], + device='cuda:0'), in_proj_covar=tensor([0.0472, 0.0389, 0.0470, 0.0331, 0.0394, 0.0408, 0.0402, 0.0432], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:00:51,639 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116103.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:00:52,826 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116104.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:00:55,571 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.457e+02 4.972e+02 6.494e+02 8.085e+02 1.604e+03, threshold=1.299e+03, percent-clipped=1.0 +2023-04-02 11:01:15,439 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116123.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:01:18,718 INFO [train.py:903] (0/4) Epoch 18, batch 50, loss[loss=0.1668, simple_loss=0.2378, pruned_loss=0.04792, over 15941.00 frames. ], tot_loss[loss=0.215, simple_loss=0.2915, pruned_loss=0.06922, over 859248.24 frames. ], batch size: 35, lr: 4.66e-03, grad_scale: 8.0 +2023-04-02 11:01:21,404 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116128.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:01:23,481 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116129.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:01:30,152 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116134.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:01:52,471 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 11:01:59,224 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5572, 1.2501, 1.4948, 1.2467, 2.2185, 1.0012, 2.0748, 2.4316], + device='cuda:0'), covar=tensor([0.0705, 0.2741, 0.2626, 0.1678, 0.0837, 0.2068, 0.1001, 0.0479], + device='cuda:0'), in_proj_covar=tensor([0.0386, 0.0354, 0.0370, 0.0339, 0.0359, 0.0343, 0.0357, 0.0381], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:02:21,175 INFO [train.py:903] (0/4) Epoch 18, batch 100, loss[loss=0.2128, simple_loss=0.3014, pruned_loss=0.06206, over 19752.00 frames. ], tot_loss[loss=0.2159, simple_loss=0.294, pruned_loss=0.06888, over 1519747.53 frames. ], batch size: 63, lr: 4.66e-03, grad_scale: 8.0 +2023-04-02 11:02:32,224 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 11:02:41,859 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5974, 4.1998, 2.6671, 3.7069, 0.7982, 4.0578, 4.0093, 4.0916], + device='cuda:0'), covar=tensor([0.0600, 0.0933, 0.1902, 0.0772, 0.4175, 0.0647, 0.0834, 0.0920], + device='cuda:0'), in_proj_covar=tensor([0.0472, 0.0390, 0.0472, 0.0332, 0.0395, 0.0408, 0.0403, 0.0433], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:02:58,312 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.917e+02 4.838e+02 6.090e+02 7.458e+02 2.009e+03, threshold=1.218e+03, percent-clipped=2.0 +2023-04-02 11:03:21,607 INFO [train.py:903] (0/4) Epoch 18, batch 150, loss[loss=0.2455, simple_loss=0.3283, pruned_loss=0.08133, over 19453.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2934, pruned_loss=0.06783, over 2037301.20 frames. ], batch size: 64, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:03:56,074 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=116254.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 11:04:20,457 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 11:04:21,608 INFO [train.py:903] (0/4) Epoch 18, batch 200, loss[loss=0.1676, simple_loss=0.2525, pruned_loss=0.04134, over 19766.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.2923, pruned_loss=0.06778, over 2442348.46 frames. ], batch size: 48, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:05:01,397 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.017e+02 4.750e+02 5.613e+02 7.153e+02 1.890e+03, threshold=1.123e+03, percent-clipped=2.0 +2023-04-02 11:05:24,085 INFO [train.py:903] (0/4) Epoch 18, batch 250, loss[loss=0.2309, simple_loss=0.3221, pruned_loss=0.06981, over 19348.00 frames. ], tot_loss[loss=0.2155, simple_loss=0.2936, pruned_loss=0.06871, over 2744642.37 frames. ], batch size: 70, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:05:43,806 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116341.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:05:57,162 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-04-02 11:06:05,938 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8459, 1.1498, 1.5028, 0.6130, 2.0788, 2.3681, 2.0464, 2.4700], + device='cuda:0'), covar=tensor([0.1719, 0.3851, 0.3329, 0.2805, 0.0640, 0.0308, 0.0386, 0.0384], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0312, 0.0340, 0.0258, 0.0232, 0.0176, 0.0210, 0.0237], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 11:06:18,041 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116369.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 11:06:19,021 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2465, 1.7859, 2.0777, 2.9767, 2.0917, 2.3976, 2.4193, 2.3974], + device='cuda:0'), covar=tensor([0.0799, 0.1094, 0.0988, 0.0775, 0.0921, 0.0872, 0.0960, 0.0710], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0220, 0.0223, 0.0240, 0.0225, 0.0209, 0.0186, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 11:06:25,429 INFO [train.py:903] (0/4) Epoch 18, batch 300, loss[loss=0.253, simple_loss=0.3298, pruned_loss=0.08812, over 19624.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2939, pruned_loss=0.0693, over 2995698.32 frames. ], batch size: 57, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:06:25,572 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=116376.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:07:03,942 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.432e+02 5.241e+02 6.705e+02 8.261e+02 1.478e+03, threshold=1.341e+03, percent-clipped=3.0 +2023-04-02 11:07:28,469 INFO [train.py:903] (0/4) Epoch 18, batch 350, loss[loss=0.252, simple_loss=0.3252, pruned_loss=0.08944, over 19310.00 frames. ], tot_loss[loss=0.2155, simple_loss=0.2931, pruned_loss=0.06891, over 3193474.09 frames. ], batch size: 66, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:07:33,277 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 11:07:44,270 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-04-02 11:08:16,131 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116464.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:08:19,403 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=116467.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:08:29,490 INFO [train.py:903] (0/4) Epoch 18, batch 400, loss[loss=0.2116, simple_loss=0.2822, pruned_loss=0.07049, over 19626.00 frames. ], tot_loss[loss=0.216, simple_loss=0.2934, pruned_loss=0.06926, over 3325346.06 frames. ], batch size: 50, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:08:31,842 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=116478.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:08:47,814 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116491.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:09:08,776 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.099e+02 4.874e+02 5.859e+02 7.069e+02 1.370e+03, threshold=1.172e+03, percent-clipped=1.0 +2023-04-02 11:09:27,678 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116523.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:09:31,057 INFO [train.py:903] (0/4) Epoch 18, batch 450, loss[loss=0.1992, simple_loss=0.2766, pruned_loss=0.06095, over 19599.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.294, pruned_loss=0.06928, over 3426617.74 frames. ], batch size: 50, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:10:06,938 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 11:10:08,082 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 11:10:36,073 INFO [train.py:903] (0/4) Epoch 18, batch 500, loss[loss=0.1777, simple_loss=0.2594, pruned_loss=0.04797, over 19752.00 frames. ], tot_loss[loss=0.2154, simple_loss=0.2929, pruned_loss=0.06896, over 3505726.55 frames. ], batch size: 51, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:10:43,256 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116582.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:10:57,197 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116593.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:11:13,492 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.162e+02 5.090e+02 6.291e+02 8.243e+02 1.843e+03, threshold=1.258e+03, percent-clipped=5.0 +2023-04-02 11:11:38,118 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116625.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 11:11:38,806 INFO [train.py:903] (0/4) Epoch 18, batch 550, loss[loss=0.2266, simple_loss=0.2944, pruned_loss=0.07943, over 19436.00 frames. ], tot_loss[loss=0.215, simple_loss=0.293, pruned_loss=0.06849, over 3568979.65 frames. ], batch size: 48, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:11:51,096 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116636.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:12:07,775 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116650.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 11:12:07,824 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2629, 1.3205, 1.6622, 1.4704, 2.2041, 1.9164, 2.1883, 0.9722], + device='cuda:0'), covar=tensor([0.2495, 0.4265, 0.2597, 0.2023, 0.1587, 0.2333, 0.1492, 0.4308], + device='cuda:0'), in_proj_covar=tensor([0.0512, 0.0612, 0.0671, 0.0463, 0.0611, 0.0516, 0.0648, 0.0522], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 11:12:40,996 INFO [train.py:903] (0/4) Epoch 18, batch 600, loss[loss=0.2215, simple_loss=0.3025, pruned_loss=0.07023, over 19669.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2938, pruned_loss=0.06869, over 3635928.76 frames. ], batch size: 55, lr: 4.65e-03, grad_scale: 8.0 +2023-04-02 11:12:51,630 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=116685.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:13:18,872 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.849e+02 4.880e+02 6.214e+02 8.095e+02 1.532e+03, threshold=1.243e+03, percent-clipped=4.0 +2023-04-02 11:13:21,141 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 11:13:28,752 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9565, 1.8502, 1.6187, 2.0631, 1.7641, 1.8076, 1.6123, 1.9353], + device='cuda:0'), covar=tensor([0.1078, 0.1420, 0.1475, 0.0938, 0.1378, 0.0544, 0.1352, 0.0712], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0356, 0.0302, 0.0247, 0.0299, 0.0247, 0.0296, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:13:37,834 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116722.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:13:42,022 INFO [train.py:903] (0/4) Epoch 18, batch 650, loss[loss=0.2869, simple_loss=0.358, pruned_loss=0.1079, over 17401.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2941, pruned_loss=0.06902, over 3683748.88 frames. ], batch size: 101, lr: 4.64e-03, grad_scale: 8.0 +2023-04-02 11:13:55,017 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5568, 1.0768, 1.3103, 1.2710, 2.2025, 0.9324, 1.9556, 2.4312], + device='cuda:0'), covar=tensor([0.0704, 0.2822, 0.2892, 0.1643, 0.0828, 0.2150, 0.1153, 0.0482], + device='cuda:0'), in_proj_covar=tensor([0.0390, 0.0354, 0.0374, 0.0340, 0.0362, 0.0345, 0.0359, 0.0384], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:14:08,712 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116747.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:14:34,415 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.63 vs. limit=5.0 +2023-04-02 11:14:38,772 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116772.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:14:43,011 INFO [train.py:903] (0/4) Epoch 18, batch 700, loss[loss=0.2064, simple_loss=0.289, pruned_loss=0.06186, over 19710.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2943, pruned_loss=0.069, over 3713864.60 frames. ], batch size: 59, lr: 4.64e-03, grad_scale: 8.0 +2023-04-02 11:15:15,694 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116800.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:15:23,661 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.148e+02 4.855e+02 5.777e+02 7.000e+02 1.472e+03, threshold=1.155e+03, percent-clipped=1.0 +2023-04-02 11:15:24,929 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=116808.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:15:27,849 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-02 11:15:39,750 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=116820.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:15:47,509 INFO [train.py:903] (0/4) Epoch 18, batch 750, loss[loss=0.1913, simple_loss=0.2676, pruned_loss=0.05745, over 19735.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2929, pruned_loss=0.06808, over 3751780.96 frames. ], batch size: 46, lr: 4.64e-03, grad_scale: 8.0 +2023-04-02 11:16:03,692 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116838.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:16:16,536 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=116849.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:16:33,983 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116863.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:16:39,876 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=116867.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:16:50,618 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=116874.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:16:52,573 INFO [train.py:903] (0/4) Epoch 18, batch 800, loss[loss=0.2109, simple_loss=0.2972, pruned_loss=0.06226, over 19766.00 frames. ], tot_loss[loss=0.2155, simple_loss=0.2938, pruned_loss=0.06858, over 3760059.88 frames. ], batch size: 56, lr: 4.64e-03, grad_scale: 8.0 +2023-04-02 11:17:06,586 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 11:17:31,391 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.2582, 5.5780, 3.0677, 5.0028, 1.4494, 5.6876, 5.6170, 5.8230], + device='cuda:0'), covar=tensor([0.0362, 0.0984, 0.1943, 0.0656, 0.3726, 0.0540, 0.0708, 0.0821], + device='cuda:0'), in_proj_covar=tensor([0.0472, 0.0388, 0.0470, 0.0331, 0.0390, 0.0407, 0.0401, 0.0431], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:17:32,327 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.570e+02 5.513e+02 6.326e+02 7.669e+02 1.889e+03, threshold=1.265e+03, percent-clipped=5.0 +2023-04-02 11:17:51,775 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116923.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:17:55,027 INFO [train.py:903] (0/4) Epoch 18, batch 850, loss[loss=0.2228, simple_loss=0.3116, pruned_loss=0.067, over 19685.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2946, pruned_loss=0.06921, over 3753629.63 frames. ], batch size: 58, lr: 4.64e-03, grad_scale: 4.0 +2023-04-02 11:18:11,531 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2263, 1.3508, 1.7500, 1.1008, 2.5524, 3.3267, 3.0071, 3.5121], + device='cuda:0'), covar=tensor([0.1542, 0.3691, 0.3170, 0.2493, 0.0559, 0.0225, 0.0229, 0.0257], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0312, 0.0339, 0.0256, 0.0231, 0.0177, 0.0209, 0.0236], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 11:18:48,152 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 11:18:56,388 INFO [train.py:903] (0/4) Epoch 18, batch 900, loss[loss=0.2325, simple_loss=0.315, pruned_loss=0.07507, over 19739.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2943, pruned_loss=0.06938, over 3758148.33 frames. ], batch size: 63, lr: 4.64e-03, grad_scale: 4.0 +2023-04-02 11:19:01,313 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=116980.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:19:03,989 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2289, 1.4046, 1.9100, 1.6769, 3.0713, 4.7570, 4.6771, 5.1094], + device='cuda:0'), covar=tensor([0.1634, 0.3585, 0.3043, 0.2019, 0.0542, 0.0147, 0.0134, 0.0131], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0311, 0.0339, 0.0256, 0.0231, 0.0176, 0.0209, 0.0235], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 11:19:04,001 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=116982.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:19:38,589 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.269e+02 4.671e+02 5.637e+02 7.276e+02 1.422e+03, threshold=1.127e+03, percent-clipped=2.0 +2023-04-02 11:20:00,613 INFO [train.py:903] (0/4) Epoch 18, batch 950, loss[loss=0.2437, simple_loss=0.3158, pruned_loss=0.08584, over 13542.00 frames. ], tot_loss[loss=0.2154, simple_loss=0.2937, pruned_loss=0.06858, over 3774597.63 frames. ], batch size: 136, lr: 4.64e-03, grad_scale: 4.0 +2023-04-02 11:20:02,912 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 11:20:38,595 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=117056.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:20:51,115 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=117066.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:21:03,206 INFO [train.py:903] (0/4) Epoch 18, batch 1000, loss[loss=0.2501, simple_loss=0.3243, pruned_loss=0.08796, over 19599.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2935, pruned_loss=0.06839, over 3794403.49 frames. ], batch size: 57, lr: 4.64e-03, grad_scale: 4.0 +2023-04-02 11:21:11,171 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=117081.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:21:27,305 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117095.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:21:43,056 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.648e+02 4.999e+02 6.181e+02 7.829e+02 2.221e+03, threshold=1.236e+03, percent-clipped=4.0 +2023-04-02 11:21:56,750 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 11:22:07,203 INFO [train.py:903] (0/4) Epoch 18, batch 1050, loss[loss=0.2284, simple_loss=0.303, pruned_loss=0.07688, over 19577.00 frames. ], tot_loss[loss=0.2153, simple_loss=0.2937, pruned_loss=0.06845, over 3806429.57 frames. ], batch size: 52, lr: 4.64e-03, grad_scale: 4.0 +2023-04-02 11:22:40,243 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 11:22:55,009 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=117164.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:23:09,072 INFO [train.py:903] (0/4) Epoch 18, batch 1100, loss[loss=0.2005, simple_loss=0.2816, pruned_loss=0.05963, over 19621.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.2934, pruned_loss=0.06822, over 3813081.12 frames. ], batch size: 50, lr: 4.64e-03, grad_scale: 4.0 +2023-04-02 11:23:11,061 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.58 vs. limit=5.0 +2023-04-02 11:23:13,162 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=117179.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:23:15,431 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117181.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:23:44,494 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=117204.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:23:49,512 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.961e+02 5.150e+02 6.225e+02 7.900e+02 1.283e+03, threshold=1.245e+03, percent-clipped=2.0 +2023-04-02 11:24:11,116 INFO [train.py:903] (0/4) Epoch 18, batch 1150, loss[loss=0.1926, simple_loss=0.2595, pruned_loss=0.06284, over 19774.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2924, pruned_loss=0.06753, over 3829267.44 frames. ], batch size: 48, lr: 4.63e-03, grad_scale: 4.0 +2023-04-02 11:24:27,252 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=117238.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:24:28,265 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=117239.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:24:58,336 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=117263.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:25:06,699 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-04-02 11:25:14,031 INFO [train.py:903] (0/4) Epoch 18, batch 1200, loss[loss=0.1585, simple_loss=0.2399, pruned_loss=0.03862, over 19281.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2929, pruned_loss=0.06786, over 3841102.83 frames. ], batch size: 44, lr: 4.63e-03, grad_scale: 8.0 +2023-04-02 11:25:18,947 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117279.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:25:31,186 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.2998, 5.2411, 6.1190, 6.0762, 1.8127, 5.7887, 4.8554, 5.7665], + device='cuda:0'), covar=tensor([0.1443, 0.0703, 0.0476, 0.0498, 0.6296, 0.0622, 0.0584, 0.0990], + device='cuda:0'), in_proj_covar=tensor([0.0748, 0.0698, 0.0909, 0.0787, 0.0804, 0.0657, 0.0546, 0.0836], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 11:25:36,121 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1862, 1.8116, 1.4584, 1.1702, 1.6242, 1.1303, 1.1900, 1.7312], + device='cuda:0'), covar=tensor([0.0692, 0.0763, 0.1019, 0.0759, 0.0529, 0.1295, 0.0581, 0.0363], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0312, 0.0331, 0.0259, 0.0243, 0.0332, 0.0290, 0.0269], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:25:50,718 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 11:25:54,167 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.294e+02 4.775e+02 5.862e+02 7.562e+02 1.280e+03, threshold=1.172e+03, percent-clipped=1.0 +2023-04-02 11:26:18,174 INFO [train.py:903] (0/4) Epoch 18, batch 1250, loss[loss=0.1697, simple_loss=0.2451, pruned_loss=0.04712, over 19714.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2926, pruned_loss=0.06813, over 3815043.79 frames. ], batch size: 46, lr: 4.63e-03, grad_scale: 8.0 +2023-04-02 11:26:43,736 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=117347.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:26:49,476 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=117351.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:27:20,895 INFO [train.py:903] (0/4) Epoch 18, batch 1300, loss[loss=0.2113, simple_loss=0.2917, pruned_loss=0.06545, over 19671.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2918, pruned_loss=0.06749, over 3817645.39 frames. ], batch size: 58, lr: 4.63e-03, grad_scale: 8.0 +2023-04-02 11:27:21,310 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=117376.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:28:01,377 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.655e+02 5.418e+02 6.594e+02 8.357e+02 1.516e+03, threshold=1.319e+03, percent-clipped=5.0 +2023-04-02 11:28:22,255 INFO [train.py:903] (0/4) Epoch 18, batch 1350, loss[loss=0.1683, simple_loss=0.2534, pruned_loss=0.04161, over 19805.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2908, pruned_loss=0.06711, over 3817861.84 frames. ], batch size: 48, lr: 4.63e-03, grad_scale: 8.0 +2023-04-02 11:28:36,242 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=117437.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:29:07,724 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=117462.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:29:24,520 INFO [train.py:903] (0/4) Epoch 18, batch 1400, loss[loss=0.2072, simple_loss=0.2856, pruned_loss=0.06441, over 19582.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2919, pruned_loss=0.06778, over 3816224.22 frames. ], batch size: 52, lr: 4.63e-03, grad_scale: 8.0 +2023-04-02 11:30:04,549 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.042e+02 5.443e+02 6.741e+02 8.791e+02 2.167e+03, threshold=1.348e+03, percent-clipped=5.0 +2023-04-02 11:30:28,256 INFO [train.py:903] (0/4) Epoch 18, batch 1450, loss[loss=0.1968, simple_loss=0.2838, pruned_loss=0.05487, over 19652.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2919, pruned_loss=0.06752, over 3823987.26 frames. ], batch size: 58, lr: 4.63e-03, grad_scale: 4.0 +2023-04-02 11:30:29,452 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 11:30:40,096 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=117535.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:31:11,204 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=117560.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:31:30,928 INFO [train.py:903] (0/4) Epoch 18, batch 1500, loss[loss=0.2312, simple_loss=0.2973, pruned_loss=0.08251, over 19579.00 frames. ], tot_loss[loss=0.2154, simple_loss=0.2936, pruned_loss=0.06859, over 3822842.93 frames. ], batch size: 52, lr: 4.63e-03, grad_scale: 4.0 +2023-04-02 11:31:39,041 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=117583.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:31:45,169 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3168, 1.3224, 1.7754, 1.2223, 2.6349, 3.5974, 3.2809, 3.7760], + device='cuda:0'), covar=tensor([0.1515, 0.3693, 0.3095, 0.2318, 0.0556, 0.0190, 0.0206, 0.0227], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0309, 0.0339, 0.0257, 0.0231, 0.0178, 0.0209, 0.0236], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 11:32:11,895 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.089e+02 4.657e+02 6.062e+02 7.787e+02 1.498e+03, threshold=1.212e+03, percent-clipped=2.0 +2023-04-02 11:32:32,136 INFO [train.py:903] (0/4) Epoch 18, batch 1550, loss[loss=0.1891, simple_loss=0.263, pruned_loss=0.05757, over 19349.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2935, pruned_loss=0.0684, over 3830576.06 frames. ], batch size: 47, lr: 4.63e-03, grad_scale: 4.0 +2023-04-02 11:33:03,840 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6403, 1.4218, 1.4547, 1.9406, 1.4551, 1.8477, 1.8861, 1.6475], + device='cuda:0'), covar=tensor([0.0814, 0.0954, 0.1042, 0.0754, 0.0861, 0.0715, 0.0855, 0.0703], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0220, 0.0222, 0.0240, 0.0226, 0.0209, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 11:33:34,531 INFO [train.py:903] (0/4) Epoch 18, batch 1600, loss[loss=0.195, simple_loss=0.2682, pruned_loss=0.06093, over 19468.00 frames. ], tot_loss[loss=0.2146, simple_loss=0.2928, pruned_loss=0.06822, over 3833124.68 frames. ], batch size: 49, lr: 4.63e-03, grad_scale: 8.0 +2023-04-02 11:33:54,943 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=117691.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:34:01,918 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 11:34:03,320 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117698.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:34:15,618 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.549e+02 4.964e+02 5.954e+02 7.051e+02 1.393e+03, threshold=1.191e+03, percent-clipped=4.0 +2023-04-02 11:34:37,833 INFO [train.py:903] (0/4) Epoch 18, batch 1650, loss[loss=0.2581, simple_loss=0.3219, pruned_loss=0.09712, over 19828.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2924, pruned_loss=0.06833, over 3831107.90 frames. ], batch size: 52, lr: 4.63e-03, grad_scale: 8.0 +2023-04-02 11:34:50,758 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0103, 2.0404, 2.2562, 2.6520, 1.8836, 2.4602, 2.3145, 2.1077], + device='cuda:0'), covar=tensor([0.4018, 0.3797, 0.1723, 0.2200, 0.3981, 0.1976, 0.4355, 0.3125], + device='cuda:0'), in_proj_covar=tensor([0.0858, 0.0912, 0.0689, 0.0911, 0.0838, 0.0775, 0.0818, 0.0756], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 11:35:36,411 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5670, 1.3866, 1.3950, 2.0873, 1.4558, 1.7468, 1.8560, 1.6345], + device='cuda:0'), covar=tensor([0.0842, 0.0986, 0.1065, 0.0760, 0.0912, 0.0787, 0.0899, 0.0712], + device='cuda:0'), in_proj_covar=tensor([0.0208, 0.0220, 0.0222, 0.0240, 0.0225, 0.0208, 0.0187, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 11:35:39,488 INFO [train.py:903] (0/4) Epoch 18, batch 1700, loss[loss=0.2268, simple_loss=0.3103, pruned_loss=0.07168, over 19622.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.2931, pruned_loss=0.06835, over 3834884.90 frames. ], batch size: 57, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:35:55,765 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5847, 1.1359, 1.4438, 1.2073, 2.2475, 0.9956, 2.0463, 2.4410], + device='cuda:0'), covar=tensor([0.0727, 0.2854, 0.2711, 0.1750, 0.0863, 0.2074, 0.0983, 0.0528], + device='cuda:0'), in_proj_covar=tensor([0.0394, 0.0356, 0.0376, 0.0341, 0.0367, 0.0345, 0.0363, 0.0385], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:36:16,616 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=117806.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:36:19,946 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.954e+02 5.098e+02 6.258e+02 7.253e+02 1.524e+03, threshold=1.252e+03, percent-clipped=2.0 +2023-04-02 11:36:21,131 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 11:36:40,233 INFO [train.py:903] (0/4) Epoch 18, batch 1750, loss[loss=0.2175, simple_loss=0.3046, pruned_loss=0.06523, over 19686.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2939, pruned_loss=0.06918, over 3801613.60 frames. ], batch size: 53, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:37:43,032 INFO [train.py:903] (0/4) Epoch 18, batch 1800, loss[loss=0.2152, simple_loss=0.2916, pruned_loss=0.06935, over 19751.00 frames. ], tot_loss[loss=0.215, simple_loss=0.293, pruned_loss=0.06844, over 3813466.78 frames. ], batch size: 54, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:38:23,380 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.984e+02 4.803e+02 6.041e+02 7.952e+02 1.877e+03, threshold=1.208e+03, percent-clipped=3.0 +2023-04-02 11:38:42,058 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 11:38:45,252 INFO [train.py:903] (0/4) Epoch 18, batch 1850, loss[loss=0.1968, simple_loss=0.2876, pruned_loss=0.05299, over 19304.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.293, pruned_loss=0.0684, over 3819431.43 frames. ], batch size: 70, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:39:18,908 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 11:39:19,282 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=117954.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:39:47,632 INFO [train.py:903] (0/4) Epoch 18, batch 1900, loss[loss=0.1833, simple_loss=0.2758, pruned_loss=0.04544, over 19587.00 frames. ], tot_loss[loss=0.2142, simple_loss=0.2926, pruned_loss=0.06795, over 3826859.96 frames. ], batch size: 57, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:39:51,607 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=117979.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:40:03,248 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 11:40:07,853 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 11:40:15,686 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-118000.pt +2023-04-02 11:40:27,936 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.102e+02 5.120e+02 6.309e+02 7.973e+02 1.539e+03, threshold=1.262e+03, percent-clipped=6.0 +2023-04-02 11:40:34,575 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 11:40:39,489 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.63 vs. limit=2.0 +2023-04-02 11:40:45,014 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6556, 2.3965, 1.8289, 1.6463, 2.2335, 1.5370, 1.3547, 1.9325], + device='cuda:0'), covar=tensor([0.1057, 0.0739, 0.1004, 0.0815, 0.0490, 0.1137, 0.0793, 0.0504], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0314, 0.0335, 0.0263, 0.0246, 0.0335, 0.0293, 0.0271], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:40:48,203 INFO [train.py:903] (0/4) Epoch 18, batch 1950, loss[loss=0.244, simple_loss=0.315, pruned_loss=0.08649, over 19324.00 frames. ], tot_loss[loss=0.2157, simple_loss=0.2938, pruned_loss=0.06883, over 3818287.95 frames. ], batch size: 66, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:41:28,980 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118058.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:41:33,522 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=118062.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:41:49,955 INFO [train.py:903] (0/4) Epoch 18, batch 2000, loss[loss=0.1836, simple_loss=0.2637, pruned_loss=0.05176, over 19760.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2936, pruned_loss=0.06876, over 3815880.20 frames. ], batch size: 46, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:42:05,375 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=118087.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:42:31,795 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.263e+02 5.367e+02 6.632e+02 8.072e+02 1.503e+03, threshold=1.326e+03, percent-clipped=5.0 +2023-04-02 11:42:42,569 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-02 11:42:46,431 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 11:42:54,014 INFO [train.py:903] (0/4) Epoch 18, batch 2050, loss[loss=0.2477, simple_loss=0.3171, pruned_loss=0.0892, over 13528.00 frames. ], tot_loss[loss=0.2144, simple_loss=0.2926, pruned_loss=0.06812, over 3812459.36 frames. ], batch size: 136, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:42:58,526 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0886, 5.4033, 3.2079, 4.7681, 1.1020, 5.5396, 5.3781, 5.5687], + device='cuda:0'), covar=tensor([0.0370, 0.0945, 0.1811, 0.0671, 0.4181, 0.0487, 0.0734, 0.0888], + device='cuda:0'), in_proj_covar=tensor([0.0480, 0.0396, 0.0481, 0.0339, 0.0400, 0.0415, 0.0411, 0.0442], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:43:06,257 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 11:43:07,430 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 11:43:26,112 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 11:43:55,358 INFO [train.py:903] (0/4) Epoch 18, batch 2100, loss[loss=0.2171, simple_loss=0.3055, pruned_loss=0.06437, over 19091.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2934, pruned_loss=0.06847, over 3819841.75 frames. ], batch size: 69, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:44:07,037 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118186.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:44:21,271 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 11:44:29,248 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-04-02 11:44:36,110 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.283e+02 5.186e+02 6.689e+02 8.493e+02 1.656e+03, threshold=1.338e+03, percent-clipped=6.0 +2023-04-02 11:44:44,896 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 11:44:56,300 INFO [train.py:903] (0/4) Epoch 18, batch 2150, loss[loss=0.2175, simple_loss=0.2993, pruned_loss=0.06788, over 19610.00 frames. ], tot_loss[loss=0.214, simple_loss=0.2923, pruned_loss=0.06783, over 3827420.05 frames. ], batch size: 57, lr: 4.62e-03, grad_scale: 8.0 +2023-04-02 11:45:57,805 INFO [train.py:903] (0/4) Epoch 18, batch 2200, loss[loss=0.2281, simple_loss=0.3134, pruned_loss=0.07141, over 19741.00 frames. ], tot_loss[loss=0.2146, simple_loss=0.2929, pruned_loss=0.06814, over 3836358.81 frames. ], batch size: 63, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:46:13,129 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118287.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:46:37,320 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118307.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:46:39,293 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.599e+02 5.366e+02 7.280e+02 9.450e+02 2.114e+03, threshold=1.456e+03, percent-clipped=6.0 +2023-04-02 11:47:01,197 INFO [train.py:903] (0/4) Epoch 18, batch 2250, loss[loss=0.2096, simple_loss=0.2937, pruned_loss=0.06275, over 19572.00 frames. ], tot_loss[loss=0.216, simple_loss=0.294, pruned_loss=0.06902, over 3822115.67 frames. ], batch size: 52, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:47:21,583 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9186, 2.7506, 2.0901, 2.1321, 1.8750, 2.4128, 1.1027, 2.0448], + device='cuda:0'), covar=tensor([0.0576, 0.0582, 0.0669, 0.1020, 0.1090, 0.0968, 0.1218, 0.0897], + device='cuda:0'), in_proj_covar=tensor([0.0351, 0.0347, 0.0346, 0.0373, 0.0448, 0.0380, 0.0327, 0.0334], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 11:47:27,273 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118347.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:47:54,144 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-02 11:48:02,754 INFO [train.py:903] (0/4) Epoch 18, batch 2300, loss[loss=0.1835, simple_loss=0.2648, pruned_loss=0.0511, over 19752.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2947, pruned_loss=0.06909, over 3824270.17 frames. ], batch size: 51, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:48:15,073 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 11:48:34,716 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=118402.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:48:44,940 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.375e+02 5.253e+02 6.243e+02 7.561e+02 1.558e+03, threshold=1.249e+03, percent-clipped=2.0 +2023-04-02 11:49:05,628 INFO [train.py:903] (0/4) Epoch 18, batch 2350, loss[loss=0.2035, simple_loss=0.2874, pruned_loss=0.05983, over 19667.00 frames. ], tot_loss[loss=0.2154, simple_loss=0.2941, pruned_loss=0.06834, over 3822596.68 frames. ], batch size: 59, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:49:12,760 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1376, 1.3232, 1.6922, 1.1684, 2.5400, 3.3838, 3.0478, 3.5659], + device='cuda:0'), covar=tensor([0.1610, 0.3647, 0.3188, 0.2446, 0.0543, 0.0194, 0.0232, 0.0244], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0311, 0.0341, 0.0259, 0.0233, 0.0179, 0.0211, 0.0238], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 11:49:46,252 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 11:50:02,111 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 11:50:06,554 INFO [train.py:903] (0/4) Epoch 18, batch 2400, loss[loss=0.2069, simple_loss=0.293, pruned_loss=0.06038, over 19368.00 frames. ], tot_loss[loss=0.2146, simple_loss=0.2931, pruned_loss=0.06802, over 3833852.80 frames. ], batch size: 70, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:50:06,903 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3749, 2.2635, 2.1998, 2.5210, 2.2742, 2.0052, 2.1963, 2.3626], + device='cuda:0'), covar=tensor([0.0776, 0.1220, 0.1060, 0.0779, 0.1071, 0.0455, 0.0970, 0.0542], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0354, 0.0303, 0.0250, 0.0300, 0.0248, 0.0297, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:50:48,686 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.320e+02 4.821e+02 5.593e+02 7.730e+02 1.750e+03, threshold=1.119e+03, percent-clipped=3.0 +2023-04-02 11:50:58,486 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118517.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:51:10,680 INFO [train.py:903] (0/4) Epoch 18, batch 2450, loss[loss=0.2607, simple_loss=0.3288, pruned_loss=0.09629, over 13552.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.2932, pruned_loss=0.06824, over 3811473.42 frames. ], batch size: 135, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:51:16,268 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=118530.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:52:12,594 INFO [train.py:903] (0/4) Epoch 18, batch 2500, loss[loss=0.2616, simple_loss=0.3305, pruned_loss=0.09632, over 18979.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.294, pruned_loss=0.06882, over 3792194.57 frames. ], batch size: 75, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:52:53,545 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.320e+02 5.081e+02 6.040e+02 7.266e+02 1.380e+03, threshold=1.208e+03, percent-clipped=1.0 +2023-04-02 11:53:13,676 INFO [train.py:903] (0/4) Epoch 18, batch 2550, loss[loss=0.2735, simple_loss=0.3387, pruned_loss=0.1041, over 13162.00 frames. ], tot_loss[loss=0.2161, simple_loss=0.2942, pruned_loss=0.06895, over 3810612.74 frames. ], batch size: 136, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:53:19,800 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=118631.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:53:37,329 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.52 vs. limit=5.0 +2023-04-02 11:53:37,917 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118645.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:53:42,634 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=118649.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:53:45,542 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=118651.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:54:06,487 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 11:54:15,893 INFO [train.py:903] (0/4) Epoch 18, batch 2600, loss[loss=0.2056, simple_loss=0.2862, pruned_loss=0.06247, over 19729.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.292, pruned_loss=0.06778, over 3826728.75 frames. ], batch size: 51, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:54:34,614 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=118691.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:54:56,095 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.395e+02 4.742e+02 5.749e+02 7.114e+02 1.367e+03, threshold=1.150e+03, percent-clipped=3.0 +2023-04-02 11:55:16,574 INFO [train.py:903] (0/4) Epoch 18, batch 2650, loss[loss=0.2126, simple_loss=0.277, pruned_loss=0.07408, over 19081.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.2922, pruned_loss=0.06802, over 3829202.29 frames. ], batch size: 42, lr: 4.61e-03, grad_scale: 8.0 +2023-04-02 11:55:33,739 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 11:55:42,225 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118746.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:56:06,112 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118766.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:56:15,292 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=118773.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:56:18,246 INFO [train.py:903] (0/4) Epoch 18, batch 2700, loss[loss=0.1958, simple_loss=0.2776, pruned_loss=0.05697, over 19597.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2927, pruned_loss=0.06817, over 3827023.36 frames. ], batch size: 52, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 11:56:44,770 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=118798.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:56:49,042 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.87 vs. limit=5.0 +2023-04-02 11:56:55,783 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=118806.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:56:59,036 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.246e+02 5.095e+02 6.154e+02 8.195e+02 1.746e+03, threshold=1.231e+03, percent-clipped=5.0 +2023-04-02 11:57:04,011 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.9955, 5.3680, 3.2680, 4.7253, 0.8144, 5.4863, 5.3765, 5.4735], + device='cuda:0'), covar=tensor([0.0385, 0.0803, 0.1590, 0.0652, 0.4399, 0.0494, 0.0703, 0.0894], + device='cuda:0'), in_proj_covar=tensor([0.0483, 0.0394, 0.0480, 0.0341, 0.0402, 0.0419, 0.0413, 0.0445], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:57:06,542 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0244, 2.2096, 1.5742, 2.1257, 2.3542, 1.4957, 1.6742, 1.9519], + device='cuda:0'), covar=tensor([0.1245, 0.1650, 0.1984, 0.1259, 0.1406, 0.1146, 0.1868, 0.1046], + device='cuda:0'), in_proj_covar=tensor([0.0263, 0.0353, 0.0299, 0.0248, 0.0298, 0.0246, 0.0295, 0.0250], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:57:20,475 INFO [train.py:903] (0/4) Epoch 18, batch 2750, loss[loss=0.1841, simple_loss=0.2538, pruned_loss=0.0572, over 19719.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2919, pruned_loss=0.06776, over 3830977.54 frames. ], batch size: 46, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 11:58:15,357 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4361, 1.4389, 1.7466, 1.6453, 2.5505, 2.2210, 2.6687, 1.1210], + device='cuda:0'), covar=tensor([0.2573, 0.4531, 0.2781, 0.2057, 0.1656, 0.2288, 0.1612, 0.4518], + device='cuda:0'), in_proj_covar=tensor([0.0519, 0.0622, 0.0679, 0.0468, 0.0615, 0.0521, 0.0655, 0.0530], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 11:58:23,106 INFO [train.py:903] (0/4) Epoch 18, batch 2800, loss[loss=0.2238, simple_loss=0.3046, pruned_loss=0.07147, over 19676.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2917, pruned_loss=0.06782, over 3829345.09 frames. ], batch size: 58, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 11:58:41,697 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2482, 2.1062, 1.9783, 1.8298, 1.5776, 1.8219, 0.6521, 1.1730], + device='cuda:0'), covar=tensor([0.0555, 0.0592, 0.0432, 0.0705, 0.1245, 0.0828, 0.1204, 0.0958], + device='cuda:0'), in_proj_covar=tensor([0.0349, 0.0348, 0.0347, 0.0374, 0.0448, 0.0380, 0.0329, 0.0335], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 11:58:44,011 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8276, 1.5202, 1.7139, 1.6941, 4.3901, 1.1656, 2.6050, 4.7232], + device='cuda:0'), covar=tensor([0.0455, 0.2744, 0.2856, 0.1984, 0.0693, 0.2591, 0.1399, 0.0170], + device='cuda:0'), in_proj_covar=tensor([0.0399, 0.0357, 0.0378, 0.0342, 0.0365, 0.0345, 0.0366, 0.0389], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 11:58:54,266 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=118901.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 11:59:03,986 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.182e+02 4.804e+02 6.148e+02 8.494e+02 1.418e+03, threshold=1.230e+03, percent-clipped=5.0 +2023-04-02 11:59:24,485 INFO [train.py:903] (0/4) Epoch 18, batch 2850, loss[loss=0.2269, simple_loss=0.307, pruned_loss=0.07342, over 19778.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2921, pruned_loss=0.06821, over 3811337.07 frames. ], batch size: 56, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 11:59:24,913 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=118926.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:00:22,862 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 12:00:26,012 INFO [train.py:903] (0/4) Epoch 18, batch 2900, loss[loss=0.2018, simple_loss=0.2663, pruned_loss=0.06867, over 19764.00 frames. ], tot_loss[loss=0.2142, simple_loss=0.2923, pruned_loss=0.06803, over 3827665.34 frames. ], batch size: 47, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 12:00:27,551 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6643, 1.6903, 1.5813, 1.3764, 1.2939, 1.4285, 0.2829, 0.6650], + device='cuda:0'), covar=tensor([0.0586, 0.0560, 0.0372, 0.0555, 0.1050, 0.0697, 0.1201, 0.0968], + device='cuda:0'), in_proj_covar=tensor([0.0351, 0.0349, 0.0348, 0.0375, 0.0449, 0.0381, 0.0330, 0.0337], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 12:00:46,045 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=118993.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:00:54,343 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-04-02 12:00:54,927 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3285, 1.3716, 1.6268, 1.5998, 2.4678, 2.0829, 2.6483, 1.0955], + device='cuda:0'), covar=tensor([0.2676, 0.4514, 0.2801, 0.2122, 0.1707, 0.2436, 0.1626, 0.4740], + device='cuda:0'), in_proj_covar=tensor([0.0520, 0.0625, 0.0682, 0.0469, 0.0618, 0.0524, 0.0659, 0.0533], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 12:00:56,050 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119002.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:00:58,541 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-02 12:01:05,215 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.034e+02 5.127e+02 5.908e+02 8.371e+02 2.467e+03, threshold=1.182e+03, percent-clipped=10.0 +2023-04-02 12:01:21,042 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119022.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:01:25,113 INFO [train.py:903] (0/4) Epoch 18, batch 2950, loss[loss=0.2475, simple_loss=0.3126, pruned_loss=0.09116, over 19591.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2932, pruned_loss=0.06849, over 3819581.56 frames. ], batch size: 52, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 12:01:26,618 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119027.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:01:50,443 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119047.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:02:08,427 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119062.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:02:24,519 INFO [train.py:903] (0/4) Epoch 18, batch 3000, loss[loss=0.1928, simple_loss=0.2828, pruned_loss=0.05143, over 19794.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2932, pruned_loss=0.06857, over 3829159.00 frames. ], batch size: 56, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 12:02:24,520 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 12:02:37,007 INFO [train.py:937] (0/4) Epoch 18, validation: loss=0.1707, simple_loss=0.2711, pruned_loss=0.03521, over 944034.00 frames. +2023-04-02 12:02:37,008 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 12:02:37,363 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5657, 1.1633, 1.3566, 1.2397, 2.1978, 0.9719, 2.0163, 2.4244], + device='cuda:0'), covar=tensor([0.0680, 0.2810, 0.2836, 0.1671, 0.0860, 0.2103, 0.1053, 0.0502], + device='cuda:0'), in_proj_covar=tensor([0.0398, 0.0359, 0.0379, 0.0343, 0.0366, 0.0347, 0.0367, 0.0390], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 12:02:40,530 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 12:02:50,778 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119087.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:02:57,654 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=119093.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:02:58,916 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0029, 1.9534, 1.8016, 1.6087, 1.3046, 1.5676, 0.5775, 0.9684], + device='cuda:0'), covar=tensor([0.0830, 0.0728, 0.0506, 0.0855, 0.1515, 0.1074, 0.1351, 0.1243], + device='cuda:0'), in_proj_covar=tensor([0.0351, 0.0348, 0.0349, 0.0374, 0.0449, 0.0381, 0.0329, 0.0338], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 12:03:17,168 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=119108.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:03:17,950 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.980e+02 5.504e+02 6.538e+02 8.295e+02 4.074e+03, threshold=1.308e+03, percent-clipped=8.0 +2023-04-02 12:03:20,402 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=119111.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:03:37,845 INFO [train.py:903] (0/4) Epoch 18, batch 3050, loss[loss=0.2038, simple_loss=0.2893, pruned_loss=0.05917, over 19670.00 frames. ], tot_loss[loss=0.2144, simple_loss=0.2928, pruned_loss=0.06804, over 3837556.28 frames. ], batch size: 58, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 12:04:37,425 INFO [train.py:903] (0/4) Epoch 18, batch 3100, loss[loss=0.1923, simple_loss=0.2697, pruned_loss=0.05745, over 17744.00 frames. ], tot_loss[loss=0.2157, simple_loss=0.294, pruned_loss=0.06873, over 3815797.69 frames. ], batch size: 39, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 12:05:18,222 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.373e+02 4.828e+02 5.979e+02 7.400e+02 1.693e+03, threshold=1.196e+03, percent-clipped=2.0 +2023-04-02 12:05:39,401 INFO [train.py:903] (0/4) Epoch 18, batch 3150, loss[loss=0.199, simple_loss=0.2797, pruned_loss=0.05916, over 19698.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2939, pruned_loss=0.06884, over 3813515.74 frames. ], batch size: 53, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 12:05:57,686 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5693, 1.7019, 2.0840, 1.9415, 3.5090, 2.9395, 3.8945, 1.7496], + device='cuda:0'), covar=tensor([0.2376, 0.4143, 0.2675, 0.1722, 0.1252, 0.1812, 0.1257, 0.3863], + device='cuda:0'), in_proj_covar=tensor([0.0517, 0.0619, 0.0676, 0.0466, 0.0611, 0.0517, 0.0652, 0.0528], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 12:06:06,960 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 12:06:35,346 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1984, 1.1664, 1.2072, 1.3154, 1.0677, 1.2896, 1.3583, 1.2244], + device='cuda:0'), covar=tensor([0.0929, 0.1029, 0.1088, 0.0705, 0.0847, 0.0874, 0.0804, 0.0830], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0224, 0.0226, 0.0245, 0.0228, 0.0211, 0.0190, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 12:06:39,658 INFO [train.py:903] (0/4) Epoch 18, batch 3200, loss[loss=0.2067, simple_loss=0.284, pruned_loss=0.06471, over 12995.00 frames. ], tot_loss[loss=0.2165, simple_loss=0.2944, pruned_loss=0.06924, over 3811569.17 frames. ], batch size: 136, lr: 4.60e-03, grad_scale: 8.0 +2023-04-02 12:06:45,759 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6287, 2.6019, 2.4249, 2.7074, 2.5382, 2.3267, 2.2292, 2.7312], + device='cuda:0'), covar=tensor([0.0888, 0.1408, 0.1289, 0.1010, 0.1418, 0.0498, 0.1267, 0.0585], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0355, 0.0303, 0.0250, 0.0301, 0.0248, 0.0298, 0.0251], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 12:07:18,203 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.241e+02 4.710e+02 5.964e+02 8.137e+02 2.705e+03, threshold=1.193e+03, percent-clipped=4.0 +2023-04-02 12:07:39,013 INFO [train.py:903] (0/4) Epoch 18, batch 3250, loss[loss=0.1943, simple_loss=0.271, pruned_loss=0.05882, over 19377.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2938, pruned_loss=0.06888, over 3812715.13 frames. ], batch size: 48, lr: 4.59e-03, grad_scale: 8.0 +2023-04-02 12:08:24,454 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119364.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:08:37,660 INFO [train.py:903] (0/4) Epoch 18, batch 3300, loss[loss=0.2185, simple_loss=0.3033, pruned_loss=0.06691, over 18658.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.2928, pruned_loss=0.06853, over 3813311.95 frames. ], batch size: 74, lr: 4.59e-03, grad_scale: 8.0 +2023-04-02 12:08:42,298 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 12:08:53,579 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119389.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:09:16,517 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.2249, 5.6941, 3.0744, 4.8855, 1.4058, 5.7693, 5.5945, 5.7893], + device='cuda:0'), covar=tensor([0.0346, 0.0809, 0.1701, 0.0696, 0.3536, 0.0466, 0.0744, 0.0863], + device='cuda:0'), in_proj_covar=tensor([0.0480, 0.0393, 0.0478, 0.0340, 0.0397, 0.0415, 0.0410, 0.0442], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 12:09:17,496 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.159e+02 5.369e+02 6.623e+02 8.148e+02 1.799e+03, threshold=1.325e+03, percent-clipped=5.0 +2023-04-02 12:09:37,823 INFO [train.py:903] (0/4) Epoch 18, batch 3350, loss[loss=0.2014, simple_loss=0.2721, pruned_loss=0.06535, over 19770.00 frames. ], tot_loss[loss=0.2147, simple_loss=0.2926, pruned_loss=0.06836, over 3806914.71 frames. ], batch size: 48, lr: 4.59e-03, grad_scale: 8.0 +2023-04-02 12:09:50,645 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=119437.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:10:11,908 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=119455.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:10:37,463 INFO [train.py:903] (0/4) Epoch 18, batch 3400, loss[loss=0.2058, simple_loss=0.2855, pruned_loss=0.06309, over 19686.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2932, pruned_loss=0.06862, over 3808603.97 frames. ], batch size: 60, lr: 4.59e-03, grad_scale: 4.0 +2023-04-02 12:11:18,512 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.251e+02 4.894e+02 5.911e+02 7.983e+02 1.559e+03, threshold=1.182e+03, percent-clipped=2.0 +2023-04-02 12:11:37,556 INFO [train.py:903] (0/4) Epoch 18, batch 3450, loss[loss=0.2031, simple_loss=0.2752, pruned_loss=0.06549, over 19756.00 frames. ], tot_loss[loss=0.2154, simple_loss=0.2935, pruned_loss=0.06862, over 3799522.71 frames. ], batch size: 48, lr: 4.59e-03, grad_scale: 4.0 +2023-04-02 12:11:43,125 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 12:11:45,595 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=119532.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:12:09,581 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=119552.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:12:31,768 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=119570.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:12:38,213 INFO [train.py:903] (0/4) Epoch 18, batch 3500, loss[loss=0.2415, simple_loss=0.3299, pruned_loss=0.07651, over 19655.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.2936, pruned_loss=0.06816, over 3808217.84 frames. ], batch size: 60, lr: 4.59e-03, grad_scale: 4.0 +2023-04-02 12:13:20,028 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.151e+02 4.884e+02 6.516e+02 7.989e+02 1.670e+03, threshold=1.303e+03, percent-clipped=4.0 +2023-04-02 12:13:39,060 INFO [train.py:903] (0/4) Epoch 18, batch 3550, loss[loss=0.2562, simple_loss=0.3275, pruned_loss=0.09245, over 19138.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2922, pruned_loss=0.06758, over 3813459.44 frames. ], batch size: 69, lr: 4.59e-03, grad_scale: 4.0 +2023-04-02 12:14:15,094 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-02 12:14:39,534 INFO [train.py:903] (0/4) Epoch 18, batch 3600, loss[loss=0.1889, simple_loss=0.269, pruned_loss=0.05435, over 19353.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.2923, pruned_loss=0.06777, over 3801931.22 frames. ], batch size: 47, lr: 4.59e-03, grad_scale: 8.0 +2023-04-02 12:14:47,953 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=119683.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:15:20,526 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.297e+02 4.936e+02 6.005e+02 7.350e+02 2.220e+03, threshold=1.201e+03, percent-clipped=3.0 +2023-04-02 12:15:39,373 INFO [train.py:903] (0/4) Epoch 18, batch 3650, loss[loss=0.2231, simple_loss=0.288, pruned_loss=0.07913, over 19795.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2914, pruned_loss=0.06728, over 3803584.37 frames. ], batch size: 47, lr: 4.59e-03, grad_scale: 8.0 +2023-04-02 12:16:40,016 INFO [train.py:903] (0/4) Epoch 18, batch 3700, loss[loss=0.2104, simple_loss=0.2799, pruned_loss=0.07047, over 19316.00 frames. ], tot_loss[loss=0.214, simple_loss=0.2922, pruned_loss=0.06788, over 3795521.90 frames. ], batch size: 44, lr: 4.59e-03, grad_scale: 8.0 +2023-04-02 12:16:58,974 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 12:17:19,657 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119808.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:17:21,528 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.455e+02 5.017e+02 5.995e+02 7.636e+02 1.424e+03, threshold=1.199e+03, percent-clipped=3.0 +2023-04-02 12:17:32,236 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9273, 1.2152, 1.4823, 0.6761, 2.0146, 2.3797, 2.1053, 2.5971], + device='cuda:0'), covar=tensor([0.1685, 0.3697, 0.3439, 0.2738, 0.0622, 0.0322, 0.0357, 0.0371], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0309, 0.0340, 0.0258, 0.0233, 0.0178, 0.0211, 0.0239], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 12:17:40,480 INFO [train.py:903] (0/4) Epoch 18, batch 3750, loss[loss=0.2342, simple_loss=0.3114, pruned_loss=0.07848, over 19646.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2934, pruned_loss=0.06845, over 3791104.22 frames. ], batch size: 55, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:17:40,907 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=119826.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:17:48,788 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119833.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:17:58,558 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8869, 1.9860, 1.8560, 2.8467, 1.9053, 2.5833, 2.0416, 1.6043], + device='cuda:0'), covar=tensor([0.4802, 0.4160, 0.2644, 0.2736, 0.4401, 0.2291, 0.5776, 0.4952], + device='cuda:0'), in_proj_covar=tensor([0.0870, 0.0926, 0.0696, 0.0921, 0.0850, 0.0789, 0.0827, 0.0766], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 12:18:10,599 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=119851.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:18:20,874 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5203, 4.1114, 4.2441, 4.2582, 1.7345, 4.0078, 3.4873, 3.9472], + device='cuda:0'), covar=tensor([0.1656, 0.0802, 0.0567, 0.0672, 0.5715, 0.0894, 0.0685, 0.1095], + device='cuda:0'), in_proj_covar=tensor([0.0757, 0.0696, 0.0905, 0.0790, 0.0806, 0.0656, 0.0546, 0.0839], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 12:18:33,192 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3800, 1.4348, 1.7238, 1.6251, 2.2642, 2.0457, 2.3259, 1.1080], + device='cuda:0'), covar=tensor([0.2310, 0.3949, 0.2413, 0.1797, 0.1457, 0.2091, 0.1346, 0.4063], + device='cuda:0'), in_proj_covar=tensor([0.0518, 0.0618, 0.0678, 0.0467, 0.0611, 0.0519, 0.0654, 0.0530], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 12:18:39,735 INFO [train.py:903] (0/4) Epoch 18, batch 3800, loss[loss=0.2403, simple_loss=0.3259, pruned_loss=0.07737, over 18818.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.2926, pruned_loss=0.06777, over 3790915.27 frames. ], batch size: 74, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:18:39,891 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=119876.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:19:12,059 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 12:19:22,040 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.281e+02 5.499e+02 6.857e+02 8.596e+02 2.059e+03, threshold=1.371e+03, percent-clipped=8.0 +2023-04-02 12:19:41,466 INFO [train.py:903] (0/4) Epoch 18, batch 3850, loss[loss=0.1863, simple_loss=0.2712, pruned_loss=0.05067, over 19676.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2917, pruned_loss=0.0669, over 3813470.81 frames. ], batch size: 58, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:20:40,740 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0 +2023-04-02 12:20:43,324 INFO [train.py:903] (0/4) Epoch 18, batch 3900, loss[loss=0.2092, simple_loss=0.287, pruned_loss=0.0657, over 19524.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2908, pruned_loss=0.06634, over 3800332.49 frames. ], batch size: 56, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:21:02,815 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=119991.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:21:14,032 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-120000.pt +2023-04-02 12:21:26,390 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.199e+02 4.757e+02 5.722e+02 7.251e+02 1.425e+03, threshold=1.144e+03, percent-clipped=2.0 +2023-04-02 12:21:45,630 INFO [train.py:903] (0/4) Epoch 18, batch 3950, loss[loss=0.19, simple_loss=0.2606, pruned_loss=0.05968, over 19723.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2916, pruned_loss=0.06676, over 3800472.14 frames. ], batch size: 46, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:21:46,900 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=120027.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:21:49,998 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 12:22:05,628 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0311, 5.4730, 3.1358, 4.7858, 0.9884, 5.6389, 5.4231, 5.6421], + device='cuda:0'), covar=tensor([0.0356, 0.0755, 0.1714, 0.0776, 0.4151, 0.0471, 0.0718, 0.0855], + device='cuda:0'), in_proj_covar=tensor([0.0480, 0.0392, 0.0479, 0.0341, 0.0396, 0.0415, 0.0408, 0.0442], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 12:22:47,957 INFO [train.py:903] (0/4) Epoch 18, batch 4000, loss[loss=0.2256, simple_loss=0.3103, pruned_loss=0.07051, over 19672.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2914, pruned_loss=0.06669, over 3807147.50 frames. ], batch size: 60, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:23:29,303 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.243e+02 4.630e+02 5.764e+02 7.444e+02 1.534e+03, threshold=1.153e+03, percent-clipped=2.0 +2023-04-02 12:23:36,024 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 12:23:49,504 INFO [train.py:903] (0/4) Epoch 18, batch 4050, loss[loss=0.1783, simple_loss=0.2621, pruned_loss=0.04722, over 19668.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2914, pruned_loss=0.06679, over 3805101.01 frames. ], batch size: 53, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:24:08,092 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=120142.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:24:47,334 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2353, 1.1670, 1.1529, 1.3591, 1.0771, 1.2989, 1.2970, 1.2605], + device='cuda:0'), covar=tensor([0.0924, 0.1032, 0.1139, 0.0688, 0.0873, 0.0830, 0.0837, 0.0796], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0222, 0.0224, 0.0244, 0.0227, 0.0208, 0.0188, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 12:24:49,139 INFO [train.py:903] (0/4) Epoch 18, batch 4100, loss[loss=0.2358, simple_loss=0.311, pruned_loss=0.08031, over 17961.00 frames. ], tot_loss[loss=0.214, simple_loss=0.2924, pruned_loss=0.06777, over 3810127.78 frames. ], batch size: 83, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:25:24,882 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 12:25:29,250 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.453e+02 5.087e+02 6.190e+02 7.903e+02 1.294e+03, threshold=1.238e+03, percent-clipped=4.0 +2023-04-02 12:25:48,079 INFO [train.py:903] (0/4) Epoch 18, batch 4150, loss[loss=0.2299, simple_loss=0.3124, pruned_loss=0.07373, over 19250.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2923, pruned_loss=0.06759, over 3814175.96 frames. ], batch size: 66, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:26:14,748 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=120247.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:26:46,342 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=120272.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:26:50,729 INFO [train.py:903] (0/4) Epoch 18, batch 4200, loss[loss=0.218, simple_loss=0.2988, pruned_loss=0.06856, over 18207.00 frames. ], tot_loss[loss=0.2136, simple_loss=0.2923, pruned_loss=0.06744, over 3803384.14 frames. ], batch size: 83, lr: 4.58e-03, grad_scale: 8.0 +2023-04-02 12:26:57,166 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 12:27:06,674 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0738, 1.9815, 1.8140, 1.5935, 1.4143, 1.6042, 0.5226, 1.0537], + device='cuda:0'), covar=tensor([0.0540, 0.0569, 0.0439, 0.0720, 0.1161, 0.0892, 0.1209, 0.0977], + device='cuda:0'), in_proj_covar=tensor([0.0348, 0.0345, 0.0346, 0.0372, 0.0448, 0.0379, 0.0325, 0.0334], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 12:27:30,830 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.475e+02 4.968e+02 6.385e+02 7.954e+02 1.571e+03, threshold=1.277e+03, percent-clipped=3.0 +2023-04-02 12:27:51,828 INFO [train.py:903] (0/4) Epoch 18, batch 4250, loss[loss=0.2117, simple_loss=0.287, pruned_loss=0.06816, over 19635.00 frames. ], tot_loss[loss=0.2136, simple_loss=0.2923, pruned_loss=0.06746, over 3802232.07 frames. ], batch size: 50, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:28:09,653 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 12:28:18,896 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 12:28:19,572 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.56 vs. limit=5.0 +2023-04-02 12:28:51,544 INFO [train.py:903] (0/4) Epoch 18, batch 4300, loss[loss=0.26, simple_loss=0.3287, pruned_loss=0.09564, over 19665.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.2933, pruned_loss=0.06832, over 3813853.69 frames. ], batch size: 53, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:29:19,840 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=120398.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:29:34,125 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.947e+02 5.084e+02 5.969e+02 7.631e+02 1.294e+03, threshold=1.194e+03, percent-clipped=1.0 +2023-04-02 12:29:46,151 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 12:29:49,854 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=120423.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:29:52,876 INFO [train.py:903] (0/4) Epoch 18, batch 4350, loss[loss=0.1816, simple_loss=0.2556, pruned_loss=0.05379, over 19722.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.2914, pruned_loss=0.06715, over 3826596.77 frames. ], batch size: 46, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:29:56,626 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0118, 1.1650, 1.5905, 1.0901, 2.4210, 3.2802, 3.0226, 3.5403], + device='cuda:0'), covar=tensor([0.1870, 0.4101, 0.3499, 0.2556, 0.0651, 0.0206, 0.0265, 0.0275], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0311, 0.0339, 0.0258, 0.0234, 0.0179, 0.0211, 0.0238], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 12:30:20,690 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7478, 1.6508, 1.9418, 2.0227, 4.1131, 1.3751, 2.7763, 4.5363], + device='cuda:0'), covar=tensor([0.0440, 0.2837, 0.2689, 0.1786, 0.0762, 0.2639, 0.1302, 0.0234], + device='cuda:0'), in_proj_covar=tensor([0.0394, 0.0354, 0.0374, 0.0338, 0.0363, 0.0343, 0.0364, 0.0385], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 12:30:55,182 INFO [train.py:903] (0/4) Epoch 18, batch 4400, loss[loss=0.2471, simple_loss=0.3144, pruned_loss=0.08991, over 19350.00 frames. ], tot_loss[loss=0.2132, simple_loss=0.2914, pruned_loss=0.06754, over 3792211.62 frames. ], batch size: 70, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:31:19,628 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 12:31:28,329 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 12:31:35,103 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.222e+02 5.092e+02 6.306e+02 8.223e+02 1.474e+03, threshold=1.261e+03, percent-clipped=7.0 +2023-04-02 12:31:55,629 INFO [train.py:903] (0/4) Epoch 18, batch 4450, loss[loss=0.1851, simple_loss=0.2584, pruned_loss=0.05597, over 15699.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2915, pruned_loss=0.06779, over 3788989.64 frames. ], batch size: 34, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:32:55,857 INFO [train.py:903] (0/4) Epoch 18, batch 4500, loss[loss=0.2298, simple_loss=0.3213, pruned_loss=0.06915, over 19674.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2914, pruned_loss=0.0673, over 3799205.62 frames. ], batch size: 60, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:33:37,597 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.517e+02 5.202e+02 6.117e+02 7.756e+02 2.072e+03, threshold=1.223e+03, percent-clipped=4.0 +2023-04-02 12:33:56,192 INFO [train.py:903] (0/4) Epoch 18, batch 4550, loss[loss=0.2217, simple_loss=0.3117, pruned_loss=0.06582, over 19602.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.2926, pruned_loss=0.06776, over 3803110.87 frames. ], batch size: 61, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:34:05,861 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 12:34:29,490 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 12:34:56,267 INFO [train.py:903] (0/4) Epoch 18, batch 4600, loss[loss=0.1971, simple_loss=0.2706, pruned_loss=0.06183, over 19789.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2921, pruned_loss=0.06746, over 3812576.67 frames. ], batch size: 49, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:35:17,287 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=120694.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:35:27,824 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3757, 1.2955, 1.3185, 1.6934, 1.3268, 1.6481, 1.6840, 1.4593], + device='cuda:0'), covar=tensor([0.0865, 0.0989, 0.1049, 0.0718, 0.0880, 0.0763, 0.0822, 0.0744], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0224, 0.0226, 0.0245, 0.0229, 0.0211, 0.0189, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 12:35:35,196 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.215e+02 4.865e+02 6.169e+02 7.995e+02 1.948e+03, threshold=1.234e+03, percent-clipped=9.0 +2023-04-02 12:35:55,081 INFO [train.py:903] (0/4) Epoch 18, batch 4650, loss[loss=0.199, simple_loss=0.2744, pruned_loss=0.06186, over 19720.00 frames. ], tot_loss[loss=0.2136, simple_loss=0.2922, pruned_loss=0.06754, over 3828450.73 frames. ], batch size: 45, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:36:14,447 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 12:36:23,040 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-02 12:36:24,605 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 12:36:55,850 INFO [train.py:903] (0/4) Epoch 18, batch 4700, loss[loss=0.2151, simple_loss=0.2984, pruned_loss=0.06589, over 19659.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.2915, pruned_loss=0.06711, over 3826119.69 frames. ], batch size: 58, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:37:18,750 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 12:37:37,345 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.471e+02 5.224e+02 6.544e+02 8.077e+02 2.112e+03, threshold=1.309e+03, percent-clipped=4.0 +2023-04-02 12:37:51,204 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6420, 1.4936, 1.5038, 2.0074, 1.5502, 1.8375, 1.9229, 1.7081], + device='cuda:0'), covar=tensor([0.0809, 0.0932, 0.1037, 0.0763, 0.0863, 0.0743, 0.0816, 0.0717], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0223, 0.0225, 0.0244, 0.0227, 0.0209, 0.0189, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 12:37:54,545 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-02 12:37:56,117 INFO [train.py:903] (0/4) Epoch 18, batch 4750, loss[loss=0.222, simple_loss=0.2907, pruned_loss=0.07659, over 19777.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2915, pruned_loss=0.06741, over 3823697.37 frames. ], batch size: 56, lr: 4.57e-03, grad_scale: 8.0 +2023-04-02 12:38:57,285 INFO [train.py:903] (0/4) Epoch 18, batch 4800, loss[loss=0.259, simple_loss=0.3248, pruned_loss=0.09659, over 18810.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.2917, pruned_loss=0.06799, over 3805748.93 frames. ], batch size: 74, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:39:38,286 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.835e+02 5.063e+02 6.517e+02 8.294e+02 1.429e+03, threshold=1.303e+03, percent-clipped=1.0 +2023-04-02 12:39:57,795 INFO [train.py:903] (0/4) Epoch 18, batch 4850, loss[loss=0.2019, simple_loss=0.2924, pruned_loss=0.05574, over 19533.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2915, pruned_loss=0.06733, over 3802187.89 frames. ], batch size: 56, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:40:19,477 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 12:40:38,938 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 12:40:44,414 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 12:40:45,486 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 12:40:55,312 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 12:40:57,465 INFO [train.py:903] (0/4) Epoch 18, batch 4900, loss[loss=0.1807, simple_loss=0.2631, pruned_loss=0.0492, over 19840.00 frames. ], tot_loss[loss=0.215, simple_loss=0.2931, pruned_loss=0.0684, over 3807575.58 frames. ], batch size: 52, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:41:15,403 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 12:41:38,441 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.463e+02 5.379e+02 6.693e+02 8.152e+02 1.326e+03, threshold=1.339e+03, percent-clipped=1.0 +2023-04-02 12:41:50,618 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121021.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:41:56,822 INFO [train.py:903] (0/4) Epoch 18, batch 4950, loss[loss=0.1885, simple_loss=0.2667, pruned_loss=0.05514, over 19417.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2927, pruned_loss=0.06794, over 3821429.66 frames. ], batch size: 48, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:42:10,889 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=121038.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:42:13,941 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 12:42:36,736 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 12:42:55,214 INFO [train.py:903] (0/4) Epoch 18, batch 5000, loss[loss=0.2401, simple_loss=0.3136, pruned_loss=0.08329, over 19589.00 frames. ], tot_loss[loss=0.2142, simple_loss=0.2925, pruned_loss=0.06797, over 3824463.38 frames. ], batch size: 61, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:43:05,407 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 12:43:11,791 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8995, 1.6355, 1.7209, 2.6286, 1.8064, 2.2335, 2.2783, 1.9582], + device='cuda:0'), covar=tensor([0.0881, 0.1015, 0.1097, 0.0858, 0.1010, 0.0764, 0.0948, 0.0747], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0224, 0.0226, 0.0244, 0.0229, 0.0210, 0.0189, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 12:43:16,148 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 12:43:36,120 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.358e+02 5.049e+02 6.044e+02 8.198e+02 1.739e+03, threshold=1.209e+03, percent-clipped=8.0 +2023-04-02 12:43:55,697 INFO [train.py:903] (0/4) Epoch 18, batch 5050, loss[loss=0.1851, simple_loss=0.2768, pruned_loss=0.04673, over 19779.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2931, pruned_loss=0.06774, over 3829847.46 frames. ], batch size: 56, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:44:27,536 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=121153.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:44:29,513 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 12:44:55,825 INFO [train.py:903] (0/4) Epoch 18, batch 5100, loss[loss=0.176, simple_loss=0.2582, pruned_loss=0.04693, over 19614.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.2931, pruned_loss=0.0683, over 3819291.34 frames. ], batch size: 50, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:45:03,061 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121182.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:45:05,058 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 12:45:08,125 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 12:45:12,784 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 12:45:37,241 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.930e+02 4.928e+02 6.063e+02 8.354e+02 2.244e+03, threshold=1.213e+03, percent-clipped=8.0 +2023-04-02 12:45:54,392 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9240, 1.3540, 1.0611, 1.0210, 1.1673, 0.9918, 1.0680, 1.2437], + device='cuda:0'), covar=tensor([0.0592, 0.0850, 0.1109, 0.0751, 0.0595, 0.1354, 0.0549, 0.0494], + device='cuda:0'), in_proj_covar=tensor([0.0296, 0.0313, 0.0326, 0.0259, 0.0247, 0.0332, 0.0292, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 12:45:56,107 INFO [train.py:903] (0/4) Epoch 18, batch 5150, loss[loss=0.2557, simple_loss=0.3297, pruned_loss=0.09087, over 19663.00 frames. ], tot_loss[loss=0.2163, simple_loss=0.2942, pruned_loss=0.06922, over 3811755.08 frames. ], batch size: 55, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:46:05,296 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 12:46:40,463 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 12:46:55,814 INFO [train.py:903] (0/4) Epoch 18, batch 5200, loss[loss=0.2005, simple_loss=0.2706, pruned_loss=0.0652, over 19041.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2933, pruned_loss=0.06857, over 3826616.21 frames. ], batch size: 42, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:47:03,754 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4001, 1.1655, 1.3999, 1.6012, 2.9831, 1.1601, 2.2289, 3.2799], + device='cuda:0'), covar=tensor([0.0510, 0.2932, 0.2942, 0.1692, 0.0707, 0.2405, 0.1327, 0.0324], + device='cuda:0'), in_proj_covar=tensor([0.0394, 0.0355, 0.0376, 0.0341, 0.0365, 0.0347, 0.0366, 0.0387], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 12:47:08,979 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 12:47:36,015 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.274e+02 5.110e+02 6.332e+02 8.392e+02 3.036e+03, threshold=1.266e+03, percent-clipped=7.0 +2023-04-02 12:47:39,706 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0711, 1.3172, 1.6283, 1.0731, 2.5124, 3.3836, 3.0956, 3.6188], + device='cuda:0'), covar=tensor([0.1677, 0.3614, 0.3287, 0.2472, 0.0568, 0.0171, 0.0223, 0.0218], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0312, 0.0343, 0.0259, 0.0235, 0.0179, 0.0212, 0.0243], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 12:47:49,050 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 12:47:50,668 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3565, 1.4380, 1.7129, 1.5854, 2.5332, 2.1605, 2.6849, 1.0921], + device='cuda:0'), covar=tensor([0.2246, 0.3960, 0.2449, 0.1790, 0.1441, 0.2018, 0.1398, 0.4014], + device='cuda:0'), in_proj_covar=tensor([0.0517, 0.0621, 0.0680, 0.0466, 0.0612, 0.0519, 0.0653, 0.0531], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 12:47:54,556 INFO [train.py:903] (0/4) Epoch 18, batch 5250, loss[loss=0.2272, simple_loss=0.3124, pruned_loss=0.07103, over 19654.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2944, pruned_loss=0.06915, over 3814142.69 frames. ], batch size: 53, lr: 4.56e-03, grad_scale: 8.0 +2023-04-02 12:47:59,197 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121329.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 12:48:22,542 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4839, 1.5558, 1.7364, 1.7559, 2.6508, 2.3917, 2.7155, 1.1240], + device='cuda:0'), covar=tensor([0.2308, 0.4224, 0.2635, 0.1763, 0.1386, 0.1898, 0.1393, 0.4150], + device='cuda:0'), in_proj_covar=tensor([0.0517, 0.0620, 0.0680, 0.0466, 0.0611, 0.0518, 0.0653, 0.0531], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 12:48:40,804 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=121365.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:48:54,019 INFO [train.py:903] (0/4) Epoch 18, batch 5300, loss[loss=0.194, simple_loss=0.2653, pruned_loss=0.0613, over 19740.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2941, pruned_loss=0.06912, over 3817407.05 frames. ], batch size: 46, lr: 4.56e-03, grad_scale: 4.0 +2023-04-02 12:49:10,547 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 12:49:14,149 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5980, 1.6976, 1.9026, 2.0273, 1.5275, 1.9299, 1.9626, 1.7765], + device='cuda:0'), covar=tensor([0.3925, 0.3468, 0.1819, 0.2123, 0.3620, 0.1964, 0.4569, 0.3128], + device='cuda:0'), in_proj_covar=tensor([0.0870, 0.0928, 0.0695, 0.0923, 0.0851, 0.0788, 0.0826, 0.0764], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 12:49:33,462 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=121409.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:49:35,383 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.570e+02 5.404e+02 6.878e+02 8.269e+02 2.518e+03, threshold=1.376e+03, percent-clipped=11.0 +2023-04-02 12:49:52,480 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121425.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:49:53,395 INFO [train.py:903] (0/4) Epoch 18, batch 5350, loss[loss=0.21, simple_loss=0.2994, pruned_loss=0.06033, over 18838.00 frames. ], tot_loss[loss=0.216, simple_loss=0.2946, pruned_loss=0.06869, over 3816923.41 frames. ], batch size: 74, lr: 4.55e-03, grad_scale: 4.0 +2023-04-02 12:50:03,586 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=121434.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:50:26,219 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 12:50:53,959 INFO [train.py:903] (0/4) Epoch 18, batch 5400, loss[loss=0.2078, simple_loss=0.2914, pruned_loss=0.06216, over 19622.00 frames. ], tot_loss[loss=0.2164, simple_loss=0.2946, pruned_loss=0.06909, over 3813978.64 frames. ], batch size: 57, lr: 4.55e-03, grad_scale: 4.0 +2023-04-02 12:50:59,577 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=121480.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:51:35,379 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.204e+02 5.292e+02 6.238e+02 7.901e+02 1.766e+03, threshold=1.248e+03, percent-clipped=4.0 +2023-04-02 12:51:54,669 INFO [train.py:903] (0/4) Epoch 18, batch 5450, loss[loss=0.2656, simple_loss=0.3383, pruned_loss=0.0964, over 19564.00 frames. ], tot_loss[loss=0.2155, simple_loss=0.2937, pruned_loss=0.06862, over 3823214.52 frames. ], batch size: 61, lr: 4.55e-03, grad_scale: 4.0 +2023-04-02 12:51:54,831 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=121526.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:51:56,510 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-02 12:52:54,551 INFO [train.py:903] (0/4) Epoch 18, batch 5500, loss[loss=0.2332, simple_loss=0.3158, pruned_loss=0.07526, over 18734.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2934, pruned_loss=0.06836, over 3834371.67 frames. ], batch size: 74, lr: 4.55e-03, grad_scale: 4.0 +2023-04-02 12:53:09,252 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0168, 1.8838, 1.6726, 2.0734, 1.8525, 1.7530, 1.6386, 1.9040], + device='cuda:0'), covar=tensor([0.1058, 0.1507, 0.1390, 0.1016, 0.1278, 0.0550, 0.1357, 0.0745], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0355, 0.0302, 0.0249, 0.0299, 0.0247, 0.0297, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 12:53:19,888 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 12:53:37,492 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.086e+02 4.754e+02 5.891e+02 7.661e+02 1.861e+03, threshold=1.178e+03, percent-clipped=6.0 +2023-04-02 12:53:55,068 INFO [train.py:903] (0/4) Epoch 18, batch 5550, loss[loss=0.1911, simple_loss=0.2805, pruned_loss=0.05083, over 19670.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2934, pruned_loss=0.06853, over 3835154.57 frames. ], batch size: 53, lr: 4.55e-03, grad_scale: 4.0 +2023-04-02 12:54:03,885 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 12:54:13,226 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=121641.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:54:52,422 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 12:54:52,536 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=121673.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 12:54:55,722 INFO [train.py:903] (0/4) Epoch 18, batch 5600, loss[loss=0.2292, simple_loss=0.3091, pruned_loss=0.07467, over 19365.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2932, pruned_loss=0.06856, over 3833889.44 frames. ], batch size: 66, lr: 4.55e-03, grad_scale: 8.0 +2023-04-02 12:55:15,344 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.3837, 5.7218, 3.5891, 5.0573, 1.9196, 5.8702, 5.8552, 5.9218], + device='cuda:0'), covar=tensor([0.0365, 0.0822, 0.1781, 0.0807, 0.3547, 0.0513, 0.0687, 0.0934], + device='cuda:0'), in_proj_covar=tensor([0.0484, 0.0391, 0.0479, 0.0340, 0.0392, 0.0418, 0.0408, 0.0439], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 12:55:37,555 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.280e+02 4.851e+02 5.729e+02 6.791e+02 1.695e+03, threshold=1.146e+03, percent-clipped=1.0 +2023-04-02 12:55:56,731 INFO [train.py:903] (0/4) Epoch 18, batch 5650, loss[loss=0.2567, simple_loss=0.3305, pruned_loss=0.09148, over 19508.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.2935, pruned_loss=0.06846, over 3826308.06 frames. ], batch size: 64, lr: 4.55e-03, grad_scale: 8.0 +2023-04-02 12:56:08,931 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=121736.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:56:37,750 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=121761.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:56:44,588 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 12:56:48,209 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=121769.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:56:50,041 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.95 vs. limit=5.0 +2023-04-02 12:56:57,050 INFO [train.py:903] (0/4) Epoch 18, batch 5700, loss[loss=0.2122, simple_loss=0.2915, pruned_loss=0.06646, over 19787.00 frames. ], tot_loss[loss=0.2147, simple_loss=0.2929, pruned_loss=0.0682, over 3824096.84 frames. ], batch size: 56, lr: 4.55e-03, grad_scale: 8.0 +2023-04-02 12:57:10,997 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=121788.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 12:57:39,340 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.986e+02 4.821e+02 5.859e+02 7.608e+02 1.521e+03, threshold=1.172e+03, percent-clipped=4.0 +2023-04-02 12:57:57,043 INFO [train.py:903] (0/4) Epoch 18, batch 5750, loss[loss=0.2237, simple_loss=0.3026, pruned_loss=0.07243, over 19648.00 frames. ], tot_loss[loss=0.2149, simple_loss=0.2932, pruned_loss=0.06831, over 3821843.32 frames. ], batch size: 58, lr: 4.55e-03, grad_scale: 8.0 +2023-04-02 12:57:58,074 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 12:58:05,923 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 12:58:11,392 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 12:58:42,346 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0160, 2.0061, 1.8296, 1.6767, 1.5993, 1.7026, 0.4921, 1.0033], + device='cuda:0'), covar=tensor([0.0483, 0.0507, 0.0361, 0.0555, 0.0952, 0.0641, 0.1047, 0.0879], + device='cuda:0'), in_proj_covar=tensor([0.0351, 0.0347, 0.0349, 0.0375, 0.0451, 0.0379, 0.0329, 0.0336], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 12:58:43,856 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-02 12:58:57,512 INFO [train.py:903] (0/4) Epoch 18, batch 5800, loss[loss=0.2082, simple_loss=0.2802, pruned_loss=0.06805, over 19361.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2935, pruned_loss=0.06834, over 3831433.89 frames. ], batch size: 47, lr: 4.55e-03, grad_scale: 8.0 +2023-04-02 12:59:08,067 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=121884.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:59:24,587 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=121897.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:59:39,870 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.096e+02 5.174e+02 6.386e+02 8.157e+02 2.937e+03, threshold=1.277e+03, percent-clipped=7.0 +2023-04-02 12:59:54,653 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=121922.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 12:59:59,639 INFO [train.py:903] (0/4) Epoch 18, batch 5850, loss[loss=0.1742, simple_loss=0.2559, pruned_loss=0.04632, over 19796.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2925, pruned_loss=0.06759, over 3820728.73 frames. ], batch size: 49, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:00:05,813 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5279, 2.3853, 1.7701, 1.6556, 2.2090, 1.3776, 1.4607, 1.9338], + device='cuda:0'), covar=tensor([0.1059, 0.0747, 0.1018, 0.0789, 0.0530, 0.1235, 0.0738, 0.0486], + device='cuda:0'), in_proj_covar=tensor([0.0297, 0.0313, 0.0327, 0.0257, 0.0245, 0.0331, 0.0291, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:00:51,267 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=121968.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:01:00,886 INFO [train.py:903] (0/4) Epoch 18, batch 5900, loss[loss=0.1767, simple_loss=0.2601, pruned_loss=0.04668, over 19731.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2918, pruned_loss=0.06717, over 3820581.13 frames. ], batch size: 47, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:01:02,082 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 13:01:23,287 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 13:01:29,125 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-122000.pt +2023-04-02 13:01:39,543 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0311, 1.8333, 1.9233, 1.6748, 4.5318, 0.9346, 2.6083, 5.0325], + device='cuda:0'), covar=tensor([0.0356, 0.2482, 0.2579, 0.1934, 0.0700, 0.2842, 0.1362, 0.0148], + device='cuda:0'), in_proj_covar=tensor([0.0390, 0.0352, 0.0373, 0.0338, 0.0364, 0.0346, 0.0363, 0.0385], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:01:43,749 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.064e+02 4.950e+02 6.415e+02 8.144e+02 2.513e+03, threshold=1.283e+03, percent-clipped=4.0 +2023-04-02 13:02:01,671 INFO [train.py:903] (0/4) Epoch 18, batch 5950, loss[loss=0.1739, simple_loss=0.2618, pruned_loss=0.04299, over 19760.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2916, pruned_loss=0.06714, over 3807653.44 frames. ], batch size: 54, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:02:24,075 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122044.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 13:02:42,870 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-04-02 13:02:53,760 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122069.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 13:03:01,813 INFO [train.py:903] (0/4) Epoch 18, batch 6000, loss[loss=0.3102, simple_loss=0.3572, pruned_loss=0.1316, over 12851.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2913, pruned_loss=0.06704, over 3809554.35 frames. ], batch size: 135, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:03:01,813 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 13:03:14,295 INFO [train.py:937] (0/4) Epoch 18, validation: loss=0.1702, simple_loss=0.2706, pruned_loss=0.03489, over 944034.00 frames. +2023-04-02 13:03:14,296 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 13:03:22,595 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8806, 1.2441, 1.6024, 0.5750, 1.9646, 2.4722, 2.1012, 2.5778], + device='cuda:0'), covar=tensor([0.1651, 0.3820, 0.3241, 0.2686, 0.0613, 0.0262, 0.0350, 0.0349], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0313, 0.0343, 0.0259, 0.0235, 0.0180, 0.0212, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 13:03:57,709 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.792e+02 5.055e+02 6.106e+02 7.549e+02 1.634e+03, threshold=1.221e+03, percent-clipped=4.0 +2023-04-02 13:04:15,901 INFO [train.py:903] (0/4) Epoch 18, batch 6050, loss[loss=0.2061, simple_loss=0.2939, pruned_loss=0.05919, over 19766.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2915, pruned_loss=0.06688, over 3821374.85 frames. ], batch size: 54, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:04:27,115 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122135.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:04:33,657 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122140.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:05:04,389 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122165.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:05:18,140 INFO [train.py:903] (0/4) Epoch 18, batch 6100, loss[loss=0.2211, simple_loss=0.2809, pruned_loss=0.08064, over 19768.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2912, pruned_loss=0.06694, over 3831474.01 frames. ], batch size: 47, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:05:59,975 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.364e+02 5.309e+02 6.592e+02 8.010e+02 1.726e+03, threshold=1.318e+03, percent-clipped=1.0 +2023-04-02 13:06:03,471 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122213.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:06:18,827 INFO [train.py:903] (0/4) Epoch 18, batch 6150, loss[loss=0.1896, simple_loss=0.2696, pruned_loss=0.05476, over 19727.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2911, pruned_loss=0.06687, over 3835258.82 frames. ], batch size: 51, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:06:46,724 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 13:07:20,188 INFO [train.py:903] (0/4) Epoch 18, batch 6200, loss[loss=0.2172, simple_loss=0.2991, pruned_loss=0.06765, over 19787.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2915, pruned_loss=0.06706, over 3832845.74 frames. ], batch size: 56, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:08:04,096 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.932e+02 4.549e+02 5.662e+02 6.707e+02 1.660e+03, threshold=1.132e+03, percent-clipped=3.0 +2023-04-02 13:08:05,408 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=122312.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:08:17,460 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-02 13:08:21,490 INFO [train.py:903] (0/4) Epoch 18, batch 6250, loss[loss=0.2204, simple_loss=0.3044, pruned_loss=0.06821, over 19658.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2913, pruned_loss=0.06686, over 3831466.18 frames. ], batch size: 55, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:08:35,709 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122337.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:08:53,452 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 13:09:23,686 INFO [train.py:903] (0/4) Epoch 18, batch 6300, loss[loss=0.229, simple_loss=0.3044, pruned_loss=0.0768, over 19628.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2916, pruned_loss=0.06698, over 3822556.92 frames. ], batch size: 50, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:10:06,342 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.668e+02 4.826e+02 5.839e+02 7.420e+02 1.796e+03, threshold=1.168e+03, percent-clipped=3.0 +2023-04-02 13:10:25,333 INFO [train.py:903] (0/4) Epoch 18, batch 6350, loss[loss=0.2541, simple_loss=0.3272, pruned_loss=0.09047, over 19749.00 frames. ], tot_loss[loss=0.2144, simple_loss=0.2929, pruned_loss=0.06796, over 3794381.34 frames. ], batch size: 51, lr: 4.54e-03, grad_scale: 8.0 +2023-04-02 13:10:26,809 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122427.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:11:25,783 INFO [train.py:903] (0/4) Epoch 18, batch 6400, loss[loss=0.1966, simple_loss=0.281, pruned_loss=0.05613, over 18055.00 frames. ], tot_loss[loss=0.2162, simple_loss=0.2944, pruned_loss=0.06902, over 3796719.39 frames. ], batch size: 83, lr: 4.53e-03, grad_scale: 8.0 +2023-04-02 13:11:29,340 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=122479.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:11:38,259 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122486.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:12:08,690 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.263e+02 4.572e+02 5.352e+02 7.001e+02 2.095e+03, threshold=1.070e+03, percent-clipped=5.0 +2023-04-02 13:12:27,158 INFO [train.py:903] (0/4) Epoch 18, batch 6450, loss[loss=0.2208, simple_loss=0.3152, pruned_loss=0.06322, over 19539.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2941, pruned_loss=0.06851, over 3794664.39 frames. ], batch size: 54, lr: 4.53e-03, grad_scale: 8.0 +2023-04-02 13:13:05,515 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=122557.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:13:09,889 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 13:13:27,311 INFO [train.py:903] (0/4) Epoch 18, batch 6500, loss[loss=0.1976, simple_loss=0.2761, pruned_loss=0.05953, over 19817.00 frames. ], tot_loss[loss=0.2145, simple_loss=0.2933, pruned_loss=0.06783, over 3801570.56 frames. ], batch size: 49, lr: 4.53e-03, grad_scale: 8.0 +2023-04-02 13:13:32,630 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 13:13:50,437 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122594.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:14:10,392 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.928e+02 4.824e+02 5.508e+02 6.898e+02 1.222e+03, threshold=1.102e+03, percent-clipped=3.0 +2023-04-02 13:14:28,809 INFO [train.py:903] (0/4) Epoch 18, batch 6550, loss[loss=0.2501, simple_loss=0.3244, pruned_loss=0.08795, over 17727.00 frames. ], tot_loss[loss=0.2144, simple_loss=0.2934, pruned_loss=0.06775, over 3819939.40 frames. ], batch size: 101, lr: 4.53e-03, grad_scale: 8.0 +2023-04-02 13:15:24,485 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122672.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:15:29,442 INFO [train.py:903] (0/4) Epoch 18, batch 6600, loss[loss=0.2118, simple_loss=0.2985, pruned_loss=0.06257, over 17207.00 frames. ], tot_loss[loss=0.215, simple_loss=0.2936, pruned_loss=0.06815, over 3814207.81 frames. ], batch size: 101, lr: 4.53e-03, grad_scale: 8.0 +2023-04-02 13:15:35,553 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=122681.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:15:37,943 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122683.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:16:08,806 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122708.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:16:11,859 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.339e+02 4.976e+02 6.114e+02 8.137e+02 1.508e+03, threshold=1.223e+03, percent-clipped=10.0 +2023-04-02 13:16:29,837 INFO [train.py:903] (0/4) Epoch 18, batch 6650, loss[loss=0.2481, simple_loss=0.3258, pruned_loss=0.08522, over 19660.00 frames. ], tot_loss[loss=0.2142, simple_loss=0.2928, pruned_loss=0.06774, over 3801370.96 frames. ], batch size: 58, lr: 4.53e-03, grad_scale: 8.0 +2023-04-02 13:17:30,798 INFO [train.py:903] (0/4) Epoch 18, batch 6700, loss[loss=0.176, simple_loss=0.2535, pruned_loss=0.0493, over 19789.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2919, pruned_loss=0.06749, over 3802740.17 frames. ], batch size: 49, lr: 4.53e-03, grad_scale: 8.0 +2023-04-02 13:17:55,725 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122796.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:18:12,046 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.354e+02 4.791e+02 5.844e+02 7.603e+02 1.856e+03, threshold=1.169e+03, percent-clipped=2.0 +2023-04-02 13:18:28,394 INFO [train.py:903] (0/4) Epoch 18, batch 6750, loss[loss=0.2206, simple_loss=0.3109, pruned_loss=0.06517, over 19536.00 frames. ], tot_loss[loss=0.2144, simple_loss=0.2926, pruned_loss=0.0681, over 3818614.34 frames. ], batch size: 54, lr: 4.53e-03, grad_scale: 8.0 +2023-04-02 13:18:33,056 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=122830.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:18:55,631 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122850.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:19:24,218 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122875.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:19:25,055 INFO [train.py:903] (0/4) Epoch 18, batch 6800, loss[loss=0.1878, simple_loss=0.2746, pruned_loss=0.05048, over 19481.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2917, pruned_loss=0.06782, over 3813658.95 frames. ], batch size: 64, lr: 4.53e-03, grad_scale: 8.0 +2023-04-02 13:19:40,829 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=122890.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:19:54,378 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-18.pt +2023-04-02 13:20:09,289 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 13:20:09,734 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 13:20:12,595 INFO [train.py:903] (0/4) Epoch 19, batch 0, loss[loss=0.2173, simple_loss=0.2985, pruned_loss=0.06807, over 19398.00 frames. ], tot_loss[loss=0.2173, simple_loss=0.2985, pruned_loss=0.06807, over 19398.00 frames. ], batch size: 66, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:20:12,596 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 13:20:24,046 INFO [train.py:937] (0/4) Epoch 19, validation: loss=0.171, simple_loss=0.2713, pruned_loss=0.03533, over 944034.00 frames. +2023-04-02 13:20:24,047 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 13:20:32,696 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.525e+02 4.987e+02 6.075e+02 7.792e+02 1.350e+03, threshold=1.215e+03, percent-clipped=4.0 +2023-04-02 13:20:38,460 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 13:20:53,676 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=122928.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:21:13,140 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.53 vs. limit=2.0 +2023-04-02 13:21:13,966 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=122945.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:21:24,643 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=122953.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:21:25,464 INFO [train.py:903] (0/4) Epoch 19, batch 50, loss[loss=0.1905, simple_loss=0.286, pruned_loss=0.04752, over 19092.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.2933, pruned_loss=0.06727, over 880240.17 frames. ], batch size: 69, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:21:51,203 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.39 vs. limit=5.0 +2023-04-02 13:22:04,207 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 13:22:16,579 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3951, 2.1227, 1.5687, 1.4401, 1.9679, 1.2565, 1.3920, 1.8321], + device='cuda:0'), covar=tensor([0.1028, 0.0793, 0.1084, 0.0806, 0.0528, 0.1307, 0.0637, 0.0451], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0314, 0.0331, 0.0259, 0.0247, 0.0334, 0.0291, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:22:27,934 INFO [train.py:903] (0/4) Epoch 19, batch 100, loss[loss=0.1899, simple_loss=0.2781, pruned_loss=0.05085, over 19679.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.289, pruned_loss=0.06582, over 1542959.43 frames. ], batch size: 53, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:22:35,845 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.196e+02 4.975e+02 5.926e+02 8.151e+02 1.966e+03, threshold=1.185e+03, percent-clipped=7.0 +2023-04-02 13:22:37,373 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9439, 2.7501, 2.2166, 2.0713, 2.0440, 2.3534, 0.9218, 2.0114], + device='cuda:0'), covar=tensor([0.0621, 0.0523, 0.0600, 0.0980, 0.0895, 0.0997, 0.1279, 0.0885], + device='cuda:0'), in_proj_covar=tensor([0.0350, 0.0347, 0.0347, 0.0374, 0.0450, 0.0380, 0.0328, 0.0333], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 13:22:40,149 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 13:23:25,695 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=123052.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:23:26,591 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6248, 4.1967, 2.7467, 3.6887, 0.9216, 4.0967, 4.0327, 4.0830], + device='cuda:0'), covar=tensor([0.0623, 0.0987, 0.1865, 0.0862, 0.4186, 0.0702, 0.0784, 0.1121], + device='cuda:0'), in_proj_covar=tensor([0.0485, 0.0395, 0.0483, 0.0342, 0.0397, 0.0421, 0.0410, 0.0446], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:23:27,453 INFO [train.py:903] (0/4) Epoch 19, batch 150, loss[loss=0.2117, simple_loss=0.2917, pruned_loss=0.06579, over 19601.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2914, pruned_loss=0.06756, over 2054391.72 frames. ], batch size: 52, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:23:55,509 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=123077.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:24:27,281 INFO [train.py:903] (0/4) Epoch 19, batch 200, loss[loss=0.1973, simple_loss=0.2697, pruned_loss=0.06241, over 19767.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2909, pruned_loss=0.06695, over 2452073.56 frames. ], batch size: 47, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:24:29,571 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 13:24:35,482 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.274e+02 5.223e+02 5.868e+02 7.012e+02 1.944e+03, threshold=1.174e+03, percent-clipped=7.0 +2023-04-02 13:24:36,959 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1089, 1.2483, 1.5479, 1.4532, 2.7004, 1.0542, 2.1499, 3.0607], + device='cuda:0'), covar=tensor([0.0575, 0.2854, 0.2725, 0.1731, 0.0769, 0.2442, 0.1173, 0.0328], + device='cuda:0'), in_proj_covar=tensor([0.0396, 0.0356, 0.0378, 0.0341, 0.0367, 0.0349, 0.0369, 0.0386], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:24:58,726 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9599, 3.5955, 2.5940, 3.2682, 1.0296, 3.5480, 3.4280, 3.5189], + device='cuda:0'), covar=tensor([0.0868, 0.1173, 0.1887, 0.0910, 0.3957, 0.0848, 0.0962, 0.1216], + device='cuda:0'), in_proj_covar=tensor([0.0485, 0.0396, 0.0483, 0.0342, 0.0396, 0.0420, 0.0410, 0.0445], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:25:27,083 INFO [train.py:903] (0/4) Epoch 19, batch 250, loss[loss=0.2528, simple_loss=0.3299, pruned_loss=0.08787, over 19747.00 frames. ], tot_loss[loss=0.215, simple_loss=0.293, pruned_loss=0.06853, over 2754599.66 frames. ], batch size: 63, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:26:25,119 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=123201.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:26:28,234 INFO [train.py:903] (0/4) Epoch 19, batch 300, loss[loss=0.222, simple_loss=0.3028, pruned_loss=0.07061, over 19674.00 frames. ], tot_loss[loss=0.2144, simple_loss=0.2925, pruned_loss=0.06812, over 2998095.45 frames. ], batch size: 60, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:26:37,167 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.064e+02 4.833e+02 6.226e+02 7.852e+02 1.722e+03, threshold=1.245e+03, percent-clipped=4.0 +2023-04-02 13:26:55,589 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=123226.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:27:04,591 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=123234.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:27:27,986 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.70 vs. limit=5.0 +2023-04-02 13:27:29,488 INFO [train.py:903] (0/4) Epoch 19, batch 350, loss[loss=0.2467, simple_loss=0.3157, pruned_loss=0.08881, over 13808.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2904, pruned_loss=0.06693, over 3188695.33 frames. ], batch size: 136, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:27:35,243 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 13:27:46,485 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7550, 1.2614, 1.5411, 1.7552, 3.3235, 1.1835, 2.4386, 3.7454], + device='cuda:0'), covar=tensor([0.0484, 0.2990, 0.2915, 0.1727, 0.0771, 0.2666, 0.1310, 0.0249], + device='cuda:0'), in_proj_covar=tensor([0.0397, 0.0357, 0.0380, 0.0342, 0.0367, 0.0350, 0.0370, 0.0388], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:28:10,784 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6835, 1.4279, 1.4379, 2.2867, 1.7396, 1.6813, 1.9357, 1.7070], + device='cuda:0'), covar=tensor([0.0939, 0.1184, 0.1181, 0.0773, 0.0909, 0.1005, 0.0972, 0.0861], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0219, 0.0224, 0.0241, 0.0226, 0.0210, 0.0186, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 13:28:30,168 INFO [train.py:903] (0/4) Epoch 19, batch 400, loss[loss=0.1824, simple_loss=0.26, pruned_loss=0.05243, over 19366.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2899, pruned_loss=0.06634, over 3335767.80 frames. ], batch size: 47, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:28:37,933 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.570e+02 5.151e+02 6.306e+02 7.944e+02 1.366e+03, threshold=1.261e+03, percent-clipped=2.0 +2023-04-02 13:29:04,090 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=123332.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:29:24,919 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=123349.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:29:30,457 INFO [train.py:903] (0/4) Epoch 19, batch 450, loss[loss=0.2376, simple_loss=0.312, pruned_loss=0.08162, over 19777.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2886, pruned_loss=0.06556, over 3449748.86 frames. ], batch size: 56, lr: 4.40e-03, grad_scale: 8.0 +2023-04-02 13:30:04,818 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 13:30:05,750 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 13:30:30,934 INFO [train.py:903] (0/4) Epoch 19, batch 500, loss[loss=0.2138, simple_loss=0.2856, pruned_loss=0.07098, over 19391.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2902, pruned_loss=0.06686, over 3533692.06 frames. ], batch size: 48, lr: 4.40e-03, grad_scale: 16.0 +2023-04-02 13:30:39,752 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.237e+02 5.352e+02 7.180e+02 8.361e+02 2.088e+03, threshold=1.436e+03, percent-clipped=3.0 +2023-04-02 13:31:21,136 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=123446.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:31:22,325 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3445, 1.4759, 1.7090, 1.6932, 2.9825, 1.3679, 2.3843, 3.3339], + device='cuda:0'), covar=tensor([0.0503, 0.2562, 0.2526, 0.1715, 0.0675, 0.2235, 0.1202, 0.0284], + device='cuda:0'), in_proj_covar=tensor([0.0400, 0.0359, 0.0380, 0.0344, 0.0369, 0.0350, 0.0371, 0.0389], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:31:30,113 INFO [train.py:903] (0/4) Epoch 19, batch 550, loss[loss=0.187, simple_loss=0.2755, pruned_loss=0.04924, over 19782.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2903, pruned_loss=0.06696, over 3594901.99 frames. ], batch size: 56, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:32:30,276 INFO [train.py:903] (0/4) Epoch 19, batch 600, loss[loss=0.2518, simple_loss=0.3272, pruned_loss=0.0882, over 19479.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2897, pruned_loss=0.06652, over 3653640.86 frames. ], batch size: 64, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:32:39,899 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.253e+02 4.960e+02 5.982e+02 8.370e+02 1.865e+03, threshold=1.196e+03, percent-clipped=4.0 +2023-04-02 13:32:49,074 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0061, 4.5240, 2.5435, 3.8998, 0.8146, 4.4345, 4.3467, 4.4584], + device='cuda:0'), covar=tensor([0.0532, 0.0932, 0.2113, 0.0874, 0.4295, 0.0639, 0.0854, 0.0951], + device='cuda:0'), in_proj_covar=tensor([0.0480, 0.0396, 0.0480, 0.0339, 0.0394, 0.0418, 0.0407, 0.0443], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:33:14,013 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 13:33:30,844 INFO [train.py:903] (0/4) Epoch 19, batch 650, loss[loss=0.2049, simple_loss=0.2779, pruned_loss=0.06598, over 16319.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.289, pruned_loss=0.06614, over 3699380.60 frames. ], batch size: 36, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:34:07,507 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4449, 2.1848, 1.6142, 1.5491, 2.1100, 1.3166, 1.3274, 1.8977], + device='cuda:0'), covar=tensor([0.1189, 0.0821, 0.1022, 0.0813, 0.0503, 0.1262, 0.0792, 0.0472], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0316, 0.0335, 0.0261, 0.0248, 0.0337, 0.0293, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:34:30,416 INFO [train.py:903] (0/4) Epoch 19, batch 700, loss[loss=0.2275, simple_loss=0.3104, pruned_loss=0.07227, over 18818.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2909, pruned_loss=0.06735, over 3725845.91 frames. ], batch size: 74, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:34:31,910 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=123605.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:34:41,234 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.354e+02 5.460e+02 6.000e+02 7.674e+02 1.249e+03, threshold=1.200e+03, percent-clipped=2.0 +2023-04-02 13:35:02,771 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=123630.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:35:31,845 INFO [train.py:903] (0/4) Epoch 19, batch 750, loss[loss=0.2474, simple_loss=0.3279, pruned_loss=0.08342, over 19361.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2917, pruned_loss=0.06786, over 3724444.23 frames. ], batch size: 70, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:35:59,275 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=123676.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:36:34,036 INFO [train.py:903] (0/4) Epoch 19, batch 800, loss[loss=0.2088, simple_loss=0.2894, pruned_loss=0.06403, over 19799.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.2923, pruned_loss=0.06794, over 3736848.81 frames. ], batch size: 56, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:36:37,839 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0111, 1.9562, 1.7022, 2.1059, 1.9762, 1.7819, 1.6751, 1.9863], + device='cuda:0'), covar=tensor([0.1156, 0.1567, 0.1519, 0.1106, 0.1309, 0.0590, 0.1437, 0.0732], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0352, 0.0303, 0.0249, 0.0298, 0.0248, 0.0297, 0.0251], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:36:43,873 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.634e+02 4.695e+02 6.255e+02 7.449e+02 1.390e+03, threshold=1.251e+03, percent-clipped=2.0 +2023-04-02 13:36:47,185 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 13:37:21,007 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0834, 3.3392, 1.9590, 2.1344, 3.0704, 1.7190, 1.5128, 2.2269], + device='cuda:0'), covar=tensor([0.1276, 0.0653, 0.1039, 0.0740, 0.0500, 0.1162, 0.0917, 0.0631], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0313, 0.0332, 0.0259, 0.0245, 0.0335, 0.0291, 0.0271], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:37:34,010 INFO [train.py:903] (0/4) Epoch 19, batch 850, loss[loss=0.1738, simple_loss=0.2535, pruned_loss=0.04707, over 19761.00 frames. ], tot_loss[loss=0.2155, simple_loss=0.2935, pruned_loss=0.06878, over 3760889.31 frames. ], batch size: 45, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:38:17,250 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=123790.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:38:18,592 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=123791.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:38:23,843 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 13:38:34,411 INFO [train.py:903] (0/4) Epoch 19, batch 900, loss[loss=0.1812, simple_loss=0.27, pruned_loss=0.04621, over 19595.00 frames. ], tot_loss[loss=0.215, simple_loss=0.2932, pruned_loss=0.06835, over 3771465.79 frames. ], batch size: 52, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:38:44,953 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.405e+02 4.993e+02 6.523e+02 8.112e+02 2.572e+03, threshold=1.305e+03, percent-clipped=9.0 +2023-04-02 13:39:35,384 INFO [train.py:903] (0/4) Epoch 19, batch 950, loss[loss=0.2177, simple_loss=0.2963, pruned_loss=0.0696, over 19603.00 frames. ], tot_loss[loss=0.2157, simple_loss=0.2939, pruned_loss=0.06873, over 3780717.93 frames. ], batch size: 57, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:39:35,398 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 13:39:42,755 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 13:39:55,781 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-02 13:40:18,686 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=123890.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:40:35,887 INFO [train.py:903] (0/4) Epoch 19, batch 1000, loss[loss=0.2147, simple_loss=0.2961, pruned_loss=0.0667, over 19370.00 frames. ], tot_loss[loss=0.2142, simple_loss=0.2924, pruned_loss=0.068, over 3794105.86 frames. ], batch size: 70, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:40:37,252 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=123905.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:40:44,459 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.613e+02 5.533e+02 6.903e+02 9.244e+02 2.435e+03, threshold=1.381e+03, percent-clipped=8.0 +2023-04-02 13:41:23,925 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 13:41:33,875 INFO [train.py:903] (0/4) Epoch 19, batch 1050, loss[loss=0.2257, simple_loss=0.2943, pruned_loss=0.07855, over 19668.00 frames. ], tot_loss[loss=0.2158, simple_loss=0.2938, pruned_loss=0.06891, over 3808974.43 frames. ], batch size: 55, lr: 4.39e-03, grad_scale: 8.0 +2023-04-02 13:42:03,304 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 13:42:17,131 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9840, 5.0874, 5.8532, 5.8496, 2.0706, 5.5052, 4.6691, 5.4750], + device='cuda:0'), covar=tensor([0.1762, 0.0781, 0.0584, 0.0628, 0.5845, 0.0732, 0.0589, 0.1223], + device='cuda:0'), in_proj_covar=tensor([0.0758, 0.0704, 0.0912, 0.0800, 0.0811, 0.0661, 0.0551, 0.0838], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 13:42:30,090 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-124000.pt +2023-04-02 13:42:36,015 INFO [train.py:903] (0/4) Epoch 19, batch 1100, loss[loss=0.2438, simple_loss=0.3125, pruned_loss=0.08751, over 19759.00 frames. ], tot_loss[loss=0.2159, simple_loss=0.2938, pruned_loss=0.06897, over 3803223.34 frames. ], batch size: 54, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:42:45,327 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.024e+02 4.737e+02 5.703e+02 7.685e+02 1.238e+03, threshold=1.141e+03, percent-clipped=0.0 +2023-04-02 13:43:16,758 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3043, 1.3836, 1.5839, 1.6105, 1.2029, 1.5891, 1.6206, 1.4893], + device='cuda:0'), covar=tensor([0.3299, 0.2873, 0.1535, 0.1775, 0.3092, 0.1595, 0.3923, 0.2663], + device='cuda:0'), in_proj_covar=tensor([0.0871, 0.0931, 0.0697, 0.0918, 0.0854, 0.0789, 0.0824, 0.0764], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 13:43:28,345 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124047.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:43:36,792 INFO [train.py:903] (0/4) Epoch 19, batch 1150, loss[loss=0.1668, simple_loss=0.2465, pruned_loss=0.04362, over 18983.00 frames. ], tot_loss[loss=0.2152, simple_loss=0.293, pruned_loss=0.06865, over 3800115.28 frames. ], batch size: 42, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:43:37,089 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124054.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:43:59,149 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124072.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:44:08,936 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124080.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:44:37,476 INFO [train.py:903] (0/4) Epoch 19, batch 1200, loss[loss=0.1977, simple_loss=0.2719, pruned_loss=0.06175, over 19367.00 frames. ], tot_loss[loss=0.215, simple_loss=0.2928, pruned_loss=0.06858, over 3805638.65 frames. ], batch size: 47, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:44:48,188 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.573e+02 5.076e+02 6.121e+02 8.217e+02 1.455e+03, threshold=1.224e+03, percent-clipped=6.0 +2023-04-02 13:44:50,874 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9719, 1.7261, 1.6812, 1.9828, 1.7263, 1.6988, 1.5450, 1.9230], + device='cuda:0'), covar=tensor([0.1031, 0.1432, 0.1363, 0.0995, 0.1320, 0.0555, 0.1396, 0.0682], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0350, 0.0302, 0.0247, 0.0297, 0.0245, 0.0296, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:45:01,825 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4080, 1.4307, 1.7441, 1.6726, 2.9434, 2.2908, 3.0471, 1.2838], + device='cuda:0'), covar=tensor([0.2304, 0.4223, 0.2647, 0.1847, 0.1307, 0.2102, 0.1346, 0.4101], + device='cuda:0'), in_proj_covar=tensor([0.0515, 0.0621, 0.0682, 0.0469, 0.0613, 0.0519, 0.0653, 0.0530], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 13:45:09,226 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 13:45:38,204 INFO [train.py:903] (0/4) Epoch 19, batch 1250, loss[loss=0.223, simple_loss=0.2961, pruned_loss=0.07498, over 19359.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2927, pruned_loss=0.06792, over 3821978.14 frames. ], batch size: 70, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:45:47,139 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124161.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:46:16,847 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124186.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:46:38,209 INFO [train.py:903] (0/4) Epoch 19, batch 1300, loss[loss=0.2128, simple_loss=0.2898, pruned_loss=0.06789, over 19113.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2927, pruned_loss=0.06794, over 3826179.27 frames. ], batch size: 42, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:46:38,534 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2254, 1.1856, 1.6159, 1.6376, 2.7334, 4.4302, 4.3254, 5.0214], + device='cuda:0'), covar=tensor([0.1796, 0.5050, 0.4522, 0.2311, 0.0753, 0.0228, 0.0205, 0.0199], + device='cuda:0'), in_proj_covar=tensor([0.0264, 0.0312, 0.0343, 0.0260, 0.0237, 0.0179, 0.0212, 0.0242], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 13:46:47,487 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1731, 3.3789, 3.6604, 3.6630, 1.9939, 3.4080, 3.1164, 3.4533], + device='cuda:0'), covar=tensor([0.1387, 0.2855, 0.0697, 0.0747, 0.4479, 0.1326, 0.0622, 0.1049], + device='cuda:0'), in_proj_covar=tensor([0.0757, 0.0705, 0.0909, 0.0795, 0.0808, 0.0660, 0.0550, 0.0838], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 13:46:48,381 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.504e+02 4.873e+02 5.866e+02 7.986e+02 1.872e+03, threshold=1.173e+03, percent-clipped=5.0 +2023-04-02 13:47:15,086 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=124234.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:47:37,770 INFO [train.py:903] (0/4) Epoch 19, batch 1350, loss[loss=0.1848, simple_loss=0.2601, pruned_loss=0.05477, over 19754.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2921, pruned_loss=0.06751, over 3837476.75 frames. ], batch size: 47, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:48:22,154 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124290.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:48:39,028 INFO [train.py:903] (0/4) Epoch 19, batch 1400, loss[loss=0.2419, simple_loss=0.3182, pruned_loss=0.08283, over 19587.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2921, pruned_loss=0.06739, over 3828181.94 frames. ], batch size: 52, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:48:48,825 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.237e+02 5.508e+02 6.829e+02 9.566e+02 2.163e+03, threshold=1.366e+03, percent-clipped=9.0 +2023-04-02 13:48:58,161 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4962, 2.1895, 1.6882, 1.5603, 2.0575, 1.3850, 1.3500, 1.8357], + device='cuda:0'), covar=tensor([0.1082, 0.0806, 0.1003, 0.0764, 0.0588, 0.1210, 0.0757, 0.0515], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0314, 0.0334, 0.0259, 0.0246, 0.0336, 0.0292, 0.0272], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:49:00,431 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9346, 2.0311, 2.1791, 2.7736, 2.0410, 2.5983, 2.2554, 2.0368], + device='cuda:0'), covar=tensor([0.3936, 0.3531, 0.1792, 0.1964, 0.3699, 0.1773, 0.4369, 0.3072], + device='cuda:0'), in_proj_covar=tensor([0.0873, 0.0934, 0.0700, 0.0920, 0.0856, 0.0793, 0.0827, 0.0768], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 13:49:28,390 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8184, 4.3636, 2.7788, 3.9074, 0.9004, 4.2438, 4.1508, 4.2727], + device='cuda:0'), covar=tensor([0.0518, 0.1003, 0.1884, 0.0810, 0.4115, 0.0655, 0.0822, 0.0971], + device='cuda:0'), in_proj_covar=tensor([0.0489, 0.0395, 0.0484, 0.0340, 0.0397, 0.0423, 0.0411, 0.0446], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:49:32,999 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124349.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:49:38,316 INFO [train.py:903] (0/4) Epoch 19, batch 1450, loss[loss=0.1855, simple_loss=0.2659, pruned_loss=0.05253, over 19402.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2917, pruned_loss=0.06727, over 3834128.89 frames. ], batch size: 48, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:49:40,266 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 13:49:53,664 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124366.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 13:50:32,826 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=124398.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:50:40,521 INFO [train.py:903] (0/4) Epoch 19, batch 1500, loss[loss=0.2129, simple_loss=0.2971, pruned_loss=0.06432, over 17344.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2915, pruned_loss=0.06675, over 3830073.95 frames. ], batch size: 101, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:50:50,227 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.987e+02 4.822e+02 6.235e+02 8.378e+02 1.519e+03, threshold=1.247e+03, percent-clipped=2.0 +2023-04-02 13:50:58,616 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.56 vs. limit=2.0 +2023-04-02 13:51:04,420 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=124424.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:51:35,967 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124450.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:51:39,950 INFO [train.py:903] (0/4) Epoch 19, batch 1550, loss[loss=0.2388, simple_loss=0.3116, pruned_loss=0.08298, over 19757.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2915, pruned_loss=0.06704, over 3821588.08 frames. ], batch size: 54, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:52:40,759 INFO [train.py:903] (0/4) Epoch 19, batch 1600, loss[loss=0.2119, simple_loss=0.2934, pruned_loss=0.06518, over 19675.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2922, pruned_loss=0.0674, over 3815147.71 frames. ], batch size: 60, lr: 4.38e-03, grad_scale: 8.0 +2023-04-02 13:52:51,838 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.803e+02 4.800e+02 6.281e+02 8.115e+02 1.566e+03, threshold=1.256e+03, percent-clipped=2.0 +2023-04-02 13:52:52,207 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124513.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:53:06,488 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 13:53:23,364 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124539.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:53:40,466 INFO [train.py:903] (0/4) Epoch 19, batch 1650, loss[loss=0.2159, simple_loss=0.2981, pruned_loss=0.06686, over 19330.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.2925, pruned_loss=0.06784, over 3814619.97 frames. ], batch size: 66, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 13:54:43,096 INFO [train.py:903] (0/4) Epoch 19, batch 1700, loss[loss=0.2001, simple_loss=0.2739, pruned_loss=0.06311, over 19470.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2915, pruned_loss=0.06699, over 3815984.69 frames. ], batch size: 49, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 13:54:44,650 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124605.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:54:53,185 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.721e+02 4.942e+02 5.736e+02 7.226e+02 1.444e+03, threshold=1.147e+03, percent-clipped=3.0 +2023-04-02 13:55:14,619 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124630.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:55:18,825 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=124634.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:55:22,856 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 13:55:42,973 INFO [train.py:903] (0/4) Epoch 19, batch 1750, loss[loss=0.2112, simple_loss=0.2785, pruned_loss=0.07198, over 19465.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2922, pruned_loss=0.06734, over 3817510.80 frames. ], batch size: 49, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 13:55:45,637 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124656.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:56:16,371 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124681.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:56:20,687 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9893, 0.9528, 0.9769, 1.0367, 0.8844, 1.0089, 1.0171, 1.0049], + device='cuda:0'), covar=tensor([0.0703, 0.0771, 0.0851, 0.0559, 0.0693, 0.0740, 0.0748, 0.0625], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0220, 0.0225, 0.0243, 0.0227, 0.0211, 0.0189, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 13:56:44,205 INFO [train.py:903] (0/4) Epoch 19, batch 1800, loss[loss=0.1782, simple_loss=0.2621, pruned_loss=0.04714, over 19485.00 frames. ], tot_loss[loss=0.2132, simple_loss=0.2924, pruned_loss=0.06699, over 3826437.49 frames. ], batch size: 49, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 13:56:51,808 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=124710.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 13:56:54,991 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.267e+02 5.111e+02 6.286e+02 7.731e+02 1.656e+03, threshold=1.257e+03, percent-clipped=2.0 +2023-04-02 13:57:38,758 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 13:57:39,005 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124749.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:57:40,094 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7770, 1.4679, 1.6268, 1.6610, 4.3146, 1.1956, 2.3311, 4.5414], + device='cuda:0'), covar=tensor([0.0472, 0.2962, 0.3095, 0.2097, 0.0743, 0.2737, 0.1642, 0.0230], + device='cuda:0'), in_proj_covar=tensor([0.0398, 0.0357, 0.0377, 0.0342, 0.0366, 0.0349, 0.0369, 0.0388], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 13:57:44,336 INFO [train.py:903] (0/4) Epoch 19, batch 1850, loss[loss=0.2271, simple_loss=0.3084, pruned_loss=0.07295, over 19688.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.293, pruned_loss=0.06737, over 3813373.67 frames. ], batch size: 60, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 13:58:03,265 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124769.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:58:17,480 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 13:58:33,192 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=124794.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:58:33,420 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124794.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:58:34,542 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=124795.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:58:46,479 INFO [train.py:903] (0/4) Epoch 19, batch 1900, loss[loss=0.2034, simple_loss=0.2909, pruned_loss=0.05794, over 19336.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.292, pruned_loss=0.06656, over 3824398.10 frames. ], batch size: 66, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 13:58:56,687 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.301e+02 4.862e+02 5.969e+02 7.822e+02 1.490e+03, threshold=1.194e+03, percent-clipped=1.0 +2023-04-02 13:59:01,914 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 13:59:05,761 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=124820.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:59:07,790 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 13:59:11,246 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124825.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 13:59:26,838 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=124838.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 13:59:32,012 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 13:59:46,594 INFO [train.py:903] (0/4) Epoch 19, batch 1950, loss[loss=0.2023, simple_loss=0.2959, pruned_loss=0.05429, over 19756.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2929, pruned_loss=0.06693, over 3826793.87 frames. ], batch size: 63, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 14:00:47,546 INFO [train.py:903] (0/4) Epoch 19, batch 2000, loss[loss=0.2638, simple_loss=0.3306, pruned_loss=0.09848, over 13021.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2918, pruned_loss=0.0667, over 3825962.26 frames. ], batch size: 136, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 14:00:54,327 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=124909.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:00:58,634 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.950e+02 4.950e+02 6.179e+02 7.428e+02 1.573e+03, threshold=1.236e+03, percent-clipped=5.0 +2023-04-02 14:01:07,242 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 14:01:45,077 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 14:01:48,491 INFO [train.py:903] (0/4) Epoch 19, batch 2050, loss[loss=0.2357, simple_loss=0.3227, pruned_loss=0.07428, over 19568.00 frames. ], tot_loss[loss=0.214, simple_loss=0.2929, pruned_loss=0.06751, over 3814362.85 frames. ], batch size: 61, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 14:02:06,436 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 14:02:07,322 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 14:02:25,215 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 14:02:44,766 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=125000.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:02:50,765 INFO [train.py:903] (0/4) Epoch 19, batch 2100, loss[loss=0.2053, simple_loss=0.2917, pruned_loss=0.05943, over 19534.00 frames. ], tot_loss[loss=0.2136, simple_loss=0.2925, pruned_loss=0.06736, over 3799055.85 frames. ], batch size: 54, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 14:02:52,325 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125005.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:02:52,530 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.53 vs. limit=2.0 +2023-04-02 14:03:00,992 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.657e+02 4.808e+02 5.766e+02 7.881e+02 2.968e+03, threshold=1.153e+03, percent-clipped=4.0 +2023-04-02 14:03:15,323 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=125025.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:03:15,652 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0115, 2.0429, 2.2893, 2.8108, 2.0009, 2.6111, 2.3288, 2.1091], + device='cuda:0'), covar=tensor([0.4216, 0.3973, 0.1833, 0.2202, 0.4144, 0.2047, 0.4770, 0.3257], + device='cuda:0'), in_proj_covar=tensor([0.0869, 0.0927, 0.0696, 0.0912, 0.0853, 0.0786, 0.0824, 0.0763], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 14:03:18,743 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 14:03:19,093 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9718, 1.7872, 1.6289, 2.0615, 1.7513, 1.7113, 1.6038, 1.9151], + device='cuda:0'), covar=tensor([0.0982, 0.1294, 0.1437, 0.0822, 0.1209, 0.0570, 0.1343, 0.0701], + device='cuda:0'), in_proj_covar=tensor([0.0268, 0.0356, 0.0307, 0.0250, 0.0299, 0.0249, 0.0300, 0.0251], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:03:22,171 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125030.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:03:30,110 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4919, 1.6733, 2.1063, 1.9169, 3.2841, 2.5312, 3.4908, 1.6935], + device='cuda:0'), covar=tensor([0.2559, 0.4337, 0.2889, 0.1900, 0.1520, 0.2237, 0.1589, 0.4038], + device='cuda:0'), in_proj_covar=tensor([0.0515, 0.0624, 0.0684, 0.0467, 0.0614, 0.0518, 0.0653, 0.0532], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 14:03:39,537 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 14:03:49,806 INFO [train.py:903] (0/4) Epoch 19, batch 2150, loss[loss=0.2774, simple_loss=0.3386, pruned_loss=0.1081, over 13040.00 frames. ], tot_loss[loss=0.2136, simple_loss=0.2926, pruned_loss=0.06736, over 3788372.21 frames. ], batch size: 136, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 14:04:22,837 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125081.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 14:04:50,320 INFO [train.py:903] (0/4) Epoch 19, batch 2200, loss[loss=0.2102, simple_loss=0.2945, pruned_loss=0.06291, over 18015.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2915, pruned_loss=0.067, over 3795384.96 frames. ], batch size: 83, lr: 4.37e-03, grad_scale: 8.0 +2023-04-02 14:04:53,006 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125106.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 14:05:01,336 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.808e+02 4.810e+02 5.592e+02 6.977e+02 1.826e+03, threshold=1.118e+03, percent-clipped=4.0 +2023-04-02 14:05:04,001 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125115.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:05:33,638 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125140.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:05:40,923 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0 +2023-04-02 14:05:41,554 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125146.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:05:51,114 INFO [train.py:903] (0/4) Epoch 19, batch 2250, loss[loss=0.2062, simple_loss=0.2913, pruned_loss=0.06054, over 19485.00 frames. ], tot_loss[loss=0.213, simple_loss=0.292, pruned_loss=0.06697, over 3800437.89 frames. ], batch size: 49, lr: 4.36e-03, grad_scale: 8.0 +2023-04-02 14:05:52,659 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8935, 1.7544, 1.5380, 2.0245, 1.6833, 1.6369, 1.5869, 1.7671], + device='cuda:0'), covar=tensor([0.1013, 0.1452, 0.1486, 0.0914, 0.1288, 0.0565, 0.1328, 0.0759], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0356, 0.0307, 0.0249, 0.0299, 0.0249, 0.0300, 0.0251], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:06:05,776 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125165.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:06:25,680 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=125182.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:06:35,049 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125190.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:06:52,195 INFO [train.py:903] (0/4) Epoch 19, batch 2300, loss[loss=0.2584, simple_loss=0.3309, pruned_loss=0.09291, over 17300.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2916, pruned_loss=0.06666, over 3815378.94 frames. ], batch size: 101, lr: 4.36e-03, grad_scale: 4.0 +2023-04-02 14:07:04,347 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.756e+02 5.218e+02 6.176e+02 8.185e+02 2.110e+03, threshold=1.235e+03, percent-clipped=6.0 +2023-04-02 14:07:06,714 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 14:07:52,829 INFO [train.py:903] (0/4) Epoch 19, batch 2350, loss[loss=0.1807, simple_loss=0.2609, pruned_loss=0.05026, over 19371.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2907, pruned_loss=0.06644, over 3810877.87 frames. ], batch size: 47, lr: 4.36e-03, grad_scale: 4.0 +2023-04-02 14:08:33,595 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 14:08:45,060 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125297.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:08:49,235 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 14:08:52,752 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2212, 1.2977, 1.4327, 1.5433, 1.2176, 1.5093, 1.4900, 1.3592], + device='cuda:0'), covar=tensor([0.2663, 0.2193, 0.1307, 0.1470, 0.2390, 0.1345, 0.3111, 0.2267], + device='cuda:0'), in_proj_covar=tensor([0.0874, 0.0934, 0.0700, 0.0921, 0.0859, 0.0789, 0.0831, 0.0768], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 14:08:53,418 INFO [train.py:903] (0/4) Epoch 19, batch 2400, loss[loss=0.2973, simple_loss=0.356, pruned_loss=0.1194, over 13851.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2923, pruned_loss=0.06734, over 3797107.54 frames. ], batch size: 137, lr: 4.36e-03, grad_scale: 8.0 +2023-04-02 14:08:54,899 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8452, 3.3206, 3.3461, 3.3650, 1.4211, 3.2241, 2.8236, 3.1061], + device='cuda:0'), covar=tensor([0.1635, 0.0877, 0.0804, 0.0910, 0.5076, 0.0950, 0.0802, 0.1349], + device='cuda:0'), in_proj_covar=tensor([0.0765, 0.0710, 0.0916, 0.0806, 0.0816, 0.0666, 0.0554, 0.0849], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 14:09:05,368 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.876e+02 4.925e+02 6.150e+02 7.515e+02 1.529e+03, threshold=1.230e+03, percent-clipped=3.0 +2023-04-02 14:09:45,494 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-04-02 14:09:53,248 INFO [train.py:903] (0/4) Epoch 19, batch 2450, loss[loss=0.1936, simple_loss=0.2696, pruned_loss=0.05882, over 19401.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2903, pruned_loss=0.06625, over 3802711.01 frames. ], batch size: 48, lr: 4.36e-03, grad_scale: 8.0 +2023-04-02 14:10:14,728 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125371.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:10:17,759 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125374.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:10:41,851 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125394.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:10:44,299 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125396.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:10:44,329 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125396.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:10:53,819 INFO [train.py:903] (0/4) Epoch 19, batch 2500, loss[loss=0.2179, simple_loss=0.3004, pruned_loss=0.06764, over 19796.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2911, pruned_loss=0.067, over 3797807.13 frames. ], batch size: 63, lr: 4.36e-03, grad_scale: 8.0 +2023-04-02 14:11:05,671 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.142e+02 4.898e+02 6.460e+02 8.264e+02 2.020e+03, threshold=1.292e+03, percent-clipped=4.0 +2023-04-02 14:11:13,967 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125421.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:11:35,802 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125439.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:11:54,216 INFO [train.py:903] (0/4) Epoch 19, batch 2550, loss[loss=0.2114, simple_loss=0.2961, pruned_loss=0.06332, over 19577.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.291, pruned_loss=0.06691, over 3814847.90 frames. ], batch size: 61, lr: 4.36e-03, grad_scale: 8.0 +2023-04-02 14:11:56,653 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3288, 3.0666, 2.1588, 2.7550, 0.8557, 2.9427, 2.8592, 2.9717], + device='cuda:0'), covar=tensor([0.1158, 0.1446, 0.2376, 0.1090, 0.3888, 0.1059, 0.1194, 0.1673], + device='cuda:0'), in_proj_covar=tensor([0.0486, 0.0396, 0.0480, 0.0337, 0.0394, 0.0420, 0.0411, 0.0445], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:12:38,052 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=125490.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:12:47,108 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 14:12:53,797 INFO [train.py:903] (0/4) Epoch 19, batch 2600, loss[loss=0.2307, simple_loss=0.3126, pruned_loss=0.07442, over 19440.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2907, pruned_loss=0.0666, over 3824886.48 frames. ], batch size: 64, lr: 4.36e-03, grad_scale: 8.0 +2023-04-02 14:13:05,879 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.683e+02 4.815e+02 5.841e+02 7.665e+02 1.339e+03, threshold=1.168e+03, percent-clipped=2.0 +2023-04-02 14:13:21,432 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1050, 1.3713, 1.7371, 1.2187, 2.5956, 3.5794, 3.2977, 3.7831], + device='cuda:0'), covar=tensor([0.1600, 0.3620, 0.3201, 0.2398, 0.0583, 0.0180, 0.0208, 0.0228], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0311, 0.0343, 0.0260, 0.0237, 0.0179, 0.0212, 0.0244], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 14:13:48,578 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.50 vs. limit=5.0 +2023-04-02 14:13:52,883 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125553.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:13:53,662 INFO [train.py:903] (0/4) Epoch 19, batch 2650, loss[loss=0.2108, simple_loss=0.2911, pruned_loss=0.06525, over 19575.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2921, pruned_loss=0.0677, over 3834606.93 frames. ], batch size: 52, lr: 4.36e-03, grad_scale: 8.0 +2023-04-02 14:14:15,462 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 14:14:23,600 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125578.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:14:54,493 INFO [train.py:903] (0/4) Epoch 19, batch 2700, loss[loss=0.224, simple_loss=0.3051, pruned_loss=0.07144, over 19527.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2902, pruned_loss=0.06664, over 3833363.74 frames. ], batch size: 56, lr: 4.36e-03, grad_scale: 8.0 +2023-04-02 14:14:55,974 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125605.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:15:07,179 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.594e+02 4.817e+02 5.964e+02 7.468e+02 1.608e+03, threshold=1.193e+03, percent-clipped=5.0 +2023-04-02 14:15:15,684 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125621.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:15:31,466 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8138, 0.8921, 0.8600, 0.7465, 0.7287, 0.7504, 0.0704, 0.2371], + device='cuda:0'), covar=tensor([0.0463, 0.0444, 0.0297, 0.0419, 0.0737, 0.0487, 0.1039, 0.0731], + device='cuda:0'), in_proj_covar=tensor([0.0354, 0.0349, 0.0351, 0.0374, 0.0451, 0.0385, 0.0331, 0.0339], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 14:15:35,557 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0792, 1.5874, 2.1721, 1.6009, 4.5949, 0.9461, 2.4504, 4.9125], + device='cuda:0'), covar=tensor([0.0382, 0.2641, 0.2385, 0.1976, 0.0663, 0.2741, 0.1424, 0.0176], + device='cuda:0'), in_proj_covar=tensor([0.0397, 0.0358, 0.0376, 0.0342, 0.0365, 0.0349, 0.0368, 0.0387], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:15:56,194 INFO [train.py:903] (0/4) Epoch 19, batch 2750, loss[loss=0.216, simple_loss=0.2873, pruned_loss=0.0723, over 19773.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2911, pruned_loss=0.06715, over 3824027.89 frames. ], batch size: 54, lr: 4.36e-03, grad_scale: 8.0 +2023-04-02 14:16:05,817 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3849, 1.5262, 1.8938, 1.6667, 2.5977, 2.1351, 2.6766, 1.2619], + device='cuda:0'), covar=tensor([0.2574, 0.4238, 0.2495, 0.2049, 0.1571, 0.2269, 0.1589, 0.4305], + device='cuda:0'), in_proj_covar=tensor([0.0518, 0.0625, 0.0688, 0.0470, 0.0614, 0.0521, 0.0655, 0.0534], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 14:16:55,539 INFO [train.py:903] (0/4) Epoch 19, batch 2800, loss[loss=0.2055, simple_loss=0.2839, pruned_loss=0.06359, over 19660.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2908, pruned_loss=0.06736, over 3823520.41 frames. ], batch size: 53, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:17:08,445 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.000e+02 5.188e+02 6.338e+02 7.813e+02 1.733e+03, threshold=1.268e+03, percent-clipped=8.0 +2023-04-02 14:17:12,864 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=125718.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:17:37,004 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=125738.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:17:39,791 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-02 14:17:42,815 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=125743.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:17:55,254 INFO [train.py:903] (0/4) Epoch 19, batch 2850, loss[loss=0.1743, simple_loss=0.2543, pruned_loss=0.04713, over 19393.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2902, pruned_loss=0.06702, over 3832504.53 frames. ], batch size: 48, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:18:25,424 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1925, 1.3379, 1.8459, 1.3704, 2.8326, 3.7018, 3.4629, 3.9070], + device='cuda:0'), covar=tensor([0.1646, 0.3712, 0.3105, 0.2345, 0.0599, 0.0238, 0.0201, 0.0270], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0312, 0.0343, 0.0260, 0.0238, 0.0179, 0.0212, 0.0243], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 14:18:30,662 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=125783.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:18:56,092 INFO [train.py:903] (0/4) Epoch 19, batch 2900, loss[loss=0.1789, simple_loss=0.2593, pruned_loss=0.0492, over 19590.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2912, pruned_loss=0.06714, over 3836827.00 frames. ], batch size: 50, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:18:56,101 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 14:19:01,432 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3033, 3.8952, 2.6405, 3.5067, 1.0944, 3.8534, 3.7626, 3.7977], + device='cuda:0'), covar=tensor([0.0759, 0.1083, 0.2013, 0.0846, 0.3721, 0.0715, 0.0862, 0.1186], + device='cuda:0'), in_proj_covar=tensor([0.0487, 0.0398, 0.0481, 0.0338, 0.0395, 0.0419, 0.0410, 0.0446], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:19:09,044 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.234e+02 4.625e+02 5.511e+02 7.350e+02 1.619e+03, threshold=1.102e+03, percent-clipped=3.0 +2023-04-02 14:19:31,527 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125833.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:19:55,773 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125853.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:19:56,475 INFO [train.py:903] (0/4) Epoch 19, batch 2950, loss[loss=0.2018, simple_loss=0.2906, pruned_loss=0.05656, over 19263.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2908, pruned_loss=0.06659, over 3841334.58 frames. ], batch size: 70, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:20:05,702 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=125861.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:20:35,710 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=125886.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:20:50,774 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=125898.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:20:57,388 INFO [train.py:903] (0/4) Epoch 19, batch 3000, loss[loss=0.181, simple_loss=0.257, pruned_loss=0.05243, over 19391.00 frames. ], tot_loss[loss=0.2136, simple_loss=0.2921, pruned_loss=0.06758, over 3810125.85 frames. ], batch size: 48, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:20:57,389 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 14:21:10,743 INFO [train.py:937] (0/4) Epoch 19, validation: loss=0.1696, simple_loss=0.2702, pruned_loss=0.03451, over 944034.00 frames. +2023-04-02 14:21:10,745 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 14:21:10,803 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 14:21:24,049 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.456e+02 4.999e+02 6.816e+02 8.693e+02 1.814e+03, threshold=1.363e+03, percent-clipped=12.0 +2023-04-02 14:22:11,566 INFO [train.py:903] (0/4) Epoch 19, batch 3050, loss[loss=0.2433, simple_loss=0.3191, pruned_loss=0.08375, over 14021.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2916, pruned_loss=0.06729, over 3808739.66 frames. ], batch size: 136, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:22:24,990 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=125965.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:23:08,041 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-126000.pt +2023-04-02 14:23:13,568 INFO [train.py:903] (0/4) Epoch 19, batch 3100, loss[loss=0.2096, simple_loss=0.2894, pruned_loss=0.06491, over 19738.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2907, pruned_loss=0.06643, over 3818748.31 frames. ], batch size: 63, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:23:14,309 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-02 14:23:26,817 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.089e+02 4.651e+02 5.591e+02 6.916e+02 1.279e+03, threshold=1.118e+03, percent-clipped=0.0 +2023-04-02 14:23:28,003 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=126016.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:24:13,741 INFO [train.py:903] (0/4) Epoch 19, batch 3150, loss[loss=0.1639, simple_loss=0.2428, pruned_loss=0.04256, over 19048.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.2896, pruned_loss=0.06567, over 3826822.05 frames. ], batch size: 42, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:24:40,336 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 14:24:45,984 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126080.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:24:48,473 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-02 14:24:53,523 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=126087.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:24:55,901 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126089.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:25:14,240 INFO [train.py:903] (0/4) Epoch 19, batch 3200, loss[loss=0.1956, simple_loss=0.2749, pruned_loss=0.05813, over 19483.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2896, pruned_loss=0.06559, over 3817991.22 frames. ], batch size: 49, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:25:21,274 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126109.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:25:26,792 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=126114.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:25:27,594 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.283e+02 4.817e+02 6.226e+02 7.515e+02 1.545e+03, threshold=1.245e+03, percent-clipped=7.0 +2023-04-02 14:25:51,008 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=126134.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:26:15,153 INFO [train.py:903] (0/4) Epoch 19, batch 3250, loss[loss=0.2288, simple_loss=0.3109, pruned_loss=0.07334, over 19659.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2902, pruned_loss=0.06602, over 3816363.48 frames. ], batch size: 60, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:26:15,589 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126154.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:26:32,405 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0059, 1.6228, 1.8505, 2.4272, 1.9307, 2.2904, 2.3417, 2.0580], + device='cuda:0'), covar=tensor([0.0795, 0.0925, 0.0966, 0.0930, 0.0850, 0.0712, 0.0827, 0.0666], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0227, 0.0245, 0.0227, 0.0212, 0.0189, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 14:26:46,058 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=126179.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:27:14,808 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126202.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:27:16,669 INFO [train.py:903] (0/4) Epoch 19, batch 3300, loss[loss=0.2274, simple_loss=0.3044, pruned_loss=0.07519, over 19767.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2894, pruned_loss=0.06539, over 3832678.21 frames. ], batch size: 56, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:27:20,149 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 14:27:30,250 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.439e+02 5.162e+02 6.410e+02 7.971e+02 2.422e+03, threshold=1.282e+03, percent-clipped=4.0 +2023-04-02 14:27:36,966 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3768, 3.9809, 2.7315, 3.5268, 0.9580, 3.8911, 3.7609, 3.9025], + device='cuda:0'), covar=tensor([0.0702, 0.1029, 0.1843, 0.0911, 0.4000, 0.0705, 0.0955, 0.1113], + device='cuda:0'), in_proj_covar=tensor([0.0488, 0.0394, 0.0479, 0.0337, 0.0394, 0.0420, 0.0410, 0.0446], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:28:17,436 INFO [train.py:903] (0/4) Epoch 19, batch 3350, loss[loss=0.2566, simple_loss=0.3176, pruned_loss=0.09778, over 13210.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2901, pruned_loss=0.06586, over 3837607.64 frames. ], batch size: 136, lr: 4.35e-03, grad_scale: 8.0 +2023-04-02 14:29:08,277 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0568, 2.8333, 2.1522, 2.0486, 1.8458, 2.3738, 0.9417, 2.0250], + device='cuda:0'), covar=tensor([0.0676, 0.0598, 0.0695, 0.1182, 0.1128, 0.1044, 0.1368, 0.1035], + device='cuda:0'), in_proj_covar=tensor([0.0357, 0.0350, 0.0352, 0.0377, 0.0453, 0.0385, 0.0332, 0.0341], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 14:29:17,957 INFO [train.py:903] (0/4) Epoch 19, batch 3400, loss[loss=0.2068, simple_loss=0.2929, pruned_loss=0.06038, over 19676.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.29, pruned_loss=0.06578, over 3835799.20 frames. ], batch size: 58, lr: 4.34e-03, grad_scale: 8.0 +2023-04-02 14:29:25,173 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3436, 1.4719, 1.8198, 1.6013, 2.6006, 2.2364, 2.7575, 1.0503], + device='cuda:0'), covar=tensor([0.2462, 0.4178, 0.2563, 0.1959, 0.1627, 0.2112, 0.1432, 0.4437], + device='cuda:0'), in_proj_covar=tensor([0.0519, 0.0627, 0.0689, 0.0474, 0.0617, 0.0521, 0.0656, 0.0536], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 14:29:31,306 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.222e+02 5.180e+02 6.014e+02 8.021e+02 1.733e+03, threshold=1.203e+03, percent-clipped=4.0 +2023-04-02 14:29:57,167 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126336.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:30:18,214 INFO [train.py:903] (0/4) Epoch 19, batch 3450, loss[loss=0.239, simple_loss=0.3117, pruned_loss=0.08316, over 19789.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2905, pruned_loss=0.06641, over 3840794.92 frames. ], batch size: 56, lr: 4.34e-03, grad_scale: 4.0 +2023-04-02 14:30:22,539 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 14:30:27,158 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=126360.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:30:28,561 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=126361.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:31:13,294 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=126398.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:31:20,718 INFO [train.py:903] (0/4) Epoch 19, batch 3500, loss[loss=0.2206, simple_loss=0.3068, pruned_loss=0.06719, over 19686.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2904, pruned_loss=0.0662, over 3836906.61 frames. ], batch size: 60, lr: 4.34e-03, grad_scale: 4.0 +2023-04-02 14:31:34,959 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.238e+02 4.908e+02 6.053e+02 7.325e+02 1.346e+03, threshold=1.211e+03, percent-clipped=1.0 +2023-04-02 14:32:21,699 INFO [train.py:903] (0/4) Epoch 19, batch 3550, loss[loss=0.2038, simple_loss=0.2867, pruned_loss=0.06049, over 19503.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2898, pruned_loss=0.06604, over 3827155.83 frames. ], batch size: 64, lr: 4.34e-03, grad_scale: 4.0 +2023-04-02 14:32:23,009 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2983, 1.5374, 1.8971, 1.3669, 2.6384, 3.3416, 3.1712, 3.4959], + device='cuda:0'), covar=tensor([0.1634, 0.3466, 0.3041, 0.2416, 0.0700, 0.0284, 0.0217, 0.0313], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0313, 0.0344, 0.0260, 0.0238, 0.0181, 0.0214, 0.0245], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 14:32:26,660 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126458.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:32:38,970 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=126468.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:32:48,119 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126475.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:32:57,946 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=126483.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:33:21,942 INFO [train.py:903] (0/4) Epoch 19, batch 3600, loss[loss=0.2459, simple_loss=0.3338, pruned_loss=0.07899, over 19457.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2906, pruned_loss=0.06628, over 3811865.42 frames. ], batch size: 70, lr: 4.34e-03, grad_scale: 8.0 +2023-04-02 14:33:37,205 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.060e+02 4.926e+02 5.826e+02 7.456e+02 2.258e+03, threshold=1.165e+03, percent-clipped=2.0 +2023-04-02 14:34:04,292 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2114, 1.8508, 1.5175, 1.2516, 1.6699, 1.2349, 1.2453, 1.7333], + device='cuda:0'), covar=tensor([0.0669, 0.0803, 0.1062, 0.0785, 0.0528, 0.1237, 0.0562, 0.0381], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0311, 0.0332, 0.0260, 0.0243, 0.0333, 0.0289, 0.0271], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:34:23,092 INFO [train.py:903] (0/4) Epoch 19, batch 3650, loss[loss=0.2155, simple_loss=0.2881, pruned_loss=0.07146, over 19591.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2904, pruned_loss=0.06646, over 3811194.47 frames. ], batch size: 52, lr: 4.34e-03, grad_scale: 8.0 +2023-04-02 14:35:24,555 INFO [train.py:903] (0/4) Epoch 19, batch 3700, loss[loss=0.1899, simple_loss=0.2611, pruned_loss=0.0594, over 19767.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2907, pruned_loss=0.06705, over 3812482.22 frames. ], batch size: 48, lr: 4.34e-03, grad_scale: 8.0 +2023-04-02 14:35:38,476 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.032e+02 5.326e+02 6.409e+02 8.349e+02 1.648e+03, threshold=1.282e+03, percent-clipped=7.0 +2023-04-02 14:36:23,996 INFO [train.py:903] (0/4) Epoch 19, batch 3750, loss[loss=0.2156, simple_loss=0.2995, pruned_loss=0.06587, over 19789.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2919, pruned_loss=0.0678, over 3811305.57 frames. ], batch size: 56, lr: 4.34e-03, grad_scale: 8.0 +2023-04-02 14:36:26,350 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2875, 1.2993, 1.4634, 1.3970, 1.9457, 1.7489, 1.9448, 1.0364], + device='cuda:0'), covar=tensor([0.1728, 0.3083, 0.1915, 0.1418, 0.1114, 0.1663, 0.1037, 0.3567], + device='cuda:0'), in_proj_covar=tensor([0.0518, 0.0627, 0.0690, 0.0473, 0.0615, 0.0521, 0.0656, 0.0535], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 14:37:25,131 INFO [train.py:903] (0/4) Epoch 19, batch 3800, loss[loss=0.1856, simple_loss=0.275, pruned_loss=0.04809, over 19667.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2916, pruned_loss=0.06748, over 3801843.64 frames. ], batch size: 55, lr: 4.34e-03, grad_scale: 8.0 +2023-04-02 14:37:40,984 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.076e+02 4.757e+02 5.693e+02 7.302e+02 1.543e+03, threshold=1.139e+03, percent-clipped=1.0 +2023-04-02 14:37:57,319 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 14:37:58,864 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=126731.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:38:12,086 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=126742.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:38:24,575 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9088, 4.4591, 2.6484, 3.8493, 0.7745, 4.3359, 4.2615, 4.3438], + device='cuda:0'), covar=tensor([0.0612, 0.1003, 0.2000, 0.0826, 0.4380, 0.0725, 0.0896, 0.1046], + device='cuda:0'), in_proj_covar=tensor([0.0491, 0.0397, 0.0483, 0.0340, 0.0399, 0.0421, 0.0411, 0.0448], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:38:26,704 INFO [train.py:903] (0/4) Epoch 19, batch 3850, loss[loss=0.2164, simple_loss=0.3016, pruned_loss=0.06564, over 19530.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2922, pruned_loss=0.06743, over 3816260.38 frames. ], batch size: 56, lr: 4.34e-03, grad_scale: 8.0 +2023-04-02 14:38:30,264 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=126756.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:39:28,456 INFO [train.py:903] (0/4) Epoch 19, batch 3900, loss[loss=0.2525, simple_loss=0.3263, pruned_loss=0.08932, over 19783.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.2914, pruned_loss=0.06724, over 3816593.40 frames. ], batch size: 56, lr: 4.34e-03, grad_scale: 8.0 +2023-04-02 14:39:34,823 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.75 vs. limit=2.0 +2023-04-02 14:39:37,770 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=126812.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:39:42,911 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.374e+02 5.082e+02 6.454e+02 7.734e+02 3.345e+03, threshold=1.291e+03, percent-clipped=6.0 +2023-04-02 14:40:04,537 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=126834.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:40:29,133 INFO [train.py:903] (0/4) Epoch 19, batch 3950, loss[loss=0.2277, simple_loss=0.3056, pruned_loss=0.07491, over 19523.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.2916, pruned_loss=0.06712, over 3802839.51 frames. ], batch size: 56, lr: 4.34e-03, grad_scale: 8.0 +2023-04-02 14:40:33,507 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126857.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:40:35,252 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 14:41:14,050 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.67 vs. limit=5.0 +2023-04-02 14:41:29,518 INFO [train.py:903] (0/4) Epoch 19, batch 4000, loss[loss=0.2386, simple_loss=0.3127, pruned_loss=0.08228, over 19735.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2914, pruned_loss=0.06761, over 3802800.87 frames. ], batch size: 63, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:41:43,556 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.579e+02 4.980e+02 6.258e+02 9.023e+02 1.716e+03, threshold=1.252e+03, percent-clipped=4.0 +2023-04-02 14:41:57,623 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=126927.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:42:16,841 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 14:42:30,358 INFO [train.py:903] (0/4) Epoch 19, batch 4050, loss[loss=0.2196, simple_loss=0.2874, pruned_loss=0.07592, over 19409.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2924, pruned_loss=0.06808, over 3791520.48 frames. ], batch size: 48, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:42:53,012 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.79 vs. limit=5.0 +2023-04-02 14:43:30,606 INFO [train.py:903] (0/4) Epoch 19, batch 4100, loss[loss=0.2014, simple_loss=0.2827, pruned_loss=0.06012, over 19676.00 frames. ], tot_loss[loss=0.2142, simple_loss=0.2927, pruned_loss=0.06782, over 3804495.85 frames. ], batch size: 53, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:43:45,831 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.326e+02 4.699e+02 5.681e+02 7.096e+02 1.300e+03, threshold=1.136e+03, percent-clipped=1.0 +2023-04-02 14:44:06,972 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 14:44:31,635 INFO [train.py:903] (0/4) Epoch 19, batch 4150, loss[loss=0.2151, simple_loss=0.2986, pruned_loss=0.06582, over 19773.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.2926, pruned_loss=0.06775, over 3799020.78 frames. ], batch size: 54, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:44:44,492 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.91 vs. limit=2.0 +2023-04-02 14:45:04,800 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0505, 5.0509, 5.9153, 5.9227, 2.0291, 5.5725, 4.7269, 5.5380], + device='cuda:0'), covar=tensor([0.1555, 0.0975, 0.0532, 0.0555, 0.6192, 0.0697, 0.0581, 0.1109], + device='cuda:0'), in_proj_covar=tensor([0.0770, 0.0716, 0.0921, 0.0808, 0.0821, 0.0675, 0.0556, 0.0858], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 14:45:32,546 INFO [train.py:903] (0/4) Epoch 19, batch 4200, loss[loss=0.2478, simple_loss=0.332, pruned_loss=0.0818, over 17644.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2919, pruned_loss=0.06712, over 3804018.04 frames. ], batch size: 102, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:45:35,888 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 14:45:42,373 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-02 14:45:43,054 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127113.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:45:46,907 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.729e+02 4.915e+02 5.762e+02 6.825e+02 1.362e+03, threshold=1.152e+03, percent-clipped=8.0 +2023-04-02 14:46:01,602 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.8136, 5.2662, 3.1023, 4.6241, 1.0186, 5.3226, 5.1752, 5.3584], + device='cuda:0'), covar=tensor([0.0393, 0.0783, 0.1834, 0.0671, 0.4074, 0.0520, 0.0724, 0.0927], + device='cuda:0'), in_proj_covar=tensor([0.0488, 0.0394, 0.0482, 0.0337, 0.0397, 0.0419, 0.0410, 0.0445], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:46:14,862 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127138.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:46:32,718 INFO [train.py:903] (0/4) Epoch 19, batch 4250, loss[loss=0.2252, simple_loss=0.3019, pruned_loss=0.07425, over 19601.00 frames. ], tot_loss[loss=0.214, simple_loss=0.293, pruned_loss=0.06751, over 3816359.22 frames. ], batch size: 52, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:46:50,062 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 14:47:01,523 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 14:47:03,640 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=127178.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:47:09,520 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127183.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:47:20,953 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-04-02 14:47:34,699 INFO [train.py:903] (0/4) Epoch 19, batch 4300, loss[loss=0.1854, simple_loss=0.2747, pruned_loss=0.04804, over 19707.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2919, pruned_loss=0.06671, over 3819270.09 frames. ], batch size: 59, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:47:40,383 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127208.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:47:50,084 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.800e+02 4.893e+02 5.914e+02 7.996e+02 1.682e+03, threshold=1.183e+03, percent-clipped=7.0 +2023-04-02 14:48:17,091 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-04-02 14:48:28,064 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 14:48:35,537 INFO [train.py:903] (0/4) Epoch 19, batch 4350, loss[loss=0.1866, simple_loss=0.2648, pruned_loss=0.05418, over 19397.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2923, pruned_loss=0.06719, over 3815326.42 frames. ], batch size: 48, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:48:51,139 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127267.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:49:23,180 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=127293.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:49:35,822 INFO [train.py:903] (0/4) Epoch 19, batch 4400, loss[loss=0.2135, simple_loss=0.2973, pruned_loss=0.06481, over 19494.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2924, pruned_loss=0.06758, over 3804097.33 frames. ], batch size: 64, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:49:45,411 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5627, 1.1782, 1.3350, 1.2537, 2.2142, 0.9927, 1.9721, 2.5150], + device='cuda:0'), covar=tensor([0.0688, 0.2685, 0.2793, 0.1674, 0.0897, 0.2180, 0.1079, 0.0423], + device='cuda:0'), in_proj_covar=tensor([0.0394, 0.0357, 0.0376, 0.0342, 0.0366, 0.0348, 0.0368, 0.0385], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:49:49,568 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.177e+02 5.239e+02 6.175e+02 6.853e+02 1.670e+03, threshold=1.235e+03, percent-clipped=2.0 +2023-04-02 14:50:02,112 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 14:50:12,009 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 14:50:22,516 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4840, 1.7053, 2.0183, 1.7466, 3.1635, 2.5385, 3.4602, 1.5521], + device='cuda:0'), covar=tensor([0.2431, 0.4044, 0.2507, 0.1887, 0.1528, 0.2074, 0.1540, 0.4085], + device='cuda:0'), in_proj_covar=tensor([0.0520, 0.0626, 0.0691, 0.0472, 0.0616, 0.0518, 0.0659, 0.0536], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 14:50:36,323 INFO [train.py:903] (0/4) Epoch 19, batch 4450, loss[loss=0.1849, simple_loss=0.2712, pruned_loss=0.04927, over 19844.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2919, pruned_loss=0.0673, over 3803107.19 frames. ], batch size: 52, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:50:46,010 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-02 14:51:38,005 INFO [train.py:903] (0/4) Epoch 19, batch 4500, loss[loss=0.2132, simple_loss=0.2964, pruned_loss=0.06503, over 19574.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2907, pruned_loss=0.0665, over 3810060.30 frames. ], batch size: 52, lr: 4.33e-03, grad_scale: 8.0 +2023-04-02 14:51:52,896 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.369e+02 5.116e+02 6.133e+02 7.767e+02 1.446e+03, threshold=1.227e+03, percent-clipped=3.0 +2023-04-02 14:51:57,310 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1068, 1.1795, 1.4018, 1.3468, 2.7234, 1.0830, 2.1727, 3.0567], + device='cuda:0'), covar=tensor([0.0573, 0.2863, 0.2916, 0.1904, 0.0751, 0.2447, 0.1195, 0.0328], + device='cuda:0'), in_proj_covar=tensor([0.0393, 0.0357, 0.0376, 0.0341, 0.0367, 0.0347, 0.0368, 0.0385], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:52:28,114 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127445.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 14:52:39,719 INFO [train.py:903] (0/4) Epoch 19, batch 4550, loss[loss=0.1609, simple_loss=0.2405, pruned_loss=0.04068, over 19771.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2906, pruned_loss=0.0665, over 3810268.32 frames. ], batch size: 47, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 14:52:48,347 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 14:53:11,956 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 14:53:40,574 INFO [train.py:903] (0/4) Epoch 19, batch 4600, loss[loss=0.1968, simple_loss=0.2784, pruned_loss=0.05759, over 19754.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2904, pruned_loss=0.06661, over 3802661.44 frames. ], batch size: 54, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 14:53:48,015 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 14:53:52,390 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8647, 1.7418, 1.7744, 2.2044, 1.9291, 1.9931, 2.0175, 1.9678], + device='cuda:0'), covar=tensor([0.0687, 0.0768, 0.0826, 0.0618, 0.0805, 0.0700, 0.0800, 0.0573], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0223, 0.0227, 0.0246, 0.0230, 0.0213, 0.0190, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 14:53:54,244 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.388e+02 5.018e+02 6.286e+02 8.427e+02 2.189e+03, threshold=1.257e+03, percent-clipped=8.0 +2023-04-02 14:54:32,443 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127547.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:54:34,745 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127549.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:54:39,912 INFO [train.py:903] (0/4) Epoch 19, batch 4650, loss[loss=0.249, simple_loss=0.3252, pruned_loss=0.08645, over 19731.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2912, pruned_loss=0.06689, over 3802160.47 frames. ], batch size: 63, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 14:54:56,755 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 14:55:02,163 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2571, 2.3922, 2.5653, 3.0550, 2.2235, 2.9219, 2.5636, 2.2372], + device='cuda:0'), covar=tensor([0.4355, 0.3724, 0.1875, 0.2368, 0.4308, 0.2080, 0.4733, 0.3435], + device='cuda:0'), in_proj_covar=tensor([0.0879, 0.0937, 0.0702, 0.0924, 0.0861, 0.0796, 0.0831, 0.0768], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 14:55:05,404 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=127574.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:55:07,390 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 14:55:40,717 INFO [train.py:903] (0/4) Epoch 19, batch 4700, loss[loss=0.2011, simple_loss=0.2832, pruned_loss=0.05951, over 19660.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2896, pruned_loss=0.06615, over 3802878.69 frames. ], batch size: 58, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 14:55:41,052 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3780, 1.0441, 1.2484, 2.1072, 1.5055, 1.3781, 1.6078, 1.3464], + device='cuda:0'), covar=tensor([0.1209, 0.1765, 0.1403, 0.1019, 0.1170, 0.1398, 0.1328, 0.1108], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0222, 0.0226, 0.0245, 0.0229, 0.0211, 0.0189, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 14:55:50,499 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=127611.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:55:55,863 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.385e+02 5.036e+02 6.203e+02 8.078e+02 1.735e+03, threshold=1.241e+03, percent-clipped=3.0 +2023-04-02 14:56:02,615 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 14:56:41,689 INFO [train.py:903] (0/4) Epoch 19, batch 4750, loss[loss=0.1909, simple_loss=0.282, pruned_loss=0.04989, over 19588.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.29, pruned_loss=0.06637, over 3817570.62 frames. ], batch size: 57, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 14:57:13,610 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127680.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:57:33,554 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5913, 1.0843, 1.3353, 1.3403, 2.2344, 1.0439, 2.0843, 2.4149], + device='cuda:0'), covar=tensor([0.0710, 0.2816, 0.2808, 0.1563, 0.0843, 0.2069, 0.1030, 0.0497], + device='cuda:0'), in_proj_covar=tensor([0.0395, 0.0359, 0.0377, 0.0341, 0.0367, 0.0347, 0.0369, 0.0387], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:57:41,809 INFO [train.py:903] (0/4) Epoch 19, batch 4800, loss[loss=0.2114, simple_loss=0.2722, pruned_loss=0.07526, over 19300.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2899, pruned_loss=0.06664, over 3808736.36 frames. ], batch size: 44, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 14:57:55,391 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.353e+02 4.990e+02 6.329e+02 8.030e+02 1.437e+03, threshold=1.266e+03, percent-clipped=1.0 +2023-04-02 14:58:07,600 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=127726.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:58:40,155 INFO [train.py:903] (0/4) Epoch 19, batch 4850, loss[loss=0.1924, simple_loss=0.2662, pruned_loss=0.05932, over 18339.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2912, pruned_loss=0.06699, over 3815464.44 frames. ], batch size: 40, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 14:59:04,712 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 14:59:22,981 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=127789.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 14:59:25,083 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 14:59:30,832 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 14:59:30,852 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 14:59:32,266 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127797.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 14:59:35,945 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0142, 1.9190, 1.6378, 2.1576, 1.8452, 1.8050, 1.6588, 1.8831], + device='cuda:0'), covar=tensor([0.1029, 0.1312, 0.1480, 0.0891, 0.1237, 0.0519, 0.1308, 0.0732], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0351, 0.0305, 0.0248, 0.0296, 0.0246, 0.0296, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 14:59:40,801 INFO [train.py:903] (0/4) Epoch 19, batch 4900, loss[loss=0.2264, simple_loss=0.3089, pruned_loss=0.07201, over 19318.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2906, pruned_loss=0.06612, over 3821117.15 frames. ], batch size: 66, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 14:59:40,839 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 14:59:55,910 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.219e+02 4.848e+02 5.865e+02 7.992e+02 2.664e+03, threshold=1.173e+03, percent-clipped=3.0 +2023-04-02 15:00:01,796 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 15:00:41,586 INFO [train.py:903] (0/4) Epoch 19, batch 4950, loss[loss=0.1954, simple_loss=0.2721, pruned_loss=0.05932, over 19432.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2908, pruned_loss=0.06617, over 3821523.12 frames. ], batch size: 48, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 15:00:58,818 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127868.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:00:59,650 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 15:01:22,184 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 15:01:23,565 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1445, 1.2514, 1.6801, 1.1551, 2.4033, 3.3817, 3.0907, 3.5742], + device='cuda:0'), covar=tensor([0.1729, 0.3812, 0.3301, 0.2442, 0.0603, 0.0177, 0.0236, 0.0272], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0315, 0.0344, 0.0260, 0.0237, 0.0181, 0.0213, 0.0245], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 15:01:24,516 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=127890.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:01:26,348 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=127891.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:01:41,518 INFO [train.py:903] (0/4) Epoch 19, batch 5000, loss[loss=0.212, simple_loss=0.2916, pruned_loss=0.06619, over 19717.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2914, pruned_loss=0.06668, over 3808917.87 frames. ], batch size: 59, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 15:01:41,893 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=127904.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 15:01:51,192 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 15:01:55,669 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.309e+02 4.675e+02 5.614e+02 6.818e+02 2.294e+03, threshold=1.123e+03, percent-clipped=3.0 +2023-04-02 15:02:03,230 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 15:02:04,133 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.89 vs. limit=2.0 +2023-04-02 15:02:22,572 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-04-02 15:02:41,853 INFO [train.py:903] (0/4) Epoch 19, batch 5050, loss[loss=0.2418, simple_loss=0.3228, pruned_loss=0.08038, over 19594.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2918, pruned_loss=0.06717, over 3809159.71 frames. ], batch size: 61, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 15:03:16,213 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=127982.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:03:18,095 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 15:03:36,860 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-128000.pt +2023-04-02 15:03:43,138 INFO [train.py:903] (0/4) Epoch 19, batch 5100, loss[loss=0.1941, simple_loss=0.2848, pruned_loss=0.05172, over 19672.00 frames. ], tot_loss[loss=0.2122, simple_loss=0.291, pruned_loss=0.06671, over 3814895.50 frames. ], batch size: 59, lr: 4.32e-03, grad_scale: 8.0 +2023-04-02 15:03:45,772 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128006.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:03:47,019 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128007.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:03:47,355 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-02 15:03:56,492 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 15:03:58,283 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.139e+02 4.818e+02 5.706e+02 8.227e+02 1.561e+03, threshold=1.141e+03, percent-clipped=7.0 +2023-04-02 15:04:00,657 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 15:04:04,158 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 15:04:07,804 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=128024.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:04:43,093 INFO [train.py:903] (0/4) Epoch 19, batch 5150, loss[loss=0.2407, simple_loss=0.3116, pruned_loss=0.08487, over 19382.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2903, pruned_loss=0.06649, over 3810676.78 frames. ], batch size: 70, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:04:51,937 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0127, 1.9425, 1.8141, 1.5896, 1.5562, 1.6363, 0.4082, 0.9746], + device='cuda:0'), covar=tensor([0.0519, 0.0561, 0.0368, 0.0631, 0.1103, 0.0775, 0.1219, 0.0940], + device='cuda:0'), in_proj_covar=tensor([0.0352, 0.0346, 0.0348, 0.0372, 0.0447, 0.0380, 0.0328, 0.0336], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 15:04:57,229 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 15:05:31,571 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 15:05:44,863 INFO [train.py:903] (0/4) Epoch 19, batch 5200, loss[loss=0.2152, simple_loss=0.2985, pruned_loss=0.06601, over 19606.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2906, pruned_loss=0.06632, over 3811819.82 frames. ], batch size: 57, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:05:59,012 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.969e+02 4.920e+02 6.208e+02 7.921e+02 1.726e+03, threshold=1.242e+03, percent-clipped=7.0 +2023-04-02 15:05:59,070 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 15:06:16,658 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-02 15:06:28,650 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128139.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:06:29,667 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1191, 1.2492, 1.4891, 1.3470, 2.7034, 1.0064, 2.0893, 3.0560], + device='cuda:0'), covar=tensor([0.0590, 0.2819, 0.2715, 0.1851, 0.0795, 0.2487, 0.1252, 0.0329], + device='cuda:0'), in_proj_covar=tensor([0.0397, 0.0358, 0.0378, 0.0344, 0.0369, 0.0349, 0.0371, 0.0390], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:06:30,653 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=128141.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:06:41,119 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 15:06:45,758 INFO [train.py:903] (0/4) Epoch 19, batch 5250, loss[loss=0.1777, simple_loss=0.2598, pruned_loss=0.04777, over 19405.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2909, pruned_loss=0.06643, over 3819611.23 frames. ], batch size: 48, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:06:53,696 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128160.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 15:07:23,148 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128185.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 15:07:45,451 INFO [train.py:903] (0/4) Epoch 19, batch 5300, loss[loss=0.2075, simple_loss=0.2931, pruned_loss=0.06098, over 19523.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2917, pruned_loss=0.06722, over 3827041.56 frames. ], batch size: 54, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:07:54,714 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=128212.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:07:59,668 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.291e+02 5.001e+02 6.088e+02 7.600e+02 1.403e+03, threshold=1.218e+03, percent-clipped=2.0 +2023-04-02 15:08:00,872 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 15:08:21,801 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=128234.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:08:34,560 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4523, 1.4240, 1.7590, 1.6592, 2.5974, 2.3339, 2.7484, 1.0469], + device='cuda:0'), covar=tensor([0.2395, 0.4275, 0.2570, 0.1885, 0.1475, 0.1935, 0.1393, 0.4425], + device='cuda:0'), in_proj_covar=tensor([0.0522, 0.0629, 0.0692, 0.0475, 0.0615, 0.0521, 0.0660, 0.0536], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 15:08:46,735 INFO [train.py:903] (0/4) Epoch 19, batch 5350, loss[loss=0.2454, simple_loss=0.3188, pruned_loss=0.08603, over 18096.00 frames. ], tot_loss[loss=0.2123, simple_loss=0.2912, pruned_loss=0.06675, over 3818621.74 frames. ], batch size: 83, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:08:50,303 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0998, 1.4178, 1.8565, 1.5191, 2.8301, 4.5148, 4.4468, 5.0090], + device='cuda:0'), covar=tensor([0.1810, 0.3768, 0.3436, 0.2311, 0.0690, 0.0221, 0.0185, 0.0193], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0315, 0.0346, 0.0260, 0.0238, 0.0181, 0.0213, 0.0246], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 15:08:50,315 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128256.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:08:57,115 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128262.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:09:20,022 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 15:09:26,895 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128287.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:09:47,392 INFO [train.py:903] (0/4) Epoch 19, batch 5400, loss[loss=0.1931, simple_loss=0.2695, pruned_loss=0.05837, over 19587.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.2924, pruned_loss=0.06767, over 3792834.14 frames. ], batch size: 52, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:10:00,905 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0312, 3.6564, 2.5056, 3.2857, 0.8969, 3.5607, 3.4774, 3.6197], + device='cuda:0'), covar=tensor([0.0771, 0.1074, 0.1969, 0.0912, 0.3875, 0.0782, 0.0861, 0.1098], + device='cuda:0'), in_proj_covar=tensor([0.0489, 0.0394, 0.0482, 0.0340, 0.0396, 0.0420, 0.0410, 0.0446], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:10:01,756 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.288e+02 4.631e+02 5.571e+02 7.152e+02 1.493e+03, threshold=1.114e+03, percent-clipped=2.0 +2023-04-02 15:10:15,370 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128327.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:10:29,858 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128338.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:10:42,687 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128349.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:10:48,543 INFO [train.py:903] (0/4) Epoch 19, batch 5450, loss[loss=0.2714, simple_loss=0.3291, pruned_loss=0.1069, over 13022.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2923, pruned_loss=0.06762, over 3794302.70 frames. ], batch size: 135, lr: 4.31e-03, grad_scale: 16.0 +2023-04-02 15:11:39,578 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128395.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:11:50,174 INFO [train.py:903] (0/4) Epoch 19, batch 5500, loss[loss=0.2195, simple_loss=0.2955, pruned_loss=0.07179, over 19707.00 frames. ], tot_loss[loss=0.2141, simple_loss=0.2928, pruned_loss=0.06764, over 3797858.95 frames. ], batch size: 59, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:12:06,853 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.088e+02 5.166e+02 6.121e+02 7.872e+02 1.632e+03, threshold=1.224e+03, percent-clipped=5.0 +2023-04-02 15:12:10,622 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128420.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:12:12,705 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 15:12:16,375 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0715, 1.4251, 1.8277, 1.6037, 2.9522, 4.4761, 4.3726, 4.9489], + device='cuda:0'), covar=tensor([0.1735, 0.3793, 0.3392, 0.2220, 0.0660, 0.0187, 0.0178, 0.0177], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0315, 0.0346, 0.0260, 0.0238, 0.0181, 0.0214, 0.0246], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 15:12:23,819 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128431.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:12:50,265 INFO [train.py:903] (0/4) Epoch 19, batch 5550, loss[loss=0.2143, simple_loss=0.2927, pruned_loss=0.06797, over 19395.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2924, pruned_loss=0.06747, over 3797381.45 frames. ], batch size: 47, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:12:56,475 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 15:13:44,849 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 15:13:51,308 INFO [train.py:903] (0/4) Epoch 19, batch 5600, loss[loss=0.2794, simple_loss=0.3391, pruned_loss=0.1098, over 13384.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2921, pruned_loss=0.06732, over 3807376.74 frames. ], batch size: 136, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:14:01,884 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128512.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:14:07,037 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.064e+02 4.843e+02 5.877e+02 7.578e+02 1.194e+03, threshold=1.175e+03, percent-clipped=0.0 +2023-04-02 15:14:32,655 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128537.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:14:52,040 INFO [train.py:903] (0/4) Epoch 19, batch 5650, loss[loss=0.2016, simple_loss=0.2683, pruned_loss=0.06752, over 18574.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2906, pruned_loss=0.06624, over 3818749.17 frames. ], batch size: 41, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:15:27,808 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128583.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:15:37,266 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 15:15:53,141 INFO [train.py:903] (0/4) Epoch 19, batch 5700, loss[loss=0.1971, simple_loss=0.2683, pruned_loss=0.06298, over 17281.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2903, pruned_loss=0.06606, over 3835882.44 frames. ], batch size: 38, lr: 4.31e-03, grad_scale: 8.0 +2023-04-02 15:15:54,805 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=128605.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:15:58,262 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128608.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:16:08,028 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.931e+02 5.156e+02 6.108e+02 7.232e+02 1.309e+03, threshold=1.222e+03, percent-clipped=4.0 +2023-04-02 15:16:24,868 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=128630.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:16:53,492 INFO [train.py:903] (0/4) Epoch 19, batch 5750, loss[loss=0.2174, simple_loss=0.3029, pruned_loss=0.06601, over 19700.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2896, pruned_loss=0.0655, over 3844228.43 frames. ], batch size: 63, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:16:55,750 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 15:17:05,272 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 15:17:09,593 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 15:17:26,653 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.47 vs. limit=2.0 +2023-04-02 15:17:27,291 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=128682.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:17:42,109 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2267, 2.3262, 2.4870, 2.9929, 2.1890, 2.9214, 2.5477, 2.2871], + device='cuda:0'), covar=tensor([0.4176, 0.3874, 0.1744, 0.2573, 0.4431, 0.2124, 0.4605, 0.3279], + device='cuda:0'), in_proj_covar=tensor([0.0874, 0.0935, 0.0700, 0.0923, 0.0859, 0.0794, 0.0833, 0.0767], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 15:17:55,080 INFO [train.py:903] (0/4) Epoch 19, batch 5800, loss[loss=0.2384, simple_loss=0.3151, pruned_loss=0.08084, over 19689.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.29, pruned_loss=0.06623, over 3843522.72 frames. ], batch size: 60, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:18:10,461 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.713e+02 4.671e+02 6.414e+02 7.787e+02 1.302e+03, threshold=1.283e+03, percent-clipped=2.0 +2023-04-02 15:18:24,061 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5173, 2.1643, 1.7246, 1.2397, 2.1202, 1.2359, 1.3207, 1.9330], + device='cuda:0'), covar=tensor([0.0821, 0.0601, 0.0763, 0.0788, 0.0441, 0.1012, 0.0634, 0.0420], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0309, 0.0327, 0.0259, 0.0242, 0.0332, 0.0289, 0.0271], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:18:55,625 INFO [train.py:903] (0/4) Epoch 19, batch 5850, loss[loss=0.2086, simple_loss=0.2915, pruned_loss=0.06284, over 19545.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2897, pruned_loss=0.06598, over 3843251.17 frames. ], batch size: 54, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:19:20,629 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=128775.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:19:48,422 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128797.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:19:50,609 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128799.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:19:55,907 INFO [train.py:903] (0/4) Epoch 19, batch 5900, loss[loss=0.2107, simple_loss=0.2826, pruned_loss=0.0694, over 19619.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2901, pruned_loss=0.06647, over 3842118.53 frames. ], batch size: 50, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:20:02,588 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 15:20:11,629 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.258e+02 4.778e+02 5.849e+02 7.721e+02 1.320e+03, threshold=1.170e+03, percent-clipped=1.0 +2023-04-02 15:20:23,109 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 15:20:56,176 INFO [train.py:903] (0/4) Epoch 19, batch 5950, loss[loss=0.2036, simple_loss=0.288, pruned_loss=0.05961, over 19547.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2897, pruned_loss=0.06643, over 3833046.59 frames. ], batch size: 56, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:20:57,899 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 15:21:41,415 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=128890.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:21:57,512 INFO [train.py:903] (0/4) Epoch 19, batch 6000, loss[loss=0.2067, simple_loss=0.2964, pruned_loss=0.05851, over 19709.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2897, pruned_loss=0.06625, over 3840940.99 frames. ], batch size: 59, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:21:57,513 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 15:22:07,208 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8938, 1.5830, 1.4887, 1.7912, 1.4793, 1.6131, 1.4049, 1.7161], + device='cuda:0'), covar=tensor([0.1048, 0.1229, 0.1537, 0.1096, 0.1445, 0.0562, 0.1668, 0.0726], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0355, 0.0308, 0.0249, 0.0298, 0.0248, 0.0300, 0.0251], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:22:12,603 INFO [train.py:937] (0/4) Epoch 19, validation: loss=0.1702, simple_loss=0.2702, pruned_loss=0.03514, over 944034.00 frames. +2023-04-02 15:22:12,604 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 15:22:17,263 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=128908.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 15:22:27,998 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.370e+02 5.136e+02 6.485e+02 9.043e+02 2.174e+03, threshold=1.297e+03, percent-clipped=7.0 +2023-04-02 15:23:13,564 INFO [train.py:903] (0/4) Epoch 19, batch 6050, loss[loss=0.2373, simple_loss=0.3204, pruned_loss=0.07705, over 19743.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.291, pruned_loss=0.06699, over 3841467.32 frames. ], batch size: 63, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:23:33,155 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7808, 1.9083, 2.1372, 2.4034, 1.7708, 2.2669, 2.1353, 1.9358], + device='cuda:0'), covar=tensor([0.3971, 0.3561, 0.1763, 0.2111, 0.3693, 0.2000, 0.4647, 0.3242], + device='cuda:0'), in_proj_covar=tensor([0.0875, 0.0934, 0.0699, 0.0923, 0.0859, 0.0793, 0.0831, 0.0766], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 15:24:14,689 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3591, 1.4782, 1.8002, 1.6013, 2.9965, 2.4999, 3.3082, 1.6222], + device='cuda:0'), covar=tensor([0.2520, 0.4426, 0.2800, 0.2062, 0.1662, 0.2111, 0.1546, 0.4134], + device='cuda:0'), in_proj_covar=tensor([0.0523, 0.0632, 0.0695, 0.0476, 0.0620, 0.0524, 0.0664, 0.0540], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 15:24:15,354 INFO [train.py:903] (0/4) Epoch 19, batch 6100, loss[loss=0.2312, simple_loss=0.3158, pruned_loss=0.07337, over 19464.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2912, pruned_loss=0.06684, over 3842626.47 frames. ], batch size: 64, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:24:30,770 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.431e+02 4.994e+02 6.076e+02 7.380e+02 1.472e+03, threshold=1.215e+03, percent-clipped=4.0 +2023-04-02 15:25:14,909 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.8078, 5.2753, 2.9711, 4.6247, 1.0206, 5.3064, 5.2053, 5.3652], + device='cuda:0'), covar=tensor([0.0405, 0.0823, 0.1966, 0.0719, 0.4161, 0.0519, 0.0761, 0.1070], + device='cuda:0'), in_proj_covar=tensor([0.0493, 0.0397, 0.0486, 0.0343, 0.0399, 0.0422, 0.0413, 0.0450], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:25:15,123 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=129053.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:25:15,830 INFO [train.py:903] (0/4) Epoch 19, batch 6150, loss[loss=0.2018, simple_loss=0.2944, pruned_loss=0.05457, over 18802.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.2916, pruned_loss=0.06714, over 3843264.72 frames. ], batch size: 74, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:25:37,314 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-02 15:25:44,284 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 15:25:44,629 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=129078.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:26:16,321 INFO [train.py:903] (0/4) Epoch 19, batch 6200, loss[loss=0.2093, simple_loss=0.284, pruned_loss=0.06726, over 19865.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2904, pruned_loss=0.06619, over 3849561.42 frames. ], batch size: 52, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:26:32,068 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.839e+02 4.815e+02 6.250e+02 7.621e+02 1.523e+03, threshold=1.250e+03, percent-clipped=7.0 +2023-04-02 15:26:53,197 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8214, 1.9364, 2.1999, 2.4333, 1.7518, 2.3372, 2.1984, 1.9935], + device='cuda:0'), covar=tensor([0.4174, 0.3567, 0.1750, 0.2145, 0.3753, 0.1863, 0.4531, 0.3187], + device='cuda:0'), in_proj_covar=tensor([0.0874, 0.0934, 0.0701, 0.0922, 0.0858, 0.0793, 0.0829, 0.0766], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 15:27:04,010 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=129143.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:27:07,652 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=129146.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:27:17,155 INFO [train.py:903] (0/4) Epoch 19, batch 6250, loss[loss=0.2429, simple_loss=0.328, pruned_loss=0.07888, over 17161.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2905, pruned_loss=0.06611, over 3834550.66 frames. ], batch size: 101, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:27:38,409 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=129171.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:27:47,710 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 15:28:17,840 INFO [train.py:903] (0/4) Epoch 19, batch 6300, loss[loss=0.2365, simple_loss=0.3137, pruned_loss=0.07964, over 19498.00 frames. ], tot_loss[loss=0.2123, simple_loss=0.2913, pruned_loss=0.0667, over 3826664.69 frames. ], batch size: 64, lr: 4.30e-03, grad_scale: 8.0 +2023-04-02 15:28:33,725 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.168e+02 5.181e+02 6.373e+02 8.503e+02 1.874e+03, threshold=1.275e+03, percent-clipped=7.0 +2023-04-02 15:29:17,490 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=129252.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 15:29:19,480 INFO [train.py:903] (0/4) Epoch 19, batch 6350, loss[loss=0.211, simple_loss=0.2937, pruned_loss=0.06414, over 19670.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2909, pruned_loss=0.06637, over 3823119.51 frames. ], batch size: 55, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:29:25,240 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=129258.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:30:20,709 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 15:30:21,272 INFO [train.py:903] (0/4) Epoch 19, batch 6400, loss[loss=0.1707, simple_loss=0.2513, pruned_loss=0.04504, over 19314.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2913, pruned_loss=0.06676, over 3823746.52 frames. ], batch size: 44, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:30:36,857 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.331e+02 4.991e+02 6.008e+02 7.727e+02 1.608e+03, threshold=1.202e+03, percent-clipped=4.0 +2023-04-02 15:30:58,495 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3303, 2.0279, 1.5430, 1.4368, 1.8255, 1.1986, 1.2940, 1.7817], + device='cuda:0'), covar=tensor([0.0894, 0.0807, 0.1064, 0.0742, 0.0510, 0.1229, 0.0649, 0.0437], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0310, 0.0328, 0.0259, 0.0241, 0.0331, 0.0288, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:31:22,261 INFO [train.py:903] (0/4) Epoch 19, batch 6450, loss[loss=0.1913, simple_loss=0.2781, pruned_loss=0.05222, over 19536.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2907, pruned_loss=0.06603, over 3823423.45 frames. ], batch size: 64, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:31:38,193 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=129367.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 15:31:48,959 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8882, 1.3403, 1.0568, 1.0344, 1.1493, 1.0251, 1.0443, 1.2196], + device='cuda:0'), covar=tensor([0.0573, 0.0824, 0.1104, 0.0690, 0.0541, 0.1234, 0.0534, 0.0479], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0310, 0.0329, 0.0260, 0.0242, 0.0332, 0.0289, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:32:06,808 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 15:32:07,095 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0612, 1.7077, 1.9987, 1.9709, 4.5184, 1.3641, 2.7285, 4.9915], + device='cuda:0'), covar=tensor([0.0387, 0.2670, 0.2564, 0.1761, 0.0733, 0.2504, 0.1258, 0.0171], + device='cuda:0'), in_proj_covar=tensor([0.0396, 0.0358, 0.0377, 0.0340, 0.0367, 0.0348, 0.0370, 0.0391], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:32:22,359 INFO [train.py:903] (0/4) Epoch 19, batch 6500, loss[loss=0.1993, simple_loss=0.2774, pruned_loss=0.06057, over 19380.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2908, pruned_loss=0.06593, over 3815926.28 frames. ], batch size: 48, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:32:29,740 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 15:32:38,768 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.146e+02 5.340e+02 6.897e+02 8.888e+02 1.987e+03, threshold=1.379e+03, percent-clipped=7.0 +2023-04-02 15:32:47,856 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6021, 1.1986, 1.3427, 1.2036, 2.2405, 1.0439, 2.0705, 2.5502], + device='cuda:0'), covar=tensor([0.0678, 0.2683, 0.2874, 0.1685, 0.0907, 0.2106, 0.0955, 0.0420], + device='cuda:0'), in_proj_covar=tensor([0.0396, 0.0357, 0.0377, 0.0340, 0.0367, 0.0348, 0.0370, 0.0391], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:32:56,832 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=129431.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 15:33:24,206 INFO [train.py:903] (0/4) Epoch 19, batch 6550, loss[loss=0.2085, simple_loss=0.2926, pruned_loss=0.06222, over 19590.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2909, pruned_loss=0.06565, over 3833083.74 frames. ], batch size: 61, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:34:24,937 INFO [train.py:903] (0/4) Epoch 19, batch 6600, loss[loss=0.2048, simple_loss=0.294, pruned_loss=0.05784, over 19694.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2916, pruned_loss=0.06616, over 3835388.30 frames. ], batch size: 59, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:34:37,588 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=129514.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:34:40,453 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.539e+02 5.847e+02 6.807e+02 8.552e+02 1.538e+03, threshold=1.361e+03, percent-clipped=4.0 +2023-04-02 15:35:03,129 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.90 vs. limit=5.0 +2023-04-02 15:35:03,893 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5923, 1.4744, 1.4664, 1.9924, 1.5380, 2.0033, 1.9121, 1.6477], + device='cuda:0'), covar=tensor([0.0823, 0.0934, 0.1022, 0.0774, 0.0902, 0.0667, 0.0831, 0.0731], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0219, 0.0225, 0.0245, 0.0227, 0.0209, 0.0189, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 15:35:07,235 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=129539.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:35:25,979 INFO [train.py:903] (0/4) Epoch 19, batch 6650, loss[loss=0.1809, simple_loss=0.2578, pruned_loss=0.05202, over 19751.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2905, pruned_loss=0.06613, over 3831987.15 frames. ], batch size: 46, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:35:32,231 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4721, 2.4456, 2.6977, 3.3123, 2.4653, 3.0758, 2.8973, 2.5558], + device='cuda:0'), covar=tensor([0.3852, 0.3561, 0.1632, 0.2280, 0.3959, 0.1943, 0.4175, 0.2858], + device='cuda:0'), in_proj_covar=tensor([0.0868, 0.0927, 0.0697, 0.0917, 0.0854, 0.0787, 0.0826, 0.0761], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 15:36:26,350 INFO [train.py:903] (0/4) Epoch 19, batch 6700, loss[loss=0.218, simple_loss=0.3042, pruned_loss=0.06589, over 19745.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2905, pruned_loss=0.06652, over 3825347.92 frames. ], batch size: 63, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:36:42,862 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.775e+02 4.923e+02 5.649e+02 7.598e+02 1.428e+03, threshold=1.130e+03, percent-clipped=1.0 +2023-04-02 15:36:51,037 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=129623.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 15:36:58,742 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=129630.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:37:03,803 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-02 15:37:04,553 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6487, 1.7859, 1.7070, 2.4899, 1.8151, 2.3854, 1.8084, 1.3967], + device='cuda:0'), covar=tensor([0.5307, 0.4421, 0.3068, 0.2807, 0.4295, 0.2244, 0.6722, 0.5690], + device='cuda:0'), in_proj_covar=tensor([0.0869, 0.0929, 0.0697, 0.0918, 0.0855, 0.0788, 0.0826, 0.0762], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 15:37:18,815 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=129648.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 15:37:25,090 INFO [train.py:903] (0/4) Epoch 19, batch 6750, loss[loss=0.1772, simple_loss=0.2615, pruned_loss=0.04645, over 19487.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.2894, pruned_loss=0.06579, over 3843708.97 frames. ], batch size: 49, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:38:20,255 INFO [train.py:903] (0/4) Epoch 19, batch 6800, loss[loss=0.1933, simple_loss=0.259, pruned_loss=0.06381, over 19787.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2884, pruned_loss=0.06581, over 3841359.26 frames. ], batch size: 47, lr: 4.29e-03, grad_scale: 8.0 +2023-04-02 15:38:34,410 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.328e+02 4.884e+02 6.226e+02 8.201e+02 1.689e+03, threshold=1.245e+03, percent-clipped=11.0 +2023-04-02 15:38:49,821 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-19.pt +2023-04-02 15:39:05,072 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 15:39:06,153 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 15:39:08,368 INFO [train.py:903] (0/4) Epoch 20, batch 0, loss[loss=0.2249, simple_loss=0.2982, pruned_loss=0.07581, over 19728.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.2982, pruned_loss=0.07581, over 19728.00 frames. ], batch size: 51, lr: 4.18e-03, grad_scale: 8.0 +2023-04-02 15:39:08,369 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 15:39:19,745 INFO [train.py:937] (0/4) Epoch 20, validation: loss=0.1695, simple_loss=0.2703, pruned_loss=0.03432, over 944034.00 frames. +2023-04-02 15:39:19,745 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 15:39:31,873 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 15:40:12,660 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=129775.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 15:40:20,206 INFO [train.py:903] (0/4) Epoch 20, batch 50, loss[loss=0.2011, simple_loss=0.2889, pruned_loss=0.05666, over 19693.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2901, pruned_loss=0.06588, over 875020.00 frames. ], batch size: 53, lr: 4.18e-03, grad_scale: 8.0 +2023-04-02 15:40:51,307 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=129809.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:40:54,560 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 15:41:03,011 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.371e+02 5.543e+02 6.891e+02 8.835e+02 1.770e+03, threshold=1.378e+03, percent-clipped=8.0 +2023-04-02 15:41:05,559 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8173, 1.5550, 1.5594, 1.4688, 3.3828, 0.9734, 2.4458, 3.8771], + device='cuda:0'), covar=tensor([0.0434, 0.2430, 0.2625, 0.1829, 0.0659, 0.2641, 0.1207, 0.0203], + device='cuda:0'), in_proj_covar=tensor([0.0394, 0.0355, 0.0375, 0.0337, 0.0365, 0.0345, 0.0369, 0.0388], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:41:20,212 INFO [train.py:903] (0/4) Epoch 20, batch 100, loss[loss=0.2305, simple_loss=0.3044, pruned_loss=0.07823, over 19520.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2903, pruned_loss=0.06625, over 1546789.83 frames. ], batch size: 64, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:41:31,345 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 15:42:14,137 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2766, 1.2958, 1.7986, 1.2881, 2.7414, 3.6389, 3.3761, 3.8834], + device='cuda:0'), covar=tensor([0.1635, 0.3898, 0.3248, 0.2355, 0.0588, 0.0180, 0.0226, 0.0264], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0316, 0.0346, 0.0261, 0.0237, 0.0181, 0.0213, 0.0247], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 15:42:21,224 INFO [train.py:903] (0/4) Epoch 20, batch 150, loss[loss=0.218, simple_loss=0.3043, pruned_loss=0.06582, over 19528.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2918, pruned_loss=0.06685, over 2049930.86 frames. ], batch size: 54, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:42:30,167 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=129890.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 15:43:03,459 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.452e+02 4.799e+02 5.935e+02 7.467e+02 3.197e+03, threshold=1.187e+03, percent-clipped=3.0 +2023-04-02 15:43:21,677 INFO [train.py:903] (0/4) Epoch 20, batch 200, loss[loss=0.1948, simple_loss=0.2754, pruned_loss=0.05707, over 19688.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2931, pruned_loss=0.06727, over 2444943.77 frames. ], batch size: 53, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:43:22,848 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 15:43:36,798 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-04-02 15:44:13,724 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=129974.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:44:23,539 INFO [train.py:903] (0/4) Epoch 20, batch 250, loss[loss=0.1973, simple_loss=0.2825, pruned_loss=0.056, over 19672.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2917, pruned_loss=0.06657, over 2747680.59 frames. ], batch size: 58, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:44:45,556 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-130000.pt +2023-04-02 15:45:06,935 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.070e+02 5.298e+02 6.354e+02 8.098e+02 1.543e+03, threshold=1.271e+03, percent-clipped=7.0 +2023-04-02 15:45:08,335 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130018.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:45:10,543 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130020.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:45:25,487 INFO [train.py:903] (0/4) Epoch 20, batch 300, loss[loss=0.2237, simple_loss=0.3, pruned_loss=0.07368, over 19670.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2914, pruned_loss=0.06699, over 2996452.71 frames. ], batch size: 60, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:45:57,485 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5278, 4.0954, 4.2559, 4.2650, 1.7817, 4.0263, 3.4846, 3.9692], + device='cuda:0'), covar=tensor([0.1641, 0.0913, 0.0638, 0.0669, 0.5582, 0.0886, 0.0674, 0.1137], + device='cuda:0'), in_proj_covar=tensor([0.0766, 0.0719, 0.0923, 0.0805, 0.0817, 0.0679, 0.0555, 0.0856], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 15:46:25,935 INFO [train.py:903] (0/4) Epoch 20, batch 350, loss[loss=0.2261, simple_loss=0.3216, pruned_loss=0.06529, over 19675.00 frames. ], tot_loss[loss=0.2132, simple_loss=0.2923, pruned_loss=0.06709, over 3186080.19 frames. ], batch size: 58, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:46:35,001 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 15:46:35,300 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130089.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:46:38,868 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3769, 1.2937, 1.5018, 1.4752, 2.9263, 1.2079, 2.3434, 3.3980], + device='cuda:0'), covar=tensor([0.0554, 0.2796, 0.2914, 0.1921, 0.0799, 0.2464, 0.1253, 0.0284], + device='cuda:0'), in_proj_covar=tensor([0.0399, 0.0359, 0.0378, 0.0341, 0.0369, 0.0349, 0.0373, 0.0393], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:46:50,450 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3869, 2.0831, 2.1073, 2.8628, 1.7827, 2.5288, 2.3900, 2.3252], + device='cuda:0'), covar=tensor([0.0716, 0.0831, 0.0922, 0.0769, 0.0971, 0.0708, 0.0925, 0.0612], + device='cuda:0'), in_proj_covar=tensor([0.0207, 0.0219, 0.0224, 0.0242, 0.0225, 0.0209, 0.0187, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 15:47:08,653 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.196e+02 5.230e+02 6.397e+02 7.792e+02 1.393e+03, threshold=1.279e+03, percent-clipped=3.0 +2023-04-02 15:47:26,635 INFO [train.py:903] (0/4) Epoch 20, batch 400, loss[loss=0.2079, simple_loss=0.2921, pruned_loss=0.06185, over 17346.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2909, pruned_loss=0.0665, over 3328327.02 frames. ], batch size: 101, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:47:41,012 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.41 vs. limit=5.0 +2023-04-02 15:47:42,916 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130146.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 15:47:53,120 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=130153.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:48:14,976 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130171.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 15:48:27,090 INFO [train.py:903] (0/4) Epoch 20, batch 450, loss[loss=0.2239, simple_loss=0.3076, pruned_loss=0.07008, over 19722.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2918, pruned_loss=0.06681, over 3432284.71 frames. ], batch size: 59, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:48:41,193 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2057, 2.1161, 1.8683, 1.6906, 1.7325, 1.7469, 0.6991, 1.1120], + device='cuda:0'), covar=tensor([0.0538, 0.0569, 0.0460, 0.0685, 0.1051, 0.0773, 0.1102, 0.0921], + device='cuda:0'), in_proj_covar=tensor([0.0358, 0.0350, 0.0353, 0.0376, 0.0451, 0.0383, 0.0333, 0.0338], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 15:49:03,630 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 15:49:04,562 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 15:49:09,165 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.114e+02 5.003e+02 6.443e+02 8.043e+02 1.786e+03, threshold=1.289e+03, percent-clipped=5.0 +2023-04-02 15:49:18,399 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5841, 1.6190, 1.7904, 1.6945, 2.4029, 2.1378, 2.4118, 1.4549], + device='cuda:0'), covar=tensor([0.1773, 0.3123, 0.2055, 0.1508, 0.1217, 0.1681, 0.1135, 0.3442], + device='cuda:0'), in_proj_covar=tensor([0.0526, 0.0632, 0.0697, 0.0475, 0.0618, 0.0526, 0.0661, 0.0541], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 15:49:27,173 INFO [train.py:903] (0/4) Epoch 20, batch 500, loss[loss=0.2124, simple_loss=0.2896, pruned_loss=0.0676, over 19356.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2913, pruned_loss=0.06675, over 3516366.45 frames. ], batch size: 48, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:49:50,923 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.66 vs. limit=2.0 +2023-04-02 15:50:10,052 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130268.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:50:27,989 INFO [train.py:903] (0/4) Epoch 20, batch 550, loss[loss=0.18, simple_loss=0.2527, pruned_loss=0.05363, over 19759.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2901, pruned_loss=0.06598, over 3586998.49 frames. ], batch size: 45, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:51:11,221 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.373e+02 5.062e+02 6.327e+02 8.479e+02 2.088e+03, threshold=1.265e+03, percent-clipped=6.0 +2023-04-02 15:51:28,462 INFO [train.py:903] (0/4) Epoch 20, batch 600, loss[loss=0.206, simple_loss=0.2955, pruned_loss=0.05825, over 19703.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2905, pruned_loss=0.06602, over 3648113.65 frames. ], batch size: 59, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:51:44,867 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130345.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:52:06,337 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=130362.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:52:08,686 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=130364.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:52:15,668 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 15:52:16,043 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130370.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:52:30,047 INFO [train.py:903] (0/4) Epoch 20, batch 650, loss[loss=0.2262, simple_loss=0.3027, pruned_loss=0.07482, over 19657.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2905, pruned_loss=0.06619, over 3701058.03 frames. ], batch size: 53, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:52:40,964 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9283, 1.3319, 1.0492, 0.9552, 1.1342, 0.9711, 0.8930, 1.2294], + device='cuda:0'), covar=tensor([0.0578, 0.0724, 0.0990, 0.0652, 0.0518, 0.1198, 0.0575, 0.0455], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0313, 0.0330, 0.0261, 0.0244, 0.0336, 0.0290, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 15:52:56,026 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130402.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:53:13,825 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.680e+02 5.222e+02 6.307e+02 8.250e+02 2.391e+03, threshold=1.261e+03, percent-clipped=5.0 +2023-04-02 15:53:31,357 INFO [train.py:903] (0/4) Epoch 20, batch 700, loss[loss=0.2336, simple_loss=0.3101, pruned_loss=0.07852, over 19679.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2903, pruned_loss=0.0658, over 3733445.31 frames. ], batch size: 53, lr: 4.17e-03, grad_scale: 8.0 +2023-04-02 15:53:55,089 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130450.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:54:02,068 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130456.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:54:28,618 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130477.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:54:30,948 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130479.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:54:35,655 INFO [train.py:903] (0/4) Epoch 20, batch 750, loss[loss=0.2363, simple_loss=0.313, pruned_loss=0.0798, over 13054.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2908, pruned_loss=0.0661, over 3745699.00 frames. ], batch size: 135, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 15:54:58,168 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-02 15:55:19,214 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.433e+02 4.790e+02 6.042e+02 7.311e+02 1.890e+03, threshold=1.208e+03, percent-clipped=4.0 +2023-04-02 15:55:28,430 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130524.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:55:37,192 INFO [train.py:903] (0/4) Epoch 20, batch 800, loss[loss=0.1993, simple_loss=0.2867, pruned_loss=0.05592, over 19676.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2909, pruned_loss=0.06598, over 3755591.05 frames. ], batch size: 60, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 15:55:53,517 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 15:55:58,722 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130549.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:55:59,953 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3900, 1.3212, 1.7506, 1.4113, 2.7551, 3.7343, 3.4526, 3.9374], + device='cuda:0'), covar=tensor([0.1501, 0.3625, 0.3227, 0.2284, 0.0540, 0.0186, 0.0206, 0.0237], + device='cuda:0'), in_proj_covar=tensor([0.0268, 0.0316, 0.0347, 0.0262, 0.0238, 0.0182, 0.0213, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 15:56:21,742 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130566.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:56:27,625 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3121, 1.3578, 1.4771, 1.4395, 1.7487, 1.7874, 1.7490, 0.6319], + device='cuda:0'), covar=tensor([0.2225, 0.4008, 0.2588, 0.1843, 0.1567, 0.2155, 0.1383, 0.4478], + device='cuda:0'), in_proj_covar=tensor([0.0526, 0.0633, 0.0696, 0.0476, 0.0616, 0.0526, 0.0659, 0.0540], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 15:56:40,733 INFO [train.py:903] (0/4) Epoch 20, batch 850, loss[loss=0.2227, simple_loss=0.2982, pruned_loss=0.07362, over 19769.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2895, pruned_loss=0.06528, over 3769479.87 frames. ], batch size: 54, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 15:56:41,903 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=130583.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:57:07,797 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7479, 1.5855, 1.6733, 2.2568, 1.7434, 2.0188, 1.9989, 1.8128], + device='cuda:0'), covar=tensor([0.0779, 0.0904, 0.0949, 0.0664, 0.0779, 0.0711, 0.0831, 0.0694], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0220, 0.0226, 0.0243, 0.0227, 0.0209, 0.0188, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 15:57:25,283 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.130e+02 4.920e+02 5.760e+02 7.852e+02 1.760e+03, threshold=1.152e+03, percent-clipped=6.0 +2023-04-02 15:57:33,264 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 15:57:40,813 INFO [train.py:903] (0/4) Epoch 20, batch 900, loss[loss=0.2017, simple_loss=0.2874, pruned_loss=0.05799, over 19541.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2904, pruned_loss=0.06617, over 3773867.98 frames. ], batch size: 56, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 15:58:44,421 INFO [train.py:903] (0/4) Epoch 20, batch 950, loss[loss=0.2403, simple_loss=0.321, pruned_loss=0.07978, over 19514.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2906, pruned_loss=0.06629, over 3773990.51 frames. ], batch size: 64, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 15:58:47,635 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 15:59:28,713 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.021e+02 4.928e+02 5.917e+02 7.294e+02 1.421e+03, threshold=1.183e+03, percent-clipped=3.0 +2023-04-02 15:59:46,666 INFO [train.py:903] (0/4) Epoch 20, batch 1000, loss[loss=0.2412, simple_loss=0.3184, pruned_loss=0.08199, over 19751.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2912, pruned_loss=0.06653, over 3791385.69 frames. ], batch size: 54, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 15:59:48,240 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130733.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 15:59:50,473 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=130735.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:00:03,661 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=130746.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:00:16,552 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-02 16:00:18,718 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130758.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:00:20,899 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=130760.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:00:38,540 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 16:00:48,485 INFO [train.py:903] (0/4) Epoch 20, batch 1050, loss[loss=0.1965, simple_loss=0.2682, pruned_loss=0.06234, over 19621.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2922, pruned_loss=0.06714, over 3800851.98 frames. ], batch size: 50, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 16:01:02,832 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=130794.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:01:10,718 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=130800.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:01:20,573 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 16:01:33,077 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.731e+02 5.562e+02 6.742e+02 8.268e+02 2.102e+03, threshold=1.348e+03, percent-clipped=2.0 +2023-04-02 16:01:49,838 INFO [train.py:903] (0/4) Epoch 20, batch 1100, loss[loss=0.2075, simple_loss=0.2737, pruned_loss=0.07071, over 19753.00 frames. ], tot_loss[loss=0.2122, simple_loss=0.2912, pruned_loss=0.06655, over 3813094.59 frames. ], batch size: 46, lr: 4.16e-03, grad_scale: 4.0 +2023-04-02 16:02:27,128 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130861.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:02:36,684 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.08 vs. limit=2.0 +2023-04-02 16:02:47,563 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.50 vs. limit=2.0 +2023-04-02 16:02:52,364 INFO [train.py:903] (0/4) Epoch 20, batch 1150, loss[loss=0.2044, simple_loss=0.2909, pruned_loss=0.05897, over 19531.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2915, pruned_loss=0.06693, over 3794992.82 frames. ], batch size: 64, lr: 4.16e-03, grad_scale: 4.0 +2023-04-02 16:03:26,689 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130909.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:03:27,633 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=130910.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:03:33,808 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=130915.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:03:39,055 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.314e+02 5.063e+02 6.056e+02 7.993e+02 1.743e+03, threshold=1.211e+03, percent-clipped=5.0 +2023-04-02 16:03:50,403 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=130927.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:03:55,681 INFO [train.py:903] (0/4) Epoch 20, batch 1200, loss[loss=0.2203, simple_loss=0.2984, pruned_loss=0.07112, over 19515.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.29, pruned_loss=0.06623, over 3812121.54 frames. ], batch size: 64, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 16:04:23,976 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 16:04:56,111 INFO [train.py:903] (0/4) Epoch 20, batch 1250, loss[loss=0.1766, simple_loss=0.2543, pruned_loss=0.04942, over 18591.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2904, pruned_loss=0.06644, over 3818964.47 frames. ], batch size: 41, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 16:05:42,761 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.304e+02 5.121e+02 6.297e+02 7.673e+02 2.016e+03, threshold=1.259e+03, percent-clipped=4.0 +2023-04-02 16:05:50,218 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131025.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:05:52,242 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0242, 1.1788, 1.6690, 1.1756, 2.6153, 3.6477, 3.3714, 3.8497], + device='cuda:0'), covar=tensor([0.1766, 0.3968, 0.3406, 0.2477, 0.0629, 0.0198, 0.0202, 0.0251], + device='cuda:0'), in_proj_covar=tensor([0.0268, 0.0317, 0.0348, 0.0262, 0.0238, 0.0183, 0.0213, 0.0249], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 16:05:58,785 INFO [train.py:903] (0/4) Epoch 20, batch 1300, loss[loss=0.1999, simple_loss=0.2889, pruned_loss=0.05541, over 19605.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2901, pruned_loss=0.06597, over 3832936.30 frames. ], batch size: 57, lr: 4.16e-03, grad_scale: 8.0 +2023-04-02 16:06:12,053 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131042.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:06:32,908 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8772, 4.3913, 4.6268, 4.6434, 1.7474, 4.3039, 3.7544, 4.2974], + device='cuda:0'), covar=tensor([0.1754, 0.0767, 0.0618, 0.0699, 0.5978, 0.0928, 0.0677, 0.1238], + device='cuda:0'), in_proj_covar=tensor([0.0758, 0.0716, 0.0917, 0.0799, 0.0815, 0.0675, 0.0551, 0.0856], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 16:06:59,462 INFO [train.py:903] (0/4) Epoch 20, batch 1350, loss[loss=0.2386, simple_loss=0.309, pruned_loss=0.0841, over 14097.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2908, pruned_loss=0.06653, over 3816672.34 frames. ], batch size: 136, lr: 4.15e-03, grad_scale: 8.0 +2023-04-02 16:07:19,072 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 16:07:43,051 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131117.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:07:44,787 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.358e+02 5.090e+02 6.517e+02 8.267e+02 2.193e+03, threshold=1.303e+03, percent-clipped=8.0 +2023-04-02 16:08:02,290 INFO [train.py:903] (0/4) Epoch 20, batch 1400, loss[loss=0.233, simple_loss=0.3048, pruned_loss=0.08063, over 19483.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2915, pruned_loss=0.06687, over 3812392.74 frames. ], batch size: 49, lr: 4.15e-03, grad_scale: 8.0 +2023-04-02 16:08:15,121 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4936, 1.4034, 1.4265, 1.9758, 1.5787, 1.8298, 1.8550, 1.5352], + device='cuda:0'), covar=tensor([0.0844, 0.0923, 0.0995, 0.0712, 0.0803, 0.0721, 0.0796, 0.0735], + device='cuda:0'), in_proj_covar=tensor([0.0209, 0.0219, 0.0225, 0.0243, 0.0226, 0.0209, 0.0188, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 16:08:15,148 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131142.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:08:42,679 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131165.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:08:50,665 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131171.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:09:03,825 INFO [train.py:903] (0/4) Epoch 20, batch 1450, loss[loss=0.18, simple_loss=0.2546, pruned_loss=0.05269, over 19764.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2912, pruned_loss=0.06708, over 3816243.37 frames. ], batch size: 47, lr: 4.15e-03, grad_scale: 8.0 +2023-04-02 16:09:06,058 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 16:09:14,520 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131190.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:09:16,743 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=131192.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:09:21,563 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131196.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:09:50,840 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.524e+02 5.028e+02 6.181e+02 7.641e+02 1.699e+03, threshold=1.236e+03, percent-clipped=6.0 +2023-04-02 16:10:06,715 INFO [train.py:903] (0/4) Epoch 20, batch 1500, loss[loss=0.2178, simple_loss=0.289, pruned_loss=0.07331, over 19563.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2916, pruned_loss=0.06702, over 3824634.45 frames. ], batch size: 61, lr: 4.15e-03, grad_scale: 8.0 +2023-04-02 16:10:21,846 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7925, 4.3621, 2.7714, 3.8311, 0.8331, 4.3077, 4.1754, 4.3076], + device='cuda:0'), covar=tensor([0.0570, 0.0833, 0.1793, 0.0765, 0.3935, 0.0616, 0.0810, 0.1013], + device='cuda:0'), in_proj_covar=tensor([0.0494, 0.0397, 0.0485, 0.0345, 0.0400, 0.0423, 0.0416, 0.0447], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 16:10:49,080 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.44 vs. limit=5.0 +2023-04-02 16:11:07,013 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131281.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:11:07,713 INFO [train.py:903] (0/4) Epoch 20, batch 1550, loss[loss=0.2227, simple_loss=0.282, pruned_loss=0.08166, over 19750.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2925, pruned_loss=0.06747, over 3814092.96 frames. ], batch size: 47, lr: 4.15e-03, grad_scale: 8.0 +2023-04-02 16:11:29,222 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131298.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:11:38,375 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131306.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:11:53,753 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.770e+02 5.087e+02 6.243e+02 7.473e+02 1.350e+03, threshold=1.249e+03, percent-clipped=1.0 +2023-04-02 16:11:58,674 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131323.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:12:07,309 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5841, 1.1737, 1.4149, 1.3987, 2.2455, 1.0892, 2.2311, 2.4702], + device='cuda:0'), covar=tensor([0.0663, 0.2697, 0.2743, 0.1492, 0.0831, 0.1945, 0.0894, 0.0465], + device='cuda:0'), in_proj_covar=tensor([0.0399, 0.0357, 0.0377, 0.0339, 0.0368, 0.0348, 0.0371, 0.0393], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 16:12:10,214 INFO [train.py:903] (0/4) Epoch 20, batch 1600, loss[loss=0.1792, simple_loss=0.27, pruned_loss=0.04421, over 19695.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2925, pruned_loss=0.06741, over 3812865.18 frames. ], batch size: 53, lr: 4.15e-03, grad_scale: 8.0 +2023-04-02 16:12:36,103 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 16:13:12,817 INFO [train.py:903] (0/4) Epoch 20, batch 1650, loss[loss=0.1894, simple_loss=0.2805, pruned_loss=0.04919, over 18203.00 frames. ], tot_loss[loss=0.2148, simple_loss=0.2936, pruned_loss=0.06804, over 3790095.45 frames. ], batch size: 83, lr: 4.15e-03, grad_scale: 8.0 +2023-04-02 16:13:59,208 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.697e+02 5.218e+02 6.304e+02 8.075e+02 1.501e+03, threshold=1.261e+03, percent-clipped=5.0 +2023-04-02 16:14:08,577 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=131427.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:14:15,058 INFO [train.py:903] (0/4) Epoch 20, batch 1700, loss[loss=0.2008, simple_loss=0.2869, pruned_loss=0.05739, over 19768.00 frames. ], tot_loss[loss=0.2142, simple_loss=0.293, pruned_loss=0.06774, over 3802747.21 frames. ], batch size: 56, lr: 4.15e-03, grad_scale: 8.0 +2023-04-02 16:14:17,619 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=131434.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:14:55,853 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 16:15:16,171 INFO [train.py:903] (0/4) Epoch 20, batch 1750, loss[loss=0.1534, simple_loss=0.2373, pruned_loss=0.03477, over 19298.00 frames. ], tot_loss[loss=0.2132, simple_loss=0.292, pruned_loss=0.06723, over 3807382.28 frames. ], batch size: 44, lr: 4.15e-03, grad_scale: 8.0 +2023-04-02 16:15:52,691 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=131510.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:16:02,690 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.468e+02 5.359e+02 6.412e+02 7.691e+02 1.507e+03, threshold=1.282e+03, percent-clipped=3.0 +2023-04-02 16:16:18,693 INFO [train.py:903] (0/4) Epoch 20, batch 1800, loss[loss=0.189, simple_loss=0.2726, pruned_loss=0.0527, over 19749.00 frames. ], tot_loss[loss=0.2122, simple_loss=0.2913, pruned_loss=0.06654, over 3824093.57 frames. ], batch size: 51, lr: 4.15e-03, grad_scale: 4.0 +2023-04-02 16:16:24,470 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=131536.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:17:16,070 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 16:17:22,062 INFO [train.py:903] (0/4) Epoch 20, batch 1850, loss[loss=0.2172, simple_loss=0.2946, pruned_loss=0.06995, over 19483.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2915, pruned_loss=0.06687, over 3809689.95 frames. ], batch size: 49, lr: 4.15e-03, grad_scale: 4.0 +2023-04-02 16:17:33,547 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2006, 1.0150, 1.4541, 1.2609, 2.2503, 3.1962, 2.9237, 3.5730], + device='cuda:0'), covar=tensor([0.1826, 0.5248, 0.4597, 0.2524, 0.0780, 0.0266, 0.0323, 0.0279], + device='cuda:0'), in_proj_covar=tensor([0.0269, 0.0317, 0.0348, 0.0263, 0.0238, 0.0183, 0.0214, 0.0250], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 16:17:54,263 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 16:18:09,336 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4743, 1.6002, 2.2071, 1.8359, 3.0418, 2.4545, 3.2116, 1.5720], + device='cuda:0'), covar=tensor([0.2636, 0.4604, 0.2686, 0.1998, 0.1631, 0.2389, 0.1953, 0.4427], + device='cuda:0'), in_proj_covar=tensor([0.0526, 0.0631, 0.0694, 0.0474, 0.0612, 0.0526, 0.0658, 0.0539], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 16:18:09,977 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.335e+02 4.826e+02 6.634e+02 9.045e+02 2.049e+03, threshold=1.327e+03, percent-clipped=6.0 +2023-04-02 16:18:25,162 INFO [train.py:903] (0/4) Epoch 20, batch 1900, loss[loss=0.2253, simple_loss=0.3026, pruned_loss=0.07396, over 17291.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.2921, pruned_loss=0.0669, over 3809779.77 frames. ], batch size: 101, lr: 4.15e-03, grad_scale: 4.0 +2023-04-02 16:18:40,086 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 16:18:45,471 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 16:18:47,918 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131651.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:19:10,717 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 16:19:25,446 INFO [train.py:903] (0/4) Epoch 20, batch 1950, loss[loss=0.1946, simple_loss=0.2835, pruned_loss=0.05285, over 19756.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2922, pruned_loss=0.06703, over 3814409.51 frames. ], batch size: 54, lr: 4.15e-03, grad_scale: 4.0 +2023-04-02 16:19:49,104 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8437, 1.3492, 1.0275, 0.9395, 1.1550, 0.9457, 0.9653, 1.2118], + device='cuda:0'), covar=tensor([0.0676, 0.0766, 0.1137, 0.0746, 0.0565, 0.1337, 0.0594, 0.0508], + device='cuda:0'), in_proj_covar=tensor([0.0306, 0.0316, 0.0335, 0.0263, 0.0248, 0.0337, 0.0293, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 16:20:03,185 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0475, 2.5132, 1.8118, 1.7977, 2.3532, 1.7417, 1.6792, 2.1136], + device='cuda:0'), covar=tensor([0.1035, 0.0855, 0.0798, 0.0732, 0.0485, 0.0897, 0.0710, 0.0567], + device='cuda:0'), in_proj_covar=tensor([0.0305, 0.0315, 0.0335, 0.0263, 0.0247, 0.0336, 0.0292, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 16:20:13,302 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.458e+02 5.129e+02 6.435e+02 8.344e+02 2.370e+03, threshold=1.287e+03, percent-clipped=2.0 +2023-04-02 16:20:28,634 INFO [train.py:903] (0/4) Epoch 20, batch 2000, loss[loss=0.2213, simple_loss=0.3038, pruned_loss=0.06944, over 19279.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2922, pruned_loss=0.06721, over 3802896.09 frames. ], batch size: 66, lr: 4.14e-03, grad_scale: 8.0 +2023-04-02 16:21:18,583 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=131771.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:21:26,462 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 16:21:28,767 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=131778.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:21:33,340 INFO [train.py:903] (0/4) Epoch 20, batch 2050, loss[loss=0.1933, simple_loss=0.2815, pruned_loss=0.05257, over 19682.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.2918, pruned_loss=0.06701, over 3790358.29 frames. ], batch size: 59, lr: 4.14e-03, grad_scale: 8.0 +2023-04-02 16:21:45,609 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 16:21:46,767 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 16:22:07,705 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 16:22:22,550 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.239e+02 4.827e+02 6.005e+02 7.777e+02 1.829e+03, threshold=1.201e+03, percent-clipped=4.0 +2023-04-02 16:22:25,301 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3028, 1.2731, 1.2837, 1.3630, 1.0279, 1.3803, 1.2980, 1.3746], + device='cuda:0'), covar=tensor([0.0939, 0.1012, 0.1091, 0.0695, 0.0927, 0.0861, 0.0894, 0.0799], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0225, 0.0243, 0.0228, 0.0211, 0.0187, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 16:22:35,848 INFO [train.py:903] (0/4) Epoch 20, batch 2100, loss[loss=0.2257, simple_loss=0.3069, pruned_loss=0.07221, over 19481.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.2921, pruned_loss=0.06742, over 3791383.83 frames. ], batch size: 64, lr: 4.14e-03, grad_scale: 4.0 +2023-04-02 16:23:02,414 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 16:23:02,523 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=131854.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:23:26,816 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 16:23:29,473 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=131875.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:23:37,612 INFO [train.py:903] (0/4) Epoch 20, batch 2150, loss[loss=0.2216, simple_loss=0.2945, pruned_loss=0.07438, over 19486.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2901, pruned_loss=0.06633, over 3804035.46 frames. ], batch size: 49, lr: 4.14e-03, grad_scale: 4.0 +2023-04-02 16:23:42,649 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131886.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:23:51,930 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131893.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:24:09,638 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=131907.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:24:16,678 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1995, 1.2787, 1.2317, 1.0210, 1.0680, 1.0372, 0.1098, 0.3591], + device='cuda:0'), covar=tensor([0.0727, 0.0678, 0.0432, 0.0628, 0.1273, 0.0679, 0.1319, 0.1091], + device='cuda:0'), in_proj_covar=tensor([0.0359, 0.0351, 0.0356, 0.0382, 0.0455, 0.0386, 0.0335, 0.0338], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 16:24:26,289 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.293e+02 5.118e+02 6.039e+02 8.265e+02 1.505e+03, threshold=1.208e+03, percent-clipped=4.0 +2023-04-02 16:24:39,695 INFO [train.py:903] (0/4) Epoch 20, batch 2200, loss[loss=0.2641, simple_loss=0.3348, pruned_loss=0.09668, over 19684.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2911, pruned_loss=0.06721, over 3793193.94 frames. ], batch size: 59, lr: 4.14e-03, grad_scale: 4.0 +2023-04-02 16:24:40,118 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=131932.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:25:17,293 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1593, 1.9889, 1.7662, 2.1176, 1.9476, 1.9054, 1.6329, 2.0553], + device='cuda:0'), covar=tensor([0.1014, 0.1461, 0.1490, 0.1078, 0.1370, 0.0523, 0.1436, 0.0678], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0356, 0.0309, 0.0250, 0.0300, 0.0249, 0.0305, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 16:25:26,135 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=131969.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:25:42,868 INFO [train.py:903] (0/4) Epoch 20, batch 2250, loss[loss=0.1959, simple_loss=0.2747, pruned_loss=0.0586, over 19612.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2917, pruned_loss=0.06759, over 3806851.48 frames. ], batch size: 50, lr: 4.14e-03, grad_scale: 4.0 +2023-04-02 16:26:04,599 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-132000.pt +2023-04-02 16:26:31,962 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.368e+02 4.978e+02 6.272e+02 7.965e+02 1.499e+03, threshold=1.254e+03, percent-clipped=2.0 +2023-04-02 16:26:44,539 INFO [train.py:903] (0/4) Epoch 20, batch 2300, loss[loss=0.2481, simple_loss=0.3216, pruned_loss=0.08725, over 19644.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2927, pruned_loss=0.06797, over 3808670.94 frames. ], batch size: 58, lr: 4.14e-03, grad_scale: 4.0 +2023-04-02 16:26:58,039 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 16:27:05,368 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=132049.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:27:34,684 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=132072.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:27:47,305 INFO [train.py:903] (0/4) Epoch 20, batch 2350, loss[loss=0.1874, simple_loss=0.2678, pruned_loss=0.05348, over 19851.00 frames. ], tot_loss[loss=0.2151, simple_loss=0.2936, pruned_loss=0.06828, over 3802446.58 frames. ], batch size: 52, lr: 4.14e-03, grad_scale: 4.0 +2023-04-02 16:28:26,794 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 16:28:35,983 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.260e+02 4.795e+02 5.776e+02 7.778e+02 1.972e+03, threshold=1.155e+03, percent-clipped=8.0 +2023-04-02 16:28:42,717 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 16:28:49,774 INFO [train.py:903] (0/4) Epoch 20, batch 2400, loss[loss=0.1874, simple_loss=0.2774, pruned_loss=0.04868, over 19841.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2922, pruned_loss=0.06724, over 3814614.88 frames. ], batch size: 52, lr: 4.14e-03, grad_scale: 8.0 +2023-04-02 16:29:03,424 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132142.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:29:11,458 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132149.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:29:32,881 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132167.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:29:41,666 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132174.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:29:51,981 INFO [train.py:903] (0/4) Epoch 20, batch 2450, loss[loss=0.1971, simple_loss=0.2677, pruned_loss=0.06329, over 19754.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2907, pruned_loss=0.06635, over 3819431.58 frames. ], batch size: 46, lr: 4.14e-03, grad_scale: 8.0 +2023-04-02 16:30:07,511 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.49 vs. limit=5.0 +2023-04-02 16:30:38,830 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=132219.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:30:41,020 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.150e+02 4.995e+02 6.263e+02 8.063e+02 1.363e+03, threshold=1.253e+03, percent-clipped=5.0 +2023-04-02 16:30:47,013 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132225.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:30:54,280 INFO [train.py:903] (0/4) Epoch 20, batch 2500, loss[loss=0.2169, simple_loss=0.2945, pruned_loss=0.06961, over 19500.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2902, pruned_loss=0.06637, over 3819977.35 frames. ], batch size: 64, lr: 4.14e-03, grad_scale: 8.0 +2023-04-02 16:31:11,671 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.70 vs. limit=5.0 +2023-04-02 16:31:15,973 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132250.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:31:56,368 INFO [train.py:903] (0/4) Epoch 20, batch 2550, loss[loss=0.2224, simple_loss=0.2937, pruned_loss=0.07552, over 19609.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2902, pruned_loss=0.06659, over 3822083.25 frames. ], batch size: 50, lr: 4.14e-03, grad_scale: 8.0 +2023-04-02 16:32:45,610 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.513e+02 4.985e+02 5.787e+02 8.066e+02 1.995e+03, threshold=1.157e+03, percent-clipped=4.0 +2023-04-02 16:32:52,674 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 16:32:58,506 INFO [train.py:903] (0/4) Epoch 20, batch 2600, loss[loss=0.216, simple_loss=0.2995, pruned_loss=0.06618, over 18159.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2905, pruned_loss=0.06657, over 3818381.36 frames. ], batch size: 83, lr: 4.14e-03, grad_scale: 8.0 +2023-04-02 16:33:02,427 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=132334.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:33:39,103 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6210, 1.3985, 1.4387, 2.1351, 1.6235, 1.9161, 1.9381, 1.6618], + device='cuda:0'), covar=tensor([0.0851, 0.1004, 0.1092, 0.0758, 0.0897, 0.0758, 0.0909, 0.0721], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0220, 0.0224, 0.0242, 0.0228, 0.0210, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 16:34:01,399 INFO [train.py:903] (0/4) Epoch 20, batch 2650, loss[loss=0.2391, simple_loss=0.3235, pruned_loss=0.0773, over 18285.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2906, pruned_loss=0.06618, over 3835346.65 frames. ], batch size: 83, lr: 4.13e-03, grad_scale: 8.0 +2023-04-02 16:34:15,314 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=132393.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:34:23,137 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 16:34:44,187 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=132416.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:34:50,449 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.050e+02 4.865e+02 6.149e+02 7.368e+02 1.585e+03, threshold=1.230e+03, percent-clipped=4.0 +2023-04-02 16:35:04,061 INFO [train.py:903] (0/4) Epoch 20, batch 2700, loss[loss=0.2259, simple_loss=0.3039, pruned_loss=0.07395, over 19596.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.291, pruned_loss=0.06627, over 3839604.94 frames. ], batch size: 52, lr: 4.13e-03, grad_scale: 8.0 +2023-04-02 16:35:14,882 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7416, 1.7106, 1.6275, 1.3929, 1.2921, 1.4659, 0.2194, 0.6889], + device='cuda:0'), covar=tensor([0.0560, 0.0560, 0.0349, 0.0561, 0.1108, 0.0653, 0.1165, 0.0947], + device='cuda:0'), in_proj_covar=tensor([0.0357, 0.0351, 0.0354, 0.0381, 0.0454, 0.0383, 0.0333, 0.0336], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 16:36:06,699 INFO [train.py:903] (0/4) Epoch 20, batch 2750, loss[loss=0.1964, simple_loss=0.2805, pruned_loss=0.05617, over 19658.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2904, pruned_loss=0.06623, over 3821091.94 frames. ], batch size: 58, lr: 4.13e-03, grad_scale: 8.0 +2023-04-02 16:36:39,128 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=132508.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:36:55,659 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.965e+02 5.186e+02 6.180e+02 7.968e+02 1.505e+03, threshold=1.236e+03, percent-clipped=2.0 +2023-04-02 16:37:07,814 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=132531.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:37:08,659 INFO [train.py:903] (0/4) Epoch 20, batch 2800, loss[loss=0.272, simple_loss=0.3335, pruned_loss=0.1052, over 13019.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2915, pruned_loss=0.0669, over 3810962.25 frames. ], batch size: 135, lr: 4.13e-03, grad_scale: 8.0 +2023-04-02 16:37:27,250 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2051, 2.2153, 2.5376, 3.1428, 2.2658, 2.9274, 2.6330, 2.2373], + device='cuda:0'), covar=tensor([0.4223, 0.4162, 0.1785, 0.2261, 0.4323, 0.2031, 0.4245, 0.3305], + device='cuda:0'), in_proj_covar=tensor([0.0883, 0.0942, 0.0704, 0.0926, 0.0862, 0.0797, 0.0833, 0.0773], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 16:38:13,003 INFO [train.py:903] (0/4) Epoch 20, batch 2850, loss[loss=0.2445, simple_loss=0.3228, pruned_loss=0.0831, over 19702.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2915, pruned_loss=0.06683, over 3814902.12 frames. ], batch size: 59, lr: 4.13e-03, grad_scale: 4.0 +2023-04-02 16:38:22,397 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132590.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:38:26,809 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0685, 2.2120, 1.6730, 2.1361, 2.2757, 1.6436, 1.6786, 1.9820], + device='cuda:0'), covar=tensor([0.1165, 0.1563, 0.1851, 0.1323, 0.1384, 0.0959, 0.1786, 0.0995], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0356, 0.0311, 0.0251, 0.0300, 0.0249, 0.0305, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 16:38:52,959 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132615.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:38:54,131 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3444, 2.4421, 2.1631, 2.5768, 2.4355, 2.1009, 2.0378, 2.4636], + device='cuda:0'), covar=tensor([0.0963, 0.1403, 0.1302, 0.0983, 0.1208, 0.0513, 0.1289, 0.0634], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0355, 0.0309, 0.0249, 0.0299, 0.0248, 0.0304, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 16:39:01,716 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.639e+02 4.861e+02 5.815e+02 7.642e+02 3.357e+03, threshold=1.163e+03, percent-clipped=7.0 +2023-04-02 16:39:14,638 INFO [train.py:903] (0/4) Epoch 20, batch 2900, loss[loss=0.1954, simple_loss=0.2782, pruned_loss=0.05629, over 19488.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2907, pruned_loss=0.0665, over 3802962.07 frames. ], batch size: 64, lr: 4.13e-03, grad_scale: 4.0 +2023-04-02 16:39:15,642 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 16:40:13,482 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=132678.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:40:18,618 INFO [train.py:903] (0/4) Epoch 20, batch 2950, loss[loss=0.1711, simple_loss=0.2648, pruned_loss=0.03867, over 19530.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2894, pruned_loss=0.06544, over 3811112.67 frames. ], batch size: 56, lr: 4.13e-03, grad_scale: 4.0 +2023-04-02 16:40:21,815 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-02 16:41:09,140 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.010e+02 4.627e+02 5.679e+02 7.371e+02 2.153e+03, threshold=1.136e+03, percent-clipped=3.0 +2023-04-02 16:41:20,813 INFO [train.py:903] (0/4) Epoch 20, batch 3000, loss[loss=0.2371, simple_loss=0.3139, pruned_loss=0.08014, over 19552.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2884, pruned_loss=0.06473, over 3820485.62 frames. ], batch size: 61, lr: 4.13e-03, grad_scale: 4.0 +2023-04-02 16:41:20,813 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 16:41:34,266 INFO [train.py:937] (0/4) Epoch 20, validation: loss=0.1695, simple_loss=0.2697, pruned_loss=0.03462, over 944034.00 frames. +2023-04-02 16:41:34,268 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 16:41:40,268 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 16:42:10,887 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9051, 4.3157, 4.6469, 4.6407, 1.7625, 4.3496, 3.7948, 4.3516], + device='cuda:0'), covar=tensor([0.1674, 0.0876, 0.0618, 0.0632, 0.6157, 0.0949, 0.0634, 0.1186], + device='cuda:0'), in_proj_covar=tensor([0.0765, 0.0723, 0.0919, 0.0806, 0.0819, 0.0682, 0.0555, 0.0859], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 16:42:12,911 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=132764.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:42:13,059 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132764.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:42:15,794 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.09 vs. limit=2.0 +2023-04-02 16:42:35,229 INFO [train.py:903] (0/4) Epoch 20, batch 3050, loss[loss=0.2014, simple_loss=0.2845, pruned_loss=0.05912, over 19531.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2894, pruned_loss=0.06542, over 3812407.24 frames. ], batch size: 54, lr: 4.13e-03, grad_scale: 4.0 +2023-04-02 16:42:41,476 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=132787.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:42:43,575 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132789.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:43:13,147 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=132812.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:43:24,233 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.615e+02 4.785e+02 6.187e+02 7.720e+02 1.879e+03, threshold=1.237e+03, percent-clipped=7.0 +2023-04-02 16:43:28,313 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9033, 2.0024, 2.1836, 2.6260, 1.8918, 2.4398, 2.2473, 1.9719], + device='cuda:0'), covar=tensor([0.4191, 0.3946, 0.1892, 0.2273, 0.4117, 0.2108, 0.4681, 0.3441], + device='cuda:0'), in_proj_covar=tensor([0.0886, 0.0948, 0.0708, 0.0931, 0.0866, 0.0800, 0.0837, 0.0776], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 16:43:37,016 INFO [train.py:903] (0/4) Epoch 20, batch 3100, loss[loss=0.236, simple_loss=0.3179, pruned_loss=0.07708, over 17260.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.2895, pruned_loss=0.06569, over 3807499.36 frames. ], batch size: 100, lr: 4.13e-03, grad_scale: 4.0 +2023-04-02 16:43:39,637 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3248, 2.1534, 2.0356, 1.9010, 1.6785, 1.8457, 0.6043, 1.2881], + device='cuda:0'), covar=tensor([0.0529, 0.0555, 0.0445, 0.0775, 0.1024, 0.0846, 0.1227, 0.0964], + device='cuda:0'), in_proj_covar=tensor([0.0360, 0.0352, 0.0357, 0.0383, 0.0457, 0.0386, 0.0334, 0.0339], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 16:44:33,776 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8871, 4.3256, 4.6147, 4.6535, 1.6980, 4.3480, 3.7684, 4.3404], + device='cuda:0'), covar=tensor([0.1726, 0.0864, 0.0636, 0.0688, 0.6155, 0.0915, 0.0667, 0.1230], + device='cuda:0'), in_proj_covar=tensor([0.0766, 0.0722, 0.0921, 0.0807, 0.0821, 0.0682, 0.0556, 0.0859], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 16:44:40,221 INFO [train.py:903] (0/4) Epoch 20, batch 3150, loss[loss=0.2886, simple_loss=0.3525, pruned_loss=0.1124, over 13059.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2896, pruned_loss=0.0658, over 3812850.32 frames. ], batch size: 136, lr: 4.13e-03, grad_scale: 4.0 +2023-04-02 16:45:07,813 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 16:45:11,639 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5546, 1.6260, 1.9062, 1.7898, 2.6453, 2.3110, 2.7534, 1.3971], + device='cuda:0'), covar=tensor([0.2258, 0.3804, 0.2483, 0.1789, 0.1391, 0.2011, 0.1395, 0.4027], + device='cuda:0'), in_proj_covar=tensor([0.0524, 0.0630, 0.0694, 0.0476, 0.0612, 0.0523, 0.0656, 0.0540], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 16:45:29,824 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.089e+02 5.025e+02 5.951e+02 7.011e+02 1.371e+03, threshold=1.190e+03, percent-clipped=2.0 +2023-04-02 16:45:42,509 INFO [train.py:903] (0/4) Epoch 20, batch 3200, loss[loss=0.1857, simple_loss=0.266, pruned_loss=0.05272, over 19583.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2891, pruned_loss=0.06575, over 3829035.25 frames. ], batch size: 52, lr: 4.13e-03, grad_scale: 8.0 +2023-04-02 16:46:13,797 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7392, 1.7220, 1.6998, 1.4633, 1.3583, 1.5000, 0.2287, 0.7558], + device='cuda:0'), covar=tensor([0.0664, 0.0606, 0.0366, 0.0600, 0.1211, 0.0676, 0.1188, 0.1008], + device='cuda:0'), in_proj_covar=tensor([0.0360, 0.0352, 0.0358, 0.0383, 0.0458, 0.0386, 0.0334, 0.0339], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 16:46:14,351 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-04-02 16:46:46,021 INFO [train.py:903] (0/4) Epoch 20, batch 3250, loss[loss=0.2403, simple_loss=0.3091, pruned_loss=0.08572, over 19849.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2892, pruned_loss=0.06576, over 3832399.53 frames. ], batch size: 52, lr: 4.13e-03, grad_scale: 8.0 +2023-04-02 16:47:37,650 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.383e+02 4.867e+02 6.333e+02 8.818e+02 1.782e+03, threshold=1.267e+03, percent-clipped=7.0 +2023-04-02 16:47:37,813 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=133022.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:47:49,136 INFO [train.py:903] (0/4) Epoch 20, batch 3300, loss[loss=0.2112, simple_loss=0.2955, pruned_loss=0.0634, over 18869.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2913, pruned_loss=0.067, over 3831149.95 frames. ], batch size: 74, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:47:57,159 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 16:48:03,452 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4934, 1.3025, 1.8282, 1.3325, 2.6200, 3.5902, 3.2981, 3.8179], + device='cuda:0'), covar=tensor([0.1455, 0.3734, 0.3131, 0.2248, 0.0562, 0.0187, 0.0212, 0.0272], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0319, 0.0349, 0.0263, 0.0240, 0.0184, 0.0215, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 16:48:54,792 INFO [train.py:903] (0/4) Epoch 20, batch 3350, loss[loss=0.2035, simple_loss=0.2922, pruned_loss=0.0574, over 19678.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2902, pruned_loss=0.06648, over 3822335.78 frames. ], batch size: 53, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:49:27,411 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=133108.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:49:45,520 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.359e+02 4.956e+02 6.093e+02 7.175e+02 1.819e+03, threshold=1.219e+03, percent-clipped=1.0 +2023-04-02 16:49:57,526 INFO [train.py:903] (0/4) Epoch 20, batch 3400, loss[loss=0.2046, simple_loss=0.2921, pruned_loss=0.05854, over 19700.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2902, pruned_loss=0.06598, over 3818294.78 frames. ], batch size: 59, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:50:06,193 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=133137.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:51:02,179 INFO [train.py:903] (0/4) Epoch 20, batch 3450, loss[loss=0.2039, simple_loss=0.2953, pruned_loss=0.05626, over 19670.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2917, pruned_loss=0.06657, over 3825508.80 frames. ], batch size: 58, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:51:08,047 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 16:51:41,614 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4198, 2.0718, 1.5591, 1.2918, 1.9840, 1.1963, 1.3018, 1.8880], + device='cuda:0'), covar=tensor([0.1090, 0.0766, 0.1136, 0.0926, 0.0519, 0.1365, 0.0800, 0.0479], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0313, 0.0333, 0.0259, 0.0245, 0.0335, 0.0291, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 16:51:52,696 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.981e+02 4.746e+02 5.634e+02 7.504e+02 1.582e+03, threshold=1.127e+03, percent-clipped=2.0 +2023-04-02 16:51:54,138 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=133223.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:52:04,226 INFO [train.py:903] (0/4) Epoch 20, batch 3500, loss[loss=0.2131, simple_loss=0.2999, pruned_loss=0.06318, over 19680.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.292, pruned_loss=0.06714, over 3827187.26 frames. ], batch size: 59, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:53:08,045 INFO [train.py:903] (0/4) Epoch 20, batch 3550, loss[loss=0.2339, simple_loss=0.3055, pruned_loss=0.08115, over 19704.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2921, pruned_loss=0.06781, over 3822070.56 frames. ], batch size: 59, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:53:15,827 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.09 vs. limit=2.0 +2023-04-02 16:53:52,740 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-02 16:53:58,721 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.076e+02 5.017e+02 5.933e+02 7.980e+02 2.795e+03, threshold=1.187e+03, percent-clipped=11.0 +2023-04-02 16:54:10,371 INFO [train.py:903] (0/4) Epoch 20, batch 3600, loss[loss=0.2163, simple_loss=0.3007, pruned_loss=0.06589, over 19837.00 frames. ], tot_loss[loss=0.2138, simple_loss=0.2923, pruned_loss=0.0677, over 3819565.72 frames. ], batch size: 52, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:54:48,395 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0158, 2.0963, 2.3490, 2.7159, 2.0049, 2.5989, 2.3704, 2.1288], + device='cuda:0'), covar=tensor([0.4279, 0.4031, 0.1802, 0.2420, 0.4202, 0.2134, 0.4761, 0.3316], + device='cuda:0'), in_proj_covar=tensor([0.0883, 0.0945, 0.0705, 0.0928, 0.0864, 0.0799, 0.0834, 0.0771], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 16:55:15,902 INFO [train.py:903] (0/4) Epoch 20, batch 3650, loss[loss=0.2698, simple_loss=0.3537, pruned_loss=0.09294, over 19281.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2918, pruned_loss=0.06747, over 3808794.45 frames. ], batch size: 70, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:55:28,068 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=133392.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 16:55:29,466 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=133393.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:55:30,464 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=133394.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:55:32,920 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8795, 1.7989, 1.4706, 1.8180, 1.6995, 1.4666, 1.4565, 1.7420], + device='cuda:0'), covar=tensor([0.1138, 0.1492, 0.1743, 0.1234, 0.1457, 0.0943, 0.1717, 0.0934], + device='cuda:0'), in_proj_covar=tensor([0.0261, 0.0350, 0.0302, 0.0245, 0.0294, 0.0244, 0.0298, 0.0248], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 16:56:02,498 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=133418.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:56:06,667 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.934e+02 5.113e+02 6.456e+02 7.889e+02 1.610e+03, threshold=1.291e+03, percent-clipped=2.0 +2023-04-02 16:56:18,443 INFO [train.py:903] (0/4) Epoch 20, batch 3700, loss[loss=0.1775, simple_loss=0.2598, pruned_loss=0.04763, over 19624.00 frames. ], tot_loss[loss=0.2135, simple_loss=0.292, pruned_loss=0.06748, over 3805575.36 frames. ], batch size: 50, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:57:02,955 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-04-02 16:57:19,791 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=133479.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:57:23,016 INFO [train.py:903] (0/4) Epoch 20, batch 3750, loss[loss=0.2352, simple_loss=0.3116, pruned_loss=0.07936, over 19330.00 frames. ], tot_loss[loss=0.2148, simple_loss=0.293, pruned_loss=0.06829, over 3802492.56 frames. ], batch size: 66, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:57:34,881 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=133492.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:57:39,326 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6336, 1.7348, 1.9778, 2.0203, 1.5039, 1.8857, 2.0110, 1.8788], + device='cuda:0'), covar=tensor([0.3859, 0.3151, 0.1704, 0.2016, 0.3326, 0.1979, 0.4588, 0.3048], + device='cuda:0'), in_proj_covar=tensor([0.0881, 0.0942, 0.0703, 0.0923, 0.0861, 0.0794, 0.0832, 0.0768], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 16:57:50,915 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=133504.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 16:58:13,508 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.259e+02 5.077e+02 5.935e+02 8.206e+02 1.595e+03, threshold=1.187e+03, percent-clipped=3.0 +2023-04-02 16:58:24,813 INFO [train.py:903] (0/4) Epoch 20, batch 3800, loss[loss=0.2786, simple_loss=0.3411, pruned_loss=0.1081, over 19665.00 frames. ], tot_loss[loss=0.215, simple_loss=0.2935, pruned_loss=0.0683, over 3805620.00 frames. ], batch size: 59, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:58:58,170 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 16:59:30,203 INFO [train.py:903] (0/4) Epoch 20, batch 3850, loss[loss=0.1778, simple_loss=0.264, pruned_loss=0.04576, over 19778.00 frames. ], tot_loss[loss=0.2139, simple_loss=0.2923, pruned_loss=0.06774, over 3809916.34 frames. ], batch size: 48, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 16:59:31,738 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8777, 0.8295, 0.8489, 0.9042, 0.7922, 0.9476, 0.9620, 0.9037], + device='cuda:0'), covar=tensor([0.0687, 0.0750, 0.0800, 0.0587, 0.0764, 0.0671, 0.0687, 0.0600], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0221, 0.0224, 0.0242, 0.0227, 0.0210, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 16:59:48,047 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.38 vs. limit=5.0 +2023-04-02 17:00:20,972 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.481e+02 5.079e+02 6.219e+02 7.261e+02 1.808e+03, threshold=1.244e+03, percent-clipped=5.0 +2023-04-02 17:00:32,688 INFO [train.py:903] (0/4) Epoch 20, batch 3900, loss[loss=0.2469, simple_loss=0.3241, pruned_loss=0.08486, over 18038.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2918, pruned_loss=0.0671, over 3816135.50 frames. ], batch size: 83, lr: 4.12e-03, grad_scale: 8.0 +2023-04-02 17:01:21,113 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=133670.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:01:37,403 INFO [train.py:903] (0/4) Epoch 20, batch 3950, loss[loss=0.2014, simple_loss=0.2868, pruned_loss=0.05803, over 19647.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2918, pruned_loss=0.06711, over 3798132.89 frames. ], batch size: 55, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:01:42,253 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 17:01:50,353 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-04-02 17:02:26,843 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.611e+02 5.167e+02 6.244e+02 7.613e+02 1.189e+03, threshold=1.249e+03, percent-clipped=0.0 +2023-04-02 17:02:38,784 INFO [train.py:903] (0/4) Epoch 20, batch 4000, loss[loss=0.2147, simple_loss=0.2959, pruned_loss=0.06677, over 19415.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2904, pruned_loss=0.06608, over 3814055.83 frames. ], batch size: 70, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:02:43,766 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=133736.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 17:02:46,990 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=133738.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:02:53,757 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-02 17:03:26,442 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 17:03:41,991 INFO [train.py:903] (0/4) Epoch 20, batch 4050, loss[loss=0.2391, simple_loss=0.3104, pruned_loss=0.08391, over 19533.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2901, pruned_loss=0.06613, over 3804347.62 frames. ], batch size: 56, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:04:30,941 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.559e+02 4.971e+02 6.413e+02 8.150e+02 1.897e+03, threshold=1.283e+03, percent-clipped=7.0 +2023-04-02 17:04:42,288 INFO [train.py:903] (0/4) Epoch 20, batch 4100, loss[loss=0.2175, simple_loss=0.3001, pruned_loss=0.06743, over 19522.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2913, pruned_loss=0.06696, over 3798444.84 frames. ], batch size: 56, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:04:47,156 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=133836.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:04:58,532 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0737, 1.7536, 1.8413, 2.7739, 2.0285, 2.3652, 2.5012, 2.0918], + device='cuda:0'), covar=tensor([0.0828, 0.0934, 0.0989, 0.0807, 0.0853, 0.0727, 0.0792, 0.0699], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0219, 0.0223, 0.0240, 0.0225, 0.0208, 0.0186, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 17:05:06,743 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=133851.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 17:05:09,020 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=133853.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:05:18,894 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 17:05:35,459 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9505, 2.0494, 2.2736, 2.7544, 2.0848, 2.5949, 2.3469, 2.0733], + device='cuda:0'), covar=tensor([0.4249, 0.3843, 0.1837, 0.2159, 0.3891, 0.1982, 0.4433, 0.3209], + device='cuda:0'), in_proj_covar=tensor([0.0887, 0.0948, 0.0707, 0.0931, 0.0868, 0.0801, 0.0836, 0.0773], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 17:05:45,978 INFO [train.py:903] (0/4) Epoch 20, batch 4150, loss[loss=0.1801, simple_loss=0.2547, pruned_loss=0.05278, over 19723.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2906, pruned_loss=0.0668, over 3803036.83 frames. ], batch size: 46, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:06:00,788 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-04-02 17:06:19,805 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5325, 1.6140, 1.9671, 1.7244, 3.0907, 2.6064, 3.5756, 1.5359], + device='cuda:0'), covar=tensor([0.2329, 0.3997, 0.2578, 0.1788, 0.1455, 0.1998, 0.1442, 0.4041], + device='cuda:0'), in_proj_covar=tensor([0.0525, 0.0633, 0.0698, 0.0477, 0.0613, 0.0524, 0.0658, 0.0541], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 17:06:35,678 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.109e+02 4.757e+02 5.877e+02 6.653e+02 1.329e+03, threshold=1.175e+03, percent-clipped=1.0 +2023-04-02 17:06:47,879 INFO [train.py:903] (0/4) Epoch 20, batch 4200, loss[loss=0.1978, simple_loss=0.2747, pruned_loss=0.0604, over 19605.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2898, pruned_loss=0.06619, over 3810433.91 frames. ], batch size: 50, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:06:51,427 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 17:07:12,220 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=133951.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:07:30,706 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3659, 3.9733, 2.6760, 3.4903, 0.7582, 3.9279, 3.8048, 3.9054], + device='cuda:0'), covar=tensor([0.0728, 0.1023, 0.1982, 0.0875, 0.4136, 0.0757, 0.0872, 0.1109], + device='cuda:0'), in_proj_covar=tensor([0.0492, 0.0398, 0.0483, 0.0342, 0.0399, 0.0423, 0.0415, 0.0449], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 17:07:31,264 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-04-02 17:07:50,933 INFO [train.py:903] (0/4) Epoch 20, batch 4250, loss[loss=0.2322, simple_loss=0.3106, pruned_loss=0.07693, over 19622.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2897, pruned_loss=0.06571, over 3824633.90 frames. ], batch size: 57, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:08:08,173 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 17:08:12,876 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-134000.pt +2023-04-02 17:08:20,523 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 17:08:32,199 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=134014.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:08:33,454 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=134015.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:08:36,367 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0 +2023-04-02 17:08:41,181 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.039e+02 5.008e+02 5.790e+02 6.980e+02 1.679e+03, threshold=1.158e+03, percent-clipped=4.0 +2023-04-02 17:08:54,645 INFO [train.py:903] (0/4) Epoch 20, batch 4300, loss[loss=0.197, simple_loss=0.2652, pruned_loss=0.06442, over 19355.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2889, pruned_loss=0.06549, over 3833588.44 frames. ], batch size: 47, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:09:27,709 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-04-02 17:09:50,646 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 17:09:56,296 INFO [train.py:903] (0/4) Epoch 20, batch 4350, loss[loss=0.2058, simple_loss=0.2917, pruned_loss=0.05993, over 19678.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.2891, pruned_loss=0.06527, over 3833773.99 frames. ], batch size: 53, lr: 4.11e-03, grad_scale: 4.0 +2023-04-02 17:10:30,261 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=134107.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 17:10:32,489 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=134109.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:10:48,201 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.336e+02 5.216e+02 6.315e+02 8.361e+02 2.012e+03, threshold=1.263e+03, percent-clipped=10.0 +2023-04-02 17:10:57,734 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134129.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:11:00,848 INFO [train.py:903] (0/4) Epoch 20, batch 4400, loss[loss=0.2241, simple_loss=0.3025, pruned_loss=0.07284, over 19792.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2892, pruned_loss=0.06561, over 3821881.74 frames. ], batch size: 56, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:11:01,273 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=134132.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 17:11:03,606 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=134134.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:11:18,430 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-04-02 17:11:29,061 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 17:11:37,426 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 17:12:05,041 INFO [train.py:903] (0/4) Epoch 20, batch 4450, loss[loss=0.2355, simple_loss=0.3017, pruned_loss=0.08469, over 13514.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2898, pruned_loss=0.06572, over 3804112.73 frames. ], batch size: 135, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:12:36,932 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=134207.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:12:56,171 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.329e+02 4.942e+02 5.989e+02 7.537e+02 1.405e+03, threshold=1.198e+03, percent-clipped=2.0 +2023-04-02 17:13:08,112 INFO [train.py:903] (0/4) Epoch 20, batch 4500, loss[loss=0.272, simple_loss=0.3359, pruned_loss=0.104, over 13992.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.29, pruned_loss=0.06585, over 3810262.37 frames. ], batch size: 136, lr: 4.11e-03, grad_scale: 8.0 +2023-04-02 17:13:08,517 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=134232.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:14:10,220 INFO [train.py:903] (0/4) Epoch 20, batch 4550, loss[loss=0.2066, simple_loss=0.2789, pruned_loss=0.06718, over 19057.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2905, pruned_loss=0.06668, over 3815087.75 frames. ], batch size: 42, lr: 4.11e-03, grad_scale: 4.0 +2023-04-02 17:14:19,190 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 17:14:42,913 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 17:15:01,018 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=134322.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:15:04,301 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.805e+02 4.924e+02 5.886e+02 8.898e+02 2.816e+03, threshold=1.177e+03, percent-clipped=9.0 +2023-04-02 17:15:09,041 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=134327.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:15:10,239 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9683, 2.1248, 1.7299, 2.0684, 2.1088, 1.6984, 1.6890, 1.8687], + device='cuda:0'), covar=tensor([0.0973, 0.1212, 0.1388, 0.0891, 0.1044, 0.0837, 0.1410, 0.0796], + device='cuda:0'), in_proj_covar=tensor([0.0266, 0.0355, 0.0309, 0.0250, 0.0299, 0.0248, 0.0304, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 17:15:14,419 INFO [train.py:903] (0/4) Epoch 20, batch 4600, loss[loss=0.2245, simple_loss=0.3034, pruned_loss=0.07281, over 19617.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2911, pruned_loss=0.06701, over 3803593.72 frames. ], batch size: 61, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:15:47,425 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=134359.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:16:15,891 INFO [train.py:903] (0/4) Epoch 20, batch 4650, loss[loss=0.2184, simple_loss=0.3005, pruned_loss=0.06818, over 19618.00 frames. ], tot_loss[loss=0.213, simple_loss=0.2918, pruned_loss=0.0671, over 3801792.13 frames. ], batch size: 57, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:16:20,718 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=134385.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:16:32,931 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 17:16:44,589 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 17:16:52,767 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=134410.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:17:09,009 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.247e+02 4.575e+02 5.939e+02 8.134e+02 1.295e+03, threshold=1.188e+03, percent-clipped=3.0 +2023-04-02 17:17:14,341 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.41 vs. limit=5.0 +2023-04-02 17:17:19,254 INFO [train.py:903] (0/4) Epoch 20, batch 4700, loss[loss=0.1913, simple_loss=0.2688, pruned_loss=0.05692, over 19587.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2902, pruned_loss=0.06618, over 3801380.79 frames. ], batch size: 52, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:17:23,222 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5967, 2.3136, 1.6963, 1.5757, 2.1280, 1.3522, 1.3706, 1.8969], + device='cuda:0'), covar=tensor([0.1096, 0.0828, 0.0965, 0.0780, 0.0511, 0.1202, 0.0760, 0.0494], + device='cuda:0'), in_proj_covar=tensor([0.0294, 0.0310, 0.0331, 0.0256, 0.0244, 0.0334, 0.0286, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 17:17:41,459 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 17:18:11,952 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134474.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:18:20,723 INFO [train.py:903] (0/4) Epoch 20, batch 4750, loss[loss=0.2223, simple_loss=0.2949, pruned_loss=0.07481, over 19470.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2906, pruned_loss=0.06648, over 3805494.38 frames. ], batch size: 49, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:18:58,065 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-02 17:19:09,037 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 17:19:14,949 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.087e+02 5.001e+02 5.955e+02 7.090e+02 1.974e+03, threshold=1.191e+03, percent-clipped=7.0 +2023-04-02 17:19:15,401 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3261, 1.3347, 1.4827, 1.5096, 1.7285, 1.8515, 1.7375, 0.5484], + device='cuda:0'), covar=tensor([0.2315, 0.4271, 0.2555, 0.1905, 0.1688, 0.2253, 0.1483, 0.4747], + device='cuda:0'), in_proj_covar=tensor([0.0525, 0.0635, 0.0697, 0.0477, 0.0615, 0.0523, 0.0658, 0.0543], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 17:19:25,462 INFO [train.py:903] (0/4) Epoch 20, batch 4800, loss[loss=0.1947, simple_loss=0.2747, pruned_loss=0.05733, over 19750.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2894, pruned_loss=0.0657, over 3820838.33 frames. ], batch size: 54, lr: 4.10e-03, grad_scale: 8.0 +2023-04-02 17:20:15,904 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6598, 1.6194, 1.5545, 2.1244, 1.6182, 2.0134, 1.8516, 1.6812], + device='cuda:0'), covar=tensor([0.0810, 0.0864, 0.0957, 0.0686, 0.0847, 0.0686, 0.0896, 0.0694], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0220, 0.0222, 0.0241, 0.0224, 0.0208, 0.0185, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 17:20:26,805 INFO [train.py:903] (0/4) Epoch 20, batch 4850, loss[loss=0.2018, simple_loss=0.2727, pruned_loss=0.06545, over 19419.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2894, pruned_loss=0.06554, over 3817892.45 frames. ], batch size: 48, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:20:50,078 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 17:21:13,058 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 17:21:18,750 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 17:21:18,786 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 17:21:21,958 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.496e+02 4.787e+02 5.710e+02 7.760e+02 1.554e+03, threshold=1.142e+03, percent-clipped=3.0 +2023-04-02 17:21:31,384 INFO [train.py:903] (0/4) Epoch 20, batch 4900, loss[loss=0.1961, simple_loss=0.2684, pruned_loss=0.06189, over 19805.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.2891, pruned_loss=0.06524, over 3815444.62 frames. ], batch size: 48, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:21:31,400 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 17:21:51,449 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 17:21:52,995 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3001, 2.3126, 2.5445, 3.0997, 2.2891, 3.0041, 2.5538, 2.3391], + device='cuda:0'), covar=tensor([0.4002, 0.3946, 0.1794, 0.2515, 0.4327, 0.2031, 0.4478, 0.3132], + device='cuda:0'), in_proj_covar=tensor([0.0883, 0.0947, 0.0707, 0.0926, 0.0866, 0.0801, 0.0834, 0.0773], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 17:22:14,358 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=134666.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:22:21,026 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=134671.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:22:33,361 INFO [train.py:903] (0/4) Epoch 20, batch 4950, loss[loss=0.2618, simple_loss=0.3299, pruned_loss=0.09689, over 19524.00 frames. ], tot_loss[loss=0.2133, simple_loss=0.2918, pruned_loss=0.06736, over 3798982.22 frames. ], batch size: 56, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:22:49,195 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 17:23:15,465 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 17:23:28,332 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.953e+02 4.876e+02 5.783e+02 7.285e+02 1.244e+03, threshold=1.157e+03, percent-clipped=2.0 +2023-04-02 17:23:34,564 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=134730.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:23:37,532 INFO [train.py:903] (0/4) Epoch 20, batch 5000, loss[loss=0.2528, simple_loss=0.317, pruned_loss=0.09435, over 13195.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2914, pruned_loss=0.0669, over 3806653.38 frames. ], batch size: 137, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:23:45,504 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 17:23:56,617 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 17:24:06,086 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=134755.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:24:37,268 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3262, 1.7539, 2.0019, 2.0608, 3.9088, 1.3712, 2.7019, 4.1414], + device='cuda:0'), covar=tensor([0.0521, 0.2726, 0.2643, 0.1735, 0.0803, 0.2633, 0.1523, 0.0246], + device='cuda:0'), in_proj_covar=tensor([0.0400, 0.0361, 0.0381, 0.0342, 0.0371, 0.0346, 0.0371, 0.0396], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 17:24:37,308 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134781.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:24:38,191 INFO [train.py:903] (0/4) Epoch 20, batch 5050, loss[loss=0.218, simple_loss=0.3002, pruned_loss=0.06796, over 17254.00 frames. ], tot_loss[loss=0.2143, simple_loss=0.2931, pruned_loss=0.06779, over 3808492.66 frames. ], batch size: 101, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:24:44,175 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=134786.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:24:56,567 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3053, 1.2649, 1.4735, 1.4621, 1.8432, 1.7772, 1.7718, 0.6095], + device='cuda:0'), covar=tensor([0.2637, 0.4609, 0.2802, 0.2175, 0.1672, 0.2594, 0.1544, 0.5001], + device='cuda:0'), in_proj_covar=tensor([0.0523, 0.0631, 0.0693, 0.0476, 0.0612, 0.0521, 0.0654, 0.0540], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 17:25:13,075 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 17:25:31,564 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.532e+02 4.915e+02 6.341e+02 8.226e+02 2.739e+03, threshold=1.268e+03, percent-clipped=9.0 +2023-04-02 17:25:41,326 INFO [train.py:903] (0/4) Epoch 20, batch 5100, loss[loss=0.2574, simple_loss=0.3297, pruned_loss=0.09259, over 19404.00 frames. ], tot_loss[loss=0.2137, simple_loss=0.2925, pruned_loss=0.06745, over 3790563.44 frames. ], batch size: 70, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:25:42,885 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6717, 1.7951, 1.7933, 2.5943, 1.8009, 2.3820, 1.8950, 1.5613], + device='cuda:0'), covar=tensor([0.4751, 0.4366, 0.2757, 0.2589, 0.4306, 0.2245, 0.5981, 0.4987], + device='cuda:0'), in_proj_covar=tensor([0.0882, 0.0946, 0.0707, 0.0926, 0.0865, 0.0799, 0.0834, 0.0773], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 17:25:50,461 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 17:25:53,835 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 17:25:58,181 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 17:26:26,106 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3387, 3.0668, 2.4039, 2.8285, 0.7196, 3.0307, 2.8903, 2.9845], + device='cuda:0'), covar=tensor([0.1076, 0.1373, 0.1960, 0.1033, 0.4083, 0.0942, 0.1073, 0.1449], + device='cuda:0'), in_proj_covar=tensor([0.0490, 0.0399, 0.0486, 0.0341, 0.0398, 0.0422, 0.0415, 0.0450], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 17:26:41,255 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=134880.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:26:43,294 INFO [train.py:903] (0/4) Epoch 20, batch 5150, loss[loss=0.1904, simple_loss=0.2593, pruned_loss=0.06076, over 19741.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2913, pruned_loss=0.06669, over 3807183.65 frames. ], batch size: 46, lr: 4.10e-03, grad_scale: 4.0 +2023-04-02 17:26:55,951 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 17:27:07,080 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 17:27:32,541 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 17:27:37,090 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.618e+02 4.964e+02 6.321e+02 8.056e+02 1.479e+03, threshold=1.264e+03, percent-clipped=3.0 +2023-04-02 17:27:46,172 INFO [train.py:903] (0/4) Epoch 20, batch 5200, loss[loss=0.2373, simple_loss=0.3134, pruned_loss=0.08062, over 19666.00 frames. ], tot_loss[loss=0.2122, simple_loss=0.2915, pruned_loss=0.06644, over 3812548.70 frames. ], batch size: 60, lr: 4.10e-03, grad_scale: 8.0 +2023-04-02 17:28:00,843 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 17:28:33,739 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.44 vs. limit=2.0 +2023-04-02 17:28:46,967 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 17:28:49,186 INFO [train.py:903] (0/4) Epoch 20, batch 5250, loss[loss=0.1932, simple_loss=0.2669, pruned_loss=0.05978, over 19403.00 frames. ], tot_loss[loss=0.2123, simple_loss=0.2917, pruned_loss=0.06646, over 3804643.90 frames. ], batch size: 48, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:29:42,796 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.326e+02 4.845e+02 5.899e+02 8.450e+02 1.811e+03, threshold=1.180e+03, percent-clipped=4.0 +2023-04-02 17:29:51,999 INFO [train.py:903] (0/4) Epoch 20, batch 5300, loss[loss=0.1756, simple_loss=0.2489, pruned_loss=0.05121, over 19764.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2921, pruned_loss=0.06656, over 3805146.61 frames. ], batch size: 45, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:29:59,059 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=135037.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:30:04,893 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=135042.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:30:11,576 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 17:30:24,893 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4694, 2.5186, 2.6482, 3.1525, 2.5940, 3.0282, 2.7264, 2.5260], + device='cuda:0'), covar=tensor([0.3243, 0.2893, 0.1422, 0.1971, 0.3161, 0.1545, 0.3314, 0.2358], + device='cuda:0'), in_proj_covar=tensor([0.0885, 0.0947, 0.0708, 0.0928, 0.0867, 0.0800, 0.0837, 0.0773], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 17:30:30,726 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=135062.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:30:36,496 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=135067.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:30:54,695 INFO [train.py:903] (0/4) Epoch 20, batch 5350, loss[loss=0.2538, simple_loss=0.3247, pruned_loss=0.09142, over 19297.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2901, pruned_loss=0.06567, over 3813560.87 frames. ], batch size: 66, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:31:29,720 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 17:31:45,352 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9283, 1.7574, 1.5527, 1.9525, 1.6911, 1.6691, 1.5853, 1.8208], + device='cuda:0'), covar=tensor([0.1087, 0.1400, 0.1490, 0.0957, 0.1331, 0.0569, 0.1363, 0.0755], + device='cuda:0'), in_proj_covar=tensor([0.0262, 0.0350, 0.0304, 0.0247, 0.0296, 0.0247, 0.0302, 0.0250], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 17:31:48,520 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.364e+02 5.248e+02 6.375e+02 8.534e+02 1.946e+03, threshold=1.275e+03, percent-clipped=9.0 +2023-04-02 17:31:57,760 INFO [train.py:903] (0/4) Epoch 20, batch 5400, loss[loss=0.168, simple_loss=0.2471, pruned_loss=0.04443, over 19487.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2897, pruned_loss=0.06552, over 3822787.58 frames. ], batch size: 49, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:33:00,878 INFO [train.py:903] (0/4) Epoch 20, batch 5450, loss[loss=0.2288, simple_loss=0.2948, pruned_loss=0.0814, over 19584.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2894, pruned_loss=0.0654, over 3829809.95 frames. ], batch size: 52, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:33:09,002 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=135189.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:33:36,826 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2871, 1.3381, 1.7603, 1.3486, 2.7579, 3.5793, 3.3410, 3.7550], + device='cuda:0'), covar=tensor([0.1599, 0.3591, 0.3141, 0.2337, 0.0675, 0.0221, 0.0200, 0.0276], + device='cuda:0'), in_proj_covar=tensor([0.0265, 0.0316, 0.0344, 0.0262, 0.0237, 0.0183, 0.0213, 0.0249], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 17:33:52,673 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=135224.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:33:54,891 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.649e+02 5.255e+02 6.010e+02 7.558e+02 1.824e+03, threshold=1.202e+03, percent-clipped=3.0 +2023-04-02 17:34:03,095 INFO [train.py:903] (0/4) Epoch 20, batch 5500, loss[loss=0.2445, simple_loss=0.3197, pruned_loss=0.08466, over 19672.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2899, pruned_loss=0.06585, over 3833790.16 frames. ], batch size: 58, lr: 4.09e-03, grad_scale: 4.0 +2023-04-02 17:34:29,182 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 17:35:05,523 INFO [train.py:903] (0/4) Epoch 20, batch 5550, loss[loss=0.2211, simple_loss=0.2847, pruned_loss=0.07875, over 19780.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2895, pruned_loss=0.06567, over 3836837.91 frames. ], batch size: 48, lr: 4.09e-03, grad_scale: 4.0 +2023-04-02 17:35:13,923 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 17:35:47,243 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8315, 4.4236, 2.8033, 3.9065, 0.7405, 4.3299, 4.2005, 4.3328], + device='cuda:0'), covar=tensor([0.0562, 0.0896, 0.1781, 0.0854, 0.4389, 0.0623, 0.0863, 0.0987], + device='cuda:0'), in_proj_covar=tensor([0.0492, 0.0401, 0.0488, 0.0344, 0.0401, 0.0424, 0.0419, 0.0454], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 17:36:00,656 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.142e+02 4.884e+02 6.418e+02 8.039e+02 2.322e+03, threshold=1.284e+03, percent-clipped=8.0 +2023-04-02 17:36:03,999 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 17:36:07,361 INFO [train.py:903] (0/4) Epoch 20, batch 5600, loss[loss=0.2147, simple_loss=0.2812, pruned_loss=0.07411, over 19370.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2898, pruned_loss=0.06582, over 3837853.27 frames. ], batch size: 47, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:36:18,135 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=135339.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:36:29,866 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4896, 1.5455, 1.8384, 1.7135, 2.7063, 2.3426, 2.8914, 1.3169], + device='cuda:0'), covar=tensor([0.2319, 0.4168, 0.2630, 0.1858, 0.1357, 0.1982, 0.1229, 0.4193], + device='cuda:0'), in_proj_covar=tensor([0.0526, 0.0633, 0.0697, 0.0477, 0.0616, 0.0523, 0.0657, 0.0543], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 17:37:11,251 INFO [train.py:903] (0/4) Epoch 20, batch 5650, loss[loss=0.1987, simple_loss=0.2687, pruned_loss=0.0643, over 19010.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2892, pruned_loss=0.06541, over 3828172.45 frames. ], batch size: 42, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:38:01,349 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 17:38:05,709 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.932e+02 4.892e+02 5.590e+02 6.975e+02 1.698e+03, threshold=1.118e+03, percent-clipped=3.0 +2023-04-02 17:38:12,540 INFO [train.py:903] (0/4) Epoch 20, batch 5700, loss[loss=0.1941, simple_loss=0.2668, pruned_loss=0.06071, over 19303.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2893, pruned_loss=0.06553, over 3835411.19 frames. ], batch size: 44, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:38:47,470 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6734, 1.4020, 1.5032, 2.1986, 1.5954, 1.9509, 2.0012, 1.7136], + device='cuda:0'), covar=tensor([0.0950, 0.1133, 0.1070, 0.0810, 0.0990, 0.0822, 0.1019, 0.0823], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0222, 0.0225, 0.0243, 0.0227, 0.0211, 0.0186, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 17:38:48,671 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9008, 3.3409, 1.9059, 1.6405, 3.0349, 1.5078, 1.2835, 2.3057], + device='cuda:0'), covar=tensor([0.1336, 0.0509, 0.0952, 0.1046, 0.0480, 0.1294, 0.1052, 0.0627], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0311, 0.0331, 0.0257, 0.0244, 0.0333, 0.0289, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 17:39:14,106 INFO [train.py:903] (0/4) Epoch 20, batch 5750, loss[loss=0.2576, simple_loss=0.3347, pruned_loss=0.09023, over 19665.00 frames. ], tot_loss[loss=0.2122, simple_loss=0.2912, pruned_loss=0.06661, over 3825072.67 frames. ], batch size: 60, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:39:17,235 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 17:39:25,436 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 17:39:31,255 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 17:40:10,278 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.984e+02 5.062e+02 5.931e+02 7.729e+02 1.708e+03, threshold=1.186e+03, percent-clipped=6.0 +2023-04-02 17:40:18,214 INFO [train.py:903] (0/4) Epoch 20, batch 5800, loss[loss=0.2135, simple_loss=0.2975, pruned_loss=0.06473, over 19342.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2907, pruned_loss=0.0662, over 3819455.61 frames. ], batch size: 66, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:40:19,513 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=135533.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:40:21,363 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-02 17:40:56,842 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 17:41:20,579 INFO [train.py:903] (0/4) Epoch 20, batch 5850, loss[loss=0.2435, simple_loss=0.318, pruned_loss=0.08447, over 18139.00 frames. ], tot_loss[loss=0.2126, simple_loss=0.2922, pruned_loss=0.06654, over 3825731.26 frames. ], batch size: 83, lr: 4.09e-03, grad_scale: 8.0 +2023-04-02 17:41:35,346 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=135595.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:42:03,224 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-04-02 17:42:08,892 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=135620.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:42:15,863 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.765e+02 5.164e+02 6.346e+02 8.557e+02 2.066e+03, threshold=1.269e+03, percent-clipped=9.0 +2023-04-02 17:42:23,021 INFO [train.py:903] (0/4) Epoch 20, batch 5900, loss[loss=0.2004, simple_loss=0.2843, pruned_loss=0.05826, over 19539.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2911, pruned_loss=0.06605, over 3832748.97 frames. ], batch size: 56, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:42:25,417 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 17:42:42,980 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=135648.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:42:43,044 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=135648.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:42:46,308 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 17:43:19,396 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6827, 1.4924, 1.5626, 1.9160, 1.5015, 1.8019, 1.7866, 1.6529], + device='cuda:0'), covar=tensor([0.0750, 0.0907, 0.0931, 0.0594, 0.0752, 0.0738, 0.0801, 0.0683], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0223, 0.0226, 0.0244, 0.0227, 0.0212, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 17:43:24,699 INFO [train.py:903] (0/4) Epoch 20, batch 5950, loss[loss=0.2079, simple_loss=0.2898, pruned_loss=0.06294, over 19776.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2908, pruned_loss=0.06653, over 3824438.88 frames. ], batch size: 56, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:44:00,309 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2622, 2.2888, 2.5359, 2.9954, 2.2199, 2.8613, 2.6379, 2.2934], + device='cuda:0'), covar=tensor([0.4109, 0.3893, 0.1719, 0.2544, 0.4340, 0.2063, 0.4370, 0.3217], + device='cuda:0'), in_proj_covar=tensor([0.0881, 0.0945, 0.0703, 0.0925, 0.0863, 0.0794, 0.0831, 0.0770], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 17:44:19,051 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.081e+02 5.126e+02 6.708e+02 1.004e+03 2.382e+03, threshold=1.342e+03, percent-clipped=11.0 +2023-04-02 17:44:19,670 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-02 17:44:27,256 INFO [train.py:903] (0/4) Epoch 20, batch 6000, loss[loss=0.1948, simple_loss=0.2789, pruned_loss=0.05533, over 19526.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2898, pruned_loss=0.06577, over 3830224.13 frames. ], batch size: 54, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:44:27,257 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 17:44:39,928 INFO [train.py:937] (0/4) Epoch 20, validation: loss=0.1697, simple_loss=0.2697, pruned_loss=0.0349, over 944034.00 frames. +2023-04-02 17:44:39,929 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 17:45:41,594 INFO [train.py:903] (0/4) Epoch 20, batch 6050, loss[loss=0.2146, simple_loss=0.3084, pruned_loss=0.0604, over 19368.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2908, pruned_loss=0.06633, over 3824394.83 frames. ], batch size: 70, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:46:36,913 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.594e+02 4.873e+02 5.806e+02 7.519e+02 1.887e+03, threshold=1.161e+03, percent-clipped=2.0 +2023-04-02 17:46:43,929 INFO [train.py:903] (0/4) Epoch 20, batch 6100, loss[loss=0.2071, simple_loss=0.2922, pruned_loss=0.06101, over 19559.00 frames. ], tot_loss[loss=0.2124, simple_loss=0.2912, pruned_loss=0.06684, over 3817792.94 frames. ], batch size: 56, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:47:46,920 INFO [train.py:903] (0/4) Epoch 20, batch 6150, loss[loss=0.2234, simple_loss=0.2838, pruned_loss=0.08155, over 19717.00 frames. ], tot_loss[loss=0.2134, simple_loss=0.2917, pruned_loss=0.06752, over 3806650.31 frames. ], batch size: 51, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:48:15,302 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=135904.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:48:16,099 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 17:48:41,891 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.532e+02 4.701e+02 6.211e+02 7.118e+02 1.417e+03, threshold=1.242e+03, percent-clipped=3.0 +2023-04-02 17:48:45,706 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=135929.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:48:49,668 INFO [train.py:903] (0/4) Epoch 20, batch 6200, loss[loss=0.2057, simple_loss=0.2905, pruned_loss=0.06041, over 19524.00 frames. ], tot_loss[loss=0.2127, simple_loss=0.2915, pruned_loss=0.06701, over 3819308.64 frames. ], batch size: 54, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:49:18,685 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4107, 1.2787, 1.2926, 1.8304, 1.4164, 1.5608, 1.6957, 1.4000], + device='cuda:0'), covar=tensor([0.0942, 0.1032, 0.1139, 0.0658, 0.0872, 0.0841, 0.0883, 0.0791], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0225, 0.0243, 0.0226, 0.0211, 0.0186, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 17:49:52,112 INFO [train.py:903] (0/4) Epoch 20, batch 6250, loss[loss=0.2499, simple_loss=0.3224, pruned_loss=0.08871, over 18072.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.292, pruned_loss=0.06691, over 3822297.24 frames. ], batch size: 83, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:50:04,667 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=135992.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:50:14,613 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-136000.pt +2023-04-02 17:50:15,875 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136000.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:50:24,777 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 17:50:30,680 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136012.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:50:48,348 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.357e+02 5.185e+02 6.758e+02 8.208e+02 2.074e+03, threshold=1.352e+03, percent-clipped=8.0 +2023-04-02 17:50:55,342 INFO [train.py:903] (0/4) Epoch 20, batch 6300, loss[loss=0.1772, simple_loss=0.2592, pruned_loss=0.04758, over 19765.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2923, pruned_loss=0.06701, over 3820046.98 frames. ], batch size: 48, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:51:26,707 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7157, 4.2442, 4.4478, 4.4428, 1.7623, 4.1568, 3.6347, 4.1412], + device='cuda:0'), covar=tensor([0.1714, 0.0856, 0.0614, 0.0724, 0.5857, 0.0915, 0.0760, 0.1164], + device='cuda:0'), in_proj_covar=tensor([0.0770, 0.0724, 0.0927, 0.0810, 0.0820, 0.0685, 0.0557, 0.0856], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 17:51:58,448 INFO [train.py:903] (0/4) Epoch 20, batch 6350, loss[loss=0.1773, simple_loss=0.255, pruned_loss=0.04982, over 19619.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2914, pruned_loss=0.066, over 3833267.40 frames. ], batch size: 50, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:52:29,665 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136107.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:52:53,613 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.408e+02 5.045e+02 5.932e+02 7.156e+02 1.987e+03, threshold=1.186e+03, percent-clipped=3.0 +2023-04-02 17:53:01,256 INFO [train.py:903] (0/4) Epoch 20, batch 6400, loss[loss=0.2805, simple_loss=0.3447, pruned_loss=0.1082, over 13443.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2925, pruned_loss=0.06687, over 3829167.77 frames. ], batch size: 136, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:53:36,330 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136160.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:54:02,560 INFO [train.py:903] (0/4) Epoch 20, batch 6450, loss[loss=0.2559, simple_loss=0.32, pruned_loss=0.09588, over 19087.00 frames. ], tot_loss[loss=0.2131, simple_loss=0.2927, pruned_loss=0.06673, over 3836768.68 frames. ], batch size: 69, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:54:51,293 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 17:54:59,003 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.358e+02 5.012e+02 6.099e+02 8.089e+02 3.011e+03, threshold=1.220e+03, percent-clipped=7.0 +2023-04-02 17:55:07,145 INFO [train.py:903] (0/4) Epoch 20, batch 6500, loss[loss=0.1922, simple_loss=0.2805, pruned_loss=0.05192, over 19672.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2904, pruned_loss=0.06526, over 3847421.03 frames. ], batch size: 53, lr: 4.08e-03, grad_scale: 8.0 +2023-04-02 17:55:12,966 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 17:55:44,587 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.51 vs. limit=2.0 +2023-04-02 17:56:09,970 INFO [train.py:903] (0/4) Epoch 20, batch 6550, loss[loss=0.2222, simple_loss=0.3052, pruned_loss=0.06964, over 19689.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2903, pruned_loss=0.06525, over 3832348.91 frames. ], batch size: 59, lr: 4.07e-03, grad_scale: 8.0 +2023-04-02 17:57:06,361 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.205e+02 5.319e+02 6.594e+02 8.030e+02 1.579e+03, threshold=1.319e+03, percent-clipped=2.0 +2023-04-02 17:57:14,431 INFO [train.py:903] (0/4) Epoch 20, batch 6600, loss[loss=0.245, simple_loss=0.33, pruned_loss=0.07997, over 19304.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2899, pruned_loss=0.06516, over 3829235.55 frames. ], batch size: 66, lr: 4.07e-03, grad_scale: 8.0 +2023-04-02 17:57:28,427 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=136344.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:57:41,115 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136353.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:57:45,647 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=136356.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:57:53,909 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136363.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:58:16,828 INFO [train.py:903] (0/4) Epoch 20, batch 6650, loss[loss=0.2549, simple_loss=0.3272, pruned_loss=0.09134, over 19648.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2904, pruned_loss=0.06558, over 3824799.76 frames. ], batch size: 58, lr: 4.07e-03, grad_scale: 8.0 +2023-04-02 17:58:26,135 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=136388.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 17:58:56,601 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.32 vs. limit=2.0 +2023-04-02 17:59:14,408 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.147e+02 4.763e+02 5.879e+02 7.861e+02 2.647e+03, threshold=1.176e+03, percent-clipped=6.0 +2023-04-02 17:59:22,069 INFO [train.py:903] (0/4) Epoch 20, batch 6700, loss[loss=0.1821, simple_loss=0.2625, pruned_loss=0.05088, over 19361.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2901, pruned_loss=0.06573, over 3825143.77 frames. ], batch size: 47, lr: 4.07e-03, grad_scale: 8.0 +2023-04-02 17:59:55,242 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136459.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:00:08,586 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136471.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:00:20,886 INFO [train.py:903] (0/4) Epoch 20, batch 6750, loss[loss=0.2156, simple_loss=0.2999, pruned_loss=0.06562, over 19597.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.2893, pruned_loss=0.06582, over 3826942.09 frames. ], batch size: 61, lr: 4.07e-03, grad_scale: 8.0 +2023-04-02 18:00:45,714 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=136504.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:01:11,793 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.251e+02 5.547e+02 6.932e+02 9.039e+02 1.788e+03, threshold=1.386e+03, percent-clipped=9.0 +2023-04-02 18:01:19,069 INFO [train.py:903] (0/4) Epoch 20, batch 6800, loss[loss=0.2557, simple_loss=0.3332, pruned_loss=0.08908, over 19317.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2897, pruned_loss=0.06629, over 3808722.95 frames. ], batch size: 66, lr: 4.07e-03, grad_scale: 8.0 +2023-04-02 18:01:48,959 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-20.pt +2023-04-02 18:02:04,538 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 18:02:04,990 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 18:02:07,961 INFO [train.py:903] (0/4) Epoch 21, batch 0, loss[loss=0.2156, simple_loss=0.2909, pruned_loss=0.07016, over 19734.00 frames. ], tot_loss[loss=0.2156, simple_loss=0.2909, pruned_loss=0.07016, over 19734.00 frames. ], batch size: 45, lr: 3.97e-03, grad_scale: 8.0 +2023-04-02 18:02:07,961 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 18:02:18,721 INFO [train.py:937] (0/4) Epoch 21, validation: loss=0.1691, simple_loss=0.2696, pruned_loss=0.03427, over 944034.00 frames. +2023-04-02 18:02:18,722 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18677MB +2023-04-02 18:02:30,980 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 18:03:20,706 INFO [train.py:903] (0/4) Epoch 21, batch 50, loss[loss=0.2389, simple_loss=0.3158, pruned_loss=0.08103, over 19655.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2883, pruned_loss=0.06347, over 865964.31 frames. ], batch size: 55, lr: 3.97e-03, grad_scale: 4.0 +2023-04-02 18:03:33,306 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136619.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:03:43,288 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.449e+02 4.947e+02 6.173e+02 6.953e+02 1.295e+03, threshold=1.235e+03, percent-clipped=0.0 +2023-04-02 18:03:54,604 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 18:04:02,463 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.00 vs. limit=5.0 +2023-04-02 18:04:22,856 INFO [train.py:903] (0/4) Epoch 21, batch 100, loss[loss=0.1831, simple_loss=0.2682, pruned_loss=0.04901, over 19662.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2888, pruned_loss=0.06449, over 1522623.57 frames. ], batch size: 53, lr: 3.97e-03, grad_scale: 4.0 +2023-04-02 18:04:25,205 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136661.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:04:34,151 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 18:05:09,044 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=136697.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:05:25,009 INFO [train.py:903] (0/4) Epoch 21, batch 150, loss[loss=0.2158, simple_loss=0.2882, pruned_loss=0.07173, over 19698.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2879, pruned_loss=0.06499, over 2037311.72 frames. ], batch size: 53, lr: 3.97e-03, grad_scale: 4.0 +2023-04-02 18:05:31,955 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136715.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:05:45,589 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.079e+02 5.031e+02 5.954e+02 7.665e+02 1.668e+03, threshold=1.191e+03, percent-clipped=3.0 +2023-04-02 18:05:45,976 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136727.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:06:02,666 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=136740.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:06:03,843 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7791, 1.5757, 1.5306, 2.3208, 1.8909, 2.1263, 2.1197, 1.7693], + device='cuda:0'), covar=tensor([0.0826, 0.0933, 0.1003, 0.0694, 0.0764, 0.0720, 0.0881, 0.0703], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0222, 0.0225, 0.0242, 0.0226, 0.0212, 0.0187, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 18:06:18,602 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=136752.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:06:25,311 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 18:06:27,565 INFO [train.py:903] (0/4) Epoch 21, batch 200, loss[loss=0.243, simple_loss=0.3207, pruned_loss=0.08264, over 18342.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2885, pruned_loss=0.0649, over 2441580.84 frames. ], batch size: 84, lr: 3.97e-03, grad_scale: 4.0 +2023-04-02 18:06:29,103 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6286, 1.5078, 1.5296, 2.0085, 1.5587, 1.8702, 1.7926, 1.6069], + device='cuda:0'), covar=tensor([0.0832, 0.0997, 0.0981, 0.0702, 0.0822, 0.0750, 0.0877, 0.0722], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0223, 0.0226, 0.0243, 0.0226, 0.0213, 0.0188, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 18:06:35,855 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7806, 4.3339, 2.7929, 3.8436, 0.8658, 4.2324, 4.1301, 4.3453], + device='cuda:0'), covar=tensor([0.0579, 0.1055, 0.1954, 0.0839, 0.4313, 0.0730, 0.0947, 0.1026], + device='cuda:0'), in_proj_covar=tensor([0.0496, 0.0403, 0.0491, 0.0343, 0.0403, 0.0426, 0.0422, 0.0458], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 18:06:43,862 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5790, 1.3689, 1.3951, 2.1731, 1.7679, 1.9076, 1.8285, 1.5326], + device='cuda:0'), covar=tensor([0.0897, 0.1074, 0.1077, 0.0736, 0.0797, 0.0759, 0.0927, 0.0789], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0222, 0.0226, 0.0243, 0.0226, 0.0213, 0.0188, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 18:07:12,680 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-02 18:07:29,898 INFO [train.py:903] (0/4) Epoch 21, batch 250, loss[loss=0.2208, simple_loss=0.3035, pruned_loss=0.06904, over 18138.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2898, pruned_loss=0.06555, over 2752423.09 frames. ], batch size: 84, lr: 3.97e-03, grad_scale: 4.0 +2023-04-02 18:07:32,575 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=136812.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:07:47,462 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=136824.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:07:52,059 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.077e+02 4.815e+02 6.209e+02 8.068e+02 1.278e+03, threshold=1.242e+03, percent-clipped=1.0 +2023-04-02 18:08:31,137 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4961, 2.3067, 1.8240, 1.4585, 2.1633, 1.4263, 1.3862, 2.0568], + device='cuda:0'), covar=tensor([0.0997, 0.0804, 0.0955, 0.0903, 0.0481, 0.1217, 0.0773, 0.0471], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0314, 0.0338, 0.0260, 0.0247, 0.0336, 0.0292, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 18:08:33,037 INFO [train.py:903] (0/4) Epoch 21, batch 300, loss[loss=0.218, simple_loss=0.3065, pruned_loss=0.06472, over 19660.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.29, pruned_loss=0.06554, over 2986279.27 frames. ], batch size: 55, lr: 3.97e-03, grad_scale: 4.0 +2023-04-02 18:08:53,704 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=136875.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:09:23,331 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=136900.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:09:36,406 INFO [train.py:903] (0/4) Epoch 21, batch 350, loss[loss=0.2335, simple_loss=0.3151, pruned_loss=0.07601, over 19618.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2897, pruned_loss=0.06571, over 3175133.29 frames. ], batch size: 61, lr: 3.97e-03, grad_scale: 4.0 +2023-04-02 18:09:38,661 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 18:09:56,978 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.395e+02 4.835e+02 5.943e+02 7.281e+02 1.741e+03, threshold=1.189e+03, percent-clipped=3.0 +2023-04-02 18:10:39,349 INFO [train.py:903] (0/4) Epoch 21, batch 400, loss[loss=0.1924, simple_loss=0.2647, pruned_loss=0.06003, over 19737.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.291, pruned_loss=0.06633, over 3318465.11 frames. ], batch size: 46, lr: 3.97e-03, grad_scale: 8.0 +2023-04-02 18:11:02,227 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.42 vs. limit=2.0 +2023-04-02 18:11:35,373 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=137005.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:11:41,217 INFO [train.py:903] (0/4) Epoch 21, batch 450, loss[loss=0.189, simple_loss=0.2666, pruned_loss=0.05572, over 19408.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2898, pruned_loss=0.06573, over 3429743.78 frames. ], batch size: 48, lr: 3.96e-03, grad_scale: 4.0 +2023-04-02 18:11:59,603 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6680, 3.0215, 3.1849, 3.2612, 1.3506, 2.9841, 2.6526, 2.7546], + device='cuda:0'), covar=tensor([0.2800, 0.1751, 0.1432, 0.1719, 0.7304, 0.2351, 0.1550, 0.2572], + device='cuda:0'), in_proj_covar=tensor([0.0767, 0.0719, 0.0928, 0.0817, 0.0819, 0.0687, 0.0560, 0.0860], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 18:12:00,846 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4391, 1.5014, 1.7831, 1.6970, 2.6732, 2.3084, 2.8299, 1.3115], + device='cuda:0'), covar=tensor([0.2430, 0.4142, 0.2657, 0.1903, 0.1606, 0.2045, 0.1526, 0.4132], + device='cuda:0'), in_proj_covar=tensor([0.0526, 0.0634, 0.0700, 0.0479, 0.0618, 0.0525, 0.0657, 0.0540], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 18:12:03,759 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.868e+02 4.613e+02 5.973e+02 7.527e+02 1.521e+03, threshold=1.195e+03, percent-clipped=6.0 +2023-04-02 18:12:12,950 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 18:12:14,061 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 18:12:43,473 INFO [train.py:903] (0/4) Epoch 21, batch 500, loss[loss=0.2254, simple_loss=0.2865, pruned_loss=0.08219, over 19765.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2884, pruned_loss=0.06499, over 3528121.66 frames. ], batch size: 45, lr: 3.96e-03, grad_scale: 4.0 +2023-04-02 18:12:53,144 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=137068.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:13:25,140 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=137093.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:13:43,701 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.2386, 5.2708, 6.0985, 6.1214, 2.1717, 5.7422, 4.9297, 5.7548], + device='cuda:0'), covar=tensor([0.1569, 0.0694, 0.0456, 0.0502, 0.5510, 0.0686, 0.0553, 0.0968], + device='cuda:0'), in_proj_covar=tensor([0.0769, 0.0720, 0.0929, 0.0817, 0.0820, 0.0688, 0.0560, 0.0860], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 18:13:45,883 INFO [train.py:903] (0/4) Epoch 21, batch 550, loss[loss=0.2026, simple_loss=0.2821, pruned_loss=0.06154, over 19581.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2872, pruned_loss=0.06432, over 3609433.90 frames. ], batch size: 52, lr: 3.96e-03, grad_scale: 4.0 +2023-04-02 18:13:59,942 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=137120.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:14:09,847 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.012e+02 5.365e+02 6.823e+02 8.347e+02 2.113e+03, threshold=1.365e+03, percent-clipped=7.0 +2023-04-02 18:14:33,015 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5795, 1.1317, 1.4147, 1.1772, 2.2069, 0.9790, 2.1031, 2.4464], + device='cuda:0'), covar=tensor([0.0704, 0.2856, 0.2765, 0.1837, 0.0889, 0.2204, 0.1058, 0.0501], + device='cuda:0'), in_proj_covar=tensor([0.0404, 0.0361, 0.0382, 0.0344, 0.0372, 0.0347, 0.0373, 0.0398], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 18:14:48,996 INFO [train.py:903] (0/4) Epoch 21, batch 600, loss[loss=0.2232, simple_loss=0.2968, pruned_loss=0.07483, over 19743.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2882, pruned_loss=0.0644, over 3666779.36 frames. ], batch size: 51, lr: 3.96e-03, grad_scale: 4.0 +2023-04-02 18:15:00,534 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=137168.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:15:29,956 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 18:15:47,666 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0854, 1.7699, 1.8257, 2.8272, 1.9893, 2.2680, 2.2824, 2.0699], + device='cuda:0'), covar=tensor([0.0890, 0.1116, 0.1107, 0.0856, 0.0996, 0.0916, 0.1019, 0.0808], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0220, 0.0224, 0.0241, 0.0224, 0.0210, 0.0186, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 18:15:53,982 INFO [train.py:903] (0/4) Epoch 21, batch 650, loss[loss=0.2057, simple_loss=0.2764, pruned_loss=0.06749, over 19365.00 frames. ], tot_loss[loss=0.2097, simple_loss=0.2892, pruned_loss=0.06511, over 3697931.10 frames. ], batch size: 47, lr: 3.96e-03, grad_scale: 4.0 +2023-04-02 18:16:16,601 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.985e+02 4.775e+02 5.968e+02 8.002e+02 1.696e+03, threshold=1.194e+03, percent-clipped=7.0 +2023-04-02 18:16:56,145 INFO [train.py:903] (0/4) Epoch 21, batch 700, loss[loss=0.201, simple_loss=0.2877, pruned_loss=0.0571, over 19528.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2908, pruned_loss=0.06588, over 3726388.75 frames. ], batch size: 56, lr: 3.96e-03, grad_scale: 4.0 +2023-04-02 18:17:03,358 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2667, 1.8692, 1.8444, 2.1993, 1.9131, 1.8869, 1.7971, 2.2170], + device='cuda:0'), covar=tensor([0.0904, 0.1472, 0.1333, 0.0926, 0.1213, 0.0519, 0.1306, 0.0606], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0354, 0.0309, 0.0249, 0.0298, 0.0251, 0.0307, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 18:17:26,116 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=137283.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:18:00,857 INFO [train.py:903] (0/4) Epoch 21, batch 750, loss[loss=0.196, simple_loss=0.2822, pruned_loss=0.05491, over 19670.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.289, pruned_loss=0.06465, over 3764586.02 frames. ], batch size: 53, lr: 3.96e-03, grad_scale: 4.0 +2023-04-02 18:18:18,164 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7502, 1.8457, 2.0798, 2.3339, 1.7408, 2.2814, 2.1373, 1.9571], + device='cuda:0'), covar=tensor([0.4024, 0.3541, 0.1796, 0.2153, 0.3672, 0.1871, 0.4633, 0.3149], + device='cuda:0'), in_proj_covar=tensor([0.0888, 0.0950, 0.0709, 0.0927, 0.0870, 0.0800, 0.0835, 0.0775], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 18:18:22,899 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.306e+02 4.851e+02 6.236e+02 7.648e+02 2.101e+03, threshold=1.247e+03, percent-clipped=5.0 +2023-04-02 18:18:42,017 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 18:19:01,638 INFO [train.py:903] (0/4) Epoch 21, batch 800, loss[loss=0.2279, simple_loss=0.3103, pruned_loss=0.07277, over 19613.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2898, pruned_loss=0.0655, over 3741175.51 frames. ], batch size: 57, lr: 3.96e-03, grad_scale: 8.0 +2023-04-02 18:19:07,857 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 18:19:22,466 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=137376.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:19:53,492 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=137401.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:20:05,538 INFO [train.py:903] (0/4) Epoch 21, batch 850, loss[loss=0.2083, simple_loss=0.2904, pruned_loss=0.06312, over 19608.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2896, pruned_loss=0.0654, over 3767709.99 frames. ], batch size: 61, lr: 3.96e-03, grad_scale: 8.0 +2023-04-02 18:20:27,118 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.313e+02 5.583e+02 6.647e+02 8.818e+02 2.027e+03, threshold=1.329e+03, percent-clipped=5.0 +2023-04-02 18:20:48,462 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 18:21:06,665 INFO [train.py:903] (0/4) Epoch 21, batch 900, loss[loss=0.2304, simple_loss=0.3077, pruned_loss=0.07661, over 19537.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2908, pruned_loss=0.06614, over 3777557.41 frames. ], batch size: 56, lr: 3.96e-03, grad_scale: 8.0 +2023-04-02 18:21:59,587 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 18:22:08,625 INFO [train.py:903] (0/4) Epoch 21, batch 950, loss[loss=0.2276, simple_loss=0.3032, pruned_loss=0.07601, over 19595.00 frames. ], tot_loss[loss=0.2121, simple_loss=0.2914, pruned_loss=0.06636, over 3788044.97 frames. ], batch size: 52, lr: 3.96e-03, grad_scale: 8.0 +2023-04-02 18:22:31,546 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.353e+02 5.256e+02 6.264e+02 7.895e+02 1.664e+03, threshold=1.253e+03, percent-clipped=2.0 +2023-04-02 18:22:45,756 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=137539.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:23:10,897 INFO [train.py:903] (0/4) Epoch 21, batch 1000, loss[loss=0.1879, simple_loss=0.2628, pruned_loss=0.05648, over 19740.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2906, pruned_loss=0.06627, over 3794194.01 frames. ], batch size: 45, lr: 3.96e-03, grad_scale: 8.0 +2023-04-02 18:23:15,602 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=137564.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:23:55,417 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 18:23:55,871 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 18:24:14,004 INFO [train.py:903] (0/4) Epoch 21, batch 1050, loss[loss=0.2524, simple_loss=0.3185, pruned_loss=0.0931, over 13052.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2911, pruned_loss=0.06642, over 3804807.02 frames. ], batch size: 136, lr: 3.96e-03, grad_scale: 8.0 +2023-04-02 18:24:35,334 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.928e+02 5.318e+02 6.486e+02 8.521e+02 3.216e+03, threshold=1.297e+03, percent-clipped=7.0 +2023-04-02 18:24:36,546 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 18:25:17,802 INFO [train.py:903] (0/4) Epoch 21, batch 1100, loss[loss=0.2275, simple_loss=0.307, pruned_loss=0.074, over 19369.00 frames. ], tot_loss[loss=0.2122, simple_loss=0.2912, pruned_loss=0.0666, over 3805246.84 frames. ], batch size: 66, lr: 3.96e-03, grad_scale: 8.0 +2023-04-02 18:26:19,703 INFO [train.py:903] (0/4) Epoch 21, batch 1150, loss[loss=0.1859, simple_loss=0.2785, pruned_loss=0.04667, over 19781.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2909, pruned_loss=0.06595, over 3818994.74 frames. ], batch size: 56, lr: 3.95e-03, grad_scale: 4.0 +2023-04-02 18:26:43,796 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.088e+02 5.127e+02 6.263e+02 7.580e+02 1.245e+03, threshold=1.253e+03, percent-clipped=0.0 +2023-04-02 18:27:22,172 INFO [train.py:903] (0/4) Epoch 21, batch 1200, loss[loss=0.2056, simple_loss=0.2799, pruned_loss=0.06562, over 19754.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.29, pruned_loss=0.06544, over 3828252.19 frames. ], batch size: 51, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:27:46,791 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 18:28:10,900 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-02 18:28:25,533 INFO [train.py:903] (0/4) Epoch 21, batch 1250, loss[loss=0.215, simple_loss=0.2997, pruned_loss=0.06512, over 17561.00 frames. ], tot_loss[loss=0.2096, simple_loss=0.2892, pruned_loss=0.06499, over 3822645.35 frames. ], batch size: 101, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:28:48,825 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.413e+02 5.084e+02 5.991e+02 7.123e+02 1.423e+03, threshold=1.198e+03, percent-clipped=5.0 +2023-04-02 18:29:27,971 INFO [train.py:903] (0/4) Epoch 21, batch 1300, loss[loss=0.2019, simple_loss=0.2706, pruned_loss=0.06655, over 19771.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2895, pruned_loss=0.06556, over 3819913.24 frames. ], batch size: 54, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:29:41,249 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7787, 1.6427, 1.6904, 2.2865, 1.7457, 2.1008, 1.9998, 1.7972], + device='cuda:0'), covar=tensor([0.0838, 0.0945, 0.0984, 0.0751, 0.0869, 0.0743, 0.0919, 0.0736], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0225, 0.0227, 0.0245, 0.0227, 0.0214, 0.0190, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 18:30:19,739 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=5.28 vs. limit=5.0 +2023-04-02 18:30:30,622 INFO [train.py:903] (0/4) Epoch 21, batch 1350, loss[loss=0.1999, simple_loss=0.2785, pruned_loss=0.06071, over 19677.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2892, pruned_loss=0.06544, over 3821554.31 frames. ], batch size: 53, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:30:30,911 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=137910.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:30:34,446 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=137913.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:30:54,796 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.119e+02 5.098e+02 6.340e+02 8.452e+02 2.491e+03, threshold=1.268e+03, percent-clipped=6.0 +2023-04-02 18:31:33,355 INFO [train.py:903] (0/4) Epoch 21, batch 1400, loss[loss=0.1986, simple_loss=0.2827, pruned_loss=0.05729, over 19763.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2896, pruned_loss=0.06546, over 3827032.08 frames. ], batch size: 54, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:32:08,989 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6082, 1.5275, 1.5650, 1.9959, 1.5760, 1.8754, 1.9024, 1.6645], + device='cuda:0'), covar=tensor([0.0858, 0.0939, 0.0990, 0.0749, 0.0823, 0.0749, 0.0824, 0.0729], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0224, 0.0227, 0.0244, 0.0227, 0.0213, 0.0189, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 18:32:23,939 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-138000.pt +2023-04-02 18:32:28,139 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 18:32:37,232 INFO [train.py:903] (0/4) Epoch 21, batch 1450, loss[loss=0.2049, simple_loss=0.2896, pruned_loss=0.06008, over 19511.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2901, pruned_loss=0.06557, over 3821802.96 frames. ], batch size: 56, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:32:41,596 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.31 vs. limit=2.0 +2023-04-02 18:32:45,903 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5089, 1.4797, 1.4884, 1.8479, 1.4600, 1.7074, 1.7095, 1.5479], + device='cuda:0'), covar=tensor([0.0883, 0.0958, 0.1027, 0.0719, 0.0848, 0.0819, 0.0868, 0.0755], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0224, 0.0228, 0.0244, 0.0227, 0.0213, 0.0189, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 18:33:01,289 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.354e+02 4.676e+02 5.543e+02 6.978e+02 2.034e+03, threshold=1.109e+03, percent-clipped=2.0 +2023-04-02 18:33:39,045 INFO [train.py:903] (0/4) Epoch 21, batch 1500, loss[loss=0.2094, simple_loss=0.2918, pruned_loss=0.06346, over 19485.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.2894, pruned_loss=0.06509, over 3828170.08 frames. ], batch size: 64, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:34:17,988 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8919, 1.2004, 1.6466, 0.5842, 1.9799, 2.4074, 2.1042, 2.5570], + device='cuda:0'), covar=tensor([0.1601, 0.3737, 0.3046, 0.2638, 0.0614, 0.0298, 0.0347, 0.0365], + device='cuda:0'), in_proj_covar=tensor([0.0269, 0.0320, 0.0349, 0.0264, 0.0241, 0.0185, 0.0215, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 18:34:42,049 INFO [train.py:903] (0/4) Epoch 21, batch 1550, loss[loss=0.2213, simple_loss=0.293, pruned_loss=0.07476, over 19687.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2891, pruned_loss=0.06534, over 3833907.02 frames. ], batch size: 53, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:34:43,703 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9936, 2.1125, 2.3242, 2.7011, 2.0559, 2.6711, 2.3665, 2.1090], + device='cuda:0'), covar=tensor([0.4321, 0.3881, 0.1865, 0.2388, 0.4170, 0.2027, 0.4717, 0.3363], + device='cuda:0'), in_proj_covar=tensor([0.0888, 0.0952, 0.0711, 0.0927, 0.0869, 0.0800, 0.0833, 0.0777], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 18:34:56,473 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9120, 1.6631, 1.8855, 1.4227, 4.4274, 1.0874, 2.5032, 4.8716], + device='cuda:0'), covar=tensor([0.0438, 0.2680, 0.2702, 0.2185, 0.0733, 0.2752, 0.1498, 0.0166], + device='cuda:0'), in_proj_covar=tensor([0.0408, 0.0365, 0.0385, 0.0346, 0.0374, 0.0349, 0.0377, 0.0400], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 18:35:05,478 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.935e+02 5.280e+02 6.228e+02 7.726e+02 2.313e+03, threshold=1.246e+03, percent-clipped=5.0 +2023-04-02 18:35:38,370 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9977, 5.0587, 5.8099, 5.8207, 2.1671, 5.5359, 4.6627, 5.4479], + device='cuda:0'), covar=tensor([0.1625, 0.0845, 0.0535, 0.0612, 0.5762, 0.0719, 0.0578, 0.1226], + device='cuda:0'), in_proj_covar=tensor([0.0770, 0.0721, 0.0929, 0.0816, 0.0817, 0.0687, 0.0561, 0.0862], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 18:35:44,862 INFO [train.py:903] (0/4) Epoch 21, batch 1600, loss[loss=0.2063, simple_loss=0.2719, pruned_loss=0.07029, over 19719.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.2889, pruned_loss=0.06495, over 3823579.14 frames. ], batch size: 46, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:36:07,164 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 18:36:48,268 INFO [train.py:903] (0/4) Epoch 21, batch 1650, loss[loss=0.2399, simple_loss=0.3125, pruned_loss=0.0837, over 13180.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2883, pruned_loss=0.06485, over 3795242.71 frames. ], batch size: 136, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:36:50,165 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-04-02 18:37:12,972 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.826e+02 4.591e+02 5.832e+02 7.172e+02 1.632e+03, threshold=1.166e+03, percent-clipped=1.0 +2023-04-02 18:37:43,137 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=138254.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:37:46,572 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=138257.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:37:49,815 INFO [train.py:903] (0/4) Epoch 21, batch 1700, loss[loss=0.2173, simple_loss=0.2937, pruned_loss=0.0705, over 19655.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2897, pruned_loss=0.06558, over 3800920.01 frames. ], batch size: 60, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:38:24,230 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138288.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:38:28,573 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 18:38:53,713 INFO [train.py:903] (0/4) Epoch 21, batch 1750, loss[loss=0.209, simple_loss=0.2909, pruned_loss=0.06351, over 19661.00 frames. ], tot_loss[loss=0.2097, simple_loss=0.2893, pruned_loss=0.0651, over 3822386.09 frames. ], batch size: 58, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:39:16,057 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.511e+02 4.819e+02 5.989e+02 8.047e+02 2.111e+03, threshold=1.198e+03, percent-clipped=8.0 +2023-04-02 18:39:30,928 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138340.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:39:55,290 INFO [train.py:903] (0/4) Epoch 21, batch 1800, loss[loss=0.2176, simple_loss=0.3036, pruned_loss=0.06579, over 19697.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2888, pruned_loss=0.0645, over 3824165.00 frames. ], batch size: 59, lr: 3.95e-03, grad_scale: 8.0 +2023-04-02 18:40:05,943 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=138369.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:40:10,874 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=138372.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:40:54,328 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 18:40:57,827 INFO [train.py:903] (0/4) Epoch 21, batch 1850, loss[loss=0.1962, simple_loss=0.2706, pruned_loss=0.06096, over 19399.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2891, pruned_loss=0.06461, over 3821179.96 frames. ], batch size: 48, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:41:22,804 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.428e+02 4.784e+02 5.677e+02 7.800e+02 1.333e+03, threshold=1.135e+03, percent-clipped=2.0 +2023-04-02 18:41:34,213 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 18:42:01,833 INFO [train.py:903] (0/4) Epoch 21, batch 1900, loss[loss=0.232, simple_loss=0.3126, pruned_loss=0.07568, over 17468.00 frames. ], tot_loss[loss=0.2117, simple_loss=0.2912, pruned_loss=0.06613, over 3792434.69 frames. ], batch size: 101, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:42:18,834 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 18:42:19,482 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 18:42:23,577 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 18:42:48,358 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 18:43:04,463 INFO [train.py:903] (0/4) Epoch 21, batch 1950, loss[loss=0.2119, simple_loss=0.3064, pruned_loss=0.0587, over 19532.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2907, pruned_loss=0.06585, over 3806004.90 frames. ], batch size: 54, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:43:27,758 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.334e+02 4.783e+02 6.007e+02 7.405e+02 3.008e+03, threshold=1.201e+03, percent-clipped=4.0 +2023-04-02 18:44:06,915 INFO [train.py:903] (0/4) Epoch 21, batch 2000, loss[loss=0.2116, simple_loss=0.303, pruned_loss=0.06013, over 18110.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2906, pruned_loss=0.06567, over 3801331.24 frames. ], batch size: 83, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:44:14,587 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-02 18:45:03,885 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 18:45:08,463 INFO [train.py:903] (0/4) Epoch 21, batch 2050, loss[loss=0.1665, simple_loss=0.2469, pruned_loss=0.04304, over 19625.00 frames. ], tot_loss[loss=0.2096, simple_loss=0.2897, pruned_loss=0.0648, over 3808021.79 frames. ], batch size: 50, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:45:22,125 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 18:45:23,301 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 18:45:28,161 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=138625.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:45:32,734 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=138628.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:45:33,541 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.887e+02 4.818e+02 6.151e+02 8.119e+02 1.355e+03, threshold=1.230e+03, percent-clipped=5.0 +2023-04-02 18:45:38,227 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=138632.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:45:45,212 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 18:46:00,056 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=138650.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:46:03,534 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=138653.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:46:12,220 INFO [train.py:903] (0/4) Epoch 21, batch 2100, loss[loss=0.2052, simple_loss=0.2833, pruned_loss=0.06353, over 19610.00 frames. ], tot_loss[loss=0.2096, simple_loss=0.2894, pruned_loss=0.06487, over 3799805.07 frames. ], batch size: 50, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:46:13,822 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9765, 1.3421, 1.0578, 0.9659, 1.1474, 0.9549, 1.0087, 1.2632], + device='cuda:0'), covar=tensor([0.0561, 0.0845, 0.1138, 0.0762, 0.0568, 0.1327, 0.0544, 0.0476], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0313, 0.0334, 0.0260, 0.0245, 0.0336, 0.0290, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 18:46:39,348 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 18:46:41,754 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=138684.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:47:00,945 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138699.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:47:02,059 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 18:47:06,941 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138704.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:47:14,541 INFO [train.py:903] (0/4) Epoch 21, batch 2150, loss[loss=0.2234, simple_loss=0.302, pruned_loss=0.0724, over 19614.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.29, pruned_loss=0.06548, over 3806054.42 frames. ], batch size: 50, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:47:21,563 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138715.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:47:37,629 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.976e+02 5.033e+02 5.895e+02 7.227e+02 1.459e+03, threshold=1.179e+03, percent-clipped=3.0 +2023-04-02 18:48:00,969 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=138747.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:48:03,228 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7708, 3.2387, 3.2949, 3.3215, 1.3675, 3.1626, 2.7529, 3.0407], + device='cuda:0'), covar=tensor([0.1859, 0.1272, 0.0875, 0.0960, 0.5439, 0.1207, 0.0921, 0.1433], + device='cuda:0'), in_proj_covar=tensor([0.0781, 0.0735, 0.0941, 0.0824, 0.0829, 0.0696, 0.0570, 0.0870], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 18:48:18,134 INFO [train.py:903] (0/4) Epoch 21, batch 2200, loss[loss=0.2457, simple_loss=0.3251, pruned_loss=0.08321, over 19479.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2904, pruned_loss=0.06576, over 3804389.64 frames. ], batch size: 64, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:48:39,537 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4321, 2.1901, 2.2115, 2.4852, 2.2435, 2.0284, 2.1433, 2.4392], + device='cuda:0'), covar=tensor([0.0769, 0.1335, 0.1077, 0.0826, 0.1118, 0.0473, 0.1124, 0.0526], + device='cuda:0'), in_proj_covar=tensor([0.0268, 0.0358, 0.0311, 0.0250, 0.0301, 0.0251, 0.0309, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 18:49:08,075 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=138799.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:49:20,496 INFO [train.py:903] (0/4) Epoch 21, batch 2250, loss[loss=0.2205, simple_loss=0.3184, pruned_loss=0.06136, over 19676.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2914, pruned_loss=0.06567, over 3808698.18 frames. ], batch size: 60, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:49:41,356 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0522, 1.3230, 1.7864, 1.2627, 2.7397, 3.7452, 3.4925, 3.9204], + device='cuda:0'), covar=tensor([0.1673, 0.3676, 0.3144, 0.2374, 0.0614, 0.0179, 0.0192, 0.0247], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0321, 0.0350, 0.0264, 0.0241, 0.0185, 0.0216, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 18:49:44,480 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.352e+02 5.230e+02 6.732e+02 8.271e+02 2.316e+03, threshold=1.346e+03, percent-clipped=7.0 +2023-04-02 18:50:12,140 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6450, 4.0134, 4.4943, 4.5175, 1.8491, 4.2096, 3.5366, 3.8726], + device='cuda:0'), covar=tensor([0.2287, 0.1506, 0.0935, 0.1152, 0.7143, 0.1793, 0.1201, 0.1940], + device='cuda:0'), in_proj_covar=tensor([0.0780, 0.0733, 0.0942, 0.0823, 0.0826, 0.0694, 0.0570, 0.0870], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 18:50:23,832 INFO [train.py:903] (0/4) Epoch 21, batch 2300, loss[loss=0.1971, simple_loss=0.2747, pruned_loss=0.05976, over 19627.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2905, pruned_loss=0.06597, over 3809612.48 frames. ], batch size: 50, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:50:38,385 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 18:51:27,178 INFO [train.py:903] (0/4) Epoch 21, batch 2350, loss[loss=0.2061, simple_loss=0.2964, pruned_loss=0.05793, over 19684.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2906, pruned_loss=0.06595, over 3807153.33 frames. ], batch size: 59, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:51:48,927 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=138927.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:51:50,723 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.927e+02 4.807e+02 6.646e+02 8.268e+02 1.737e+03, threshold=1.329e+03, percent-clipped=3.0 +2023-04-02 18:52:08,314 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 18:52:25,718 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 18:52:31,431 INFO [train.py:903] (0/4) Epoch 21, batch 2400, loss[loss=0.2433, simple_loss=0.3213, pruned_loss=0.08261, over 19621.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2912, pruned_loss=0.06628, over 3804246.41 frames. ], batch size: 50, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:53:26,790 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139003.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:53:34,426 INFO [train.py:903] (0/4) Epoch 21, batch 2450, loss[loss=0.2383, simple_loss=0.3193, pruned_loss=0.07859, over 19097.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2905, pruned_loss=0.06594, over 3799628.11 frames. ], batch size: 69, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:53:58,371 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139028.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:53:59,945 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.106e+02 5.027e+02 6.383e+02 8.090e+02 1.476e+03, threshold=1.277e+03, percent-clipped=1.0 +2023-04-02 18:54:17,224 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=139043.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:54:22,823 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=139048.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:54:31,523 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139055.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:54:37,040 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=139059.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:54:39,177 INFO [train.py:903] (0/4) Epoch 21, batch 2500, loss[loss=0.2437, simple_loss=0.3072, pruned_loss=0.09006, over 19724.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2901, pruned_loss=0.06619, over 3805577.38 frames. ], batch size: 51, lr: 3.94e-03, grad_scale: 8.0 +2023-04-02 18:55:05,336 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139080.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:55:20,465 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.47 vs. limit=2.0 +2023-04-02 18:55:34,451 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.71 vs. limit=5.0 +2023-04-02 18:55:42,984 INFO [train.py:903] (0/4) Epoch 21, batch 2550, loss[loss=0.2153, simple_loss=0.2963, pruned_loss=0.06708, over 19320.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.289, pruned_loss=0.06535, over 3801524.69 frames. ], batch size: 66, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 18:55:53,912 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7157, 1.7120, 1.5935, 1.3432, 1.2775, 1.3837, 0.3418, 0.6842], + device='cuda:0'), covar=tensor([0.0647, 0.0659, 0.0417, 0.0645, 0.1324, 0.0758, 0.1263, 0.1096], + device='cuda:0'), in_proj_covar=tensor([0.0358, 0.0352, 0.0357, 0.0381, 0.0458, 0.0386, 0.0333, 0.0339], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 18:56:06,421 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.656e+02 4.926e+02 6.170e+02 7.459e+02 2.294e+03, threshold=1.234e+03, percent-clipped=3.0 +2023-04-02 18:56:14,904 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2457, 1.3202, 1.7101, 1.2973, 2.6908, 3.7057, 3.4167, 3.9262], + device='cuda:0'), covar=tensor([0.1667, 0.3831, 0.3482, 0.2539, 0.0617, 0.0220, 0.0212, 0.0259], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0322, 0.0351, 0.0264, 0.0242, 0.0186, 0.0217, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 18:56:35,492 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 18:56:43,471 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139158.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:56:45,330 INFO [train.py:903] (0/4) Epoch 21, batch 2600, loss[loss=0.2286, simple_loss=0.3067, pruned_loss=0.07525, over 19502.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2895, pruned_loss=0.06585, over 3801937.35 frames. ], batch size: 64, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 18:56:49,270 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139163.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:57:02,136 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139174.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:57:47,164 INFO [train.py:903] (0/4) Epoch 21, batch 2650, loss[loss=0.2124, simple_loss=0.2868, pruned_loss=0.06897, over 19857.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.2902, pruned_loss=0.0661, over 3803277.69 frames. ], batch size: 52, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 18:58:08,412 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 18:58:13,098 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.645e+02 4.962e+02 5.879e+02 7.846e+02 2.263e+03, threshold=1.176e+03, percent-clipped=6.0 +2023-04-02 18:58:45,336 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139256.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:58:49,505 INFO [train.py:903] (0/4) Epoch 21, batch 2700, loss[loss=0.2134, simple_loss=0.2895, pruned_loss=0.06865, over 19860.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2904, pruned_loss=0.06592, over 3796246.25 frames. ], batch size: 52, lr: 3.93e-03, grad_scale: 4.0 +2023-04-02 18:59:04,241 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=139271.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:59:26,353 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-04-02 18:59:33,178 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139294.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:59:49,897 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139307.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 18:59:52,995 INFO [train.py:903] (0/4) Epoch 21, batch 2750, loss[loss=0.1952, simple_loss=0.2657, pruned_loss=0.06238, over 19336.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2906, pruned_loss=0.06606, over 3807801.76 frames. ], batch size: 44, lr: 3.93e-03, grad_scale: 4.0 +2023-04-02 19:00:18,050 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.927e+02 4.705e+02 5.796e+02 7.407e+02 1.811e+03, threshold=1.159e+03, percent-clipped=2.0 +2023-04-02 19:00:55,535 INFO [train.py:903] (0/4) Epoch 21, batch 2800, loss[loss=0.205, simple_loss=0.289, pruned_loss=0.06047, over 19796.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2905, pruned_loss=0.06566, over 3821840.09 frames. ], batch size: 56, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 19:01:08,569 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 19:01:27,962 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139386.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:01:45,346 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-02 19:01:58,454 INFO [train.py:903] (0/4) Epoch 21, batch 2850, loss[loss=0.2029, simple_loss=0.2776, pruned_loss=0.06416, over 19746.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2899, pruned_loss=0.06534, over 3810153.86 frames. ], batch size: 47, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 19:02:03,716 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139414.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:02:09,510 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139419.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:02:23,960 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.945e+02 5.282e+02 6.184e+02 7.725e+02 1.552e+03, threshold=1.237e+03, percent-clipped=4.0 +2023-04-02 19:02:24,350 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139430.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:02:35,364 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139439.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:02:42,178 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139444.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:02:54,851 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139455.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:02:57,840 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 19:03:00,105 INFO [train.py:903] (0/4) Epoch 21, batch 2900, loss[loss=0.216, simple_loss=0.2961, pruned_loss=0.06794, over 19383.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2901, pruned_loss=0.06528, over 3809016.60 frames. ], batch size: 70, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 19:03:31,305 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7656, 1.2956, 1.5635, 1.6493, 3.2218, 1.2683, 2.4940, 3.6583], + device='cuda:0'), covar=tensor([0.0549, 0.2978, 0.2839, 0.1894, 0.0796, 0.2583, 0.1274, 0.0293], + device='cuda:0'), in_proj_covar=tensor([0.0406, 0.0361, 0.0380, 0.0342, 0.0368, 0.0345, 0.0372, 0.0398], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:04:04,518 INFO [train.py:903] (0/4) Epoch 21, batch 2950, loss[loss=0.2009, simple_loss=0.2876, pruned_loss=0.05713, over 19669.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.288, pruned_loss=0.06439, over 3823620.01 frames. ], batch size: 53, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 19:04:28,835 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.677e+02 4.684e+02 5.947e+02 7.442e+02 1.403e+03, threshold=1.189e+03, percent-clipped=5.0 +2023-04-02 19:05:01,584 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1464, 1.2605, 1.6556, 1.1969, 2.6519, 3.5125, 3.2315, 3.6311], + device='cuda:0'), covar=tensor([0.1663, 0.3788, 0.3380, 0.2420, 0.0587, 0.0192, 0.0205, 0.0266], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0320, 0.0350, 0.0264, 0.0242, 0.0186, 0.0216, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 19:05:06,911 INFO [train.py:903] (0/4) Epoch 21, batch 3000, loss[loss=0.2071, simple_loss=0.2917, pruned_loss=0.06123, over 19701.00 frames. ], tot_loss[loss=0.2093, simple_loss=0.2887, pruned_loss=0.06493, over 3838509.52 frames. ], batch size: 59, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 19:05:06,912 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 19:05:14,028 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8040, 3.5311, 2.6889, 3.2725, 0.8576, 3.5290, 3.2698, 3.5878], + device='cuda:0'), covar=tensor([0.0694, 0.0727, 0.1779, 0.0755, 0.3788, 0.0711, 0.0722, 0.0862], + device='cuda:0'), in_proj_covar=tensor([0.0495, 0.0407, 0.0488, 0.0342, 0.0398, 0.0424, 0.0419, 0.0455], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:05:20,609 INFO [train.py:937] (0/4) Epoch 21, validation: loss=0.1693, simple_loss=0.2693, pruned_loss=0.03465, over 944034.00 frames. +2023-04-02 19:05:20,610 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-02 19:05:24,369 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 19:05:57,511 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139589.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:06:05,905 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.94 vs. limit=2.0 +2023-04-02 19:06:12,376 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=139600.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:06:12,515 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139600.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:06:23,830 INFO [train.py:903] (0/4) Epoch 21, batch 3050, loss[loss=0.2276, simple_loss=0.3042, pruned_loss=0.07554, over 19527.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.2885, pruned_loss=0.06522, over 3825143.67 frames. ], batch size: 54, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 19:06:35,341 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-04-02 19:06:48,012 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.185e+02 5.171e+02 6.119e+02 7.965e+02 1.649e+03, threshold=1.224e+03, percent-clipped=3.0 +2023-04-02 19:06:57,089 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=139638.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:07:03,665 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139642.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:07:14,485 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=139651.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:07:24,505 INFO [train.py:903] (0/4) Epoch 21, batch 3100, loss[loss=0.1761, simple_loss=0.2608, pruned_loss=0.0457, over 19726.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2874, pruned_loss=0.06501, over 3821965.36 frames. ], batch size: 51, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 19:07:34,147 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139667.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:07:43,943 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8184, 1.3466, 1.4856, 1.7711, 3.3799, 1.2285, 2.3705, 3.8478], + device='cuda:0'), covar=tensor([0.0475, 0.2895, 0.2954, 0.1757, 0.0731, 0.2535, 0.1430, 0.0233], + device='cuda:0'), in_proj_covar=tensor([0.0407, 0.0363, 0.0381, 0.0342, 0.0369, 0.0346, 0.0373, 0.0399], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:08:25,945 INFO [train.py:903] (0/4) Epoch 21, batch 3150, loss[loss=0.2573, simple_loss=0.3212, pruned_loss=0.09665, over 13922.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2875, pruned_loss=0.06525, over 3812861.22 frames. ], batch size: 136, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 19:08:32,303 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139715.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:08:51,319 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.442e+02 5.084e+02 6.643e+02 9.166e+02 2.493e+03, threshold=1.329e+03, percent-clipped=12.0 +2023-04-02 19:08:52,525 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 19:09:19,205 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139753.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:09:28,041 INFO [train.py:903] (0/4) Epoch 21, batch 3200, loss[loss=0.206, simple_loss=0.2891, pruned_loss=0.06152, over 19525.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2885, pruned_loss=0.06564, over 3818134.00 frames. ], batch size: 54, lr: 3.93e-03, grad_scale: 8.0 +2023-04-02 19:09:36,911 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=139766.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:10:01,984 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139787.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:10:06,558 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3869, 2.1497, 1.6350, 1.4313, 1.9872, 1.3060, 1.4403, 1.9027], + device='cuda:0'), covar=tensor([0.1042, 0.0759, 0.1085, 0.0835, 0.0520, 0.1252, 0.0686, 0.0446], + device='cuda:0'), in_proj_covar=tensor([0.0298, 0.0309, 0.0332, 0.0258, 0.0241, 0.0332, 0.0286, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:10:27,823 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139807.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:10:30,933 INFO [train.py:903] (0/4) Epoch 21, batch 3250, loss[loss=0.225, simple_loss=0.3038, pruned_loss=0.07306, over 19587.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.289, pruned_loss=0.06547, over 3830180.81 frames. ], batch size: 61, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:10:46,073 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139822.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:10:48,403 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139824.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:10:48,533 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9933, 1.9106, 1.8086, 1.6674, 1.4540, 1.6190, 0.4966, 0.8850], + device='cuda:0'), covar=tensor([0.0597, 0.0638, 0.0425, 0.0673, 0.1227, 0.0825, 0.1252, 0.1073], + device='cuda:0'), in_proj_covar=tensor([0.0355, 0.0351, 0.0355, 0.0379, 0.0456, 0.0384, 0.0331, 0.0338], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 19:10:55,284 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.059e+02 5.030e+02 6.558e+02 8.674e+02 2.471e+03, threshold=1.312e+03, percent-clipped=9.0 +2023-04-02 19:11:32,327 INFO [train.py:903] (0/4) Epoch 21, batch 3300, loss[loss=0.209, simple_loss=0.2937, pruned_loss=0.06222, over 19674.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.289, pruned_loss=0.06542, over 3825295.63 frames. ], batch size: 53, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:11:37,041 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 19:11:39,024 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-02 19:11:43,190 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=139868.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:12:34,959 INFO [train.py:903] (0/4) Epoch 21, batch 3350, loss[loss=0.1929, simple_loss=0.2673, pruned_loss=0.0592, over 15559.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.2883, pruned_loss=0.06521, over 3815768.69 frames. ], batch size: 34, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:13:00,887 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.203e+02 4.734e+02 5.660e+02 7.256e+02 1.171e+03, threshold=1.132e+03, percent-clipped=0.0 +2023-04-02 19:13:04,626 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=139933.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:13:15,672 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-04-02 19:13:17,284 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=139944.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:13:37,533 INFO [train.py:903] (0/4) Epoch 21, batch 3400, loss[loss=0.2284, simple_loss=0.3089, pruned_loss=0.07397, over 18074.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2879, pruned_loss=0.06511, over 3815226.63 frames. ], batch size: 83, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:13:52,373 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=139971.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:14:22,124 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=139996.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:14:27,648 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-140000.pt +2023-04-02 19:14:41,302 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140009.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:14:42,108 INFO [train.py:903] (0/4) Epoch 21, batch 3450, loss[loss=0.2081, simple_loss=0.2902, pruned_loss=0.06296, over 19314.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2899, pruned_loss=0.06589, over 3821509.92 frames. ], batch size: 66, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:14:43,268 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 19:14:56,546 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140022.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:14:57,031 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 19:14:58,850 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8921, 1.2252, 0.9921, 0.8957, 1.0925, 0.8793, 0.9309, 1.1715], + device='cuda:0'), covar=tensor([0.0548, 0.0805, 0.1058, 0.0696, 0.0519, 0.1255, 0.0519, 0.0457], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0312, 0.0335, 0.0260, 0.0244, 0.0334, 0.0287, 0.0272], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:15:06,286 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.793e+02 5.068e+02 6.577e+02 8.644e+02 2.362e+03, threshold=1.315e+03, percent-clipped=9.0 +2023-04-02 19:15:11,058 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140034.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:15:27,194 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140047.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:15:28,349 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140048.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:15:41,562 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140059.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:15:42,380 INFO [train.py:903] (0/4) Epoch 21, batch 3500, loss[loss=0.2226, simple_loss=0.3017, pruned_loss=0.0718, over 17337.00 frames. ], tot_loss[loss=0.2119, simple_loss=0.2909, pruned_loss=0.06645, over 3831897.11 frames. ], batch size: 101, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:16:45,071 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-04-02 19:16:45,566 INFO [train.py:903] (0/4) Epoch 21, batch 3550, loss[loss=0.2276, simple_loss=0.2958, pruned_loss=0.07977, over 19489.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2911, pruned_loss=0.06644, over 3834592.82 frames. ], batch size: 64, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:16:59,737 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140121.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:17:08,439 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140128.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:17:10,317 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.090e+02 5.095e+02 6.549e+02 8.089e+02 2.006e+03, threshold=1.310e+03, percent-clipped=2.0 +2023-04-02 19:17:11,692 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140131.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:17:36,893 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140151.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:17:47,020 INFO [train.py:903] (0/4) Epoch 21, batch 3600, loss[loss=0.2245, simple_loss=0.3082, pruned_loss=0.07041, over 19529.00 frames. ], tot_loss[loss=0.2112, simple_loss=0.29, pruned_loss=0.0662, over 3842487.61 frames. ], batch size: 56, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:17:56,425 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140166.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:17:58,696 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140168.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:18:51,572 INFO [train.py:903] (0/4) Epoch 21, batch 3650, loss[loss=0.2218, simple_loss=0.3049, pruned_loss=0.06936, over 19773.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2898, pruned_loss=0.06594, over 3838764.93 frames. ], batch size: 54, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:18:54,130 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140212.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:19:15,568 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.814e+02 5.123e+02 6.079e+02 7.474e+02 1.635e+03, threshold=1.216e+03, percent-clipped=1.0 +2023-04-02 19:19:36,491 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140246.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:19:54,182 INFO [train.py:903] (0/4) Epoch 21, batch 3700, loss[loss=0.1821, simple_loss=0.2649, pruned_loss=0.04967, over 19794.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2902, pruned_loss=0.06585, over 3845975.76 frames. ], batch size: 47, lr: 3.92e-03, grad_scale: 8.0 +2023-04-02 19:20:01,678 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140266.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:20:15,857 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140277.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:20:20,499 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140281.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:20:22,748 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140283.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:20:49,158 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140304.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:20:55,390 INFO [train.py:903] (0/4) Epoch 21, batch 3750, loss[loss=0.1862, simple_loss=0.2644, pruned_loss=0.05404, over 19830.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2906, pruned_loss=0.06631, over 3817072.41 frames. ], batch size: 52, lr: 3.92e-03, grad_scale: 4.0 +2023-04-02 19:21:02,636 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140315.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:21:13,149 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-04-02 19:21:17,849 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140327.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:21:20,319 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140329.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:21:22,188 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.949e+02 4.554e+02 6.128e+02 7.118e+02 1.255e+03, threshold=1.226e+03, percent-clipped=1.0 +2023-04-02 19:21:31,154 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-04-02 19:21:33,844 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140340.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:21:44,250 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4803, 2.4012, 2.1950, 2.5937, 2.2841, 2.2673, 2.2303, 2.4552], + device='cuda:0'), covar=tensor([0.1049, 0.1634, 0.1386, 0.1090, 0.1401, 0.0498, 0.1259, 0.0668], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0360, 0.0312, 0.0251, 0.0301, 0.0252, 0.0310, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:21:57,847 INFO [train.py:903] (0/4) Epoch 21, batch 3800, loss[loss=0.2267, simple_loss=0.307, pruned_loss=0.07318, over 19402.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2901, pruned_loss=0.06571, over 3833859.97 frames. ], batch size: 66, lr: 3.92e-03, grad_scale: 4.0 +2023-04-02 19:22:30,450 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 19:22:54,714 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140405.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:23:01,180 INFO [train.py:903] (0/4) Epoch 21, batch 3850, loss[loss=0.1982, simple_loss=0.2796, pruned_loss=0.05842, over 17318.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2898, pruned_loss=0.0655, over 3826230.00 frames. ], batch size: 38, lr: 3.92e-03, grad_scale: 4.0 +2023-04-02 19:23:19,346 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.45 vs. limit=2.0 +2023-04-02 19:23:25,867 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.131e+02 5.035e+02 6.153e+02 7.922e+02 1.662e+03, threshold=1.231e+03, percent-clipped=3.0 +2023-04-02 19:24:03,191 INFO [train.py:903] (0/4) Epoch 21, batch 3900, loss[loss=0.1784, simple_loss=0.2564, pruned_loss=0.05017, over 19755.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2896, pruned_loss=0.06556, over 3826879.30 frames. ], batch size: 46, lr: 3.92e-03, grad_scale: 4.0 +2023-04-02 19:24:09,090 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140465.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:24:17,273 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140472.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:24:55,703 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140502.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:25:01,336 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140507.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:25:04,333 INFO [train.py:903] (0/4) Epoch 21, batch 3950, loss[loss=0.199, simple_loss=0.2901, pruned_loss=0.05394, over 19482.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2897, pruned_loss=0.06547, over 3821968.66 frames. ], batch size: 64, lr: 3.91e-03, grad_scale: 4.0 +2023-04-02 19:25:10,222 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 19:25:20,283 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140522.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:25:26,803 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140527.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:25:26,909 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140527.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:25:31,859 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.897e+02 4.763e+02 5.824e+02 7.544e+02 1.413e+03, threshold=1.165e+03, percent-clipped=2.0 +2023-04-02 19:25:40,198 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140537.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:25:42,497 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140539.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:25:51,816 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140547.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:26:07,097 INFO [train.py:903] (0/4) Epoch 21, batch 4000, loss[loss=0.2083, simple_loss=0.2892, pruned_loss=0.06377, over 19592.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2898, pruned_loss=0.06548, over 3824267.66 frames. ], batch size: 57, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:26:09,718 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140562.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:26:13,081 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140564.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:26:14,609 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.50 vs. limit=2.0 +2023-04-02 19:26:33,167 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140580.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:26:36,731 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140583.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:26:41,305 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140587.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:26:55,163 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 19:27:09,309 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140608.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:27:11,209 INFO [train.py:903] (0/4) Epoch 21, batch 4050, loss[loss=0.2318, simple_loss=0.3112, pruned_loss=0.07616, over 19583.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2898, pruned_loss=0.06502, over 3823691.44 frames. ], batch size: 61, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:27:25,062 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140621.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:27:36,533 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.169e+02 5.165e+02 6.710e+02 8.332e+02 1.443e+03, threshold=1.342e+03, percent-clipped=3.0 +2023-04-02 19:28:13,809 INFO [train.py:903] (0/4) Epoch 21, batch 4100, loss[loss=0.284, simple_loss=0.3385, pruned_loss=0.1147, over 13410.00 frames. ], tot_loss[loss=0.2096, simple_loss=0.2891, pruned_loss=0.06503, over 3833618.42 frames. ], batch size: 135, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:28:33,319 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.81 vs. limit=5.0 +2023-04-02 19:28:52,810 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 19:29:15,741 INFO [train.py:903] (0/4) Epoch 21, batch 4150, loss[loss=0.2165, simple_loss=0.2982, pruned_loss=0.06744, over 19654.00 frames. ], tot_loss[loss=0.2095, simple_loss=0.2895, pruned_loss=0.06474, over 3829728.35 frames. ], batch size: 60, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:29:42,680 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.451e+02 5.318e+02 6.390e+02 7.957e+02 1.686e+03, threshold=1.278e+03, percent-clipped=3.0 +2023-04-02 19:29:50,042 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140736.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:29:55,881 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140741.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:30:05,092 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140749.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:30:17,540 INFO [train.py:903] (0/4) Epoch 21, batch 4200, loss[loss=0.2323, simple_loss=0.3101, pruned_loss=0.07725, over 19507.00 frames. ], tot_loss[loss=0.2097, simple_loss=0.2894, pruned_loss=0.065, over 3826251.28 frames. ], batch size: 64, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:30:24,373 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 19:30:41,732 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7605, 2.5820, 2.5533, 2.8248, 2.5077, 2.3316, 2.2206, 2.7155], + device='cuda:0'), covar=tensor([0.0888, 0.1488, 0.1221, 0.0895, 0.1247, 0.0489, 0.1317, 0.0574], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0360, 0.0311, 0.0250, 0.0301, 0.0252, 0.0310, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:31:10,596 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6275, 1.3607, 1.5494, 1.4435, 3.2298, 1.0245, 2.5076, 3.5961], + device='cuda:0'), covar=tensor([0.0483, 0.2649, 0.2788, 0.1954, 0.0701, 0.2635, 0.1097, 0.0267], + device='cuda:0'), in_proj_covar=tensor([0.0405, 0.0363, 0.0383, 0.0343, 0.0370, 0.0347, 0.0374, 0.0398], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:31:21,343 INFO [train.py:903] (0/4) Epoch 21, batch 4250, loss[loss=0.2214, simple_loss=0.3029, pruned_loss=0.06998, over 19447.00 frames. ], tot_loss[loss=0.2096, simple_loss=0.2893, pruned_loss=0.06493, over 3839724.20 frames. ], batch size: 64, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:31:25,367 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.09 vs. limit=2.0 +2023-04-02 19:31:39,325 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7117, 4.2519, 4.4761, 4.4838, 1.6549, 4.1822, 3.6710, 4.1772], + device='cuda:0'), covar=tensor([0.1794, 0.0819, 0.0637, 0.0696, 0.6196, 0.1020, 0.0679, 0.1188], + device='cuda:0'), in_proj_covar=tensor([0.0783, 0.0733, 0.0947, 0.0830, 0.0829, 0.0702, 0.0571, 0.0876], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 19:31:40,145 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 19:31:42,786 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140827.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:31:47,367 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.155e+02 5.291e+02 6.401e+02 8.566e+02 1.506e+03, threshold=1.280e+03, percent-clipped=6.0 +2023-04-02 19:31:50,845 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 19:31:53,755 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140836.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:32:01,546 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140843.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:32:12,614 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140851.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:32:24,303 INFO [train.py:903] (0/4) Epoch 21, batch 4300, loss[loss=0.204, simple_loss=0.2828, pruned_loss=0.06264, over 19522.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2888, pruned_loss=0.06473, over 3831156.88 frames. ], batch size: 54, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:32:25,851 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140861.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:32:29,397 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140864.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:32:33,861 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=140868.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:32:37,166 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=140871.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:33:18,642 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 19:33:24,433 INFO [train.py:903] (0/4) Epoch 21, batch 4350, loss[loss=0.1958, simple_loss=0.2858, pruned_loss=0.05287, over 19676.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2885, pruned_loss=0.06472, over 3831735.00 frames. ], batch size: 60, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:33:51,508 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.395e+02 4.814e+02 5.846e+02 6.905e+02 1.613e+03, threshold=1.169e+03, percent-clipped=2.0 +2023-04-02 19:33:56,408 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140935.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:34:25,325 INFO [train.py:903] (0/4) Epoch 21, batch 4400, loss[loss=0.2087, simple_loss=0.2941, pruned_loss=0.06166, over 18265.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2891, pruned_loss=0.06535, over 3825491.66 frames. ], batch size: 84, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:34:29,685 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3747, 3.9756, 2.6387, 3.5293, 1.1401, 3.9258, 3.8579, 3.8966], + device='cuda:0'), covar=tensor([0.0632, 0.1001, 0.1838, 0.0809, 0.3500, 0.0666, 0.0781, 0.1217], + device='cuda:0'), in_proj_covar=tensor([0.0494, 0.0405, 0.0485, 0.0343, 0.0395, 0.0425, 0.0418, 0.0453], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:34:32,107 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=140965.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:34:33,344 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140966.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:34:53,850 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 19:34:58,756 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=140986.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:35:02,873 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 19:35:05,662 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=140992.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:35:27,476 INFO [train.py:903] (0/4) Epoch 21, batch 4450, loss[loss=0.1839, simple_loss=0.2611, pruned_loss=0.05335, over 19616.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2889, pruned_loss=0.06548, over 3825590.34 frames. ], batch size: 50, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:35:37,881 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141017.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:35:54,628 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.268e+02 5.226e+02 6.450e+02 8.222e+02 2.218e+03, threshold=1.290e+03, percent-clipped=9.0 +2023-04-02 19:36:06,100 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-02 19:36:32,489 INFO [train.py:903] (0/4) Epoch 21, batch 4500, loss[loss=0.1843, simple_loss=0.2794, pruned_loss=0.04463, over 19613.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2891, pruned_loss=0.06552, over 3824987.51 frames. ], batch size: 57, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:37:01,368 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=141085.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:37:34,668 INFO [train.py:903] (0/4) Epoch 21, batch 4550, loss[loss=0.197, simple_loss=0.282, pruned_loss=0.05595, over 18097.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.2898, pruned_loss=0.06565, over 3825913.55 frames. ], batch size: 83, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:37:46,075 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 19:37:46,482 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141120.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:37:59,727 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.292e+02 5.246e+02 6.195e+02 7.478e+02 1.454e+03, threshold=1.239e+03, percent-clipped=4.0 +2023-04-02 19:38:10,838 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 19:38:18,670 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141145.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:38:28,863 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0107, 3.6290, 2.3217, 3.2650, 0.8347, 3.5503, 3.5120, 3.5343], + device='cuda:0'), covar=tensor([0.0837, 0.1210, 0.2192, 0.0947, 0.3970, 0.0810, 0.0981, 0.1227], + device='cuda:0'), in_proj_covar=tensor([0.0499, 0.0408, 0.0489, 0.0345, 0.0397, 0.0428, 0.0420, 0.0455], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:38:35,234 INFO [train.py:903] (0/4) Epoch 21, batch 4600, loss[loss=0.2246, simple_loss=0.2953, pruned_loss=0.0769, over 19696.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2907, pruned_loss=0.06619, over 3821098.80 frames. ], batch size: 60, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:38:36,709 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141161.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:38:48,706 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=141171.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:39:25,208 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141200.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:39:36,153 INFO [train.py:903] (0/4) Epoch 21, batch 4650, loss[loss=0.2139, simple_loss=0.299, pruned_loss=0.0644, over 19769.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2903, pruned_loss=0.06583, over 3823168.27 frames. ], batch size: 54, lr: 3.91e-03, grad_scale: 8.0 +2023-04-02 19:39:53,019 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141222.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:39:56,094 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 19:40:02,331 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.173e+02 4.513e+02 5.853e+02 7.712e+02 1.984e+03, threshold=1.171e+03, percent-clipped=6.0 +2023-04-02 19:40:05,980 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 19:40:12,928 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0191, 3.6619, 2.5283, 3.3055, 1.0997, 3.5945, 3.5123, 3.6378], + device='cuda:0'), covar=tensor([0.0780, 0.1070, 0.2024, 0.0976, 0.3712, 0.0754, 0.0932, 0.1137], + device='cuda:0'), in_proj_covar=tensor([0.0495, 0.0406, 0.0486, 0.0343, 0.0395, 0.0424, 0.0419, 0.0451], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:40:15,513 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141242.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:40:18,306 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-04-02 19:40:21,237 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141247.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:40:37,899 INFO [train.py:903] (0/4) Epoch 21, batch 4700, loss[loss=0.1875, simple_loss=0.261, pruned_loss=0.05698, over 19781.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.2899, pruned_loss=0.06559, over 3823764.45 frames. ], batch size: 48, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:40:45,391 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-02 19:40:47,257 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141267.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:41:00,529 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 19:41:00,630 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=141279.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:41:08,822 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141286.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:41:38,490 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=141309.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:41:39,482 INFO [train.py:903] (0/4) Epoch 21, batch 4750, loss[loss=0.2253, simple_loss=0.3126, pruned_loss=0.06895, over 19513.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2902, pruned_loss=0.06559, over 3823593.09 frames. ], batch size: 64, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:41:59,001 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6957, 2.4472, 2.4207, 2.6694, 2.4658, 2.3056, 2.2734, 2.6167], + device='cuda:0'), covar=tensor([0.0905, 0.1532, 0.1199, 0.0968, 0.1264, 0.0488, 0.1208, 0.0601], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0360, 0.0312, 0.0251, 0.0301, 0.0253, 0.0309, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:42:03,070 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.546e+02 5.235e+02 6.308e+02 8.195e+02 2.468e+03, threshold=1.262e+03, percent-clipped=8.0 +2023-04-02 19:42:24,995 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141347.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:42:30,013 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-02 19:42:39,822 INFO [train.py:903] (0/4) Epoch 21, batch 4800, loss[loss=0.1918, simple_loss=0.2677, pruned_loss=0.05799, over 19666.00 frames. ], tot_loss[loss=0.2114, simple_loss=0.2906, pruned_loss=0.06611, over 3822757.61 frames. ], batch size: 53, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:43:23,134 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141394.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:43:41,794 INFO [train.py:903] (0/4) Epoch 21, batch 4850, loss[loss=0.2269, simple_loss=0.3026, pruned_loss=0.07564, over 18141.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2906, pruned_loss=0.06613, over 3828306.04 frames. ], batch size: 83, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:44:01,418 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141424.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:44:07,725 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 19:44:08,842 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.291e+02 5.192e+02 6.534e+02 8.985e+02 2.500e+03, threshold=1.307e+03, percent-clipped=12.0 +2023-04-02 19:44:13,622 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141435.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 19:44:28,284 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 19:44:33,280 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 19:44:34,448 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 19:44:39,583 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141456.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:44:44,215 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141459.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:44:45,061 INFO [train.py:903] (0/4) Epoch 21, batch 4900, loss[loss=0.2186, simple_loss=0.3034, pruned_loss=0.06691, over 19459.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2902, pruned_loss=0.06575, over 3831993.97 frames. ], batch size: 64, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:44:47,103 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 19:45:05,776 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 19:45:10,378 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141481.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:45:18,453 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2139, 2.1389, 1.9799, 1.8020, 1.6680, 1.8139, 0.5882, 1.1582], + device='cuda:0'), covar=tensor([0.0606, 0.0531, 0.0399, 0.0685, 0.1079, 0.0755, 0.1216, 0.0986], + device='cuda:0'), in_proj_covar=tensor([0.0354, 0.0350, 0.0353, 0.0378, 0.0457, 0.0383, 0.0332, 0.0338], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 19:45:40,778 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=141505.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:45:46,097 INFO [train.py:903] (0/4) Epoch 21, batch 4950, loss[loss=0.2355, simple_loss=0.3166, pruned_loss=0.07723, over 19508.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2902, pruned_loss=0.06595, over 3804253.82 frames. ], batch size: 64, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:46:03,691 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 19:46:10,479 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.127e+02 5.116e+02 6.031e+02 7.721e+02 1.403e+03, threshold=1.206e+03, percent-clipped=1.0 +2023-04-02 19:46:26,548 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141542.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:46:28,465 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 19:46:46,926 INFO [train.py:903] (0/4) Epoch 21, batch 5000, loss[loss=0.2002, simple_loss=0.2939, pruned_loss=0.05329, over 19603.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2898, pruned_loss=0.06549, over 3808731.95 frames. ], batch size: 57, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:46:53,584 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 19:46:55,134 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141567.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:47:08,259 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 19:47:48,192 INFO [train.py:903] (0/4) Epoch 21, batch 5050, loss[loss=0.2132, simple_loss=0.3026, pruned_loss=0.06187, over 19766.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2908, pruned_loss=0.0657, over 3811540.22 frames. ], batch size: 54, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:48:00,894 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2150, 1.2668, 1.6788, 1.2076, 2.6199, 3.5056, 3.2683, 3.6815], + device='cuda:0'), covar=tensor([0.1652, 0.3845, 0.3419, 0.2486, 0.0627, 0.0192, 0.0211, 0.0267], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0321, 0.0349, 0.0264, 0.0241, 0.0184, 0.0215, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 19:48:03,155 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141620.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:48:16,453 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.325e+02 4.754e+02 5.832e+02 6.826e+02 1.423e+03, threshold=1.166e+03, percent-clipped=2.0 +2023-04-02 19:48:25,907 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 19:48:39,224 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141650.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:48:53,405 INFO [train.py:903] (0/4) Epoch 21, batch 5100, loss[loss=0.2022, simple_loss=0.2908, pruned_loss=0.05675, over 19517.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2901, pruned_loss=0.06497, over 3818635.46 frames. ], batch size: 56, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:49:04,603 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 19:49:07,002 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 19:49:11,630 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 19:49:11,955 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141675.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:49:17,701 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141680.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:49:23,377 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3032, 3.8360, 3.9442, 3.9385, 1.5961, 3.7789, 3.2575, 3.6980], + device='cuda:0'), covar=tensor([0.1659, 0.0889, 0.0626, 0.0713, 0.5759, 0.0883, 0.0715, 0.1088], + device='cuda:0'), in_proj_covar=tensor([0.0789, 0.0740, 0.0955, 0.0835, 0.0837, 0.0706, 0.0576, 0.0881], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 19:49:30,089 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=141691.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:49:50,662 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141705.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:49:56,348 INFO [train.py:903] (0/4) Epoch 21, batch 5150, loss[loss=0.2298, simple_loss=0.3019, pruned_loss=0.07882, over 18792.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2907, pruned_loss=0.06576, over 3805113.61 frames. ], batch size: 74, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:50:09,410 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 19:50:20,923 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.369e+02 4.996e+02 6.424e+02 8.131e+02 1.633e+03, threshold=1.285e+03, percent-clipped=6.0 +2023-04-02 19:50:46,536 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 19:50:56,442 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-02 19:50:58,074 INFO [train.py:903] (0/4) Epoch 21, batch 5200, loss[loss=0.2383, simple_loss=0.3126, pruned_loss=0.08197, over 19672.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2905, pruned_loss=0.06548, over 3806029.91 frames. ], batch size: 58, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:51:14,046 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 19:51:22,176 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=141779.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 19:51:51,405 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=141803.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:51:54,942 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141806.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:51:58,264 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 19:51:59,455 INFO [train.py:903] (0/4) Epoch 21, batch 5250, loss[loss=0.1836, simple_loss=0.2553, pruned_loss=0.05592, over 19790.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2884, pruned_loss=0.06455, over 3828403.28 frames. ], batch size: 48, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:52:28,863 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.823e+02 4.884e+02 6.185e+02 7.574e+02 1.457e+03, threshold=1.237e+03, percent-clipped=1.0 +2023-04-02 19:53:02,552 INFO [train.py:903] (0/4) Epoch 21, batch 5300, loss[loss=0.1957, simple_loss=0.2677, pruned_loss=0.06187, over 19408.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.289, pruned_loss=0.06527, over 3817099.80 frames. ], batch size: 48, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:53:23,461 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=141876.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:53:24,187 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 19:53:44,797 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141894.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 19:53:50,270 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9671, 1.7314, 1.6109, 1.9662, 1.6297, 1.7385, 1.5916, 1.8785], + device='cuda:0'), covar=tensor([0.1067, 0.1383, 0.1507, 0.0989, 0.1366, 0.0556, 0.1412, 0.0721], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0358, 0.0311, 0.0251, 0.0299, 0.0252, 0.0310, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 19:53:55,954 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=141901.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:54:06,903 INFO [train.py:903] (0/4) Epoch 21, batch 5350, loss[loss=0.237, simple_loss=0.322, pruned_loss=0.076, over 19596.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2892, pruned_loss=0.06528, over 3824588.48 frames. ], batch size: 61, lr: 3.90e-03, grad_scale: 8.0 +2023-04-02 19:54:16,560 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=141918.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:54:32,497 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.261e+02 4.997e+02 5.942e+02 7.628e+02 1.099e+03, threshold=1.188e+03, percent-clipped=0.0 +2023-04-02 19:54:35,137 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141934.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:54:43,732 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 19:54:51,430 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=141946.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:55:07,484 INFO [train.py:903] (0/4) Epoch 21, batch 5400, loss[loss=0.1834, simple_loss=0.2615, pruned_loss=0.05267, over 19767.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2899, pruned_loss=0.06586, over 3832836.22 frames. ], batch size: 48, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 19:55:14,487 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4986, 3.7137, 4.0166, 4.0146, 2.3310, 3.7540, 3.4522, 3.8203], + device='cuda:0'), covar=tensor([0.1374, 0.3002, 0.0650, 0.0717, 0.4488, 0.1374, 0.0572, 0.1023], + device='cuda:0'), in_proj_covar=tensor([0.0781, 0.0735, 0.0946, 0.0827, 0.0828, 0.0701, 0.0569, 0.0871], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 19:55:57,949 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-142000.pt +2023-04-02 19:56:10,561 INFO [train.py:903] (0/4) Epoch 21, batch 5450, loss[loss=0.188, simple_loss=0.2725, pruned_loss=0.05177, over 19828.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2899, pruned_loss=0.06593, over 3817658.97 frames. ], batch size: 52, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 19:56:39,062 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8827, 1.3422, 1.6440, 0.5621, 1.9339, 2.4391, 2.1127, 2.5914], + device='cuda:0'), covar=tensor([0.1660, 0.3670, 0.3246, 0.2843, 0.0658, 0.0287, 0.0353, 0.0363], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0321, 0.0349, 0.0263, 0.0242, 0.0184, 0.0215, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 19:56:39,855 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.702e+02 5.354e+02 6.392e+02 8.231e+02 1.329e+03, threshold=1.278e+03, percent-clipped=1.0 +2023-04-02 19:57:14,049 INFO [train.py:903] (0/4) Epoch 21, batch 5500, loss[loss=0.2169, simple_loss=0.2978, pruned_loss=0.06804, over 19520.00 frames. ], tot_loss[loss=0.2113, simple_loss=0.2901, pruned_loss=0.06628, over 3815549.94 frames. ], batch size: 56, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 19:57:18,001 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=142062.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:57:42,006 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 19:57:48,309 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=142087.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:58:15,211 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-02 19:58:17,740 INFO [train.py:903] (0/4) Epoch 21, batch 5550, loss[loss=0.2576, simple_loss=0.3166, pruned_loss=0.09928, over 13049.00 frames. ], tot_loss[loss=0.2118, simple_loss=0.2907, pruned_loss=0.06652, over 3811910.96 frames. ], batch size: 136, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 19:58:26,200 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 19:58:27,680 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7774, 4.1954, 4.4488, 4.4674, 1.7509, 4.2061, 3.6432, 4.1792], + device='cuda:0'), covar=tensor([0.1576, 0.1016, 0.0590, 0.0632, 0.5935, 0.0916, 0.0676, 0.1105], + device='cuda:0'), in_proj_covar=tensor([0.0781, 0.0736, 0.0944, 0.0826, 0.0827, 0.0702, 0.0570, 0.0874], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 19:58:34,914 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5219, 1.5837, 1.7523, 1.7097, 2.5982, 2.2959, 2.6328, 1.2009], + device='cuda:0'), covar=tensor([0.2368, 0.4186, 0.2577, 0.1820, 0.1330, 0.2009, 0.1297, 0.4239], + device='cuda:0'), in_proj_covar=tensor([0.0531, 0.0639, 0.0709, 0.0483, 0.0617, 0.0531, 0.0660, 0.0547], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 19:58:43,636 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.159e+02 4.904e+02 5.778e+02 7.569e+02 2.193e+03, threshold=1.156e+03, percent-clipped=4.0 +2023-04-02 19:59:07,890 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=142150.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 19:59:15,792 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 19:59:19,356 INFO [train.py:903] (0/4) Epoch 21, batch 5600, loss[loss=0.1851, simple_loss=0.276, pruned_loss=0.0471, over 19674.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2901, pruned_loss=0.06604, over 3825530.33 frames. ], batch size: 55, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 19:59:36,298 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=142174.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 19:59:37,405 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=142175.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 19:59:58,342 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3484, 1.2364, 1.2911, 1.3415, 1.0661, 1.3961, 1.4254, 1.3582], + device='cuda:0'), covar=tensor([0.0848, 0.0991, 0.1041, 0.0683, 0.0850, 0.0886, 0.0845, 0.0775], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0226, 0.0240, 0.0224, 0.0211, 0.0186, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 20:00:08,574 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=142199.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:00:20,402 INFO [train.py:903] (0/4) Epoch 21, batch 5650, loss[loss=0.1843, simple_loss=0.2641, pruned_loss=0.05224, over 19389.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2902, pruned_loss=0.06595, over 3820691.37 frames. ], batch size: 47, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:00:49,291 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.717e+02 4.896e+02 5.835e+02 7.915e+02 1.772e+03, threshold=1.167e+03, percent-clipped=3.0 +2023-04-02 20:01:05,167 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5391, 2.4641, 2.3708, 2.6503, 2.4092, 2.2003, 1.9563, 2.6311], + device='cuda:0'), covar=tensor([0.0961, 0.1504, 0.1294, 0.0957, 0.1275, 0.0519, 0.1412, 0.0625], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0356, 0.0311, 0.0249, 0.0298, 0.0251, 0.0309, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:01:10,294 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 20:01:22,640 INFO [train.py:903] (0/4) Epoch 21, batch 5700, loss[loss=0.2223, simple_loss=0.3025, pruned_loss=0.0711, over 19776.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2901, pruned_loss=0.06588, over 3838168.11 frames. ], batch size: 56, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:01:45,761 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=142278.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:01:59,755 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=142290.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:02:09,014 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3780, 0.9855, 1.1778, 2.2309, 1.5897, 1.3312, 1.8156, 1.2534], + device='cuda:0'), covar=tensor([0.1169, 0.1746, 0.1385, 0.0850, 0.1061, 0.1429, 0.1162, 0.1123], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0226, 0.0239, 0.0224, 0.0210, 0.0186, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 20:02:26,062 INFO [train.py:903] (0/4) Epoch 21, batch 5750, loss[loss=0.2125, simple_loss=0.2941, pruned_loss=0.06548, over 19784.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2902, pruned_loss=0.06597, over 3819639.29 frames. ], batch size: 56, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:02:28,353 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 20:02:36,343 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 20:02:41,007 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 20:02:51,480 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.086e+02 5.170e+02 6.700e+02 8.460e+02 1.665e+03, threshold=1.340e+03, percent-clipped=6.0 +2023-04-02 20:03:26,602 INFO [train.py:903] (0/4) Epoch 21, batch 5800, loss[loss=0.2171, simple_loss=0.2808, pruned_loss=0.07667, over 19741.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2891, pruned_loss=0.06557, over 3826721.97 frames. ], batch size: 45, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:04:08,885 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=142393.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:04:22,739 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=142405.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:04:28,053 INFO [train.py:903] (0/4) Epoch 21, batch 5850, loss[loss=0.209, simple_loss=0.2905, pruned_loss=0.06371, over 19666.00 frames. ], tot_loss[loss=0.2102, simple_loss=0.2897, pruned_loss=0.06541, over 3841233.23 frames. ], batch size: 58, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:04:57,526 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.411e+02 4.867e+02 6.014e+02 7.504e+02 1.855e+03, threshold=1.203e+03, percent-clipped=4.0 +2023-04-02 20:05:31,500 INFO [train.py:903] (0/4) Epoch 21, batch 5900, loss[loss=0.2046, simple_loss=0.2928, pruned_loss=0.05817, over 17289.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2896, pruned_loss=0.0652, over 3836774.25 frames. ], batch size: 101, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:05:35,059 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 20:05:58,770 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 20:06:36,834 INFO [train.py:903] (0/4) Epoch 21, batch 5950, loss[loss=0.2121, simple_loss=0.2756, pruned_loss=0.0743, over 19090.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.2895, pruned_loss=0.06501, over 3842088.00 frames. ], batch size: 42, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:07:02,052 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.165e+02 4.752e+02 5.691e+02 7.550e+02 2.003e+03, threshold=1.138e+03, percent-clipped=5.0 +2023-04-02 20:07:37,103 INFO [train.py:903] (0/4) Epoch 21, batch 6000, loss[loss=0.2529, simple_loss=0.3166, pruned_loss=0.09457, over 14074.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2905, pruned_loss=0.06581, over 3831000.11 frames. ], batch size: 136, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:07:37,104 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 20:07:50,393 INFO [train.py:937] (0/4) Epoch 21, validation: loss=0.1692, simple_loss=0.2693, pruned_loss=0.03459, over 944034.00 frames. +2023-04-02 20:07:50,394 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-02 20:08:26,799 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=142591.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:08:52,466 INFO [train.py:903] (0/4) Epoch 21, batch 6050, loss[loss=0.2077, simple_loss=0.2887, pruned_loss=0.06331, over 18698.00 frames. ], tot_loss[loss=0.2116, simple_loss=0.2909, pruned_loss=0.06614, over 3830798.64 frames. ], batch size: 74, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:09:18,887 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.159e+02 4.812e+02 5.741e+02 7.692e+02 1.541e+03, threshold=1.148e+03, percent-clipped=3.0 +2023-04-02 20:09:41,449 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=142649.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:09:53,724 INFO [train.py:903] (0/4) Epoch 21, batch 6100, loss[loss=0.1877, simple_loss=0.2698, pruned_loss=0.05282, over 19660.00 frames. ], tot_loss[loss=0.2115, simple_loss=0.2907, pruned_loss=0.06611, over 3820009.67 frames. ], batch size: 53, lr: 3.89e-03, grad_scale: 8.0 +2023-04-02 20:09:55,343 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=142661.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:10:11,615 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=142674.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:10:17,587 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-02 20:10:28,115 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=142686.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:10:55,895 INFO [train.py:903] (0/4) Epoch 21, batch 6150, loss[loss=0.2186, simple_loss=0.2965, pruned_loss=0.07036, over 19532.00 frames. ], tot_loss[loss=0.212, simple_loss=0.2914, pruned_loss=0.06635, over 3823415.30 frames. ], batch size: 54, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:11:24,998 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.643e+02 5.511e+02 6.918e+02 9.659e+02 2.206e+03, threshold=1.384e+03, percent-clipped=13.0 +2023-04-02 20:11:26,176 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 20:11:36,759 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=142742.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:11:59,801 INFO [train.py:903] (0/4) Epoch 21, batch 6200, loss[loss=0.1737, simple_loss=0.2622, pruned_loss=0.04259, over 19848.00 frames. ], tot_loss[loss=0.2128, simple_loss=0.2922, pruned_loss=0.06666, over 3794329.16 frames. ], batch size: 52, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:12:15,917 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5158, 1.6484, 2.0920, 1.7763, 3.1445, 2.5449, 3.5553, 1.6232], + device='cuda:0'), covar=tensor([0.2468, 0.4242, 0.2634, 0.1892, 0.1478, 0.2141, 0.1492, 0.4109], + device='cuda:0'), in_proj_covar=tensor([0.0530, 0.0643, 0.0710, 0.0485, 0.0619, 0.0532, 0.0664, 0.0548], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 20:13:02,732 INFO [train.py:903] (0/4) Epoch 21, batch 6250, loss[loss=0.2106, simple_loss=0.2881, pruned_loss=0.06655, over 13499.00 frames. ], tot_loss[loss=0.2123, simple_loss=0.2917, pruned_loss=0.06641, over 3785116.77 frames. ], batch size: 136, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:13:28,469 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.899e+02 5.305e+02 6.117e+02 7.859e+02 2.157e+03, threshold=1.223e+03, percent-clipped=2.0 +2023-04-02 20:13:30,732 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 20:13:46,848 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2136, 1.2590, 1.7810, 1.0624, 2.3563, 3.0951, 2.8221, 3.2841], + device='cuda:0'), covar=tensor([0.1522, 0.3678, 0.3072, 0.2325, 0.0577, 0.0229, 0.0257, 0.0296], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0319, 0.0349, 0.0263, 0.0240, 0.0184, 0.0214, 0.0252], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 20:14:05,176 INFO [train.py:903] (0/4) Epoch 21, batch 6300, loss[loss=0.2274, simple_loss=0.3045, pruned_loss=0.07509, over 19083.00 frames. ], tot_loss[loss=0.2106, simple_loss=0.2899, pruned_loss=0.06562, over 3795450.17 frames. ], batch size: 69, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:14:47,699 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1908, 2.0968, 1.9160, 1.7172, 1.6652, 1.7547, 0.5139, 1.0583], + device='cuda:0'), covar=tensor([0.0580, 0.0607, 0.0442, 0.0751, 0.1160, 0.0839, 0.1301, 0.1072], + device='cuda:0'), in_proj_covar=tensor([0.0356, 0.0355, 0.0354, 0.0382, 0.0459, 0.0386, 0.0335, 0.0341], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 20:15:07,132 INFO [train.py:903] (0/4) Epoch 21, batch 6350, loss[loss=0.251, simple_loss=0.324, pruned_loss=0.08899, over 19757.00 frames. ], tot_loss[loss=0.2097, simple_loss=0.2892, pruned_loss=0.06514, over 3799174.45 frames. ], batch size: 63, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:15:36,285 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.232e+02 4.679e+02 5.550e+02 7.220e+02 1.923e+03, threshold=1.110e+03, percent-clipped=1.0 +2023-04-02 20:15:39,968 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=142935.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:16:11,261 INFO [train.py:903] (0/4) Epoch 21, batch 6400, loss[loss=0.2312, simple_loss=0.3074, pruned_loss=0.07748, over 13194.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.289, pruned_loss=0.06468, over 3805687.09 frames. ], batch size: 136, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:17:14,730 INFO [train.py:903] (0/4) Epoch 21, batch 6450, loss[loss=0.1907, simple_loss=0.2728, pruned_loss=0.05428, over 19746.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2887, pruned_loss=0.06419, over 3810117.50 frames. ], batch size: 51, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:17:40,515 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.231e+02 4.995e+02 6.270e+02 8.275e+02 2.312e+03, threshold=1.254e+03, percent-clipped=6.0 +2023-04-02 20:18:01,476 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 20:18:04,895 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=143050.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:18:16,201 INFO [train.py:903] (0/4) Epoch 21, batch 6500, loss[loss=0.2014, simple_loss=0.2705, pruned_loss=0.06616, over 19763.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2886, pruned_loss=0.06422, over 3798204.37 frames. ], batch size: 47, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:18:23,436 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 20:18:24,238 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-04-02 20:18:48,992 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=143086.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:19:16,584 INFO [train.py:903] (0/4) Epoch 21, batch 6550, loss[loss=0.1883, simple_loss=0.2706, pruned_loss=0.05295, over 19564.00 frames. ], tot_loss[loss=0.209, simple_loss=0.289, pruned_loss=0.06451, over 3802500.86 frames. ], batch size: 52, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:19:44,580 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.797e+02 5.073e+02 6.169e+02 7.633e+02 2.146e+03, threshold=1.234e+03, percent-clipped=4.0 +2023-04-02 20:20:01,334 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6938, 2.3254, 2.2378, 2.5627, 2.3316, 2.1783, 2.0337, 2.5454], + device='cuda:0'), covar=tensor([0.0870, 0.1594, 0.1364, 0.1015, 0.1374, 0.0531, 0.1392, 0.0625], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0356, 0.0308, 0.0248, 0.0299, 0.0251, 0.0310, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:20:19,906 INFO [train.py:903] (0/4) Epoch 21, batch 6600, loss[loss=0.1986, simple_loss=0.2715, pruned_loss=0.06281, over 19472.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2884, pruned_loss=0.06404, over 3821166.68 frames. ], batch size: 49, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:21:10,256 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=143201.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:21:20,391 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9803, 1.9685, 1.9071, 1.7272, 1.5856, 1.7883, 1.2026, 1.3880], + device='cuda:0'), covar=tensor([0.0680, 0.0725, 0.0491, 0.0801, 0.1221, 0.1029, 0.1265, 0.1057], + device='cuda:0'), in_proj_covar=tensor([0.0360, 0.0359, 0.0360, 0.0385, 0.0465, 0.0390, 0.0338, 0.0344], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 20:21:22,421 INFO [train.py:903] (0/4) Epoch 21, batch 6650, loss[loss=0.1971, simple_loss=0.2625, pruned_loss=0.06586, over 19142.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2875, pruned_loss=0.06382, over 3821960.94 frames. ], batch size: 42, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:21:47,868 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.522e+02 4.940e+02 5.672e+02 7.065e+02 1.538e+03, threshold=1.134e+03, percent-clipped=2.0 +2023-04-02 20:22:23,655 INFO [train.py:903] (0/4) Epoch 21, batch 6700, loss[loss=0.1901, simple_loss=0.2627, pruned_loss=0.05869, over 19400.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2871, pruned_loss=0.0635, over 3824585.34 frames. ], batch size: 47, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:22:52,030 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.93 vs. limit=2.0 +2023-04-02 20:23:19,526 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=143306.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:23:23,650 INFO [train.py:903] (0/4) Epoch 21, batch 6750, loss[loss=0.1766, simple_loss=0.263, pruned_loss=0.04514, over 19848.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2865, pruned_loss=0.0633, over 3821601.45 frames. ], batch size: 52, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:23:48,075 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=143331.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:23:48,849 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.668e+02 4.995e+02 6.197e+02 7.772e+02 2.067e+03, threshold=1.239e+03, percent-clipped=6.0 +2023-04-02 20:24:20,248 INFO [train.py:903] (0/4) Epoch 21, batch 6800, loss[loss=0.2096, simple_loss=0.2968, pruned_loss=0.06113, over 19688.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2862, pruned_loss=0.06352, over 3831605.10 frames. ], batch size: 60, lr: 3.88e-03, grad_scale: 8.0 +2023-04-02 20:24:50,784 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-21.pt +2023-04-02 20:25:06,360 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 20:25:07,469 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 20:25:09,779 INFO [train.py:903] (0/4) Epoch 22, batch 0, loss[loss=0.2129, simple_loss=0.297, pruned_loss=0.06437, over 19291.00 frames. ], tot_loss[loss=0.2129, simple_loss=0.297, pruned_loss=0.06437, over 19291.00 frames. ], batch size: 66, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:25:09,780 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 20:25:18,164 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4047, 1.3529, 1.4099, 1.7995, 1.4658, 1.5941, 1.5515, 1.5439], + device='cuda:0'), covar=tensor([0.0764, 0.0942, 0.0903, 0.0609, 0.0948, 0.0868, 0.1013, 0.0703], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0223, 0.0228, 0.0240, 0.0227, 0.0213, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 20:25:20,457 INFO [train.py:937] (0/4) Epoch 22, validation: loss=0.1683, simple_loss=0.2691, pruned_loss=0.03373, over 944034.00 frames. +2023-04-02 20:25:20,458 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-02 20:25:31,910 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 20:25:55,279 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=143418.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 20:26:14,247 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.150e+02 4.908e+02 5.891e+02 8.006e+02 1.582e+03, threshold=1.178e+03, percent-clipped=4.0 +2023-04-02 20:26:21,017 INFO [train.py:903] (0/4) Epoch 22, batch 50, loss[loss=0.2153, simple_loss=0.2958, pruned_loss=0.06742, over 19332.00 frames. ], tot_loss[loss=0.217, simple_loss=0.2954, pruned_loss=0.06932, over 851964.27 frames. ], batch size: 70, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:26:24,464 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1193, 1.9127, 1.7670, 2.0477, 1.7781, 1.8236, 1.6501, 2.0197], + device='cuda:0'), covar=tensor([0.0943, 0.1381, 0.1501, 0.1047, 0.1300, 0.0541, 0.1481, 0.0667], + device='cuda:0'), in_proj_covar=tensor([0.0267, 0.0353, 0.0308, 0.0246, 0.0296, 0.0248, 0.0307, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:26:42,120 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=143457.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:26:53,911 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 20:27:13,715 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=143482.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:27:19,995 INFO [train.py:903] (0/4) Epoch 22, batch 100, loss[loss=0.221, simple_loss=0.3063, pruned_loss=0.0678, over 19748.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2917, pruned_loss=0.06669, over 1499904.66 frames. ], batch size: 63, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:27:23,803 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=143491.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 20:27:31,530 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 20:27:56,513 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5760, 1.5302, 1.4734, 1.9647, 1.6264, 1.8145, 1.9500, 1.6564], + device='cuda:0'), covar=tensor([0.0838, 0.0898, 0.1021, 0.0693, 0.0801, 0.0762, 0.0767, 0.0691], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0222, 0.0226, 0.0239, 0.0225, 0.0212, 0.0186, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 20:28:12,239 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.507e+02 5.227e+02 6.391e+02 8.671e+02 1.540e+03, threshold=1.278e+03, percent-clipped=3.0 +2023-04-02 20:28:19,047 INFO [train.py:903] (0/4) Epoch 22, batch 150, loss[loss=0.2454, simple_loss=0.3152, pruned_loss=0.08784, over 17434.00 frames. ], tot_loss[loss=0.2125, simple_loss=0.2913, pruned_loss=0.06688, over 2017076.76 frames. ], batch size: 101, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:28:30,206 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0259, 3.6867, 2.5018, 3.3121, 1.0894, 3.6284, 3.5461, 3.5301], + device='cuda:0'), covar=tensor([0.0768, 0.1129, 0.2058, 0.0978, 0.3886, 0.0858, 0.0999, 0.1329], + device='cuda:0'), in_proj_covar=tensor([0.0499, 0.0405, 0.0490, 0.0343, 0.0401, 0.0427, 0.0421, 0.0457], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:29:18,917 INFO [train.py:903] (0/4) Epoch 22, batch 200, loss[loss=0.2435, simple_loss=0.3181, pruned_loss=0.08443, over 18080.00 frames. ], tot_loss[loss=0.2122, simple_loss=0.2907, pruned_loss=0.06688, over 2420575.89 frames. ], batch size: 83, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:29:18,963 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 20:30:12,489 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.534e+02 5.144e+02 6.083e+02 7.742e+02 1.350e+03, threshold=1.217e+03, percent-clipped=1.0 +2023-04-02 20:30:20,919 INFO [train.py:903] (0/4) Epoch 22, batch 250, loss[loss=0.2331, simple_loss=0.3096, pruned_loss=0.07829, over 18405.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2891, pruned_loss=0.06559, over 2737814.75 frames. ], batch size: 84, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:31:20,919 INFO [train.py:903] (0/4) Epoch 22, batch 300, loss[loss=0.2281, simple_loss=0.3095, pruned_loss=0.07333, over 19448.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2896, pruned_loss=0.0652, over 2975530.79 frames. ], batch size: 70, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:31:56,667 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-02 20:32:15,079 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.028e+02 5.065e+02 6.247e+02 8.237e+02 1.383e+03, threshold=1.249e+03, percent-clipped=3.0 +2023-04-02 20:32:22,209 INFO [train.py:903] (0/4) Epoch 22, batch 350, loss[loss=0.2053, simple_loss=0.2757, pruned_loss=0.06746, over 19633.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2888, pruned_loss=0.0645, over 3172729.67 frames. ], batch size: 50, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:32:29,137 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 20:32:51,084 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=143762.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 20:32:52,726 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-04-02 20:33:20,944 INFO [train.py:903] (0/4) Epoch 22, batch 400, loss[loss=0.1821, simple_loss=0.2579, pruned_loss=0.05315, over 19768.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2882, pruned_loss=0.06495, over 3308671.90 frames. ], batch size: 47, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:34:15,302 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.174e+02 5.215e+02 6.557e+02 8.093e+02 2.351e+03, threshold=1.311e+03, percent-clipped=8.0 +2023-04-02 20:34:17,778 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=143835.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 20:34:20,915 INFO [train.py:903] (0/4) Epoch 22, batch 450, loss[loss=0.2126, simple_loss=0.2903, pruned_loss=0.06742, over 19669.00 frames. ], tot_loss[loss=0.2096, simple_loss=0.2888, pruned_loss=0.06515, over 3429897.33 frames. ], batch size: 58, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:34:57,883 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 20:34:58,983 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 20:35:08,553 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=143877.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 20:35:22,927 INFO [train.py:903] (0/4) Epoch 22, batch 500, loss[loss=0.2172, simple_loss=0.3023, pruned_loss=0.06603, over 19563.00 frames. ], tot_loss[loss=0.2097, simple_loss=0.2888, pruned_loss=0.06534, over 3523889.19 frames. ], batch size: 61, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:36:17,497 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.793e+02 5.123e+02 6.359e+02 8.434e+02 1.804e+03, threshold=1.272e+03, percent-clipped=4.0 +2023-04-02 20:36:23,288 INFO [train.py:903] (0/4) Epoch 22, batch 550, loss[loss=0.2519, simple_loss=0.3219, pruned_loss=0.09097, over 18322.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2898, pruned_loss=0.06577, over 3591892.18 frames. ], batch size: 83, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:36:37,249 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=143950.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 20:37:00,075 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5589, 1.3816, 1.4261, 2.0432, 1.5949, 1.8449, 1.9587, 1.6355], + device='cuda:0'), covar=tensor([0.0947, 0.1055, 0.1105, 0.0772, 0.0878, 0.0805, 0.0837, 0.0755], + device='cuda:0'), in_proj_covar=tensor([0.0214, 0.0223, 0.0227, 0.0240, 0.0226, 0.0213, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 20:37:23,305 INFO [train.py:903] (0/4) Epoch 22, batch 600, loss[loss=0.1936, simple_loss=0.284, pruned_loss=0.05161, over 19659.00 frames. ], tot_loss[loss=0.2101, simple_loss=0.2897, pruned_loss=0.0653, over 3651143.59 frames. ], batch size: 55, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:37:36,828 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-144000.pt +2023-04-02 20:38:02,248 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9889, 3.6526, 2.4619, 3.1883, 0.9829, 3.5680, 3.4132, 3.5427], + device='cuda:0'), covar=tensor([0.0828, 0.1013, 0.2118, 0.0976, 0.3900, 0.0787, 0.1105, 0.1215], + device='cuda:0'), in_proj_covar=tensor([0.0500, 0.0408, 0.0492, 0.0345, 0.0402, 0.0430, 0.0426, 0.0459], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:38:06,649 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 20:38:17,798 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.931e+02 4.914e+02 6.190e+02 8.004e+02 1.732e+03, threshold=1.238e+03, percent-clipped=3.0 +2023-04-02 20:38:23,572 INFO [train.py:903] (0/4) Epoch 22, batch 650, loss[loss=0.2024, simple_loss=0.2901, pruned_loss=0.05734, over 19614.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.2891, pruned_loss=0.06484, over 3703067.62 frames. ], batch size: 57, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:39:26,363 INFO [train.py:903] (0/4) Epoch 22, batch 700, loss[loss=0.1908, simple_loss=0.2679, pruned_loss=0.05685, over 19796.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2876, pruned_loss=0.06362, over 3741066.11 frames. ], batch size: 49, lr: 3.78e-03, grad_scale: 8.0 +2023-04-02 20:40:19,658 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.965e+02 4.796e+02 6.107e+02 7.975e+02 1.533e+03, threshold=1.221e+03, percent-clipped=5.0 +2023-04-02 20:40:20,113 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=144133.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 20:40:21,147 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9159, 1.3478, 1.6732, 0.6300, 2.0409, 2.4700, 2.1282, 2.6324], + device='cuda:0'), covar=tensor([0.1521, 0.3568, 0.3083, 0.2613, 0.0610, 0.0251, 0.0342, 0.0360], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0321, 0.0350, 0.0265, 0.0243, 0.0185, 0.0215, 0.0254], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 20:40:26,351 INFO [train.py:903] (0/4) Epoch 22, batch 750, loss[loss=0.1978, simple_loss=0.2737, pruned_loss=0.06092, over 19629.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2879, pruned_loss=0.06409, over 3765107.67 frames. ], batch size: 50, lr: 3.77e-03, grad_scale: 8.0 +2023-04-02 20:40:49,207 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=144158.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 20:41:21,371 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.09 vs. limit=2.0 +2023-04-02 20:41:26,342 INFO [train.py:903] (0/4) Epoch 22, batch 800, loss[loss=0.207, simple_loss=0.2925, pruned_loss=0.0607, over 18211.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2882, pruned_loss=0.06375, over 3790235.57 frames. ], batch size: 83, lr: 3.77e-03, grad_scale: 8.0 +2023-04-02 20:41:44,760 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 20:41:47,230 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=144206.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 20:42:19,086 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=144231.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 20:42:20,959 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.163e+02 5.021e+02 6.351e+02 8.019e+02 1.751e+03, threshold=1.270e+03, percent-clipped=5.0 +2023-04-02 20:42:26,670 INFO [train.py:903] (0/4) Epoch 22, batch 850, loss[loss=0.1951, simple_loss=0.2838, pruned_loss=0.05317, over 18810.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2881, pruned_loss=0.06381, over 3786487.30 frames. ], batch size: 74, lr: 3.77e-03, grad_scale: 8.0 +2023-04-02 20:43:19,873 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 20:43:26,414 INFO [train.py:903] (0/4) Epoch 22, batch 900, loss[loss=0.1626, simple_loss=0.2416, pruned_loss=0.04184, over 17813.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2885, pruned_loss=0.06406, over 3782503.85 frames. ], batch size: 39, lr: 3.77e-03, grad_scale: 8.0 +2023-04-02 20:43:42,705 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0466, 5.4545, 3.1106, 4.8283, 0.9508, 5.6962, 5.4420, 5.6757], + device='cuda:0'), covar=tensor([0.0406, 0.0893, 0.1799, 0.0780, 0.4406, 0.0483, 0.0763, 0.0941], + device='cuda:0'), in_proj_covar=tensor([0.0497, 0.0408, 0.0488, 0.0343, 0.0400, 0.0426, 0.0421, 0.0457], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:44:05,037 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6394, 2.4394, 2.3008, 2.6501, 2.4576, 2.1949, 2.2704, 2.4403], + device='cuda:0'), covar=tensor([0.0919, 0.1567, 0.1349, 0.1057, 0.1329, 0.0501, 0.1260, 0.0701], + device='cuda:0'), in_proj_covar=tensor([0.0269, 0.0357, 0.0312, 0.0250, 0.0300, 0.0249, 0.0308, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:44:21,529 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.404e+02 5.111e+02 6.361e+02 7.451e+02 1.172e+03, threshold=1.272e+03, percent-clipped=0.0 +2023-04-02 20:44:26,104 INFO [train.py:903] (0/4) Epoch 22, batch 950, loss[loss=0.2344, simple_loss=0.3159, pruned_loss=0.0764, over 19329.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2892, pruned_loss=0.0643, over 3794079.87 frames. ], batch size: 70, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:44:30,630 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 20:45:27,344 INFO [train.py:903] (0/4) Epoch 22, batch 1000, loss[loss=0.2045, simple_loss=0.2922, pruned_loss=0.05844, over 19463.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2885, pruned_loss=0.06429, over 3799138.99 frames. ], batch size: 64, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:46:17,109 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=144429.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:46:17,960 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 20:46:22,209 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.871e+02 5.215e+02 6.539e+02 8.059e+02 1.779e+03, threshold=1.308e+03, percent-clipped=4.0 +2023-04-02 20:46:26,887 INFO [train.py:903] (0/4) Epoch 22, batch 1050, loss[loss=0.2133, simple_loss=0.2928, pruned_loss=0.06691, over 19673.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2876, pruned_loss=0.06413, over 3804716.38 frames. ], batch size: 55, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:46:53,388 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0355, 5.0528, 5.8411, 5.8524, 1.7361, 5.4516, 4.6893, 5.4616], + device='cuda:0'), covar=tensor([0.1681, 0.0862, 0.0578, 0.0595, 0.6796, 0.0777, 0.0619, 0.1166], + device='cuda:0'), in_proj_covar=tensor([0.0771, 0.0731, 0.0933, 0.0821, 0.0823, 0.0695, 0.0564, 0.0865], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 20:47:00,685 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 20:47:07,840 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5918, 4.1813, 2.6393, 3.7035, 0.8569, 4.1293, 4.0213, 4.1128], + device='cuda:0'), covar=tensor([0.0643, 0.0954, 0.2015, 0.0856, 0.4307, 0.0638, 0.0955, 0.1211], + device='cuda:0'), in_proj_covar=tensor([0.0496, 0.0407, 0.0486, 0.0342, 0.0400, 0.0427, 0.0421, 0.0457], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:47:26,633 INFO [train.py:903] (0/4) Epoch 22, batch 1100, loss[loss=0.1938, simple_loss=0.2764, pruned_loss=0.05562, over 17979.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2892, pruned_loss=0.06457, over 3808312.11 frames. ], batch size: 83, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:48:21,826 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.380e+02 5.103e+02 6.169e+02 7.547e+02 2.403e+03, threshold=1.234e+03, percent-clipped=2.0 +2023-04-02 20:48:27,964 INFO [train.py:903] (0/4) Epoch 22, batch 1150, loss[loss=0.1844, simple_loss=0.2645, pruned_loss=0.05219, over 19730.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2884, pruned_loss=0.06432, over 3821150.77 frames. ], batch size: 51, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:49:28,349 INFO [train.py:903] (0/4) Epoch 22, batch 1200, loss[loss=0.2519, simple_loss=0.33, pruned_loss=0.08693, over 19665.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.289, pruned_loss=0.0643, over 3815645.23 frames. ], batch size: 59, lr: 3.77e-03, grad_scale: 8.0 +2023-04-02 20:49:59,941 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 20:50:23,752 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.253e+02 4.877e+02 6.112e+02 7.869e+02 2.071e+03, threshold=1.222e+03, percent-clipped=4.0 +2023-04-02 20:50:27,117 INFO [train.py:903] (0/4) Epoch 22, batch 1250, loss[loss=0.2587, simple_loss=0.3291, pruned_loss=0.0942, over 19749.00 frames. ], tot_loss[loss=0.2093, simple_loss=0.2894, pruned_loss=0.0646, over 3816243.12 frames. ], batch size: 63, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:50:33,355 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0631, 1.5700, 1.6311, 1.9096, 1.5020, 1.7098, 1.5345, 1.8733], + device='cuda:0'), covar=tensor([0.0927, 0.1287, 0.1355, 0.0812, 0.1232, 0.0536, 0.1404, 0.0678], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0357, 0.0312, 0.0249, 0.0299, 0.0250, 0.0308, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:51:26,211 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0357, 1.2204, 1.6723, 1.2243, 2.6302, 3.6292, 3.4023, 3.8780], + device='cuda:0'), covar=tensor([0.1768, 0.4015, 0.3484, 0.2541, 0.0616, 0.0181, 0.0209, 0.0253], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0320, 0.0349, 0.0263, 0.0241, 0.0184, 0.0214, 0.0253], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 20:51:28,173 INFO [train.py:903] (0/4) Epoch 22, batch 1300, loss[loss=0.2237, simple_loss=0.3024, pruned_loss=0.07253, over 18734.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2881, pruned_loss=0.06413, over 3827850.95 frames. ], batch size: 74, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:51:33,945 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7464, 1.5898, 1.7439, 1.6533, 4.2981, 1.0900, 2.5312, 4.6147], + device='cuda:0'), covar=tensor([0.0440, 0.2694, 0.2918, 0.1998, 0.0750, 0.2768, 0.1482, 0.0187], + device='cuda:0'), in_proj_covar=tensor([0.0404, 0.0361, 0.0381, 0.0343, 0.0370, 0.0347, 0.0374, 0.0398], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:51:46,040 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.55 vs. limit=5.0 +2023-04-02 20:51:59,492 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.90 vs. limit=2.0 +2023-04-02 20:52:26,776 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.265e+02 4.679e+02 5.951e+02 8.140e+02 2.957e+03, threshold=1.190e+03, percent-clipped=7.0 +2023-04-02 20:52:30,257 INFO [train.py:903] (0/4) Epoch 22, batch 1350, loss[loss=0.2219, simple_loss=0.2827, pruned_loss=0.08058, over 19484.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2888, pruned_loss=0.06477, over 3814219.33 frames. ], batch size: 49, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:52:39,372 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7762, 1.7660, 1.6691, 1.4417, 1.4350, 1.4704, 0.2915, 0.6798], + device='cuda:0'), covar=tensor([0.0586, 0.0580, 0.0384, 0.0516, 0.1081, 0.0651, 0.1165, 0.0987], + device='cuda:0'), in_proj_covar=tensor([0.0354, 0.0355, 0.0357, 0.0381, 0.0457, 0.0385, 0.0334, 0.0339], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 20:53:12,534 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=144773.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:53:20,643 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-02 20:53:31,338 INFO [train.py:903] (0/4) Epoch 22, batch 1400, loss[loss=0.2362, simple_loss=0.3078, pruned_loss=0.08232, over 19658.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2885, pruned_loss=0.0645, over 3803308.84 frames. ], batch size: 60, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:54:22,356 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.88 vs. limit=5.0 +2023-04-02 20:54:28,427 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.383e+02 4.808e+02 5.945e+02 7.380e+02 1.517e+03, threshold=1.189e+03, percent-clipped=2.0 +2023-04-02 20:54:29,515 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 20:54:31,676 INFO [train.py:903] (0/4) Epoch 22, batch 1450, loss[loss=0.2268, simple_loss=0.3033, pruned_loss=0.07514, over 19730.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2876, pruned_loss=0.06405, over 3821098.42 frames. ], batch size: 63, lr: 3.77e-03, grad_scale: 4.0 +2023-04-02 20:55:23,555 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-04-02 20:55:30,804 INFO [train.py:903] (0/4) Epoch 22, batch 1500, loss[loss=0.2316, simple_loss=0.3063, pruned_loss=0.07844, over 17474.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.2878, pruned_loss=0.06426, over 3806902.80 frames. ], batch size: 101, lr: 3.76e-03, grad_scale: 4.0 +2023-04-02 20:55:31,145 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=144888.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:56:27,869 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.316e+02 4.957e+02 5.987e+02 8.036e+02 1.770e+03, threshold=1.197e+03, percent-clipped=5.0 +2023-04-02 20:56:31,399 INFO [train.py:903] (0/4) Epoch 22, batch 1550, loss[loss=0.2453, simple_loss=0.3212, pruned_loss=0.08466, over 19332.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2885, pruned_loss=0.06491, over 3805285.75 frames. ], batch size: 66, lr: 3.76e-03, grad_scale: 4.0 +2023-04-02 20:57:30,408 INFO [train.py:903] (0/4) Epoch 22, batch 1600, loss[loss=0.2263, simple_loss=0.3045, pruned_loss=0.07407, over 18273.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2881, pruned_loss=0.06449, over 3804353.27 frames. ], batch size: 84, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 20:57:48,910 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1354, 1.2897, 1.4824, 1.4059, 2.7619, 1.0789, 2.2690, 3.1167], + device='cuda:0'), covar=tensor([0.0557, 0.2659, 0.2877, 0.1784, 0.0744, 0.2360, 0.1132, 0.0315], + device='cuda:0'), in_proj_covar=tensor([0.0408, 0.0365, 0.0388, 0.0349, 0.0375, 0.0351, 0.0378, 0.0404], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 20:57:50,812 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 20:58:02,378 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145014.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:58:27,678 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.160e+02 4.884e+02 5.870e+02 7.908e+02 1.403e+03, threshold=1.174e+03, percent-clipped=3.0 +2023-04-02 20:58:31,145 INFO [train.py:903] (0/4) Epoch 22, batch 1650, loss[loss=0.2259, simple_loss=0.3151, pruned_loss=0.06831, over 18740.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2885, pruned_loss=0.0646, over 3811494.91 frames. ], batch size: 74, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 20:58:39,506 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145045.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:59:27,008 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145084.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 20:59:31,360 INFO [train.py:903] (0/4) Epoch 22, batch 1700, loss[loss=0.1673, simple_loss=0.2424, pruned_loss=0.04612, over 19773.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2883, pruned_loss=0.06477, over 3809622.91 frames. ], batch size: 47, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 20:59:52,393 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.83 vs. limit=5.0 +2023-04-02 21:00:08,581 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 21:00:28,005 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.426e+02 4.845e+02 6.228e+02 7.810e+02 2.223e+03, threshold=1.246e+03, percent-clipped=2.0 +2023-04-02 21:00:33,039 INFO [train.py:903] (0/4) Epoch 22, batch 1750, loss[loss=0.2148, simple_loss=0.2778, pruned_loss=0.07586, over 19291.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.2887, pruned_loss=0.06508, over 3812831.76 frames. ], batch size: 44, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 21:00:40,308 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145144.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:01:09,158 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145169.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:01:31,687 INFO [train.py:903] (0/4) Epoch 22, batch 1800, loss[loss=0.2117, simple_loss=0.2932, pruned_loss=0.06511, over 19385.00 frames. ], tot_loss[loss=0.2095, simple_loss=0.2889, pruned_loss=0.06502, over 3818735.52 frames. ], batch size: 70, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 21:01:55,238 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1816, 1.3919, 1.7986, 1.7505, 2.9906, 4.5141, 4.4105, 4.9452], + device='cuda:0'), covar=tensor([0.1717, 0.3848, 0.3479, 0.2211, 0.0619, 0.0210, 0.0166, 0.0179], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0321, 0.0352, 0.0265, 0.0243, 0.0186, 0.0216, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 21:02:27,946 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.845e+02 5.086e+02 5.993e+02 7.804e+02 1.410e+03, threshold=1.199e+03, percent-clipped=2.0 +2023-04-02 21:02:27,977 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 21:02:31,524 INFO [train.py:903] (0/4) Epoch 22, batch 1850, loss[loss=0.2587, simple_loss=0.324, pruned_loss=0.09674, over 12937.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2898, pruned_loss=0.06554, over 3825238.45 frames. ], batch size: 137, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 21:03:04,109 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 21:03:30,820 INFO [train.py:903] (0/4) Epoch 22, batch 1900, loss[loss=0.2327, simple_loss=0.3137, pruned_loss=0.07586, over 19410.00 frames. ], tot_loss[loss=0.2105, simple_loss=0.29, pruned_loss=0.06553, over 3820439.06 frames. ], batch size: 70, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 21:03:44,232 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6780, 4.2120, 4.4804, 4.4704, 1.6872, 4.1897, 3.6042, 4.1871], + device='cuda:0'), covar=tensor([0.1687, 0.0899, 0.0618, 0.0629, 0.6226, 0.0914, 0.0693, 0.1117], + device='cuda:0'), in_proj_covar=tensor([0.0775, 0.0737, 0.0934, 0.0825, 0.0825, 0.0698, 0.0561, 0.0872], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 21:03:48,269 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 21:03:52,773 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 21:04:15,285 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 21:04:26,521 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.830e+02 5.286e+02 6.049e+02 6.874e+02 1.450e+03, threshold=1.210e+03, percent-clipped=2.0 +2023-04-02 21:04:30,791 INFO [train.py:903] (0/4) Epoch 22, batch 1950, loss[loss=0.2296, simple_loss=0.3051, pruned_loss=0.07708, over 18752.00 frames. ], tot_loss[loss=0.2107, simple_loss=0.2901, pruned_loss=0.0657, over 3805975.21 frames. ], batch size: 74, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 21:04:44,257 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8139, 1.5591, 1.4549, 1.7931, 1.4913, 1.5951, 1.4741, 1.7212], + device='cuda:0'), covar=tensor([0.1081, 0.1375, 0.1520, 0.0988, 0.1296, 0.0575, 0.1424, 0.0765], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0356, 0.0312, 0.0249, 0.0300, 0.0250, 0.0309, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:04:55,830 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=145358.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:05:31,553 INFO [train.py:903] (0/4) Epoch 22, batch 2000, loss[loss=0.2005, simple_loss=0.2726, pruned_loss=0.06422, over 19374.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2893, pruned_loss=0.06535, over 3810135.21 frames. ], batch size: 48, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 21:05:32,826 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=145389.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:05:32,966 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145389.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:06:19,669 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=145428.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:06:27,594 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.585e+02 4.801e+02 6.060e+02 7.916e+02 1.266e+03, threshold=1.212e+03, percent-clipped=1.0 +2023-04-02 21:06:27,628 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 21:06:30,125 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5449, 1.3119, 1.8590, 1.4847, 2.7007, 3.8534, 3.5579, 4.0600], + device='cuda:0'), covar=tensor([0.1423, 0.3773, 0.3138, 0.2203, 0.0604, 0.0166, 0.0193, 0.0234], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0321, 0.0352, 0.0265, 0.0243, 0.0186, 0.0216, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 21:06:30,885 INFO [train.py:903] (0/4) Epoch 22, batch 2050, loss[loss=0.1975, simple_loss=0.2716, pruned_loss=0.06173, over 19401.00 frames. ], tot_loss[loss=0.2093, simple_loss=0.289, pruned_loss=0.06484, over 3814330.15 frames. ], batch size: 47, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 21:06:46,539 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 21:06:46,573 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 21:07:06,386 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 21:07:13,309 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=145473.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:07:30,908 INFO [train.py:903] (0/4) Epoch 22, batch 2100, loss[loss=0.2173, simple_loss=0.2974, pruned_loss=0.06863, over 19535.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2899, pruned_loss=0.0651, over 3825605.46 frames. ], batch size: 54, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 21:07:51,369 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=145504.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:08:01,438 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 21:08:22,540 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 21:08:27,087 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.088e+02 4.926e+02 6.113e+02 7.931e+02 1.598e+03, threshold=1.223e+03, percent-clipped=5.0 +2023-04-02 21:08:30,642 INFO [train.py:903] (0/4) Epoch 22, batch 2150, loss[loss=0.2493, simple_loss=0.3194, pruned_loss=0.08961, over 13022.00 frames. ], tot_loss[loss=0.2096, simple_loss=0.2895, pruned_loss=0.06484, over 3816667.03 frames. ], batch size: 135, lr: 3.76e-03, grad_scale: 8.0 +2023-04-02 21:08:38,086 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=145543.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:09:32,290 INFO [train.py:903] (0/4) Epoch 22, batch 2200, loss[loss=0.1873, simple_loss=0.2602, pruned_loss=0.05721, over 19758.00 frames. ], tot_loss[loss=0.211, simple_loss=0.2905, pruned_loss=0.06574, over 3824369.72 frames. ], batch size: 46, lr: 3.76e-03, grad_scale: 4.0 +2023-04-02 21:10:00,376 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145612.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:10:08,473 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6941, 1.5535, 1.5349, 2.1807, 1.7317, 2.0684, 2.1241, 1.8363], + device='cuda:0'), covar=tensor([0.0848, 0.0893, 0.1029, 0.0706, 0.0804, 0.0661, 0.0795, 0.0660], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0221, 0.0226, 0.0240, 0.0225, 0.0212, 0.0186, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 21:10:29,863 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.244e+02 5.056e+02 6.179e+02 8.064e+02 2.249e+03, threshold=1.236e+03, percent-clipped=3.0 +2023-04-02 21:10:32,067 INFO [train.py:903] (0/4) Epoch 22, batch 2250, loss[loss=0.2292, simple_loss=0.3114, pruned_loss=0.0735, over 19518.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2893, pruned_loss=0.06459, over 3830453.52 frames. ], batch size: 54, lr: 3.76e-03, grad_scale: 4.0 +2023-04-02 21:11:01,341 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145661.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:11:33,806 INFO [train.py:903] (0/4) Epoch 22, batch 2300, loss[loss=0.206, simple_loss=0.2901, pruned_loss=0.0609, over 19757.00 frames. ], tot_loss[loss=0.2095, simple_loss=0.2895, pruned_loss=0.06476, over 3809060.88 frames. ], batch size: 54, lr: 3.75e-03, grad_scale: 4.0 +2023-04-02 21:11:45,981 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 21:12:22,729 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145729.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:12:27,059 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=145733.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:12:30,302 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.518e+02 4.899e+02 6.109e+02 7.402e+02 2.135e+03, threshold=1.222e+03, percent-clipped=2.0 +2023-04-02 21:12:32,754 INFO [train.py:903] (0/4) Epoch 22, batch 2350, loss[loss=0.2423, simple_loss=0.3171, pruned_loss=0.08376, over 19574.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2888, pruned_loss=0.06481, over 3792304.76 frames. ], batch size: 61, lr: 3.75e-03, grad_scale: 4.0 +2023-04-02 21:12:54,257 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145754.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:12:56,382 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5994, 4.1530, 2.8433, 3.6603, 0.8993, 4.1224, 4.0133, 4.1366], + device='cuda:0'), covar=tensor([0.0680, 0.1126, 0.1668, 0.0858, 0.4198, 0.0674, 0.0934, 0.0973], + device='cuda:0'), in_proj_covar=tensor([0.0502, 0.0410, 0.0493, 0.0345, 0.0403, 0.0432, 0.0426, 0.0460], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:13:01,050 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145760.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:13:07,028 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-02 21:13:14,114 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 21:13:31,578 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 21:13:31,980 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145785.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:13:35,674 INFO [train.py:903] (0/4) Epoch 22, batch 2400, loss[loss=0.2629, simple_loss=0.3365, pruned_loss=0.09463, over 19664.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2887, pruned_loss=0.06476, over 3806211.20 frames. ], batch size: 53, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:13:48,776 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=145799.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:14:20,086 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=145824.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:14:34,018 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.210e+02 4.826e+02 5.747e+02 7.009e+02 1.532e+03, threshold=1.149e+03, percent-clipped=5.0 +2023-04-02 21:14:35,665 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7696, 1.8850, 2.1014, 2.3324, 1.7174, 2.2267, 2.0858, 1.9296], + device='cuda:0'), covar=tensor([0.4134, 0.3870, 0.1899, 0.2290, 0.3913, 0.2130, 0.5084, 0.3458], + device='cuda:0'), in_proj_covar=tensor([0.0893, 0.0957, 0.0714, 0.0929, 0.0873, 0.0811, 0.0840, 0.0780], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 21:14:36,518 INFO [train.py:903] (0/4) Epoch 22, batch 2450, loss[loss=0.2134, simple_loss=0.2979, pruned_loss=0.06443, over 19472.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2884, pruned_loss=0.06444, over 3816316.19 frames. ], batch size: 64, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:14:37,985 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=145839.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 21:14:49,280 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=145848.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:14:52,763 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7404, 1.3217, 1.5458, 1.5093, 3.3304, 1.0590, 2.3464, 3.8084], + device='cuda:0'), covar=tensor([0.0512, 0.2924, 0.2930, 0.1962, 0.0741, 0.2703, 0.1395, 0.0225], + device='cuda:0'), in_proj_covar=tensor([0.0409, 0.0366, 0.0387, 0.0350, 0.0375, 0.0350, 0.0380, 0.0406], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:15:28,950 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3942, 1.4352, 1.7047, 1.6065, 2.4813, 2.1358, 2.6601, 1.0610], + device='cuda:0'), covar=tensor([0.2509, 0.4391, 0.2756, 0.1948, 0.1454, 0.2186, 0.1331, 0.4643], + device='cuda:0'), in_proj_covar=tensor([0.0535, 0.0643, 0.0712, 0.0481, 0.0618, 0.0529, 0.0663, 0.0549], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 21:15:37,464 INFO [train.py:903] (0/4) Epoch 22, batch 2500, loss[loss=0.2199, simple_loss=0.3104, pruned_loss=0.06469, over 19656.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2885, pruned_loss=0.06405, over 3824565.66 frames. ], batch size: 60, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:16:25,710 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6254, 4.0063, 4.2340, 4.2363, 1.6605, 3.9881, 3.5079, 3.9823], + device='cuda:0'), covar=tensor([0.1718, 0.1028, 0.0694, 0.0756, 0.6049, 0.1149, 0.0724, 0.1233], + device='cuda:0'), in_proj_covar=tensor([0.0781, 0.0741, 0.0941, 0.0828, 0.0830, 0.0704, 0.0565, 0.0878], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 21:16:34,429 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.066e+02 4.836e+02 5.791e+02 7.519e+02 1.267e+03, threshold=1.158e+03, percent-clipped=1.0 +2023-04-02 21:16:36,601 INFO [train.py:903] (0/4) Epoch 22, batch 2550, loss[loss=0.2241, simple_loss=0.3149, pruned_loss=0.06663, over 19519.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2887, pruned_loss=0.064, over 3836536.12 frames. ], batch size: 54, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:16:40,550 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.74 vs. limit=5.0 +2023-04-02 21:16:59,313 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=145956.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:17:15,003 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6818, 4.2555, 2.7712, 3.7926, 0.9863, 4.2262, 4.1210, 4.2087], + device='cuda:0'), covar=tensor([0.0570, 0.0897, 0.1911, 0.0860, 0.3922, 0.0646, 0.0815, 0.0978], + device='cuda:0'), in_proj_covar=tensor([0.0501, 0.0407, 0.0489, 0.0342, 0.0399, 0.0427, 0.0422, 0.0456], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:17:33,993 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 21:17:38,147 INFO [train.py:903] (0/4) Epoch 22, batch 2600, loss[loss=0.2192, simple_loss=0.2953, pruned_loss=0.07158, over 19481.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.2886, pruned_loss=0.06406, over 3827654.46 frames. ], batch size: 64, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:17:52,879 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-146000.pt +2023-04-02 21:17:59,427 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=146005.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:18:28,305 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8293, 1.3382, 1.0833, 0.9651, 1.1783, 1.0094, 0.8492, 1.2130], + device='cuda:0'), covar=tensor([0.0660, 0.0817, 0.1148, 0.0727, 0.0561, 0.1283, 0.0669, 0.0512], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0315, 0.0338, 0.0264, 0.0248, 0.0336, 0.0290, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:18:38,079 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.365e+02 5.022e+02 6.231e+02 7.783e+02 1.698e+03, threshold=1.246e+03, percent-clipped=5.0 +2023-04-02 21:18:40,365 INFO [train.py:903] (0/4) Epoch 22, batch 2650, loss[loss=0.2024, simple_loss=0.2899, pruned_loss=0.05746, over 17981.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.288, pruned_loss=0.06408, over 3830291.73 frames. ], batch size: 83, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:19:00,427 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 21:19:21,317 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=146071.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:19:35,523 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-02 21:19:41,307 INFO [train.py:903] (0/4) Epoch 22, batch 2700, loss[loss=0.2257, simple_loss=0.3031, pruned_loss=0.07417, over 19684.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2886, pruned_loss=0.06445, over 3821210.44 frames. ], batch size: 53, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:20:00,899 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=146104.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:20:20,723 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=146120.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:20:32,009 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=146129.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:20:39,332 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.068e+02 4.787e+02 6.288e+02 8.148e+02 2.582e+03, threshold=1.258e+03, percent-clipped=4.0 +2023-04-02 21:20:41,745 INFO [train.py:903] (0/4) Epoch 22, batch 2750, loss[loss=0.1736, simple_loss=0.2654, pruned_loss=0.04088, over 19683.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2886, pruned_loss=0.06428, over 3821244.38 frames. ], batch size: 53, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:21:05,842 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.64 vs. limit=5.0 +2023-04-02 21:21:08,980 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0517, 1.1875, 1.6008, 1.0644, 2.3183, 3.0021, 2.8218, 3.3469], + device='cuda:0'), covar=tensor([0.1790, 0.4735, 0.4404, 0.2541, 0.0660, 0.0276, 0.0306, 0.0336], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0320, 0.0352, 0.0265, 0.0242, 0.0186, 0.0215, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 21:21:18,163 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=146167.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:21:37,421 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=146183.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 21:21:43,817 INFO [train.py:903] (0/4) Epoch 22, batch 2800, loss[loss=0.2432, simple_loss=0.317, pruned_loss=0.08474, over 13308.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2889, pruned_loss=0.06401, over 3807697.77 frames. ], batch size: 136, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:21:47,751 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.04 vs. limit=5.0 +2023-04-02 21:22:42,896 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.980e+02 4.555e+02 5.863e+02 7.335e+02 1.249e+03, threshold=1.173e+03, percent-clipped=1.0 +2023-04-02 21:22:45,118 INFO [train.py:903] (0/4) Epoch 22, batch 2850, loss[loss=0.1846, simple_loss=0.2629, pruned_loss=0.05322, over 18181.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2877, pruned_loss=0.06337, over 3814783.74 frames. ], batch size: 40, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:23:42,916 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 21:23:45,159 INFO [train.py:903] (0/4) Epoch 22, batch 2900, loss[loss=0.2367, simple_loss=0.3085, pruned_loss=0.08252, over 19529.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2875, pruned_loss=0.06353, over 3818971.07 frames. ], batch size: 54, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:23:51,888 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6540, 1.4054, 1.4509, 2.0771, 1.6613, 1.9854, 2.0830, 1.7338], + device='cuda:0'), covar=tensor([0.0876, 0.1046, 0.1086, 0.0830, 0.0947, 0.0830, 0.0867, 0.0739], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0224, 0.0239, 0.0225, 0.0210, 0.0186, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 21:23:57,291 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=146298.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 21:24:33,183 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=146327.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:24:43,684 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.365e+02 4.789e+02 5.854e+02 7.393e+02 1.532e+03, threshold=1.171e+03, percent-clipped=5.0 +2023-04-02 21:24:45,861 INFO [train.py:903] (0/4) Epoch 22, batch 2950, loss[loss=0.2139, simple_loss=0.2997, pruned_loss=0.064, over 19657.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2874, pruned_loss=0.06341, over 3830746.18 frames. ], batch size: 55, lr: 3.75e-03, grad_scale: 8.0 +2023-04-02 21:25:04,171 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=146352.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:25:32,537 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=146376.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:25:33,962 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.92 vs. limit=2.0 +2023-04-02 21:25:46,771 INFO [train.py:903] (0/4) Epoch 22, batch 3000, loss[loss=0.2031, simple_loss=0.2874, pruned_loss=0.05942, over 19680.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2875, pruned_loss=0.06358, over 3825449.57 frames. ], batch size: 60, lr: 3.75e-03, grad_scale: 4.0 +2023-04-02 21:25:46,772 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 21:25:59,186 INFO [train.py:937] (0/4) Epoch 22, validation: loss=0.1687, simple_loss=0.2687, pruned_loss=0.0344, over 944034.00 frames. +2023-04-02 21:25:59,186 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-02 21:26:02,608 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 21:26:15,295 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=146401.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:26:58,610 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.135e+02 5.066e+02 6.686e+02 8.533e+02 1.871e+03, threshold=1.337e+03, percent-clipped=6.0 +2023-04-02 21:26:59,734 INFO [train.py:903] (0/4) Epoch 22, batch 3050, loss[loss=0.178, simple_loss=0.2608, pruned_loss=0.04761, over 19592.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2879, pruned_loss=0.06377, over 3824607.44 frames. ], batch size: 52, lr: 3.74e-03, grad_scale: 4.0 +2023-04-02 21:28:00,901 INFO [train.py:903] (0/4) Epoch 22, batch 3100, loss[loss=0.1884, simple_loss=0.28, pruned_loss=0.04844, over 18069.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2875, pruned_loss=0.06331, over 3836125.30 frames. ], batch size: 83, lr: 3.74e-03, grad_scale: 4.0 +2023-04-02 21:28:15,879 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2577, 1.5299, 2.0315, 1.7144, 3.2267, 4.8014, 4.6465, 5.1736], + device='cuda:0'), covar=tensor([0.1669, 0.3645, 0.3117, 0.2157, 0.0549, 0.0199, 0.0156, 0.0191], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0321, 0.0352, 0.0265, 0.0243, 0.0186, 0.0215, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 21:28:27,613 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=146511.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:28:38,294 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4822, 1.3469, 1.3174, 1.8328, 1.4132, 1.6852, 1.7237, 1.5172], + device='cuda:0'), covar=tensor([0.0799, 0.0910, 0.1015, 0.0604, 0.0775, 0.0723, 0.0771, 0.0685], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0225, 0.0239, 0.0225, 0.0209, 0.0186, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 21:28:59,179 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.441e+02 5.077e+02 6.322e+02 8.082e+02 1.628e+03, threshold=1.264e+03, percent-clipped=2.0 +2023-04-02 21:29:00,360 INFO [train.py:903] (0/4) Epoch 22, batch 3150, loss[loss=0.1848, simple_loss=0.2666, pruned_loss=0.05149, over 19845.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2886, pruned_loss=0.06445, over 3825820.36 frames. ], batch size: 52, lr: 3.74e-03, grad_scale: 4.0 +2023-04-02 21:29:19,882 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=146554.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 21:29:29,190 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 21:29:30,739 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8647, 2.7357, 2.1375, 2.1343, 1.9620, 2.4237, 1.1190, 1.9724], + device='cuda:0'), covar=tensor([0.0684, 0.0626, 0.0695, 0.1084, 0.1065, 0.1024, 0.1370, 0.1062], + device='cuda:0'), in_proj_covar=tensor([0.0358, 0.0357, 0.0359, 0.0384, 0.0460, 0.0388, 0.0338, 0.0343], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 21:29:47,823 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8751, 1.9682, 2.2032, 2.5973, 1.9561, 2.5410, 2.2338, 2.0833], + device='cuda:0'), covar=tensor([0.4432, 0.3750, 0.2002, 0.2250, 0.3975, 0.1974, 0.5005, 0.3365], + device='cuda:0'), in_proj_covar=tensor([0.0897, 0.0961, 0.0717, 0.0930, 0.0878, 0.0814, 0.0844, 0.0781], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 21:29:51,107 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=146579.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 21:30:00,733 INFO [train.py:903] (0/4) Epoch 22, batch 3200, loss[loss=0.2222, simple_loss=0.301, pruned_loss=0.07168, over 19526.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2881, pruned_loss=0.06417, over 3830737.46 frames. ], batch size: 54, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:30:28,099 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=146609.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:30:47,445 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=146626.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:31:01,671 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.447e+02 4.719e+02 5.878e+02 7.469e+02 1.229e+03, threshold=1.176e+03, percent-clipped=0.0 +2023-04-02 21:31:02,527 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 21:31:02,818 INFO [train.py:903] (0/4) Epoch 22, batch 3250, loss[loss=0.2114, simple_loss=0.2972, pruned_loss=0.06278, over 19531.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2873, pruned_loss=0.06387, over 3843201.63 frames. ], batch size: 56, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:31:10,856 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=146644.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:32:03,196 INFO [train.py:903] (0/4) Epoch 22, batch 3300, loss[loss=0.1861, simple_loss=0.2607, pruned_loss=0.05575, over 19307.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2877, pruned_loss=0.06437, over 3823304.61 frames. ], batch size: 44, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:32:08,318 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-02 21:32:09,862 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 21:32:59,758 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.231e+02 5.268e+02 6.638e+02 8.509e+02 1.642e+03, threshold=1.328e+03, percent-clipped=7.0 +2023-04-02 21:33:00,760 INFO [train.py:903] (0/4) Epoch 22, batch 3350, loss[loss=0.2291, simple_loss=0.3039, pruned_loss=0.07711, over 18665.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2886, pruned_loss=0.06475, over 3825747.83 frames. ], batch size: 74, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:34:00,056 INFO [train.py:903] (0/4) Epoch 22, batch 3400, loss[loss=0.2022, simple_loss=0.28, pruned_loss=0.06223, over 19755.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2882, pruned_loss=0.06479, over 3811992.25 frames. ], batch size: 46, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:34:59,708 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.964e+02 5.013e+02 6.159e+02 8.066e+02 2.491e+03, threshold=1.232e+03, percent-clipped=4.0 +2023-04-02 21:35:00,913 INFO [train.py:903] (0/4) Epoch 22, batch 3450, loss[loss=0.2334, simple_loss=0.3111, pruned_loss=0.07785, over 19730.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2882, pruned_loss=0.06465, over 3818861.59 frames. ], batch size: 63, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:35:04,227 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 21:35:28,806 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=146862.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:35:54,044 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=146882.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:36:01,446 INFO [train.py:903] (0/4) Epoch 22, batch 3500, loss[loss=0.1816, simple_loss=0.2657, pruned_loss=0.04871, over 19846.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2876, pruned_loss=0.06418, over 3815604.78 frames. ], batch size: 52, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:36:23,424 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=146907.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:37:00,107 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.850e+02 4.615e+02 6.325e+02 8.235e+02 2.059e+03, threshold=1.265e+03, percent-clipped=6.0 +2023-04-02 21:37:01,338 INFO [train.py:903] (0/4) Epoch 22, batch 3550, loss[loss=0.1704, simple_loss=0.2467, pruned_loss=0.04708, over 19731.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2875, pruned_loss=0.06396, over 3823585.34 frames. ], batch size: 46, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:37:18,263 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=146953.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:37:40,836 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.14 vs. limit=5.0 +2023-04-02 21:38:02,340 INFO [train.py:903] (0/4) Epoch 22, batch 3600, loss[loss=0.195, simple_loss=0.2746, pruned_loss=0.05771, over 19631.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.288, pruned_loss=0.06428, over 3798824.04 frames. ], batch size: 50, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:38:02,532 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=146988.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:39:01,661 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.391e+02 4.948e+02 6.297e+02 8.667e+02 2.605e+03, threshold=1.259e+03, percent-clipped=8.0 +2023-04-02 21:39:02,705 INFO [train.py:903] (0/4) Epoch 22, batch 3650, loss[loss=0.2584, simple_loss=0.3261, pruned_loss=0.0954, over 19614.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2886, pruned_loss=0.06465, over 3802493.14 frames. ], batch size: 57, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:39:39,070 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147068.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:40:03,858 INFO [train.py:903] (0/4) Epoch 22, batch 3700, loss[loss=0.2916, simple_loss=0.3503, pruned_loss=0.1164, over 13341.00 frames. ], tot_loss[loss=0.2104, simple_loss=0.2898, pruned_loss=0.06545, over 3796502.04 frames. ], batch size: 135, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:40:21,179 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147103.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:40:58,545 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5753, 4.1824, 2.6258, 3.6587, 0.9944, 4.1306, 4.0176, 4.1027], + device='cuda:0'), covar=tensor([0.0636, 0.0997, 0.2014, 0.0839, 0.3976, 0.0724, 0.0928, 0.1252], + device='cuda:0'), in_proj_covar=tensor([0.0503, 0.0410, 0.0492, 0.0343, 0.0401, 0.0431, 0.0425, 0.0459], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:41:02,894 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.261e+02 4.816e+02 5.960e+02 7.144e+02 1.653e+03, threshold=1.192e+03, percent-clipped=4.0 +2023-04-02 21:41:04,071 INFO [train.py:903] (0/4) Epoch 22, batch 3750, loss[loss=0.1982, simple_loss=0.2655, pruned_loss=0.06539, over 19755.00 frames. ], tot_loss[loss=0.2108, simple_loss=0.2901, pruned_loss=0.06578, over 3809248.56 frames. ], batch size: 47, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:41:21,574 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4889, 1.4944, 1.7669, 1.7047, 2.6977, 2.3027, 2.8826, 1.3261], + device='cuda:0'), covar=tensor([0.2445, 0.4305, 0.2655, 0.1918, 0.1435, 0.2071, 0.1392, 0.4195], + device='cuda:0'), in_proj_covar=tensor([0.0536, 0.0643, 0.0714, 0.0482, 0.0618, 0.0531, 0.0663, 0.0549], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 21:42:04,510 INFO [train.py:903] (0/4) Epoch 22, batch 3800, loss[loss=0.186, simple_loss=0.2613, pruned_loss=0.05537, over 19765.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.2892, pruned_loss=0.06514, over 3820261.70 frames. ], batch size: 45, lr: 3.74e-03, grad_scale: 8.0 +2023-04-02 21:42:26,697 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=147206.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:42:38,889 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-02 21:43:02,579 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.867e+02 5.077e+02 5.970e+02 7.548e+02 1.289e+03, threshold=1.194e+03, percent-clipped=1.0 +2023-04-02 21:43:03,538 INFO [train.py:903] (0/4) Epoch 22, batch 3850, loss[loss=0.2051, simple_loss=0.2851, pruned_loss=0.0625, over 19678.00 frames. ], tot_loss[loss=0.2097, simple_loss=0.2891, pruned_loss=0.06511, over 3815976.38 frames. ], batch size: 53, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:43:09,367 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0219, 1.9606, 1.5093, 1.9630, 1.8472, 1.5072, 1.4284, 1.8264], + device='cuda:0'), covar=tensor([0.1188, 0.1612, 0.1851, 0.1217, 0.1571, 0.0832, 0.1868, 0.0927], + device='cuda:0'), in_proj_covar=tensor([0.0269, 0.0354, 0.0309, 0.0249, 0.0301, 0.0251, 0.0308, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:43:31,162 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147259.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:44:05,365 INFO [train.py:903] (0/4) Epoch 22, batch 3900, loss[loss=0.2168, simple_loss=0.2977, pruned_loss=0.06801, over 19747.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.2891, pruned_loss=0.06522, over 3816113.63 frames. ], batch size: 54, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:44:28,436 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.47 vs. limit=5.0 +2023-04-02 21:44:41,083 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8856, 1.7936, 2.0373, 1.6621, 4.4131, 1.1029, 2.5408, 4.8303], + device='cuda:0'), covar=tensor([0.0407, 0.2658, 0.2669, 0.2035, 0.0754, 0.2646, 0.1489, 0.0158], + device='cuda:0'), in_proj_covar=tensor([0.0406, 0.0364, 0.0385, 0.0348, 0.0372, 0.0347, 0.0379, 0.0402], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:44:46,192 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147321.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:44:49,772 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147324.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:44:55,095 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147328.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:45:04,849 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.696e+02 4.890e+02 6.799e+02 8.609e+02 1.784e+03, threshold=1.360e+03, percent-clipped=9.0 +2023-04-02 21:45:05,887 INFO [train.py:903] (0/4) Epoch 22, batch 3950, loss[loss=0.1794, simple_loss=0.2569, pruned_loss=0.05094, over 16459.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2881, pruned_loss=0.06467, over 3810167.25 frames. ], batch size: 36, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:45:08,145 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-02 21:45:18,206 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147349.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:45:30,127 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147359.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:46:00,798 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147384.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:46:04,829 INFO [train.py:903] (0/4) Epoch 22, batch 4000, loss[loss=0.2056, simple_loss=0.2831, pruned_loss=0.06405, over 19621.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2872, pruned_loss=0.0641, over 3827208.94 frames. ], batch size: 50, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:46:50,141 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147425.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:46:52,234 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-02 21:46:54,077 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-02 21:47:03,821 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.354e+02 4.920e+02 5.925e+02 7.748e+02 1.160e+03, threshold=1.185e+03, percent-clipped=0.0 +2023-04-02 21:47:05,864 INFO [train.py:903] (0/4) Epoch 22, batch 4050, loss[loss=0.2085, simple_loss=0.2798, pruned_loss=0.06855, over 19484.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2877, pruned_loss=0.06418, over 3829945.69 frames. ], batch size: 49, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:47:24,647 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147452.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:47:45,016 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-02 21:48:07,528 INFO [train.py:903] (0/4) Epoch 22, batch 4100, loss[loss=0.2709, simple_loss=0.3297, pruned_loss=0.106, over 19585.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2882, pruned_loss=0.06498, over 3813879.32 frames. ], batch size: 61, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:48:44,884 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-02 21:49:07,919 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.284e+02 5.254e+02 6.349e+02 7.493e+02 1.711e+03, threshold=1.270e+03, percent-clipped=4.0 +2023-04-02 21:49:09,115 INFO [train.py:903] (0/4) Epoch 22, batch 4150, loss[loss=0.2198, simple_loss=0.3046, pruned_loss=0.06746, over 19733.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2879, pruned_loss=0.06494, over 3816424.83 frames. ], batch size: 63, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:49:56,959 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147577.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:50:09,104 INFO [train.py:903] (0/4) Epoch 22, batch 4200, loss[loss=0.2247, simple_loss=0.3051, pruned_loss=0.07216, over 19687.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2867, pruned_loss=0.06424, over 3817162.33 frames. ], batch size: 59, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:50:13,785 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-02 21:50:26,564 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147602.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:50:27,413 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=147603.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:51:09,770 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.286e+02 4.933e+02 5.671e+02 7.351e+02 2.024e+03, threshold=1.134e+03, percent-clipped=5.0 +2023-04-02 21:51:10,932 INFO [train.py:903] (0/4) Epoch 22, batch 4250, loss[loss=0.2014, simple_loss=0.2875, pruned_loss=0.05763, over 19775.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2861, pruned_loss=0.06404, over 3815995.38 frames. ], batch size: 56, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:51:25,901 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-02 21:51:38,183 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-02 21:51:51,755 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=147672.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:51:57,369 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147677.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:52:11,224 INFO [train.py:903] (0/4) Epoch 22, batch 4300, loss[loss=0.1871, simple_loss=0.2748, pruned_loss=0.04971, over 19682.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2847, pruned_loss=0.06332, over 3825563.88 frames. ], batch size: 58, lr: 3.73e-03, grad_scale: 4.0 +2023-04-02 21:52:47,204 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147718.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:53:02,952 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-02 21:53:11,621 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.973e+02 5.055e+02 6.219e+02 8.115e+02 2.735e+03, threshold=1.244e+03, percent-clipped=11.0 +2023-04-02 21:53:11,639 INFO [train.py:903] (0/4) Epoch 22, batch 4350, loss[loss=0.2259, simple_loss=0.3013, pruned_loss=0.07528, over 19672.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2851, pruned_loss=0.06349, over 3846619.75 frames. ], batch size: 55, lr: 3.73e-03, grad_scale: 4.0 +2023-04-02 21:53:48,686 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=147769.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:54:11,180 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147787.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:54:12,000 INFO [train.py:903] (0/4) Epoch 22, batch 4400, loss[loss=0.2769, simple_loss=0.3433, pruned_loss=0.1053, over 12685.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2856, pruned_loss=0.06336, over 3837920.19 frames. ], batch size: 136, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:54:20,676 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=147796.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:54:37,134 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-02 21:54:46,704 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-02 21:55:06,292 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147833.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:55:12,509 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.305e+02 5.137e+02 6.374e+02 7.743e+02 1.534e+03, threshold=1.275e+03, percent-clipped=3.0 +2023-04-02 21:55:12,527 INFO [train.py:903] (0/4) Epoch 22, batch 4450, loss[loss=0.2174, simple_loss=0.3008, pruned_loss=0.06696, over 19517.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2864, pruned_loss=0.06367, over 3835867.67 frames. ], batch size: 64, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:55:57,419 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147875.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:55:57,452 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1339, 1.2887, 1.5361, 1.3225, 2.7582, 1.0064, 2.0831, 3.1084], + device='cuda:0'), covar=tensor([0.0638, 0.2883, 0.2872, 0.1996, 0.0797, 0.2589, 0.1396, 0.0340], + device='cuda:0'), in_proj_covar=tensor([0.0408, 0.0366, 0.0387, 0.0348, 0.0373, 0.0349, 0.0382, 0.0404], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:56:08,753 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147884.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:56:12,927 INFO [train.py:903] (0/4) Epoch 22, batch 4500, loss[loss=0.217, simple_loss=0.2987, pruned_loss=0.06766, over 19788.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2881, pruned_loss=0.06459, over 3830422.75 frames. ], batch size: 56, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:56:41,878 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=147911.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:57:03,777 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=147929.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:57:15,408 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.490e+02 4.535e+02 5.624e+02 7.235e+02 1.683e+03, threshold=1.125e+03, percent-clipped=3.0 +2023-04-02 21:57:15,427 INFO [train.py:903] (0/4) Epoch 22, batch 4550, loss[loss=0.2018, simple_loss=0.2947, pruned_loss=0.05451, over 19774.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2883, pruned_loss=0.06457, over 3840464.09 frames. ], batch size: 54, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:57:23,458 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-02 21:57:46,092 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-02 21:57:58,525 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=147974.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:58:15,826 INFO [train.py:903] (0/4) Epoch 22, batch 4600, loss[loss=0.2547, simple_loss=0.3268, pruned_loss=0.09125, over 19487.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2886, pruned_loss=0.06463, over 3841638.08 frames. ], batch size: 64, lr: 3.73e-03, grad_scale: 8.0 +2023-04-02 21:58:28,667 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=147999.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:58:29,490 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-148000.pt +2023-04-02 21:58:56,656 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=148021.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:59:04,092 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-02 21:59:12,790 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6266, 1.3678, 1.5908, 1.4684, 3.2411, 1.0945, 2.4235, 3.6233], + device='cuda:0'), covar=tensor([0.0522, 0.2672, 0.2775, 0.1864, 0.0717, 0.2478, 0.1163, 0.0257], + device='cuda:0'), in_proj_covar=tensor([0.0408, 0.0366, 0.0387, 0.0347, 0.0374, 0.0348, 0.0381, 0.0403], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 21:59:16,086 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.327e+02 4.857e+02 6.032e+02 8.227e+02 1.754e+03, threshold=1.206e+03, percent-clipped=4.0 +2023-04-02 21:59:16,108 INFO [train.py:903] (0/4) Epoch 22, batch 4650, loss[loss=0.2379, simple_loss=0.3115, pruned_loss=0.08213, over 13632.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2883, pruned_loss=0.06452, over 3842029.40 frames. ], batch size: 135, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 21:59:22,644 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148043.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 21:59:32,295 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-02 21:59:43,962 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-02 21:59:53,347 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148068.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:00:16,030 INFO [train.py:903] (0/4) Epoch 22, batch 4700, loss[loss=0.1899, simple_loss=0.2696, pruned_loss=0.05511, over 19829.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2879, pruned_loss=0.064, over 3840038.48 frames. ], batch size: 52, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:00:39,956 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-02 22:01:15,694 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148136.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:01:17,614 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.017e+02 5.448e+02 6.322e+02 7.572e+02 1.580e+03, threshold=1.264e+03, percent-clipped=4.0 +2023-04-02 22:01:17,634 INFO [train.py:903] (0/4) Epoch 22, batch 4750, loss[loss=0.2177, simple_loss=0.3021, pruned_loss=0.06666, over 19362.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2884, pruned_loss=0.06443, over 3822968.00 frames. ], batch size: 70, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:01:20,378 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148140.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:01:34,514 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-02 22:01:50,443 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148165.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:01:52,769 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148167.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:02:05,545 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=148177.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:02:06,208 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-04-02 22:02:18,760 INFO [train.py:903] (0/4) Epoch 22, batch 4800, loss[loss=0.2221, simple_loss=0.2995, pruned_loss=0.0724, over 19304.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2882, pruned_loss=0.06407, over 3823438.89 frames. ], batch size: 66, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:02:23,650 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148192.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:02:56,872 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=148219.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:03:18,986 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.136e+02 4.799e+02 5.785e+02 7.279e+02 1.291e+03, threshold=1.157e+03, percent-clipped=1.0 +2023-04-02 22:03:19,005 INFO [train.py:903] (0/4) Epoch 22, batch 4850, loss[loss=0.2262, simple_loss=0.3059, pruned_loss=0.0732, over 19584.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2878, pruned_loss=0.06424, over 3831935.98 frames. ], batch size: 61, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:03:44,130 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-02 22:04:01,907 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=148273.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:04:02,912 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-02 22:04:08,108 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-02 22:04:09,257 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-02 22:04:18,711 INFO [train.py:903] (0/4) Epoch 22, batch 4900, loss[loss=0.2171, simple_loss=0.2831, pruned_loss=0.07551, over 19603.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2886, pruned_loss=0.06492, over 3814525.24 frames. ], batch size: 50, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:04:18,727 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-02 22:04:24,279 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148292.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:04:39,207 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-02 22:04:40,586 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148305.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:05:14,451 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148334.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:05:19,520 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.031e+02 5.396e+02 6.572e+02 8.433e+02 1.736e+03, threshold=1.314e+03, percent-clipped=6.0 +2023-04-02 22:05:19,538 INFO [train.py:903] (0/4) Epoch 22, batch 4950, loss[loss=0.1995, simple_loss=0.2851, pruned_loss=0.05695, over 17428.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2894, pruned_loss=0.06525, over 3804102.54 frames. ], batch size: 101, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:05:26,429 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.33 vs. limit=5.0 +2023-04-02 22:05:35,928 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-02 22:06:01,175 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-02 22:06:20,895 INFO [train.py:903] (0/4) Epoch 22, batch 5000, loss[loss=0.2249, simple_loss=0.3029, pruned_loss=0.07342, over 19589.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2879, pruned_loss=0.06438, over 3820179.45 frames. ], batch size: 61, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:06:21,260 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148388.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:06:25,704 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148392.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:06:29,627 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-02 22:06:40,548 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-02 22:06:55,431 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148417.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:07:19,244 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.597e+02 4.828e+02 6.337e+02 7.923e+02 1.456e+03, threshold=1.267e+03, percent-clipped=1.0 +2023-04-02 22:07:19,262 INFO [train.py:903] (0/4) Epoch 22, batch 5050, loss[loss=0.2419, simple_loss=0.3176, pruned_loss=0.08313, over 19462.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2885, pruned_loss=0.06477, over 3812776.86 frames. ], batch size: 64, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:07:30,037 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148447.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:07:46,919 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3533, 3.0640, 2.2439, 2.7554, 0.9096, 3.0408, 2.9155, 3.0102], + device='cuda:0'), covar=tensor([0.1126, 0.1371, 0.2045, 0.1125, 0.3755, 0.0962, 0.1167, 0.1322], + device='cuda:0'), in_proj_covar=tensor([0.0502, 0.0409, 0.0492, 0.0344, 0.0398, 0.0432, 0.0424, 0.0458], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:07:54,473 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-02 22:08:19,394 INFO [train.py:903] (0/4) Epoch 22, batch 5100, loss[loss=0.2255, simple_loss=0.3078, pruned_loss=0.07155, over 19581.00 frames. ], tot_loss[loss=0.2093, simple_loss=0.2892, pruned_loss=0.0647, over 3819429.11 frames. ], batch size: 61, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:08:21,057 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-02 22:08:30,458 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-02 22:08:33,789 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-02 22:08:39,193 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-02 22:08:53,279 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148516.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:09:19,535 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.928e+02 5.468e+02 6.941e+02 9.893e+02 2.948e+03, threshold=1.388e+03, percent-clipped=12.0 +2023-04-02 22:09:19,553 INFO [train.py:903] (0/4) Epoch 22, batch 5150, loss[loss=0.1721, simple_loss=0.2579, pruned_loss=0.04313, over 19671.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2898, pruned_loss=0.06503, over 3815440.30 frames. ], batch size: 53, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:09:31,376 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-02 22:09:32,953 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148548.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:09:40,842 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-02 22:10:02,797 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148573.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:10:05,888 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 22:10:20,789 INFO [train.py:903] (0/4) Epoch 22, batch 5200, loss[loss=0.1751, simple_loss=0.2619, pruned_loss=0.04414, over 19861.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2886, pruned_loss=0.06468, over 3812700.07 frames. ], batch size: 52, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:10:23,562 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148590.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:10:33,163 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-02 22:10:53,687 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148615.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:10:58,724 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.69 vs. limit=2.0 +2023-04-02 22:11:17,526 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-02 22:11:21,000 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.230e+02 4.558e+02 5.776e+02 7.251e+02 2.001e+03, threshold=1.155e+03, percent-clipped=2.0 +2023-04-02 22:11:21,018 INFO [train.py:903] (0/4) Epoch 22, batch 5250, loss[loss=0.1987, simple_loss=0.2748, pruned_loss=0.06128, over 19564.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2887, pruned_loss=0.0648, over 3795550.67 frames. ], batch size: 52, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:11:27,758 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1350, 1.3893, 1.4883, 1.5148, 2.7691, 1.2176, 2.1464, 3.0935], + device='cuda:0'), covar=tensor([0.0529, 0.2500, 0.2840, 0.1693, 0.0702, 0.2198, 0.1231, 0.0313], + device='cuda:0'), in_proj_covar=tensor([0.0410, 0.0367, 0.0387, 0.0348, 0.0375, 0.0351, 0.0384, 0.0405], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:11:27,866 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=148644.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:11:33,739 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=148649.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:11:35,097 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3223, 1.4416, 1.9895, 1.4047, 2.7774, 3.7514, 3.4790, 3.9130], + device='cuda:0'), covar=tensor([0.1578, 0.3606, 0.2983, 0.2344, 0.0555, 0.0168, 0.0186, 0.0233], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0326, 0.0357, 0.0268, 0.0248, 0.0190, 0.0218, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 22:11:58,275 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148669.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:11:58,383 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=148669.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:12:20,268 INFO [train.py:903] (0/4) Epoch 22, batch 5300, loss[loss=0.2059, simple_loss=0.3006, pruned_loss=0.05559, over 19681.00 frames. ], tot_loss[loss=0.2098, simple_loss=0.2898, pruned_loss=0.06485, over 3808249.96 frames. ], batch size: 55, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:12:39,136 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-02 22:13:17,781 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=148734.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:13:22,173 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.679e+02 5.309e+02 6.457e+02 8.011e+02 2.116e+03, threshold=1.291e+03, percent-clipped=5.0 +2023-04-02 22:13:22,191 INFO [train.py:903] (0/4) Epoch 22, batch 5350, loss[loss=0.1981, simple_loss=0.2779, pruned_loss=0.05918, over 19496.00 frames. ], tot_loss[loss=0.2109, simple_loss=0.2909, pruned_loss=0.06547, over 3791030.31 frames. ], batch size: 49, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:13:47,456 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9469, 2.0010, 2.3200, 2.6329, 1.9576, 2.4807, 2.3655, 2.0819], + device='cuda:0'), covar=tensor([0.4248, 0.4051, 0.1934, 0.2307, 0.4058, 0.2163, 0.4627, 0.3450], + device='cuda:0'), in_proj_covar=tensor([0.0897, 0.0961, 0.0716, 0.0929, 0.0877, 0.0814, 0.0840, 0.0780], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 22:13:50,697 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0444, 1.7871, 2.1874, 2.1026, 4.4592, 1.6318, 2.9276, 4.9405], + device='cuda:0'), covar=tensor([0.0449, 0.2893, 0.2456, 0.1863, 0.0732, 0.2402, 0.1254, 0.0170], + device='cuda:0'), in_proj_covar=tensor([0.0411, 0.0368, 0.0387, 0.0347, 0.0375, 0.0350, 0.0383, 0.0405], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:13:53,897 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148764.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:13:55,486 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-02 22:14:03,701 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1843, 1.4834, 1.6406, 1.4112, 2.8306, 1.0372, 2.2786, 3.2255], + device='cuda:0'), covar=tensor([0.0453, 0.2304, 0.2353, 0.1723, 0.0645, 0.2284, 0.1027, 0.0230], + device='cuda:0'), in_proj_covar=tensor([0.0412, 0.0369, 0.0388, 0.0348, 0.0376, 0.0351, 0.0384, 0.0406], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:14:24,407 INFO [train.py:903] (0/4) Epoch 22, batch 5400, loss[loss=0.2078, simple_loss=0.2936, pruned_loss=0.06104, over 19289.00 frames. ], tot_loss[loss=0.2099, simple_loss=0.2898, pruned_loss=0.065, over 3789610.71 frames. ], batch size: 66, lr: 3.72e-03, grad_scale: 8.0 +2023-04-02 22:14:28,071 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=148791.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:15:24,091 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.599e+02 4.882e+02 5.931e+02 8.184e+02 2.288e+03, threshold=1.186e+03, percent-clipped=7.0 +2023-04-02 22:15:24,109 INFO [train.py:903] (0/4) Epoch 22, batch 5450, loss[loss=0.2044, simple_loss=0.2872, pruned_loss=0.06079, over 19661.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2897, pruned_loss=0.06509, over 3800090.11 frames. ], batch size: 58, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:15:50,186 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=148860.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:16:24,044 INFO [train.py:903] (0/4) Epoch 22, batch 5500, loss[loss=0.2051, simple_loss=0.2936, pruned_loss=0.05831, over 19545.00 frames. ], tot_loss[loss=0.2111, simple_loss=0.2907, pruned_loss=0.06577, over 3795966.78 frames. ], batch size: 56, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:16:47,559 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148906.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:16:49,405 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-02 22:16:50,917 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9356, 1.2028, 1.4703, 0.5596, 2.0222, 2.4199, 2.1509, 2.5835], + device='cuda:0'), covar=tensor([0.1654, 0.3873, 0.3581, 0.2895, 0.0629, 0.0285, 0.0342, 0.0363], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0324, 0.0356, 0.0266, 0.0247, 0.0189, 0.0216, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 22:17:25,271 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.265e+02 4.956e+02 6.031e+02 8.410e+02 1.981e+03, threshold=1.206e+03, percent-clipped=11.0 +2023-04-02 22:17:25,290 INFO [train.py:903] (0/4) Epoch 22, batch 5550, loss[loss=0.1705, simple_loss=0.247, pruned_loss=0.04701, over 17280.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2887, pruned_loss=0.06437, over 3800210.93 frames. ], batch size: 38, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:17:33,836 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-02 22:17:57,713 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5542, 1.2218, 1.3704, 1.1573, 2.1949, 0.9817, 2.1012, 2.5075], + device='cuda:0'), covar=tensor([0.0772, 0.2798, 0.2999, 0.1893, 0.0873, 0.2261, 0.1059, 0.0467], + device='cuda:0'), in_proj_covar=tensor([0.0409, 0.0367, 0.0386, 0.0347, 0.0375, 0.0351, 0.0382, 0.0406], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:18:10,822 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=148975.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:18:21,581 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-02 22:18:27,007 INFO [train.py:903] (0/4) Epoch 22, batch 5600, loss[loss=0.1709, simple_loss=0.252, pruned_loss=0.04489, over 19615.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2893, pruned_loss=0.06456, over 3813113.04 frames. ], batch size: 50, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:18:56,760 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=149013.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:19:06,015 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=149020.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:19:10,575 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-02 22:19:27,601 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.511e+02 5.096e+02 6.059e+02 8.103e+02 1.757e+03, threshold=1.212e+03, percent-clipped=10.0 +2023-04-02 22:19:27,619 INFO [train.py:903] (0/4) Epoch 22, batch 5650, loss[loss=0.1974, simple_loss=0.2786, pruned_loss=0.0581, over 19592.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.2889, pruned_loss=0.06399, over 3824839.92 frames. ], batch size: 50, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:19:30,322 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2676, 2.9917, 2.3491, 2.2879, 2.2354, 2.6144, 1.2194, 2.1296], + device='cuda:0'), covar=tensor([0.0624, 0.0601, 0.0706, 0.1172, 0.1076, 0.1036, 0.1382, 0.1073], + device='cuda:0'), in_proj_covar=tensor([0.0356, 0.0357, 0.0359, 0.0382, 0.0462, 0.0390, 0.0337, 0.0341], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 22:19:35,857 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=149045.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:20:15,158 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-02 22:20:16,290 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=149078.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:20:28,281 INFO [train.py:903] (0/4) Epoch 22, batch 5700, loss[loss=0.2011, simple_loss=0.2735, pruned_loss=0.06433, over 19366.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2871, pruned_loss=0.06307, over 3836151.37 frames. ], batch size: 47, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:21:17,800 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=149128.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:21:29,596 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.928e+02 4.618e+02 6.077e+02 7.635e+02 1.470e+03, threshold=1.215e+03, percent-clipped=6.0 +2023-04-02 22:21:29,614 INFO [train.py:903] (0/4) Epoch 22, batch 5750, loss[loss=0.1905, simple_loss=0.2642, pruned_loss=0.05836, over 19775.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2883, pruned_loss=0.06314, over 3848412.68 frames. ], batch size: 47, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:21:30,799 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-02 22:21:39,631 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-02 22:21:46,332 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-02 22:21:59,230 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=149162.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:22:03,954 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0454, 2.0942, 2.3682, 2.7032, 2.0018, 2.6027, 2.3720, 2.1687], + device='cuda:0'), covar=tensor([0.4234, 0.3837, 0.1977, 0.2404, 0.4190, 0.2085, 0.5006, 0.3419], + device='cuda:0'), in_proj_covar=tensor([0.0897, 0.0962, 0.0718, 0.0933, 0.0880, 0.0814, 0.0841, 0.0781], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 22:22:29,799 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=149187.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:22:30,643 INFO [train.py:903] (0/4) Epoch 22, batch 5800, loss[loss=0.2715, simple_loss=0.3269, pruned_loss=0.1081, over 13724.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2874, pruned_loss=0.0632, over 3850700.44 frames. ], batch size: 135, lr: 3.71e-03, grad_scale: 4.0 +2023-04-02 22:22:37,196 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=149193.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:23:23,057 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=149231.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:23:30,410 INFO [train.py:903] (0/4) Epoch 22, batch 5850, loss[loss=0.2153, simple_loss=0.2995, pruned_loss=0.06556, over 19601.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.2884, pruned_loss=0.06393, over 3856241.58 frames. ], batch size: 61, lr: 3.71e-03, grad_scale: 4.0 +2023-04-02 22:23:31,585 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.414e+02 5.174e+02 6.346e+02 7.936e+02 1.645e+03, threshold=1.269e+03, percent-clipped=7.0 +2023-04-02 22:23:39,918 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4464, 1.2886, 1.2291, 1.4628, 1.2410, 1.2577, 1.2248, 1.3688], + device='cuda:0'), covar=tensor([0.0846, 0.1131, 0.1173, 0.0830, 0.1041, 0.0494, 0.1151, 0.0673], + device='cuda:0'), in_proj_covar=tensor([0.0268, 0.0354, 0.0312, 0.0250, 0.0300, 0.0250, 0.0309, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:23:52,890 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=149256.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:24:19,376 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6655, 1.5218, 1.5648, 2.0934, 1.5886, 1.9772, 2.1038, 1.7440], + device='cuda:0'), covar=tensor([0.0882, 0.0957, 0.1001, 0.0801, 0.0900, 0.0763, 0.0785, 0.0696], + device='cuda:0'), in_proj_covar=tensor([0.0214, 0.0223, 0.0226, 0.0241, 0.0229, 0.0213, 0.0187, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 22:24:30,285 INFO [train.py:903] (0/4) Epoch 22, batch 5900, loss[loss=0.1859, simple_loss=0.2596, pruned_loss=0.05607, over 19477.00 frames. ], tot_loss[loss=0.2093, simple_loss=0.2891, pruned_loss=0.06472, over 3840691.18 frames. ], batch size: 49, lr: 3.71e-03, grad_scale: 4.0 +2023-04-02 22:24:35,575 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-02 22:24:56,388 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-02 22:25:31,743 INFO [train.py:903] (0/4) Epoch 22, batch 5950, loss[loss=0.1947, simple_loss=0.2665, pruned_loss=0.06142, over 19617.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.289, pruned_loss=0.06484, over 3824734.15 frames. ], batch size: 50, lr: 3.71e-03, grad_scale: 4.0 +2023-04-02 22:25:32,887 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.792e+02 4.957e+02 5.985e+02 7.132e+02 1.534e+03, threshold=1.197e+03, percent-clipped=1.0 +2023-04-02 22:26:26,169 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 22:26:28,453 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=149384.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:26:33,222 INFO [train.py:903] (0/4) Epoch 22, batch 6000, loss[loss=0.1919, simple_loss=0.2728, pruned_loss=0.05547, over 19496.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2884, pruned_loss=0.06455, over 3831133.68 frames. ], batch size: 49, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:26:33,223 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 22:26:40,161 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4683, 1.4825, 1.5033, 1.8444, 1.4859, 1.6312, 1.6326, 1.6323], + device='cuda:0'), covar=tensor([0.0916, 0.0986, 0.1008, 0.0643, 0.0946, 0.0900, 0.0975, 0.0714], + device='cuda:0'), in_proj_covar=tensor([0.0214, 0.0223, 0.0226, 0.0241, 0.0229, 0.0214, 0.0188, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 22:26:43,625 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2040, 3.7851, 3.8501, 3.8621, 1.7284, 3.6039, 3.2938, 3.5890], + device='cuda:0'), covar=tensor([0.1810, 0.0690, 0.0686, 0.0708, 0.6094, 0.0945, 0.0719, 0.1230], + device='cuda:0'), in_proj_covar=tensor([0.0783, 0.0743, 0.0949, 0.0833, 0.0835, 0.0714, 0.0566, 0.0883], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 22:26:46,899 INFO [train.py:937] (0/4) Epoch 22, validation: loss=0.1681, simple_loss=0.2682, pruned_loss=0.03398, over 944034.00 frames. +2023-04-02 22:26:46,901 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-02 22:27:13,614 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=149409.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:27:48,665 INFO [train.py:903] (0/4) Epoch 22, batch 6050, loss[loss=0.1584, simple_loss=0.2452, pruned_loss=0.03578, over 19387.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.288, pruned_loss=0.06389, over 3841651.74 frames. ], batch size: 48, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:27:49,810 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.518e+02 4.880e+02 5.766e+02 7.280e+02 1.810e+03, threshold=1.153e+03, percent-clipped=3.0 +2023-04-02 22:27:52,759 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.63 vs. limit=5.0 +2023-04-02 22:28:02,640 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=149449.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:28:07,331 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3850, 3.6592, 2.1723, 2.2388, 3.3296, 1.9944, 1.6955, 2.3460], + device='cuda:0'), covar=tensor([0.1303, 0.0650, 0.1101, 0.0893, 0.0523, 0.1199, 0.0971, 0.0762], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0318, 0.0340, 0.0266, 0.0248, 0.0338, 0.0291, 0.0276], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:28:32,559 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=149474.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:28:50,322 INFO [train.py:903] (0/4) Epoch 22, batch 6100, loss[loss=0.2008, simple_loss=0.2868, pruned_loss=0.05745, over 18732.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.287, pruned_loss=0.06332, over 3838383.29 frames. ], batch size: 74, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:28:59,759 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8033, 1.7507, 1.7202, 1.6153, 1.5086, 1.6299, 0.7954, 1.2334], + device='cuda:0'), covar=tensor([0.0635, 0.0659, 0.0390, 0.0630, 0.0983, 0.0865, 0.1261, 0.0918], + device='cuda:0'), in_proj_covar=tensor([0.0356, 0.0355, 0.0358, 0.0381, 0.0462, 0.0389, 0.0336, 0.0341], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 22:29:02,457 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.59 vs. limit=5.0 +2023-04-02 22:29:42,756 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=149532.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:29:48,971 INFO [train.py:903] (0/4) Epoch 22, batch 6150, loss[loss=0.2178, simple_loss=0.2811, pruned_loss=0.07728, over 19358.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.287, pruned_loss=0.06371, over 3833799.32 frames. ], batch size: 44, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:29:49,331 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4369, 1.3985, 1.3423, 1.8339, 1.3448, 1.5936, 1.7042, 1.5142], + device='cuda:0'), covar=tensor([0.0943, 0.0982, 0.1099, 0.0648, 0.0889, 0.0813, 0.0810, 0.0776], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0222, 0.0224, 0.0240, 0.0227, 0.0212, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 22:29:50,016 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.191e+02 4.761e+02 6.063e+02 7.648e+02 1.908e+03, threshold=1.213e+03, percent-clipped=8.0 +2023-04-02 22:29:50,445 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2078, 2.0394, 1.7941, 2.1774, 1.9031, 1.8822, 1.7966, 2.0711], + device='cuda:0'), covar=tensor([0.0976, 0.1430, 0.1555, 0.1027, 0.1408, 0.0569, 0.1404, 0.0730], + device='cuda:0'), in_proj_covar=tensor([0.0269, 0.0354, 0.0312, 0.0250, 0.0301, 0.0250, 0.0309, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:30:15,388 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0484, 5.1019, 5.8638, 5.8588, 2.0309, 5.5683, 4.6453, 5.5020], + device='cuda:0'), covar=tensor([0.1642, 0.0874, 0.0567, 0.0576, 0.6040, 0.0768, 0.0626, 0.1196], + device='cuda:0'), in_proj_covar=tensor([0.0779, 0.0740, 0.0943, 0.0827, 0.0831, 0.0708, 0.0563, 0.0877], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 22:30:19,811 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-02 22:30:29,462 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.86 vs. limit=5.0 +2023-04-02 22:30:43,268 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=149583.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:30:49,418 INFO [train.py:903] (0/4) Epoch 22, batch 6200, loss[loss=0.221, simple_loss=0.3056, pruned_loss=0.06821, over 19473.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2871, pruned_loss=0.06364, over 3840102.63 frames. ], batch size: 64, lr: 3.71e-03, grad_scale: 8.0 +2023-04-02 22:31:51,302 INFO [train.py:903] (0/4) Epoch 22, batch 6250, loss[loss=0.1712, simple_loss=0.2607, pruned_loss=0.04086, over 19773.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2864, pruned_loss=0.06318, over 3842702.41 frames. ], batch size: 54, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:31:52,391 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.896e+02 5.068e+02 6.178e+02 8.268e+02 1.694e+03, threshold=1.236e+03, percent-clipped=5.0 +2023-04-02 22:32:21,865 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-02 22:32:52,473 INFO [train.py:903] (0/4) Epoch 22, batch 6300, loss[loss=0.2166, simple_loss=0.298, pruned_loss=0.06757, over 19368.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.2867, pruned_loss=0.06319, over 3840153.47 frames. ], batch size: 70, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:32:59,392 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6731, 1.6053, 1.5015, 2.1095, 1.5280, 1.9773, 2.0156, 1.7661], + device='cuda:0'), covar=tensor([0.0846, 0.0918, 0.1028, 0.0719, 0.0875, 0.0720, 0.0831, 0.0679], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0220, 0.0222, 0.0238, 0.0226, 0.0211, 0.0186, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 22:33:51,829 INFO [train.py:903] (0/4) Epoch 22, batch 6350, loss[loss=0.2352, simple_loss=0.3194, pruned_loss=0.07548, over 19297.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2856, pruned_loss=0.06268, over 3844086.04 frames. ], batch size: 66, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:33:52,934 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.101e+02 5.551e+02 6.531e+02 8.044e+02 1.579e+03, threshold=1.306e+03, percent-clipped=6.0 +2023-04-02 22:34:23,929 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=149764.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 22:34:52,496 INFO [train.py:903] (0/4) Epoch 22, batch 6400, loss[loss=0.2414, simple_loss=0.3152, pruned_loss=0.08377, over 19765.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2872, pruned_loss=0.06321, over 3846638.89 frames. ], batch size: 63, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:35:54,292 INFO [train.py:903] (0/4) Epoch 22, batch 6450, loss[loss=0.1891, simple_loss=0.264, pruned_loss=0.0571, over 19400.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2877, pruned_loss=0.06361, over 3836205.18 frames. ], batch size: 48, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:35:55,267 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.826e+02 4.788e+02 5.699e+02 7.070e+02 1.580e+03, threshold=1.140e+03, percent-clipped=2.0 +2023-04-02 22:36:39,134 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-02 22:36:40,526 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=149876.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:36:54,748 INFO [train.py:903] (0/4) Epoch 22, batch 6500, loss[loss=0.2196, simple_loss=0.3137, pruned_loss=0.06268, over 19619.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.287, pruned_loss=0.06309, over 3849543.75 frames. ], batch size: 57, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:37:00,216 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-02 22:37:25,352 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5640, 4.0812, 4.2588, 4.2426, 1.7065, 3.9795, 3.4683, 3.9813], + device='cuda:0'), covar=tensor([0.1683, 0.0798, 0.0622, 0.0726, 0.5679, 0.0930, 0.0714, 0.1140], + device='cuda:0'), in_proj_covar=tensor([0.0784, 0.0743, 0.0950, 0.0827, 0.0835, 0.0710, 0.0565, 0.0879], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 22:37:41,908 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=149927.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:37:55,100 INFO [train.py:903] (0/4) Epoch 22, batch 6550, loss[loss=0.2284, simple_loss=0.3098, pruned_loss=0.07343, over 19683.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2882, pruned_loss=0.06378, over 3834249.59 frames. ], batch size: 59, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:37:56,255 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.115e+02 4.654e+02 5.933e+02 7.304e+02 1.667e+03, threshold=1.187e+03, percent-clipped=4.0 +2023-04-02 22:38:45,109 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8284, 1.5775, 1.4662, 1.7767, 1.5024, 1.5564, 1.5093, 1.6791], + device='cuda:0'), covar=tensor([0.1069, 0.1348, 0.1558, 0.0996, 0.1266, 0.0594, 0.1419, 0.0796], + device='cuda:0'), in_proj_covar=tensor([0.0270, 0.0354, 0.0312, 0.0250, 0.0302, 0.0251, 0.0308, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:38:55,835 INFO [train.py:903] (0/4) Epoch 22, batch 6600, loss[loss=0.1849, simple_loss=0.2764, pruned_loss=0.04666, over 19680.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2894, pruned_loss=0.06436, over 3834782.96 frames. ], batch size: 53, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:38:59,608 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=149991.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:39:11,411 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-150000.pt +2023-04-02 22:39:59,893 INFO [train.py:903] (0/4) Epoch 22, batch 6650, loss[loss=0.2121, simple_loss=0.2759, pruned_loss=0.07414, over 19790.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2883, pruned_loss=0.0638, over 3832097.34 frames. ], batch size: 48, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:40:01,053 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.517e+02 4.673e+02 5.867e+02 7.414e+02 1.313e+03, threshold=1.173e+03, percent-clipped=2.0 +2023-04-02 22:40:04,720 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=150042.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:40:59,360 INFO [train.py:903] (0/4) Epoch 22, batch 6700, loss[loss=0.2009, simple_loss=0.2875, pruned_loss=0.05719, over 19732.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2879, pruned_loss=0.06381, over 3838914.60 frames. ], batch size: 63, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:41:10,387 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=150097.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:41:10,938 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.84 vs. limit=2.0 +2023-04-02 22:41:23,633 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=150108.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 22:41:57,256 INFO [train.py:903] (0/4) Epoch 22, batch 6750, loss[loss=0.2002, simple_loss=0.2833, pruned_loss=0.05856, over 19672.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2879, pruned_loss=0.06395, over 3839903.54 frames. ], batch size: 58, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:41:58,371 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.145e+02 4.879e+02 6.504e+02 7.654e+02 1.720e+03, threshold=1.301e+03, percent-clipped=5.0 +2023-04-02 22:42:17,374 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9589, 1.8301, 1.5055, 1.8849, 1.7504, 1.4370, 1.4785, 1.7832], + device='cuda:0'), covar=tensor([0.1129, 0.1534, 0.1774, 0.1195, 0.1489, 0.0838, 0.1808, 0.0903], + device='cuda:0'), in_proj_covar=tensor([0.0269, 0.0352, 0.0312, 0.0249, 0.0301, 0.0250, 0.0308, 0.0255], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:42:53,133 INFO [train.py:903] (0/4) Epoch 22, batch 6800, loss[loss=0.1897, simple_loss=0.2732, pruned_loss=0.05313, over 19664.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.287, pruned_loss=0.0634, over 3829368.83 frames. ], batch size: 55, lr: 3.70e-03, grad_scale: 8.0 +2023-04-02 22:43:23,023 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-22.pt +2023-04-02 22:43:39,240 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-02 22:43:39,707 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-02 22:43:42,623 INFO [train.py:903] (0/4) Epoch 23, batch 0, loss[loss=0.2004, simple_loss=0.2664, pruned_loss=0.0672, over 19036.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2664, pruned_loss=0.0672, over 19036.00 frames. ], batch size: 42, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:43:42,624 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 22:43:54,251 INFO [train.py:937] (0/4) Epoch 23, validation: loss=0.1688, simple_loss=0.2693, pruned_loss=0.03418, over 944034.00 frames. +2023-04-02 22:43:54,251 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-02 22:43:54,671 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3755, 1.4612, 1.6641, 1.5991, 2.4891, 2.0992, 2.5388, 1.0544], + device='cuda:0'), covar=tensor([0.2560, 0.4337, 0.2589, 0.1941, 0.1469, 0.2278, 0.1427, 0.4486], + device='cuda:0'), in_proj_covar=tensor([0.0536, 0.0643, 0.0714, 0.0483, 0.0620, 0.0531, 0.0662, 0.0548], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 22:44:03,545 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=150223.0, num_to_drop=1, layers_to_drop={1} +2023-04-02 22:44:06,626 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-02 22:44:21,434 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.019e+02 4.848e+02 5.561e+02 7.527e+02 1.735e+03, threshold=1.112e+03, percent-clipped=5.0 +2023-04-02 22:44:31,829 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=150247.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:44:55,831 INFO [train.py:903] (0/4) Epoch 23, batch 50, loss[loss=0.2282, simple_loss=0.3083, pruned_loss=0.07408, over 19468.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2864, pruned_loss=0.06261, over 870087.41 frames. ], batch size: 64, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:45:03,001 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=150272.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:45:27,179 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=150292.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:45:30,280 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-02 22:45:35,174 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=150298.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:45:57,947 INFO [train.py:903] (0/4) Epoch 23, batch 100, loss[loss=0.2309, simple_loss=0.3081, pruned_loss=0.07688, over 13167.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2853, pruned_loss=0.06231, over 1526992.47 frames. ], batch size: 135, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:46:06,475 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=150323.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:46:07,258 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-02 22:46:26,505 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.179e+02 4.910e+02 5.630e+02 7.676e+02 1.557e+03, threshold=1.126e+03, percent-clipped=7.0 +2023-04-02 22:46:59,546 INFO [train.py:903] (0/4) Epoch 23, batch 150, loss[loss=0.1885, simple_loss=0.2776, pruned_loss=0.04975, over 19525.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2854, pruned_loss=0.06248, over 2043211.72 frames. ], batch size: 54, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:47:22,245 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9018, 3.3207, 3.6404, 3.6413, 1.7458, 3.4383, 2.9752, 3.1700], + device='cuda:0'), covar=tensor([0.2744, 0.2229, 0.1196, 0.1658, 0.7334, 0.2266, 0.1407, 0.2161], + device='cuda:0'), in_proj_covar=tensor([0.0789, 0.0747, 0.0955, 0.0836, 0.0839, 0.0715, 0.0569, 0.0882], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 22:47:23,460 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6725, 1.5654, 1.5341, 2.1405, 1.5980, 2.0196, 2.0728, 1.7784], + device='cuda:0'), covar=tensor([0.0877, 0.0960, 0.1034, 0.0794, 0.0919, 0.0752, 0.0866, 0.0718], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0222, 0.0222, 0.0240, 0.0227, 0.0213, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 22:47:59,877 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-02 22:48:00,265 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2763, 1.9944, 2.0970, 2.7745, 2.0371, 2.4063, 2.5151, 2.2812], + device='cuda:0'), covar=tensor([0.0749, 0.0909, 0.0889, 0.0827, 0.0835, 0.0778, 0.0924, 0.0648], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0222, 0.0239, 0.0226, 0.0212, 0.0187, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 22:48:01,004 INFO [train.py:903] (0/4) Epoch 23, batch 200, loss[loss=0.1764, simple_loss=0.2688, pruned_loss=0.04198, over 19763.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2863, pruned_loss=0.06377, over 2448264.03 frames. ], batch size: 54, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:48:01,405 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5987, 1.5505, 1.5177, 1.9752, 1.5768, 1.8606, 1.9619, 1.7494], + device='cuda:0'), covar=tensor([0.0824, 0.0911, 0.0991, 0.0755, 0.0800, 0.0752, 0.0781, 0.0672], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0222, 0.0239, 0.0226, 0.0212, 0.0187, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 22:48:26,084 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-04-02 22:48:30,853 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.764e+02 5.350e+02 6.653e+02 9.712e+02 2.771e+03, threshold=1.331e+03, percent-clipped=16.0 +2023-04-02 22:48:33,196 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=150441.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:48:41,458 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4040, 2.1704, 1.6266, 1.3795, 2.0201, 1.2959, 1.4395, 1.8414], + device='cuda:0'), covar=tensor([0.1081, 0.0844, 0.1101, 0.0854, 0.0543, 0.1324, 0.0679, 0.0524], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0317, 0.0339, 0.0267, 0.0247, 0.0336, 0.0291, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:49:02,670 INFO [train.py:903] (0/4) Epoch 23, batch 250, loss[loss=0.2032, simple_loss=0.2887, pruned_loss=0.05888, over 18247.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.287, pruned_loss=0.06457, over 2754167.31 frames. ], batch size: 83, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:49:20,036 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=150479.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 22:49:36,975 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8691, 4.3280, 4.6218, 4.6168, 1.7854, 4.3250, 3.7943, 4.3143], + device='cuda:0'), covar=tensor([0.1709, 0.0989, 0.0655, 0.0685, 0.6101, 0.0978, 0.0696, 0.1184], + device='cuda:0'), in_proj_covar=tensor([0.0784, 0.0744, 0.0951, 0.0832, 0.0836, 0.0712, 0.0566, 0.0878], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 22:49:49,516 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7905, 4.3051, 2.8700, 3.7512, 0.8841, 4.2596, 4.1452, 4.3372], + device='cuda:0'), covar=tensor([0.0679, 0.1202, 0.1915, 0.0936, 0.4296, 0.0724, 0.0927, 0.1307], + device='cuda:0'), in_proj_covar=tensor([0.0504, 0.0412, 0.0497, 0.0345, 0.0400, 0.0433, 0.0425, 0.0460], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:49:49,699 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=150504.0, num_to_drop=1, layers_to_drop={0} +2023-04-02 22:50:05,860 INFO [train.py:903] (0/4) Epoch 23, batch 300, loss[loss=0.19, simple_loss=0.2758, pruned_loss=0.0521, over 17464.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2852, pruned_loss=0.06292, over 3004967.00 frames. ], batch size: 101, lr: 3.61e-03, grad_scale: 4.0 +2023-04-02 22:50:34,492 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.361e+02 5.024e+02 5.928e+02 7.198e+02 2.066e+03, threshold=1.186e+03, percent-clipped=3.0 +2023-04-02 22:50:54,914 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=150555.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:50:56,018 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=150556.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:50:59,515 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2428, 1.3103, 1.2851, 1.0654, 1.0834, 1.1149, 0.0710, 0.3800], + device='cuda:0'), covar=tensor([0.0676, 0.0632, 0.0404, 0.0541, 0.1236, 0.0623, 0.1253, 0.1085], + device='cuda:0'), in_proj_covar=tensor([0.0355, 0.0354, 0.0358, 0.0380, 0.0458, 0.0387, 0.0334, 0.0339], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 22:51:07,066 INFO [train.py:903] (0/4) Epoch 23, batch 350, loss[loss=0.1977, simple_loss=0.2834, pruned_loss=0.05602, over 19643.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2859, pruned_loss=0.06344, over 3183956.38 frames. ], batch size: 53, lr: 3.61e-03, grad_scale: 4.0 +2023-04-02 22:51:11,934 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-02 22:51:24,685 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8113, 4.2912, 4.5093, 4.5065, 1.7912, 4.1941, 3.7407, 4.2438], + device='cuda:0'), covar=tensor([0.1583, 0.0804, 0.0607, 0.0648, 0.5696, 0.0838, 0.0645, 0.1093], + device='cuda:0'), in_proj_covar=tensor([0.0787, 0.0745, 0.0954, 0.0832, 0.0838, 0.0713, 0.0568, 0.0880], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 22:51:53,186 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2948, 2.4090, 2.5230, 2.9712, 2.4566, 2.8772, 2.5453, 2.3979], + device='cuda:0'), covar=tensor([0.3615, 0.3116, 0.1567, 0.1965, 0.3335, 0.1670, 0.3819, 0.2609], + device='cuda:0'), in_proj_covar=tensor([0.0903, 0.0968, 0.0719, 0.0932, 0.0882, 0.0819, 0.0844, 0.0785], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 22:52:09,969 INFO [train.py:903] (0/4) Epoch 23, batch 400, loss[loss=0.2031, simple_loss=0.2649, pruned_loss=0.07066, over 19761.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2867, pruned_loss=0.06336, over 3324608.70 frames. ], batch size: 47, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:52:18,810 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3490, 2.1028, 1.6901, 1.4595, 1.9192, 1.4023, 1.3912, 1.8561], + device='cuda:0'), covar=tensor([0.0939, 0.0776, 0.0983, 0.0830, 0.0563, 0.1247, 0.0649, 0.0458], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0316, 0.0339, 0.0266, 0.0246, 0.0336, 0.0290, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 22:52:36,586 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=150636.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:52:40,894 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.962e+02 5.093e+02 6.486e+02 8.008e+02 1.724e+03, threshold=1.297e+03, percent-clipped=3.0 +2023-04-02 22:53:11,936 INFO [train.py:903] (0/4) Epoch 23, batch 450, loss[loss=0.165, simple_loss=0.2398, pruned_loss=0.04507, over 19025.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2867, pruned_loss=0.06347, over 3429677.83 frames. ], batch size: 42, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:53:46,065 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-02 22:53:46,092 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-02 22:54:10,481 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5760, 1.6306, 1.9272, 1.7819, 2.7549, 2.3724, 2.8639, 1.4050], + device='cuda:0'), covar=tensor([0.2479, 0.4262, 0.2610, 0.1924, 0.1445, 0.2093, 0.1387, 0.4338], + device='cuda:0'), in_proj_covar=tensor([0.0535, 0.0644, 0.0713, 0.0485, 0.0617, 0.0532, 0.0662, 0.0548], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 22:54:15,729 INFO [train.py:903] (0/4) Epoch 23, batch 500, loss[loss=0.1836, simple_loss=0.2731, pruned_loss=0.04703, over 19839.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.286, pruned_loss=0.06289, over 3529970.50 frames. ], batch size: 52, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:54:45,188 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.118e+02 5.191e+02 6.635e+02 8.528e+02 2.142e+03, threshold=1.327e+03, percent-clipped=5.0 +2023-04-02 22:54:56,598 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.59 vs. limit=2.0 +2023-04-02 22:54:57,358 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=150750.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:54:58,595 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=150751.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:55:17,411 INFO [train.py:903] (0/4) Epoch 23, batch 550, loss[loss=0.2359, simple_loss=0.3136, pruned_loss=0.07914, over 19486.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2867, pruned_loss=0.06335, over 3612525.53 frames. ], batch size: 64, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:56:14,501 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=150812.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:56:18,591 INFO [train.py:903] (0/4) Epoch 23, batch 600, loss[loss=0.1797, simple_loss=0.255, pruned_loss=0.05215, over 19378.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2875, pruned_loss=0.06402, over 3658175.88 frames. ], batch size: 47, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:56:45,478 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=150837.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:56:48,650 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.714e+02 5.000e+02 5.859e+02 6.998e+02 1.831e+03, threshold=1.172e+03, percent-clipped=2.0 +2023-04-02 22:56:59,201 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-02 22:57:21,232 INFO [train.py:903] (0/4) Epoch 23, batch 650, loss[loss=0.1996, simple_loss=0.2851, pruned_loss=0.05703, over 19670.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2872, pruned_loss=0.06375, over 3699010.73 frames. ], batch size: 53, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:58:02,420 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=150899.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:58:23,646 INFO [train.py:903] (0/4) Epoch 23, batch 700, loss[loss=0.1902, simple_loss=0.2714, pruned_loss=0.05453, over 19681.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.2864, pruned_loss=0.06327, over 3731026.97 frames. ], batch size: 53, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:58:26,208 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=150918.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 22:58:52,734 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.226e+02 5.019e+02 5.794e+02 7.164e+02 1.349e+03, threshold=1.159e+03, percent-clipped=1.0 +2023-04-02 22:59:26,033 INFO [train.py:903] (0/4) Epoch 23, batch 750, loss[loss=0.1597, simple_loss=0.2374, pruned_loss=0.04096, over 19756.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2863, pruned_loss=0.06326, over 3758917.47 frames. ], batch size: 46, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 22:59:36,730 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0283, 2.0633, 2.3455, 2.7492, 2.0066, 2.6246, 2.4424, 2.1386], + device='cuda:0'), covar=tensor([0.4154, 0.3852, 0.1849, 0.2346, 0.4008, 0.1999, 0.4703, 0.3342], + device='cuda:0'), in_proj_covar=tensor([0.0902, 0.0968, 0.0718, 0.0932, 0.0884, 0.0819, 0.0845, 0.0784], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 23:00:06,239 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9682, 4.1048, 4.6922, 4.7661, 1.8287, 4.4423, 3.7248, 4.1245], + device='cuda:0'), covar=tensor([0.1970, 0.1349, 0.0897, 0.0946, 0.7109, 0.1595, 0.1092, 0.1822], + device='cuda:0'), in_proj_covar=tensor([0.0789, 0.0752, 0.0958, 0.0836, 0.0844, 0.0715, 0.0570, 0.0889], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 23:00:17,543 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151007.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:00:26,933 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151014.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:00:28,666 INFO [train.py:903] (0/4) Epoch 23, batch 800, loss[loss=0.2699, simple_loss=0.3392, pruned_loss=0.1003, over 19665.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2875, pruned_loss=0.06346, over 3786611.76 frames. ], batch size: 60, lr: 3.61e-03, grad_scale: 8.0 +2023-04-02 23:00:46,706 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-02 23:00:48,191 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151032.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:00:57,135 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.089e+02 5.089e+02 6.161e+02 7.478e+02 1.780e+03, threshold=1.232e+03, percent-clipped=6.0 +2023-04-02 23:01:29,767 INFO [train.py:903] (0/4) Epoch 23, batch 850, loss[loss=0.2066, simple_loss=0.295, pruned_loss=0.05909, over 19687.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.2886, pruned_loss=0.06378, over 3806402.56 frames. ], batch size: 59, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:01:49,507 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1641, 2.0737, 2.0260, 2.3409, 2.0960, 1.8946, 1.9870, 2.1918], + device='cuda:0'), covar=tensor([0.0903, 0.1239, 0.1226, 0.0777, 0.1066, 0.0519, 0.1143, 0.0619], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0357, 0.0316, 0.0252, 0.0306, 0.0254, 0.0312, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:02:04,867 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=151094.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:02:25,394 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-02 23:02:31,810 INFO [train.py:903] (0/4) Epoch 23, batch 900, loss[loss=0.2035, simple_loss=0.2726, pruned_loss=0.06717, over 16008.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2881, pruned_loss=0.06379, over 3811441.38 frames. ], batch size: 35, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:03:02,019 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.084e+02 5.070e+02 6.543e+02 8.443e+02 1.332e+03, threshold=1.309e+03, percent-clipped=3.0 +2023-04-02 23:03:32,678 INFO [train.py:903] (0/4) Epoch 23, batch 950, loss[loss=0.2238, simple_loss=0.3017, pruned_loss=0.07295, over 19609.00 frames. ], tot_loss[loss=0.2103, simple_loss=0.2899, pruned_loss=0.0654, over 3812974.98 frames. ], batch size: 57, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:03:39,562 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-02 23:04:17,796 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151202.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:04:27,592 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151209.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:04:35,177 INFO [train.py:903] (0/4) Epoch 23, batch 1000, loss[loss=0.2255, simple_loss=0.3053, pruned_loss=0.07283, over 19770.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2899, pruned_loss=0.06509, over 3815514.53 frames. ], batch size: 56, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:05:04,887 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.333e+02 5.146e+02 6.401e+02 7.951e+02 1.702e+03, threshold=1.280e+03, percent-clipped=4.0 +2023-04-02 23:05:32,293 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-02 23:05:33,601 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=151262.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:05:37,849 INFO [train.py:903] (0/4) Epoch 23, batch 1050, loss[loss=0.2056, simple_loss=0.2816, pruned_loss=0.06481, over 19749.00 frames. ], tot_loss[loss=0.21, simple_loss=0.2898, pruned_loss=0.06508, over 3828226.29 frames. ], batch size: 51, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:05:42,793 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151270.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:06:12,791 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-02 23:06:14,343 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151295.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:06:21,346 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151300.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:06:40,104 INFO [train.py:903] (0/4) Epoch 23, batch 1100, loss[loss=0.1931, simple_loss=0.263, pruned_loss=0.06156, over 19736.00 frames. ], tot_loss[loss=0.2094, simple_loss=0.2894, pruned_loss=0.06473, over 3835681.29 frames. ], batch size: 45, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:06:40,432 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151316.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:06:57,240 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6300, 1.7821, 2.1768, 1.9467, 3.2839, 2.6826, 3.5808, 1.7663], + device='cuda:0'), covar=tensor([0.2510, 0.4251, 0.2737, 0.1892, 0.1560, 0.2091, 0.1594, 0.4114], + device='cuda:0'), in_proj_covar=tensor([0.0536, 0.0644, 0.0714, 0.0486, 0.0618, 0.0529, 0.0661, 0.0550], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 23:07:09,142 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.937e+02 5.070e+02 6.152e+02 7.617e+02 1.362e+03, threshold=1.230e+03, percent-clipped=2.0 +2023-04-02 23:07:40,862 INFO [train.py:903] (0/4) Epoch 23, batch 1150, loss[loss=0.2051, simple_loss=0.2713, pruned_loss=0.06947, over 19744.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2885, pruned_loss=0.06407, over 3828871.78 frames. ], batch size: 47, lr: 3.60e-03, grad_scale: 4.0 +2023-04-02 23:07:52,344 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9693, 1.3543, 1.1168, 0.9545, 1.2539, 0.8973, 1.0248, 1.2282], + device='cuda:0'), covar=tensor([0.0593, 0.0639, 0.0676, 0.0624, 0.0432, 0.0985, 0.0465, 0.0400], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0318, 0.0340, 0.0266, 0.0247, 0.0338, 0.0292, 0.0276], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:07:55,725 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151377.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:08:43,939 INFO [train.py:903] (0/4) Epoch 23, batch 1200, loss[loss=0.2818, simple_loss=0.3403, pruned_loss=0.1116, over 13559.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2881, pruned_loss=0.06418, over 3818392.49 frames. ], batch size: 136, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:09:10,455 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6216, 1.2200, 1.2681, 1.5051, 1.1254, 1.3688, 1.2403, 1.4571], + device='cuda:0'), covar=tensor([0.1016, 0.1206, 0.1452, 0.0906, 0.1232, 0.0611, 0.1426, 0.0754], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0355, 0.0314, 0.0251, 0.0303, 0.0252, 0.0309, 0.0256], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:09:14,770 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.106e+02 4.896e+02 6.001e+02 7.643e+02 1.247e+03, threshold=1.200e+03, percent-clipped=2.0 +2023-04-02 23:09:18,087 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-02 23:09:45,194 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151465.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:09:46,024 INFO [train.py:903] (0/4) Epoch 23, batch 1250, loss[loss=0.2063, simple_loss=0.2928, pruned_loss=0.05986, over 18375.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2887, pruned_loss=0.0646, over 3823902.35 frames. ], batch size: 84, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:10:16,513 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151490.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:10:27,795 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-02 23:10:46,698 INFO [train.py:903] (0/4) Epoch 23, batch 1300, loss[loss=0.2554, simple_loss=0.3244, pruned_loss=0.09317, over 19649.00 frames. ], tot_loss[loss=0.2093, simple_loss=0.289, pruned_loss=0.06484, over 3816125.00 frames. ], batch size: 55, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:10:55,127 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3130, 1.3347, 1.4877, 1.4505, 1.7688, 1.8211, 1.8195, 0.6260], + device='cuda:0'), covar=tensor([0.2484, 0.4265, 0.2664, 0.1983, 0.1648, 0.2290, 0.1423, 0.4673], + device='cuda:0'), in_proj_covar=tensor([0.0539, 0.0648, 0.0718, 0.0488, 0.0621, 0.0533, 0.0666, 0.0554], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 23:10:55,327 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-02 23:11:16,028 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.826e+02 5.118e+02 6.046e+02 8.032e+02 1.744e+03, threshold=1.209e+03, percent-clipped=5.0 +2023-04-02 23:11:22,857 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=151546.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:11:25,567 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4310, 1.4323, 1.7008, 1.6255, 2.3624, 2.1104, 2.4137, 1.0108], + device='cuda:0'), covar=tensor([0.2555, 0.4397, 0.2710, 0.1930, 0.1587, 0.2204, 0.1655, 0.4636], + device='cuda:0'), in_proj_covar=tensor([0.0540, 0.0650, 0.0719, 0.0490, 0.0622, 0.0534, 0.0668, 0.0555], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 23:11:46,626 INFO [train.py:903] (0/4) Epoch 23, batch 1350, loss[loss=0.2137, simple_loss=0.3011, pruned_loss=0.06311, over 19592.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2885, pruned_loss=0.06465, over 3819930.27 frames. ], batch size: 57, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:12:48,237 INFO [train.py:903] (0/4) Epoch 23, batch 1400, loss[loss=0.21, simple_loss=0.298, pruned_loss=0.06095, over 19699.00 frames. ], tot_loss[loss=0.2092, simple_loss=0.2889, pruned_loss=0.06471, over 3820326.14 frames. ], batch size: 59, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:13:08,471 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151633.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:13:17,803 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.750e+02 5.043e+02 6.240e+02 8.237e+02 1.280e+03, threshold=1.248e+03, percent-clipped=3.0 +2023-04-02 23:13:21,398 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=151644.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:13:38,316 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151658.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:13:40,474 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=151660.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:13:41,838 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151661.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:13:48,339 INFO [train.py:903] (0/4) Epoch 23, batch 1450, loss[loss=0.2263, simple_loss=0.3055, pruned_loss=0.07352, over 19418.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2886, pruned_loss=0.06427, over 3825662.00 frames. ], batch size: 70, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:13:48,372 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-02 23:14:36,938 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151706.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:14:49,057 INFO [train.py:903] (0/4) Epoch 23, batch 1500, loss[loss=0.1671, simple_loss=0.2497, pruned_loss=0.04227, over 19765.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2879, pruned_loss=0.06386, over 3826109.41 frames. ], batch size: 47, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:15:18,487 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.327e+02 4.905e+02 6.054e+02 7.299e+02 2.065e+03, threshold=1.211e+03, percent-clipped=4.0 +2023-04-02 23:15:40,356 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151759.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:15:47,614 INFO [train.py:903] (0/4) Epoch 23, batch 1550, loss[loss=0.2113, simple_loss=0.2946, pruned_loss=0.06395, over 19477.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2885, pruned_loss=0.06436, over 3819166.93 frames. ], batch size: 64, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:15:59,948 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=151775.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:16:50,043 INFO [train.py:903] (0/4) Epoch 23, batch 1600, loss[loss=0.2062, simple_loss=0.2852, pruned_loss=0.06359, over 18232.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2881, pruned_loss=0.06415, over 3805415.56 frames. ], batch size: 83, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:16:53,576 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=151819.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:17:10,293 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-02 23:17:20,214 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.366e+02 4.863e+02 5.887e+02 6.951e+02 2.426e+03, threshold=1.177e+03, percent-clipped=3.0 +2023-04-02 23:17:50,202 INFO [train.py:903] (0/4) Epoch 23, batch 1650, loss[loss=0.2008, simple_loss=0.2868, pruned_loss=0.05741, over 19604.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2887, pruned_loss=0.06441, over 3818953.03 frames. ], batch size: 57, lr: 3.60e-03, grad_scale: 8.0 +2023-04-02 23:18:39,713 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2258, 1.3323, 1.7673, 1.3544, 2.7825, 3.8009, 3.5580, 4.0582], + device='cuda:0'), covar=tensor([0.1608, 0.3729, 0.3262, 0.2396, 0.0586, 0.0183, 0.0198, 0.0225], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0323, 0.0355, 0.0265, 0.0244, 0.0189, 0.0217, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 23:18:51,777 INFO [train.py:903] (0/4) Epoch 23, batch 1700, loss[loss=0.1876, simple_loss=0.2652, pruned_loss=0.05502, over 19427.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2881, pruned_loss=0.06397, over 3808768.32 frames. ], batch size: 48, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:18:53,366 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=151917.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:19:21,472 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.136e+02 5.018e+02 6.073e+02 7.613e+02 1.748e+03, threshold=1.215e+03, percent-clipped=5.0 +2023-04-02 23:19:23,180 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=151942.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:19:26,202 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-02 23:19:52,372 INFO [train.py:903] (0/4) Epoch 23, batch 1750, loss[loss=0.1904, simple_loss=0.2738, pruned_loss=0.05351, over 19728.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2879, pruned_loss=0.06377, over 3818442.82 frames. ], batch size: 51, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:20:33,144 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-152000.pt +2023-04-02 23:20:53,712 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152015.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:20:54,414 INFO [train.py:903] (0/4) Epoch 23, batch 1800, loss[loss=0.1914, simple_loss=0.2812, pruned_loss=0.05078, over 17329.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2857, pruned_loss=0.06277, over 3813887.19 frames. ], batch size: 101, lr: 3.59e-03, grad_scale: 4.0 +2023-04-02 23:21:13,130 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152031.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:21:24,598 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152040.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:21:26,673 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.151e+02 5.200e+02 6.601e+02 8.823e+02 1.720e+03, threshold=1.320e+03, percent-clipped=12.0 +2023-04-02 23:21:36,352 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=152050.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:21:40,797 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0915, 1.3995, 1.6905, 1.1522, 2.6187, 3.3650, 3.0509, 3.5531], + device='cuda:0'), covar=tensor([0.1703, 0.3682, 0.3364, 0.2514, 0.0552, 0.0183, 0.0239, 0.0259], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0324, 0.0355, 0.0265, 0.0245, 0.0188, 0.0217, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 23:21:44,275 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152056.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:21:48,439 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-02 23:21:55,072 INFO [train.py:903] (0/4) Epoch 23, batch 1850, loss[loss=0.2126, simple_loss=0.2966, pruned_loss=0.0643, over 18120.00 frames. ], tot_loss[loss=0.206, simple_loss=0.286, pruned_loss=0.06298, over 3820143.81 frames. ], batch size: 83, lr: 3.59e-03, grad_scale: 4.0 +2023-04-02 23:22:26,622 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-02 23:22:54,485 INFO [train.py:903] (0/4) Epoch 23, batch 1900, loss[loss=0.2186, simple_loss=0.2785, pruned_loss=0.07936, over 19773.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2886, pruned_loss=0.06465, over 3819651.07 frames. ], batch size: 47, lr: 3.59e-03, grad_scale: 4.0 +2023-04-02 23:23:09,860 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-02 23:23:16,371 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-02 23:23:26,804 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.505e+02 4.994e+02 5.968e+02 7.712e+02 2.482e+03, threshold=1.194e+03, percent-clipped=3.0 +2023-04-02 23:23:36,562 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5074, 1.5955, 1.7913, 1.7600, 2.6850, 2.4075, 2.8471, 1.1523], + device='cuda:0'), covar=tensor([0.2473, 0.4345, 0.2750, 0.1890, 0.1534, 0.2045, 0.1436, 0.4474], + device='cuda:0'), in_proj_covar=tensor([0.0534, 0.0643, 0.0715, 0.0486, 0.0618, 0.0530, 0.0663, 0.0549], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 23:23:41,404 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-02 23:23:52,764 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=152163.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:23:55,363 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=152165.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:23:56,144 INFO [train.py:903] (0/4) Epoch 23, batch 1950, loss[loss=0.1862, simple_loss=0.2773, pruned_loss=0.04757, over 19723.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.287, pruned_loss=0.06365, over 3814573.22 frames. ], batch size: 63, lr: 3.59e-03, grad_scale: 4.0 +2023-04-02 23:24:58,529 INFO [train.py:903] (0/4) Epoch 23, batch 2000, loss[loss=0.2263, simple_loss=0.3133, pruned_loss=0.06965, over 19593.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2879, pruned_loss=0.06393, over 3820139.97 frames. ], batch size: 57, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:25:06,473 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.61 vs. limit=5.0 +2023-04-02 23:25:28,733 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.131e+02 4.661e+02 5.613e+02 7.041e+02 1.127e+03, threshold=1.123e+03, percent-clipped=0.0 +2023-04-02 23:25:30,268 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152243.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:25:53,680 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-02 23:25:58,367 INFO [train.py:903] (0/4) Epoch 23, batch 2050, loss[loss=0.2002, simple_loss=0.288, pruned_loss=0.05626, over 19807.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2879, pruned_loss=0.06363, over 3813974.73 frames. ], batch size: 56, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:26:13,106 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-02 23:26:13,467 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=152278.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:26:14,198 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-02 23:26:29,846 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152291.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:26:36,222 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-02 23:26:58,781 INFO [train.py:903] (0/4) Epoch 23, batch 2100, loss[loss=0.1916, simple_loss=0.2784, pruned_loss=0.05238, over 19669.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.287, pruned_loss=0.06336, over 3818004.49 frames. ], batch size: 53, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:27:27,810 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-02 23:27:31,218 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.686e+02 4.808e+02 5.705e+02 7.050e+02 1.568e+03, threshold=1.141e+03, percent-clipped=6.0 +2023-04-02 23:27:48,004 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-02 23:27:59,774 INFO [train.py:903] (0/4) Epoch 23, batch 2150, loss[loss=0.229, simple_loss=0.3052, pruned_loss=0.07641, over 19669.00 frames. ], tot_loss[loss=0.208, simple_loss=0.288, pruned_loss=0.06395, over 3792044.62 frames. ], batch size: 53, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:28:24,386 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2186, 1.9718, 1.8638, 2.1427, 1.9257, 1.8705, 1.8120, 2.0543], + device='cuda:0'), covar=tensor([0.0972, 0.1431, 0.1375, 0.1047, 0.1297, 0.0553, 0.1386, 0.0673], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0355, 0.0314, 0.0253, 0.0304, 0.0252, 0.0310, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:28:35,465 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-04-02 23:29:00,654 INFO [train.py:903] (0/4) Epoch 23, batch 2200, loss[loss=0.2396, simple_loss=0.3191, pruned_loss=0.08008, over 19661.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2873, pruned_loss=0.06357, over 3806891.42 frames. ], batch size: 55, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:29:07,429 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152421.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:29:31,743 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.170e+02 5.084e+02 6.029e+02 8.181e+02 1.825e+03, threshold=1.206e+03, percent-clipped=10.0 +2023-04-02 23:29:37,615 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152446.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:29:58,449 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152463.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:30:01,458 INFO [train.py:903] (0/4) Epoch 23, batch 2250, loss[loss=0.2172, simple_loss=0.2981, pruned_loss=0.06815, over 19859.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2875, pruned_loss=0.06365, over 3809732.94 frames. ], batch size: 52, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:31:01,764 INFO [train.py:903] (0/4) Epoch 23, batch 2300, loss[loss=0.2275, simple_loss=0.3063, pruned_loss=0.07436, over 19472.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2869, pruned_loss=0.06325, over 3816529.67 frames. ], batch size: 64, lr: 3.59e-03, grad_scale: 4.0 +2023-04-02 23:31:17,224 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-02 23:31:25,329 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152534.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:31:36,142 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.429e+02 4.927e+02 5.902e+02 7.617e+02 2.113e+03, threshold=1.180e+03, percent-clipped=5.0 +2023-04-02 23:31:55,711 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152559.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:32:04,337 INFO [train.py:903] (0/4) Epoch 23, batch 2350, loss[loss=0.1966, simple_loss=0.2846, pruned_loss=0.05433, over 19481.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2861, pruned_loss=0.06249, over 3829570.37 frames. ], batch size: 64, lr: 3.59e-03, grad_scale: 4.0 +2023-04-02 23:32:30,240 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=152587.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:32:39,671 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-02 23:32:44,139 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-02 23:32:45,492 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152600.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:32:54,402 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152607.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:33:03,099 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-02 23:33:05,157 INFO [train.py:903] (0/4) Epoch 23, batch 2400, loss[loss=0.3627, simple_loss=0.4017, pruned_loss=0.1619, over 13800.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.286, pruned_loss=0.06276, over 3819436.15 frames. ], batch size: 136, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:33:28,191 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=152635.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:33:38,293 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.633e+02 5.347e+02 6.485e+02 7.646e+02 1.871e+03, threshold=1.297e+03, percent-clipped=3.0 +2023-04-02 23:34:06,660 INFO [train.py:903] (0/4) Epoch 23, batch 2450, loss[loss=0.1895, simple_loss=0.2687, pruned_loss=0.05514, over 17327.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2859, pruned_loss=0.06289, over 3823415.94 frames. ], batch size: 38, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:34:37,228 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6624, 1.6174, 1.6346, 1.4376, 3.2219, 1.1469, 2.4396, 3.6882], + device='cuda:0'), covar=tensor([0.0465, 0.2555, 0.2633, 0.1918, 0.0662, 0.2523, 0.1286, 0.0224], + device='cuda:0'), in_proj_covar=tensor([0.0413, 0.0369, 0.0390, 0.0350, 0.0377, 0.0355, 0.0385, 0.0408], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:34:51,410 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=152702.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:35:06,552 INFO [train.py:903] (0/4) Epoch 23, batch 2500, loss[loss=0.1637, simple_loss=0.2374, pruned_loss=0.04502, over 19732.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.2862, pruned_loss=0.06327, over 3814415.94 frames. ], batch size: 45, lr: 3.59e-03, grad_scale: 8.0 +2023-04-02 23:35:12,378 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0188, 1.7214, 1.5920, 1.8780, 1.6378, 1.6927, 1.5229, 1.8687], + device='cuda:0'), covar=tensor([0.1082, 0.1439, 0.1610, 0.1094, 0.1408, 0.0574, 0.1525, 0.0756], + device='cuda:0'), in_proj_covar=tensor([0.0271, 0.0355, 0.0313, 0.0252, 0.0302, 0.0250, 0.0309, 0.0257], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:35:29,547 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9930, 3.6624, 2.6170, 3.2019, 0.8693, 3.5936, 3.4254, 3.5686], + device='cuda:0'), covar=tensor([0.0897, 0.1186, 0.2058, 0.1038, 0.4258, 0.0829, 0.1011, 0.1528], + device='cuda:0'), in_proj_covar=tensor([0.0515, 0.0416, 0.0502, 0.0351, 0.0404, 0.0440, 0.0431, 0.0464], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:35:40,605 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.829e+02 4.968e+02 5.949e+02 7.714e+02 2.745e+03, threshold=1.190e+03, percent-clipped=5.0 +2023-04-02 23:35:42,075 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=152744.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:35:48,939 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=152750.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:36:08,245 INFO [train.py:903] (0/4) Epoch 23, batch 2550, loss[loss=0.1912, simple_loss=0.2672, pruned_loss=0.05762, over 19365.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2855, pruned_loss=0.06248, over 3810493.39 frames. ], batch size: 47, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:36:28,459 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1501, 1.7407, 1.3693, 1.1890, 1.6008, 1.1506, 1.1635, 1.5508], + device='cuda:0'), covar=tensor([0.0826, 0.0747, 0.1000, 0.0784, 0.0539, 0.1179, 0.0610, 0.0438], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0314, 0.0334, 0.0265, 0.0246, 0.0336, 0.0288, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:36:57,882 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=152807.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:37:01,071 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-02 23:37:08,446 INFO [train.py:903] (0/4) Epoch 23, batch 2600, loss[loss=0.1867, simple_loss=0.2735, pruned_loss=0.04994, over 19554.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2865, pruned_loss=0.0631, over 3796402.38 frames. ], batch size: 61, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:37:37,485 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9568, 0.9521, 1.0521, 1.0618, 1.3025, 1.2807, 1.2623, 0.5630], + device='cuda:0'), covar=tensor([0.1762, 0.3119, 0.1920, 0.1481, 0.1212, 0.1698, 0.1100, 0.4008], + device='cuda:0'), in_proj_covar=tensor([0.0536, 0.0647, 0.0718, 0.0487, 0.0620, 0.0533, 0.0664, 0.0552], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 23:37:40,425 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.218e+02 4.724e+02 5.510e+02 7.301e+02 1.657e+03, threshold=1.102e+03, percent-clipped=4.0 +2023-04-02 23:38:08,554 INFO [train.py:903] (0/4) Epoch 23, batch 2650, loss[loss=0.2797, simple_loss=0.3379, pruned_loss=0.1107, over 12957.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2867, pruned_loss=0.0633, over 3798848.56 frames. ], batch size: 136, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:38:27,765 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-02 23:39:08,680 INFO [train.py:903] (0/4) Epoch 23, batch 2700, loss[loss=0.2336, simple_loss=0.3084, pruned_loss=0.07943, over 19517.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2857, pruned_loss=0.06279, over 3808293.71 frames. ], batch size: 56, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:39:16,566 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=152922.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:39:42,350 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.879e+02 4.895e+02 5.730e+02 7.672e+02 1.465e+03, threshold=1.146e+03, percent-clipped=5.0 +2023-04-02 23:39:43,655 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=152944.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:39:51,723 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=152951.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:40:00,053 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=152958.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:40:09,443 INFO [train.py:903] (0/4) Epoch 23, batch 2750, loss[loss=0.2261, simple_loss=0.3139, pruned_loss=0.06913, over 19768.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2862, pruned_loss=0.06303, over 3800572.25 frames. ], batch size: 54, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:40:23,468 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1037, 1.2638, 1.5018, 1.3309, 2.7556, 1.0628, 2.1099, 3.0923], + device='cuda:0'), covar=tensor([0.0610, 0.2826, 0.2935, 0.1869, 0.0739, 0.2428, 0.1296, 0.0315], + device='cuda:0'), in_proj_covar=tensor([0.0414, 0.0370, 0.0390, 0.0350, 0.0375, 0.0354, 0.0383, 0.0407], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:40:29,164 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7721, 1.5344, 1.6384, 2.3497, 1.6397, 2.0874, 2.0376, 1.8597], + device='cuda:0'), covar=tensor([0.0836, 0.0996, 0.1002, 0.0738, 0.0896, 0.0776, 0.0889, 0.0698], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0220, 0.0224, 0.0238, 0.0227, 0.0211, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 23:40:31,170 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=152983.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:40:57,914 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153006.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:41:10,187 INFO [train.py:903] (0/4) Epoch 23, batch 2800, loss[loss=0.2157, simple_loss=0.2906, pruned_loss=0.07042, over 19675.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2859, pruned_loss=0.06294, over 3808086.26 frames. ], batch size: 53, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:41:27,917 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153031.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:41:42,353 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.106e+02 4.940e+02 6.140e+02 7.865e+02 1.529e+03, threshold=1.228e+03, percent-clipped=3.0 +2023-04-02 23:41:54,223 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-02 23:42:02,476 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153059.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:42:10,881 INFO [train.py:903] (0/4) Epoch 23, batch 2850, loss[loss=0.2096, simple_loss=0.298, pruned_loss=0.06062, over 19609.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2864, pruned_loss=0.06297, over 3816137.80 frames. ], batch size: 50, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:42:11,224 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153066.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:42:33,222 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2881, 3.8188, 3.9346, 3.9231, 1.6925, 3.7337, 3.2816, 3.6824], + device='cuda:0'), covar=tensor([0.1770, 0.1051, 0.0664, 0.0795, 0.5632, 0.1026, 0.0721, 0.1177], + device='cuda:0'), in_proj_covar=tensor([0.0786, 0.0750, 0.0957, 0.0840, 0.0844, 0.0718, 0.0570, 0.0892], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 23:42:36,382 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=153088.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:42:51,840 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0674, 2.6380, 2.8010, 3.3568, 2.5682, 3.0656, 2.9735, 2.9220], + device='cuda:0'), covar=tensor([0.0535, 0.0685, 0.0669, 0.0599, 0.0716, 0.0603, 0.0777, 0.0495], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0221, 0.0224, 0.0239, 0.0228, 0.0212, 0.0188, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-02 23:43:09,824 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-02 23:43:11,002 INFO [train.py:903] (0/4) Epoch 23, batch 2900, loss[loss=0.2284, simple_loss=0.3047, pruned_loss=0.076, over 19758.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2866, pruned_loss=0.0634, over 3794982.32 frames. ], batch size: 63, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:43:12,400 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7687, 1.3309, 1.5604, 1.5374, 3.3847, 1.2499, 2.3853, 3.7893], + device='cuda:0'), covar=tensor([0.0528, 0.2764, 0.2846, 0.1890, 0.0685, 0.2466, 0.1324, 0.0238], + device='cuda:0'), in_proj_covar=tensor([0.0413, 0.0368, 0.0390, 0.0350, 0.0374, 0.0354, 0.0383, 0.0407], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:43:34,519 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=153135.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:43:45,165 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.545e+02 5.074e+02 6.117e+02 7.609e+02 1.538e+03, threshold=1.223e+03, percent-clipped=2.0 +2023-04-02 23:44:10,299 INFO [train.py:903] (0/4) Epoch 23, batch 2950, loss[loss=0.2524, simple_loss=0.3226, pruned_loss=0.09111, over 19786.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.287, pruned_loss=0.06379, over 3801221.60 frames. ], batch size: 56, lr: 3.58e-03, grad_scale: 4.0 +2023-04-02 23:44:23,197 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6384, 4.1184, 4.2794, 4.2568, 1.5917, 4.0524, 3.5314, 4.0128], + device='cuda:0'), covar=tensor([0.1545, 0.0836, 0.0591, 0.0706, 0.5928, 0.0893, 0.0686, 0.1018], + device='cuda:0'), in_proj_covar=tensor([0.0780, 0.0743, 0.0950, 0.0833, 0.0838, 0.0711, 0.0566, 0.0885], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-02 23:44:25,524 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153178.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:44:54,204 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153203.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:44:54,244 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153203.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:45:09,887 INFO [train.py:903] (0/4) Epoch 23, batch 3000, loss[loss=0.205, simple_loss=0.2846, pruned_loss=0.06268, over 19679.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2879, pruned_loss=0.06423, over 3802267.83 frames. ], batch size: 59, lr: 3.58e-03, grad_scale: 4.0 +2023-04-02 23:45:09,888 INFO [train.py:928] (0/4) Computing validation loss +2023-04-02 23:45:23,394 INFO [train.py:937] (0/4) Epoch 23, validation: loss=0.1686, simple_loss=0.2685, pruned_loss=0.03441, over 944034.00 frames. +2023-04-02 23:45:23,395 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-02 23:45:26,708 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-02 23:45:27,896 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9295, 4.5015, 2.6139, 3.8925, 0.9134, 4.4582, 4.3079, 4.4288], + device='cuda:0'), covar=tensor([0.0561, 0.0879, 0.2098, 0.0866, 0.4135, 0.0590, 0.0904, 0.1042], + device='cuda:0'), in_proj_covar=tensor([0.0511, 0.0414, 0.0498, 0.0349, 0.0402, 0.0437, 0.0429, 0.0463], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:45:40,632 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-02 23:45:55,393 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1376, 2.1931, 2.4516, 2.9812, 2.2363, 2.8086, 2.4839, 2.1121], + device='cuda:0'), covar=tensor([0.4600, 0.4224, 0.1983, 0.2739, 0.4515, 0.2356, 0.5225, 0.3589], + device='cuda:0'), in_proj_covar=tensor([0.0908, 0.0975, 0.0721, 0.0934, 0.0885, 0.0822, 0.0848, 0.0787], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 23:45:57,202 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.963e+02 5.132e+02 6.544e+02 7.997e+02 1.730e+03, threshold=1.309e+03, percent-clipped=4.0 +2023-04-02 23:46:23,492 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0 +2023-04-02 23:46:24,013 INFO [train.py:903] (0/4) Epoch 23, batch 3050, loss[loss=0.1778, simple_loss=0.2528, pruned_loss=0.05139, over 19401.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2878, pruned_loss=0.06408, over 3796795.45 frames. ], batch size: 47, lr: 3.58e-03, grad_scale: 4.0 +2023-04-02 23:46:26,459 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8202, 4.4060, 2.8226, 3.8447, 0.9551, 4.3452, 4.2161, 4.3031], + device='cuda:0'), covar=tensor([0.0567, 0.0824, 0.1767, 0.0783, 0.3983, 0.0618, 0.0843, 0.1182], + device='cuda:0'), in_proj_covar=tensor([0.0511, 0.0414, 0.0497, 0.0349, 0.0401, 0.0437, 0.0430, 0.0463], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:47:00,333 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=153296.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:47:05,917 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4462, 1.5392, 1.8694, 1.7732, 2.7815, 2.3904, 3.0033, 1.2883], + device='cuda:0'), covar=tensor([0.2540, 0.4384, 0.2665, 0.1826, 0.1534, 0.2128, 0.1505, 0.4468], + device='cuda:0'), in_proj_covar=tensor([0.0535, 0.0644, 0.0715, 0.0486, 0.0617, 0.0531, 0.0662, 0.0551], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 23:47:25,736 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153315.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:47:26,469 INFO [train.py:903] (0/4) Epoch 23, batch 3100, loss[loss=0.2151, simple_loss=0.3032, pruned_loss=0.06348, over 19318.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2879, pruned_loss=0.06399, over 3807104.53 frames. ], batch size: 70, lr: 3.58e-03, grad_scale: 4.0 +2023-04-02 23:47:33,570 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153322.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:47:54,325 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153340.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:47:59,280 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.375e+02 4.897e+02 6.414e+02 9.491e+02 6.432e+03, threshold=1.283e+03, percent-clipped=11.0 +2023-04-02 23:48:01,113 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.72 vs. limit=2.0 +2023-04-02 23:48:03,099 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153347.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:48:13,731 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2504, 2.0488, 1.8459, 2.1983, 2.1859, 1.7087, 1.6899, 2.1878], + device='cuda:0'), covar=tensor([0.1166, 0.1848, 0.1823, 0.1296, 0.1510, 0.0944, 0.1935, 0.0917], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0357, 0.0314, 0.0254, 0.0305, 0.0252, 0.0310, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:48:25,988 INFO [train.py:903] (0/4) Epoch 23, batch 3150, loss[loss=0.164, simple_loss=0.2468, pruned_loss=0.04064, over 19735.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2866, pruned_loss=0.06294, over 3824899.71 frames. ], batch size: 51, lr: 3.58e-03, grad_scale: 4.0 +2023-04-02 23:48:54,115 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-02 23:48:57,789 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3273, 1.4547, 1.6174, 1.6390, 2.9680, 1.2032, 2.4807, 3.2811], + device='cuda:0'), covar=tensor([0.0570, 0.2826, 0.2845, 0.1776, 0.0690, 0.2492, 0.1255, 0.0331], + device='cuda:0'), in_proj_covar=tensor([0.0415, 0.0370, 0.0392, 0.0352, 0.0376, 0.0354, 0.0385, 0.0408], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:49:03,481 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5892, 1.7699, 2.0815, 1.9082, 3.2538, 2.7665, 3.7812, 1.7499], + device='cuda:0'), covar=tensor([0.2436, 0.4274, 0.2754, 0.1825, 0.1461, 0.2004, 0.1360, 0.4180], + device='cuda:0'), in_proj_covar=tensor([0.0535, 0.0645, 0.0716, 0.0486, 0.0617, 0.0532, 0.0662, 0.0552], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-02 23:49:03,699 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.46 vs. limit=5.0 +2023-04-02 23:49:26,056 INFO [train.py:903] (0/4) Epoch 23, batch 3200, loss[loss=0.1947, simple_loss=0.2781, pruned_loss=0.0556, over 19672.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2865, pruned_loss=0.06296, over 3826020.44 frames. ], batch size: 53, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:49:54,849 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=153439.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:50:00,122 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.031e+02 5.043e+02 6.093e+02 8.078e+02 1.420e+03, threshold=1.219e+03, percent-clipped=2.0 +2023-04-02 23:50:17,427 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153459.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:50:26,644 INFO [train.py:903] (0/4) Epoch 23, batch 3250, loss[loss=0.2487, simple_loss=0.3248, pruned_loss=0.08632, over 19407.00 frames. ], tot_loss[loss=0.207, simple_loss=0.2872, pruned_loss=0.06343, over 3826506.00 frames. ], batch size: 70, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:50:43,125 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=153479.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:50:48,912 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153484.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:51:05,693 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-02 23:51:27,761 INFO [train.py:903] (0/4) Epoch 23, batch 3300, loss[loss=0.1922, simple_loss=0.2799, pruned_loss=0.05227, over 19617.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2877, pruned_loss=0.06364, over 3829999.25 frames. ], batch size: 57, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:51:34,810 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-02 23:52:00,748 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.814e+02 5.081e+02 6.216e+02 8.009e+02 2.047e+03, threshold=1.243e+03, percent-clipped=5.0 +2023-04-02 23:52:26,359 INFO [train.py:903] (0/4) Epoch 23, batch 3350, loss[loss=0.2197, simple_loss=0.2931, pruned_loss=0.07316, over 19494.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2879, pruned_loss=0.06393, over 3830731.08 frames. ], batch size: 49, lr: 3.58e-03, grad_scale: 8.0 +2023-04-02 23:52:56,022 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5786, 2.2621, 1.6124, 1.5912, 2.0626, 1.3388, 1.4465, 1.9028], + device='cuda:0'), covar=tensor([0.1095, 0.0750, 0.1180, 0.0798, 0.0609, 0.1315, 0.0736, 0.0539], + device='cuda:0'), in_proj_covar=tensor([0.0307, 0.0319, 0.0344, 0.0269, 0.0252, 0.0343, 0.0295, 0.0280], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:53:00,302 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153594.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:53:26,238 INFO [train.py:903] (0/4) Epoch 23, batch 3400, loss[loss=0.1972, simple_loss=0.2762, pruned_loss=0.0591, over 19668.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2886, pruned_loss=0.06458, over 3815667.51 frames. ], batch size: 55, lr: 3.57e-03, grad_scale: 8.0 +2023-04-02 23:53:37,516 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.46 vs. limit=2.0 +2023-04-02 23:53:47,128 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4360, 2.1040, 2.0940, 3.1337, 2.1486, 2.6398, 2.4681, 2.5402], + device='cuda:0'), covar=tensor([0.0687, 0.0800, 0.0859, 0.0654, 0.0800, 0.0668, 0.0850, 0.0547], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0219, 0.0223, 0.0237, 0.0225, 0.0211, 0.0186, 0.0202], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-02 23:53:56,978 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=153640.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:53:58,246 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3221, 1.5446, 1.9843, 1.3807, 3.0896, 4.7311, 4.6067, 5.1321], + device='cuda:0'), covar=tensor([0.1646, 0.3631, 0.3266, 0.2395, 0.0581, 0.0185, 0.0171, 0.0202], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0324, 0.0353, 0.0266, 0.0245, 0.0189, 0.0218, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-02 23:54:01,368 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.785e+02 5.295e+02 6.743e+02 8.549e+02 2.424e+03, threshold=1.349e+03, percent-clipped=5.0 +2023-04-02 23:54:17,531 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0012, 1.7378, 1.6782, 2.0847, 1.7327, 1.7465, 1.5442, 1.9007], + device='cuda:0'), covar=tensor([0.1111, 0.1611, 0.1574, 0.1017, 0.1394, 0.0588, 0.1547, 0.0806], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0360, 0.0317, 0.0256, 0.0308, 0.0254, 0.0313, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:54:28,040 INFO [train.py:903] (0/4) Epoch 23, batch 3450, loss[loss=0.2153, simple_loss=0.2988, pruned_loss=0.06588, over 19511.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2882, pruned_loss=0.06422, over 3809805.64 frames. ], batch size: 54, lr: 3.57e-03, grad_scale: 8.0 +2023-04-02 23:54:31,533 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-02 23:54:52,913 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0345, 2.1135, 2.3673, 2.7895, 2.0619, 2.6053, 2.4620, 2.1875], + device='cuda:0'), covar=tensor([0.4415, 0.4219, 0.1978, 0.2580, 0.4325, 0.2258, 0.4873, 0.3451], + device='cuda:0'), in_proj_covar=tensor([0.0907, 0.0972, 0.0719, 0.0934, 0.0882, 0.0819, 0.0847, 0.0788], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-02 23:55:30,145 INFO [train.py:903] (0/4) Epoch 23, batch 3500, loss[loss=0.2545, simple_loss=0.3233, pruned_loss=0.09284, over 18193.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2873, pruned_loss=0.06354, over 3810487.92 frames. ], batch size: 83, lr: 3.57e-03, grad_scale: 8.0 +2023-04-02 23:56:02,449 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.496e+02 4.938e+02 5.821e+02 7.521e+02 2.332e+03, threshold=1.164e+03, percent-clipped=1.0 +2023-04-02 23:56:17,935 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153755.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:56:30,178 INFO [train.py:903] (0/4) Epoch 23, batch 3550, loss[loss=0.1955, simple_loss=0.2801, pruned_loss=0.05548, over 19767.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2877, pruned_loss=0.06381, over 3802608.43 frames. ], batch size: 54, lr: 3.57e-03, grad_scale: 8.0 +2023-04-02 23:56:50,046 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=153783.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:57:30,137 INFO [train.py:903] (0/4) Epoch 23, batch 3600, loss[loss=0.2203, simple_loss=0.2942, pruned_loss=0.07317, over 19353.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.2869, pruned_loss=0.06307, over 3807244.08 frames. ], batch size: 70, lr: 3.57e-03, grad_scale: 8.0 +2023-04-02 23:57:47,205 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-02 23:58:05,061 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.445e+02 5.155e+02 6.351e+02 8.015e+02 2.586e+03, threshold=1.270e+03, percent-clipped=6.0 +2023-04-02 23:58:12,348 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=153850.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:58:30,879 INFO [train.py:903] (0/4) Epoch 23, batch 3650, loss[loss=0.2454, simple_loss=0.3255, pruned_loss=0.08262, over 19466.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.2866, pruned_loss=0.06304, over 3807941.45 frames. ], batch size: 64, lr: 3.57e-03, grad_scale: 8.0 +2023-04-02 23:58:42,969 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=153875.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:59:08,904 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=153898.0, num_to_drop=0, layers_to_drop=set() +2023-04-02 23:59:31,666 INFO [train.py:903] (0/4) Epoch 23, batch 3700, loss[loss=0.1835, simple_loss=0.2547, pruned_loss=0.05612, over 19753.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2863, pruned_loss=0.06292, over 3818112.85 frames. ], batch size: 46, lr: 3.57e-03, grad_scale: 8.0 +2023-04-02 23:59:57,424 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0120, 0.9107, 1.3441, 1.2425, 2.4635, 1.0284, 2.1976, 2.8021], + device='cuda:0'), covar=tensor([0.0804, 0.3957, 0.3314, 0.2249, 0.1203, 0.2823, 0.1331, 0.0523], + device='cuda:0'), in_proj_covar=tensor([0.0413, 0.0369, 0.0391, 0.0352, 0.0375, 0.0352, 0.0385, 0.0406], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-02 23:59:57,955 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.82 vs. limit=2.0 +2023-04-03 00:00:00,727 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=153941.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:00:04,690 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.006e+02 4.617e+02 5.510e+02 6.874e+02 2.344e+03, threshold=1.102e+03, percent-clipped=3.0 +2023-04-03 00:00:31,961 INFO [train.py:903] (0/4) Epoch 23, batch 3750, loss[loss=0.1786, simple_loss=0.2599, pruned_loss=0.04868, over 19792.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2871, pruned_loss=0.0633, over 3801886.30 frames. ], batch size: 47, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:01:13,574 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-154000.pt +2023-04-03 00:01:24,685 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-03 00:01:27,882 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154011.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:01:33,255 INFO [train.py:903] (0/4) Epoch 23, batch 3800, loss[loss=0.2097, simple_loss=0.2841, pruned_loss=0.06761, over 19673.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2862, pruned_loss=0.06283, over 3801970.15 frames. ], batch size: 53, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:01:59,820 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154036.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:02:05,110 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-03 00:02:08,273 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.228e+02 4.923e+02 6.103e+02 7.526e+02 2.694e+03, threshold=1.221e+03, percent-clipped=9.0 +2023-04-03 00:02:33,013 INFO [train.py:903] (0/4) Epoch 23, batch 3850, loss[loss=0.2641, simple_loss=0.3327, pruned_loss=0.0977, over 18771.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.2867, pruned_loss=0.06314, over 3807907.23 frames. ], batch size: 74, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:02:35,275 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154067.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 00:03:01,621 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.86 vs. limit=2.0 +2023-04-03 00:03:09,332 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154094.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:03:18,567 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154102.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:03:35,952 INFO [train.py:903] (0/4) Epoch 23, batch 3900, loss[loss=0.2172, simple_loss=0.296, pruned_loss=0.06923, over 19499.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2864, pruned_loss=0.06265, over 3819693.58 frames. ], batch size: 64, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:04:09,484 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.867e+02 4.608e+02 5.656e+02 7.392e+02 1.919e+03, threshold=1.131e+03, percent-clipped=3.0 +2023-04-03 00:04:22,810 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154154.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:04:37,505 INFO [train.py:903] (0/4) Epoch 23, batch 3950, loss[loss=0.1923, simple_loss=0.2668, pruned_loss=0.05891, over 19627.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2869, pruned_loss=0.06328, over 3822555.97 frames. ], batch size: 50, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:04:44,239 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-03 00:04:52,288 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154179.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:05:37,012 INFO [train.py:903] (0/4) Epoch 23, batch 4000, loss[loss=0.1786, simple_loss=0.2584, pruned_loss=0.04942, over 19404.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2878, pruned_loss=0.06369, over 3815891.87 frames. ], batch size: 47, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:06:06,076 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154239.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:06:12,359 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.243e+02 5.130e+02 6.145e+02 8.525e+02 2.203e+03, threshold=1.229e+03, percent-clipped=9.0 +2023-04-03 00:06:27,014 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-03 00:06:37,054 INFO [train.py:903] (0/4) Epoch 23, batch 4050, loss[loss=0.2265, simple_loss=0.307, pruned_loss=0.07301, over 17400.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.2865, pruned_loss=0.06303, over 3825975.24 frames. ], batch size: 101, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:07:01,630 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154285.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:07:17,517 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5019, 1.0939, 1.3665, 1.2011, 2.0880, 1.0208, 2.1144, 2.4575], + device='cuda:0'), covar=tensor([0.0932, 0.3196, 0.3040, 0.1906, 0.1154, 0.2254, 0.1186, 0.0530], + device='cuda:0'), in_proj_covar=tensor([0.0416, 0.0371, 0.0394, 0.0355, 0.0377, 0.0354, 0.0387, 0.0409], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 00:07:20,311 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-04-03 00:07:23,111 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154304.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:07:29,627 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154309.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:07:37,604 INFO [train.py:903] (0/4) Epoch 23, batch 4100, loss[loss=0.239, simple_loss=0.3179, pruned_loss=0.08007, over 19631.00 frames. ], tot_loss[loss=0.207, simple_loss=0.2871, pruned_loss=0.06346, over 3833769.85 frames. ], batch size: 57, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:08:11,156 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.231e+02 5.105e+02 5.940e+02 7.682e+02 1.555e+03, threshold=1.188e+03, percent-clipped=4.0 +2023-04-03 00:08:13,567 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-03 00:08:16,177 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-03 00:08:25,562 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154355.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 00:08:39,412 INFO [train.py:903] (0/4) Epoch 23, batch 4150, loss[loss=0.2186, simple_loss=0.2999, pruned_loss=0.06862, over 19686.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2864, pruned_loss=0.06314, over 3814555.99 frames. ], batch size: 60, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:09:22,099 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154400.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:09:33,964 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154411.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 00:09:39,114 INFO [train.py:903] (0/4) Epoch 23, batch 4200, loss[loss=0.2116, simple_loss=0.2917, pruned_loss=0.06572, over 19325.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2869, pruned_loss=0.06327, over 3814885.01 frames. ], batch size: 70, lr: 3.57e-03, grad_scale: 8.0 +2023-04-03 00:09:41,433 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-03 00:10:07,054 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154438.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:10:14,760 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.103e+02 4.744e+02 5.863e+02 7.378e+02 1.705e+03, threshold=1.173e+03, percent-clipped=3.0 +2023-04-03 00:10:17,212 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154446.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:10:40,149 INFO [train.py:903] (0/4) Epoch 23, batch 4250, loss[loss=0.1843, simple_loss=0.272, pruned_loss=0.04827, over 19753.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2861, pruned_loss=0.06291, over 3812220.11 frames. ], batch size: 51, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:10:54,220 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-03 00:11:05,308 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-03 00:11:12,043 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154492.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:11:40,260 INFO [train.py:903] (0/4) Epoch 23, batch 4300, loss[loss=0.207, simple_loss=0.2877, pruned_loss=0.06316, over 19682.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.286, pruned_loss=0.06333, over 3813084.01 frames. ], batch size: 60, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:11:53,648 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154526.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 00:12:02,711 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0865, 5.0910, 5.8385, 5.8704, 2.0816, 5.5896, 4.7290, 5.5153], + device='cuda:0'), covar=tensor([0.1624, 0.0832, 0.0555, 0.0608, 0.6009, 0.0801, 0.0621, 0.1107], + device='cuda:0'), in_proj_covar=tensor([0.0784, 0.0745, 0.0950, 0.0835, 0.0835, 0.0712, 0.0568, 0.0883], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 00:12:13,328 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.268e+02 4.652e+02 5.888e+02 7.584e+02 1.931e+03, threshold=1.178e+03, percent-clipped=3.0 +2023-04-03 00:12:24,496 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154553.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:12:33,752 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-03 00:12:35,188 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154561.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:12:41,455 INFO [train.py:903] (0/4) Epoch 23, batch 4350, loss[loss=0.2082, simple_loss=0.2959, pruned_loss=0.06023, over 19520.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.286, pruned_loss=0.06324, over 3823068.52 frames. ], batch size: 54, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:13:01,183 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154583.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:13:40,254 INFO [train.py:903] (0/4) Epoch 23, batch 4400, loss[loss=0.2304, simple_loss=0.3138, pruned_loss=0.07355, over 19676.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2871, pruned_loss=0.06364, over 3822153.66 frames. ], batch size: 55, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:13:46,129 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2097, 1.3030, 1.7306, 1.4307, 2.7202, 3.7611, 3.4660, 4.0212], + device='cuda:0'), covar=tensor([0.1733, 0.3942, 0.3467, 0.2402, 0.0634, 0.0200, 0.0223, 0.0252], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0322, 0.0351, 0.0263, 0.0243, 0.0187, 0.0215, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 00:14:04,361 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-03 00:14:14,043 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.271e+02 5.197e+02 6.555e+02 7.915e+02 1.480e+03, threshold=1.311e+03, percent-clipped=6.0 +2023-04-03 00:14:15,161 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-03 00:14:16,465 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154646.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:14:18,497 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154648.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:14:23,872 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154653.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:14:27,560 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154656.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:14:38,570 INFO [train.py:903] (0/4) Epoch 23, batch 4450, loss[loss=0.2608, simple_loss=0.3296, pruned_loss=0.09598, over 19570.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2867, pruned_loss=0.06343, over 3826652.57 frames. ], batch size: 61, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:14:58,401 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154681.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:15:18,468 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154698.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:15:19,363 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154699.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 00:15:38,748 INFO [train.py:903] (0/4) Epoch 23, batch 4500, loss[loss=0.2504, simple_loss=0.3251, pruned_loss=0.08783, over 17507.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2885, pruned_loss=0.0645, over 3821959.84 frames. ], batch size: 101, lr: 3.56e-03, grad_scale: 4.0 +2023-04-03 00:15:59,640 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7406, 1.5472, 1.5387, 2.2267, 1.5345, 2.0879, 2.0210, 1.8672], + device='cuda:0'), covar=tensor([0.0810, 0.0955, 0.1008, 0.0722, 0.0911, 0.0732, 0.0843, 0.0659], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0221, 0.0226, 0.0240, 0.0227, 0.0212, 0.0187, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-03 00:16:01,943 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2813, 1.3601, 1.6127, 1.5021, 2.2521, 1.9840, 2.3227, 0.9403], + device='cuda:0'), covar=tensor([0.2772, 0.4652, 0.2947, 0.2152, 0.1576, 0.2453, 0.1510, 0.4850], + device='cuda:0'), in_proj_covar=tensor([0.0539, 0.0647, 0.0720, 0.0488, 0.0620, 0.0534, 0.0660, 0.0554], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 00:16:06,361 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154738.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:16:13,910 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.080e+02 4.922e+02 6.448e+02 7.735e+02 1.395e+03, threshold=1.290e+03, percent-clipped=1.0 +2023-04-03 00:16:37,119 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154763.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:16:41,139 INFO [train.py:903] (0/4) Epoch 23, batch 4550, loss[loss=0.1958, simple_loss=0.2791, pruned_loss=0.05626, over 19774.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2881, pruned_loss=0.06443, over 3820327.17 frames. ], batch size: 54, lr: 3.56e-03, grad_scale: 4.0 +2023-04-03 00:16:43,802 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154768.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:16:48,238 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-03 00:17:00,064 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154782.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 00:17:11,893 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-03 00:17:31,839 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154807.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 00:17:34,145 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154809.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:17:39,823 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154814.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 00:17:41,844 INFO [train.py:903] (0/4) Epoch 23, batch 4600, loss[loss=0.2031, simple_loss=0.2863, pruned_loss=0.05994, over 19696.00 frames. ], tot_loss[loss=0.2087, simple_loss=0.2885, pruned_loss=0.06449, over 3814342.12 frames. ], batch size: 59, lr: 3.56e-03, grad_scale: 4.0 +2023-04-03 00:17:43,457 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154817.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:18:02,716 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154834.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:18:04,525 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154836.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:18:13,443 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154842.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:18:17,274 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.766e+02 4.757e+02 5.456e+02 7.137e+02 2.039e+03, threshold=1.091e+03, percent-clipped=4.0 +2023-04-03 00:18:41,886 INFO [train.py:903] (0/4) Epoch 23, batch 4650, loss[loss=0.2074, simple_loss=0.2909, pruned_loss=0.06189, over 19722.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2875, pruned_loss=0.06371, over 3806874.94 frames. ], batch size: 63, lr: 3.56e-03, grad_scale: 4.0 +2023-04-03 00:18:57,533 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-03 00:19:09,934 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-03 00:19:42,546 INFO [train.py:903] (0/4) Epoch 23, batch 4700, loss[loss=0.2285, simple_loss=0.3063, pruned_loss=0.07539, over 18388.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2867, pruned_loss=0.06343, over 3794080.62 frames. ], batch size: 84, lr: 3.56e-03, grad_scale: 4.0 +2023-04-03 00:20:04,430 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-03 00:20:17,977 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.052e+02 4.550e+02 5.511e+02 7.065e+02 1.410e+03, threshold=1.102e+03, percent-clipped=2.0 +2023-04-03 00:20:25,028 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=154951.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:20:28,237 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=154954.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:20:37,193 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1323, 1.3400, 1.5574, 1.4346, 2.7235, 1.0970, 2.3007, 3.0947], + device='cuda:0'), covar=tensor([0.0569, 0.2804, 0.2695, 0.1767, 0.0747, 0.2339, 0.1073, 0.0321], + device='cuda:0'), in_proj_covar=tensor([0.0412, 0.0367, 0.0390, 0.0350, 0.0374, 0.0351, 0.0384, 0.0406], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 00:20:44,143 INFO [train.py:903] (0/4) Epoch 23, batch 4750, loss[loss=0.1842, simple_loss=0.266, pruned_loss=0.05121, over 19466.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2866, pruned_loss=0.06311, over 3809423.41 frames. ], batch size: 49, lr: 3.56e-03, grad_scale: 4.0 +2023-04-03 00:21:00,237 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=154979.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:21:11,339 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8944, 1.1350, 1.4458, 0.6387, 1.8937, 2.1918, 1.9597, 2.3298], + device='cuda:0'), covar=tensor([0.1623, 0.3751, 0.3258, 0.2853, 0.0793, 0.0385, 0.0394, 0.0440], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0324, 0.0354, 0.0264, 0.0246, 0.0189, 0.0217, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 00:21:12,322 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=154990.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:21:17,039 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=154994.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:21:45,322 INFO [train.py:903] (0/4) Epoch 23, batch 4800, loss[loss=0.1994, simple_loss=0.2832, pruned_loss=0.05778, over 19337.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2876, pruned_loss=0.0636, over 3798697.62 frames. ], batch size: 66, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:21:49,034 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155019.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:21:54,235 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155024.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:22:18,855 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155044.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:22:19,564 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.470e+02 5.324e+02 6.216e+02 7.674e+02 2.163e+03, threshold=1.243e+03, percent-clipped=8.0 +2023-04-03 00:22:26,085 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155049.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:22:44,538 INFO [train.py:903] (0/4) Epoch 23, batch 4850, loss[loss=0.228, simple_loss=0.3144, pruned_loss=0.07083, over 19390.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2874, pruned_loss=0.06359, over 3791784.02 frames. ], batch size: 70, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:22:49,269 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155070.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 00:23:03,350 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=155082.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:23:09,627 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-03 00:23:11,841 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155087.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:23:21,471 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155095.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 00:23:28,250 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1578, 1.8567, 2.0578, 2.8716, 1.9830, 2.4288, 2.4558, 2.2357], + device='cuda:0'), covar=tensor([0.0773, 0.0891, 0.0911, 0.0726, 0.0864, 0.0761, 0.0894, 0.0624], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0222, 0.0226, 0.0241, 0.0227, 0.0213, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 00:23:29,054 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-03 00:23:32,628 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155105.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:23:34,406 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-03 00:23:34,436 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-03 00:23:44,625 INFO [train.py:903] (0/4) Epoch 23, batch 4900, loss[loss=0.1988, simple_loss=0.2877, pruned_loss=0.05496, over 18852.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2886, pruned_loss=0.06449, over 3792269.59 frames. ], batch size: 74, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:23:44,639 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-03 00:24:04,394 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-03 00:24:20,234 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.924e+02 5.163e+02 5.938e+02 7.647e+02 1.407e+03, threshold=1.188e+03, percent-clipped=5.0 +2023-04-03 00:24:46,212 INFO [train.py:903] (0/4) Epoch 23, batch 4950, loss[loss=0.2024, simple_loss=0.2901, pruned_loss=0.05736, over 19638.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2888, pruned_loss=0.06442, over 3789670.27 frames. ], batch size: 55, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:25:01,074 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-03 00:25:21,570 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155197.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:25:22,303 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-03 00:25:34,809 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155207.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:25:44,533 INFO [train.py:903] (0/4) Epoch 23, batch 5000, loss[loss=0.1824, simple_loss=0.2563, pruned_loss=0.0543, over 19743.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2881, pruned_loss=0.06455, over 3790227.23 frames. ], batch size: 46, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:25:52,527 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-03 00:26:02,909 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155232.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:26:03,634 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-03 00:26:19,060 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.096e+02 4.751e+02 5.889e+02 7.363e+02 1.722e+03, threshold=1.178e+03, percent-clipped=5.0 +2023-04-03 00:26:43,568 INFO [train.py:903] (0/4) Epoch 23, batch 5050, loss[loss=0.2217, simple_loss=0.3024, pruned_loss=0.07053, over 19761.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2881, pruned_loss=0.06455, over 3794200.23 frames. ], batch size: 54, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:27:17,599 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-03 00:27:42,609 INFO [train.py:903] (0/4) Epoch 23, batch 5100, loss[loss=0.2236, simple_loss=0.3043, pruned_loss=0.07143, over 19737.00 frames. ], tot_loss[loss=0.209, simple_loss=0.2886, pruned_loss=0.06467, over 3789938.38 frames. ], batch size: 63, lr: 3.56e-03, grad_scale: 8.0 +2023-04-03 00:27:53,109 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-03 00:27:56,475 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-03 00:28:01,511 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-03 00:28:10,591 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=155338.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:28:11,044 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.61 vs. limit=5.0 +2023-04-03 00:28:18,268 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.450e+02 5.084e+02 6.467e+02 7.878e+02 1.414e+03, threshold=1.293e+03, percent-clipped=6.0 +2023-04-03 00:28:36,963 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155361.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:28:43,672 INFO [train.py:903] (0/4) Epoch 23, batch 5150, loss[loss=0.2303, simple_loss=0.3067, pruned_loss=0.07696, over 19762.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2887, pruned_loss=0.06458, over 3804029.86 frames. ], batch size: 51, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:28:56,737 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-03 00:29:08,611 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155386.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:29:30,423 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 00:29:45,071 INFO [train.py:903] (0/4) Epoch 23, batch 5200, loss[loss=0.2463, simple_loss=0.3232, pruned_loss=0.08471, over 19657.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.2881, pruned_loss=0.06436, over 3809949.03 frames. ], batch size: 55, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:29:58,666 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-03 00:30:02,277 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=155431.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:30:19,768 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.141e+02 5.305e+02 6.432e+02 7.969e+02 2.733e+03, threshold=1.286e+03, percent-clipped=6.0 +2023-04-03 00:30:30,719 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155453.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:30:30,789 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155453.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:30:34,230 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-03 00:30:41,477 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-03 00:30:44,906 INFO [train.py:903] (0/4) Epoch 23, batch 5250, loss[loss=0.1753, simple_loss=0.2486, pruned_loss=0.05097, over 19793.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.2874, pruned_loss=0.06393, over 3822595.38 frames. ], batch size: 48, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:30:55,505 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155475.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:30:59,135 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155478.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:31:45,245 INFO [train.py:903] (0/4) Epoch 23, batch 5300, loss[loss=0.2377, simple_loss=0.3165, pruned_loss=0.07947, over 19723.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2874, pruned_loss=0.06403, over 3810833.90 frames. ], batch size: 63, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:32:03,696 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-03 00:32:21,415 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.105e+02 4.703e+02 5.856e+02 7.687e+02 1.612e+03, threshold=1.171e+03, percent-clipped=4.0 +2023-04-03 00:32:22,928 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155546.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:32:46,436 INFO [train.py:903] (0/4) Epoch 23, batch 5350, loss[loss=0.2125, simple_loss=0.2927, pruned_loss=0.06618, over 19350.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2868, pruned_loss=0.06329, over 3807721.63 frames. ], batch size: 66, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:33:18,084 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-03 00:33:46,942 INFO [train.py:903] (0/4) Epoch 23, batch 5400, loss[loss=0.2445, simple_loss=0.317, pruned_loss=0.08599, over 19654.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2869, pruned_loss=0.06325, over 3818870.31 frames. ], batch size: 55, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:33:56,238 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155623.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:34:21,898 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.892e+02 4.747e+02 5.806e+02 7.220e+02 1.360e+03, threshold=1.161e+03, percent-clipped=1.0 +2023-04-03 00:34:48,075 INFO [train.py:903] (0/4) Epoch 23, batch 5450, loss[loss=0.2377, simple_loss=0.3133, pruned_loss=0.08102, over 19491.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2871, pruned_loss=0.06339, over 3816152.41 frames. ], batch size: 64, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:34:58,337 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155675.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:35:39,944 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155709.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:35:47,576 INFO [train.py:903] (0/4) Epoch 23, batch 5500, loss[loss=0.2487, simple_loss=0.3252, pruned_loss=0.08604, over 19566.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2884, pruned_loss=0.06425, over 3817116.94 frames. ], batch size: 61, lr: 3.55e-03, grad_scale: 4.0 +2023-04-03 00:36:09,966 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155734.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:36:13,479 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-03 00:36:24,173 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.224e+02 5.057e+02 6.298e+02 8.158e+02 1.659e+03, threshold=1.260e+03, percent-clipped=6.0 +2023-04-03 00:36:25,533 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8045, 3.4686, 2.5252, 3.1687, 0.9598, 3.4424, 3.3380, 3.4076], + device='cuda:0'), covar=tensor([0.0807, 0.1125, 0.1829, 0.0934, 0.3740, 0.0774, 0.0976, 0.1135], + device='cuda:0'), in_proj_covar=tensor([0.0518, 0.0418, 0.0502, 0.0352, 0.0405, 0.0442, 0.0433, 0.0467], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 00:36:46,709 INFO [train.py:903] (0/4) Epoch 23, batch 5550, loss[loss=0.1938, simple_loss=0.2655, pruned_loss=0.06099, over 19055.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2886, pruned_loss=0.06454, over 3822408.88 frames. ], batch size: 42, lr: 3.55e-03, grad_scale: 4.0 +2023-04-03 00:36:56,203 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-03 00:37:30,487 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=155802.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:37:30,623 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=155802.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:37:42,233 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-03 00:37:48,755 INFO [train.py:903] (0/4) Epoch 23, batch 5600, loss[loss=0.1957, simple_loss=0.2809, pruned_loss=0.05526, over 19612.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2877, pruned_loss=0.06416, over 3820001.76 frames. ], batch size: 57, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:37:52,342 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=155819.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:38:02,600 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=155827.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:38:23,384 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.767e+02 5.365e+02 7.033e+02 8.601e+02 1.530e+03, threshold=1.407e+03, percent-clipped=6.0 +2023-04-03 00:38:29,809 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3473, 1.9569, 1.5394, 1.0943, 1.9497, 1.0528, 1.2086, 1.8971], + device='cuda:0'), covar=tensor([0.1037, 0.0796, 0.1060, 0.1115, 0.0571, 0.1452, 0.0827, 0.0444], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0314, 0.0337, 0.0265, 0.0247, 0.0339, 0.0290, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 00:38:48,661 INFO [train.py:903] (0/4) Epoch 23, batch 5650, loss[loss=0.2198, simple_loss=0.2981, pruned_loss=0.07076, over 19666.00 frames. ], tot_loss[loss=0.208, simple_loss=0.2873, pruned_loss=0.06437, over 3821396.45 frames. ], batch size: 55, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:39:12,712 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.43 vs. limit=2.0 +2023-04-03 00:39:33,337 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-03 00:39:47,798 INFO [train.py:903] (0/4) Epoch 23, batch 5700, loss[loss=0.2411, simple_loss=0.3144, pruned_loss=0.0839, over 19571.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.286, pruned_loss=0.06347, over 3829748.76 frames. ], batch size: 61, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:40:04,254 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-04-03 00:40:10,823 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=155934.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:40:24,776 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.238e+02 4.906e+02 6.184e+02 7.924e+02 2.131e+03, threshold=1.237e+03, percent-clipped=2.0 +2023-04-03 00:40:47,843 INFO [train.py:903] (0/4) Epoch 23, batch 5750, loss[loss=0.1759, simple_loss=0.2632, pruned_loss=0.04424, over 19588.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2864, pruned_loss=0.06344, over 3827825.69 frames. ], batch size: 52, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:40:49,188 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=155967.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:40:51,093 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-03 00:40:58,779 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-03 00:41:04,139 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-03 00:41:29,692 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-156000.pt +2023-04-03 00:41:48,390 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4555, 1.5620, 1.8088, 1.7018, 2.4079, 2.2028, 2.5646, 1.0881], + device='cuda:0'), covar=tensor([0.2510, 0.4253, 0.2660, 0.1968, 0.1584, 0.2249, 0.1456, 0.4498], + device='cuda:0'), in_proj_covar=tensor([0.0535, 0.0645, 0.0718, 0.0488, 0.0620, 0.0534, 0.0657, 0.0551], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 00:41:50,978 INFO [train.py:903] (0/4) Epoch 23, batch 5800, loss[loss=0.2168, simple_loss=0.2943, pruned_loss=0.06963, over 19614.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2872, pruned_loss=0.06359, over 3839970.33 frames. ], batch size: 61, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:41:54,502 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=156019.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:42:19,973 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9539, 1.9351, 1.7964, 1.5590, 1.5142, 1.5664, 0.3378, 0.8655], + device='cuda:0'), covar=tensor([0.0676, 0.0646, 0.0478, 0.0784, 0.1267, 0.0934, 0.1434, 0.1182], + device='cuda:0'), in_proj_covar=tensor([0.0361, 0.0357, 0.0363, 0.0387, 0.0466, 0.0393, 0.0339, 0.0345], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 00:42:25,081 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.381e+02 4.982e+02 6.301e+02 7.857e+02 1.493e+03, threshold=1.260e+03, percent-clipped=3.0 +2023-04-03 00:42:50,183 INFO [train.py:903] (0/4) Epoch 23, batch 5850, loss[loss=0.1937, simple_loss=0.2702, pruned_loss=0.05862, over 19587.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2873, pruned_loss=0.06354, over 3839624.31 frames. ], batch size: 52, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:42:52,029 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.63 vs. limit=5.0 +2023-04-03 00:43:08,238 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156082.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:43:14,839 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156088.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:43:35,999 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0996, 1.3290, 1.6893, 1.2946, 2.7735, 3.6845, 3.4293, 4.0100], + device='cuda:0'), covar=tensor([0.1792, 0.4059, 0.3653, 0.2578, 0.0660, 0.0209, 0.0250, 0.0264], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0326, 0.0355, 0.0265, 0.0247, 0.0191, 0.0218, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 00:43:48,541 INFO [train.py:903] (0/4) Epoch 23, batch 5900, loss[loss=0.2376, simple_loss=0.3128, pruned_loss=0.0812, over 19690.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2875, pruned_loss=0.06356, over 3833472.01 frames. ], batch size: 53, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:43:52,961 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-03 00:44:10,841 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156134.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:44:14,056 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-03 00:44:24,575 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.271e+02 5.043e+02 6.365e+02 8.179e+02 2.050e+03, threshold=1.273e+03, percent-clipped=8.0 +2023-04-03 00:44:25,633 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=156146.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:44:48,118 INFO [train.py:903] (0/4) Epoch 23, batch 5950, loss[loss=0.2255, simple_loss=0.2998, pruned_loss=0.07563, over 19111.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2885, pruned_loss=0.06453, over 3819271.04 frames. ], batch size: 69, lr: 3.55e-03, grad_scale: 8.0 +2023-04-03 00:44:50,267 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-04-03 00:44:55,112 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7640, 1.8594, 1.9369, 2.5018, 1.8737, 2.4039, 2.0927, 1.6915], + device='cuda:0'), covar=tensor([0.4243, 0.3938, 0.2413, 0.2610, 0.3884, 0.2144, 0.5408, 0.4536], + device='cuda:0'), in_proj_covar=tensor([0.0906, 0.0976, 0.0723, 0.0935, 0.0889, 0.0824, 0.0846, 0.0788], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 00:45:19,170 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156190.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:45:30,628 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-03 00:45:33,831 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 00:45:34,640 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6393, 1.5027, 1.4806, 2.0564, 1.5025, 1.9388, 1.9439, 1.7426], + device='cuda:0'), covar=tensor([0.0819, 0.0945, 0.1035, 0.0749, 0.0952, 0.0746, 0.0841, 0.0703], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0222, 0.0225, 0.0241, 0.0226, 0.0212, 0.0189, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 00:45:47,971 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156215.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:45:48,678 INFO [train.py:903] (0/4) Epoch 23, batch 6000, loss[loss=0.1756, simple_loss=0.2536, pruned_loss=0.04882, over 19754.00 frames. ], tot_loss[loss=0.208, simple_loss=0.288, pruned_loss=0.06399, over 3830444.87 frames. ], batch size: 47, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:45:48,679 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 00:46:01,143 INFO [train.py:937] (0/4) Epoch 23, validation: loss=0.1686, simple_loss=0.2684, pruned_loss=0.03439, over 944034.00 frames. +2023-04-03 00:46:01,143 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 00:46:23,279 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156234.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:46:30,901 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7241, 1.5384, 1.5552, 2.1851, 1.5134, 2.0677, 2.0189, 1.8881], + device='cuda:0'), covar=tensor([0.0818, 0.0918, 0.1019, 0.0765, 0.0915, 0.0734, 0.0826, 0.0660], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0221, 0.0225, 0.0240, 0.0226, 0.0212, 0.0188, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 00:46:37,246 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.373e+02 4.873e+02 6.527e+02 8.069e+02 1.468e+03, threshold=1.305e+03, percent-clipped=4.0 +2023-04-03 00:46:55,676 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156261.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:47:00,357 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4001, 1.3807, 1.5717, 1.5527, 1.7188, 1.9525, 1.8339, 0.5188], + device='cuda:0'), covar=tensor([0.2395, 0.4256, 0.2677, 0.1904, 0.1703, 0.2158, 0.1414, 0.4853], + device='cuda:0'), in_proj_covar=tensor([0.0540, 0.0652, 0.0725, 0.0492, 0.0626, 0.0538, 0.0661, 0.0556], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 00:47:01,914 INFO [train.py:903] (0/4) Epoch 23, batch 6050, loss[loss=0.1905, simple_loss=0.2774, pruned_loss=0.05179, over 19800.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2873, pruned_loss=0.0641, over 3812753.83 frames. ], batch size: 56, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:47:23,879 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.08 vs. limit=2.0 +2023-04-03 00:47:53,130 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156308.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:48:02,112 INFO [train.py:903] (0/4) Epoch 23, batch 6100, loss[loss=0.2838, simple_loss=0.3384, pruned_loss=0.1146, over 13673.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2865, pruned_loss=0.06384, over 3807805.60 frames. ], batch size: 136, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:48:28,001 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156338.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:48:37,367 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.018e+02 4.768e+02 6.249e+02 8.138e+02 1.749e+03, threshold=1.250e+03, percent-clipped=2.0 +2023-04-03 00:48:58,826 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156363.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:49:01,847 INFO [train.py:903] (0/4) Epoch 23, batch 6150, loss[loss=0.328, simple_loss=0.3839, pruned_loss=0.1361, over 19460.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2879, pruned_loss=0.0649, over 3797670.60 frames. ], batch size: 64, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:49:31,008 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156390.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:49:31,820 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-03 00:49:58,228 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6854, 1.5826, 1.5548, 1.9644, 1.5089, 1.9496, 1.9239, 1.7544], + device='cuda:0'), covar=tensor([0.0798, 0.0909, 0.0972, 0.0731, 0.0870, 0.0714, 0.0817, 0.0687], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0222, 0.0225, 0.0240, 0.0227, 0.0212, 0.0188, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-03 00:50:00,606 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156415.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:50:01,347 INFO [train.py:903] (0/4) Epoch 23, batch 6200, loss[loss=0.1851, simple_loss=0.2761, pruned_loss=0.04704, over 19776.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2875, pruned_loss=0.0641, over 3798544.29 frames. ], batch size: 54, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:50:22,375 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=156432.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:50:38,864 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.999e+02 4.815e+02 5.704e+02 6.895e+02 2.552e+03, threshold=1.141e+03, percent-clipped=3.0 +2023-04-03 00:51:02,807 INFO [train.py:903] (0/4) Epoch 23, batch 6250, loss[loss=0.2417, simple_loss=0.3213, pruned_loss=0.08102, over 19599.00 frames. ], tot_loss[loss=0.2073, simple_loss=0.2872, pruned_loss=0.06368, over 3797223.59 frames. ], batch size: 61, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:51:32,679 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-03 00:52:02,820 INFO [train.py:903] (0/4) Epoch 23, batch 6300, loss[loss=0.2372, simple_loss=0.3179, pruned_loss=0.07823, over 19657.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2869, pruned_loss=0.06331, over 3819412.44 frames. ], batch size: 55, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:52:04,422 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156517.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:52:34,279 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156542.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:52:39,263 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.268e+02 5.353e+02 6.743e+02 8.019e+02 1.408e+03, threshold=1.349e+03, percent-clipped=4.0 +2023-04-03 00:52:40,687 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156547.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:53:03,411 INFO [train.py:903] (0/4) Epoch 23, batch 6350, loss[loss=0.2017, simple_loss=0.2973, pruned_loss=0.05308, over 19545.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2866, pruned_loss=0.0629, over 3814303.63 frames. ], batch size: 56, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:53:17,247 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=156578.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:53:28,411 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5630, 1.6775, 2.1438, 1.8679, 3.2465, 2.7142, 3.4936, 1.7342], + device='cuda:0'), covar=tensor([0.2544, 0.4486, 0.2761, 0.1939, 0.1491, 0.2093, 0.1537, 0.4186], + device='cuda:0'), in_proj_covar=tensor([0.0539, 0.0651, 0.0723, 0.0492, 0.0624, 0.0536, 0.0660, 0.0554], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 00:54:02,672 INFO [train.py:903] (0/4) Epoch 23, batch 6400, loss[loss=0.2039, simple_loss=0.2951, pruned_loss=0.05632, over 19608.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2886, pruned_loss=0.06428, over 3791700.80 frames. ], batch size: 57, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:54:39,391 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.510e+02 4.834e+02 5.927e+02 7.987e+02 2.615e+03, threshold=1.185e+03, percent-clipped=4.0 +2023-04-03 00:54:46,368 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=156652.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:55:04,069 INFO [train.py:903] (0/4) Epoch 23, batch 6450, loss[loss=0.1814, simple_loss=0.2641, pruned_loss=0.04937, over 19728.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2888, pruned_loss=0.06422, over 3789325.06 frames. ], batch size: 51, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:55:35,960 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156693.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:55:47,926 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-03 00:56:04,457 INFO [train.py:903] (0/4) Epoch 23, batch 6500, loss[loss=0.2567, simple_loss=0.307, pruned_loss=0.1032, over 19785.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2886, pruned_loss=0.06418, over 3778824.53 frames. ], batch size: 49, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:56:10,981 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-03 00:56:39,931 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.419e+02 5.077e+02 6.090e+02 8.057e+02 1.603e+03, threshold=1.218e+03, percent-clipped=6.0 +2023-04-03 00:57:04,759 INFO [train.py:903] (0/4) Epoch 23, batch 6550, loss[loss=0.216, simple_loss=0.2928, pruned_loss=0.06956, over 19625.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2884, pruned_loss=0.06425, over 3776477.69 frames. ], batch size: 50, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:57:06,351 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156767.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:57:06,419 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=156767.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:57:50,626 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156803.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:58:04,881 INFO [train.py:903] (0/4) Epoch 23, batch 6600, loss[loss=0.176, simple_loss=0.2504, pruned_loss=0.0508, over 19387.00 frames. ], tot_loss[loss=0.2084, simple_loss=0.2884, pruned_loss=0.06418, over 3789562.96 frames. ], batch size: 48, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:58:20,340 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156828.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:58:41,922 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.453e+02 5.163e+02 6.336e+02 8.010e+02 1.885e+03, threshold=1.267e+03, percent-clipped=5.0 +2023-04-03 00:59:05,157 INFO [train.py:903] (0/4) Epoch 23, batch 6650, loss[loss=0.2052, simple_loss=0.2868, pruned_loss=0.06178, over 19293.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2886, pruned_loss=0.06419, over 3788273.39 frames. ], batch size: 66, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 00:59:51,159 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=156903.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 00:59:59,079 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3699, 3.8748, 3.9851, 3.9729, 1.6215, 3.7924, 3.2743, 3.7260], + device='cuda:0'), covar=tensor([0.1651, 0.0901, 0.0700, 0.0797, 0.5693, 0.0905, 0.0770, 0.1171], + device='cuda:0'), in_proj_covar=tensor([0.0786, 0.0747, 0.0956, 0.0832, 0.0836, 0.0721, 0.0571, 0.0882], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 01:00:07,300 INFO [train.py:903] (0/4) Epoch 23, batch 6700, loss[loss=0.2834, simple_loss=0.3389, pruned_loss=0.114, over 13577.00 frames. ], tot_loss[loss=0.2069, simple_loss=0.2871, pruned_loss=0.06332, over 3792451.09 frames. ], batch size: 135, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 01:00:41,730 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.603e+02 5.252e+02 6.535e+02 7.903e+02 1.565e+03, threshold=1.307e+03, percent-clipped=2.0 +2023-04-03 01:00:45,394 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=156949.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:01:04,448 INFO [train.py:903] (0/4) Epoch 23, batch 6750, loss[loss=0.2041, simple_loss=0.2809, pruned_loss=0.0637, over 19740.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2888, pruned_loss=0.06412, over 3801315.98 frames. ], batch size: 46, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 01:01:13,586 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=156974.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:02:00,907 INFO [train.py:903] (0/4) Epoch 23, batch 6800, loss[loss=0.1938, simple_loss=0.2727, pruned_loss=0.05744, over 19671.00 frames. ], tot_loss[loss=0.2082, simple_loss=0.2885, pruned_loss=0.06394, over 3795605.29 frames. ], batch size: 53, lr: 3.54e-03, grad_scale: 8.0 +2023-04-03 01:02:09,484 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157023.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:02:14,927 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157028.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:02:30,223 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-23.pt +2023-04-03 01:02:44,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-03 01:02:45,298 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-03 01:02:48,304 INFO [train.py:903] (0/4) Epoch 24, batch 0, loss[loss=0.1649, simple_loss=0.2425, pruned_loss=0.04364, over 19040.00 frames. ], tot_loss[loss=0.1649, simple_loss=0.2425, pruned_loss=0.04364, over 19040.00 frames. ], batch size: 42, lr: 3.46e-03, grad_scale: 8.0 +2023-04-03 01:02:48,304 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 01:02:59,926 INFO [train.py:937] (0/4) Epoch 24, validation: loss=0.1683, simple_loss=0.2685, pruned_loss=0.03408, over 944034.00 frames. +2023-04-03 01:02:59,927 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 01:03:03,177 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.276e+02 5.212e+02 6.445e+02 8.399e+02 3.393e+03, threshold=1.289e+03, percent-clipped=7.0 +2023-04-03 01:03:05,668 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157048.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:03:12,276 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-03 01:03:57,980 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0248, 2.1615, 2.4759, 2.2987, 3.7086, 2.9660, 3.8958, 2.2125], + device='cuda:0'), covar=tensor([0.2259, 0.3768, 0.2427, 0.1673, 0.1313, 0.1905, 0.1347, 0.3457], + device='cuda:0'), in_proj_covar=tensor([0.0537, 0.0646, 0.0718, 0.0489, 0.0618, 0.0533, 0.0658, 0.0551], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 01:04:00,822 INFO [train.py:903] (0/4) Epoch 24, batch 50, loss[loss=0.1886, simple_loss=0.281, pruned_loss=0.04809, over 19653.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2851, pruned_loss=0.06257, over 858695.71 frames. ], batch size: 60, lr: 3.46e-03, grad_scale: 8.0 +2023-04-03 01:04:20,652 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=157111.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:04:32,482 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-03 01:05:01,205 INFO [train.py:903] (0/4) Epoch 24, batch 100, loss[loss=0.2231, simple_loss=0.3042, pruned_loss=0.07103, over 19533.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.2859, pruned_loss=0.06355, over 1515448.40 frames. ], batch size: 56, lr: 3.46e-03, grad_scale: 8.0 +2023-04-03 01:05:03,493 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.245e+02 5.500e+02 6.534e+02 8.918e+02 1.825e+03, threshold=1.307e+03, percent-clipped=7.0 +2023-04-03 01:05:11,334 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-03 01:05:19,869 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157160.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:06:02,079 INFO [train.py:903] (0/4) Epoch 24, batch 150, loss[loss=0.2225, simple_loss=0.308, pruned_loss=0.06852, over 19608.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2852, pruned_loss=0.06269, over 2037421.31 frames. ], batch size: 57, lr: 3.46e-03, grad_scale: 8.0 +2023-04-03 01:06:42,420 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157226.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:07:01,366 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-03 01:07:02,476 INFO [train.py:903] (0/4) Epoch 24, batch 200, loss[loss=0.2125, simple_loss=0.3008, pruned_loss=0.0621, over 19784.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2864, pruned_loss=0.06291, over 2431977.05 frames. ], batch size: 56, lr: 3.46e-03, grad_scale: 8.0 +2023-04-03 01:07:04,623 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.193e+02 4.992e+02 5.973e+02 7.088e+02 2.080e+03, threshold=1.195e+03, percent-clipped=2.0 +2023-04-03 01:07:05,909 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=157247.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:08:03,058 INFO [train.py:903] (0/4) Epoch 24, batch 250, loss[loss=0.1788, simple_loss=0.26, pruned_loss=0.0488, over 19432.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2851, pruned_loss=0.06276, over 2746783.95 frames. ], batch size: 48, lr: 3.46e-03, grad_scale: 8.0 +2023-04-03 01:08:10,680 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.57 vs. limit=5.0 +2023-04-03 01:09:03,278 INFO [train.py:903] (0/4) Epoch 24, batch 300, loss[loss=0.2388, simple_loss=0.3002, pruned_loss=0.08871, over 19470.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2854, pruned_loss=0.06289, over 2989761.28 frames. ], batch size: 49, lr: 3.46e-03, grad_scale: 8.0 +2023-04-03 01:09:06,234 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.399e+02 5.405e+02 6.557e+02 9.024e+02 1.464e+03, threshold=1.311e+03, percent-clipped=9.0 +2023-04-03 01:09:25,587 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157362.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:09:36,369 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=157372.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:09:39,461 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157374.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:10:05,065 INFO [train.py:903] (0/4) Epoch 24, batch 350, loss[loss=0.2772, simple_loss=0.3356, pruned_loss=0.1094, over 13261.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2862, pruned_loss=0.06353, over 3175550.15 frames. ], batch size: 137, lr: 3.46e-03, grad_scale: 8.0 +2023-04-03 01:10:10,685 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 01:10:12,244 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157400.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:10:27,188 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8494, 1.9044, 2.2144, 2.4211, 1.8233, 2.3125, 2.1897, 2.0195], + device='cuda:0'), covar=tensor([0.4268, 0.4076, 0.1960, 0.2499, 0.4256, 0.2286, 0.4996, 0.3500], + device='cuda:0'), in_proj_covar=tensor([0.0906, 0.0975, 0.0722, 0.0932, 0.0887, 0.0823, 0.0846, 0.0787], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 01:10:39,433 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157422.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 01:11:05,163 INFO [train.py:903] (0/4) Epoch 24, batch 400, loss[loss=0.2322, simple_loss=0.305, pruned_loss=0.0797, over 18715.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2872, pruned_loss=0.06408, over 3327415.46 frames. ], batch size: 74, lr: 3.46e-03, grad_scale: 8.0 +2023-04-03 01:11:07,648 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.933e+02 4.825e+02 6.674e+02 8.153e+02 1.427e+03, threshold=1.335e+03, percent-clipped=2.0 +2023-04-03 01:11:52,582 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157482.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:11:58,232 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157487.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:12:05,698 INFO [train.py:903] (0/4) Epoch 24, batch 450, loss[loss=0.25, simple_loss=0.317, pruned_loss=0.09148, over 19270.00 frames. ], tot_loss[loss=0.2095, simple_loss=0.2892, pruned_loss=0.06487, over 3431921.42 frames. ], batch size: 66, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:12:11,725 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8486, 1.8024, 1.9210, 1.6371, 3.4837, 1.2260, 2.6224, 3.8731], + device='cuda:0'), covar=tensor([0.0480, 0.2349, 0.2410, 0.1886, 0.0655, 0.2472, 0.1164, 0.0229], + device='cuda:0'), in_proj_covar=tensor([0.0416, 0.0371, 0.0391, 0.0352, 0.0375, 0.0352, 0.0386, 0.0408], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:12:20,134 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=157504.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:12:22,668 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3551, 1.9494, 2.0261, 1.6782, 3.9646, 1.3710, 2.8065, 4.1324], + device='cuda:0'), covar=tensor([0.0530, 0.2440, 0.2605, 0.2145, 0.0771, 0.2624, 0.1555, 0.0259], + device='cuda:0'), in_proj_covar=tensor([0.0416, 0.0371, 0.0391, 0.0352, 0.0375, 0.0352, 0.0385, 0.0407], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:12:23,944 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157507.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:12:40,658 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-03 01:12:40,690 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-03 01:12:44,467 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157524.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 01:13:08,943 INFO [train.py:903] (0/4) Epoch 24, batch 500, loss[loss=0.1882, simple_loss=0.268, pruned_loss=0.05417, over 19386.00 frames. ], tot_loss[loss=0.2089, simple_loss=0.2886, pruned_loss=0.06462, over 3524706.19 frames. ], batch size: 47, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:13:12,136 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.617e+02 5.203e+02 6.096e+02 8.720e+02 1.456e+03, threshold=1.219e+03, percent-clipped=3.0 +2023-04-03 01:14:02,838 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157586.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:14:12,207 INFO [train.py:903] (0/4) Epoch 24, batch 550, loss[loss=0.2228, simple_loss=0.2911, pruned_loss=0.07723, over 19619.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2879, pruned_loss=0.06391, over 3585630.44 frames. ], batch size: 50, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:14:21,008 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8419, 4.4110, 2.7860, 3.8269, 1.0344, 4.3303, 4.2522, 4.3550], + device='cuda:0'), covar=tensor([0.0598, 0.1016, 0.1900, 0.0847, 0.3998, 0.0674, 0.0864, 0.1004], + device='cuda:0'), in_proj_covar=tensor([0.0515, 0.0415, 0.0499, 0.0349, 0.0403, 0.0439, 0.0433, 0.0469], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:14:41,093 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157618.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:14:42,110 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157619.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:15:13,043 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157643.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:15:13,768 INFO [train.py:903] (0/4) Epoch 24, batch 600, loss[loss=0.202, simple_loss=0.2783, pruned_loss=0.0628, over 19576.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2875, pruned_loss=0.0639, over 3630435.83 frames. ], batch size: 52, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:15:15,911 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.272e+02 4.646e+02 5.574e+02 6.767e+02 1.170e+03, threshold=1.115e+03, percent-clipped=0.0 +2023-04-03 01:15:53,025 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-03 01:16:14,759 INFO [train.py:903] (0/4) Epoch 24, batch 650, loss[loss=0.169, simple_loss=0.2434, pruned_loss=0.04726, over 19364.00 frames. ], tot_loss[loss=0.2088, simple_loss=0.2887, pruned_loss=0.0644, over 3672320.26 frames. ], batch size: 47, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:16:45,908 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=157718.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:17:07,759 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4752, 1.5375, 1.7860, 1.7163, 2.8003, 2.2826, 2.8933, 1.2847], + device='cuda:0'), covar=tensor([0.2724, 0.4590, 0.3003, 0.2124, 0.1492, 0.2347, 0.1496, 0.4625], + device='cuda:0'), in_proj_covar=tensor([0.0542, 0.0651, 0.0725, 0.0494, 0.0623, 0.0537, 0.0664, 0.0555], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 01:17:14,675 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157743.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:17:15,389 INFO [train.py:903] (0/4) Epoch 24, batch 700, loss[loss=0.2291, simple_loss=0.3045, pruned_loss=0.07681, over 18147.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2879, pruned_loss=0.06388, over 3703014.88 frames. ], batch size: 83, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:17:15,552 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=157744.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:17:20,745 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.032e+02 5.232e+02 6.869e+02 8.195e+02 1.483e+03, threshold=1.374e+03, percent-clipped=7.0 +2023-04-03 01:17:44,055 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=157766.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 01:17:46,345 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157768.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:18:19,077 INFO [train.py:903] (0/4) Epoch 24, batch 750, loss[loss=0.2006, simple_loss=0.2698, pruned_loss=0.06573, over 19794.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2883, pruned_loss=0.06413, over 3724010.93 frames. ], batch size: 47, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:19:06,769 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157833.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:19:19,815 INFO [train.py:903] (0/4) Epoch 24, batch 800, loss[loss=0.1869, simple_loss=0.2602, pruned_loss=0.05685, over 19629.00 frames. ], tot_loss[loss=0.2085, simple_loss=0.2885, pruned_loss=0.06428, over 3738370.33 frames. ], batch size: 50, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:19:22,466 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4766, 2.3365, 2.1175, 2.6617, 2.3544, 2.1132, 1.8920, 2.4688], + device='cuda:0'), covar=tensor([0.1002, 0.1642, 0.1509, 0.1051, 0.1375, 0.0565, 0.1549, 0.0717], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0356, 0.0315, 0.0254, 0.0305, 0.0255, 0.0314, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:19:23,271 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.261e+02 5.074e+02 6.358e+02 8.526e+02 1.766e+03, threshold=1.272e+03, percent-clipped=4.0 +2023-04-03 01:19:30,236 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-03 01:19:31,726 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2352, 1.9821, 1.8169, 2.1057, 1.8744, 1.8429, 1.6550, 2.0720], + device='cuda:0'), covar=tensor([0.0963, 0.1410, 0.1428, 0.0927, 0.1379, 0.0564, 0.1521, 0.0698], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0357, 0.0315, 0.0254, 0.0305, 0.0255, 0.0314, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:19:37,379 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157859.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:19:48,481 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=157868.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 01:19:52,843 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=157871.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:19:59,147 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=157875.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:20:05,647 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157881.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 01:20:20,136 INFO [train.py:903] (0/4) Epoch 24, batch 850, loss[loss=0.188, simple_loss=0.2695, pruned_loss=0.05324, over 19618.00 frames. ], tot_loss[loss=0.2086, simple_loss=0.2888, pruned_loss=0.06419, over 3745198.45 frames. ], batch size: 50, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:20:27,388 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=157900.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:21:05,393 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=157930.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:21:08,626 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-03 01:21:21,347 INFO [train.py:903] (0/4) Epoch 24, batch 900, loss[loss=0.2125, simple_loss=0.2985, pruned_loss=0.06322, over 19681.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2882, pruned_loss=0.06372, over 3766265.56 frames. ], batch size: 60, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:21:25,793 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.893e+02 4.800e+02 5.808e+02 7.910e+02 1.683e+03, threshold=1.162e+03, percent-clipped=5.0 +2023-04-03 01:21:31,620 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5471, 3.0119, 3.1661, 3.2146, 1.3252, 2.9887, 2.6519, 2.7353], + device='cuda:0'), covar=tensor([0.3086, 0.2094, 0.1553, 0.1920, 0.7851, 0.2390, 0.1503, 0.2857], + device='cuda:0'), in_proj_covar=tensor([0.0789, 0.0756, 0.0965, 0.0841, 0.0844, 0.0727, 0.0579, 0.0887], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 01:22:10,104 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=157983.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 01:22:22,960 INFO [train.py:903] (0/4) Epoch 24, batch 950, loss[loss=0.2122, simple_loss=0.2999, pruned_loss=0.0622, over 18117.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2878, pruned_loss=0.06331, over 3793276.21 frames. ], batch size: 83, lr: 3.45e-03, grad_scale: 4.0 +2023-04-03 01:22:22,980 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-03 01:22:31,627 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-158000.pt +2023-04-03 01:22:38,855 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6163, 4.2251, 2.8219, 3.7155, 0.7894, 4.2261, 4.0669, 4.1514], + device='cuda:0'), covar=tensor([0.0579, 0.0941, 0.1687, 0.0816, 0.4086, 0.0607, 0.0868, 0.0986], + device='cuda:0'), in_proj_covar=tensor([0.0512, 0.0413, 0.0496, 0.0347, 0.0401, 0.0438, 0.0432, 0.0466], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:23:26,858 INFO [train.py:903] (0/4) Epoch 24, batch 1000, loss[loss=0.227, simple_loss=0.3059, pruned_loss=0.07409, over 19681.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.288, pruned_loss=0.06342, over 3800325.05 frames. ], batch size: 60, lr: 3.45e-03, grad_scale: 4.0 +2023-04-03 01:23:28,196 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158045.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:23:31,291 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.770e+02 5.075e+02 5.979e+02 8.043e+02 1.884e+03, threshold=1.196e+03, percent-clipped=5.0 +2023-04-03 01:23:34,133 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7063, 1.8070, 1.8914, 2.4488, 1.8712, 2.3311, 2.0113, 1.6172], + device='cuda:0'), covar=tensor([0.4985, 0.4598, 0.2744, 0.2951, 0.4407, 0.2488, 0.6315, 0.5087], + device='cuda:0'), in_proj_covar=tensor([0.0909, 0.0980, 0.0723, 0.0935, 0.0888, 0.0825, 0.0845, 0.0787], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 01:24:17,479 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-03 01:24:22,236 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158089.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:24:27,537 INFO [train.py:903] (0/4) Epoch 24, batch 1050, loss[loss=0.1792, simple_loss=0.2684, pruned_loss=0.04495, over 19847.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2875, pruned_loss=0.06343, over 3814203.62 frames. ], batch size: 52, lr: 3.45e-03, grad_scale: 4.0 +2023-04-03 01:24:50,219 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158114.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:24:52,247 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158115.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:24:56,173 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-03 01:25:18,954 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5236, 1.5959, 1.9835, 1.7598, 2.7992, 2.3778, 2.9687, 1.3528], + device='cuda:0'), covar=tensor([0.2389, 0.4249, 0.2587, 0.1882, 0.1421, 0.2066, 0.1357, 0.4215], + device='cuda:0'), in_proj_covar=tensor([0.0538, 0.0648, 0.0720, 0.0492, 0.0619, 0.0534, 0.0660, 0.0553], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 01:25:20,095 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158137.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 01:25:23,232 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=158140.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:25:23,325 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158140.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:25:27,541 INFO [train.py:903] (0/4) Epoch 24, batch 1100, loss[loss=0.1965, simple_loss=0.2813, pruned_loss=0.05585, over 19615.00 frames. ], tot_loss[loss=0.207, simple_loss=0.2872, pruned_loss=0.06338, over 3820129.36 frames. ], batch size: 57, lr: 3.45e-03, grad_scale: 4.0 +2023-04-03 01:25:31,913 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.699e+02 5.122e+02 6.777e+02 7.992e+02 2.032e+03, threshold=1.355e+03, percent-clipped=5.0 +2023-04-03 01:25:50,614 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158162.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 01:26:28,898 INFO [train.py:903] (0/4) Epoch 24, batch 1150, loss[loss=0.2047, simple_loss=0.2788, pruned_loss=0.06533, over 19621.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2862, pruned_loss=0.06306, over 3822133.73 frames. ], batch size: 50, lr: 3.45e-03, grad_scale: 4.0 +2023-04-03 01:26:56,843 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=158215.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:27:25,336 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158239.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 01:27:31,330 INFO [train.py:903] (0/4) Epoch 24, batch 1200, loss[loss=0.2658, simple_loss=0.3296, pruned_loss=0.101, over 13246.00 frames. ], tot_loss[loss=0.206, simple_loss=0.286, pruned_loss=0.06299, over 3825851.85 frames. ], batch size: 136, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:27:37,766 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.973e+02 4.926e+02 5.852e+02 7.782e+02 1.430e+03, threshold=1.170e+03, percent-clipped=2.0 +2023-04-03 01:27:56,352 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158264.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 01:28:02,790 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-03 01:28:21,971 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9678, 2.0299, 2.2742, 2.5163, 1.9220, 2.4275, 2.2842, 2.1299], + device='cuda:0'), covar=tensor([0.4174, 0.3777, 0.1868, 0.2519, 0.4113, 0.2200, 0.4742, 0.3180], + device='cuda:0'), in_proj_covar=tensor([0.0908, 0.0981, 0.0725, 0.0939, 0.0890, 0.0826, 0.0846, 0.0788], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 01:28:34,708 INFO [train.py:903] (0/4) Epoch 24, batch 1250, loss[loss=0.1923, simple_loss=0.2869, pruned_loss=0.04882, over 19782.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2853, pruned_loss=0.0624, over 3834887.94 frames. ], batch size: 56, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:28:43,298 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158301.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:29:14,178 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158326.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:29:20,306 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158330.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:29:35,676 INFO [train.py:903] (0/4) Epoch 24, batch 1300, loss[loss=0.204, simple_loss=0.2744, pruned_loss=0.06679, over 19763.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2848, pruned_loss=0.06194, over 3839446.41 frames. ], batch size: 47, lr: 3.45e-03, grad_scale: 8.0 +2023-04-03 01:29:40,388 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.859e+02 5.283e+02 7.241e+02 8.945e+02 2.355e+03, threshold=1.448e+03, percent-clipped=9.0 +2023-04-03 01:30:36,953 INFO [train.py:903] (0/4) Epoch 24, batch 1350, loss[loss=0.2176, simple_loss=0.2965, pruned_loss=0.06934, over 19569.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2855, pruned_loss=0.06238, over 3841089.66 frames. ], batch size: 61, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:30:47,118 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=158402.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:31:38,835 INFO [train.py:903] (0/4) Epoch 24, batch 1400, loss[loss=0.1851, simple_loss=0.2646, pruned_loss=0.0528, over 19869.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2855, pruned_loss=0.0626, over 3846845.19 frames. ], batch size: 52, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:31:43,420 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.414e+02 5.106e+02 6.656e+02 8.188e+02 2.197e+03, threshold=1.331e+03, percent-clipped=4.0 +2023-04-03 01:32:03,687 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1807, 1.1842, 1.4473, 1.2819, 2.2642, 3.1886, 2.8748, 3.5103], + device='cuda:0'), covar=tensor([0.1811, 0.5213, 0.4987, 0.2660, 0.0845, 0.0299, 0.0417, 0.0421], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0322, 0.0354, 0.0263, 0.0244, 0.0190, 0.0215, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 01:32:16,281 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2548, 1.9554, 1.5708, 1.3366, 1.8430, 1.2344, 1.1590, 1.7480], + device='cuda:0'), covar=tensor([0.0994, 0.0859, 0.1070, 0.0874, 0.0573, 0.1325, 0.0753, 0.0496], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0318, 0.0340, 0.0270, 0.0249, 0.0340, 0.0292, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:32:27,022 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=158484.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:32:39,488 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4029, 2.0605, 1.5669, 1.3983, 1.9302, 1.2266, 1.2084, 1.8073], + device='cuda:0'), covar=tensor([0.1030, 0.0850, 0.1136, 0.0899, 0.0577, 0.1328, 0.0821, 0.0511], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0317, 0.0339, 0.0270, 0.0248, 0.0340, 0.0291, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:32:40,271 INFO [train.py:903] (0/4) Epoch 24, batch 1450, loss[loss=0.2089, simple_loss=0.3002, pruned_loss=0.05881, over 19297.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.286, pruned_loss=0.06315, over 3844077.67 frames. ], batch size: 70, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:32:43,457 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-03 01:33:26,517 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7724, 4.2491, 4.4926, 4.4902, 1.6569, 4.1824, 3.6861, 4.2288], + device='cuda:0'), covar=tensor([0.1734, 0.0861, 0.0634, 0.0715, 0.6184, 0.0893, 0.0688, 0.1201], + device='cuda:0'), in_proj_covar=tensor([0.0790, 0.0756, 0.0961, 0.0839, 0.0844, 0.0727, 0.0578, 0.0888], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 01:33:41,524 INFO [train.py:903] (0/4) Epoch 24, batch 1500, loss[loss=0.2055, simple_loss=0.2901, pruned_loss=0.06043, over 19671.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2856, pruned_loss=0.063, over 3850204.28 frames. ], batch size: 55, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:33:46,124 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.633e+02 4.870e+02 5.968e+02 7.477e+02 1.869e+03, threshold=1.194e+03, percent-clipped=2.0 +2023-04-03 01:34:33,871 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158586.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:34:42,444 INFO [train.py:903] (0/4) Epoch 24, batch 1550, loss[loss=0.1855, simple_loss=0.2727, pruned_loss=0.04918, over 19840.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.2855, pruned_loss=0.06293, over 3839280.72 frames. ], batch size: 52, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:34:48,557 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158599.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:35:04,342 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158611.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:35:46,041 INFO [train.py:903] (0/4) Epoch 24, batch 1600, loss[loss=0.2078, simple_loss=0.2991, pruned_loss=0.05824, over 19344.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2856, pruned_loss=0.06269, over 3832827.44 frames. ], batch size: 66, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:35:51,808 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.433e+02 4.721e+02 6.000e+02 7.264e+02 1.836e+03, threshold=1.200e+03, percent-clipped=4.0 +2023-04-03 01:36:11,982 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1038, 2.1820, 2.3445, 2.2032, 3.1329, 2.6604, 3.2375, 2.1349], + device='cuda:0'), covar=tensor([0.1960, 0.3177, 0.2202, 0.1587, 0.1305, 0.1833, 0.1361, 0.3603], + device='cuda:0'), in_proj_covar=tensor([0.0540, 0.0650, 0.0721, 0.0492, 0.0621, 0.0537, 0.0663, 0.0557], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 01:36:12,716 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-03 01:36:28,018 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5160, 1.6207, 1.8276, 1.8568, 1.4378, 1.7754, 1.8710, 1.7090], + device='cuda:0'), covar=tensor([0.3851, 0.3129, 0.1726, 0.2067, 0.3321, 0.1899, 0.4458, 0.3070], + device='cuda:0'), in_proj_covar=tensor([0.0911, 0.0981, 0.0724, 0.0937, 0.0890, 0.0827, 0.0845, 0.0789], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 01:36:50,302 INFO [train.py:903] (0/4) Epoch 24, batch 1650, loss[loss=0.1845, simple_loss=0.2665, pruned_loss=0.05126, over 19848.00 frames. ], tot_loss[loss=0.206, simple_loss=0.2861, pruned_loss=0.06293, over 3824806.09 frames. ], batch size: 52, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:37:52,882 INFO [train.py:903] (0/4) Epoch 24, batch 1700, loss[loss=0.182, simple_loss=0.2635, pruned_loss=0.05023, over 18653.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2848, pruned_loss=0.06218, over 3832685.23 frames. ], batch size: 41, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:37:55,314 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=158746.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:37:57,371 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.031e+02 4.745e+02 5.793e+02 7.168e+02 1.456e+03, threshold=1.159e+03, percent-clipped=2.0 +2023-04-03 01:38:00,945 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3245, 3.0652, 2.2868, 2.7374, 0.8972, 3.0080, 2.9041, 3.0013], + device='cuda:0'), covar=tensor([0.0975, 0.1292, 0.1928, 0.1088, 0.3538, 0.1038, 0.1125, 0.1318], + device='cuda:0'), in_proj_covar=tensor([0.0512, 0.0414, 0.0497, 0.0346, 0.0403, 0.0438, 0.0430, 0.0463], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:38:35,593 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-03 01:38:54,538 INFO [train.py:903] (0/4) Epoch 24, batch 1750, loss[loss=0.1459, simple_loss=0.2297, pruned_loss=0.03107, over 19364.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2857, pruned_loss=0.06273, over 3828894.94 frames. ], batch size: 47, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:38:56,109 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=158795.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:39:03,105 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=158801.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:39:57,555 INFO [train.py:903] (0/4) Epoch 24, batch 1800, loss[loss=0.2432, simple_loss=0.3193, pruned_loss=0.08351, over 19759.00 frames. ], tot_loss[loss=0.206, simple_loss=0.2863, pruned_loss=0.06285, over 3836356.94 frames. ], batch size: 63, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:40:02,406 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.024e+02 4.690e+02 5.979e+02 7.282e+02 2.087e+03, threshold=1.196e+03, percent-clipped=3.0 +2023-04-03 01:40:08,475 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=158853.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 01:40:10,958 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=158855.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:40:19,324 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=158861.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:40:40,502 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=158880.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:40:56,888 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-03 01:40:58,011 INFO [train.py:903] (0/4) Epoch 24, batch 1850, loss[loss=0.2201, simple_loss=0.2987, pruned_loss=0.07072, over 19536.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2866, pruned_loss=0.06311, over 3831627.66 frames. ], batch size: 64, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:41:32,409 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-03 01:41:49,530 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8401, 4.1519, 4.7006, 4.7502, 1.6527, 4.3945, 3.7418, 4.0782], + device='cuda:0'), covar=tensor([0.2351, 0.1466, 0.0926, 0.1154, 0.8099, 0.1856, 0.1231, 0.1975], + device='cuda:0'), in_proj_covar=tensor([0.0788, 0.0757, 0.0959, 0.0838, 0.0841, 0.0727, 0.0576, 0.0886], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 01:42:00,399 INFO [train.py:903] (0/4) Epoch 24, batch 1900, loss[loss=0.1946, simple_loss=0.2887, pruned_loss=0.05023, over 19667.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2863, pruned_loss=0.06294, over 3833540.40 frames. ], batch size: 58, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:42:04,924 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.445e+02 4.936e+02 5.873e+02 7.509e+02 2.125e+03, threshold=1.175e+03, percent-clipped=8.0 +2023-04-03 01:42:18,812 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-03 01:42:24,122 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-03 01:42:47,555 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-03 01:43:00,390 INFO [train.py:903] (0/4) Epoch 24, batch 1950, loss[loss=0.2413, simple_loss=0.3086, pruned_loss=0.08701, over 13511.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2874, pruned_loss=0.06374, over 3816413.43 frames. ], batch size: 135, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:44:03,842 INFO [train.py:903] (0/4) Epoch 24, batch 2000, loss[loss=0.1911, simple_loss=0.2773, pruned_loss=0.05245, over 19525.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2869, pruned_loss=0.06369, over 3819694.62 frames. ], batch size: 54, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:44:08,591 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.425e+02 5.106e+02 6.605e+02 9.481e+02 1.726e+03, threshold=1.321e+03, percent-clipped=5.0 +2023-04-03 01:44:57,594 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159087.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:44:59,845 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-03 01:45:06,694 INFO [train.py:903] (0/4) Epoch 24, batch 2050, loss[loss=0.2269, simple_loss=0.313, pruned_loss=0.07039, over 19659.00 frames. ], tot_loss[loss=0.2075, simple_loss=0.2875, pruned_loss=0.06373, over 3816027.80 frames. ], batch size: 58, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:45:19,165 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-03 01:45:20,293 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-03 01:45:35,009 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159117.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:45:41,066 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-03 01:46:02,768 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=159139.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:46:06,486 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159142.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:46:08,201 INFO [train.py:903] (0/4) Epoch 24, batch 2100, loss[loss=0.2418, simple_loss=0.3141, pruned_loss=0.0847, over 19686.00 frames. ], tot_loss[loss=0.2072, simple_loss=0.2874, pruned_loss=0.06345, over 3815487.59 frames. ], batch size: 60, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:46:10,355 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=159145.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:46:13,635 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.527e+02 4.786e+02 5.922e+02 8.131e+02 1.881e+03, threshold=1.184e+03, percent-clipped=4.0 +2023-04-03 01:46:37,332 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-03 01:46:49,747 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159177.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:46:58,696 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-03 01:47:00,963 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1214, 1.1943, 1.4050, 1.5441, 2.7198, 1.2403, 2.2041, 3.1163], + device='cuda:0'), covar=tensor([0.0617, 0.3008, 0.3144, 0.1705, 0.0738, 0.2312, 0.1289, 0.0311], + device='cuda:0'), in_proj_covar=tensor([0.0417, 0.0373, 0.0394, 0.0354, 0.0378, 0.0355, 0.0389, 0.0409], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:47:10,234 INFO [train.py:903] (0/4) Epoch 24, batch 2150, loss[loss=0.1677, simple_loss=0.2385, pruned_loss=0.04846, over 19736.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2876, pruned_loss=0.06362, over 3823084.92 frames. ], batch size: 45, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:47:13,851 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=159197.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 01:48:12,832 INFO [train.py:903] (0/4) Epoch 24, batch 2200, loss[loss=0.2475, simple_loss=0.3307, pruned_loss=0.08218, over 19529.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.287, pruned_loss=0.06318, over 3832850.47 frames. ], batch size: 54, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:48:18,018 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.316e+02 4.881e+02 6.157e+02 7.813e+02 2.191e+03, threshold=1.231e+03, percent-clipped=6.0 +2023-04-03 01:48:25,263 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159254.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:48:33,230 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159260.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:48:36,865 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1368, 2.0489, 1.8369, 1.7389, 1.3925, 1.5656, 0.6511, 1.0941], + device='cuda:0'), covar=tensor([0.0851, 0.0777, 0.0574, 0.1002, 0.1594, 0.1304, 0.1514, 0.1276], + device='cuda:0'), in_proj_covar=tensor([0.0357, 0.0357, 0.0360, 0.0385, 0.0462, 0.0390, 0.0338, 0.0342], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 01:49:14,284 INFO [train.py:903] (0/4) Epoch 24, batch 2250, loss[loss=0.2494, simple_loss=0.3164, pruned_loss=0.09122, over 13354.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2879, pruned_loss=0.06393, over 3811302.30 frames. ], batch size: 135, lr: 3.44e-03, grad_scale: 8.0 +2023-04-03 01:49:31,837 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159308.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:49:37,375 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159312.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 01:50:16,687 INFO [train.py:903] (0/4) Epoch 24, batch 2300, loss[loss=0.2013, simple_loss=0.2891, pruned_loss=0.05672, over 19692.00 frames. ], tot_loss[loss=0.2083, simple_loss=0.2887, pruned_loss=0.06393, over 3815541.56 frames. ], batch size: 60, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:50:21,057 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.893e+02 4.864e+02 6.208e+02 8.672e+02 1.812e+03, threshold=1.242e+03, percent-clipped=10.0 +2023-04-03 01:50:31,299 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-03 01:51:19,173 INFO [train.py:903] (0/4) Epoch 24, batch 2350, loss[loss=0.2622, simple_loss=0.3302, pruned_loss=0.09705, over 13737.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2881, pruned_loss=0.06381, over 3817565.84 frames. ], batch size: 136, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:52:00,177 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-03 01:52:04,796 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=159431.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:52:17,051 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-03 01:52:19,432 INFO [train.py:903] (0/4) Epoch 24, batch 2400, loss[loss=0.2002, simple_loss=0.2774, pruned_loss=0.06147, over 19720.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2869, pruned_loss=0.06313, over 3826143.35 frames. ], batch size: 51, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:52:21,048 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3023, 1.8844, 1.8962, 2.1429, 1.8371, 1.8801, 1.7294, 2.1118], + device='cuda:0'), covar=tensor([0.0898, 0.1484, 0.1329, 0.1030, 0.1343, 0.0537, 0.1440, 0.0683], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0358, 0.0313, 0.0255, 0.0304, 0.0253, 0.0315, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 01:52:25,061 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.677e+02 4.946e+02 5.943e+02 8.368e+02 2.189e+03, threshold=1.189e+03, percent-clipped=6.0 +2023-04-03 01:53:21,819 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0251, 1.2836, 1.7129, 1.1383, 2.4238, 3.2944, 2.9916, 3.5104], + device='cuda:0'), covar=tensor([0.1774, 0.3898, 0.3415, 0.2726, 0.0707, 0.0231, 0.0253, 0.0301], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0325, 0.0355, 0.0265, 0.0246, 0.0190, 0.0216, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 01:53:22,623 INFO [train.py:903] (0/4) Epoch 24, batch 2450, loss[loss=0.2042, simple_loss=0.288, pruned_loss=0.06021, over 19746.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2857, pruned_loss=0.06259, over 3822071.91 frames. ], batch size: 63, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:53:42,077 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159510.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:53:47,923 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-03 01:53:49,981 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159516.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:53:55,500 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=159521.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:54:13,538 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159535.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:54:21,689 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159541.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:54:24,657 INFO [train.py:903] (0/4) Epoch 24, batch 2500, loss[loss=0.191, simple_loss=0.2692, pruned_loss=0.05639, over 19721.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2859, pruned_loss=0.06267, over 3817150.68 frames. ], batch size: 51, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:54:27,427 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159546.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:54:29,311 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.080e+02 4.941e+02 6.110e+02 7.649e+02 1.406e+03, threshold=1.222e+03, percent-clipped=1.0 +2023-04-03 01:54:55,452 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159568.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 01:54:56,911 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.46 vs. limit=2.0 +2023-04-03 01:55:25,694 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159593.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 01:55:26,488 INFO [train.py:903] (0/4) Epoch 24, batch 2550, loss[loss=0.2085, simple_loss=0.2959, pruned_loss=0.06059, over 19752.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2862, pruned_loss=0.06275, over 3827172.33 frames. ], batch size: 54, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:56:19,777 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159636.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:56:20,610 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-03 01:56:29,397 INFO [train.py:903] (0/4) Epoch 24, batch 2600, loss[loss=0.2164, simple_loss=0.3, pruned_loss=0.06642, over 19588.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.286, pruned_loss=0.06248, over 3828051.65 frames. ], batch size: 52, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:56:34,952 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.356e+02 4.779e+02 5.928e+02 8.262e+02 1.528e+03, threshold=1.186e+03, percent-clipped=6.0 +2023-04-03 01:56:36,540 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159649.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:56:39,953 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=159652.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:57:31,606 INFO [train.py:903] (0/4) Epoch 24, batch 2650, loss[loss=0.2122, simple_loss=0.291, pruned_loss=0.06672, over 19772.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2854, pruned_loss=0.06198, over 3808265.45 frames. ], batch size: 54, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:57:34,981 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2612, 1.3049, 1.7538, 1.2947, 2.6093, 3.5311, 3.2381, 3.7694], + device='cuda:0'), covar=tensor([0.1654, 0.3846, 0.3371, 0.2580, 0.0642, 0.0199, 0.0232, 0.0276], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0324, 0.0356, 0.0264, 0.0246, 0.0190, 0.0216, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 01:57:43,910 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-03 01:58:21,001 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6579, 1.7926, 2.2276, 2.1696, 3.1342, 2.6585, 3.3350, 1.7611], + device='cuda:0'), covar=tensor([0.2695, 0.4575, 0.2999, 0.1983, 0.1784, 0.2337, 0.1890, 0.4443], + device='cuda:0'), in_proj_covar=tensor([0.0544, 0.0656, 0.0727, 0.0495, 0.0625, 0.0537, 0.0667, 0.0560], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 01:58:34,707 INFO [train.py:903] (0/4) Epoch 24, batch 2700, loss[loss=0.2036, simple_loss=0.288, pruned_loss=0.05964, over 19768.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2853, pruned_loss=0.06221, over 3824710.20 frames. ], batch size: 54, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:58:39,053 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.760e+02 5.237e+02 6.508e+02 8.466e+02 2.382e+03, threshold=1.302e+03, percent-clipped=8.0 +2023-04-03 01:59:03,865 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=159767.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 01:59:36,000 INFO [train.py:903] (0/4) Epoch 24, batch 2750, loss[loss=0.1936, simple_loss=0.2704, pruned_loss=0.05837, over 19611.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2868, pruned_loss=0.06274, over 3826061.13 frames. ], batch size: 50, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 01:59:46,796 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159802.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:00:19,499 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159827.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:00:40,385 INFO [train.py:903] (0/4) Epoch 24, batch 2800, loss[loss=0.2117, simple_loss=0.2977, pruned_loss=0.06287, over 19795.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.286, pruned_loss=0.06212, over 3814915.99 frames. ], batch size: 56, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 02:00:45,932 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.444e+02 4.661e+02 5.641e+02 7.181e+02 2.352e+03, threshold=1.128e+03, percent-clipped=2.0 +2023-04-03 02:01:10,957 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-04-03 02:01:16,147 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159873.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:01:40,841 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=159892.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:01:42,761 INFO [train.py:903] (0/4) Epoch 24, batch 2850, loss[loss=0.2079, simple_loss=0.2933, pruned_loss=0.06126, over 19616.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.2862, pruned_loss=0.06261, over 3812191.01 frames. ], batch size: 57, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 02:02:10,626 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=159917.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:02:25,874 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=159928.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:02:39,634 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-03 02:02:45,323 INFO [train.py:903] (0/4) Epoch 24, batch 2900, loss[loss=0.2362, simple_loss=0.3136, pruned_loss=0.07939, over 19297.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2867, pruned_loss=0.06306, over 3819558.29 frames. ], batch size: 66, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 02:02:51,071 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.210e+02 4.874e+02 6.501e+02 8.672e+02 1.518e+03, threshold=1.300e+03, percent-clipped=5.0 +2023-04-03 02:03:45,493 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=159993.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:03:46,486 INFO [train.py:903] (0/4) Epoch 24, batch 2950, loss[loss=0.2442, simple_loss=0.3294, pruned_loss=0.07947, over 19765.00 frames. ], tot_loss[loss=0.2079, simple_loss=0.2881, pruned_loss=0.06387, over 3818052.80 frames. ], batch size: 63, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 02:03:54,507 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-160000.pt +2023-04-03 02:04:24,681 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=160023.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:04:45,828 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-03 02:04:49,461 INFO [train.py:903] (0/4) Epoch 24, batch 3000, loss[loss=0.2088, simple_loss=0.2824, pruned_loss=0.06759, over 19588.00 frames. ], tot_loss[loss=0.2078, simple_loss=0.2878, pruned_loss=0.06389, over 3809799.19 frames. ], batch size: 52, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 02:04:49,462 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 02:05:02,000 INFO [train.py:937] (0/4) Epoch 24, validation: loss=0.1679, simple_loss=0.268, pruned_loss=0.03397, over 944034.00 frames. +2023-04-03 02:05:02,001 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 02:05:08,004 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=160048.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:05:08,850 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.319e+02 4.966e+02 6.275e+02 7.790e+02 1.988e+03, threshold=1.255e+03, percent-clipped=5.0 +2023-04-03 02:06:04,587 INFO [train.py:903] (0/4) Epoch 24, batch 3050, loss[loss=0.2339, simple_loss=0.3103, pruned_loss=0.07874, over 18113.00 frames. ], tot_loss[loss=0.207, simple_loss=0.2873, pruned_loss=0.06337, over 3812515.05 frames. ], batch size: 83, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 02:06:23,573 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=160108.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:06:37,826 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-04-03 02:07:06,754 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9690, 1.9177, 1.8095, 1.6383, 1.4921, 1.5964, 0.3826, 0.9514], + device='cuda:0'), covar=tensor([0.0605, 0.0599, 0.0421, 0.0665, 0.1235, 0.0803, 0.1304, 0.1075], + device='cuda:0'), in_proj_covar=tensor([0.0358, 0.0356, 0.0361, 0.0383, 0.0461, 0.0391, 0.0338, 0.0344], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 02:07:08,747 INFO [train.py:903] (0/4) Epoch 24, batch 3100, loss[loss=0.2509, simple_loss=0.3261, pruned_loss=0.08787, over 13308.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.2864, pruned_loss=0.0625, over 3808413.16 frames. ], batch size: 137, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 02:07:14,571 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.956e+02 4.859e+02 5.894e+02 7.109e+02 1.682e+03, threshold=1.179e+03, percent-clipped=4.0 +2023-04-03 02:07:40,543 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2219, 3.8140, 3.9202, 3.9199, 1.6202, 3.7336, 3.2349, 3.6688], + device='cuda:0'), covar=tensor([0.1886, 0.0971, 0.0747, 0.0866, 0.5927, 0.1029, 0.0785, 0.1201], + device='cuda:0'), in_proj_covar=tensor([0.0793, 0.0757, 0.0963, 0.0844, 0.0847, 0.0733, 0.0576, 0.0895], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 02:08:10,532 INFO [train.py:903] (0/4) Epoch 24, batch 3150, loss[loss=0.2284, simple_loss=0.2982, pruned_loss=0.07931, over 19606.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.285, pruned_loss=0.06188, over 3814296.93 frames. ], batch size: 50, lr: 3.43e-03, grad_scale: 8.0 +2023-04-03 02:08:36,006 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-03 02:08:39,732 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=160217.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:09:14,344 INFO [train.py:903] (0/4) Epoch 24, batch 3200, loss[loss=0.1979, simple_loss=0.2843, pruned_loss=0.05576, over 17366.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2852, pruned_loss=0.06211, over 3823938.17 frames. ], batch size: 101, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:09:20,052 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.297e+02 5.134e+02 6.599e+02 8.700e+02 2.161e+03, threshold=1.320e+03, percent-clipped=4.0 +2023-04-03 02:09:33,578 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9969, 1.3601, 1.7500, 1.1371, 2.5283, 3.5671, 3.2149, 3.7098], + device='cuda:0'), covar=tensor([0.1766, 0.3849, 0.3401, 0.2618, 0.0693, 0.0202, 0.0230, 0.0301], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0326, 0.0356, 0.0265, 0.0246, 0.0190, 0.0217, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 02:09:49,686 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=160272.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:10:17,062 INFO [train.py:903] (0/4) Epoch 24, batch 3250, loss[loss=0.2004, simple_loss=0.2822, pruned_loss=0.05932, over 19730.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2866, pruned_loss=0.06264, over 3824444.43 frames. ], batch size: 51, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:10:56,742 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=160325.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 02:11:04,747 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=160332.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:11:21,816 INFO [train.py:903] (0/4) Epoch 24, batch 3300, loss[loss=0.2441, simple_loss=0.3223, pruned_loss=0.08294, over 19683.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.287, pruned_loss=0.06298, over 3826789.99 frames. ], batch size: 58, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:11:24,384 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-03 02:11:27,804 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.182e+02 4.973e+02 5.834e+02 7.675e+02 1.997e+03, threshold=1.167e+03, percent-clipped=3.0 +2023-04-03 02:11:46,496 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=160364.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:12:15,589 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=160387.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:12:18,158 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=160389.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:12:23,804 INFO [train.py:903] (0/4) Epoch 24, batch 3350, loss[loss=0.215, simple_loss=0.305, pruned_loss=0.06255, over 19679.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.287, pruned_loss=0.06305, over 3824835.28 frames. ], batch size: 60, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:12:27,298 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-04-03 02:12:39,926 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.59 vs. limit=2.0 +2023-04-03 02:13:26,812 INFO [train.py:903] (0/4) Epoch 24, batch 3400, loss[loss=0.1723, simple_loss=0.2639, pruned_loss=0.04036, over 19658.00 frames. ], tot_loss[loss=0.206, simple_loss=0.2866, pruned_loss=0.06267, over 3824509.14 frames. ], batch size: 55, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:13:32,538 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.093e+02 5.133e+02 6.647e+02 9.203e+02 1.938e+03, threshold=1.329e+03, percent-clipped=8.0 +2023-04-03 02:13:48,736 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4018, 1.4926, 1.8072, 1.6467, 2.4087, 2.0648, 2.5272, 1.1402], + device='cuda:0'), covar=tensor([0.2630, 0.4409, 0.2680, 0.2035, 0.1651, 0.2347, 0.1505, 0.4589], + device='cuda:0'), in_proj_covar=tensor([0.0543, 0.0654, 0.0728, 0.0494, 0.0622, 0.0537, 0.0665, 0.0558], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 02:14:28,014 INFO [train.py:903] (0/4) Epoch 24, batch 3450, loss[loss=0.1918, simple_loss=0.264, pruned_loss=0.05983, over 19764.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2857, pruned_loss=0.06236, over 3828392.64 frames. ], batch size: 47, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:14:31,633 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-03 02:15:29,541 INFO [train.py:903] (0/4) Epoch 24, batch 3500, loss[loss=0.2021, simple_loss=0.2872, pruned_loss=0.05846, over 19537.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2848, pruned_loss=0.06148, over 3831952.84 frames. ], batch size: 54, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:15:38,051 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.068e+02 4.685e+02 5.887e+02 7.904e+02 2.662e+03, threshold=1.177e+03, percent-clipped=4.0 +2023-04-03 02:15:42,059 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3473, 2.0672, 1.6085, 1.3930, 1.8399, 1.2823, 1.2667, 1.8603], + device='cuda:0'), covar=tensor([0.0969, 0.0761, 0.1059, 0.0804, 0.0535, 0.1299, 0.0697, 0.0427], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0314, 0.0336, 0.0266, 0.0246, 0.0339, 0.0289, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:15:50,495 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-03 02:16:26,846 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=160588.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:16:33,573 INFO [train.py:903] (0/4) Epoch 24, batch 3550, loss[loss=0.2048, simple_loss=0.2945, pruned_loss=0.05753, over 18371.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2853, pruned_loss=0.06176, over 3820924.68 frames. ], batch size: 84, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:16:50,470 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3302, 3.8420, 3.9843, 3.9933, 1.6876, 3.8099, 3.3490, 3.7251], + device='cuda:0'), covar=tensor([0.1649, 0.0861, 0.0701, 0.0745, 0.5493, 0.0946, 0.0707, 0.1181], + device='cuda:0'), in_proj_covar=tensor([0.0793, 0.0757, 0.0963, 0.0842, 0.0846, 0.0733, 0.0576, 0.0895], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 02:16:57,349 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=160613.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:16:58,911 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-03 02:17:20,635 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-04-03 02:17:36,544 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=160643.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:17:37,259 INFO [train.py:903] (0/4) Epoch 24, batch 3600, loss[loss=0.1978, simple_loss=0.2758, pruned_loss=0.05987, over 19468.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2854, pruned_loss=0.06224, over 3818203.34 frames. ], batch size: 49, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:17:44,410 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.188e+02 4.811e+02 5.669e+02 7.209e+02 1.690e+03, threshold=1.134e+03, percent-clipped=3.0 +2023-04-03 02:18:07,930 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=160668.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:18:08,795 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=160669.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 02:18:29,979 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4135, 1.5009, 1.8295, 1.6638, 2.6067, 2.2894, 2.7530, 1.3054], + device='cuda:0'), covar=tensor([0.2665, 0.4452, 0.2820, 0.2061, 0.1628, 0.2191, 0.1562, 0.4456], + device='cuda:0'), in_proj_covar=tensor([0.0541, 0.0652, 0.0725, 0.0493, 0.0621, 0.0535, 0.0663, 0.0556], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 02:18:40,044 INFO [train.py:903] (0/4) Epoch 24, batch 3650, loss[loss=0.2533, simple_loss=0.3368, pruned_loss=0.08484, over 19338.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2844, pruned_loss=0.06184, over 3813934.78 frames. ], batch size: 66, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:18:48,125 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8254, 4.2203, 4.4563, 4.5157, 1.6179, 4.2191, 3.6482, 4.1724], + device='cuda:0'), covar=tensor([0.1614, 0.0967, 0.0644, 0.0688, 0.6209, 0.0904, 0.0731, 0.1158], + device='cuda:0'), in_proj_covar=tensor([0.0794, 0.0757, 0.0962, 0.0842, 0.0847, 0.0733, 0.0577, 0.0894], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 02:19:40,897 INFO [train.py:903] (0/4) Epoch 24, batch 3700, loss[loss=0.1961, simple_loss=0.2803, pruned_loss=0.05599, over 19630.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2857, pruned_loss=0.0632, over 3807126.96 frames. ], batch size: 57, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:19:49,425 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.079e+02 4.616e+02 6.347e+02 7.690e+02 1.972e+03, threshold=1.269e+03, percent-clipped=6.0 +2023-04-03 02:20:30,317 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=160784.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 02:20:44,492 INFO [train.py:903] (0/4) Epoch 24, batch 3750, loss[loss=0.2165, simple_loss=0.3033, pruned_loss=0.06486, over 19681.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2857, pruned_loss=0.06301, over 3815151.17 frames. ], batch size: 59, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:21:45,652 INFO [train.py:903] (0/4) Epoch 24, batch 3800, loss[loss=0.2598, simple_loss=0.3255, pruned_loss=0.09705, over 13226.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.2866, pruned_loss=0.06339, over 3798500.83 frames. ], batch size: 136, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:21:53,458 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.186e+02 4.635e+02 5.250e+02 7.046e+02 1.734e+03, threshold=1.050e+03, percent-clipped=4.0 +2023-04-03 02:22:18,584 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-03 02:22:22,488 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0673, 1.8680, 1.6858, 2.0824, 1.8053, 1.7740, 1.7020, 1.9691], + device='cuda:0'), covar=tensor([0.1048, 0.1432, 0.1507, 0.1052, 0.1284, 0.0544, 0.1445, 0.0725], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0356, 0.0313, 0.0255, 0.0305, 0.0254, 0.0313, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:22:41,752 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0989, 1.2528, 1.6713, 1.2135, 2.5494, 3.5015, 3.2297, 3.7050], + device='cuda:0'), covar=tensor([0.1743, 0.3963, 0.3432, 0.2616, 0.0622, 0.0211, 0.0219, 0.0243], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0325, 0.0355, 0.0264, 0.0245, 0.0190, 0.0216, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 02:22:47,087 INFO [train.py:903] (0/4) Epoch 24, batch 3850, loss[loss=0.1701, simple_loss=0.2531, pruned_loss=0.04353, over 19773.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2866, pruned_loss=0.06355, over 3803898.68 frames. ], batch size: 47, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:23:18,517 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=160919.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:23:48,689 INFO [train.py:903] (0/4) Epoch 24, batch 3900, loss[loss=0.2471, simple_loss=0.3112, pruned_loss=0.09156, over 13396.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2863, pruned_loss=0.06295, over 3797452.44 frames. ], batch size: 135, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:23:58,397 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.922e+02 5.134e+02 6.381e+02 7.692e+02 1.884e+03, threshold=1.276e+03, percent-clipped=12.0 +2023-04-03 02:24:38,062 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=160983.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:24:53,489 INFO [train.py:903] (0/4) Epoch 24, batch 3950, loss[loss=0.1841, simple_loss=0.2732, pruned_loss=0.04747, over 19676.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2854, pruned_loss=0.06235, over 3799893.41 frames. ], batch size: 59, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:24:57,060 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-03 02:25:51,208 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161040.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 02:25:55,364 INFO [train.py:903] (0/4) Epoch 24, batch 4000, loss[loss=0.1679, simple_loss=0.2515, pruned_loss=0.0421, over 19360.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.285, pruned_loss=0.06223, over 3805628.86 frames. ], batch size: 47, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:26:00,829 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.22 vs. limit=5.0 +2023-04-03 02:26:03,419 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.651e+02 5.074e+02 6.327e+02 7.723e+02 1.762e+03, threshold=1.265e+03, percent-clipped=6.0 +2023-04-03 02:26:08,607 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8616, 1.6913, 1.4954, 1.8976, 1.6242, 1.5810, 1.5592, 1.7585], + device='cuda:0'), covar=tensor([0.1168, 0.1500, 0.1719, 0.0996, 0.1379, 0.0667, 0.1545, 0.0856], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0357, 0.0313, 0.0255, 0.0305, 0.0254, 0.0314, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:26:21,921 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161065.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 02:26:41,780 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-03 02:26:57,071 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.11 vs. limit=2.0 +2023-04-03 02:26:57,367 INFO [train.py:903] (0/4) Epoch 24, batch 4050, loss[loss=0.2599, simple_loss=0.3295, pruned_loss=0.09512, over 19564.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2859, pruned_loss=0.06265, over 3811782.76 frames. ], batch size: 61, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:27:57,729 INFO [train.py:903] (0/4) Epoch 24, batch 4100, loss[loss=0.1883, simple_loss=0.2685, pruned_loss=0.05402, over 19483.00 frames. ], tot_loss[loss=0.2074, simple_loss=0.2876, pruned_loss=0.0636, over 3804923.39 frames. ], batch size: 49, lr: 3.42e-03, grad_scale: 8.0 +2023-04-03 02:27:58,485 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-04-03 02:28:06,051 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.204e+02 4.915e+02 6.129e+02 7.795e+02 1.333e+03, threshold=1.226e+03, percent-clipped=1.0 +2023-04-03 02:28:31,030 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-03 02:28:49,598 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.70 vs. limit=5.0 +2023-04-03 02:29:00,662 INFO [train.py:903] (0/4) Epoch 24, batch 4150, loss[loss=0.2221, simple_loss=0.2989, pruned_loss=0.07262, over 19536.00 frames. ], tot_loss[loss=0.2076, simple_loss=0.288, pruned_loss=0.06361, over 3798511.47 frames. ], batch size: 56, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:29:50,549 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161234.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:30:01,374 INFO [train.py:903] (0/4) Epoch 24, batch 4200, loss[loss=0.1698, simple_loss=0.2429, pruned_loss=0.04839, over 19718.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2867, pruned_loss=0.06319, over 3814459.97 frames. ], batch size: 46, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:30:02,573 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-03 02:30:08,523 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161249.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:30:09,255 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.885e+02 4.684e+02 6.110e+02 7.845e+02 2.290e+03, threshold=1.222e+03, percent-clipped=7.0 +2023-04-03 02:30:24,244 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=161263.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:31:03,257 INFO [train.py:903] (0/4) Epoch 24, batch 4250, loss[loss=0.2139, simple_loss=0.2983, pruned_loss=0.06477, over 19742.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2859, pruned_loss=0.06242, over 3826577.54 frames. ], batch size: 63, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:31:17,046 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-03 02:31:28,236 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-03 02:31:44,199 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=161327.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:31:47,940 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.74 vs. limit=2.0 +2023-04-03 02:32:04,737 INFO [train.py:903] (0/4) Epoch 24, batch 4300, loss[loss=0.2098, simple_loss=0.292, pruned_loss=0.06383, over 19665.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2846, pruned_loss=0.06166, over 3825272.95 frames. ], batch size: 55, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:32:09,652 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3134, 3.8087, 3.9236, 3.9437, 1.6089, 3.7617, 3.2439, 3.6585], + device='cuda:0'), covar=tensor([0.1755, 0.0992, 0.0669, 0.0816, 0.5760, 0.1027, 0.0773, 0.1161], + device='cuda:0'), in_proj_covar=tensor([0.0794, 0.0760, 0.0961, 0.0842, 0.0844, 0.0728, 0.0575, 0.0893], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 02:32:12,545 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.040e+02 4.585e+02 5.768e+02 7.257e+02 2.214e+03, threshold=1.154e+03, percent-clipped=5.0 +2023-04-03 02:32:36,957 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6885, 2.6378, 2.3603, 2.6775, 2.6033, 2.3234, 2.1785, 2.7218], + device='cuda:0'), covar=tensor([0.0921, 0.1395, 0.1320, 0.1017, 0.1230, 0.0489, 0.1376, 0.0619], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0358, 0.0314, 0.0256, 0.0306, 0.0254, 0.0315, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:32:47,562 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161378.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:32:47,633 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.0337, 2.8482, 2.2881, 2.2293, 2.0759, 2.5953, 1.0316, 2.1040], + device='cuda:0'), covar=tensor([0.0704, 0.0676, 0.0726, 0.1105, 0.1102, 0.1024, 0.1522, 0.1082], + device='cuda:0'), in_proj_covar=tensor([0.0360, 0.0360, 0.0364, 0.0387, 0.0465, 0.0393, 0.0341, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 02:32:57,530 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-03 02:33:06,255 INFO [train.py:903] (0/4) Epoch 24, batch 4350, loss[loss=0.1817, simple_loss=0.2728, pruned_loss=0.04528, over 19766.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.284, pruned_loss=0.06135, over 3831090.17 frames. ], batch size: 54, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:33:19,475 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1677, 1.9213, 2.0775, 2.9492, 1.9545, 2.4279, 2.5853, 2.2920], + device='cuda:0'), covar=tensor([0.0787, 0.0884, 0.0908, 0.0665, 0.0899, 0.0734, 0.0805, 0.0620], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0222, 0.0226, 0.0238, 0.0224, 0.0212, 0.0188, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-03 02:33:37,810 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161419.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:33:38,146 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.26 vs. limit=5.0 +2023-04-03 02:34:07,007 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161442.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:34:08,990 INFO [train.py:903] (0/4) Epoch 24, batch 4400, loss[loss=0.166, simple_loss=0.2449, pruned_loss=0.04351, over 19374.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2848, pruned_loss=0.06177, over 3844648.39 frames. ], batch size: 47, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:34:15,586 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.985e+02 5.009e+02 6.093e+02 7.233e+02 1.222e+03, threshold=1.219e+03, percent-clipped=3.0 +2023-04-03 02:34:31,964 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-03 02:34:41,590 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-03 02:34:50,943 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161478.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:35:10,538 INFO [train.py:903] (0/4) Epoch 24, batch 4450, loss[loss=0.2939, simple_loss=0.3513, pruned_loss=0.1183, over 13206.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2853, pruned_loss=0.06263, over 3833338.48 frames. ], batch size: 137, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:35:14,645 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-04-03 02:35:49,116 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0029, 1.9459, 1.8679, 1.6194, 1.6194, 1.6189, 0.3735, 0.9237], + device='cuda:0'), covar=tensor([0.0621, 0.0650, 0.0426, 0.0725, 0.1212, 0.0854, 0.1419, 0.1117], + device='cuda:0'), in_proj_covar=tensor([0.0359, 0.0359, 0.0361, 0.0385, 0.0463, 0.0391, 0.0339, 0.0345], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 02:36:10,897 INFO [train.py:903] (0/4) Epoch 24, batch 4500, loss[loss=0.2223, simple_loss=0.2991, pruned_loss=0.07278, over 19606.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.2862, pruned_loss=0.06344, over 3832479.34 frames. ], batch size: 57, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:36:17,773 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.067e+02 5.206e+02 6.185e+02 8.214e+02 2.130e+03, threshold=1.237e+03, percent-clipped=6.0 +2023-04-03 02:36:48,170 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8853, 1.7139, 1.4828, 1.8702, 1.5788, 1.5743, 1.5432, 1.7655], + device='cuda:0'), covar=tensor([0.1246, 0.1532, 0.1835, 0.1234, 0.1516, 0.0702, 0.1646, 0.0898], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0358, 0.0315, 0.0258, 0.0307, 0.0255, 0.0317, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:36:53,344 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=161578.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:36:57,977 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5689, 2.2823, 1.7719, 1.5336, 2.0488, 1.4143, 1.3309, 1.9862], + device='cuda:0'), covar=tensor([0.1158, 0.0849, 0.1088, 0.0961, 0.0602, 0.1447, 0.0819, 0.0511], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0317, 0.0338, 0.0267, 0.0248, 0.0341, 0.0292, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:37:10,263 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=161593.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:37:12,187 INFO [train.py:903] (0/4) Epoch 24, batch 4550, loss[loss=0.2093, simple_loss=0.2858, pruned_loss=0.06638, over 19590.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.2862, pruned_loss=0.06318, over 3833612.08 frames. ], batch size: 52, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:37:20,135 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-03 02:37:34,973 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5761, 1.6062, 1.8389, 1.8277, 2.6592, 2.4203, 2.8188, 1.1789], + device='cuda:0'), covar=tensor([0.2454, 0.4303, 0.2684, 0.1872, 0.1579, 0.2036, 0.1508, 0.4504], + device='cuda:0'), in_proj_covar=tensor([0.0546, 0.0656, 0.0732, 0.0497, 0.0629, 0.0540, 0.0667, 0.0560], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 02:37:44,726 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-03 02:38:01,460 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161634.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:38:15,061 INFO [train.py:903] (0/4) Epoch 24, batch 4600, loss[loss=0.1724, simple_loss=0.2538, pruned_loss=0.04548, over 19748.00 frames. ], tot_loss[loss=0.206, simple_loss=0.286, pruned_loss=0.06306, over 3828992.01 frames. ], batch size: 45, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:38:21,970 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.192e+02 4.894e+02 5.872e+02 7.852e+02 1.807e+03, threshold=1.174e+03, percent-clipped=3.0 +2023-04-03 02:38:33,827 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161659.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:39:15,601 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161693.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:39:16,434 INFO [train.py:903] (0/4) Epoch 24, batch 4650, loss[loss=0.2078, simple_loss=0.2989, pruned_loss=0.05837, over 19326.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2857, pruned_loss=0.06299, over 3828014.41 frames. ], batch size: 66, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:39:22,575 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161698.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:39:23,675 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1485, 1.2921, 1.8079, 1.3360, 2.8075, 3.7026, 3.4567, 3.9518], + device='cuda:0'), covar=tensor([0.1616, 0.3764, 0.3158, 0.2396, 0.0557, 0.0193, 0.0196, 0.0253], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0326, 0.0356, 0.0266, 0.0246, 0.0191, 0.0217, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 02:39:33,562 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-03 02:39:33,878 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161708.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:39:44,911 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-03 02:39:53,487 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161723.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:39:57,777 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3371, 2.0531, 1.6454, 1.3840, 1.8910, 1.3396, 1.2355, 1.8090], + device='cuda:0'), covar=tensor([0.0893, 0.0796, 0.1203, 0.0905, 0.0568, 0.1363, 0.0749, 0.0479], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0315, 0.0338, 0.0267, 0.0247, 0.0340, 0.0291, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:40:19,300 INFO [train.py:903] (0/4) Epoch 24, batch 4700, loss[loss=0.1932, simple_loss=0.2691, pruned_loss=0.05861, over 19758.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2863, pruned_loss=0.06304, over 3822364.87 frames. ], batch size: 51, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:40:26,425 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.995e+02 5.147e+02 6.134e+02 7.606e+02 1.537e+03, threshold=1.227e+03, percent-clipped=3.0 +2023-04-03 02:40:39,899 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-03 02:40:43,331 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=161763.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:41:20,593 INFO [train.py:903] (0/4) Epoch 24, batch 4750, loss[loss=0.2199, simple_loss=0.3007, pruned_loss=0.06957, over 19538.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2864, pruned_loss=0.06296, over 3823039.21 frames. ], batch size: 56, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:41:57,340 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=161822.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:42:09,924 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=161833.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:42:23,972 INFO [train.py:903] (0/4) Epoch 24, batch 4800, loss[loss=0.1996, simple_loss=0.2785, pruned_loss=0.06038, over 19656.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2859, pruned_loss=0.06288, over 3826928.84 frames. ], batch size: 53, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:42:31,545 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.081e+02 5.225e+02 6.101e+02 7.305e+02 1.695e+03, threshold=1.220e+03, percent-clipped=2.0 +2023-04-03 02:43:04,826 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161878.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:43:21,612 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0456, 1.7197, 1.8974, 1.6390, 4.5274, 1.1130, 2.5589, 4.9886], + device='cuda:0'), covar=tensor([0.0476, 0.2659, 0.2750, 0.2041, 0.0736, 0.2663, 0.1529, 0.0152], + device='cuda:0'), in_proj_covar=tensor([0.0414, 0.0368, 0.0390, 0.0350, 0.0373, 0.0352, 0.0387, 0.0407], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:43:25,943 INFO [train.py:903] (0/4) Epoch 24, batch 4850, loss[loss=0.2109, simple_loss=0.289, pruned_loss=0.06637, over 19691.00 frames. ], tot_loss[loss=0.2063, simple_loss=0.286, pruned_loss=0.06328, over 3823852.31 frames. ], batch size: 53, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:43:50,704 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-03 02:44:11,980 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-03 02:44:16,558 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-03 02:44:17,700 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-03 02:44:19,171 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=161937.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:44:27,632 INFO [train.py:903] (0/4) Epoch 24, batch 4900, loss[loss=0.2119, simple_loss=0.2996, pruned_loss=0.06206, over 19289.00 frames. ], tot_loss[loss=0.2077, simple_loss=0.2873, pruned_loss=0.06407, over 3818616.94 frames. ], batch size: 66, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:44:27,659 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-03 02:44:33,806 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161949.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:44:34,692 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.672e+02 5.435e+02 6.496e+02 8.047e+02 2.666e+03, threshold=1.299e+03, percent-clipped=6.0 +2023-04-03 02:44:47,022 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-03 02:44:53,262 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=161964.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:44:54,340 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7191, 4.2640, 4.4857, 4.4784, 1.7731, 4.2254, 3.7003, 4.1886], + device='cuda:0'), covar=tensor([0.1831, 0.0967, 0.0629, 0.0715, 0.6025, 0.1050, 0.0676, 0.1177], + device='cuda:0'), in_proj_covar=tensor([0.0794, 0.0759, 0.0966, 0.0848, 0.0847, 0.0731, 0.0578, 0.0894], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 02:45:05,687 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161974.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:45:23,760 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=161989.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:45:28,097 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4938, 1.2945, 1.2916, 1.5169, 1.2774, 1.3094, 1.2499, 1.4238], + device='cuda:0'), covar=tensor([0.0852, 0.1131, 0.1078, 0.0695, 0.1007, 0.0473, 0.1212, 0.0587], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0357, 0.0311, 0.0255, 0.0305, 0.0253, 0.0314, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:45:28,866 INFO [train.py:903] (0/4) Epoch 24, batch 4950, loss[loss=0.2689, simple_loss=0.3266, pruned_loss=0.1056, over 13276.00 frames. ], tot_loss[loss=0.207, simple_loss=0.287, pruned_loss=0.06353, over 3816586.29 frames. ], batch size: 137, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:45:36,805 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-162000.pt +2023-04-03 02:45:48,655 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-03 02:46:12,997 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-03 02:46:33,469 INFO [train.py:903] (0/4) Epoch 24, batch 5000, loss[loss=0.2108, simple_loss=0.3014, pruned_loss=0.06009, over 19535.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.2858, pruned_loss=0.06281, over 3831986.96 frames. ], batch size: 54, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:46:41,315 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.890e+02 4.589e+02 5.769e+02 7.322e+02 1.477e+03, threshold=1.154e+03, percent-clipped=3.0 +2023-04-03 02:46:44,817 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-03 02:46:56,731 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-03 02:47:35,227 INFO [train.py:903] (0/4) Epoch 24, batch 5050, loss[loss=0.2019, simple_loss=0.2922, pruned_loss=0.05581, over 19772.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.2869, pruned_loss=0.06335, over 3822755.99 frames. ], batch size: 56, lr: 3.41e-03, grad_scale: 8.0 +2023-04-03 02:47:36,638 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=162095.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:47:40,129 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9910, 1.5591, 1.8669, 1.7708, 4.4690, 1.3433, 2.5923, 4.8835], + device='cuda:0'), covar=tensor([0.0394, 0.2734, 0.2778, 0.1905, 0.0684, 0.2377, 0.1453, 0.0160], + device='cuda:0'), in_proj_covar=tensor([0.0414, 0.0368, 0.0388, 0.0350, 0.0372, 0.0352, 0.0386, 0.0406], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:48:14,933 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-03 02:48:25,483 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=162134.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:48:37,725 INFO [train.py:903] (0/4) Epoch 24, batch 5100, loss[loss=0.1698, simple_loss=0.2595, pruned_loss=0.03998, over 19398.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2856, pruned_loss=0.06251, over 3823883.19 frames. ], batch size: 48, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:48:44,637 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.860e+02 4.540e+02 5.777e+02 7.463e+02 1.637e+03, threshold=1.155e+03, percent-clipped=6.0 +2023-04-03 02:48:51,569 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-03 02:48:55,177 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-03 02:48:55,562 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=162159.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:48:58,697 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-03 02:49:18,933 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=162177.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:49:39,588 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=162193.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:49:40,373 INFO [train.py:903] (0/4) Epoch 24, batch 5150, loss[loss=0.2099, simple_loss=0.2987, pruned_loss=0.06055, over 19727.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2854, pruned_loss=0.06242, over 3810137.29 frames. ], batch size: 63, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:49:57,182 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-03 02:50:11,316 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=162218.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:50:31,572 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 02:50:43,109 INFO [train.py:903] (0/4) Epoch 24, batch 5200, loss[loss=0.2061, simple_loss=0.2977, pruned_loss=0.05722, over 19686.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2856, pruned_loss=0.06234, over 3798925.77 frames. ], batch size: 59, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:50:50,121 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=162249.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:50:51,036 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.192e+02 4.774e+02 5.644e+02 7.681e+02 1.514e+03, threshold=1.129e+03, percent-clipped=4.0 +2023-04-03 02:50:59,707 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-03 02:51:44,065 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-03 02:51:44,346 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=162292.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:51:46,282 INFO [train.py:903] (0/4) Epoch 24, batch 5250, loss[loss=0.1885, simple_loss=0.274, pruned_loss=0.05154, over 19730.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2863, pruned_loss=0.06261, over 3804167.47 frames. ], batch size: 51, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:52:14,368 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1872, 1.4268, 1.6926, 1.3410, 2.7987, 1.1168, 2.2261, 3.1688], + device='cuda:0'), covar=tensor([0.0575, 0.2622, 0.2543, 0.1841, 0.0712, 0.2315, 0.1147, 0.0288], + device='cuda:0'), in_proj_covar=tensor([0.0414, 0.0368, 0.0389, 0.0350, 0.0373, 0.0353, 0.0386, 0.0406], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:52:50,082 INFO [train.py:903] (0/4) Epoch 24, batch 5300, loss[loss=0.2209, simple_loss=0.3158, pruned_loss=0.06299, over 19103.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2862, pruned_loss=0.0625, over 3784817.43 frames. ], batch size: 69, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:52:57,094 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.926e+02 4.862e+02 5.825e+02 7.901e+02 2.284e+03, threshold=1.165e+03, percent-clipped=8.0 +2023-04-03 02:53:08,418 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-03 02:53:49,890 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0 +2023-04-03 02:53:51,566 INFO [train.py:903] (0/4) Epoch 24, batch 5350, loss[loss=0.1784, simple_loss=0.2563, pruned_loss=0.05028, over 19755.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2869, pruned_loss=0.06293, over 3781588.47 frames. ], batch size: 48, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:54:28,280 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-03 02:54:47,996 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=162439.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:54:54,720 INFO [train.py:903] (0/4) Epoch 24, batch 5400, loss[loss=0.2071, simple_loss=0.2872, pruned_loss=0.06355, over 19604.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.287, pruned_loss=0.06271, over 3789099.73 frames. ], batch size: 52, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:55:02,634 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.752e+02 4.703e+02 6.237e+02 7.666e+02 1.372e+03, threshold=1.247e+03, percent-clipped=3.0 +2023-04-03 02:55:16,617 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6150, 1.1953, 1.2759, 1.4754, 1.0907, 1.3694, 1.2698, 1.4770], + device='cuda:0'), covar=tensor([0.1161, 0.1280, 0.1640, 0.1151, 0.1392, 0.0642, 0.1550, 0.0883], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0359, 0.0312, 0.0256, 0.0305, 0.0255, 0.0315, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:55:56,870 INFO [train.py:903] (0/4) Epoch 24, batch 5450, loss[loss=0.1956, simple_loss=0.2851, pruned_loss=0.053, over 19628.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2869, pruned_loss=0.06261, over 3783044.59 frames. ], batch size: 57, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:56:40,384 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=162529.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:57:00,250 INFO [train.py:903] (0/4) Epoch 24, batch 5500, loss[loss=0.2172, simple_loss=0.3029, pruned_loss=0.06571, over 19518.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2858, pruned_loss=0.06206, over 3795118.41 frames. ], batch size: 56, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:57:05,406 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=162548.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:57:07,330 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.996e+02 5.077e+02 6.464e+02 7.861e+02 1.317e+03, threshold=1.293e+03, percent-clipped=1.0 +2023-04-03 02:57:12,064 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=162554.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:57:24,996 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-03 02:57:36,134 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=162573.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:58:01,088 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=162593.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 02:58:02,007 INFO [train.py:903] (0/4) Epoch 24, batch 5550, loss[loss=0.2535, simple_loss=0.3245, pruned_loss=0.09121, over 18127.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2858, pruned_loss=0.06209, over 3809985.93 frames. ], batch size: 83, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 02:58:08,905 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-03 02:58:25,391 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.33 vs. limit=5.0 +2023-04-03 02:58:29,652 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7494, 2.5361, 2.2781, 2.6579, 2.3330, 2.3159, 2.1478, 2.7594], + device='cuda:0'), covar=tensor([0.0979, 0.1579, 0.1494, 0.1088, 0.1491, 0.0529, 0.1482, 0.0666], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0359, 0.0314, 0.0257, 0.0305, 0.0255, 0.0316, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 02:58:59,596 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-03 02:59:02,952 INFO [train.py:903] (0/4) Epoch 24, batch 5600, loss[loss=0.2524, simple_loss=0.3228, pruned_loss=0.09104, over 19679.00 frames. ], tot_loss[loss=0.2066, simple_loss=0.2868, pruned_loss=0.0632, over 3804969.05 frames. ], batch size: 59, lr: 3.40e-03, grad_scale: 16.0 +2023-04-03 02:59:12,271 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.260e+02 5.034e+02 5.949e+02 8.933e+02 2.100e+03, threshold=1.190e+03, percent-clipped=10.0 +2023-04-03 02:59:52,043 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8992, 1.6341, 1.5292, 1.9536, 1.5345, 1.6639, 1.5940, 1.7589], + device='cuda:0'), covar=tensor([0.1091, 0.1485, 0.1597, 0.1103, 0.1478, 0.0593, 0.1406, 0.0820], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0359, 0.0313, 0.0257, 0.0305, 0.0255, 0.0315, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:00:07,503 INFO [train.py:903] (0/4) Epoch 24, batch 5650, loss[loss=0.1652, simple_loss=0.2531, pruned_loss=0.0386, over 19825.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2865, pruned_loss=0.06269, over 3803291.74 frames. ], batch size: 52, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 03:00:24,947 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=162708.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:00:55,015 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-03 03:01:09,541 INFO [train.py:903] (0/4) Epoch 24, batch 5700, loss[loss=0.1943, simple_loss=0.284, pruned_loss=0.05232, over 19616.00 frames. ], tot_loss[loss=0.2071, simple_loss=0.2876, pruned_loss=0.0633, over 3803069.58 frames. ], batch size: 57, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 03:01:17,479 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.412e+02 4.704e+02 5.740e+02 7.100e+02 1.656e+03, threshold=1.148e+03, percent-clipped=4.0 +2023-04-03 03:01:20,488 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 03:02:11,497 INFO [train.py:903] (0/4) Epoch 24, batch 5750, loss[loss=0.2887, simple_loss=0.3487, pruned_loss=0.1143, over 14015.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.286, pruned_loss=0.06262, over 3804765.58 frames. ], batch size: 136, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 03:02:13,882 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-03 03:02:22,211 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-03 03:02:28,833 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-03 03:02:31,393 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=162810.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:03:00,387 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=5.11 vs. limit=5.0 +2023-04-03 03:03:03,442 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=162835.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:03:13,922 INFO [train.py:903] (0/4) Epoch 24, batch 5800, loss[loss=0.1902, simple_loss=0.2766, pruned_loss=0.05188, over 19600.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2852, pruned_loss=0.06167, over 3819647.34 frames. ], batch size: 52, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 03:03:15,410 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=162845.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:03:22,908 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.123e+02 4.857e+02 6.549e+02 8.249e+02 1.553e+03, threshold=1.310e+03, percent-clipped=3.0 +2023-04-03 03:03:51,948 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=162873.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:04:02,697 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-04-03 03:04:16,911 INFO [train.py:903] (0/4) Epoch 24, batch 5850, loss[loss=0.1827, simple_loss=0.2748, pruned_loss=0.04525, over 19658.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2855, pruned_loss=0.06189, over 3811469.87 frames. ], batch size: 55, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 03:04:28,656 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4413, 2.1044, 1.5460, 1.4001, 1.9755, 1.2960, 1.3105, 1.8655], + device='cuda:0'), covar=tensor([0.1049, 0.0810, 0.1211, 0.0914, 0.0554, 0.1345, 0.0792, 0.0484], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0313, 0.0336, 0.0266, 0.0246, 0.0338, 0.0290, 0.0271], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:04:39,290 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1419, 1.9615, 1.8145, 2.2015, 1.9104, 1.8243, 1.7452, 2.0917], + device='cuda:0'), covar=tensor([0.1028, 0.1478, 0.1518, 0.1050, 0.1371, 0.0568, 0.1429, 0.0722], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0357, 0.0313, 0.0256, 0.0304, 0.0254, 0.0315, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:05:20,442 INFO [train.py:903] (0/4) Epoch 24, batch 5900, loss[loss=0.162, simple_loss=0.238, pruned_loss=0.04303, over 16053.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.285, pruned_loss=0.06173, over 3810729.79 frames. ], batch size: 35, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 03:05:26,353 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-03 03:05:28,683 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.343e+02 4.550e+02 5.410e+02 6.827e+02 1.573e+03, threshold=1.082e+03, percent-clipped=1.0 +2023-04-03 03:05:37,955 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.1407, 5.5746, 3.0312, 4.8854, 1.0897, 5.7197, 5.5382, 5.6885], + device='cuda:0'), covar=tensor([0.0339, 0.0783, 0.1925, 0.0707, 0.4022, 0.0493, 0.0759, 0.0981], + device='cuda:0'), in_proj_covar=tensor([0.0512, 0.0418, 0.0502, 0.0352, 0.0401, 0.0441, 0.0435, 0.0467], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:05:43,745 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=162964.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:05:44,422 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-03 03:06:15,204 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=162988.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:06:16,454 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=162989.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:06:21,816 INFO [train.py:903] (0/4) Epoch 24, batch 5950, loss[loss=0.1789, simple_loss=0.2609, pruned_loss=0.04842, over 19853.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2855, pruned_loss=0.06179, over 3814077.50 frames. ], batch size: 52, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 03:07:22,000 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6024, 1.5499, 1.5285, 2.0599, 1.5700, 1.8904, 1.8496, 1.6451], + device='cuda:0'), covar=tensor([0.0849, 0.0928, 0.0998, 0.0732, 0.0859, 0.0741, 0.0876, 0.0715], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0222, 0.0226, 0.0239, 0.0225, 0.0212, 0.0188, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 03:07:22,795 INFO [train.py:903] (0/4) Epoch 24, batch 6000, loss[loss=0.2168, simple_loss=0.3152, pruned_loss=0.0592, over 19681.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.286, pruned_loss=0.06209, over 3822399.26 frames. ], batch size: 58, lr: 3.40e-03, grad_scale: 8.0 +2023-04-03 03:07:22,796 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 03:07:35,178 INFO [train.py:937] (0/4) Epoch 24, validation: loss=0.1683, simple_loss=0.2679, pruned_loss=0.03436, over 944034.00 frames. +2023-04-03 03:07:35,179 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 03:07:43,477 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.958e+02 5.276e+02 6.488e+02 8.005e+02 1.643e+03, threshold=1.298e+03, percent-clipped=7.0 +2023-04-03 03:08:35,913 INFO [train.py:903] (0/4) Epoch 24, batch 6050, loss[loss=0.204, simple_loss=0.2923, pruned_loss=0.05782, over 19593.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2869, pruned_loss=0.06249, over 3824333.00 frames. ], batch size: 61, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:08:53,220 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=163108.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:09:37,874 INFO [train.py:903] (0/4) Epoch 24, batch 6100, loss[loss=0.1915, simple_loss=0.2638, pruned_loss=0.05959, over 19461.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2862, pruned_loss=0.06199, over 3834149.50 frames. ], batch size: 49, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:09:45,776 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.855e+02 5.048e+02 5.892e+02 7.607e+02 1.565e+03, threshold=1.178e+03, percent-clipped=5.0 +2023-04-03 03:10:14,677 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.08 vs. limit=2.0 +2023-04-03 03:10:33,896 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=163189.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:10:40,175 INFO [train.py:903] (0/4) Epoch 24, batch 6150, loss[loss=0.2408, simple_loss=0.3153, pruned_loss=0.08322, over 18365.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2867, pruned_loss=0.06226, over 3839261.05 frames. ], batch size: 83, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:11:10,597 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-03 03:11:43,666 INFO [train.py:903] (0/4) Epoch 24, batch 6200, loss[loss=0.2611, simple_loss=0.3279, pruned_loss=0.09714, over 13775.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2858, pruned_loss=0.06203, over 3821654.46 frames. ], batch size: 137, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:11:44,084 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=163244.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:11:51,416 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.768e+02 4.538e+02 5.825e+02 7.342e+02 1.276e+03, threshold=1.165e+03, percent-clipped=3.0 +2023-04-03 03:12:00,532 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=163258.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:12:14,128 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=163269.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:12:44,000 INFO [train.py:903] (0/4) Epoch 24, batch 6250, loss[loss=0.2335, simple_loss=0.3079, pruned_loss=0.07953, over 13411.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2863, pruned_loss=0.06248, over 3801753.87 frames. ], batch size: 136, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:12:56,463 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=163304.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:13:11,282 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-03 03:13:32,263 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0628, 5.4933, 3.0309, 4.6613, 1.1173, 5.5918, 5.4426, 5.6234], + device='cuda:0'), covar=tensor([0.0409, 0.0854, 0.1973, 0.0849, 0.4352, 0.0541, 0.0834, 0.0812], + device='cuda:0'), in_proj_covar=tensor([0.0511, 0.0418, 0.0501, 0.0351, 0.0401, 0.0442, 0.0436, 0.0466], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:13:45,520 INFO [train.py:903] (0/4) Epoch 24, batch 6300, loss[loss=0.2061, simple_loss=0.2931, pruned_loss=0.05953, over 19783.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2852, pruned_loss=0.06214, over 3794699.67 frames. ], batch size: 56, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:13:54,810 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.313e+02 5.008e+02 6.887e+02 8.761e+02 2.377e+03, threshold=1.377e+03, percent-clipped=3.0 +2023-04-03 03:14:09,404 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.38 vs. limit=5.0 +2023-04-03 03:14:48,465 INFO [train.py:903] (0/4) Epoch 24, batch 6350, loss[loss=0.2545, simple_loss=0.3244, pruned_loss=0.09229, over 18762.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2857, pruned_loss=0.06251, over 3803483.62 frames. ], batch size: 74, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:15:03,592 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.2431, 5.1395, 6.0294, 6.0510, 1.9876, 5.6307, 4.7902, 5.7196], + device='cuda:0'), covar=tensor([0.1692, 0.0794, 0.0586, 0.0595, 0.6462, 0.0735, 0.0609, 0.1181], + device='cuda:0'), in_proj_covar=tensor([0.0805, 0.0764, 0.0973, 0.0849, 0.0852, 0.0737, 0.0581, 0.0901], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 03:15:50,634 INFO [train.py:903] (0/4) Epoch 24, batch 6400, loss[loss=0.27, simple_loss=0.3325, pruned_loss=0.1038, over 13083.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2851, pruned_loss=0.06268, over 3790686.26 frames. ], batch size: 136, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:15:59,000 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.312e+02 4.869e+02 5.854e+02 7.378e+02 1.563e+03, threshold=1.171e+03, percent-clipped=2.0 +2023-04-03 03:16:00,310 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=163452.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:16:52,285 INFO [train.py:903] (0/4) Epoch 24, batch 6450, loss[loss=0.1994, simple_loss=0.2861, pruned_loss=0.05632, over 19761.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2854, pruned_loss=0.06253, over 3792167.86 frames. ], batch size: 56, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:16:59,802 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.41 vs. limit=5.0 +2023-04-03 03:17:34,394 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-03 03:17:37,158 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2460, 3.6355, 2.1863, 2.1267, 3.2118, 1.9716, 1.5204, 2.4610], + device='cuda:0'), covar=tensor([0.1305, 0.0652, 0.1107, 0.0917, 0.0613, 0.1228, 0.1018, 0.0659], + device='cuda:0'), in_proj_covar=tensor([0.0301, 0.0315, 0.0336, 0.0267, 0.0247, 0.0341, 0.0290, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:17:53,147 INFO [train.py:903] (0/4) Epoch 24, batch 6500, loss[loss=0.2436, simple_loss=0.3175, pruned_loss=0.08482, over 19345.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2841, pruned_loss=0.06191, over 3801110.68 frames. ], batch size: 70, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:17:56,720 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-03 03:18:01,367 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.063e+02 4.704e+02 6.024e+02 8.376e+02 1.457e+03, threshold=1.205e+03, percent-clipped=5.0 +2023-04-03 03:18:13,945 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=163560.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:18:15,832 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-03 03:18:22,395 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=163567.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:18:33,117 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3422, 2.2442, 2.0816, 1.8906, 1.7391, 1.8872, 0.6255, 1.3218], + device='cuda:0'), covar=tensor([0.0663, 0.0638, 0.0510, 0.0926, 0.1209, 0.1005, 0.1417, 0.1097], + device='cuda:0'), in_proj_covar=tensor([0.0357, 0.0357, 0.0361, 0.0384, 0.0461, 0.0391, 0.0338, 0.0344], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 03:18:44,284 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=163585.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:18:55,197 INFO [train.py:903] (0/4) Epoch 24, batch 6550, loss[loss=0.1951, simple_loss=0.2755, pruned_loss=0.05735, over 19745.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2846, pruned_loss=0.06202, over 3810160.43 frames. ], batch size: 51, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:19:05,566 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=163602.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:19:38,249 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7185, 1.7211, 1.6920, 1.4277, 1.3924, 1.4055, 0.3314, 0.7164], + device='cuda:0'), covar=tensor([0.0650, 0.0641, 0.0383, 0.0697, 0.1244, 0.0810, 0.1317, 0.1112], + device='cuda:0'), in_proj_covar=tensor([0.0357, 0.0357, 0.0361, 0.0384, 0.0462, 0.0391, 0.0339, 0.0344], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 03:19:57,491 INFO [train.py:903] (0/4) Epoch 24, batch 6600, loss[loss=0.2176, simple_loss=0.2986, pruned_loss=0.06835, over 19772.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2853, pruned_loss=0.06228, over 3815215.91 frames. ], batch size: 54, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:20:05,514 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.150e+02 5.011e+02 6.018e+02 7.633e+02 1.393e+03, threshold=1.204e+03, percent-clipped=8.0 +2023-04-03 03:20:57,660 INFO [train.py:903] (0/4) Epoch 24, batch 6650, loss[loss=0.1685, simple_loss=0.2507, pruned_loss=0.04318, over 19486.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2848, pruned_loss=0.06224, over 3825566.36 frames. ], batch size: 49, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:21:22,433 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3716, 1.4855, 1.7784, 1.7227, 2.9840, 4.5344, 4.3390, 5.1407], + device='cuda:0'), covar=tensor([0.1761, 0.4756, 0.4439, 0.2350, 0.0671, 0.0235, 0.0219, 0.0213], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0326, 0.0356, 0.0266, 0.0245, 0.0190, 0.0216, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 03:21:25,781 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=163717.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:21:58,179 INFO [train.py:903] (0/4) Epoch 24, batch 6700, loss[loss=0.1868, simple_loss=0.2804, pruned_loss=0.04653, over 19676.00 frames. ], tot_loss[loss=0.206, simple_loss=0.2861, pruned_loss=0.0629, over 3817060.10 frames. ], batch size: 59, lr: 3.39e-03, grad_scale: 4.0 +2023-04-03 03:22:08,769 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.807e+02 4.744e+02 6.010e+02 8.153e+02 1.593e+03, threshold=1.202e+03, percent-clipped=6.0 +2023-04-03 03:22:45,666 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4048, 2.1263, 1.6338, 1.4615, 1.9577, 1.3542, 1.3355, 1.7771], + device='cuda:0'), covar=tensor([0.0987, 0.0838, 0.1081, 0.0843, 0.0575, 0.1257, 0.0704, 0.0507], + device='cuda:0'), in_proj_covar=tensor([0.0305, 0.0318, 0.0340, 0.0270, 0.0251, 0.0344, 0.0293, 0.0276], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:22:57,589 INFO [train.py:903] (0/4) Epoch 24, batch 6750, loss[loss=0.2449, simple_loss=0.3211, pruned_loss=0.08436, over 13650.00 frames. ], tot_loss[loss=0.2068, simple_loss=0.287, pruned_loss=0.06334, over 3809634.63 frames. ], batch size: 137, lr: 3.39e-03, grad_scale: 4.0 +2023-04-03 03:23:30,574 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=163823.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:23:38,381 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1071, 1.2597, 1.7549, 1.2822, 2.7650, 3.6609, 3.3635, 3.8941], + device='cuda:0'), covar=tensor([0.1769, 0.3947, 0.3385, 0.2508, 0.0600, 0.0207, 0.0213, 0.0271], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0326, 0.0356, 0.0266, 0.0246, 0.0190, 0.0216, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 03:23:53,912 INFO [train.py:903] (0/4) Epoch 24, batch 6800, loss[loss=0.1805, simple_loss=0.2527, pruned_loss=0.05417, over 19358.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.2863, pruned_loss=0.06301, over 3802862.34 frames. ], batch size: 47, lr: 3.39e-03, grad_scale: 8.0 +2023-04-03 03:23:58,806 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=163848.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:24:03,010 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.305e+02 4.890e+02 5.869e+02 7.347e+02 2.478e+03, threshold=1.174e+03, percent-clipped=2.0 +2023-04-03 03:24:24,224 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-24.pt +2023-04-03 03:24:39,503 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-03 03:24:40,522 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-03 03:24:42,797 INFO [train.py:903] (0/4) Epoch 25, batch 0, loss[loss=0.2091, simple_loss=0.2861, pruned_loss=0.06607, over 19592.00 frames. ], tot_loss[loss=0.2091, simple_loss=0.2861, pruned_loss=0.06607, over 19592.00 frames. ], batch size: 50, lr: 3.32e-03, grad_scale: 8.0 +2023-04-03 03:24:42,798 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 03:24:54,386 INFO [train.py:937] (0/4) Epoch 25, validation: loss=0.1672, simple_loss=0.2675, pruned_loss=0.03346, over 944034.00 frames. +2023-04-03 03:24:54,387 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 03:25:06,941 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-03 03:25:40,910 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1253, 1.1915, 1.4440, 1.3955, 2.7144, 1.1302, 2.1869, 3.0762], + device='cuda:0'), covar=tensor([0.0555, 0.2967, 0.2919, 0.1845, 0.0755, 0.2332, 0.1225, 0.0326], + device='cuda:0'), in_proj_covar=tensor([0.0413, 0.0368, 0.0391, 0.0347, 0.0374, 0.0352, 0.0386, 0.0405], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:25:57,056 INFO [train.py:903] (0/4) Epoch 25, batch 50, loss[loss=0.2005, simple_loss=0.2768, pruned_loss=0.0621, over 19738.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2838, pruned_loss=0.05948, over 874197.55 frames. ], batch size: 47, lr: 3.32e-03, grad_scale: 8.0 +2023-04-03 03:26:35,506 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.099e+02 4.575e+02 5.589e+02 7.102e+02 2.434e+03, threshold=1.118e+03, percent-clipped=5.0 +2023-04-03 03:26:36,710 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-03 03:27:00,593 INFO [train.py:903] (0/4) Epoch 25, batch 100, loss[loss=0.1933, simple_loss=0.2805, pruned_loss=0.05301, over 19682.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2837, pruned_loss=0.06016, over 1537986.87 frames. ], batch size: 55, lr: 3.32e-03, grad_scale: 8.0 +2023-04-03 03:27:03,083 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=163973.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:27:15,423 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-03 03:27:33,554 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=163998.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:27:36,405 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-164000.pt +2023-04-03 03:28:05,187 INFO [train.py:903] (0/4) Epoch 25, batch 150, loss[loss=0.2203, simple_loss=0.2982, pruned_loss=0.0712, over 19606.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2834, pruned_loss=0.05987, over 2060321.41 frames. ], batch size: 61, lr: 3.32e-03, grad_scale: 8.0 +2023-04-03 03:28:42,973 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.518e+02 5.520e+02 6.265e+02 7.455e+02 1.438e+03, threshold=1.253e+03, percent-clipped=2.0 +2023-04-03 03:29:06,837 INFO [train.py:903] (0/4) Epoch 25, batch 200, loss[loss=0.248, simple_loss=0.3127, pruned_loss=0.09165, over 13811.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2853, pruned_loss=0.06127, over 2450766.21 frames. ], batch size: 136, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:29:10,500 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-03 03:29:53,425 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164108.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:30:10,589 INFO [train.py:903] (0/4) Epoch 25, batch 250, loss[loss=0.2188, simple_loss=0.3049, pruned_loss=0.06633, over 19315.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2853, pruned_loss=0.06138, over 2758831.46 frames. ], batch size: 66, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:30:41,660 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.28 vs. limit=2.0 +2023-04-03 03:30:48,736 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.799e+02 4.875e+02 6.095e+02 7.386e+02 2.001e+03, threshold=1.219e+03, percent-clipped=3.0 +2023-04-03 03:31:07,751 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-03 03:31:13,886 INFO [train.py:903] (0/4) Epoch 25, batch 300, loss[loss=0.1757, simple_loss=0.2528, pruned_loss=0.04926, over 19388.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2856, pruned_loss=0.06175, over 2992797.58 frames. ], batch size: 48, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:31:16,545 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164173.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:32:18,078 INFO [train.py:903] (0/4) Epoch 25, batch 350, loss[loss=0.1785, simple_loss=0.2552, pruned_loss=0.05091, over 19130.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2851, pruned_loss=0.06135, over 3177646.30 frames. ], batch size: 42, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:32:25,190 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 03:32:55,000 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.912e+02 5.440e+02 6.552e+02 9.332e+02 1.789e+03, threshold=1.310e+03, percent-clipped=12.0 +2023-04-03 03:33:20,720 INFO [train.py:903] (0/4) Epoch 25, batch 400, loss[loss=0.1752, simple_loss=0.2495, pruned_loss=0.05041, over 19757.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2855, pruned_loss=0.06181, over 3325712.74 frames. ], batch size: 45, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:33:53,581 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3760, 2.2856, 2.2355, 2.5312, 2.2517, 2.0389, 2.1796, 2.3704], + device='cuda:0'), covar=tensor([0.0785, 0.1237, 0.1093, 0.0755, 0.1066, 0.0489, 0.1114, 0.0553], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0359, 0.0314, 0.0256, 0.0306, 0.0254, 0.0316, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:34:24,838 INFO [train.py:903] (0/4) Epoch 25, batch 450, loss[loss=0.1666, simple_loss=0.2428, pruned_loss=0.04515, over 19050.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2862, pruned_loss=0.06226, over 3441559.33 frames. ], batch size: 42, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:34:59,262 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-03 03:35:00,479 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-03 03:35:02,820 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.977e+02 4.778e+02 5.577e+02 7.081e+02 1.680e+03, threshold=1.115e+03, percent-clipped=3.0 +2023-04-03 03:35:27,607 INFO [train.py:903] (0/4) Epoch 25, batch 500, loss[loss=0.2095, simple_loss=0.3002, pruned_loss=0.0594, over 18803.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2861, pruned_loss=0.0618, over 3534067.89 frames. ], batch size: 74, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:35:35,587 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164378.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:36:31,238 INFO [train.py:903] (0/4) Epoch 25, batch 550, loss[loss=0.1623, simple_loss=0.2433, pruned_loss=0.0407, over 19773.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2858, pruned_loss=0.06171, over 3602080.81 frames. ], batch size: 45, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:36:49,752 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2584, 2.0097, 1.5605, 1.3232, 1.8273, 1.2131, 1.2496, 1.7531], + device='cuda:0'), covar=tensor([0.1014, 0.0808, 0.1081, 0.0854, 0.0585, 0.1366, 0.0717, 0.0488], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0318, 0.0339, 0.0268, 0.0250, 0.0342, 0.0292, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:37:09,261 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.449e+02 5.163e+02 6.218e+02 7.641e+02 1.675e+03, threshold=1.244e+03, percent-clipped=5.0 +2023-04-03 03:37:09,447 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=164452.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:37:34,084 INFO [train.py:903] (0/4) Epoch 25, batch 600, loss[loss=0.1622, simple_loss=0.2399, pruned_loss=0.0423, over 19730.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2864, pruned_loss=0.06202, over 3644240.93 frames. ], batch size: 46, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:38:16,672 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-03 03:38:30,633 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=164517.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:38:36,206 INFO [train.py:903] (0/4) Epoch 25, batch 650, loss[loss=0.1882, simple_loss=0.2697, pruned_loss=0.05333, over 19772.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2857, pruned_loss=0.06172, over 3700356.39 frames. ], batch size: 54, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:39:08,495 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6718, 1.5698, 1.5578, 2.2580, 1.6173, 1.8982, 1.9583, 1.7150], + device='cuda:0'), covar=tensor([0.0838, 0.0970, 0.1026, 0.0695, 0.0856, 0.0792, 0.0911, 0.0717], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0221, 0.0226, 0.0237, 0.0225, 0.0211, 0.0188, 0.0203], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0004], + device='cuda:0') +2023-04-03 03:39:15,258 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.327e+02 4.538e+02 5.913e+02 7.820e+02 1.600e+03, threshold=1.183e+03, percent-clipped=2.0 +2023-04-03 03:39:19,274 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0457, 1.7161, 2.0082, 1.6603, 4.5773, 1.3412, 2.8179, 5.0383], + device='cuda:0'), covar=tensor([0.0449, 0.2671, 0.2606, 0.2032, 0.0726, 0.2506, 0.1285, 0.0159], + device='cuda:0'), in_proj_covar=tensor([0.0418, 0.0369, 0.0393, 0.0350, 0.0376, 0.0354, 0.0388, 0.0408], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:39:33,671 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=164567.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:39:40,283 INFO [train.py:903] (0/4) Epoch 25, batch 700, loss[loss=0.199, simple_loss=0.2723, pruned_loss=0.06287, over 19487.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2858, pruned_loss=0.06243, over 3729932.62 frames. ], batch size: 49, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:39:58,596 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164586.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:40:11,617 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164596.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:40:44,576 INFO [train.py:903] (0/4) Epoch 25, batch 750, loss[loss=0.2002, simple_loss=0.2858, pruned_loss=0.05726, over 18742.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2862, pruned_loss=0.06255, over 3748590.94 frames. ], batch size: 74, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:40:48,672 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4237, 1.5151, 1.8923, 1.7459, 2.7255, 2.3196, 2.9292, 1.2683], + device='cuda:0'), covar=tensor([0.2491, 0.4258, 0.2692, 0.1870, 0.1510, 0.2090, 0.1402, 0.4432], + device='cuda:0'), in_proj_covar=tensor([0.0545, 0.0660, 0.0734, 0.0497, 0.0633, 0.0542, 0.0669, 0.0561], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 03:40:57,977 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=164632.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:41:21,491 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.304e+02 5.157e+02 6.753e+02 8.219e+02 1.587e+03, threshold=1.351e+03, percent-clipped=10.0 +2023-04-03 03:41:28,840 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8648, 1.3043, 1.0480, 0.9535, 1.1504, 0.9692, 0.8793, 1.1970], + device='cuda:0'), covar=tensor([0.0618, 0.0850, 0.1091, 0.0771, 0.0580, 0.1303, 0.0629, 0.0514], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0319, 0.0340, 0.0269, 0.0251, 0.0345, 0.0293, 0.0276], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:41:48,143 INFO [train.py:903] (0/4) Epoch 25, batch 800, loss[loss=0.2109, simple_loss=0.2947, pruned_loss=0.06362, over 18184.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2863, pruned_loss=0.06274, over 3773209.44 frames. ], batch size: 83, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:41:52,830 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164676.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:42:03,217 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-03 03:42:50,193 INFO [train.py:903] (0/4) Epoch 25, batch 850, loss[loss=0.2078, simple_loss=0.2813, pruned_loss=0.06717, over 19571.00 frames. ], tot_loss[loss=0.2065, simple_loss=0.2869, pruned_loss=0.06298, over 3798596.80 frames. ], batch size: 52, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:42:50,358 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=164722.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:43:29,203 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.532e+02 4.928e+02 6.003e+02 7.929e+02 1.446e+03, threshold=1.201e+03, percent-clipped=1.0 +2023-04-03 03:43:43,425 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=164764.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:43:44,372 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-03 03:43:52,632 INFO [train.py:903] (0/4) Epoch 25, batch 900, loss[loss=0.2153, simple_loss=0.298, pruned_loss=0.06629, over 18220.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2859, pruned_loss=0.06261, over 3807219.43 frames. ], batch size: 83, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:44:56,069 INFO [train.py:903] (0/4) Epoch 25, batch 950, loss[loss=0.2202, simple_loss=0.3037, pruned_loss=0.06836, over 17371.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.2863, pruned_loss=0.06255, over 3809807.24 frames. ], batch size: 101, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:44:57,277 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-03 03:44:57,675 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=164823.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:45:15,717 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=164837.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:45:28,776 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=164848.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:45:32,815 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.080e+02 4.681e+02 5.489e+02 6.922e+02 1.500e+03, threshold=1.098e+03, percent-clipped=4.0 +2023-04-03 03:45:43,195 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-04-03 03:46:00,730 INFO [train.py:903] (0/4) Epoch 25, batch 1000, loss[loss=0.183, simple_loss=0.2585, pruned_loss=0.05375, over 19731.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2854, pruned_loss=0.06188, over 3828146.59 frames. ], batch size: 45, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:46:19,869 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=164888.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:46:53,157 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=164913.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:46:55,239 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-03 03:47:03,437 INFO [train.py:903] (0/4) Epoch 25, batch 1050, loss[loss=0.2072, simple_loss=0.2887, pruned_loss=0.06283, over 19541.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2848, pruned_loss=0.06158, over 3834585.69 frames. ], batch size: 54, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:47:13,153 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=164930.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:47:27,131 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=164940.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:47:36,157 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-03 03:47:41,876 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.266e+02 4.678e+02 5.688e+02 6.996e+02 1.189e+03, threshold=1.138e+03, percent-clipped=3.0 +2023-04-03 03:47:45,538 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5983, 1.2352, 1.4210, 1.4489, 3.1890, 1.1008, 2.4077, 3.6492], + device='cuda:0'), covar=tensor([0.0493, 0.2871, 0.3071, 0.1948, 0.0739, 0.2615, 0.1335, 0.0228], + device='cuda:0'), in_proj_covar=tensor([0.0417, 0.0369, 0.0394, 0.0349, 0.0377, 0.0353, 0.0388, 0.0409], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:48:06,293 INFO [train.py:903] (0/4) Epoch 25, batch 1100, loss[loss=0.1714, simple_loss=0.2486, pruned_loss=0.04707, over 19774.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2849, pruned_loss=0.06193, over 3822082.55 frames. ], batch size: 46, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:49:07,178 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=165020.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:49:09,312 INFO [train.py:903] (0/4) Epoch 25, batch 1150, loss[loss=0.2076, simple_loss=0.2895, pruned_loss=0.06279, over 19689.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2854, pruned_loss=0.06226, over 3828964.59 frames. ], batch size: 59, lr: 3.31e-03, grad_scale: 8.0 +2023-04-03 03:49:40,387 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165045.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:49:48,275 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.006e+02 5.101e+02 6.369e+02 8.453e+02 1.568e+03, threshold=1.274e+03, percent-clipped=10.0 +2023-04-03 03:49:52,159 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165055.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:50:14,743 INFO [train.py:903] (0/4) Epoch 25, batch 1200, loss[loss=0.1887, simple_loss=0.2781, pruned_loss=0.04968, over 19534.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2858, pruned_loss=0.06217, over 3828753.46 frames. ], batch size: 54, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 03:50:41,038 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165093.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:50:46,453 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-03 03:51:01,332 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=165108.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:51:14,650 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165118.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:51:18,446 INFO [train.py:903] (0/4) Epoch 25, batch 1250, loss[loss=0.2737, simple_loss=0.3474, pruned_loss=0.09999, over 19693.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2859, pruned_loss=0.06244, over 3821646.60 frames. ], batch size: 59, lr: 3.30e-03, grad_scale: 4.0 +2023-04-03 03:51:34,100 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165135.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:51:41,182 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-04-03 03:51:41,824 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7959, 4.2865, 4.5045, 4.5170, 1.7972, 4.2713, 3.7131, 4.2247], + device='cuda:0'), covar=tensor([0.1610, 0.0859, 0.0583, 0.0656, 0.5942, 0.0959, 0.0675, 0.1092], + device='cuda:0'), in_proj_covar=tensor([0.0802, 0.0766, 0.0974, 0.0850, 0.0856, 0.0736, 0.0579, 0.0903], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 03:51:41,845 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=165141.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:51:49,280 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2565, 1.5131, 2.1003, 1.6601, 3.0667, 4.8701, 4.7290, 5.2567], + device='cuda:0'), covar=tensor([0.1647, 0.3580, 0.2993, 0.2250, 0.0614, 0.0188, 0.0149, 0.0178], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0329, 0.0359, 0.0267, 0.0248, 0.0191, 0.0218, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 03:51:57,604 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.318e+02 4.844e+02 6.322e+02 8.270e+02 1.695e+03, threshold=1.264e+03, percent-clipped=2.0 +2023-04-03 03:52:21,099 INFO [train.py:903] (0/4) Epoch 25, batch 1300, loss[loss=0.2779, simple_loss=0.3368, pruned_loss=0.1095, over 13324.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2855, pruned_loss=0.0627, over 3814423.01 frames. ], batch size: 137, lr: 3.30e-03, grad_scale: 4.0 +2023-04-03 03:53:23,837 INFO [train.py:903] (0/4) Epoch 25, batch 1350, loss[loss=0.2055, simple_loss=0.2991, pruned_loss=0.05593, over 19677.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2858, pruned_loss=0.06263, over 3816171.67 frames. ], batch size: 59, lr: 3.30e-03, grad_scale: 4.0 +2023-04-03 03:53:26,668 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165223.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:54:04,603 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.257e+02 4.919e+02 6.162e+02 7.502e+02 1.875e+03, threshold=1.232e+03, percent-clipped=3.0 +2023-04-03 03:54:29,442 INFO [train.py:903] (0/4) Epoch 25, batch 1400, loss[loss=0.1867, simple_loss=0.2648, pruned_loss=0.05432, over 19386.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2847, pruned_loss=0.06204, over 3813012.03 frames. ], batch size: 48, lr: 3.30e-03, grad_scale: 4.0 +2023-04-03 03:54:51,001 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6687, 1.2202, 1.5031, 1.4330, 3.2510, 1.2730, 2.4185, 3.6705], + device='cuda:0'), covar=tensor([0.0524, 0.2933, 0.2897, 0.1973, 0.0696, 0.2411, 0.1308, 0.0248], + device='cuda:0'), in_proj_covar=tensor([0.0413, 0.0367, 0.0391, 0.0348, 0.0373, 0.0350, 0.0386, 0.0407], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:54:53,477 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0897, 1.9942, 1.9271, 1.7859, 1.5938, 1.8007, 0.6153, 1.1237], + device='cuda:0'), covar=tensor([0.0595, 0.0645, 0.0437, 0.0702, 0.1092, 0.0808, 0.1382, 0.1064], + device='cuda:0'), in_proj_covar=tensor([0.0363, 0.0359, 0.0363, 0.0388, 0.0465, 0.0394, 0.0341, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 03:55:05,163 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165301.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:55:18,713 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165311.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:55:19,119 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-04-03 03:55:32,140 INFO [train.py:903] (0/4) Epoch 25, batch 1450, loss[loss=0.287, simple_loss=0.3551, pruned_loss=0.1095, over 19667.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2849, pruned_loss=0.06203, over 3833431.58 frames. ], batch size: 59, lr: 3.30e-03, grad_scale: 4.0 +2023-04-03 03:55:33,177 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-03 03:55:36,984 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165326.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:55:48,866 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165336.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:56:10,666 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.320e+02 4.584e+02 6.100e+02 8.063e+02 1.924e+03, threshold=1.220e+03, percent-clipped=4.0 +2023-04-03 03:56:33,837 INFO [train.py:903] (0/4) Epoch 25, batch 1500, loss[loss=0.2214, simple_loss=0.3043, pruned_loss=0.06926, over 19606.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2852, pruned_loss=0.06263, over 3837382.11 frames. ], batch size: 61, lr: 3.30e-03, grad_scale: 4.0 +2023-04-03 03:56:58,480 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165391.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:57:30,072 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165416.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:57:36,920 INFO [train.py:903] (0/4) Epoch 25, batch 1550, loss[loss=0.1901, simple_loss=0.277, pruned_loss=0.0516, over 19442.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2853, pruned_loss=0.06238, over 3832316.97 frames. ], batch size: 70, lr: 3.30e-03, grad_scale: 4.0 +2023-04-03 03:58:06,753 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5630, 2.1532, 1.6686, 1.5570, 2.0480, 1.4099, 1.4546, 1.9151], + device='cuda:0'), covar=tensor([0.1006, 0.0815, 0.1088, 0.0803, 0.0597, 0.1267, 0.0739, 0.0539], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0318, 0.0340, 0.0268, 0.0249, 0.0343, 0.0293, 0.0276], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 03:58:17,718 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.229e+02 4.957e+02 6.075e+02 6.924e+02 1.069e+03, threshold=1.215e+03, percent-clipped=0.0 +2023-04-03 03:58:37,746 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-03 03:58:42,876 INFO [train.py:903] (0/4) Epoch 25, batch 1600, loss[loss=0.1825, simple_loss=0.2573, pruned_loss=0.05381, over 19059.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2846, pruned_loss=0.06176, over 3826042.38 frames. ], batch size: 42, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 03:58:51,581 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165479.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:58:59,595 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=165485.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:59:06,518 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-03 03:59:21,416 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165504.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 03:59:44,271 INFO [train.py:903] (0/4) Epoch 25, batch 1650, loss[loss=0.1879, simple_loss=0.2565, pruned_loss=0.05962, over 19771.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2854, pruned_loss=0.06214, over 3816672.11 frames. ], batch size: 47, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:00:23,731 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.160e+02 4.584e+02 6.289e+02 7.621e+02 1.672e+03, threshold=1.258e+03, percent-clipped=5.0 +2023-04-03 04:00:44,037 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=165569.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:00:47,362 INFO [train.py:903] (0/4) Epoch 25, batch 1700, loss[loss=0.1865, simple_loss=0.2774, pruned_loss=0.0478, over 19531.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2858, pruned_loss=0.062, over 3823046.82 frames. ], batch size: 54, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:01:22,679 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=165600.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:01:29,488 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-03 04:01:49,445 INFO [train.py:903] (0/4) Epoch 25, batch 1750, loss[loss=0.1746, simple_loss=0.253, pruned_loss=0.04817, over 19839.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2849, pruned_loss=0.06144, over 3831770.57 frames. ], batch size: 52, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:01:53,781 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.60 vs. limit=5.0 +2023-04-03 04:02:29,221 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.400e+02 4.689e+02 5.610e+02 7.383e+02 2.270e+03, threshold=1.122e+03, percent-clipped=4.0 +2023-04-03 04:02:53,399 INFO [train.py:903] (0/4) Epoch 25, batch 1800, loss[loss=0.1871, simple_loss=0.2752, pruned_loss=0.04953, over 19616.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2838, pruned_loss=0.06119, over 3826833.42 frames. ], batch size: 50, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:03:51,790 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-03 04:03:56,454 INFO [train.py:903] (0/4) Epoch 25, batch 1850, loss[loss=0.181, simple_loss=0.258, pruned_loss=0.052, over 19762.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2842, pruned_loss=0.06152, over 3832791.84 frames. ], batch size: 47, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:04:28,947 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-03 04:04:37,050 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.780e+02 4.826e+02 5.776e+02 6.973e+02 2.376e+03, threshold=1.155e+03, percent-clipped=4.0 +2023-04-03 04:04:56,249 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=165768.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:05:00,660 INFO [train.py:903] (0/4) Epoch 25, batch 1900, loss[loss=0.2002, simple_loss=0.2857, pruned_loss=0.05735, over 19535.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2853, pruned_loss=0.06172, over 3843818.71 frames. ], batch size: 54, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:05:16,888 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-03 04:05:21,705 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-03 04:05:35,620 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6330, 2.0326, 1.6126, 1.5924, 1.9613, 1.4656, 1.5157, 1.8677], + device='cuda:0'), covar=tensor([0.0899, 0.0741, 0.0846, 0.0740, 0.0498, 0.1043, 0.0640, 0.0473], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0318, 0.0339, 0.0267, 0.0249, 0.0342, 0.0293, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 04:05:48,080 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-03 04:05:48,787 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 04:06:02,683 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=165821.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:06:03,530 INFO [train.py:903] (0/4) Epoch 25, batch 1950, loss[loss=0.1931, simple_loss=0.2805, pruned_loss=0.05285, over 19473.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2861, pruned_loss=0.06175, over 3832782.39 frames. ], batch size: 49, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:06:44,206 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.055e+02 4.671e+02 6.009e+02 7.545e+02 1.239e+03, threshold=1.202e+03, percent-clipped=2.0 +2023-04-03 04:06:48,275 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=165856.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:06:52,229 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-03 04:07:08,064 INFO [train.py:903] (0/4) Epoch 25, batch 2000, loss[loss=0.1947, simple_loss=0.2814, pruned_loss=0.05395, over 19669.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2848, pruned_loss=0.06122, over 3827769.66 frames. ], batch size: 55, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:07:20,308 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=165881.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:08:00,825 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=165913.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:08:07,766 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-03 04:08:12,462 INFO [train.py:903] (0/4) Epoch 25, batch 2050, loss[loss=0.1849, simple_loss=0.2733, pruned_loss=0.04826, over 19484.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2832, pruned_loss=0.06043, over 3839214.32 frames. ], batch size: 49, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:08:27,905 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-03 04:08:27,942 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-03 04:08:48,982 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-03 04:08:51,223 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.115e+02 5.221e+02 6.289e+02 7.653e+02 1.251e+03, threshold=1.258e+03, percent-clipped=2.0 +2023-04-03 04:09:15,826 INFO [train.py:903] (0/4) Epoch 25, batch 2100, loss[loss=0.2357, simple_loss=0.3036, pruned_loss=0.08394, over 13287.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2839, pruned_loss=0.06088, over 3827890.95 frames. ], batch size: 136, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:09:44,597 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-03 04:09:51,704 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-166000.pt +2023-04-03 04:10:08,522 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-03 04:10:19,025 INFO [train.py:903] (0/4) Epoch 25, batch 2150, loss[loss=0.3077, simple_loss=0.3556, pruned_loss=0.1299, over 13578.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2854, pruned_loss=0.06192, over 3822926.10 frames. ], batch size: 136, lr: 3.30e-03, grad_scale: 8.0 +2023-04-03 04:10:27,567 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166028.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:10:58,800 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.324e+02 5.055e+02 6.614e+02 9.413e+02 1.694e+03, threshold=1.323e+03, percent-clipped=9.0 +2023-04-03 04:11:10,561 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8032, 1.9168, 2.1376, 2.3319, 1.8491, 2.2571, 2.1839, 1.9494], + device='cuda:0'), covar=tensor([0.3990, 0.3675, 0.1917, 0.2335, 0.3788, 0.2035, 0.4727, 0.3312], + device='cuda:0'), in_proj_covar=tensor([0.0917, 0.0991, 0.0729, 0.0937, 0.0895, 0.0829, 0.0854, 0.0793], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 04:11:21,696 INFO [train.py:903] (0/4) Epoch 25, batch 2200, loss[loss=0.2115, simple_loss=0.3015, pruned_loss=0.06071, over 19584.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2854, pruned_loss=0.06208, over 3817293.79 frames. ], batch size: 61, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:11:22,008 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166072.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 04:11:33,673 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8362, 3.3285, 3.3358, 3.3412, 1.3006, 3.2221, 2.7958, 3.1386], + device='cuda:0'), covar=tensor([0.1770, 0.1006, 0.0808, 0.0963, 0.5655, 0.1134, 0.0841, 0.1256], + device='cuda:0'), in_proj_covar=tensor([0.0796, 0.0758, 0.0968, 0.0843, 0.0840, 0.0730, 0.0573, 0.0892], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 04:11:38,417 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5854, 4.7296, 5.2713, 5.3110, 2.3527, 4.9790, 4.3284, 4.9949], + device='cuda:0'), covar=tensor([0.1662, 0.1504, 0.0580, 0.0630, 0.5508, 0.0863, 0.0612, 0.1056], + device='cuda:0'), in_proj_covar=tensor([0.0796, 0.0758, 0.0968, 0.0843, 0.0841, 0.0730, 0.0573, 0.0892], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 04:12:13,547 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=166112.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:12:26,901 INFO [train.py:903] (0/4) Epoch 25, batch 2250, loss[loss=0.2234, simple_loss=0.3088, pruned_loss=0.06904, over 18198.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2847, pruned_loss=0.06159, over 3815916.23 frames. ], batch size: 83, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:13:04,906 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.096e+02 4.777e+02 5.716e+02 7.657e+02 1.924e+03, threshold=1.143e+03, percent-clipped=2.0 +2023-04-03 04:13:21,657 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=166165.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:13:21,809 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166165.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:13:31,175 INFO [train.py:903] (0/4) Epoch 25, batch 2300, loss[loss=0.2677, simple_loss=0.3312, pruned_loss=0.1021, over 13156.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.286, pruned_loss=0.0623, over 3808727.49 frames. ], batch size: 137, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:13:42,681 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-03 04:14:33,997 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5617, 2.6006, 2.2087, 2.7133, 2.6130, 2.2488, 2.0614, 2.6641], + device='cuda:0'), covar=tensor([0.0987, 0.1468, 0.1444, 0.0991, 0.1246, 0.0513, 0.1427, 0.0614], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0353, 0.0311, 0.0253, 0.0300, 0.0251, 0.0312, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 04:14:34,777 INFO [train.py:903] (0/4) Epoch 25, batch 2350, loss[loss=0.1752, simple_loss=0.2656, pruned_loss=0.04238, over 19668.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2856, pruned_loss=0.0622, over 3805042.49 frames. ], batch size: 53, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:14:41,258 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166227.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:15:09,852 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5522, 1.4472, 1.4301, 1.8548, 1.3747, 1.6369, 1.6636, 1.5905], + device='cuda:0'), covar=tensor([0.0785, 0.0867, 0.0923, 0.0582, 0.0900, 0.0772, 0.0885, 0.0683], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0223, 0.0226, 0.0239, 0.0226, 0.0213, 0.0189, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 04:15:14,901 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.796e+02 5.104e+02 6.317e+02 8.219e+02 1.547e+03, threshold=1.263e+03, percent-clipped=3.0 +2023-04-03 04:15:17,155 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-03 04:15:33,842 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0677, 1.9918, 1.7704, 2.1851, 2.0036, 1.8224, 1.7222, 1.9812], + device='cuda:0'), covar=tensor([0.1027, 0.1409, 0.1398, 0.0923, 0.1242, 0.0542, 0.1440, 0.0683], + device='cuda:0'), in_proj_covar=tensor([0.0273, 0.0354, 0.0311, 0.0253, 0.0300, 0.0252, 0.0312, 0.0258], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 04:15:34,630 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-03 04:15:38,221 INFO [train.py:903] (0/4) Epoch 25, batch 2400, loss[loss=0.2092, simple_loss=0.2908, pruned_loss=0.0638, over 19740.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.285, pruned_loss=0.06176, over 3826782.85 frames. ], batch size: 63, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:15:48,887 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166280.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:15:55,262 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166284.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:16:08,740 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2519, 1.2048, 1.1924, 1.3667, 1.0223, 1.3981, 1.3476, 1.3043], + device='cuda:0'), covar=tensor([0.0909, 0.1002, 0.1082, 0.0696, 0.0863, 0.0808, 0.0797, 0.0784], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0222, 0.0226, 0.0238, 0.0226, 0.0212, 0.0189, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 04:16:25,813 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166309.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:16:41,763 INFO [train.py:903] (0/4) Epoch 25, batch 2450, loss[loss=0.1746, simple_loss=0.2449, pruned_loss=0.05211, over 19287.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2849, pruned_loss=0.06184, over 3829941.20 frames. ], batch size: 44, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:16:46,705 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166325.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:17:20,232 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.023e+02 4.903e+02 5.916e+02 7.490e+02 1.353e+03, threshold=1.183e+03, percent-clipped=1.0 +2023-04-03 04:17:44,924 INFO [train.py:903] (0/4) Epoch 25, batch 2500, loss[loss=0.1792, simple_loss=0.2519, pruned_loss=0.05331, over 19352.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2852, pruned_loss=0.06205, over 3828268.73 frames. ], batch size: 47, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:18:41,258 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=166416.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 04:18:48,040 INFO [train.py:903] (0/4) Epoch 25, batch 2550, loss[loss=0.2455, simple_loss=0.3114, pruned_loss=0.08983, over 12844.00 frames. ], tot_loss[loss=0.2067, simple_loss=0.287, pruned_loss=0.06321, over 3813350.53 frames. ], batch size: 135, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:18:49,968 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.56 vs. limit=5.0 +2023-04-03 04:19:28,549 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.584e+02 5.112e+02 6.516e+02 8.109e+02 2.174e+03, threshold=1.303e+03, percent-clipped=8.0 +2023-04-03 04:19:46,463 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-03 04:19:53,274 INFO [train.py:903] (0/4) Epoch 25, batch 2600, loss[loss=0.1986, simple_loss=0.286, pruned_loss=0.05564, over 19684.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2861, pruned_loss=0.06246, over 3834700.47 frames. ], batch size: 59, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:20:08,593 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166483.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:20:29,561 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166499.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:20:40,153 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166508.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:20:41,192 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=166509.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:20:58,304 INFO [train.py:903] (0/4) Epoch 25, batch 2650, loss[loss=0.1744, simple_loss=0.2659, pruned_loss=0.04148, over 19522.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2856, pruned_loss=0.06199, over 3830185.21 frames. ], batch size: 56, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:21:05,449 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166527.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:21:10,230 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166531.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 04:21:17,603 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166536.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:21:19,732 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-03 04:21:32,613 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166549.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:21:38,180 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.868e+02 4.612e+02 5.770e+02 6.756e+02 1.610e+03, threshold=1.154e+03, percent-clipped=1.0 +2023-04-03 04:21:48,104 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166561.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:21:57,434 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5832, 1.6707, 1.9134, 1.8615, 2.8104, 2.2818, 2.9451, 1.5950], + device='cuda:0'), covar=tensor([0.2484, 0.4164, 0.2706, 0.1882, 0.1500, 0.2318, 0.1528, 0.4308], + device='cuda:0'), in_proj_covar=tensor([0.0544, 0.0657, 0.0731, 0.0495, 0.0625, 0.0541, 0.0666, 0.0562], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 04:22:03,046 INFO [train.py:903] (0/4) Epoch 25, batch 2700, loss[loss=0.1877, simple_loss=0.2757, pruned_loss=0.04986, over 19610.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2851, pruned_loss=0.06146, over 3830921.31 frames. ], batch size: 50, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:22:10,233 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7769, 1.6304, 1.5604, 1.7861, 1.5709, 1.5195, 1.5005, 1.6969], + device='cuda:0'), covar=tensor([0.0878, 0.1254, 0.1205, 0.0883, 0.1141, 0.0543, 0.1338, 0.0676], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0355, 0.0311, 0.0253, 0.0302, 0.0253, 0.0314, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 04:22:24,473 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.44 vs. limit=2.0 +2023-04-03 04:22:42,846 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1081, 2.8471, 2.2492, 2.2658, 1.9836, 2.4749, 0.9119, 2.0300], + device='cuda:0'), covar=tensor([0.0706, 0.0691, 0.0722, 0.1219, 0.1317, 0.1195, 0.1615, 0.1223], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0360, 0.0363, 0.0388, 0.0467, 0.0393, 0.0341, 0.0346], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 04:22:47,693 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-04-03 04:23:06,176 INFO [train.py:903] (0/4) Epoch 25, batch 2750, loss[loss=0.2316, simple_loss=0.3161, pruned_loss=0.07352, over 17581.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2865, pruned_loss=0.06221, over 3819516.00 frames. ], batch size: 101, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:23:08,967 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166624.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:23:45,192 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.014e+02 4.918e+02 6.109e+02 7.546e+02 1.552e+03, threshold=1.222e+03, percent-clipped=5.0 +2023-04-03 04:23:46,641 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.8564, 5.3233, 3.2062, 4.5945, 1.3037, 5.4749, 5.2796, 5.4584], + device='cuda:0'), covar=tensor([0.0458, 0.0882, 0.1749, 0.0754, 0.3667, 0.0528, 0.0772, 0.1045], + device='cuda:0'), in_proj_covar=tensor([0.0520, 0.0421, 0.0507, 0.0355, 0.0404, 0.0447, 0.0440, 0.0472], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 04:24:04,554 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=166669.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:24:05,267 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.59 vs. limit=5.0 +2023-04-03 04:24:07,634 INFO [train.py:903] (0/4) Epoch 25, batch 2800, loss[loss=0.2097, simple_loss=0.2882, pruned_loss=0.06556, over 19517.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2863, pruned_loss=0.06198, over 3828192.25 frames. ], batch size: 56, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:25:04,291 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.1835, 5.2178, 5.9964, 6.0240, 2.0162, 5.6788, 4.6902, 5.6328], + device='cuda:0'), covar=tensor([0.1667, 0.0759, 0.0594, 0.0590, 0.6178, 0.0781, 0.0616, 0.1161], + device='cuda:0'), in_proj_covar=tensor([0.0801, 0.0764, 0.0973, 0.0852, 0.0850, 0.0738, 0.0578, 0.0902], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 04:25:11,690 INFO [train.py:903] (0/4) Epoch 25, batch 2850, loss[loss=0.2223, simple_loss=0.3041, pruned_loss=0.07027, over 18288.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.286, pruned_loss=0.06206, over 3820894.36 frames. ], batch size: 83, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:25:50,390 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.797e+02 4.563e+02 5.950e+02 8.060e+02 1.987e+03, threshold=1.190e+03, percent-clipped=10.0 +2023-04-03 04:25:50,650 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166753.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:26:11,843 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-03 04:26:14,215 INFO [train.py:903] (0/4) Epoch 25, batch 2900, loss[loss=0.2142, simple_loss=0.289, pruned_loss=0.06977, over 18417.00 frames. ], tot_loss[loss=0.2062, simple_loss=0.287, pruned_loss=0.06271, over 3826306.52 frames. ], batch size: 84, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:26:25,423 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-03 04:26:27,330 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166782.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:26:29,658 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166784.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:26:33,239 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166787.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 04:26:43,675 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5363, 1.6794, 1.9917, 1.8572, 3.0494, 2.7220, 3.3683, 1.6048], + device='cuda:0'), covar=tensor([0.2698, 0.4438, 0.2945, 0.1963, 0.1713, 0.2158, 0.1777, 0.4432], + device='cuda:0'), in_proj_covar=tensor([0.0547, 0.0660, 0.0735, 0.0497, 0.0629, 0.0542, 0.0669, 0.0564], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 04:27:05,052 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166812.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 04:27:07,294 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2761, 2.1913, 2.1275, 1.9416, 1.7848, 1.9716, 0.7568, 1.3292], + device='cuda:0'), covar=tensor([0.0630, 0.0641, 0.0434, 0.0719, 0.1130, 0.0890, 0.1361, 0.1031], + device='cuda:0'), in_proj_covar=tensor([0.0363, 0.0360, 0.0363, 0.0386, 0.0465, 0.0392, 0.0341, 0.0345], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 04:27:17,225 INFO [train.py:903] (0/4) Epoch 25, batch 2950, loss[loss=0.1921, simple_loss=0.2798, pruned_loss=0.05222, over 19689.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2869, pruned_loss=0.06296, over 3807849.07 frames. ], batch size: 59, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:27:44,142 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=166843.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:27:56,482 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.385e+02 4.674e+02 5.957e+02 7.713e+02 2.101e+03, threshold=1.191e+03, percent-clipped=6.0 +2023-04-03 04:28:12,816 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-04-03 04:28:18,965 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=166871.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:28:21,108 INFO [train.py:903] (0/4) Epoch 25, batch 3000, loss[loss=0.2213, simple_loss=0.3029, pruned_loss=0.0699, over 17476.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2859, pruned_loss=0.06232, over 3808957.27 frames. ], batch size: 101, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:28:21,108 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 04:28:33,781 INFO [train.py:937] (0/4) Epoch 25, validation: loss=0.1677, simple_loss=0.2674, pruned_loss=0.034, over 944034.00 frames. +2023-04-03 04:28:33,784 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 04:28:35,117 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-03 04:28:44,970 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=166880.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:29:01,375 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=166893.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:29:17,266 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=166905.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:29:38,482 INFO [train.py:903] (0/4) Epoch 25, batch 3050, loss[loss=0.2168, simple_loss=0.2961, pruned_loss=0.06871, over 19586.00 frames. ], tot_loss[loss=0.206, simple_loss=0.2869, pruned_loss=0.0626, over 3821681.12 frames. ], batch size: 61, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:30:09,924 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=166947.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:30:17,437 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.386e+02 4.942e+02 6.233e+02 8.295e+02 1.859e+03, threshold=1.247e+03, percent-clipped=9.0 +2023-04-03 04:30:23,328 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166958.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:30:41,892 INFO [train.py:903] (0/4) Epoch 25, batch 3100, loss[loss=0.2235, simple_loss=0.2988, pruned_loss=0.07413, over 19787.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2873, pruned_loss=0.06276, over 3838720.53 frames. ], batch size: 56, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:30:58,956 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=166986.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:31:27,826 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167008.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:31:44,389 INFO [train.py:903] (0/4) Epoch 25, batch 3150, loss[loss=0.2096, simple_loss=0.2962, pruned_loss=0.0615, over 19708.00 frames. ], tot_loss[loss=0.207, simple_loss=0.2874, pruned_loss=0.0633, over 3834349.71 frames. ], batch size: 59, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:32:07,978 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167040.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:32:08,795 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-03 04:32:24,694 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.516e+02 5.039e+02 6.011e+02 8.459e+02 2.094e+03, threshold=1.202e+03, percent-clipped=4.0 +2023-04-03 04:32:39,350 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167065.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:32:48,322 INFO [train.py:903] (0/4) Epoch 25, batch 3200, loss[loss=0.1941, simple_loss=0.2806, pruned_loss=0.05383, over 19673.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2867, pruned_loss=0.06249, over 3832412.89 frames. ], batch size: 60, lr: 3.29e-03, grad_scale: 8.0 +2023-04-03 04:33:20,600 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=167097.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:33:47,237 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167118.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:33:52,597 INFO [train.py:903] (0/4) Epoch 25, batch 3250, loss[loss=0.1952, simple_loss=0.2813, pruned_loss=0.05449, over 19539.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2857, pruned_loss=0.06207, over 3825783.99 frames. ], batch size: 56, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:33:52,893 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167122.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:33:57,483 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=167126.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:34:33,306 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.514e+02 4.897e+02 6.248e+02 7.764e+02 1.427e+03, threshold=1.250e+03, percent-clipped=4.0 +2023-04-03 04:34:56,638 INFO [train.py:903] (0/4) Epoch 25, batch 3300, loss[loss=0.1869, simple_loss=0.2769, pruned_loss=0.04845, over 19653.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2855, pruned_loss=0.0618, over 3822749.48 frames. ], batch size: 55, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:34:56,680 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-03 04:35:01,576 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4677, 1.3334, 1.5342, 1.5595, 3.0599, 1.2793, 2.4757, 3.4570], + device='cuda:0'), covar=tensor([0.0546, 0.2765, 0.2745, 0.1770, 0.0700, 0.2293, 0.1158, 0.0268], + device='cuda:0'), in_proj_covar=tensor([0.0417, 0.0372, 0.0393, 0.0349, 0.0376, 0.0353, 0.0390, 0.0410], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 04:35:01,647 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3360, 2.0273, 1.6400, 1.3438, 1.8484, 1.3170, 1.3116, 1.8398], + device='cuda:0'), covar=tensor([0.0930, 0.0772, 0.1027, 0.0834, 0.0564, 0.1287, 0.0631, 0.0437], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0319, 0.0340, 0.0267, 0.0249, 0.0344, 0.0292, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 04:35:30,194 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9309, 1.2472, 1.6484, 0.6008, 2.1089, 2.4758, 2.1577, 2.5970], + device='cuda:0'), covar=tensor([0.1588, 0.3639, 0.3261, 0.2660, 0.0594, 0.0290, 0.0355, 0.0393], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0329, 0.0360, 0.0268, 0.0250, 0.0192, 0.0219, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 04:35:47,894 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167212.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:35:50,094 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167214.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:35:59,034 INFO [train.py:903] (0/4) Epoch 25, batch 3350, loss[loss=0.2099, simple_loss=0.2867, pruned_loss=0.06659, over 19848.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.285, pruned_loss=0.06198, over 3832988.57 frames. ], batch size: 52, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:36:21,814 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167239.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:36:24,229 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167241.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:36:25,466 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167242.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:36:41,239 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.112e+02 4.821e+02 5.770e+02 7.565e+02 1.496e+03, threshold=1.154e+03, percent-clipped=2.0 +2023-04-03 04:36:53,507 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167264.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:36:56,875 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167267.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:37:02,319 INFO [train.py:903] (0/4) Epoch 25, batch 3400, loss[loss=0.2282, simple_loss=0.3105, pruned_loss=0.07291, over 19526.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2854, pruned_loss=0.06204, over 3831784.93 frames. ], batch size: 56, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:37:14,438 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9443, 1.2566, 1.6059, 0.6060, 2.1181, 2.4872, 2.1748, 2.6233], + device='cuda:0'), covar=tensor([0.1630, 0.3778, 0.3367, 0.2825, 0.0606, 0.0277, 0.0354, 0.0391], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0328, 0.0359, 0.0268, 0.0250, 0.0192, 0.0219, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 04:37:26,094 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167289.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:37:28,365 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=167291.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:38:08,142 INFO [train.py:903] (0/4) Epoch 25, batch 3450, loss[loss=0.2002, simple_loss=0.2779, pruned_loss=0.0612, over 19475.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2857, pruned_loss=0.06199, over 3821273.79 frames. ], batch size: 49, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:38:10,563 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-03 04:38:32,278 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5633, 1.3276, 1.4288, 2.2105, 1.6255, 1.8830, 1.8453, 1.6603], + device='cuda:0'), covar=tensor([0.0953, 0.1183, 0.1149, 0.0789, 0.0935, 0.0925, 0.0997, 0.0846], + device='cuda:0'), in_proj_covar=tensor([0.0214, 0.0224, 0.0227, 0.0240, 0.0227, 0.0215, 0.0189, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 04:38:49,780 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.112e+02 4.608e+02 5.631e+02 6.989e+02 1.333e+03, threshold=1.126e+03, percent-clipped=1.0 +2023-04-03 04:39:12,369 INFO [train.py:903] (0/4) Epoch 25, batch 3500, loss[loss=0.2338, simple_loss=0.3062, pruned_loss=0.0807, over 19651.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2861, pruned_loss=0.06226, over 3809349.44 frames. ], batch size: 55, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:39:53,211 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.77 vs. limit=2.0 +2023-04-03 04:39:55,192 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167406.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:40:07,539 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1162, 2.0965, 2.3189, 2.2238, 2.9946, 2.6568, 3.0779, 2.0044], + device='cuda:0'), covar=tensor([0.1819, 0.3129, 0.2066, 0.1537, 0.1149, 0.1708, 0.1151, 0.3455], + device='cuda:0'), in_proj_covar=tensor([0.0544, 0.0656, 0.0730, 0.0493, 0.0623, 0.0537, 0.0662, 0.0560], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 04:40:15,422 INFO [train.py:903] (0/4) Epoch 25, batch 3550, loss[loss=0.1825, simple_loss=0.2594, pruned_loss=0.05279, over 19366.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2855, pruned_loss=0.06221, over 3792712.70 frames. ], batch size: 48, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:40:38,488 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3137, 3.8513, 3.9536, 3.9646, 1.6498, 3.7839, 3.3257, 3.7175], + device='cuda:0'), covar=tensor([0.1779, 0.0931, 0.0712, 0.0815, 0.5643, 0.0928, 0.0727, 0.1200], + device='cuda:0'), in_proj_covar=tensor([0.0801, 0.0766, 0.0968, 0.0851, 0.0846, 0.0737, 0.0579, 0.0900], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 04:40:51,379 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167450.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:40:55,633 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.122e+02 5.129e+02 6.252e+02 7.981e+02 1.792e+03, threshold=1.250e+03, percent-clipped=6.0 +2023-04-03 04:41:06,569 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=167462.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:41:11,115 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=167466.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:41:13,878 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167468.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:41:17,992 INFO [train.py:903] (0/4) Epoch 25, batch 3600, loss[loss=0.2696, simple_loss=0.3571, pruned_loss=0.091, over 19521.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2847, pruned_loss=0.06132, over 3801119.84 frames. ], batch size: 64, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:41:44,827 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167493.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:41:50,856 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167497.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:42:16,930 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-04-03 04:42:20,841 INFO [train.py:903] (0/4) Epoch 25, batch 3650, loss[loss=0.1644, simple_loss=0.2419, pruned_loss=0.04352, over 18721.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2854, pruned_loss=0.06158, over 3808901.15 frames. ], batch size: 41, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:42:21,285 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167522.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:42:39,952 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.33 vs. limit=5.0 +2023-04-03 04:43:00,370 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.203e+02 5.264e+02 6.466e+02 8.043e+02 1.528e+03, threshold=1.293e+03, percent-clipped=3.0 +2023-04-03 04:43:24,179 INFO [train.py:903] (0/4) Epoch 25, batch 3700, loss[loss=0.2475, simple_loss=0.3252, pruned_loss=0.08486, over 19739.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2849, pruned_loss=0.06167, over 3817879.77 frames. ], batch size: 63, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:43:31,615 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167577.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:43:36,268 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167581.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:44:28,023 INFO [train.py:903] (0/4) Epoch 25, batch 3750, loss[loss=0.189, simple_loss=0.2741, pruned_loss=0.05198, over 19587.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2842, pruned_loss=0.06124, over 3834149.53 frames. ], batch size: 52, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:45:08,828 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.966e+02 5.285e+02 6.593e+02 8.150e+02 1.742e+03, threshold=1.319e+03, percent-clipped=4.0 +2023-04-03 04:45:19,857 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167662.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:45:31,292 INFO [train.py:903] (0/4) Epoch 25, batch 3800, loss[loss=0.225, simple_loss=0.3054, pruned_loss=0.07226, over 19673.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2844, pruned_loss=0.06129, over 3825298.65 frames. ], batch size: 60, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:45:50,865 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167687.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:46:01,954 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-03 04:46:19,189 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-04-03 04:46:35,343 INFO [train.py:903] (0/4) Epoch 25, batch 3850, loss[loss=0.2777, simple_loss=0.3477, pruned_loss=0.1038, over 19349.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2842, pruned_loss=0.06126, over 3821639.22 frames. ], batch size: 70, lr: 3.28e-03, grad_scale: 4.0 +2023-04-03 04:46:53,991 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167737.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:47:16,485 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.257e+02 4.991e+02 5.959e+02 7.597e+02 1.650e+03, threshold=1.192e+03, percent-clipped=2.0 +2023-04-03 04:47:24,053 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.45 vs. limit=5.0 +2023-04-03 04:47:39,476 INFO [train.py:903] (0/4) Epoch 25, batch 3900, loss[loss=0.2054, simple_loss=0.2949, pruned_loss=0.05796, over 19787.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2856, pruned_loss=0.06188, over 3812057.40 frames. ], batch size: 56, lr: 3.28e-03, grad_scale: 4.0 +2023-04-03 04:47:55,769 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=167785.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:47:59,106 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8326, 1.4177, 1.7165, 1.4578, 3.4052, 1.1193, 2.6531, 3.8471], + device='cuda:0'), covar=tensor([0.0485, 0.2972, 0.2764, 0.2029, 0.0685, 0.2597, 0.1120, 0.0229], + device='cuda:0'), in_proj_covar=tensor([0.0417, 0.0374, 0.0396, 0.0350, 0.0377, 0.0353, 0.0390, 0.0412], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 04:48:06,191 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=167794.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:48:43,212 INFO [train.py:903] (0/4) Epoch 25, batch 3950, loss[loss=0.2081, simple_loss=0.2907, pruned_loss=0.0627, over 19473.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2856, pruned_loss=0.06189, over 3817069.68 frames. ], batch size: 64, lr: 3.28e-03, grad_scale: 4.0 +2023-04-03 04:48:45,756 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-03 04:48:56,960 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167833.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:49:01,666 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=167837.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:49:25,131 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.090e+02 4.848e+02 5.890e+02 7.638e+02 1.655e+03, threshold=1.178e+03, percent-clipped=7.0 +2023-04-03 04:49:29,083 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167858.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:49:34,592 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=167862.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:49:46,053 INFO [train.py:903] (0/4) Epoch 25, batch 4000, loss[loss=0.2014, simple_loss=0.2894, pruned_loss=0.05669, over 18233.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2854, pruned_loss=0.06192, over 3797650.99 frames. ], batch size: 83, lr: 3.28e-03, grad_scale: 8.0 +2023-04-03 04:50:32,144 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-03 04:50:33,730 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=167909.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:50:50,808 INFO [train.py:903] (0/4) Epoch 25, batch 4050, loss[loss=0.1833, simple_loss=0.2539, pruned_loss=0.05637, over 19105.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2854, pruned_loss=0.06216, over 3794303.35 frames. ], batch size: 42, lr: 3.28e-03, grad_scale: 4.0 +2023-04-03 04:50:59,278 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6173, 1.7298, 2.0392, 1.7636, 2.6308, 3.0170, 2.8786, 3.1842], + device='cuda:0'), covar=tensor([0.1425, 0.3125, 0.2806, 0.2452, 0.1181, 0.0316, 0.0253, 0.0399], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0328, 0.0359, 0.0269, 0.0250, 0.0192, 0.0218, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 04:51:32,272 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.281e+02 5.057e+02 6.529e+02 8.113e+02 1.821e+03, threshold=1.306e+03, percent-clipped=7.0 +2023-04-03 04:51:52,934 INFO [train.py:903] (0/4) Epoch 25, batch 4100, loss[loss=0.1961, simple_loss=0.2739, pruned_loss=0.05913, over 19753.00 frames. ], tot_loss[loss=0.2059, simple_loss=0.2861, pruned_loss=0.06282, over 3784756.80 frames. ], batch size: 47, lr: 3.28e-03, grad_scale: 4.0 +2023-04-03 04:52:27,684 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-168000.pt +2023-04-03 04:52:29,047 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-03 04:52:56,839 INFO [train.py:903] (0/4) Epoch 25, batch 4150, loss[loss=0.2225, simple_loss=0.3037, pruned_loss=0.07064, over 19544.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2851, pruned_loss=0.06188, over 3801087.45 frames. ], batch size: 56, lr: 3.28e-03, grad_scale: 4.0 +2023-04-03 04:53:03,561 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8850, 4.3319, 4.6455, 4.6310, 1.8479, 4.3579, 3.7321, 4.3286], + device='cuda:0'), covar=tensor([0.1789, 0.1019, 0.0632, 0.0792, 0.6378, 0.1071, 0.0778, 0.1251], + device='cuda:0'), in_proj_covar=tensor([0.0810, 0.0778, 0.0982, 0.0862, 0.0861, 0.0747, 0.0587, 0.0914], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 04:53:36,015 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=168053.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:53:39,909 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.124e+02 5.393e+02 6.582e+02 8.080e+02 1.683e+03, threshold=1.316e+03, percent-clipped=2.0 +2023-04-03 04:53:59,619 INFO [train.py:903] (0/4) Epoch 25, batch 4200, loss[loss=0.2399, simple_loss=0.326, pruned_loss=0.07684, over 19780.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2857, pruned_loss=0.06195, over 3810461.06 frames. ], batch size: 54, lr: 3.28e-03, grad_scale: 4.0 +2023-04-03 04:54:01,993 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-03 04:54:10,944 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=168081.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:55:03,249 INFO [train.py:903] (0/4) Epoch 25, batch 4250, loss[loss=0.1928, simple_loss=0.2819, pruned_loss=0.05178, over 19474.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2858, pruned_loss=0.06182, over 3820754.59 frames. ], batch size: 64, lr: 3.27e-03, grad_scale: 4.0 +2023-04-03 04:55:13,202 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=168129.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:55:17,973 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-03 04:55:29,583 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-03 04:55:46,758 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.141e+02 4.961e+02 6.439e+02 7.740e+02 2.119e+03, threshold=1.288e+03, percent-clipped=3.0 +2023-04-03 04:55:59,617 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=168165.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:56:07,296 INFO [train.py:903] (0/4) Epoch 25, batch 4300, loss[loss=0.2024, simple_loss=0.277, pruned_loss=0.06394, over 19764.00 frames. ], tot_loss[loss=0.205, simple_loss=0.286, pruned_loss=0.06197, over 3820688.95 frames. ], batch size: 48, lr: 3.27e-03, grad_scale: 4.0 +2023-04-03 04:56:30,423 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=168190.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:56:37,434 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=168196.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:57:00,345 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-03 04:57:10,762 INFO [train.py:903] (0/4) Epoch 25, batch 4350, loss[loss=0.2097, simple_loss=0.2725, pruned_loss=0.07351, over 19754.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.285, pruned_loss=0.06187, over 3820425.42 frames. ], batch size: 46, lr: 3.27e-03, grad_scale: 4.0 +2023-04-03 04:57:38,635 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=168244.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:57:47,841 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=168251.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 04:57:53,358 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.429e+02 4.774e+02 5.735e+02 7.211e+02 1.236e+03, threshold=1.147e+03, percent-clipped=0.0 +2023-04-03 04:58:13,423 INFO [train.py:903] (0/4) Epoch 25, batch 4400, loss[loss=0.1953, simple_loss=0.2785, pruned_loss=0.05604, over 19615.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2856, pruned_loss=0.0617, over 3824276.41 frames. ], batch size: 50, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 04:58:20,092 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.46 vs. limit=5.0 +2023-04-03 04:58:40,147 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-03 04:58:50,386 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-03 04:59:16,257 INFO [train.py:903] (0/4) Epoch 25, batch 4450, loss[loss=0.229, simple_loss=0.3081, pruned_loss=0.07492, over 19679.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2855, pruned_loss=0.06142, over 3826800.33 frames. ], batch size: 59, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 04:59:59,623 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.939e+02 4.609e+02 5.804e+02 7.720e+02 1.927e+03, threshold=1.161e+03, percent-clipped=7.0 +2023-04-03 05:00:20,131 INFO [train.py:903] (0/4) Epoch 25, batch 4500, loss[loss=0.1998, simple_loss=0.2763, pruned_loss=0.06167, over 19607.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.285, pruned_loss=0.06145, over 3829965.45 frames. ], batch size: 50, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:00:23,958 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7363, 2.6100, 2.2682, 2.1290, 1.9571, 2.3772, 1.1468, 1.9575], + device='cuda:0'), covar=tensor([0.0714, 0.0656, 0.0563, 0.1001, 0.1029, 0.1063, 0.1322, 0.1028], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0363, 0.0362, 0.0389, 0.0466, 0.0396, 0.0344, 0.0346], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 05:00:52,070 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=168397.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:01:23,385 INFO [train.py:903] (0/4) Epoch 25, batch 4550, loss[loss=0.1735, simple_loss=0.2502, pruned_loss=0.04833, over 19737.00 frames. ], tot_loss[loss=0.204, simple_loss=0.285, pruned_loss=0.06148, over 3828630.77 frames. ], batch size: 46, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:01:34,636 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-03 05:01:59,987 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-03 05:02:02,774 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=168452.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:02:08,833 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.350e+02 4.803e+02 5.692e+02 6.651e+02 1.392e+03, threshold=1.138e+03, percent-clipped=3.0 +2023-04-03 05:02:27,732 INFO [train.py:903] (0/4) Epoch 25, batch 4600, loss[loss=0.1855, simple_loss=0.266, pruned_loss=0.05252, over 19681.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2852, pruned_loss=0.06155, over 3809908.03 frames. ], batch size: 53, lr: 3.27e-03, grad_scale: 4.0 +2023-04-03 05:02:35,470 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=168477.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:03:04,758 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=168500.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:03:20,621 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=168512.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:03:31,424 INFO [train.py:903] (0/4) Epoch 25, batch 4650, loss[loss=0.1728, simple_loss=0.2473, pruned_loss=0.04913, over 19758.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2848, pruned_loss=0.06142, over 3792449.69 frames. ], batch size: 47, lr: 3.27e-03, grad_scale: 4.0 +2023-04-03 05:03:35,315 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=168525.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:03:50,743 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-03 05:04:02,095 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-03 05:04:16,048 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.438e+02 5.060e+02 6.015e+02 7.632e+02 1.453e+03, threshold=1.203e+03, percent-clipped=3.0 +2023-04-03 05:04:34,774 INFO [train.py:903] (0/4) Epoch 25, batch 4700, loss[loss=0.1524, simple_loss=0.2345, pruned_loss=0.03517, over 19749.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2855, pruned_loss=0.06239, over 3774835.78 frames. ], batch size: 46, lr: 3.27e-03, grad_scale: 4.0 +2023-04-03 05:04:57,660 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-03 05:05:04,401 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=168595.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:05:15,549 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-03 05:05:38,974 INFO [train.py:903] (0/4) Epoch 25, batch 4750, loss[loss=0.2411, simple_loss=0.3174, pruned_loss=0.08236, over 19328.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2857, pruned_loss=0.06266, over 3791880.74 frames. ], batch size: 66, lr: 3.27e-03, grad_scale: 4.0 +2023-04-03 05:06:16,689 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.26 vs. limit=5.0 +2023-04-03 05:06:22,552 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.376e+02 5.029e+02 6.051e+02 7.124e+02 1.309e+03, threshold=1.210e+03, percent-clipped=1.0 +2023-04-03 05:06:40,987 INFO [train.py:903] (0/4) Epoch 25, batch 4800, loss[loss=0.1979, simple_loss=0.269, pruned_loss=0.06337, over 19751.00 frames. ], tot_loss[loss=0.2064, simple_loss=0.2868, pruned_loss=0.06302, over 3797180.43 frames. ], batch size: 46, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:07:29,213 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=168710.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:07:43,819 INFO [train.py:903] (0/4) Epoch 25, batch 4850, loss[loss=0.1544, simple_loss=0.2324, pruned_loss=0.03822, over 19099.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2863, pruned_loss=0.06269, over 3802464.30 frames. ], batch size: 42, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:08:09,441 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-03 05:08:29,299 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.525e+02 5.536e+02 6.758e+02 9.275e+02 1.787e+03, threshold=1.352e+03, percent-clipped=12.0 +2023-04-03 05:08:30,524 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-03 05:08:36,430 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-03 05:08:36,465 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-03 05:08:43,844 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=168768.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:08:47,084 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-03 05:08:48,226 INFO [train.py:903] (0/4) Epoch 25, batch 4900, loss[loss=0.207, simple_loss=0.2907, pruned_loss=0.06164, over 19759.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.286, pruned_loss=0.06229, over 3822335.12 frames. ], batch size: 54, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:08:49,804 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3968, 1.4082, 1.6645, 1.6145, 2.1014, 2.0200, 2.2491, 0.8188], + device='cuda:0'), covar=tensor([0.2554, 0.4430, 0.2755, 0.2008, 0.1736, 0.2309, 0.1566, 0.5049], + device='cuda:0'), in_proj_covar=tensor([0.0548, 0.0666, 0.0738, 0.0500, 0.0630, 0.0543, 0.0670, 0.0567], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 05:09:06,568 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-03 05:09:15,992 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=168793.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:09:39,555 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0934, 1.1799, 1.6028, 1.2860, 2.6282, 3.5623, 3.2635, 3.8309], + device='cuda:0'), covar=tensor([0.1885, 0.4247, 0.3780, 0.2675, 0.0656, 0.0204, 0.0270, 0.0296], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0327, 0.0357, 0.0268, 0.0248, 0.0191, 0.0218, 0.0267], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 05:09:52,877 INFO [train.py:903] (0/4) Epoch 25, batch 4950, loss[loss=0.2453, simple_loss=0.3103, pruned_loss=0.09011, over 13308.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2862, pruned_loss=0.06253, over 3798727.54 frames. ], batch size: 135, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:10:04,477 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-03 05:10:30,146 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-03 05:10:36,919 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.098e+02 4.693e+02 5.645e+02 7.417e+02 1.662e+03, threshold=1.129e+03, percent-clipped=1.0 +2023-04-03 05:10:55,836 INFO [train.py:903] (0/4) Epoch 25, batch 5000, loss[loss=0.184, simple_loss=0.2757, pruned_loss=0.04617, over 19648.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2851, pruned_loss=0.06149, over 3816526.30 frames. ], batch size: 55, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:11:02,557 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-03 05:11:13,727 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-03 05:11:58,385 INFO [train.py:903] (0/4) Epoch 25, batch 5050, loss[loss=0.2233, simple_loss=0.2938, pruned_loss=0.0764, over 19779.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2847, pruned_loss=0.0611, over 3827739.24 frames. ], batch size: 56, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:12:33,964 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-03 05:12:41,881 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.044e+02 4.713e+02 5.534e+02 7.099e+02 1.364e+03, threshold=1.107e+03, percent-clipped=2.0 +2023-04-03 05:12:55,393 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=168966.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:13:02,136 INFO [train.py:903] (0/4) Epoch 25, batch 5100, loss[loss=0.1682, simple_loss=0.2463, pruned_loss=0.04506, over 19399.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2844, pruned_loss=0.06098, over 3817095.40 frames. ], batch size: 47, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:13:11,271 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-03 05:13:14,765 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-03 05:13:19,272 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=168985.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:13:20,158 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-03 05:13:21,484 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=168987.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 05:13:23,824 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6005, 4.0933, 4.2627, 4.2759, 1.6519, 4.0727, 3.4911, 3.9742], + device='cuda:0'), covar=tensor([0.1620, 0.0886, 0.0619, 0.0691, 0.5928, 0.0880, 0.0715, 0.1188], + device='cuda:0'), in_proj_covar=tensor([0.0805, 0.0775, 0.0978, 0.0861, 0.0852, 0.0741, 0.0583, 0.0912], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 05:13:26,009 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=168991.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:13:45,573 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.54 vs. limit=2.0 +2023-04-03 05:14:05,076 INFO [train.py:903] (0/4) Epoch 25, batch 5150, loss[loss=0.1953, simple_loss=0.2777, pruned_loss=0.05642, over 19661.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2852, pruned_loss=0.06142, over 3818229.73 frames. ], batch size: 55, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:14:09,680 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0536, 3.7180, 2.5985, 3.3491, 0.9510, 3.6655, 3.5295, 3.6124], + device='cuda:0'), covar=tensor([0.0814, 0.1137, 0.1861, 0.0877, 0.3670, 0.0792, 0.1041, 0.1173], + device='cuda:0'), in_proj_covar=tensor([0.0525, 0.0427, 0.0512, 0.0358, 0.0412, 0.0454, 0.0447, 0.0477], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:14:12,536 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-03 05:14:16,410 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-03 05:14:48,109 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.896e+02 5.295e+02 6.704e+02 8.101e+02 2.101e+03, threshold=1.341e+03, percent-clipped=6.0 +2023-04-03 05:14:52,555 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 05:15:08,252 INFO [train.py:903] (0/4) Epoch 25, batch 5200, loss[loss=0.1632, simple_loss=0.2481, pruned_loss=0.03916, over 19722.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2857, pruned_loss=0.06156, over 3830534.69 frames. ], batch size: 47, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:15:23,489 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-03 05:15:29,718 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6926, 1.8047, 2.0547, 2.0813, 1.6027, 2.0255, 2.0546, 1.8690], + device='cuda:0'), covar=tensor([0.4216, 0.3690, 0.1995, 0.2393, 0.4026, 0.2264, 0.5152, 0.3469], + device='cuda:0'), in_proj_covar=tensor([0.0923, 0.0996, 0.0733, 0.0945, 0.0900, 0.0835, 0.0853, 0.0799], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 05:15:44,161 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=169100.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:16:08,701 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-03 05:16:13,025 INFO [train.py:903] (0/4) Epoch 25, batch 5250, loss[loss=0.2188, simple_loss=0.298, pruned_loss=0.06981, over 13042.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2866, pruned_loss=0.06193, over 3825026.48 frames. ], batch size: 136, lr: 3.27e-03, grad_scale: 8.0 +2023-04-03 05:16:26,986 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0962, 1.9862, 1.8098, 1.7027, 1.5182, 1.6446, 0.4564, 1.0829], + device='cuda:0'), covar=tensor([0.0653, 0.0693, 0.0514, 0.0792, 0.1183, 0.1069, 0.1416, 0.1118], + device='cuda:0'), in_proj_covar=tensor([0.0365, 0.0363, 0.0364, 0.0391, 0.0467, 0.0399, 0.0344, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 05:16:27,869 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=169134.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:16:56,983 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.150e+02 5.211e+02 5.791e+02 7.204e+02 1.532e+03, threshold=1.158e+03, percent-clipped=2.0 +2023-04-03 05:17:05,412 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-03 05:17:16,182 INFO [train.py:903] (0/4) Epoch 25, batch 5300, loss[loss=0.2062, simple_loss=0.292, pruned_loss=0.0602, over 19473.00 frames. ], tot_loss[loss=0.2054, simple_loss=0.2867, pruned_loss=0.06203, over 3808996.81 frames. ], batch size: 64, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:17:22,363 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=169177.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:17:24,887 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8003, 1.4424, 1.6444, 1.5714, 3.4019, 1.2580, 2.5098, 3.8221], + device='cuda:0'), covar=tensor([0.0561, 0.2965, 0.2922, 0.1911, 0.0718, 0.2451, 0.1291, 0.0254], + device='cuda:0'), in_proj_covar=tensor([0.0419, 0.0372, 0.0394, 0.0351, 0.0377, 0.0353, 0.0391, 0.0411], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:17:34,475 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-03 05:17:49,978 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9976, 1.9148, 1.7796, 1.6096, 1.5367, 1.6048, 0.4888, 0.8536], + device='cuda:0'), covar=tensor([0.0646, 0.0643, 0.0487, 0.0691, 0.1208, 0.0850, 0.1303, 0.1099], + device='cuda:0'), in_proj_covar=tensor([0.0365, 0.0363, 0.0366, 0.0392, 0.0468, 0.0400, 0.0345, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 05:18:18,923 INFO [train.py:903] (0/4) Epoch 25, batch 5350, loss[loss=0.2224, simple_loss=0.2925, pruned_loss=0.07612, over 19490.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2855, pruned_loss=0.06178, over 3809255.85 frames. ], batch size: 49, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:18:21,487 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6091, 1.5198, 1.5724, 1.8514, 1.3943, 1.7646, 1.7745, 1.6506], + device='cuda:0'), covar=tensor([0.0836, 0.0930, 0.0957, 0.0661, 0.0859, 0.0771, 0.0857, 0.0691], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0224, 0.0226, 0.0239, 0.0227, 0.0214, 0.0189, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 05:18:56,390 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-03 05:19:04,345 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.964e+02 4.992e+02 6.506e+02 8.048e+02 1.510e+03, threshold=1.301e+03, percent-clipped=6.0 +2023-04-03 05:19:24,344 INFO [train.py:903] (0/4) Epoch 25, batch 5400, loss[loss=0.1852, simple_loss=0.2607, pruned_loss=0.05485, over 19356.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2848, pruned_loss=0.06137, over 3817612.03 frames. ], batch size: 47, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:20:27,248 INFO [train.py:903] (0/4) Epoch 25, batch 5450, loss[loss=0.2025, simple_loss=0.2884, pruned_loss=0.05827, over 19566.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2842, pruned_loss=0.06104, over 3817671.23 frames. ], batch size: 52, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:20:36,563 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=169329.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:20:39,940 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=169331.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 05:21:11,192 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.08 vs. limit=5.0 +2023-04-03 05:21:11,465 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.767e+02 4.467e+02 5.289e+02 7.172e+02 1.661e+03, threshold=1.058e+03, percent-clipped=2.0 +2023-04-03 05:21:29,438 INFO [train.py:903] (0/4) Epoch 25, batch 5500, loss[loss=0.1845, simple_loss=0.2718, pruned_loss=0.04857, over 19534.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2845, pruned_loss=0.0614, over 3806291.07 frames. ], batch size: 54, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:21:57,709 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-03 05:22:03,495 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-04-03 05:22:33,002 INFO [train.py:903] (0/4) Epoch 25, batch 5550, loss[loss=0.1761, simple_loss=0.254, pruned_loss=0.04909, over 19779.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2845, pruned_loss=0.06176, over 3815183.48 frames. ], batch size: 47, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:22:43,779 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-03 05:23:00,564 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=169444.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:23:00,789 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=169444.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:23:04,030 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=169446.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 05:23:17,173 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.302e+02 5.192e+02 6.082e+02 7.576e+02 1.216e+03, threshold=1.216e+03, percent-clipped=3.0 +2023-04-03 05:23:33,243 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-03 05:23:36,942 INFO [train.py:903] (0/4) Epoch 25, batch 5600, loss[loss=0.1986, simple_loss=0.2883, pruned_loss=0.05447, over 17398.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2837, pruned_loss=0.06137, over 3808842.32 frames. ], batch size: 101, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:23:44,276 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=169478.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:24:08,196 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.00 vs. limit=5.0 +2023-04-03 05:24:38,800 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=169521.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:24:39,764 INFO [train.py:903] (0/4) Epoch 25, batch 5650, loss[loss=0.1731, simple_loss=0.2508, pruned_loss=0.04766, over 19360.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2832, pruned_loss=0.06118, over 3819352.81 frames. ], batch size: 47, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:25:24,920 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.766e+02 4.961e+02 6.158e+02 8.288e+02 1.627e+03, threshold=1.232e+03, percent-clipped=5.0 +2023-04-03 05:25:27,520 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=169559.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:25:30,437 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-03 05:25:43,204 INFO [train.py:903] (0/4) Epoch 25, batch 5700, loss[loss=0.2136, simple_loss=0.2961, pruned_loss=0.06556, over 19711.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2834, pruned_loss=0.06108, over 3818525.01 frames. ], batch size: 63, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:26:11,331 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=169593.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:26:47,655 INFO [train.py:903] (0/4) Epoch 25, batch 5750, loss[loss=0.1833, simple_loss=0.2738, pruned_loss=0.04643, over 19655.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2829, pruned_loss=0.06074, over 3821555.73 frames. ], batch size: 53, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:26:48,813 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-03 05:26:59,216 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-03 05:27:04,018 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-03 05:27:06,664 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=169636.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:27:10,213 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=169639.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:27:32,989 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.159e+02 4.765e+02 5.968e+02 8.362e+02 1.575e+03, threshold=1.194e+03, percent-clipped=5.0 +2023-04-03 05:27:52,469 INFO [train.py:903] (0/4) Epoch 25, batch 5800, loss[loss=0.2306, simple_loss=0.3144, pruned_loss=0.07337, over 18133.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2835, pruned_loss=0.06102, over 3816346.54 frames. ], batch size: 83, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:28:27,979 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=169700.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:28:30,416 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=169702.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 05:28:43,033 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.1636, 5.1938, 6.0417, 6.0433, 2.0622, 5.7047, 4.8097, 5.6619], + device='cuda:0'), covar=tensor([0.1734, 0.0725, 0.0502, 0.0585, 0.6248, 0.0814, 0.0611, 0.1087], + device='cuda:0'), in_proj_covar=tensor([0.0805, 0.0771, 0.0978, 0.0859, 0.0851, 0.0743, 0.0583, 0.0911], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 05:28:55,630 INFO [train.py:903] (0/4) Epoch 25, batch 5850, loss[loss=0.1839, simple_loss=0.2739, pruned_loss=0.04695, over 19678.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2841, pruned_loss=0.0612, over 3801169.27 frames. ], batch size: 53, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:29:00,320 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=169725.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:29:02,807 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=169727.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 05:29:02,861 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5468, 1.7668, 2.1677, 1.9815, 3.1663, 2.4644, 3.2056, 1.7213], + device='cuda:0'), covar=tensor([0.2846, 0.4930, 0.3080, 0.2127, 0.1629, 0.2652, 0.1966, 0.4775], + device='cuda:0'), in_proj_covar=tensor([0.0547, 0.0663, 0.0736, 0.0501, 0.0629, 0.0542, 0.0667, 0.0564], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 05:29:12,335 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-03 05:29:28,402 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=169747.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:29:41,205 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.382e+02 5.003e+02 6.037e+02 8.413e+02 1.989e+03, threshold=1.207e+03, percent-clipped=6.0 +2023-04-03 05:29:51,726 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=169765.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:29:56,994 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.79 vs. limit=2.0 +2023-04-03 05:29:59,829 INFO [train.py:903] (0/4) Epoch 25, batch 5900, loss[loss=0.2563, simple_loss=0.3245, pruned_loss=0.0941, over 13172.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2844, pruned_loss=0.06133, over 3806701.47 frames. ], batch size: 136, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:30:04,518 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-03 05:30:27,877 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-03 05:30:41,004 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=169804.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:30:56,622 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=169815.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:31:04,208 INFO [train.py:903] (0/4) Epoch 25, batch 5950, loss[loss=0.2062, simple_loss=0.2888, pruned_loss=0.06175, over 19681.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2852, pruned_loss=0.06193, over 3785097.79 frames. ], batch size: 59, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:31:28,374 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=169840.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:31:39,964 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=169849.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:31:49,615 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.314e+02 4.837e+02 6.128e+02 7.395e+02 1.765e+03, threshold=1.226e+03, percent-clipped=3.0 +2023-04-03 05:32:09,283 INFO [train.py:903] (0/4) Epoch 25, batch 6000, loss[loss=0.2269, simple_loss=0.2987, pruned_loss=0.07758, over 18041.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2845, pruned_loss=0.0616, over 3801074.28 frames. ], batch size: 83, lr: 3.26e-03, grad_scale: 8.0 +2023-04-03 05:32:09,284 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 05:32:21,935 INFO [train.py:937] (0/4) Epoch 25, validation: loss=0.1675, simple_loss=0.2674, pruned_loss=0.03383, over 944034.00 frames. +2023-04-03 05:32:21,936 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 05:32:24,540 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=169874.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:32:30,418 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.3081, 5.2632, 6.1776, 6.1945, 2.4296, 5.8089, 4.9163, 5.8500], + device='cuda:0'), covar=tensor([0.1626, 0.0681, 0.0517, 0.0551, 0.5628, 0.0642, 0.0621, 0.0930], + device='cuda:0'), in_proj_covar=tensor([0.0800, 0.0767, 0.0974, 0.0854, 0.0848, 0.0738, 0.0581, 0.0905], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 05:32:35,051 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3647, 1.3879, 2.1101, 1.7505, 3.0510, 4.8444, 4.6250, 5.1723], + device='cuda:0'), covar=tensor([0.1592, 0.3942, 0.3187, 0.2192, 0.0633, 0.0190, 0.0184, 0.0212], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0327, 0.0356, 0.0267, 0.0248, 0.0192, 0.0217, 0.0268], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 05:32:48,340 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=169892.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:33:19,610 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5309, 1.5504, 1.7819, 1.7114, 2.7049, 2.3869, 2.8511, 1.2048], + device='cuda:0'), covar=tensor([0.2429, 0.4296, 0.2662, 0.1991, 0.1493, 0.1993, 0.1347, 0.4552], + device='cuda:0'), in_proj_covar=tensor([0.0549, 0.0666, 0.0739, 0.0503, 0.0632, 0.0544, 0.0670, 0.0567], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 05:33:20,526 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=169917.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:33:26,959 INFO [train.py:903] (0/4) Epoch 25, batch 6050, loss[loss=0.2139, simple_loss=0.3024, pruned_loss=0.06268, over 19781.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2843, pruned_loss=0.0616, over 3814731.80 frames. ], batch size: 56, lr: 3.26e-03, grad_scale: 4.0 +2023-04-03 05:33:57,316 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.52 vs. limit=5.0 +2023-04-03 05:34:12,793 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.356e+02 4.971e+02 6.200e+02 7.955e+02 1.563e+03, threshold=1.240e+03, percent-clipped=4.0 +2023-04-03 05:34:30,155 INFO [train.py:903] (0/4) Epoch 25, batch 6100, loss[loss=0.2265, simple_loss=0.3139, pruned_loss=0.06957, over 19657.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.285, pruned_loss=0.06174, over 3807111.81 frames. ], batch size: 55, lr: 3.26e-03, grad_scale: 4.0 +2023-04-03 05:34:39,019 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.41 vs. limit=2.0 +2023-04-03 05:34:44,076 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=169983.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:34:44,436 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0457, 2.1202, 2.3969, 2.6362, 2.0639, 2.5617, 2.4592, 2.1959], + device='cuda:0'), covar=tensor([0.4353, 0.4092, 0.1944, 0.2598, 0.4284, 0.2227, 0.4884, 0.3465], + device='cuda:0'), in_proj_covar=tensor([0.0921, 0.0997, 0.0733, 0.0944, 0.0900, 0.0833, 0.0855, 0.0797], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 05:35:05,921 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-170000.pt +2023-04-03 05:35:35,276 INFO [train.py:903] (0/4) Epoch 25, batch 6150, loss[loss=0.2204, simple_loss=0.2877, pruned_loss=0.07655, over 19615.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2845, pruned_loss=0.0617, over 3801622.90 frames. ], batch size: 50, lr: 3.26e-03, grad_scale: 4.0 +2023-04-03 05:35:57,194 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.31 vs. limit=5.0 +2023-04-03 05:36:07,476 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-03 05:36:22,377 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.332e+02 4.967e+02 5.743e+02 7.363e+02 2.013e+03, threshold=1.149e+03, percent-clipped=2.0 +2023-04-03 05:36:28,743 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=170063.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:36:40,357 INFO [train.py:903] (0/4) Epoch 25, batch 6200, loss[loss=0.2372, simple_loss=0.3065, pruned_loss=0.08391, over 13350.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2857, pruned_loss=0.06244, over 3792928.28 frames. ], batch size: 137, lr: 3.26e-03, grad_scale: 4.0 +2023-04-03 05:37:04,390 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=170091.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:37:13,796 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=170098.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:37:26,457 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=170109.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:37:43,645 INFO [train.py:903] (0/4) Epoch 25, batch 6250, loss[loss=0.2035, simple_loss=0.2873, pruned_loss=0.05988, over 19324.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2855, pruned_loss=0.06236, over 3807193.34 frames. ], batch size: 66, lr: 3.26e-03, grad_scale: 4.0 +2023-04-03 05:38:16,084 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-03 05:38:16,216 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=170148.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:38:29,529 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.163e+02 5.199e+02 6.170e+02 7.836e+02 1.706e+03, threshold=1.234e+03, percent-clipped=7.0 +2023-04-03 05:38:47,601 INFO [train.py:903] (0/4) Epoch 25, batch 6300, loss[loss=0.1894, simple_loss=0.2754, pruned_loss=0.0517, over 19852.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2847, pruned_loss=0.06219, over 3790836.65 frames. ], batch size: 52, lr: 3.26e-03, grad_scale: 4.0 +2023-04-03 05:39:13,781 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.4335, 4.0825, 2.6837, 3.6217, 1.0835, 4.0501, 3.8849, 3.9376], + device='cuda:0'), covar=tensor([0.0628, 0.0987, 0.1854, 0.0802, 0.3618, 0.0686, 0.0857, 0.1066], + device='cuda:0'), in_proj_covar=tensor([0.0516, 0.0421, 0.0504, 0.0353, 0.0406, 0.0445, 0.0442, 0.0471], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:39:32,208 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=170206.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:39:51,012 INFO [train.py:903] (0/4) Epoch 25, batch 6350, loss[loss=0.1916, simple_loss=0.2626, pruned_loss=0.06031, over 19398.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2838, pruned_loss=0.06162, over 3794608.95 frames. ], batch size: 48, lr: 3.25e-03, grad_scale: 4.0 +2023-04-03 05:39:53,853 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=170224.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:40:36,692 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.666e+02 4.417e+02 5.475e+02 7.140e+02 1.571e+03, threshold=1.095e+03, percent-clipped=3.0 +2023-04-03 05:40:40,362 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3937, 3.1328, 2.3537, 2.8402, 0.8045, 3.1080, 2.9480, 3.0098], + device='cuda:0'), covar=tensor([0.0970, 0.1231, 0.1866, 0.0977, 0.3602, 0.0869, 0.1076, 0.1409], + device='cuda:0'), in_proj_covar=tensor([0.0514, 0.0418, 0.0501, 0.0351, 0.0403, 0.0442, 0.0439, 0.0469], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:40:42,874 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1107, 1.2902, 1.4306, 1.5183, 2.7584, 1.1041, 2.2379, 3.1392], + device='cuda:0'), covar=tensor([0.0650, 0.2836, 0.3015, 0.1755, 0.0800, 0.2400, 0.1244, 0.0307], + device='cuda:0'), in_proj_covar=tensor([0.0416, 0.0371, 0.0392, 0.0351, 0.0377, 0.0352, 0.0391, 0.0409], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:40:42,919 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=170263.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:40:54,618 INFO [train.py:903] (0/4) Epoch 25, batch 6400, loss[loss=0.2092, simple_loss=0.2869, pruned_loss=0.06573, over 19591.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2833, pruned_loss=0.06162, over 3794526.27 frames. ], batch size: 52, lr: 3.25e-03, grad_scale: 8.0 +2023-04-03 05:41:22,300 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.98 vs. limit=2.0 +2023-04-03 05:41:59,017 INFO [train.py:903] (0/4) Epoch 25, batch 6450, loss[loss=0.2272, simple_loss=0.3093, pruned_loss=0.07253, over 18836.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2836, pruned_loss=0.06125, over 3791048.31 frames. ], batch size: 74, lr: 3.25e-03, grad_scale: 8.0 +2023-04-03 05:42:38,962 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8492, 4.0558, 4.4060, 4.4087, 2.5713, 4.1345, 3.7531, 4.1688], + device='cuda:0'), covar=tensor([0.1433, 0.2574, 0.0652, 0.0723, 0.4823, 0.1238, 0.0613, 0.1081], + device='cuda:0'), in_proj_covar=tensor([0.0804, 0.0771, 0.0976, 0.0855, 0.0849, 0.0738, 0.0581, 0.0909], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 05:42:40,327 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=170354.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:42:44,565 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.061e+02 4.688e+02 6.468e+02 8.318e+02 2.178e+03, threshold=1.294e+03, percent-clipped=13.0 +2023-04-03 05:42:45,737 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-03 05:43:03,216 INFO [train.py:903] (0/4) Epoch 25, batch 6500, loss[loss=0.2179, simple_loss=0.298, pruned_loss=0.06884, over 13763.00 frames. ], tot_loss[loss=0.2029, simple_loss=0.2833, pruned_loss=0.06128, over 3797298.30 frames. ], batch size: 136, lr: 3.25e-03, grad_scale: 8.0 +2023-04-03 05:43:07,955 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-03 05:43:12,745 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=170379.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:43:48,121 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=170407.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:44:06,261 INFO [train.py:903] (0/4) Epoch 25, batch 6550, loss[loss=0.1605, simple_loss=0.2448, pruned_loss=0.03811, over 19601.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2831, pruned_loss=0.06154, over 3790740.40 frames. ], batch size: 50, lr: 3.25e-03, grad_scale: 8.0 +2023-04-03 05:44:52,752 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.983e+02 4.881e+02 6.158e+02 7.799e+02 1.457e+03, threshold=1.232e+03, percent-clipped=1.0 +2023-04-03 05:44:57,801 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=170462.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:45:09,732 INFO [train.py:903] (0/4) Epoch 25, batch 6600, loss[loss=0.1722, simple_loss=0.2505, pruned_loss=0.047, over 19767.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2831, pruned_loss=0.06143, over 3797719.36 frames. ], batch size: 47, lr: 3.25e-03, grad_scale: 8.0 +2023-04-03 05:45:19,112 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=170479.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:45:20,460 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=170480.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:45:28,732 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=170487.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:45:51,405 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=170505.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:46:09,044 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=170519.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:46:13,014 INFO [train.py:903] (0/4) Epoch 25, batch 6650, loss[loss=0.1975, simple_loss=0.2834, pruned_loss=0.05583, over 19668.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.283, pruned_loss=0.06118, over 3804341.37 frames. ], batch size: 55, lr: 3.25e-03, grad_scale: 8.0 +2023-04-03 05:46:13,349 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=170522.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:46:41,267 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=170544.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:46:58,413 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.718e+02 4.827e+02 5.978e+02 7.974e+02 2.215e+03, threshold=1.196e+03, percent-clipped=6.0 +2023-04-03 05:46:58,837 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3375, 1.3902, 1.8830, 1.4746, 2.7536, 3.7656, 3.4280, 3.9693], + device='cuda:0'), covar=tensor([0.1523, 0.3806, 0.3145, 0.2360, 0.0613, 0.0185, 0.0213, 0.0229], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0329, 0.0359, 0.0268, 0.0250, 0.0192, 0.0218, 0.0269], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 05:47:16,983 INFO [train.py:903] (0/4) Epoch 25, batch 6700, loss[loss=0.1656, simple_loss=0.248, pruned_loss=0.04161, over 19368.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2834, pruned_loss=0.06126, over 3809689.53 frames. ], batch size: 47, lr: 3.25e-03, grad_scale: 8.0 +2023-04-03 05:47:36,655 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=170587.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:48:05,686 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9655, 1.5905, 1.7613, 1.7729, 3.6208, 1.2358, 2.6490, 4.0771], + device='cuda:0'), covar=tensor([0.0469, 0.2643, 0.2680, 0.1814, 0.0655, 0.2498, 0.1203, 0.0190], + device='cuda:0'), in_proj_covar=tensor([0.0418, 0.0373, 0.0393, 0.0352, 0.0378, 0.0353, 0.0391, 0.0411], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:48:17,850 INFO [train.py:903] (0/4) Epoch 25, batch 6750, loss[loss=0.212, simple_loss=0.297, pruned_loss=0.06351, over 19677.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2834, pruned_loss=0.06111, over 3811343.49 frames. ], batch size: 60, lr: 3.25e-03, grad_scale: 8.0 +2023-04-03 05:48:52,178 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.55 vs. limit=5.0 +2023-04-03 05:48:59,809 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.047e+02 4.526e+02 5.801e+02 6.899e+02 1.670e+03, threshold=1.160e+03, percent-clipped=2.0 +2023-04-03 05:49:15,797 INFO [train.py:903] (0/4) Epoch 25, batch 6800, loss[loss=0.2312, simple_loss=0.315, pruned_loss=0.07368, over 19587.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2837, pruned_loss=0.06134, over 3815347.26 frames. ], batch size: 61, lr: 3.25e-03, grad_scale: 8.0 +2023-04-03 05:49:44,711 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7527, 2.1472, 1.6506, 1.6518, 2.0111, 1.6070, 1.6100, 2.0017], + device='cuda:0'), covar=tensor([0.0867, 0.0696, 0.0914, 0.0763, 0.0519, 0.1061, 0.0618, 0.0492], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0316, 0.0335, 0.0268, 0.0247, 0.0340, 0.0292, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:49:46,897 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-25.pt +2023-04-03 05:50:03,525 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-03 05:50:05,063 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-03 05:50:07,355 INFO [train.py:903] (0/4) Epoch 26, batch 0, loss[loss=0.2542, simple_loss=0.3225, pruned_loss=0.09296, over 19621.00 frames. ], tot_loss[loss=0.2542, simple_loss=0.3225, pruned_loss=0.09296, over 19621.00 frames. ], batch size: 57, lr: 3.19e-03, grad_scale: 8.0 +2023-04-03 05:50:07,356 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 05:50:19,307 INFO [train.py:937] (0/4) Epoch 26, validation: loss=0.1673, simple_loss=0.2675, pruned_loss=0.03355, over 944034.00 frames. +2023-04-03 05:50:19,308 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 05:50:26,507 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9716, 1.6888, 1.6415, 1.9757, 1.6237, 1.7820, 1.7194, 1.8442], + device='cuda:0'), covar=tensor([0.1082, 0.1522, 0.1524, 0.1045, 0.1413, 0.0554, 0.1408, 0.0770], + device='cuda:0'), in_proj_covar=tensor([0.0274, 0.0355, 0.0314, 0.0253, 0.0303, 0.0254, 0.0314, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:50:32,196 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-03 05:51:20,620 INFO [train.py:903] (0/4) Epoch 26, batch 50, loss[loss=0.1894, simple_loss=0.2709, pruned_loss=0.05397, over 19688.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2835, pruned_loss=0.0602, over 866569.54 frames. ], batch size: 53, lr: 3.19e-03, grad_scale: 8.0 +2023-04-03 05:51:30,191 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.374e+02 5.268e+02 6.216e+02 7.845e+02 1.668e+03, threshold=1.243e+03, percent-clipped=9.0 +2023-04-03 05:51:54,932 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=170778.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:51:55,735 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-03 05:52:22,097 INFO [train.py:903] (0/4) Epoch 26, batch 100, loss[loss=0.1901, simple_loss=0.2708, pruned_loss=0.05473, over 19679.00 frames. ], tot_loss[loss=0.2061, simple_loss=0.2869, pruned_loss=0.06267, over 1526461.16 frames. ], batch size: 53, lr: 3.18e-03, grad_scale: 4.0 +2023-04-03 05:52:26,016 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=170803.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:52:32,332 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-03 05:52:50,716 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=170823.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:53:12,612 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-03 05:53:24,743 INFO [train.py:903] (0/4) Epoch 26, batch 150, loss[loss=0.1922, simple_loss=0.2805, pruned_loss=0.0519, over 17363.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2849, pruned_loss=0.06164, over 2030344.78 frames. ], batch size: 101, lr: 3.18e-03, grad_scale: 4.0 +2023-04-03 05:53:36,325 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.111e+02 4.908e+02 6.470e+02 7.917e+02 1.560e+03, threshold=1.294e+03, percent-clipped=6.0 +2023-04-03 05:54:20,332 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5315, 2.3758, 2.1248, 2.6717, 2.2046, 2.2636, 2.0870, 2.4656], + device='cuda:0'), covar=tensor([0.0956, 0.1685, 0.1455, 0.0944, 0.1487, 0.0519, 0.1410, 0.0700], + device='cuda:0'), in_proj_covar=tensor([0.0272, 0.0352, 0.0312, 0.0252, 0.0302, 0.0253, 0.0312, 0.0259], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:54:25,500 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-03 05:54:26,680 INFO [train.py:903] (0/4) Epoch 26, batch 200, loss[loss=0.1952, simple_loss=0.2729, pruned_loss=0.05876, over 19728.00 frames. ], tot_loss[loss=0.2029, simple_loss=0.2841, pruned_loss=0.06082, over 2442011.62 frames. ], batch size: 51, lr: 3.18e-03, grad_scale: 4.0 +2023-04-03 05:55:05,036 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=170931.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:55:14,462 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=170938.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:55:23,820 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=170946.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:55:30,414 INFO [train.py:903] (0/4) Epoch 26, batch 250, loss[loss=0.1897, simple_loss=0.2807, pruned_loss=0.04934, over 19580.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2846, pruned_loss=0.06127, over 2742086.02 frames. ], batch size: 61, lr: 3.18e-03, grad_scale: 4.0 +2023-04-03 05:55:31,162 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.75 vs. limit=2.0 +2023-04-03 05:55:39,622 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8732, 1.3490, 1.0888, 0.9643, 1.1770, 1.0031, 1.0066, 1.2485], + device='cuda:0'), covar=tensor([0.0664, 0.0931, 0.1162, 0.0789, 0.0585, 0.1330, 0.0630, 0.0532], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0318, 0.0336, 0.0270, 0.0249, 0.0342, 0.0294, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 05:55:42,608 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.950e+02 4.848e+02 5.950e+02 8.014e+02 1.769e+03, threshold=1.190e+03, percent-clipped=1.0 +2023-04-03 05:56:35,081 INFO [train.py:903] (0/4) Epoch 26, batch 300, loss[loss=0.2025, simple_loss=0.2763, pruned_loss=0.0643, over 19475.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2845, pruned_loss=0.06127, over 2992199.18 frames. ], batch size: 49, lr: 3.18e-03, grad_scale: 4.0 +2023-04-03 05:57:34,219 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=171046.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:57:34,575 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.43 vs. limit=5.0 +2023-04-03 05:57:38,577 INFO [train.py:903] (0/4) Epoch 26, batch 350, loss[loss=0.1804, simple_loss=0.2553, pruned_loss=0.05275, over 19726.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2829, pruned_loss=0.06051, over 3178473.00 frames. ], batch size: 46, lr: 3.18e-03, grad_scale: 4.0 +2023-04-03 05:57:45,659 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 05:57:49,060 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.001e+02 4.764e+02 6.020e+02 7.720e+02 1.602e+03, threshold=1.204e+03, percent-clipped=4.0 +2023-04-03 05:58:42,177 INFO [train.py:903] (0/4) Epoch 26, batch 400, loss[loss=0.2219, simple_loss=0.303, pruned_loss=0.07041, over 19512.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2834, pruned_loss=0.06051, over 3325888.12 frames. ], batch size: 64, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 05:58:55,361 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=171110.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 05:59:43,982 INFO [train.py:903] (0/4) Epoch 26, batch 450, loss[loss=0.2556, simple_loss=0.3227, pruned_loss=0.09423, over 13941.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2837, pruned_loss=0.06063, over 3424445.37 frames. ], batch size: 136, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 05:59:56,339 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.853e+02 5.217e+02 6.577e+02 9.059e+02 2.566e+03, threshold=1.315e+03, percent-clipped=7.0 +2023-04-03 06:00:19,587 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-03 06:00:20,832 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-03 06:00:30,441 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2109, 2.0690, 2.0266, 2.2860, 2.1056, 1.9138, 1.9597, 2.1637], + device='cuda:0'), covar=tensor([0.0829, 0.1173, 0.1079, 0.0757, 0.1021, 0.0514, 0.1137, 0.0585], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0354, 0.0314, 0.0254, 0.0304, 0.0255, 0.0315, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:00:38,881 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=171194.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:00:47,846 INFO [train.py:903] (0/4) Epoch 26, batch 500, loss[loss=0.1926, simple_loss=0.2724, pruned_loss=0.05646, over 19578.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2852, pruned_loss=0.06129, over 3520304.95 frames. ], batch size: 52, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:01:05,573 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=171214.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:01:11,740 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=171219.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:01:52,229 INFO [train.py:903] (0/4) Epoch 26, batch 550, loss[loss=0.2732, simple_loss=0.3377, pruned_loss=0.1044, over 19628.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2844, pruned_loss=0.06087, over 3591868.68 frames. ], batch size: 50, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:02:03,092 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.195e+02 4.798e+02 6.471e+02 7.842e+02 1.459e+03, threshold=1.294e+03, percent-clipped=3.0 +2023-04-03 06:02:12,067 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-03 06:02:44,067 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=171290.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:02:55,673 INFO [train.py:903] (0/4) Epoch 26, batch 600, loss[loss=0.1983, simple_loss=0.2866, pruned_loss=0.05496, over 19559.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2858, pruned_loss=0.06176, over 3620306.95 frames. ], batch size: 56, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:02:58,477 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=171302.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:03:31,084 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=171327.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:03:36,397 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-03 06:03:57,549 INFO [train.py:903] (0/4) Epoch 26, batch 650, loss[loss=0.1884, simple_loss=0.264, pruned_loss=0.05641, over 19407.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2862, pruned_loss=0.06214, over 3673902.79 frames. ], batch size: 48, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:04:09,344 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.689e+02 4.759e+02 5.860e+02 7.918e+02 1.260e+03, threshold=1.172e+03, percent-clipped=0.0 +2023-04-03 06:04:15,214 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=171363.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:05:01,840 INFO [train.py:903] (0/4) Epoch 26, batch 700, loss[loss=0.1709, simple_loss=0.2526, pruned_loss=0.04462, over 19750.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2855, pruned_loss=0.06185, over 3712118.69 frames. ], batch size: 47, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:05:09,250 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=171405.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:06:08,106 INFO [train.py:903] (0/4) Epoch 26, batch 750, loss[loss=0.194, simple_loss=0.2722, pruned_loss=0.05791, over 19608.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2851, pruned_loss=0.0614, over 3742022.81 frames. ], batch size: 50, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:06:11,857 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4406, 1.4227, 1.5983, 1.5182, 3.0402, 1.2486, 2.2806, 3.4336], + device='cuda:0'), covar=tensor([0.0522, 0.2639, 0.2741, 0.1858, 0.0657, 0.2343, 0.1325, 0.0272], + device='cuda:0'), in_proj_covar=tensor([0.0420, 0.0374, 0.0394, 0.0354, 0.0379, 0.0354, 0.0392, 0.0413], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:06:12,967 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=171454.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:06:18,656 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.178e+02 5.123e+02 6.550e+02 8.686e+02 2.549e+03, threshold=1.310e+03, percent-clipped=11.0 +2023-04-03 06:07:12,531 INFO [train.py:903] (0/4) Epoch 26, batch 800, loss[loss=0.1916, simple_loss=0.2834, pruned_loss=0.04987, over 19348.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2851, pruned_loss=0.06118, over 3773414.92 frames. ], batch size: 66, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:07:25,369 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-03 06:07:32,306 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=171516.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:08:14,608 INFO [train.py:903] (0/4) Epoch 26, batch 850, loss[loss=0.2112, simple_loss=0.2957, pruned_loss=0.06341, over 19606.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.285, pruned_loss=0.06132, over 3780689.50 frames. ], batch size: 57, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:08:24,827 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=171558.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:08:25,749 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.196e+02 4.689e+02 5.699e+02 7.261e+02 1.636e+03, threshold=1.140e+03, percent-clipped=2.0 +2023-04-03 06:08:40,937 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=171569.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:09:05,302 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-03 06:09:18,365 INFO [train.py:903] (0/4) Epoch 26, batch 900, loss[loss=0.2036, simple_loss=0.2912, pruned_loss=0.05802, over 19595.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2848, pruned_loss=0.0612, over 3788190.27 frames. ], batch size: 61, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:10:17,059 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.86 vs. limit=5.0 +2023-04-03 06:10:22,197 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-03 06:10:23,374 INFO [train.py:903] (0/4) Epoch 26, batch 950, loss[loss=0.1766, simple_loss=0.2646, pruned_loss=0.04425, over 19620.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2836, pruned_loss=0.06041, over 3786199.32 frames. ], batch size: 57, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:10:34,905 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.376e+02 5.213e+02 6.211e+02 8.451e+02 1.981e+03, threshold=1.242e+03, percent-clipped=10.0 +2023-04-03 06:10:37,562 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=171661.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:10:45,884 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1665, 1.7856, 1.4730, 1.0942, 1.6512, 1.1228, 1.1412, 1.6595], + device='cuda:0'), covar=tensor([0.0770, 0.0738, 0.0918, 0.0993, 0.0504, 0.1325, 0.0644, 0.0385], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0318, 0.0337, 0.0270, 0.0250, 0.0343, 0.0295, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:10:51,827 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=171673.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:11:00,243 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8103, 1.8983, 1.4486, 1.8527, 1.7661, 1.4663, 1.4208, 1.7017], + device='cuda:0'), covar=tensor([0.1250, 0.1462, 0.1866, 0.1224, 0.1484, 0.0971, 0.1956, 0.1029], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0354, 0.0314, 0.0254, 0.0305, 0.0254, 0.0315, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:11:09,420 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=171686.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:11:27,636 INFO [train.py:903] (0/4) Epoch 26, batch 1000, loss[loss=0.1859, simple_loss=0.2655, pruned_loss=0.05319, over 19712.00 frames. ], tot_loss[loss=0.203, simple_loss=0.284, pruned_loss=0.06105, over 3788118.15 frames. ], batch size: 45, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:11:36,038 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=171707.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:12:19,822 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-03 06:12:31,841 INFO [train.py:903] (0/4) Epoch 26, batch 1050, loss[loss=0.2353, simple_loss=0.3134, pruned_loss=0.07857, over 19352.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.284, pruned_loss=0.06064, over 3803074.60 frames. ], batch size: 70, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:12:42,470 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.548e+02 5.465e+02 6.383e+02 7.807e+02 1.569e+03, threshold=1.277e+03, percent-clipped=7.0 +2023-04-03 06:12:50,005 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.56 vs. limit=2.0 +2023-04-03 06:13:01,865 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-03 06:13:34,571 INFO [train.py:903] (0/4) Epoch 26, batch 1100, loss[loss=0.1957, simple_loss=0.2854, pruned_loss=0.05306, over 19721.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2847, pruned_loss=0.0611, over 3799795.29 frames. ], batch size: 63, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:14:05,400 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=171822.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:14:09,102 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=171825.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:14:40,976 INFO [train.py:903] (0/4) Epoch 26, batch 1150, loss[loss=0.2463, simple_loss=0.315, pruned_loss=0.08874, over 13135.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2847, pruned_loss=0.06108, over 3797484.86 frames. ], batch size: 136, lr: 3.18e-03, grad_scale: 8.0 +2023-04-03 06:14:41,359 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=171850.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:14:54,267 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.309e+02 5.298e+02 6.784e+02 8.276e+02 1.649e+03, threshold=1.357e+03, percent-clipped=5.0 +2023-04-03 06:14:55,493 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=171860.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:15:46,135 INFO [train.py:903] (0/4) Epoch 26, batch 1200, loss[loss=0.2111, simple_loss=0.2974, pruned_loss=0.06239, over 19552.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2854, pruned_loss=0.06165, over 3794713.48 frames. ], batch size: 56, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:16:12,768 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-03 06:16:21,033 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=171929.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:16:29,733 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-04-03 06:16:48,474 INFO [train.py:903] (0/4) Epoch 26, batch 1250, loss[loss=0.2343, simple_loss=0.2931, pruned_loss=0.08772, over 19307.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2862, pruned_loss=0.06238, over 3799316.87 frames. ], batch size: 44, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:16:53,692 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=171954.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:16:54,754 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=171955.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:16:58,907 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.249e+02 5.182e+02 6.196e+02 7.754e+02 1.405e+03, threshold=1.239e+03, percent-clipped=1.0 +2023-04-03 06:17:21,563 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=171975.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:17:51,870 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-172000.pt +2023-04-03 06:17:52,780 INFO [train.py:903] (0/4) Epoch 26, batch 1300, loss[loss=0.1638, simple_loss=0.2473, pruned_loss=0.04016, over 19472.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2856, pruned_loss=0.06206, over 3823277.58 frames. ], batch size: 49, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:18:56,742 INFO [train.py:903] (0/4) Epoch 26, batch 1350, loss[loss=0.1929, simple_loss=0.2717, pruned_loss=0.05708, over 19758.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2852, pruned_loss=0.06178, over 3826404.77 frames. ], batch size: 51, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:18:59,475 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0056, 4.5474, 2.7374, 3.9386, 0.9408, 4.5523, 4.3629, 4.4705], + device='cuda:0'), covar=tensor([0.0506, 0.0914, 0.2007, 0.0857, 0.4047, 0.0615, 0.0939, 0.1080], + device='cuda:0'), in_proj_covar=tensor([0.0521, 0.0423, 0.0510, 0.0357, 0.0409, 0.0449, 0.0444, 0.0475], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:19:09,298 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.61 vs. limit=2.0 +2023-04-03 06:19:09,474 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.423e+02 5.487e+02 6.725e+02 8.152e+02 1.378e+03, threshold=1.345e+03, percent-clipped=4.0 +2023-04-03 06:19:16,131 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.52 vs. limit=2.0 +2023-04-03 06:19:25,563 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8692, 1.5981, 1.4810, 1.7816, 1.5337, 1.5361, 1.4373, 1.7096], + device='cuda:0'), covar=tensor([0.1099, 0.1332, 0.1553, 0.1129, 0.1358, 0.0629, 0.1562, 0.0828], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0357, 0.0318, 0.0256, 0.0308, 0.0256, 0.0318, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:19:33,987 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=172078.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:20:01,390 INFO [train.py:903] (0/4) Epoch 26, batch 1400, loss[loss=0.1914, simple_loss=0.2713, pruned_loss=0.05573, over 19842.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2832, pruned_loss=0.06104, over 3831512.67 frames. ], batch size: 52, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:20:07,036 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=172103.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:20:47,879 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=172136.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:21:05,095 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-03 06:21:06,114 INFO [train.py:903] (0/4) Epoch 26, batch 1450, loss[loss=0.1795, simple_loss=0.2494, pruned_loss=0.05477, over 19779.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2832, pruned_loss=0.06102, over 3831648.75 frames. ], batch size: 48, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:21:16,561 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.245e+02 4.778e+02 6.066e+02 7.310e+02 2.231e+03, threshold=1.213e+03, percent-clipped=2.0 +2023-04-03 06:21:22,788 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4561, 1.6184, 2.0573, 1.7553, 3.1932, 2.5971, 3.4986, 1.5955], + device='cuda:0'), covar=tensor([0.2768, 0.4638, 0.2919, 0.2084, 0.1621, 0.2228, 0.1562, 0.4612], + device='cuda:0'), in_proj_covar=tensor([0.0548, 0.0662, 0.0740, 0.0500, 0.0630, 0.0541, 0.0667, 0.0565], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 06:22:10,096 INFO [train.py:903] (0/4) Epoch 26, batch 1500, loss[loss=0.1799, simple_loss=0.2749, pruned_loss=0.04249, over 19098.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2826, pruned_loss=0.06052, over 3832409.38 frames. ], batch size: 69, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:22:33,565 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.9175, 5.3545, 3.0609, 4.6902, 0.9698, 5.5706, 5.3386, 5.5680], + device='cuda:0'), covar=tensor([0.0368, 0.0818, 0.1926, 0.0726, 0.4268, 0.0506, 0.0775, 0.0955], + device='cuda:0'), in_proj_covar=tensor([0.0522, 0.0425, 0.0513, 0.0358, 0.0410, 0.0452, 0.0446, 0.0476], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:22:51,379 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=172231.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:23:14,777 INFO [train.py:903] (0/4) Epoch 26, batch 1550, loss[loss=0.1696, simple_loss=0.2508, pruned_loss=0.04419, over 19774.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2826, pruned_loss=0.06016, over 3838230.43 frames. ], batch size: 45, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:23:23,243 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=172256.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:23:26,529 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.787e+02 4.585e+02 5.725e+02 7.044e+02 1.122e+03, threshold=1.145e+03, percent-clipped=0.0 +2023-04-03 06:23:59,061 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3516, 1.7186, 1.8754, 1.9876, 2.9114, 1.7255, 2.8300, 3.2646], + device='cuda:0'), covar=tensor([0.0728, 0.2941, 0.2809, 0.1864, 0.0932, 0.2341, 0.1967, 0.0494], + device='cuda:0'), in_proj_covar=tensor([0.0420, 0.0374, 0.0395, 0.0354, 0.0380, 0.0354, 0.0393, 0.0414], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:24:17,316 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=172299.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:24:18,407 INFO [train.py:903] (0/4) Epoch 26, batch 1600, loss[loss=0.1816, simple_loss=0.2585, pruned_loss=0.0524, over 19786.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2825, pruned_loss=0.06058, over 3822916.52 frames. ], batch size: 48, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:24:30,668 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-03 06:24:44,999 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-03 06:24:49,165 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5902, 2.3422, 1.7657, 1.6156, 2.1396, 1.4649, 1.4204, 2.0910], + device='cuda:0'), covar=tensor([0.1232, 0.1013, 0.1133, 0.0996, 0.0659, 0.1355, 0.0846, 0.0540], + device='cuda:0'), in_proj_covar=tensor([0.0307, 0.0320, 0.0340, 0.0272, 0.0251, 0.0345, 0.0294, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:25:22,890 INFO [train.py:903] (0/4) Epoch 26, batch 1650, loss[loss=0.2429, simple_loss=0.324, pruned_loss=0.08089, over 18199.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2826, pruned_loss=0.06038, over 3824868.83 frames. ], batch size: 83, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:25:32,972 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.237e+02 4.806e+02 6.116e+02 7.815e+02 1.931e+03, threshold=1.223e+03, percent-clipped=6.0 +2023-04-03 06:26:25,096 INFO [train.py:903] (0/4) Epoch 26, batch 1700, loss[loss=0.1943, simple_loss=0.2723, pruned_loss=0.05811, over 19844.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2823, pruned_loss=0.06032, over 3828912.58 frames. ], batch size: 52, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:26:27,915 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-03 06:26:43,100 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=172414.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:27:07,063 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-03 06:27:27,217 INFO [train.py:903] (0/4) Epoch 26, batch 1750, loss[loss=0.1836, simple_loss=0.255, pruned_loss=0.05608, over 19311.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2819, pruned_loss=0.06034, over 3810329.84 frames. ], batch size: 44, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:27:39,761 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.268e+02 4.883e+02 5.717e+02 7.034e+02 1.807e+03, threshold=1.143e+03, percent-clipped=3.0 +2023-04-03 06:27:52,717 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0860, 2.2393, 1.6013, 2.2292, 2.2002, 1.6428, 1.6345, 2.0690], + device='cuda:0'), covar=tensor([0.1232, 0.1677, 0.1993, 0.1268, 0.1490, 0.1070, 0.2018, 0.1056], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0356, 0.0315, 0.0254, 0.0307, 0.0254, 0.0316, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:28:06,463 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=172480.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:28:09,152 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8884, 1.2586, 1.6037, 0.5341, 1.9081, 2.4817, 2.1510, 2.6252], + device='cuda:0'), covar=tensor([0.1654, 0.3832, 0.3393, 0.2917, 0.0671, 0.0289, 0.0350, 0.0381], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0330, 0.0361, 0.0270, 0.0252, 0.0193, 0.0219, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 06:28:25,522 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8009, 4.3444, 2.6829, 3.8460, 0.8274, 4.3597, 4.2289, 4.2984], + device='cuda:0'), covar=tensor([0.0547, 0.0945, 0.2015, 0.0842, 0.4229, 0.0619, 0.0862, 0.1201], + device='cuda:0'), in_proj_covar=tensor([0.0518, 0.0420, 0.0507, 0.0355, 0.0406, 0.0447, 0.0441, 0.0472], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:28:32,022 INFO [train.py:903] (0/4) Epoch 26, batch 1800, loss[loss=0.2156, simple_loss=0.2941, pruned_loss=0.06855, over 19656.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2823, pruned_loss=0.06079, over 3793200.24 frames. ], batch size: 55, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:29:31,751 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-03 06:29:36,210 INFO [train.py:903] (0/4) Epoch 26, batch 1850, loss[loss=0.2331, simple_loss=0.3096, pruned_loss=0.07831, over 19479.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.283, pruned_loss=0.06107, over 3783801.99 frames. ], batch size: 49, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:29:46,974 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.860e+02 4.880e+02 5.794e+02 7.257e+02 1.575e+03, threshold=1.159e+03, percent-clipped=2.0 +2023-04-03 06:30:11,356 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-03 06:30:20,681 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.2565, 3.9144, 2.6635, 3.5102, 1.1738, 3.8256, 3.7497, 3.7950], + device='cuda:0'), covar=tensor([0.0761, 0.0960, 0.1918, 0.0844, 0.3534, 0.0693, 0.0952, 0.1262], + device='cuda:0'), in_proj_covar=tensor([0.0521, 0.0423, 0.0511, 0.0357, 0.0407, 0.0450, 0.0443, 0.0475], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:30:34,098 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=172595.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:30:39,960 INFO [train.py:903] (0/4) Epoch 26, batch 1900, loss[loss=0.1943, simple_loss=0.2762, pruned_loss=0.05626, over 19840.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2832, pruned_loss=0.0609, over 3788627.18 frames. ], batch size: 52, lr: 3.17e-03, grad_scale: 4.0 +2023-04-03 06:30:59,214 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-03 06:31:01,764 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=172616.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:31:05,015 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-03 06:31:23,049 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0675, 5.1353, 5.9335, 5.9558, 2.1511, 5.5489, 4.6808, 5.5858], + device='cuda:0'), covar=tensor([0.1958, 0.0860, 0.0640, 0.0637, 0.6298, 0.0814, 0.0682, 0.1375], + device='cuda:0'), in_proj_covar=tensor([0.0806, 0.0774, 0.0978, 0.0859, 0.0852, 0.0745, 0.0584, 0.0906], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 06:31:30,003 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-03 06:31:36,059 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=172643.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:31:44,207 INFO [train.py:903] (0/4) Epoch 26, batch 1950, loss[loss=0.1849, simple_loss=0.2546, pruned_loss=0.05756, over 19725.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2834, pruned_loss=0.06096, over 3793192.02 frames. ], batch size: 46, lr: 3.17e-03, grad_scale: 4.0 +2023-04-03 06:31:57,689 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.017e+02 5.105e+02 6.247e+02 7.487e+02 1.872e+03, threshold=1.249e+03, percent-clipped=7.0 +2023-04-03 06:32:12,279 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=172670.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:32:42,538 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=172695.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:32:48,412 INFO [train.py:903] (0/4) Epoch 26, batch 2000, loss[loss=0.2291, simple_loss=0.3081, pruned_loss=0.075, over 19338.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2846, pruned_loss=0.06164, over 3792448.52 frames. ], batch size: 70, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:33:14,447 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 06:33:38,926 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9381, 4.2576, 4.6157, 4.6133, 2.1878, 4.3117, 3.7656, 4.3524], + device='cuda:0'), covar=tensor([0.1618, 0.1436, 0.0582, 0.0670, 0.5507, 0.1018, 0.0669, 0.1009], + device='cuda:0'), in_proj_covar=tensor([0.0803, 0.0774, 0.0978, 0.0859, 0.0852, 0.0744, 0.0584, 0.0905], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 06:33:46,789 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-03 06:33:51,654 INFO [train.py:903] (0/4) Epoch 26, batch 2050, loss[loss=0.2097, simple_loss=0.2911, pruned_loss=0.06416, over 19777.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2852, pruned_loss=0.06182, over 3793961.39 frames. ], batch size: 56, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:33:55,236 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=172752.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:34:03,941 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.396e+02 5.217e+02 6.186e+02 8.512e+02 2.102e+03, threshold=1.237e+03, percent-clipped=6.0 +2023-04-03 06:34:06,457 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-03 06:34:07,786 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-03 06:34:27,745 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-03 06:34:54,383 INFO [train.py:903] (0/4) Epoch 26, batch 2100, loss[loss=0.1796, simple_loss=0.2662, pruned_loss=0.04652, over 19466.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2848, pruned_loss=0.06149, over 3807657.56 frames. ], batch size: 49, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:35:25,181 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-03 06:35:47,342 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-03 06:35:56,681 INFO [train.py:903] (0/4) Epoch 26, batch 2150, loss[loss=0.2015, simple_loss=0.2755, pruned_loss=0.06376, over 19408.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2853, pruned_loss=0.06148, over 3825388.91 frames. ], batch size: 48, lr: 3.17e-03, grad_scale: 8.0 +2023-04-03 06:35:58,271 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=172851.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:36:02,873 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0468, 1.7085, 1.9991, 2.1201, 4.5543, 1.4901, 2.7407, 4.9841], + device='cuda:0'), covar=tensor([0.0484, 0.2791, 0.2681, 0.1726, 0.0748, 0.2264, 0.1320, 0.0163], + device='cuda:0'), in_proj_covar=tensor([0.0417, 0.0371, 0.0392, 0.0351, 0.0378, 0.0352, 0.0389, 0.0410], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:36:10,068 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.604e+02 4.749e+02 6.121e+02 8.086e+02 1.727e+03, threshold=1.224e+03, percent-clipped=6.0 +2023-04-03 06:36:14,769 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=172863.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:36:21,373 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-04-03 06:36:31,091 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=172876.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:36:59,980 INFO [train.py:903] (0/4) Epoch 26, batch 2200, loss[loss=0.2113, simple_loss=0.2941, pruned_loss=0.06424, over 19682.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2846, pruned_loss=0.06118, over 3826407.10 frames. ], batch size: 53, lr: 3.17e-03, grad_scale: 4.0 +2023-04-03 06:38:04,010 INFO [train.py:903] (0/4) Epoch 26, batch 2250, loss[loss=0.1608, simple_loss=0.2384, pruned_loss=0.04157, over 19770.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2835, pruned_loss=0.06074, over 3830495.35 frames. ], batch size: 47, lr: 3.16e-03, grad_scale: 4.0 +2023-04-03 06:38:16,968 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=172960.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:38:17,986 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.182e+02 4.806e+02 5.924e+02 7.729e+02 1.368e+03, threshold=1.185e+03, percent-clipped=2.0 +2023-04-03 06:38:51,867 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=172987.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:39:07,993 INFO [train.py:903] (0/4) Epoch 26, batch 2300, loss[loss=0.2318, simple_loss=0.3081, pruned_loss=0.07779, over 13368.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2834, pruned_loss=0.06063, over 3817052.54 frames. ], batch size: 135, lr: 3.16e-03, grad_scale: 4.0 +2023-04-03 06:39:13,393 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-03 06:39:19,730 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-03 06:40:11,060 INFO [train.py:903] (0/4) Epoch 26, batch 2350, loss[loss=0.1783, simple_loss=0.2694, pruned_loss=0.04365, over 19446.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2842, pruned_loss=0.06087, over 3819638.58 frames. ], batch size: 70, lr: 3.16e-03, grad_scale: 4.0 +2023-04-03 06:40:25,948 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.095e+02 5.175e+02 6.518e+02 7.971e+02 2.695e+03, threshold=1.304e+03, percent-clipped=8.0 +2023-04-03 06:40:43,725 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=173075.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:40:54,421 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-03 06:41:08,275 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=173096.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:41:10,528 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-03 06:41:14,038 INFO [train.py:903] (0/4) Epoch 26, batch 2400, loss[loss=0.2787, simple_loss=0.3407, pruned_loss=0.1084, over 13163.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2843, pruned_loss=0.06114, over 3818620.61 frames. ], batch size: 135, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:41:18,171 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=173102.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:41:48,367 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0805, 1.2565, 1.7846, 1.1335, 2.5265, 3.5207, 3.2130, 3.7183], + device='cuda:0'), covar=tensor([0.1749, 0.4068, 0.3370, 0.2718, 0.0703, 0.0193, 0.0222, 0.0283], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0330, 0.0362, 0.0271, 0.0252, 0.0193, 0.0220, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 06:42:19,069 INFO [train.py:903] (0/4) Epoch 26, batch 2450, loss[loss=0.2131, simple_loss=0.2904, pruned_loss=0.06783, over 19493.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2859, pruned_loss=0.06213, over 3783832.29 frames. ], batch size: 49, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:42:32,913 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.093e+02 4.972e+02 5.973e+02 7.732e+02 1.743e+03, threshold=1.195e+03, percent-clipped=3.0 +2023-04-03 06:42:41,751 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=173168.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:43:22,795 INFO [train.py:903] (0/4) Epoch 26, batch 2500, loss[loss=0.1855, simple_loss=0.275, pruned_loss=0.04801, over 19770.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2855, pruned_loss=0.06179, over 3804505.57 frames. ], batch size: 54, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:43:24,460 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2998, 2.3948, 2.6871, 2.9750, 2.2972, 2.8274, 2.7199, 2.5020], + device='cuda:0'), covar=tensor([0.4281, 0.4200, 0.1838, 0.2735, 0.4656, 0.2368, 0.4630, 0.3229], + device='cuda:0'), in_proj_covar=tensor([0.0927, 0.1004, 0.0736, 0.0948, 0.0906, 0.0842, 0.0859, 0.0804], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 06:43:31,263 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=173207.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:43:36,090 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=173211.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:44:25,914 INFO [train.py:903] (0/4) Epoch 26, batch 2550, loss[loss=0.1654, simple_loss=0.2472, pruned_loss=0.04179, over 19370.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2845, pruned_loss=0.06114, over 3808341.24 frames. ], batch size: 47, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:44:40,216 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1777, 3.2396, 1.8617, 1.9231, 2.9452, 1.5593, 1.6155, 2.3594], + device='cuda:0'), covar=tensor([0.1302, 0.0712, 0.1150, 0.0857, 0.0578, 0.1346, 0.0947, 0.0706], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0317, 0.0337, 0.0269, 0.0248, 0.0341, 0.0291, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:44:40,985 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.733e+02 4.982e+02 5.984e+02 8.594e+02 2.255e+03, threshold=1.197e+03, percent-clipped=6.0 +2023-04-03 06:45:23,044 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-03 06:45:30,051 INFO [train.py:903] (0/4) Epoch 26, batch 2600, loss[loss=0.183, simple_loss=0.2735, pruned_loss=0.04622, over 19716.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2842, pruned_loss=0.06053, over 3823907.31 frames. ], batch size: 59, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:45:59,668 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=173322.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:46:05,380 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=173327.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:46:05,477 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4933, 2.1410, 2.1825, 3.1473, 2.0699, 2.5951, 2.5021, 2.3687], + device='cuda:0'), covar=tensor([0.0762, 0.0928, 0.0915, 0.0731, 0.0924, 0.0762, 0.0921, 0.0665], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0226, 0.0228, 0.0242, 0.0228, 0.0214, 0.0190, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 06:46:10,051 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=173331.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:46:35,579 INFO [train.py:903] (0/4) Epoch 26, batch 2650, loss[loss=0.178, simple_loss=0.2587, pruned_loss=0.04864, over 19323.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2836, pruned_loss=0.06014, over 3829898.76 frames. ], batch size: 44, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:46:41,857 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3444, 2.0488, 1.6264, 1.3901, 1.8612, 1.3058, 1.2900, 1.8217], + device='cuda:0'), covar=tensor([0.0951, 0.0787, 0.1134, 0.0864, 0.0606, 0.1309, 0.0728, 0.0492], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0316, 0.0336, 0.0269, 0.0248, 0.0340, 0.0290, 0.0272], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:46:43,033 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=173356.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:46:45,484 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=173358.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:46:49,596 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.269e+02 5.033e+02 5.962e+02 7.363e+02 1.395e+03, threshold=1.192e+03, percent-clipped=2.0 +2023-04-03 06:46:55,577 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-03 06:47:17,766 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=173383.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:47:39,266 INFO [train.py:903] (0/4) Epoch 26, batch 2700, loss[loss=0.1657, simple_loss=0.2464, pruned_loss=0.04248, over 16533.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.284, pruned_loss=0.06058, over 3811753.82 frames. ], batch size: 36, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:47:59,336 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8004, 3.3080, 3.3364, 3.3552, 1.3878, 3.2157, 2.8055, 3.1206], + device='cuda:0'), covar=tensor([0.1805, 0.1037, 0.0834, 0.0969, 0.5636, 0.1088, 0.0873, 0.1302], + device='cuda:0'), in_proj_covar=tensor([0.0806, 0.0773, 0.0978, 0.0861, 0.0856, 0.0742, 0.0584, 0.0909], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 06:48:32,626 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7009, 1.7881, 2.0595, 1.9526, 1.4954, 1.9373, 2.0357, 1.9206], + device='cuda:0'), covar=tensor([0.4291, 0.3784, 0.2012, 0.2459, 0.3961, 0.2232, 0.5213, 0.3408], + device='cuda:0'), in_proj_covar=tensor([0.0928, 0.1004, 0.0736, 0.0949, 0.0908, 0.0841, 0.0859, 0.0805], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 06:48:39,712 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.46 vs. limit=2.0 +2023-04-03 06:48:41,207 INFO [train.py:903] (0/4) Epoch 26, batch 2750, loss[loss=0.2034, simple_loss=0.2902, pruned_loss=0.05823, over 19533.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2841, pruned_loss=0.06048, over 3828009.63 frames. ], batch size: 54, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:48:54,808 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.847e+02 4.981e+02 6.163e+02 7.639e+02 1.916e+03, threshold=1.233e+03, percent-clipped=5.0 +2023-04-03 06:49:03,504 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=173467.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:49:34,150 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=173492.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:49:44,183 INFO [train.py:903] (0/4) Epoch 26, batch 2800, loss[loss=0.2183, simple_loss=0.2999, pruned_loss=0.06837, over 18699.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2849, pruned_loss=0.0611, over 3833711.85 frames. ], batch size: 74, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:49:52,943 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4951, 1.6373, 1.6581, 1.8993, 1.5498, 1.8641, 1.7636, 1.5315], + device='cuda:0'), covar=tensor([0.4719, 0.4207, 0.2865, 0.2894, 0.4211, 0.2573, 0.6395, 0.5013], + device='cuda:0'), in_proj_covar=tensor([0.0930, 0.1005, 0.0738, 0.0949, 0.0910, 0.0842, 0.0861, 0.0806], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 06:49:56,406 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=173509.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 06:49:56,869 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.70 vs. limit=5.0 +2023-04-03 06:50:00,276 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=173512.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:50:48,004 INFO [train.py:903] (0/4) Epoch 26, batch 2850, loss[loss=0.2046, simple_loss=0.2789, pruned_loss=0.06516, over 19795.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2855, pruned_loss=0.06179, over 3836502.10 frames. ], batch size: 48, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:51:01,769 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.227e+02 4.954e+02 6.202e+02 8.183e+02 1.932e+03, threshold=1.240e+03, percent-clipped=7.0 +2023-04-03 06:51:23,167 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=173578.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:51:51,086 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-03 06:51:52,292 INFO [train.py:903] (0/4) Epoch 26, batch 2900, loss[loss=0.2395, simple_loss=0.308, pruned_loss=0.08552, over 17427.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2844, pruned_loss=0.06133, over 3831761.64 frames. ], batch size: 101, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:51:57,182 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=173603.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:52:07,978 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0174, 2.0832, 2.4700, 2.5420, 1.9255, 2.5129, 2.4856, 2.2704], + device='cuda:0'), covar=tensor([0.4237, 0.4014, 0.1832, 0.2425, 0.4236, 0.2218, 0.4671, 0.3238], + device='cuda:0'), in_proj_covar=tensor([0.0928, 0.1003, 0.0736, 0.0946, 0.0907, 0.0840, 0.0859, 0.0805], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 06:52:27,641 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=173627.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:52:33,440 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=173632.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:52:56,764 INFO [train.py:903] (0/4) Epoch 26, batch 2950, loss[loss=0.2263, simple_loss=0.3109, pruned_loss=0.07084, over 18165.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2846, pruned_loss=0.06181, over 3820487.75 frames. ], batch size: 83, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:53:10,748 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.057e+02 5.287e+02 6.719e+02 8.706e+02 2.181e+03, threshold=1.344e+03, percent-clipped=5.0 +2023-04-03 06:53:23,890 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=173671.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:53:59,613 INFO [train.py:903] (0/4) Epoch 26, batch 3000, loss[loss=0.2191, simple_loss=0.2992, pruned_loss=0.06949, over 19651.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2855, pruned_loss=0.06193, over 3819555.64 frames. ], batch size: 58, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:53:59,614 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 06:54:12,259 INFO [train.py:937] (0/4) Epoch 26, validation: loss=0.1681, simple_loss=0.2675, pruned_loss=0.03435, over 944034.00 frames. +2023-04-03 06:54:12,260 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 06:54:17,263 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-03 06:55:16,504 INFO [train.py:903] (0/4) Epoch 26, batch 3050, loss[loss=0.1838, simple_loss=0.2717, pruned_loss=0.04802, over 19662.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2849, pruned_loss=0.06119, over 3825213.83 frames. ], batch size: 53, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:55:25,140 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0530, 3.6938, 2.5373, 3.3300, 1.0685, 3.7332, 3.5615, 3.6013], + device='cuda:0'), covar=tensor([0.0719, 0.1141, 0.2018, 0.0978, 0.3699, 0.0711, 0.0887, 0.1116], + device='cuda:0'), in_proj_covar=tensor([0.0522, 0.0424, 0.0510, 0.0357, 0.0408, 0.0450, 0.0447, 0.0473], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 06:55:30,856 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.198e+02 4.825e+02 5.950e+02 7.456e+02 1.374e+03, threshold=1.190e+03, percent-clipped=1.0 +2023-04-03 06:56:02,523 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=173786.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:56:20,180 INFO [train.py:903] (0/4) Epoch 26, batch 3100, loss[loss=0.1819, simple_loss=0.2732, pruned_loss=0.04536, over 19668.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2853, pruned_loss=0.06204, over 3787015.55 frames. ], batch size: 55, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:57:23,189 INFO [train.py:903] (0/4) Epoch 26, batch 3150, loss[loss=0.2427, simple_loss=0.3139, pruned_loss=0.08573, over 13625.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.285, pruned_loss=0.06159, over 3796026.91 frames. ], batch size: 136, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:57:26,868 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=173853.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 06:57:37,129 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.975e+02 5.081e+02 6.233e+02 7.406e+02 2.417e+03, threshold=1.247e+03, percent-clipped=3.0 +2023-04-03 06:57:49,125 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-03 06:58:05,121 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=173883.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:58:25,323 INFO [train.py:903] (0/4) Epoch 26, batch 3200, loss[loss=0.2376, simple_loss=0.3138, pruned_loss=0.08074, over 19632.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2848, pruned_loss=0.06164, over 3811005.69 frames. ], batch size: 58, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:58:35,857 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=173908.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 06:59:25,875 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8819, 3.9783, 4.4169, 4.4312, 2.6718, 4.1149, 3.8251, 4.1813], + device='cuda:0'), covar=tensor([0.1276, 0.3443, 0.0609, 0.0673, 0.4476, 0.1249, 0.0591, 0.1031], + device='cuda:0'), in_proj_covar=tensor([0.0804, 0.0771, 0.0974, 0.0856, 0.0851, 0.0742, 0.0579, 0.0902], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 06:59:27,979 INFO [train.py:903] (0/4) Epoch 26, batch 3250, loss[loss=0.1953, simple_loss=0.279, pruned_loss=0.05578, over 19659.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.2859, pruned_loss=0.06271, over 3809048.66 frames. ], batch size: 53, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 06:59:42,864 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.716e+02 4.798e+02 6.185e+02 8.155e+02 2.789e+03, threshold=1.237e+03, percent-clipped=7.0 +2023-04-03 06:59:52,460 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=173968.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 06:59:55,195 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.42 vs. limit=5.0 +2023-04-03 07:00:01,772 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=173976.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:00:31,435 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-174000.pt +2023-04-03 07:00:32,335 INFO [train.py:903] (0/4) Epoch 26, batch 3300, loss[loss=0.1984, simple_loss=0.2789, pruned_loss=0.05896, over 19497.00 frames. ], tot_loss[loss=0.2058, simple_loss=0.2861, pruned_loss=0.06281, over 3799765.09 frames. ], batch size: 49, lr: 3.16e-03, grad_scale: 8.0 +2023-04-03 07:00:35,706 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-03 07:01:08,025 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.12 vs. limit=5.0 +2023-04-03 07:01:09,845 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2555, 2.2766, 1.7396, 2.3467, 2.3813, 1.7670, 1.8351, 2.1167], + device='cuda:0'), covar=tensor([0.1145, 0.1630, 0.1832, 0.1190, 0.1360, 0.0917, 0.1837, 0.1004], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0358, 0.0316, 0.0256, 0.0306, 0.0255, 0.0319, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:01:13,136 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=174033.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:01:17,498 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=174035.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:01:25,956 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=174042.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:01:35,836 INFO [train.py:903] (0/4) Epoch 26, batch 3350, loss[loss=0.1871, simple_loss=0.2773, pruned_loss=0.04844, over 19652.00 frames. ], tot_loss[loss=0.2057, simple_loss=0.2863, pruned_loss=0.06253, over 3788592.64 frames. ], batch size: 58, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:01:39,736 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0080, 1.5783, 1.8499, 1.6831, 4.4249, 1.1444, 2.7262, 4.8005], + device='cuda:0'), covar=tensor([0.0521, 0.3198, 0.3011, 0.2200, 0.0844, 0.2982, 0.1464, 0.0228], + device='cuda:0'), in_proj_covar=tensor([0.0419, 0.0374, 0.0393, 0.0352, 0.0378, 0.0354, 0.0392, 0.0412], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:01:49,566 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.212e+02 4.999e+02 5.843e+02 7.881e+02 1.777e+03, threshold=1.169e+03, percent-clipped=4.0 +2023-04-03 07:01:56,794 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=174067.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:02:01,246 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4924, 1.4879, 1.4576, 1.7981, 1.2806, 1.7100, 1.6669, 1.6462], + device='cuda:0'), covar=tensor([0.0869, 0.0919, 0.0961, 0.0647, 0.0847, 0.0752, 0.0826, 0.0663], + device='cuda:0'), in_proj_covar=tensor([0.0214, 0.0224, 0.0227, 0.0241, 0.0226, 0.0213, 0.0189, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 07:02:25,281 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-03 07:02:28,230 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=174091.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:02:39,922 INFO [train.py:903] (0/4) Epoch 26, batch 3400, loss[loss=0.1916, simple_loss=0.2821, pruned_loss=0.05053, over 17283.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2858, pruned_loss=0.06192, over 3791612.72 frames. ], batch size: 101, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:02:49,790 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2516, 2.0543, 1.9174, 2.1962, 1.8747, 1.9127, 1.7227, 2.1526], + device='cuda:0'), covar=tensor([0.0940, 0.1359, 0.1382, 0.0950, 0.1382, 0.0529, 0.1525, 0.0685], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0357, 0.0316, 0.0255, 0.0305, 0.0255, 0.0318, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:03:41,121 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-03 07:03:42,261 INFO [train.py:903] (0/4) Epoch 26, batch 3450, loss[loss=0.1874, simple_loss=0.2767, pruned_loss=0.04906, over 19769.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2849, pruned_loss=0.06153, over 3786554.88 frames. ], batch size: 56, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:03:51,574 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.58 vs. limit=5.0 +2023-04-03 07:03:53,304 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=174157.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:03:57,818 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.388e+02 4.666e+02 5.900e+02 7.449e+02 1.550e+03, threshold=1.180e+03, percent-clipped=6.0 +2023-04-03 07:04:08,803 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.19 vs. limit=2.0 +2023-04-03 07:04:39,824 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.41 vs. limit=2.0 +2023-04-03 07:04:47,526 INFO [train.py:903] (0/4) Epoch 26, batch 3500, loss[loss=0.1739, simple_loss=0.2476, pruned_loss=0.0501, over 19366.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2847, pruned_loss=0.06186, over 3806463.07 frames. ], batch size: 47, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:05:18,015 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=174224.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 07:05:49,673 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=174249.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 07:05:50,378 INFO [train.py:903] (0/4) Epoch 26, batch 3550, loss[loss=0.1748, simple_loss=0.2507, pruned_loss=0.04942, over 19318.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2846, pruned_loss=0.0618, over 3824267.03 frames. ], batch size: 44, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:05:55,405 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5818, 1.6912, 2.0782, 1.8427, 3.0513, 2.6092, 3.3358, 1.6415], + device='cuda:0'), covar=tensor([0.2625, 0.4499, 0.2781, 0.2023, 0.1704, 0.2215, 0.1714, 0.4630], + device='cuda:0'), in_proj_covar=tensor([0.0546, 0.0665, 0.0741, 0.0500, 0.0629, 0.0539, 0.0665, 0.0568], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 07:06:03,179 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.093e+02 4.676e+02 6.063e+02 7.972e+02 1.969e+03, threshold=1.213e+03, percent-clipped=7.0 +2023-04-03 07:06:53,367 INFO [train.py:903] (0/4) Epoch 26, batch 3600, loss[loss=0.189, simple_loss=0.2738, pruned_loss=0.05207, over 19691.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2855, pruned_loss=0.062, over 3818286.41 frames. ], batch size: 59, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:06:54,846 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=174301.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:07:05,339 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=174309.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:07:54,083 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=174347.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:07:57,121 INFO [train.py:903] (0/4) Epoch 26, batch 3650, loss[loss=0.2262, simple_loss=0.3099, pruned_loss=0.07122, over 19537.00 frames. ], tot_loss[loss=0.2053, simple_loss=0.2857, pruned_loss=0.06241, over 3820450.38 frames. ], batch size: 56, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:08:12,055 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.231e+02 5.064e+02 6.896e+02 8.623e+02 2.807e+03, threshold=1.379e+03, percent-clipped=9.0 +2023-04-03 07:08:26,843 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=174372.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:08:32,673 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=174377.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:08:35,128 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=174379.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:08:48,509 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.58 vs. limit=2.0 +2023-04-03 07:09:02,443 INFO [train.py:903] (0/4) Epoch 26, batch 3700, loss[loss=0.1761, simple_loss=0.2469, pruned_loss=0.05266, over 19146.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2854, pruned_loss=0.06191, over 3812969.61 frames. ], batch size: 42, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:09:10,643 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=174406.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:09:31,742 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9154, 1.2035, 1.4892, 0.6008, 1.9216, 2.4344, 2.1429, 2.5673], + device='cuda:0'), covar=tensor([0.1709, 0.4131, 0.3669, 0.2992, 0.0687, 0.0291, 0.0345, 0.0405], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0328, 0.0361, 0.0269, 0.0251, 0.0193, 0.0218, 0.0270], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 07:09:34,226 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3200, 2.1780, 2.0442, 1.9296, 1.6857, 1.9014, 0.5937, 1.3226], + device='cuda:0'), covar=tensor([0.0684, 0.0661, 0.0530, 0.0882, 0.1220, 0.0914, 0.1462, 0.1121], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0361, 0.0366, 0.0390, 0.0467, 0.0396, 0.0345, 0.0346], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 07:10:07,613 INFO [train.py:903] (0/4) Epoch 26, batch 3750, loss[loss=0.23, simple_loss=0.3118, pruned_loss=0.07415, over 17500.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2846, pruned_loss=0.06139, over 3815068.11 frames. ], batch size: 101, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:10:13,056 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-03 07:10:20,544 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.722e+02 4.995e+02 5.757e+02 7.069e+02 1.518e+03, threshold=1.151e+03, percent-clipped=1.0 +2023-04-03 07:11:01,675 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=174492.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:11:04,057 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=174494.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:11:05,280 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5094, 2.2366, 1.6613, 1.5543, 2.0573, 1.3766, 1.4359, 1.9391], + device='cuda:0'), covar=tensor([0.1087, 0.0755, 0.1108, 0.0864, 0.0578, 0.1280, 0.0737, 0.0496], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0319, 0.0338, 0.0271, 0.0248, 0.0343, 0.0293, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:11:10,966 INFO [train.py:903] (0/4) Epoch 26, batch 3800, loss[loss=0.1755, simple_loss=0.2629, pruned_loss=0.04406, over 19579.00 frames. ], tot_loss[loss=0.2024, simple_loss=0.2835, pruned_loss=0.06064, over 3827580.23 frames. ], batch size: 52, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:11:12,328 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=174501.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:11:42,960 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-03 07:12:14,070 INFO [train.py:903] (0/4) Epoch 26, batch 3850, loss[loss=0.2376, simple_loss=0.3152, pruned_loss=0.08004, over 19772.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2849, pruned_loss=0.06162, over 3816033.90 frames. ], batch size: 54, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:12:27,533 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.991e+02 5.118e+02 6.382e+02 8.415e+02 1.605e+03, threshold=1.276e+03, percent-clipped=5.0 +2023-04-03 07:13:15,204 INFO [train.py:903] (0/4) Epoch 26, batch 3900, loss[loss=0.1805, simple_loss=0.2593, pruned_loss=0.05084, over 19790.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2845, pruned_loss=0.06173, over 3816036.37 frames. ], batch size: 48, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:13:16,175 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-04-03 07:13:36,329 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=174616.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:14:13,851 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=174645.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:14:20,758 INFO [train.py:903] (0/4) Epoch 26, batch 3950, loss[loss=0.1952, simple_loss=0.2705, pruned_loss=0.05993, over 15235.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2855, pruned_loss=0.06205, over 3816297.37 frames. ], batch size: 33, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:14:24,350 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-03 07:14:24,476 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=174653.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:14:34,025 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.720e+02 4.927e+02 6.085e+02 7.650e+02 1.385e+03, threshold=1.217e+03, percent-clipped=3.0 +2023-04-03 07:15:17,656 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8714, 4.5153, 3.3891, 4.0188, 2.1830, 4.4098, 4.3602, 4.4680], + device='cuda:0'), covar=tensor([0.0446, 0.0854, 0.1618, 0.0732, 0.2635, 0.0613, 0.0811, 0.1057], + device='cuda:0'), in_proj_covar=tensor([0.0523, 0.0423, 0.0509, 0.0356, 0.0410, 0.0449, 0.0446, 0.0473], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:15:24,372 INFO [train.py:903] (0/4) Epoch 26, batch 4000, loss[loss=0.2413, simple_loss=0.3121, pruned_loss=0.08524, over 13267.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2846, pruned_loss=0.06157, over 3805813.38 frames. ], batch size: 135, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:15:53,888 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=174723.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:16:13,013 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-03 07:16:24,944 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=174748.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:16:26,987 INFO [train.py:903] (0/4) Epoch 26, batch 4050, loss[loss=0.219, simple_loss=0.2849, pruned_loss=0.07656, over 19726.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2842, pruned_loss=0.06123, over 3809202.12 frames. ], batch size: 51, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:16:27,145 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=174750.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:16:27,466 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=174750.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:16:35,717 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1760, 1.8089, 1.4586, 1.2611, 1.6505, 1.2411, 1.1675, 1.6185], + device='cuda:0'), covar=tensor([0.0837, 0.0789, 0.1107, 0.0807, 0.0548, 0.1245, 0.0661, 0.0468], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0320, 0.0339, 0.0272, 0.0250, 0.0346, 0.0293, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:16:40,296 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=174760.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:16:41,108 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.277e+02 4.947e+02 6.251e+02 7.600e+02 1.203e+03, threshold=1.250e+03, percent-clipped=0.0 +2023-04-03 07:16:52,058 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=174768.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:16:59,242 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=174773.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:17:01,654 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=174775.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:17:32,202 INFO [train.py:903] (0/4) Epoch 26, batch 4100, loss[loss=0.2367, simple_loss=0.3254, pruned_loss=0.074, over 19283.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2843, pruned_loss=0.06106, over 3817993.73 frames. ], batch size: 66, lr: 3.15e-03, grad_scale: 4.0 +2023-04-03 07:18:07,632 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-03 07:18:22,237 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6526, 4.2560, 2.7774, 3.7517, 0.9555, 4.2354, 4.0896, 4.1584], + device='cuda:0'), covar=tensor([0.0611, 0.1070, 0.1848, 0.0903, 0.4050, 0.0669, 0.0931, 0.1065], + device='cuda:0'), in_proj_covar=tensor([0.0521, 0.0421, 0.0506, 0.0354, 0.0409, 0.0447, 0.0444, 0.0471], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:18:38,181 INFO [train.py:903] (0/4) Epoch 26, batch 4150, loss[loss=0.1992, simple_loss=0.2877, pruned_loss=0.05536, over 19611.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2838, pruned_loss=0.06081, over 3813364.54 frames. ], batch size: 57, lr: 3.15e-03, grad_scale: 4.0 +2023-04-03 07:18:53,312 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.134e+02 5.046e+02 6.484e+02 8.542e+02 2.236e+03, threshold=1.297e+03, percent-clipped=8.0 +2023-04-03 07:18:57,232 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=174865.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:19:05,608 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=174872.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:19:11,134 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3615, 3.1281, 2.4016, 2.5156, 2.2596, 2.8023, 0.9227, 2.2630], + device='cuda:0'), covar=tensor([0.0738, 0.0606, 0.0753, 0.1152, 0.1109, 0.1075, 0.1628, 0.1024], + device='cuda:0'), in_proj_covar=tensor([0.0365, 0.0362, 0.0366, 0.0389, 0.0469, 0.0396, 0.0345, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 07:19:38,475 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=174897.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:19:41,597 INFO [train.py:903] (0/4) Epoch 26, batch 4200, loss[loss=0.1875, simple_loss=0.2643, pruned_loss=0.0554, over 19634.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2851, pruned_loss=0.06161, over 3803478.52 frames. ], batch size: 50, lr: 3.15e-03, grad_scale: 4.0 +2023-04-03 07:19:42,845 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-03 07:20:14,702 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1779, 1.8081, 1.4419, 1.2689, 1.6062, 1.2743, 1.1752, 1.6798], + device='cuda:0'), covar=tensor([0.0851, 0.0922, 0.1197, 0.0845, 0.0611, 0.1347, 0.0667, 0.0459], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0320, 0.0339, 0.0272, 0.0250, 0.0345, 0.0293, 0.0276], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:20:36,867 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6801, 1.6022, 1.6621, 2.3362, 1.6166, 2.0750, 2.0667, 1.7801], + device='cuda:0'), covar=tensor([0.0863, 0.0917, 0.0971, 0.0661, 0.0914, 0.0733, 0.0829, 0.0695], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0223, 0.0227, 0.0240, 0.0225, 0.0212, 0.0188, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 07:20:44,728 INFO [train.py:903] (0/4) Epoch 26, batch 4250, loss[loss=0.2257, simple_loss=0.3164, pruned_loss=0.06749, over 18323.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2849, pruned_loss=0.0618, over 3790923.69 frames. ], batch size: 83, lr: 3.15e-03, grad_scale: 4.0 +2023-04-03 07:20:55,179 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-03 07:21:01,855 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.522e+02 4.804e+02 5.687e+02 7.207e+02 1.586e+03, threshold=1.137e+03, percent-clipped=5.0 +2023-04-03 07:21:07,841 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-03 07:21:16,332 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-03 07:21:30,496 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6599, 1.5221, 1.5866, 2.0190, 1.5006, 1.8602, 1.8702, 1.7184], + device='cuda:0'), covar=tensor([0.0877, 0.1000, 0.1029, 0.0704, 0.0893, 0.0819, 0.0881, 0.0716], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0223, 0.0227, 0.0240, 0.0225, 0.0213, 0.0188, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 07:21:49,596 INFO [train.py:903] (0/4) Epoch 26, batch 4300, loss[loss=0.1935, simple_loss=0.2854, pruned_loss=0.05083, over 19538.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2856, pruned_loss=0.06224, over 3800340.88 frames. ], batch size: 54, lr: 3.15e-03, grad_scale: 4.0 +2023-04-03 07:22:11,532 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=175016.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:22:20,970 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=175024.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:22:38,902 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-03 07:22:41,547 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=175041.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:22:52,832 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=175049.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:22:53,613 INFO [train.py:903] (0/4) Epoch 26, batch 4350, loss[loss=0.1951, simple_loss=0.2842, pruned_loss=0.053, over 19760.00 frames. ], tot_loss[loss=0.2045, simple_loss=0.2851, pruned_loss=0.06197, over 3805883.82 frames. ], batch size: 63, lr: 3.15e-03, grad_scale: 4.0 +2023-04-03 07:23:08,529 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.209e+02 4.980e+02 6.586e+02 7.974e+02 2.340e+03, threshold=1.317e+03, percent-clipped=8.0 +2023-04-03 07:23:14,390 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=175067.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:23:33,070 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9538, 1.8089, 1.6676, 1.9593, 1.6445, 1.7306, 1.5906, 1.8949], + device='cuda:0'), covar=tensor([0.1100, 0.1555, 0.1570, 0.1032, 0.1593, 0.0584, 0.1621, 0.0824], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0355, 0.0314, 0.0254, 0.0304, 0.0255, 0.0316, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:23:56,993 INFO [train.py:903] (0/4) Epoch 26, batch 4400, loss[loss=0.1883, simple_loss=0.2683, pruned_loss=0.05416, over 19424.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2849, pruned_loss=0.06192, over 3811277.11 frames. ], batch size: 48, lr: 3.15e-03, grad_scale: 8.0 +2023-04-03 07:24:17,222 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-03 07:24:23,152 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=175121.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:24:26,283 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-03 07:24:41,395 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9823, 3.6271, 2.6489, 3.2287, 1.0784, 3.6189, 3.4956, 3.5239], + device='cuda:0'), covar=tensor([0.0896, 0.1134, 0.1908, 0.0972, 0.3904, 0.0816, 0.1053, 0.1352], + device='cuda:0'), in_proj_covar=tensor([0.0524, 0.0423, 0.0509, 0.0356, 0.0409, 0.0449, 0.0447, 0.0473], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:24:55,276 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=175146.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:24:59,549 INFO [train.py:903] (0/4) Epoch 26, batch 4450, loss[loss=0.2371, simple_loss=0.3115, pruned_loss=0.08135, over 17018.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2839, pruned_loss=0.06158, over 3830162.49 frames. ], batch size: 101, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:25:08,185 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=175157.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:25:14,965 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.113e+02 4.830e+02 6.413e+02 8.619e+02 2.132e+03, threshold=1.283e+03, percent-clipped=8.0 +2023-04-03 07:25:41,728 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=175182.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:26:02,897 INFO [train.py:903] (0/4) Epoch 26, batch 4500, loss[loss=0.2003, simple_loss=0.2837, pruned_loss=0.05847, over 19673.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2843, pruned_loss=0.06099, over 3834526.44 frames. ], batch size: 53, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:26:06,553 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1567, 3.3953, 1.9249, 2.0908, 2.9987, 1.7415, 1.6464, 2.2980], + device='cuda:0'), covar=tensor([0.1336, 0.0678, 0.1214, 0.0949, 0.0588, 0.1363, 0.1022, 0.0677], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0319, 0.0338, 0.0271, 0.0249, 0.0344, 0.0292, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:26:36,000 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=175225.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:26:38,394 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9700, 1.7419, 1.9176, 2.7146, 1.8188, 2.1193, 2.3468, 1.9806], + device='cuda:0'), covar=tensor([0.0893, 0.0989, 0.1011, 0.0785, 0.0993, 0.0819, 0.0912, 0.0712], + device='cuda:0'), in_proj_covar=tensor([0.0215, 0.0225, 0.0229, 0.0242, 0.0227, 0.0214, 0.0190, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 07:27:08,209 INFO [train.py:903] (0/4) Epoch 26, batch 4550, loss[loss=0.194, simple_loss=0.2746, pruned_loss=0.0567, over 19745.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2845, pruned_loss=0.06101, over 3830484.06 frames. ], batch size: 47, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:27:15,063 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-03 07:27:23,517 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.068e+02 4.834e+02 6.204e+02 7.660e+02 2.009e+03, threshold=1.241e+03, percent-clipped=3.0 +2023-04-03 07:27:28,773 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3316, 1.3857, 1.5613, 1.5037, 1.6974, 1.8429, 1.7395, 0.5519], + device='cuda:0'), covar=tensor([0.2552, 0.4349, 0.2699, 0.2016, 0.1710, 0.2372, 0.1519, 0.5035], + device='cuda:0'), in_proj_covar=tensor([0.0545, 0.0661, 0.0738, 0.0500, 0.0626, 0.0540, 0.0662, 0.0564], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 07:27:40,231 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-03 07:28:13,298 INFO [train.py:903] (0/4) Epoch 26, batch 4600, loss[loss=0.1699, simple_loss=0.2518, pruned_loss=0.04399, over 19770.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2854, pruned_loss=0.0616, over 3837263.58 frames. ], batch size: 48, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:29:16,038 INFO [train.py:903] (0/4) Epoch 26, batch 4650, loss[loss=0.1539, simple_loss=0.2339, pruned_loss=0.03699, over 18161.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2835, pruned_loss=0.06071, over 3835099.12 frames. ], batch size: 40, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:29:29,900 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.676e+02 4.879e+02 5.869e+02 7.462e+02 1.692e+03, threshold=1.174e+03, percent-clipped=3.0 +2023-04-03 07:29:31,128 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-03 07:29:44,485 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-03 07:30:00,505 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5025, 1.0086, 1.2328, 1.1790, 1.9903, 1.1040, 2.0862, 2.3391], + device='cuda:0'), covar=tensor([0.0975, 0.3660, 0.3487, 0.2120, 0.1437, 0.2457, 0.1253, 0.0705], + device='cuda:0'), in_proj_covar=tensor([0.0418, 0.0372, 0.0392, 0.0350, 0.0380, 0.0355, 0.0391, 0.0412], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:30:03,994 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=175388.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:30:17,713 INFO [train.py:903] (0/4) Epoch 26, batch 4700, loss[loss=0.2, simple_loss=0.2679, pruned_loss=0.06601, over 19022.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2836, pruned_loss=0.06091, over 3836226.39 frames. ], batch size: 42, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:30:35,684 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=175412.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:30:36,042 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.66 vs. limit=5.0 +2023-04-03 07:30:40,250 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0825, 1.9324, 1.7870, 1.6886, 1.4814, 1.6482, 0.4319, 1.0484], + device='cuda:0'), covar=tensor([0.0664, 0.0671, 0.0560, 0.0911, 0.1300, 0.0960, 0.1472, 0.1138], + device='cuda:0'), in_proj_covar=tensor([0.0363, 0.0362, 0.0367, 0.0389, 0.0468, 0.0395, 0.0344, 0.0346], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 07:30:41,026 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-03 07:30:43,655 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8309, 1.5961, 1.6543, 2.3622, 1.7284, 2.0489, 2.1450, 1.8322], + device='cuda:0'), covar=tensor([0.0836, 0.0958, 0.1009, 0.0715, 0.0870, 0.0735, 0.0835, 0.0694], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0224, 0.0228, 0.0240, 0.0226, 0.0213, 0.0188, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 07:31:06,994 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=175438.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:31:18,754 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8671, 1.5814, 1.4677, 1.7778, 1.5378, 1.6009, 1.4462, 1.7151], + device='cuda:0'), covar=tensor([0.1165, 0.1370, 0.1673, 0.1070, 0.1328, 0.0608, 0.1598, 0.0843], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0356, 0.0315, 0.0255, 0.0303, 0.0255, 0.0317, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:31:21,516 INFO [train.py:903] (0/4) Epoch 26, batch 4750, loss[loss=0.1782, simple_loss=0.271, pruned_loss=0.04273, over 19659.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2846, pruned_loss=0.06112, over 3836436.83 frames. ], batch size: 55, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:31:37,185 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.357e+02 4.945e+02 6.030e+02 7.655e+02 2.128e+03, threshold=1.206e+03, percent-clipped=6.0 +2023-04-03 07:31:38,753 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=175463.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:32:22,896 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=175499.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:32:23,792 INFO [train.py:903] (0/4) Epoch 26, batch 4800, loss[loss=0.167, simple_loss=0.2547, pruned_loss=0.03968, over 19490.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2841, pruned_loss=0.06079, over 3842159.02 frames. ], batch size: 49, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:32:25,130 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=175501.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:32:29,430 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.85 vs. limit=2.0 +2023-04-03 07:32:43,081 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1892, 1.2266, 1.6812, 1.2739, 2.6342, 3.5211, 3.2933, 3.8143], + device='cuda:0'), covar=tensor([0.1684, 0.4188, 0.3528, 0.2659, 0.0620, 0.0207, 0.0218, 0.0277], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0329, 0.0363, 0.0270, 0.0253, 0.0195, 0.0219, 0.0272], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 07:33:28,507 INFO [train.py:903] (0/4) Epoch 26, batch 4850, loss[loss=0.1829, simple_loss=0.2768, pruned_loss=0.04449, over 17340.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2837, pruned_loss=0.06068, over 3834038.47 frames. ], batch size: 101, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:33:42,548 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.071e+02 4.553e+02 5.462e+02 6.461e+02 1.537e+03, threshold=1.092e+03, percent-clipped=2.0 +2023-04-03 07:33:49,690 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-03 07:33:51,026 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=175569.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:34:12,016 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-03 07:34:17,956 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-03 07:34:17,978 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-03 07:34:19,653 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5221, 1.5902, 1.8111, 1.7439, 2.4851, 2.1980, 2.5749, 1.1118], + device='cuda:0'), covar=tensor([0.2499, 0.4310, 0.2696, 0.2015, 0.1599, 0.2296, 0.1497, 0.4712], + device='cuda:0'), in_proj_covar=tensor([0.0547, 0.0662, 0.0741, 0.0503, 0.0629, 0.0542, 0.0664, 0.0565], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 07:34:27,155 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-03 07:34:30,720 INFO [train.py:903] (0/4) Epoch 26, batch 4900, loss[loss=0.2402, simple_loss=0.3164, pruned_loss=0.08202, over 19483.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2838, pruned_loss=0.06084, over 3835874.28 frames. ], batch size: 64, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:34:46,804 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-03 07:34:50,742 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=175616.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:35:31,954 INFO [train.py:903] (0/4) Epoch 26, batch 4950, loss[loss=0.1921, simple_loss=0.2764, pruned_loss=0.0539, over 19667.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2848, pruned_loss=0.06161, over 3820656.04 frames. ], batch size: 58, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:35:49,941 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.056e+02 4.878e+02 6.108e+02 7.489e+02 1.803e+03, threshold=1.222e+03, percent-clipped=10.0 +2023-04-03 07:35:49,997 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-03 07:36:13,146 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-03 07:36:15,546 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=175684.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:36:36,321 INFO [train.py:903] (0/4) Epoch 26, batch 5000, loss[loss=0.1811, simple_loss=0.2594, pruned_loss=0.05142, over 19390.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2843, pruned_loss=0.06098, over 3828294.68 frames. ], batch size: 48, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:36:46,075 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-03 07:36:56,352 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-03 07:37:10,898 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=175728.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:37:16,377 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=175732.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:37:33,259 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.44 vs. limit=2.0 +2023-04-03 07:37:39,576 INFO [train.py:903] (0/4) Epoch 26, batch 5050, loss[loss=0.1895, simple_loss=0.2676, pruned_loss=0.05565, over 19414.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.285, pruned_loss=0.06123, over 3820452.60 frames. ], batch size: 48, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:37:46,789 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=175756.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:37:53,681 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.913e+02 4.916e+02 5.717e+02 6.994e+02 1.273e+03, threshold=1.143e+03, percent-clipped=1.0 +2023-04-03 07:38:14,304 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-03 07:38:37,155 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.76 vs. limit=2.0 +2023-04-03 07:38:42,564 INFO [train.py:903] (0/4) Epoch 26, batch 5100, loss[loss=0.2191, simple_loss=0.3024, pruned_loss=0.06789, over 19664.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2843, pruned_loss=0.06091, over 3822764.67 frames. ], batch size: 55, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:38:44,171 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1311, 2.1999, 2.3419, 2.6062, 2.1982, 2.5689, 2.3942, 2.2769], + device='cuda:0'), covar=tensor([0.3311, 0.2990, 0.1623, 0.1916, 0.3146, 0.1727, 0.3560, 0.2585], + device='cuda:0'), in_proj_covar=tensor([0.0928, 0.1004, 0.0737, 0.0949, 0.0907, 0.0843, 0.0857, 0.0803], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 07:38:49,488 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-03 07:38:49,788 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2013, 1.8564, 1.7462, 2.1363, 1.7930, 1.7970, 1.6948, 2.0342], + device='cuda:0'), covar=tensor([0.0991, 0.1432, 0.1573, 0.0967, 0.1393, 0.0586, 0.1528, 0.0744], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0360, 0.0318, 0.0257, 0.0307, 0.0259, 0.0321, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:38:52,906 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-03 07:38:57,449 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-03 07:39:37,086 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=175843.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:39:40,234 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.47 vs. limit=2.0 +2023-04-03 07:39:41,974 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=175847.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:39:45,160 INFO [train.py:903] (0/4) Epoch 26, batch 5150, loss[loss=0.1673, simple_loss=0.2483, pruned_loss=0.04317, over 19751.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2843, pruned_loss=0.06111, over 3802305.82 frames. ], batch size: 45, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:39:55,536 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-03 07:39:58,799 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.59 vs. limit=2.0 +2023-04-03 07:40:01,368 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.182e+02 5.003e+02 6.332e+02 8.398e+02 1.509e+03, threshold=1.266e+03, percent-clipped=8.0 +2023-04-03 07:40:14,035 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=175871.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:40:15,238 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=175872.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:40:30,102 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 07:40:47,054 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=175897.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:40:49,293 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=175899.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:40:50,195 INFO [train.py:903] (0/4) Epoch 26, batch 5200, loss[loss=0.1827, simple_loss=0.271, pruned_loss=0.04717, over 19666.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.284, pruned_loss=0.06106, over 3802863.34 frames. ], batch size: 60, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:41:02,280 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-03 07:41:12,232 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8642, 1.8322, 1.6846, 1.6342, 1.4718, 1.5702, 0.5216, 1.0338], + device='cuda:0'), covar=tensor([0.0576, 0.0569, 0.0416, 0.0569, 0.1012, 0.0778, 0.1247, 0.0972], + device='cuda:0'), in_proj_covar=tensor([0.0363, 0.0361, 0.0366, 0.0389, 0.0469, 0.0396, 0.0344, 0.0346], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 07:41:41,503 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=175940.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:41:45,617 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-03 07:41:53,690 INFO [train.py:903] (0/4) Epoch 26, batch 5250, loss[loss=0.1696, simple_loss=0.2484, pruned_loss=0.04544, over 19756.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2844, pruned_loss=0.06144, over 3806650.93 frames. ], batch size: 45, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:42:03,327 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=175958.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:42:07,499 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.944e+02 4.675e+02 5.849e+02 7.641e+02 1.436e+03, threshold=1.170e+03, percent-clipped=2.0 +2023-04-03 07:42:08,948 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=175963.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:42:11,236 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=175965.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:42:55,056 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-176000.pt +2023-04-03 07:42:56,005 INFO [train.py:903] (0/4) Epoch 26, batch 5300, loss[loss=0.1836, simple_loss=0.272, pruned_loss=0.04764, over 19771.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2852, pruned_loss=0.06226, over 3789385.95 frames. ], batch size: 56, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:43:08,844 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-03 07:43:57,535 INFO [train.py:903] (0/4) Epoch 26, batch 5350, loss[loss=0.1919, simple_loss=0.2775, pruned_loss=0.05311, over 19648.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.285, pruned_loss=0.06225, over 3803557.23 frames. ], batch size: 55, lr: 3.14e-03, grad_scale: 8.0 +2023-04-03 07:44:14,894 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.150e+02 4.809e+02 5.990e+02 7.353e+02 1.688e+03, threshold=1.198e+03, percent-clipped=9.0 +2023-04-03 07:44:27,819 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=176072.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:44:30,103 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-03 07:45:03,466 INFO [train.py:903] (0/4) Epoch 26, batch 5400, loss[loss=0.2327, simple_loss=0.3246, pruned_loss=0.07045, over 19662.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2846, pruned_loss=0.0616, over 3815463.86 frames. ], batch size: 58, lr: 3.14e-03, grad_scale: 4.0 +2023-04-03 07:45:07,635 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=176103.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:45:36,745 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=176127.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:45:37,874 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=176128.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:46:08,008 INFO [train.py:903] (0/4) Epoch 26, batch 5450, loss[loss=0.2245, simple_loss=0.304, pruned_loss=0.0725, over 18232.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2842, pruned_loss=0.06151, over 3820420.57 frames. ], batch size: 83, lr: 3.14e-03, grad_scale: 4.0 +2023-04-03 07:46:10,710 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=176152.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:46:23,251 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.815e+02 4.882e+02 5.855e+02 6.912e+02 1.680e+03, threshold=1.171e+03, percent-clipped=1.0 +2023-04-03 07:46:55,088 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=176187.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:47:11,437 INFO [train.py:903] (0/4) Epoch 26, batch 5500, loss[loss=0.2104, simple_loss=0.2901, pruned_loss=0.06537, over 19675.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2835, pruned_loss=0.06121, over 3815030.37 frames. ], batch size: 55, lr: 3.14e-03, grad_scale: 4.0 +2023-04-03 07:47:28,806 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=176214.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:47:30,665 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-03 07:47:48,427 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6494, 1.2249, 1.2734, 1.4848, 1.0839, 1.4169, 1.2774, 1.4494], + device='cuda:0'), covar=tensor([0.1175, 0.1279, 0.1753, 0.1117, 0.1469, 0.0673, 0.1587, 0.0907], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0355, 0.0315, 0.0255, 0.0304, 0.0255, 0.0317, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 07:48:02,343 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=176239.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:48:06,791 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=176243.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:48:14,675 INFO [train.py:903] (0/4) Epoch 26, batch 5550, loss[loss=0.2409, simple_loss=0.3156, pruned_loss=0.08312, over 19372.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2845, pruned_loss=0.06159, over 3802227.28 frames. ], batch size: 70, lr: 3.14e-03, grad_scale: 4.0 +2023-04-03 07:48:17,146 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-03 07:48:29,034 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=176260.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:48:32,240 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.168e+02 4.802e+02 5.930e+02 7.241e+02 1.738e+03, threshold=1.186e+03, percent-clipped=4.0 +2023-04-03 07:48:37,816 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-03 07:48:47,655 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=176275.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:49:08,002 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-03 07:49:18,895 INFO [train.py:903] (0/4) Epoch 26, batch 5600, loss[loss=0.1955, simple_loss=0.28, pruned_loss=0.0555, over 19622.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2848, pruned_loss=0.0614, over 3808838.96 frames. ], batch size: 57, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 07:49:28,225 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=176307.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:49:48,578 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.52 vs. limit=2.0 +2023-04-03 07:50:23,079 INFO [train.py:903] (0/4) Epoch 26, batch 5650, loss[loss=0.1555, simple_loss=0.2405, pruned_loss=0.03521, over 19330.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2851, pruned_loss=0.06158, over 3814912.22 frames. ], batch size: 44, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 07:50:26,193 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 07:50:32,680 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=176358.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:50:39,208 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.976e+02 4.381e+02 5.619e+02 7.137e+02 2.187e+03, threshold=1.124e+03, percent-clipped=4.0 +2023-04-03 07:51:02,586 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-03 07:51:24,935 INFO [train.py:903] (0/4) Epoch 26, batch 5700, loss[loss=0.1675, simple_loss=0.2485, pruned_loss=0.04318, over 19400.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2838, pruned_loss=0.06088, over 3824659.12 frames. ], batch size: 48, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 07:51:52,435 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=176422.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:52:19,141 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=176443.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:52:22,346 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-03 07:52:26,806 INFO [train.py:903] (0/4) Epoch 26, batch 5750, loss[loss=0.1804, simple_loss=0.264, pruned_loss=0.0484, over 19845.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2831, pruned_loss=0.06037, over 3839557.57 frames. ], batch size: 52, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 07:52:30,358 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-03 07:52:33,945 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-03 07:52:44,006 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.352e+02 5.112e+02 6.171e+02 7.862e+02 1.795e+03, threshold=1.234e+03, percent-clipped=7.0 +2023-04-03 07:52:49,987 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=176468.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:53:09,500 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=176484.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:53:28,519 INFO [train.py:903] (0/4) Epoch 26, batch 5800, loss[loss=0.2101, simple_loss=0.3036, pruned_loss=0.05828, over 19775.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.284, pruned_loss=0.06082, over 3832072.29 frames. ], batch size: 56, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 07:54:32,079 INFO [train.py:903] (0/4) Epoch 26, batch 5850, loss[loss=0.1941, simple_loss=0.2665, pruned_loss=0.06089, over 19755.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2834, pruned_loss=0.06098, over 3826435.57 frames. ], batch size: 46, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 07:54:48,251 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.299e+02 4.827e+02 6.114e+02 8.553e+02 2.097e+03, threshold=1.223e+03, percent-clipped=6.0 +2023-04-03 07:55:30,012 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-03 07:55:33,225 INFO [train.py:903] (0/4) Epoch 26, batch 5900, loss[loss=0.2106, simple_loss=0.2944, pruned_loss=0.06343, over 19662.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2849, pruned_loss=0.06166, over 3832834.76 frames. ], batch size: 60, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 07:55:37,889 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=176604.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:55:50,029 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=176614.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:55:51,762 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-03 07:55:55,443 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=176619.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:56:22,737 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=176639.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:56:27,347 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=176643.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:56:35,014 INFO [train.py:903] (0/4) Epoch 26, batch 5950, loss[loss=0.1718, simple_loss=0.2559, pruned_loss=0.04384, over 19492.00 frames. ], tot_loss[loss=0.2049, simple_loss=0.2857, pruned_loss=0.06203, over 3842277.24 frames. ], batch size: 49, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 07:56:51,412 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.421e+02 5.079e+02 6.315e+02 7.489e+02 1.732e+03, threshold=1.263e+03, percent-clipped=4.0 +2023-04-03 07:57:12,940 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=176678.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:57:18,394 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=176683.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:57:38,452 INFO [train.py:903] (0/4) Epoch 26, batch 6000, loss[loss=0.2091, simple_loss=0.2967, pruned_loss=0.06072, over 19563.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2853, pruned_loss=0.06157, over 3843936.32 frames. ], batch size: 56, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 07:57:38,453 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 07:57:51,361 INFO [train.py:937] (0/4) Epoch 26, validation: loss=0.1675, simple_loss=0.2672, pruned_loss=0.03393, over 944034.00 frames. +2023-04-03 07:57:51,362 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 07:57:55,493 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=176703.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:58:16,475 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=176719.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:58:36,003 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=176734.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 07:58:54,246 INFO [train.py:903] (0/4) Epoch 26, batch 6050, loss[loss=0.2094, simple_loss=0.2933, pruned_loss=0.06273, over 19786.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2863, pruned_loss=0.06184, over 3831507.07 frames. ], batch size: 56, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 07:58:58,144 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0424, 1.3768, 1.7467, 0.9034, 2.3954, 3.0979, 2.8340, 3.2890], + device='cuda:0'), covar=tensor([0.1626, 0.3651, 0.3173, 0.2652, 0.0590, 0.0232, 0.0257, 0.0340], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0327, 0.0361, 0.0269, 0.0253, 0.0193, 0.0218, 0.0272], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 07:59:11,767 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.381e+02 5.094e+02 6.239e+02 7.368e+02 1.384e+03, threshold=1.248e+03, percent-clipped=1.0 +2023-04-03 07:59:58,067 INFO [train.py:903] (0/4) Epoch 26, batch 6100, loss[loss=0.2349, simple_loss=0.3065, pruned_loss=0.08167, over 13685.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2858, pruned_loss=0.06217, over 3815580.54 frames. ], batch size: 135, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 08:00:17,907 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8342, 4.4358, 2.8935, 3.8810, 1.1056, 4.3639, 4.2717, 4.3808], + device='cuda:0'), covar=tensor([0.0546, 0.0874, 0.1839, 0.0869, 0.3951, 0.0653, 0.0925, 0.1066], + device='cuda:0'), in_proj_covar=tensor([0.0525, 0.0423, 0.0511, 0.0357, 0.0410, 0.0450, 0.0446, 0.0475], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 08:00:33,088 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=176828.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:00:43,833 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-03 08:01:00,302 INFO [train.py:903] (0/4) Epoch 26, batch 6150, loss[loss=0.1862, simple_loss=0.2575, pruned_loss=0.0574, over 19777.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2847, pruned_loss=0.0618, over 3821545.30 frames. ], batch size: 45, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 08:01:18,091 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.177e+02 4.899e+02 5.851e+02 7.446e+02 2.190e+03, threshold=1.170e+03, percent-clipped=4.0 +2023-04-03 08:01:21,509 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-03 08:01:25,331 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=176871.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:01:25,547 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4401, 2.4978, 2.7418, 3.2060, 2.4571, 3.1651, 2.7173, 2.5327], + device='cuda:0'), covar=tensor([0.4139, 0.4136, 0.1791, 0.2497, 0.4442, 0.2045, 0.4567, 0.3190], + device='cuda:0'), in_proj_covar=tensor([0.0929, 0.1007, 0.0739, 0.0952, 0.0908, 0.0845, 0.0860, 0.0804], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 08:02:01,461 INFO [train.py:903] (0/4) Epoch 26, batch 6200, loss[loss=0.2038, simple_loss=0.2852, pruned_loss=0.06126, over 18804.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2843, pruned_loss=0.06121, over 3832709.14 frames. ], batch size: 74, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 08:02:13,190 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5497, 1.4445, 1.4768, 1.9337, 1.4748, 1.7229, 1.8166, 1.6222], + device='cuda:0'), covar=tensor([0.0913, 0.0968, 0.1009, 0.0715, 0.0903, 0.0811, 0.0886, 0.0734], + device='cuda:0'), in_proj_covar=tensor([0.0214, 0.0226, 0.0229, 0.0241, 0.0227, 0.0213, 0.0189, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 08:02:54,365 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=176943.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:03:02,384 INFO [train.py:903] (0/4) Epoch 26, batch 6250, loss[loss=0.202, simple_loss=0.2922, pruned_loss=0.05596, over 19800.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2851, pruned_loss=0.06136, over 3827015.78 frames. ], batch size: 56, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 08:03:20,788 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.905e+02 4.899e+02 6.025e+02 7.517e+02 2.005e+03, threshold=1.205e+03, percent-clipped=5.0 +2023-04-03 08:03:29,817 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-03 08:03:33,681 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=176975.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:03:47,527 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=176987.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:03:51,227 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=176990.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:03:52,865 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.48 vs. limit=2.0 +2023-04-03 08:04:03,474 INFO [train.py:903] (0/4) Epoch 26, batch 6300, loss[loss=0.2234, simple_loss=0.3041, pruned_loss=0.07132, over 19517.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2855, pruned_loss=0.06163, over 3833080.43 frames. ], batch size: 64, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 08:04:03,951 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=177000.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:04:23,637 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=177015.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:04:37,551 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=177027.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:05:06,334 INFO [train.py:903] (0/4) Epoch 26, batch 6350, loss[loss=0.1967, simple_loss=0.2657, pruned_loss=0.0639, over 19768.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2846, pruned_loss=0.06135, over 3839077.62 frames. ], batch size: 45, lr: 3.13e-03, grad_scale: 4.0 +2023-04-03 08:05:26,137 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.448e+02 5.061e+02 6.104e+02 7.524e+02 1.291e+03, threshold=1.221e+03, percent-clipped=2.0 +2023-04-03 08:06:11,980 INFO [train.py:903] (0/4) Epoch 26, batch 6400, loss[loss=0.1892, simple_loss=0.273, pruned_loss=0.05266, over 19758.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2854, pruned_loss=0.06171, over 3836314.93 frames. ], batch size: 54, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 08:06:14,694 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=177102.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:06:19,569 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-03 08:06:45,640 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=177127.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:06:48,189 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=177129.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:07:05,538 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=177142.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:07:14,947 INFO [train.py:903] (0/4) Epoch 26, batch 6450, loss[loss=0.188, simple_loss=0.2771, pruned_loss=0.04941, over 19764.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2843, pruned_loss=0.0612, over 3826302.74 frames. ], batch size: 54, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 08:07:32,375 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 08:07:33,750 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.467e+02 4.651e+02 5.846e+02 7.696e+02 2.286e+03, threshold=1.169e+03, percent-clipped=3.0 +2023-04-03 08:07:56,667 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-03 08:08:15,638 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=177199.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:08:16,457 INFO [train.py:903] (0/4) Epoch 26, batch 6500, loss[loss=0.1708, simple_loss=0.2487, pruned_loss=0.04643, over 19735.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2851, pruned_loss=0.06156, over 3832194.71 frames. ], batch size: 45, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 08:08:17,675 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-03 08:08:38,036 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=177215.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:08:49,246 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=177224.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:09:20,909 INFO [train.py:903] (0/4) Epoch 26, batch 6550, loss[loss=0.1498, simple_loss=0.2354, pruned_loss=0.03215, over 19805.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2843, pruned_loss=0.06122, over 3820294.58 frames. ], batch size: 48, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 08:09:28,550 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6190, 1.4424, 1.4866, 2.1525, 1.6573, 1.9470, 1.9686, 1.6827], + device='cuda:0'), covar=tensor([0.0872, 0.0978, 0.1029, 0.0745, 0.0849, 0.0753, 0.0844, 0.0724], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0224, 0.0227, 0.0239, 0.0225, 0.0212, 0.0188, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 08:09:39,906 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.007e+02 4.796e+02 6.270e+02 7.966e+02 1.683e+03, threshold=1.254e+03, percent-clipped=5.0 +2023-04-03 08:10:25,212 INFO [train.py:903] (0/4) Epoch 26, batch 6600, loss[loss=0.2417, simple_loss=0.3176, pruned_loss=0.08288, over 17352.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2843, pruned_loss=0.06107, over 3822620.18 frames. ], batch size: 101, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 08:11:02,497 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=177330.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:11:27,598 INFO [train.py:903] (0/4) Epoch 26, batch 6650, loss[loss=0.2153, simple_loss=0.303, pruned_loss=0.0638, over 18261.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2843, pruned_loss=0.06127, over 3818553.99 frames. ], batch size: 84, lr: 3.13e-03, grad_scale: 8.0 +2023-04-03 08:11:37,270 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=177358.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:11:47,337 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.622e+02 4.339e+02 5.694e+02 7.782e+02 1.307e+03, threshold=1.139e+03, percent-clipped=1.0 +2023-04-03 08:12:11,036 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=177383.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:12:28,526 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=177398.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:12:30,607 INFO [train.py:903] (0/4) Epoch 26, batch 6700, loss[loss=0.1894, simple_loss=0.2722, pruned_loss=0.05327, over 19607.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2842, pruned_loss=0.06097, over 3819067.52 frames. ], batch size: 57, lr: 3.12e-03, grad_scale: 8.0 +2023-04-03 08:12:49,321 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5428, 1.5812, 1.7721, 1.7693, 2.4456, 2.2422, 2.5855, 1.0061], + device='cuda:0'), covar=tensor([0.2571, 0.4488, 0.2977, 0.2012, 0.1652, 0.2297, 0.1558, 0.4995], + device='cuda:0'), in_proj_covar=tensor([0.0549, 0.0665, 0.0743, 0.0502, 0.0630, 0.0543, 0.0664, 0.0567], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 08:13:01,352 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=177423.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:13:02,833 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-03 08:13:31,556 INFO [train.py:903] (0/4) Epoch 26, batch 6750, loss[loss=0.1722, simple_loss=0.2494, pruned_loss=0.0475, over 19354.00 frames. ], tot_loss[loss=0.2029, simple_loss=0.2837, pruned_loss=0.06102, over 3819331.14 frames. ], batch size: 47, lr: 3.12e-03, grad_scale: 8.0 +2023-04-03 08:13:48,516 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.769e+02 4.721e+02 5.880e+02 7.244e+02 1.873e+03, threshold=1.176e+03, percent-clipped=5.0 +2023-04-03 08:13:55,574 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=177471.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:13:57,752 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=177473.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:14:07,406 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 08:14:28,156 INFO [train.py:903] (0/4) Epoch 26, batch 6800, loss[loss=0.1883, simple_loss=0.2597, pruned_loss=0.05842, over 19720.00 frames. ], tot_loss[loss=0.2024, simple_loss=0.2833, pruned_loss=0.06077, over 3831496.39 frames. ], batch size: 46, lr: 3.12e-03, grad_scale: 8.0 +2023-04-03 08:14:58,603 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-26.pt +2023-04-03 08:15:14,315 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-03 08:15:15,395 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-03 08:15:18,385 INFO [train.py:903] (0/4) Epoch 27, batch 0, loss[loss=0.2081, simple_loss=0.2867, pruned_loss=0.06473, over 19453.00 frames. ], tot_loss[loss=0.2081, simple_loss=0.2867, pruned_loss=0.06473, over 19453.00 frames. ], batch size: 49, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:15:18,385 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 08:15:30,268 INFO [train.py:937] (0/4) Epoch 27, validation: loss=0.1666, simple_loss=0.2668, pruned_loss=0.03317, over 944034.00 frames. +2023-04-03 08:15:30,269 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 08:15:35,641 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2230, 1.3192, 1.2570, 1.0831, 1.1312, 1.1047, 0.1241, 0.3957], + device='cuda:0'), covar=tensor([0.0734, 0.0751, 0.0473, 0.0625, 0.1386, 0.0716, 0.1481, 0.1201], + device='cuda:0'), in_proj_covar=tensor([0.0361, 0.0360, 0.0364, 0.0387, 0.0468, 0.0392, 0.0342, 0.0344], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 08:15:42,910 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-03 08:16:15,181 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.146e+02 4.975e+02 6.244e+02 7.696e+02 2.158e+03, threshold=1.249e+03, percent-clipped=8.0 +2023-04-03 08:16:24,582 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=177571.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:16:33,682 INFO [train.py:903] (0/4) Epoch 27, batch 50, loss[loss=0.2211, simple_loss=0.2999, pruned_loss=0.07116, over 19658.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2831, pruned_loss=0.06013, over 867336.33 frames. ], batch size: 58, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:16:43,125 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=177586.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:16:43,179 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=177586.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:16:46,393 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=177588.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:17:06,248 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-03 08:17:15,809 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=177611.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:17:35,977 INFO [train.py:903] (0/4) Epoch 27, batch 100, loss[loss=0.207, simple_loss=0.3009, pruned_loss=0.05651, over 19671.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2839, pruned_loss=0.06066, over 1532178.45 frames. ], batch size: 55, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:17:47,465 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-03 08:17:47,819 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2087, 1.2601, 1.6325, 1.2066, 2.6892, 3.6377, 3.3872, 3.8753], + device='cuda:0'), covar=tensor([0.1664, 0.3921, 0.3522, 0.2681, 0.0664, 0.0218, 0.0221, 0.0278], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0328, 0.0361, 0.0269, 0.0251, 0.0194, 0.0218, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 08:18:23,247 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.041e+02 5.210e+02 6.703e+02 8.227e+02 2.617e+03, threshold=1.341e+03, percent-clipped=11.0 +2023-04-03 08:18:39,594 INFO [train.py:903] (0/4) Epoch 27, batch 150, loss[loss=0.1667, simple_loss=0.2396, pruned_loss=0.04688, over 19748.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.286, pruned_loss=0.06132, over 2046877.57 frames. ], batch size: 47, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:18:44,576 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=177682.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:19:40,066 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-03 08:19:42,203 INFO [train.py:903] (0/4) Epoch 27, batch 200, loss[loss=0.2283, simple_loss=0.3094, pruned_loss=0.07361, over 19442.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2844, pruned_loss=0.06138, over 2429729.64 frames. ], batch size: 70, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:20:29,446 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.575e+02 4.383e+02 5.368e+02 7.234e+02 1.640e+03, threshold=1.074e+03, percent-clipped=1.0 +2023-04-03 08:20:46,585 INFO [train.py:903] (0/4) Epoch 27, batch 250, loss[loss=0.2421, simple_loss=0.325, pruned_loss=0.07961, over 19786.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2839, pruned_loss=0.06101, over 2748567.35 frames. ], batch size: 56, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:21:12,211 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=177798.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:21:32,136 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=177815.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:21:43,032 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.39 vs. limit=5.0 +2023-04-03 08:21:50,875 INFO [train.py:903] (0/4) Epoch 27, batch 300, loss[loss=0.163, simple_loss=0.2465, pruned_loss=0.03976, over 19859.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2827, pruned_loss=0.0602, over 3001390.23 frames. ], batch size: 52, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:22:08,514 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=177842.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:22:10,821 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=177844.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:22:36,373 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.100e+02 5.015e+02 6.251e+02 7.839e+02 1.329e+03, threshold=1.250e+03, percent-clipped=8.0 +2023-04-03 08:22:40,002 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=177867.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:22:42,353 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=177869.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:22:52,701 INFO [train.py:903] (0/4) Epoch 27, batch 350, loss[loss=0.1852, simple_loss=0.2733, pruned_loss=0.04854, over 19524.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2821, pruned_loss=0.06003, over 3175295.98 frames. ], batch size: 54, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:23:00,638 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 08:23:40,476 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=177915.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:23:47,932 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2424, 2.1330, 1.9942, 1.8870, 1.6558, 1.8570, 0.8085, 1.3550], + device='cuda:0'), covar=tensor([0.0633, 0.0665, 0.0538, 0.0868, 0.1238, 0.1049, 0.1424, 0.1112], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0363, 0.0367, 0.0390, 0.0472, 0.0395, 0.0346, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 08:23:56,872 INFO [train.py:903] (0/4) Epoch 27, batch 400, loss[loss=0.1698, simple_loss=0.2577, pruned_loss=0.04089, over 19835.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2815, pruned_loss=0.05984, over 3307665.80 frames. ], batch size: 52, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:24:43,706 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.926e+02 4.795e+02 5.555e+02 6.674e+02 1.146e+03, threshold=1.111e+03, percent-clipped=0.0 +2023-04-03 08:24:58,338 INFO [train.py:903] (0/4) Epoch 27, batch 450, loss[loss=0.2363, simple_loss=0.3187, pruned_loss=0.07695, over 19506.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2825, pruned_loss=0.06032, over 3429360.31 frames. ], batch size: 64, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:25:26,935 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-178000.pt +2023-04-03 08:25:39,726 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-03 08:25:40,956 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-03 08:25:42,688 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2228, 2.1433, 2.0233, 1.8807, 1.6340, 1.8264, 0.7574, 1.3730], + device='cuda:0'), covar=tensor([0.0722, 0.0688, 0.0549, 0.0941, 0.1252, 0.1008, 0.1426, 0.1116], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0363, 0.0366, 0.0390, 0.0472, 0.0394, 0.0346, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 08:25:59,675 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=178026.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:26:02,530 INFO [train.py:903] (0/4) Epoch 27, batch 500, loss[loss=0.2144, simple_loss=0.2961, pruned_loss=0.06633, over 19670.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2828, pruned_loss=0.06033, over 3529006.71 frames. ], batch size: 59, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:26:06,490 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=178030.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:26:48,564 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.047e+02 5.151e+02 6.325e+02 8.134e+02 1.856e+03, threshold=1.265e+03, percent-clipped=5.0 +2023-04-03 08:27:07,126 INFO [train.py:903] (0/4) Epoch 27, batch 550, loss[loss=0.227, simple_loss=0.3036, pruned_loss=0.07519, over 19667.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2836, pruned_loss=0.06056, over 3598743.22 frames. ], batch size: 60, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:27:11,686 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.36 vs. limit=2.0 +2023-04-03 08:28:09,862 INFO [train.py:903] (0/4) Epoch 27, batch 600, loss[loss=0.2186, simple_loss=0.2976, pruned_loss=0.06987, over 18262.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2833, pruned_loss=0.06041, over 3640211.00 frames. ], batch size: 84, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:28:25,326 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=178141.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:28:26,321 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=178142.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:28:41,231 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6468, 1.4942, 1.4967, 2.1841, 1.5565, 1.9731, 1.9323, 1.7256], + device='cuda:0'), covar=tensor([0.0852, 0.0955, 0.1043, 0.0756, 0.0911, 0.0742, 0.0861, 0.0703], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0223, 0.0227, 0.0238, 0.0225, 0.0212, 0.0186, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 08:28:48,731 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=178159.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:28:54,526 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-03 08:28:55,499 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.215e+02 5.259e+02 6.338e+02 7.640e+02 1.730e+03, threshold=1.268e+03, percent-clipped=4.0 +2023-04-03 08:29:03,142 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.29 vs. limit=2.0 +2023-04-03 08:29:11,516 INFO [train.py:903] (0/4) Epoch 27, batch 650, loss[loss=0.1855, simple_loss=0.2733, pruned_loss=0.04884, over 19537.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2847, pruned_loss=0.06132, over 3680905.34 frames. ], batch size: 56, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:29:40,643 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7842, 1.4565, 1.6324, 1.6449, 3.3335, 1.2374, 2.5055, 3.8639], + device='cuda:0'), covar=tensor([0.0426, 0.2764, 0.2857, 0.1771, 0.0659, 0.2497, 0.1232, 0.0196], + device='cuda:0'), in_proj_covar=tensor([0.0421, 0.0375, 0.0393, 0.0352, 0.0380, 0.0357, 0.0391, 0.0414], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 08:30:11,962 INFO [train.py:903] (0/4) Epoch 27, batch 700, loss[loss=0.1846, simple_loss=0.2644, pruned_loss=0.05238, over 19396.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2838, pruned_loss=0.06086, over 3716868.43 frames. ], batch size: 48, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:30:16,999 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6539, 1.4750, 1.4971, 2.1565, 1.5473, 1.8639, 1.9454, 1.6577], + device='cuda:0'), covar=tensor([0.0856, 0.0962, 0.1014, 0.0694, 0.0885, 0.0759, 0.0812, 0.0727], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0224, 0.0227, 0.0238, 0.0225, 0.0212, 0.0187, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 08:30:17,024 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0606, 1.9488, 1.7438, 2.0536, 1.8129, 1.7355, 1.7424, 1.9417], + device='cuda:0'), covar=tensor([0.1092, 0.1371, 0.1520, 0.1068, 0.1457, 0.0607, 0.1400, 0.0794], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0361, 0.0320, 0.0258, 0.0309, 0.0259, 0.0323, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 08:30:49,335 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=178257.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:30:58,533 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.944e+02 4.694e+02 6.359e+02 8.615e+02 1.569e+03, threshold=1.272e+03, percent-clipped=7.0 +2023-04-03 08:31:11,144 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=178274.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:31:16,254 INFO [train.py:903] (0/4) Epoch 27, batch 750, loss[loss=0.2151, simple_loss=0.289, pruned_loss=0.07055, over 19688.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2836, pruned_loss=0.06076, over 3736888.49 frames. ], batch size: 53, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:31:26,151 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=178286.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:31:56,679 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=178311.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:32:17,785 INFO [train.py:903] (0/4) Epoch 27, batch 800, loss[loss=0.2022, simple_loss=0.2882, pruned_loss=0.05812, over 19632.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2839, pruned_loss=0.06074, over 3751901.75 frames. ], batch size: 61, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:32:31,924 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-03 08:32:58,401 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0822, 5.1248, 5.9224, 5.9560, 2.0692, 5.6044, 4.8321, 5.5886], + device='cuda:0'), covar=tensor([0.1674, 0.0951, 0.0586, 0.0658, 0.6361, 0.0856, 0.0631, 0.1210], + device='cuda:0'), in_proj_covar=tensor([0.0820, 0.0784, 0.0989, 0.0871, 0.0863, 0.0754, 0.0582, 0.0917], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 08:33:04,022 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.236e+02 4.845e+02 6.131e+02 7.065e+02 2.334e+03, threshold=1.226e+03, percent-clipped=1.0 +2023-04-03 08:33:20,260 INFO [train.py:903] (0/4) Epoch 27, batch 850, loss[loss=0.1655, simple_loss=0.2486, pruned_loss=0.04117, over 19410.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2855, pruned_loss=0.06104, over 3780919.19 frames. ], batch size: 48, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:33:45,242 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=178397.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:33:57,573 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5971, 4.2092, 2.6660, 3.7177, 0.9460, 4.2072, 4.0222, 4.1152], + device='cuda:0'), covar=tensor([0.0659, 0.0995, 0.2003, 0.0817, 0.4179, 0.0610, 0.0910, 0.1234], + device='cuda:0'), in_proj_covar=tensor([0.0528, 0.0427, 0.0515, 0.0359, 0.0410, 0.0452, 0.0447, 0.0476], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 08:34:12,703 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-03 08:34:16,570 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=178422.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:34:22,839 INFO [train.py:903] (0/4) Epoch 27, batch 900, loss[loss=0.2636, simple_loss=0.3285, pruned_loss=0.09936, over 19687.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2858, pruned_loss=0.0612, over 3784092.45 frames. ], batch size: 53, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:35:10,869 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.108e+02 4.416e+02 5.422e+02 6.593e+02 1.258e+03, threshold=1.084e+03, percent-clipped=1.0 +2023-04-03 08:35:16,994 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=178470.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:35:28,152 INFO [train.py:903] (0/4) Epoch 27, batch 950, loss[loss=0.1799, simple_loss=0.2594, pruned_loss=0.05015, over 19395.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2844, pruned_loss=0.06063, over 3786256.21 frames. ], batch size: 48, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:35:30,672 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-03 08:36:11,756 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=178513.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:36:32,190 INFO [train.py:903] (0/4) Epoch 27, batch 1000, loss[loss=0.236, simple_loss=0.3155, pruned_loss=0.07828, over 18831.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2849, pruned_loss=0.06083, over 3801144.05 frames. ], batch size: 74, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:36:34,996 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=178530.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:36:45,450 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=178538.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:37:01,969 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-03 08:37:06,188 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=178555.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:37:19,296 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.158e+02 4.971e+02 6.338e+02 8.822e+02 2.004e+03, threshold=1.268e+03, percent-clipped=12.0 +2023-04-03 08:37:25,279 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-03 08:37:35,828 INFO [train.py:903] (0/4) Epoch 27, batch 1050, loss[loss=0.1951, simple_loss=0.2739, pruned_loss=0.05819, over 19834.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2852, pruned_loss=0.06116, over 3811150.71 frames. ], batch size: 49, lr: 3.06e-03, grad_scale: 8.0 +2023-04-03 08:38:09,355 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-03 08:38:38,428 INFO [train.py:903] (0/4) Epoch 27, batch 1100, loss[loss=0.172, simple_loss=0.2528, pruned_loss=0.04562, over 19761.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.285, pruned_loss=0.06102, over 3811352.14 frames. ], batch size: 46, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:38:49,150 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=178637.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:39:27,052 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.060e+02 4.758e+02 5.833e+02 7.524e+02 1.653e+03, threshold=1.167e+03, percent-clipped=2.0 +2023-04-03 08:39:40,766 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.38 vs. limit=5.0 +2023-04-03 08:39:42,267 INFO [train.py:903] (0/4) Epoch 27, batch 1150, loss[loss=0.2187, simple_loss=0.3004, pruned_loss=0.06852, over 18742.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2856, pruned_loss=0.06196, over 3790219.58 frames. ], batch size: 74, lr: 3.05e-03, grad_scale: 4.0 +2023-04-03 08:39:42,649 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3354, 1.2769, 1.6121, 1.1006, 2.4267, 3.2823, 3.0094, 3.5380], + device='cuda:0'), covar=tensor([0.1582, 0.4111, 0.3675, 0.2835, 0.0684, 0.0223, 0.0254, 0.0304], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0330, 0.0363, 0.0272, 0.0253, 0.0196, 0.0219, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 08:40:25,272 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=178712.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:40:28,931 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=178715.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 08:40:47,635 INFO [train.py:903] (0/4) Epoch 27, batch 1200, loss[loss=0.2189, simple_loss=0.3003, pruned_loss=0.06872, over 19361.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2858, pruned_loss=0.06233, over 3784787.98 frames. ], batch size: 70, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:41:18,700 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-03 08:41:27,872 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-03 08:41:31,205 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4132, 1.4908, 1.6239, 1.5657, 1.8354, 1.9258, 1.8470, 0.7126], + device='cuda:0'), covar=tensor([0.2296, 0.4040, 0.2579, 0.1881, 0.1565, 0.2225, 0.1484, 0.4463], + device='cuda:0'), in_proj_covar=tensor([0.0550, 0.0665, 0.0744, 0.0502, 0.0631, 0.0542, 0.0664, 0.0567], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 08:41:38,086 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.626e+02 4.854e+02 6.177e+02 8.249e+02 1.593e+03, threshold=1.235e+03, percent-clipped=5.0 +2023-04-03 08:41:53,223 INFO [train.py:903] (0/4) Epoch 27, batch 1250, loss[loss=0.1939, simple_loss=0.2834, pruned_loss=0.05222, over 17504.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2851, pruned_loss=0.06183, over 3797393.11 frames. ], batch size: 101, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:42:05,375 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=178788.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:42:39,171 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=178814.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:42:56,587 INFO [train.py:903] (0/4) Epoch 27, batch 1300, loss[loss=0.1952, simple_loss=0.275, pruned_loss=0.0577, over 19681.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2845, pruned_loss=0.06162, over 3811942.60 frames. ], batch size: 53, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:43:44,695 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.050e+02 5.004e+02 6.015e+02 7.494e+02 1.649e+03, threshold=1.203e+03, percent-clipped=2.0 +2023-04-03 08:43:48,088 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.77 vs. limit=2.0 +2023-04-03 08:43:58,885 INFO [train.py:903] (0/4) Epoch 27, batch 1350, loss[loss=0.1964, simple_loss=0.2698, pruned_loss=0.06149, over 18699.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.284, pruned_loss=0.06137, over 3818422.97 frames. ], batch size: 41, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:44:41,867 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=178912.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 08:45:03,470 INFO [train.py:903] (0/4) Epoch 27, batch 1400, loss[loss=0.1908, simple_loss=0.2783, pruned_loss=0.05164, over 18140.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2846, pruned_loss=0.06152, over 3820840.38 frames. ], batch size: 83, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:45:05,018 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=178929.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:45:09,947 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4258, 1.4350, 1.6559, 1.6046, 2.1572, 2.0880, 2.2498, 0.9736], + device='cuda:0'), covar=tensor([0.2549, 0.4588, 0.2842, 0.2055, 0.1688, 0.2277, 0.1572, 0.4899], + device='cuda:0'), in_proj_covar=tensor([0.0551, 0.0666, 0.0745, 0.0503, 0.0631, 0.0542, 0.0666, 0.0568], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 08:45:42,599 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 08:45:50,938 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.415e+02 4.777e+02 6.195e+02 8.137e+02 1.607e+03, threshold=1.239e+03, percent-clipped=6.0 +2023-04-03 08:46:03,753 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-03 08:46:04,756 INFO [train.py:903] (0/4) Epoch 27, batch 1450, loss[loss=0.29, simple_loss=0.3447, pruned_loss=0.1176, over 13623.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2847, pruned_loss=0.06122, over 3827089.88 frames. ], batch size: 136, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:46:09,309 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=178981.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:47:07,119 INFO [train.py:903] (0/4) Epoch 27, batch 1500, loss[loss=0.2041, simple_loss=0.2949, pruned_loss=0.0567, over 19665.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2845, pruned_loss=0.06125, over 3802398.11 frames. ], batch size: 59, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:47:42,378 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=179056.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:47:45,632 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=179059.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 08:47:54,377 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.273e+02 4.749e+02 5.669e+02 7.445e+02 1.551e+03, threshold=1.134e+03, percent-clipped=3.0 +2023-04-03 08:48:08,110 INFO [train.py:903] (0/4) Epoch 27, batch 1550, loss[loss=0.2019, simple_loss=0.2852, pruned_loss=0.05932, over 19765.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2849, pruned_loss=0.06165, over 3807622.37 frames. ], batch size: 54, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:48:32,723 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=179096.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:49:12,281 INFO [train.py:903] (0/4) Epoch 27, batch 1600, loss[loss=0.1925, simple_loss=0.2706, pruned_loss=0.05719, over 19475.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2857, pruned_loss=0.06235, over 3805207.38 frames. ], batch size: 49, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:49:18,211 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=179132.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:49:37,301 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-03 08:49:59,058 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.135e+02 4.994e+02 6.186e+02 7.700e+02 1.707e+03, threshold=1.237e+03, percent-clipped=5.0 +2023-04-03 08:50:06,414 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=179171.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:50:09,940 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=179174.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 08:50:10,199 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-03 08:50:12,628 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.95 vs. limit=2.0 +2023-04-03 08:50:14,096 INFO [train.py:903] (0/4) Epoch 27, batch 1650, loss[loss=0.1771, simple_loss=0.2517, pruned_loss=0.05123, over 19366.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2854, pruned_loss=0.06233, over 3821910.41 frames. ], batch size: 47, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:50:23,701 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=179185.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:50:43,608 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.74 vs. limit=2.0 +2023-04-03 08:50:55,281 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=179210.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:50:56,537 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2601, 2.2960, 2.6298, 3.0353, 2.2579, 2.8233, 2.6700, 2.4642], + device='cuda:0'), covar=tensor([0.4373, 0.4460, 0.1864, 0.2664, 0.4814, 0.2379, 0.4754, 0.3364], + device='cuda:0'), in_proj_covar=tensor([0.0926, 0.1003, 0.0737, 0.0948, 0.0904, 0.0845, 0.0855, 0.0801], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 08:51:17,038 INFO [train.py:903] (0/4) Epoch 27, batch 1700, loss[loss=0.1799, simple_loss=0.2773, pruned_loss=0.04129, over 19652.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2847, pruned_loss=0.06187, over 3810905.95 frames. ], batch size: 58, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:51:41,286 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=179247.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:51:52,230 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=179256.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 08:51:58,266 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-03 08:52:04,051 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.943e+02 5.162e+02 6.407e+02 7.875e+02 2.392e+03, threshold=1.281e+03, percent-clipped=6.0 +2023-04-03 08:52:18,186 INFO [train.py:903] (0/4) Epoch 27, batch 1750, loss[loss=0.2073, simple_loss=0.2917, pruned_loss=0.06143, over 19762.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2851, pruned_loss=0.06177, over 3830945.97 frames. ], batch size: 63, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:52:34,544 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0285, 5.1628, 5.9227, 5.9616, 2.2524, 5.5989, 4.7941, 5.5919], + device='cuda:0'), covar=tensor([0.1836, 0.0805, 0.0564, 0.0635, 0.5923, 0.0824, 0.0605, 0.1091], + device='cuda:0'), in_proj_covar=tensor([0.0803, 0.0772, 0.0976, 0.0858, 0.0850, 0.0744, 0.0571, 0.0900], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 08:52:46,631 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.42 vs. limit=2.0 +2023-04-03 08:53:05,951 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-04-03 08:53:21,040 INFO [train.py:903] (0/4) Epoch 27, batch 1800, loss[loss=0.1613, simple_loss=0.2365, pruned_loss=0.04305, over 19751.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2848, pruned_loss=0.06124, over 3828117.22 frames. ], batch size: 45, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:53:51,567 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=179352.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:54:08,222 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.712e+02 4.924e+02 6.082e+02 7.444e+02 1.664e+03, threshold=1.216e+03, percent-clipped=4.0 +2023-04-03 08:54:15,279 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=179371.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 08:54:21,562 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-03 08:54:23,167 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=179377.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:54:23,936 INFO [train.py:903] (0/4) Epoch 27, batch 1850, loss[loss=0.1972, simple_loss=0.2822, pruned_loss=0.05612, over 19346.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2861, pruned_loss=0.06192, over 3813921.67 frames. ], batch size: 66, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:54:25,204 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=179379.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:54:45,779 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6182, 1.5306, 1.5102, 2.0491, 1.5087, 1.8097, 1.8681, 1.6476], + device='cuda:0'), covar=tensor([0.0892, 0.0932, 0.1024, 0.0773, 0.0923, 0.0811, 0.0903, 0.0756], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0225, 0.0228, 0.0240, 0.0227, 0.0214, 0.0189, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 08:55:00,568 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-03 08:55:19,879 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5833, 1.6984, 2.0904, 1.8736, 3.1546, 2.7358, 3.5939, 1.6762], + device='cuda:0'), covar=tensor([0.2550, 0.4456, 0.2824, 0.1999, 0.1544, 0.2078, 0.1417, 0.4449], + device='cuda:0'), in_proj_covar=tensor([0.0552, 0.0665, 0.0746, 0.0503, 0.0632, 0.0543, 0.0667, 0.0569], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 08:55:25,703 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5952, 1.5164, 1.5003, 2.0301, 1.5322, 1.8351, 1.9106, 1.6006], + device='cuda:0'), covar=tensor([0.0838, 0.0901, 0.1003, 0.0729, 0.0914, 0.0782, 0.0881, 0.0737], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0226, 0.0228, 0.0240, 0.0227, 0.0215, 0.0189, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 08:55:25,766 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=179427.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:55:26,554 INFO [train.py:903] (0/4) Epoch 27, batch 1900, loss[loss=0.2031, simple_loss=0.2885, pruned_loss=0.05887, over 18711.00 frames. ], tot_loss[loss=0.205, simple_loss=0.2859, pruned_loss=0.06205, over 3818876.77 frames. ], batch size: 74, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:55:29,237 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=179430.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 08:55:45,921 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-03 08:55:49,563 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-03 08:55:56,463 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=179452.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:56:00,898 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=179455.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 08:56:09,162 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=179462.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 08:56:14,506 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.488e+02 5.214e+02 6.120e+02 7.486e+02 1.853e+03, threshold=1.224e+03, percent-clipped=1.0 +2023-04-03 08:56:15,780 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-03 08:56:28,540 INFO [train.py:903] (0/4) Epoch 27, batch 1950, loss[loss=0.2311, simple_loss=0.3126, pruned_loss=0.07486, over 19437.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.2857, pruned_loss=0.06185, over 3830931.51 frames. ], batch size: 70, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:57:02,023 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=179503.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:57:31,829 INFO [train.py:903] (0/4) Epoch 27, batch 2000, loss[loss=0.2122, simple_loss=0.2953, pruned_loss=0.06453, over 19751.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2852, pruned_loss=0.06152, over 3821256.77 frames. ], batch size: 63, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:57:32,239 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=179528.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 08:57:55,541 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0182, 1.8348, 1.6432, 1.9192, 1.6529, 1.6896, 1.5888, 1.8644], + device='cuda:0'), covar=tensor([0.1128, 0.1495, 0.1612, 0.1219, 0.1524, 0.0631, 0.1594, 0.0790], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0359, 0.0317, 0.0256, 0.0306, 0.0256, 0.0320, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 08:58:19,143 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.349e+02 5.376e+02 6.611e+02 8.547e+02 2.231e+03, threshold=1.322e+03, percent-clipped=11.0 +2023-04-03 08:58:33,824 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-03 08:58:34,818 INFO [train.py:903] (0/4) Epoch 27, batch 2050, loss[loss=0.1914, simple_loss=0.2684, pruned_loss=0.05717, over 19488.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2854, pruned_loss=0.0617, over 3821398.52 frames. ], batch size: 49, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 08:58:53,485 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-03 08:58:54,679 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-03 08:59:14,233 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-03 08:59:37,294 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=179627.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 08:59:37,899 INFO [train.py:903] (0/4) Epoch 27, batch 2100, loss[loss=0.221, simple_loss=0.3055, pruned_loss=0.06828, over 19765.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.284, pruned_loss=0.06128, over 3814942.30 frames. ], batch size: 54, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 09:00:07,539 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-03 09:00:07,922 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=179652.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 09:00:10,134 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=179654.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:00:25,320 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.840e+02 4.609e+02 5.792e+02 6.874e+02 1.495e+03, threshold=1.158e+03, percent-clipped=2.0 +2023-04-03 09:00:28,896 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-03 09:00:39,306 INFO [train.py:903] (0/4) Epoch 27, batch 2150, loss[loss=0.2519, simple_loss=0.3212, pruned_loss=0.0913, over 12853.00 frames. ], tot_loss[loss=0.2029, simple_loss=0.2838, pruned_loss=0.06103, over 3810678.32 frames. ], batch size: 137, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 09:01:37,657 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=179723.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:01:43,433 INFO [train.py:903] (0/4) Epoch 27, batch 2200, loss[loss=0.1917, simple_loss=0.2628, pruned_loss=0.06031, over 19309.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2833, pruned_loss=0.06083, over 3817557.72 frames. ], batch size: 44, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 09:02:30,599 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.069e+02 5.235e+02 6.025e+02 7.394e+02 1.358e+03, threshold=1.205e+03, percent-clipped=1.0 +2023-04-03 09:02:46,359 INFO [train.py:903] (0/4) Epoch 27, batch 2250, loss[loss=0.265, simple_loss=0.3311, pruned_loss=0.09942, over 19398.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2847, pruned_loss=0.06179, over 3824204.98 frames. ], batch size: 70, lr: 3.05e-03, grad_scale: 8.0 +2023-04-03 09:03:21,827 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=179806.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 09:03:23,156 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2771, 2.8895, 2.4233, 2.3928, 2.0743, 2.5691, 0.9227, 2.1243], + device='cuda:0'), covar=tensor([0.0669, 0.0657, 0.0669, 0.1162, 0.1182, 0.1176, 0.1549, 0.1158], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0363, 0.0367, 0.0391, 0.0470, 0.0397, 0.0346, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 09:03:33,047 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=179814.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:03:50,156 INFO [train.py:903] (0/4) Epoch 27, batch 2300, loss[loss=0.2065, simple_loss=0.299, pruned_loss=0.05697, over 19588.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2844, pruned_loss=0.06113, over 3820268.70 frames. ], batch size: 61, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:04:02,173 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=179838.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:04:03,041 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-03 09:04:38,194 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.712e+02 5.173e+02 6.408e+02 8.011e+02 2.024e+03, threshold=1.282e+03, percent-clipped=7.0 +2023-04-03 09:04:43,397 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.43 vs. limit=5.0 +2023-04-03 09:04:51,778 INFO [train.py:903] (0/4) Epoch 27, batch 2350, loss[loss=0.1731, simple_loss=0.2496, pruned_loss=0.0483, over 19337.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2834, pruned_loss=0.06082, over 3825064.25 frames. ], batch size: 47, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:05:19,155 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-03 09:05:34,688 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-03 09:05:44,060 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6947, 1.5885, 1.6292, 2.2472, 1.5842, 2.0309, 1.9922, 1.7930], + device='cuda:0'), covar=tensor([0.0850, 0.0890, 0.0987, 0.0738, 0.0887, 0.0732, 0.0889, 0.0697], + device='cuda:0'), in_proj_covar=tensor([0.0210, 0.0223, 0.0227, 0.0239, 0.0225, 0.0212, 0.0188, 0.0204], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 09:05:45,206 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=179921.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 09:05:48,585 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=179924.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 09:05:50,588 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-03 09:05:53,778 INFO [train.py:903] (0/4) Epoch 27, batch 2400, loss[loss=0.2983, simple_loss=0.3554, pruned_loss=0.1206, over 12820.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2836, pruned_loss=0.061, over 3826851.51 frames. ], batch size: 137, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:06:11,803 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=179942.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:06:23,115 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5228, 1.4336, 1.4534, 1.8183, 1.3057, 1.6940, 1.6431, 1.5308], + device='cuda:0'), covar=tensor([0.0839, 0.0936, 0.1033, 0.0656, 0.0851, 0.0788, 0.0878, 0.0732], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0223, 0.0227, 0.0239, 0.0225, 0.0212, 0.0188, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 09:06:41,355 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.340e+02 4.979e+02 6.417e+02 8.310e+02 2.182e+03, threshold=1.283e+03, percent-clipped=4.0 +2023-04-03 09:06:57,078 INFO [train.py:903] (0/4) Epoch 27, batch 2450, loss[loss=0.1801, simple_loss=0.2583, pruned_loss=0.05099, over 19336.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2847, pruned_loss=0.06132, over 3824470.26 frames. ], batch size: 47, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:07:21,328 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=179998.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:07:23,502 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-180000.pt +2023-04-03 09:08:00,990 INFO [train.py:903] (0/4) Epoch 27, batch 2500, loss[loss=0.2128, simple_loss=0.2963, pruned_loss=0.06466, over 19759.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2854, pruned_loss=0.0613, over 3832813.24 frames. ], batch size: 63, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:08:19,368 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=180044.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:08:48,293 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.631e+02 4.714e+02 5.802e+02 6.784e+02 1.958e+03, threshold=1.160e+03, percent-clipped=2.0 +2023-04-03 09:09:02,324 INFO [train.py:903] (0/4) Epoch 27, batch 2550, loss[loss=0.1642, simple_loss=0.2447, pruned_loss=0.04182, over 19488.00 frames. ], tot_loss[loss=0.2046, simple_loss=0.2857, pruned_loss=0.06173, over 3837633.09 frames. ], batch size: 49, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:09:06,506 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-03 09:09:11,739 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7050, 1.5188, 1.5934, 2.2255, 1.7004, 1.9186, 1.9804, 1.6922], + device='cuda:0'), covar=tensor([0.0830, 0.0969, 0.1019, 0.0679, 0.0804, 0.0778, 0.0871, 0.0741], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0224, 0.0227, 0.0239, 0.0225, 0.0212, 0.0188, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 09:09:13,013 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5821, 1.7850, 2.1191, 1.9116, 3.0778, 2.6973, 3.3427, 1.6068], + device='cuda:0'), covar=tensor([0.2611, 0.4373, 0.2853, 0.2045, 0.1654, 0.2143, 0.1673, 0.4587], + device='cuda:0'), in_proj_covar=tensor([0.0549, 0.0663, 0.0744, 0.0501, 0.0629, 0.0541, 0.0664, 0.0567], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 09:09:22,060 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=180094.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:09:47,086 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=180113.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:09:54,385 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=180119.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:09:58,723 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-03 09:10:04,108 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-03 09:10:04,536 INFO [train.py:903] (0/4) Epoch 27, batch 2600, loss[loss=0.1967, simple_loss=0.2764, pruned_loss=0.05853, over 19737.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2843, pruned_loss=0.06114, over 3849379.16 frames. ], batch size: 51, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:10:42,738 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=180158.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:10:52,191 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.132e+02 5.258e+02 6.240e+02 7.647e+02 1.495e+03, threshold=1.248e+03, percent-clipped=4.0 +2023-04-03 09:11:00,081 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.16 vs. limit=2.0 +2023-04-03 09:11:07,732 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=180177.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 09:11:08,389 INFO [train.py:903] (0/4) Epoch 27, batch 2650, loss[loss=0.1704, simple_loss=0.246, pruned_loss=0.04738, over 19784.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.283, pruned_loss=0.06075, over 3839498.67 frames. ], batch size: 47, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:11:29,717 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-03 09:11:38,056 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=180202.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 09:11:50,372 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5454, 2.3931, 2.2455, 2.7040, 2.2259, 2.1596, 2.0317, 2.4818], + device='cuda:0'), covar=tensor([0.0968, 0.1624, 0.1322, 0.0910, 0.1424, 0.0524, 0.1415, 0.0679], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0359, 0.0317, 0.0257, 0.0307, 0.0256, 0.0321, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 09:12:04,700 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=180223.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:12:11,768 INFO [train.py:903] (0/4) Epoch 27, batch 2700, loss[loss=0.2327, simple_loss=0.3136, pruned_loss=0.07594, over 19777.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2829, pruned_loss=0.0603, over 3846147.96 frames. ], batch size: 56, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:13:00,986 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.426e+02 5.233e+02 6.139e+02 8.955e+02 2.009e+03, threshold=1.228e+03, percent-clipped=9.0 +2023-04-03 09:13:01,290 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5055, 1.3503, 1.7602, 1.5002, 2.7421, 3.7440, 3.4900, 4.0485], + device='cuda:0'), covar=tensor([0.1501, 0.3983, 0.3536, 0.2354, 0.0652, 0.0219, 0.0231, 0.0283], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0330, 0.0363, 0.0269, 0.0253, 0.0196, 0.0219, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 09:13:02,155 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=180268.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 09:13:07,997 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=180273.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:13:13,279 INFO [train.py:903] (0/4) Epoch 27, batch 2750, loss[loss=0.1816, simple_loss=0.2761, pruned_loss=0.04352, over 18683.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.284, pruned_loss=0.06086, over 3822350.71 frames. ], batch size: 74, lr: 3.04e-03, grad_scale: 4.0 +2023-04-03 09:13:22,763 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=180286.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:14:15,243 INFO [train.py:903] (0/4) Epoch 27, batch 2800, loss[loss=0.2059, simple_loss=0.2957, pruned_loss=0.05806, over 18043.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2846, pruned_loss=0.06088, over 3811658.48 frames. ], batch size: 83, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:14:49,773 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=180354.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:15:04,816 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.309e+02 4.838e+02 6.554e+02 8.151e+02 1.805e+03, threshold=1.311e+03, percent-clipped=3.0 +2023-04-03 09:15:07,630 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=180369.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:15:11,405 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.50 vs. limit=2.0 +2023-04-03 09:15:18,465 INFO [train.py:903] (0/4) Epoch 27, batch 2850, loss[loss=0.2421, simple_loss=0.3237, pruned_loss=0.08027, over 19610.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2839, pruned_loss=0.06083, over 3809798.03 frames. ], batch size: 57, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:15:26,263 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=180383.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 09:15:33,056 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=180388.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:15:40,137 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=180394.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:15:48,639 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=180401.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:16:20,688 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-03 09:16:22,940 INFO [train.py:903] (0/4) Epoch 27, batch 2900, loss[loss=0.1966, simple_loss=0.2717, pruned_loss=0.06073, over 19453.00 frames. ], tot_loss[loss=0.203, simple_loss=0.284, pruned_loss=0.06098, over 3814413.87 frames. ], batch size: 49, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:17:13,071 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.869e+02 4.922e+02 6.355e+02 8.136e+02 1.738e+03, threshold=1.271e+03, percent-clipped=4.0 +2023-04-03 09:17:13,372 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3077, 3.8020, 3.9140, 3.9504, 1.6278, 3.7388, 3.2666, 3.6609], + device='cuda:0'), covar=tensor([0.1745, 0.1098, 0.0710, 0.0787, 0.5964, 0.0987, 0.0768, 0.1235], + device='cuda:0'), in_proj_covar=tensor([0.0813, 0.0777, 0.0986, 0.0865, 0.0860, 0.0748, 0.0580, 0.0915], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 09:17:13,474 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5019, 1.4396, 1.4651, 1.7228, 1.3533, 1.6671, 1.6944, 1.5574], + device='cuda:0'), covar=tensor([0.0882, 0.0932, 0.1028, 0.0726, 0.0831, 0.0782, 0.0803, 0.0740], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0225, 0.0229, 0.0240, 0.0225, 0.0213, 0.0189, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 09:17:16,930 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6575, 1.5034, 1.6421, 2.1179, 1.5073, 1.8333, 1.9307, 1.6904], + device='cuda:0'), covar=tensor([0.0855, 0.0969, 0.0980, 0.0735, 0.0889, 0.0813, 0.0866, 0.0735], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0225, 0.0229, 0.0240, 0.0226, 0.0213, 0.0189, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 09:17:26,071 INFO [train.py:903] (0/4) Epoch 27, batch 2950, loss[loss=0.2249, simple_loss=0.3014, pruned_loss=0.07422, over 19603.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2836, pruned_loss=0.0607, over 3801442.40 frames. ], batch size: 61, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:17:56,660 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=180503.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:17:56,713 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4873, 2.2113, 1.7147, 1.4982, 1.9858, 1.5107, 1.3828, 1.9259], + device='cuda:0'), covar=tensor([0.0966, 0.0819, 0.1071, 0.0909, 0.0557, 0.1195, 0.0763, 0.0462], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0322, 0.0338, 0.0274, 0.0251, 0.0345, 0.0293, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 09:18:27,183 INFO [train.py:903] (0/4) Epoch 27, batch 3000, loss[loss=0.1876, simple_loss=0.2634, pruned_loss=0.05589, over 19358.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2841, pruned_loss=0.06117, over 3788469.13 frames. ], batch size: 47, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:18:27,184 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 09:18:39,752 INFO [train.py:937] (0/4) Epoch 27, validation: loss=0.1667, simple_loss=0.2664, pruned_loss=0.03355, over 944034.00 frames. +2023-04-03 09:18:39,753 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 09:18:41,272 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=180529.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:18:43,412 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-03 09:19:11,482 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=180554.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:19:20,241 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2636, 1.3398, 1.8069, 1.4329, 2.7610, 3.7950, 3.4592, 3.9750], + device='cuda:0'), covar=tensor([0.1605, 0.4004, 0.3413, 0.2502, 0.0643, 0.0198, 0.0218, 0.0239], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0331, 0.0364, 0.0270, 0.0255, 0.0196, 0.0219, 0.0276], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 09:19:26,238 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.17 vs. limit=5.0 +2023-04-03 09:19:29,304 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=180567.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:19:30,278 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.939e+02 4.587e+02 5.827e+02 7.595e+02 1.750e+03, threshold=1.165e+03, percent-clipped=2.0 +2023-04-03 09:19:41,904 INFO [train.py:903] (0/4) Epoch 27, batch 3050, loss[loss=0.2225, simple_loss=0.301, pruned_loss=0.07206, over 19669.00 frames. ], tot_loss[loss=0.2024, simple_loss=0.2834, pruned_loss=0.06065, over 3804308.78 frames. ], batch size: 58, lr: 3.04e-03, grad_scale: 4.0 +2023-04-03 09:19:54,601 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=180588.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:20:44,775 INFO [train.py:903] (0/4) Epoch 27, batch 3100, loss[loss=0.1853, simple_loss=0.2691, pruned_loss=0.05076, over 19728.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2835, pruned_loss=0.06068, over 3800051.50 frames. ], batch size: 51, lr: 3.04e-03, grad_scale: 4.0 +2023-04-03 09:20:59,808 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=180639.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 09:21:20,334 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=180657.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:21:29,218 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=180664.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 09:21:33,516 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.182e+02 4.719e+02 5.890e+02 7.652e+02 1.677e+03, threshold=1.178e+03, percent-clipped=3.0 +2023-04-03 09:21:46,248 INFO [train.py:903] (0/4) Epoch 27, batch 3150, loss[loss=0.2058, simple_loss=0.2897, pruned_loss=0.06099, over 19755.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.284, pruned_loss=0.06074, over 3807992.55 frames. ], batch size: 63, lr: 3.04e-03, grad_scale: 4.0 +2023-04-03 09:21:52,953 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=180682.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:21:52,994 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=180682.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:22:11,317 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=180698.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:22:15,894 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-03 09:22:49,864 INFO [train.py:903] (0/4) Epoch 27, batch 3200, loss[loss=0.1909, simple_loss=0.2827, pruned_loss=0.04955, over 19591.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2838, pruned_loss=0.0606, over 3815762.41 frames. ], batch size: 61, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:23:16,671 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=180750.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:23:27,991 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=180759.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:23:39,759 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.293e+02 5.255e+02 6.624e+02 9.042e+02 3.460e+03, threshold=1.325e+03, percent-clipped=12.0 +2023-04-03 09:23:51,466 INFO [train.py:903] (0/4) Epoch 27, batch 3250, loss[loss=0.1816, simple_loss=0.257, pruned_loss=0.05308, over 19627.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.284, pruned_loss=0.06066, over 3816202.17 frames. ], batch size: 50, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:23:59,260 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7282, 4.2358, 4.4657, 4.4721, 1.7703, 4.1895, 3.6339, 4.1816], + device='cuda:0'), covar=tensor([0.1926, 0.0973, 0.0618, 0.0762, 0.6264, 0.0991, 0.0783, 0.1163], + device='cuda:0'), in_proj_covar=tensor([0.0819, 0.0781, 0.0993, 0.0872, 0.0863, 0.0752, 0.0583, 0.0918], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 09:23:59,387 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=180784.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:24:36,417 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=180813.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:24:54,208 INFO [train.py:903] (0/4) Epoch 27, batch 3300, loss[loss=0.2804, simple_loss=0.3416, pruned_loss=0.1096, over 19536.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2827, pruned_loss=0.06045, over 3826155.99 frames. ], batch size: 56, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:25:00,022 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-03 09:25:13,334 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=180843.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:25:43,903 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.436e+02 5.181e+02 6.566e+02 8.708e+02 1.878e+03, threshold=1.313e+03, percent-clipped=7.0 +2023-04-03 09:25:56,104 INFO [train.py:903] (0/4) Epoch 27, batch 3350, loss[loss=0.1784, simple_loss=0.258, pruned_loss=0.04941, over 18609.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2828, pruned_loss=0.06051, over 3817100.16 frames. ], batch size: 41, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:26:18,970 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-04-03 09:27:00,204 INFO [train.py:903] (0/4) Epoch 27, batch 3400, loss[loss=0.2455, simple_loss=0.3242, pruned_loss=0.08342, over 18209.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2834, pruned_loss=0.06084, over 3818597.38 frames. ], batch size: 83, lr: 3.04e-03, grad_scale: 8.0 +2023-04-03 09:27:05,026 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=180932.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:27:11,994 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=180938.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:27:43,939 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=180963.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:27:50,447 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.175e+02 4.823e+02 5.841e+02 8.287e+02 1.627e+03, threshold=1.168e+03, percent-clipped=5.0 +2023-04-03 09:28:02,295 INFO [train.py:903] (0/4) Epoch 27, batch 3450, loss[loss=0.2187, simple_loss=0.3029, pruned_loss=0.06724, over 19358.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2835, pruned_loss=0.06107, over 3811685.13 frames. ], batch size: 70, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:28:06,948 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-03 09:29:04,451 INFO [train.py:903] (0/4) Epoch 27, batch 3500, loss[loss=0.2267, simple_loss=0.3098, pruned_loss=0.07178, over 19541.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.284, pruned_loss=0.06123, over 3815424.92 frames. ], batch size: 56, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:29:09,326 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4063, 1.4881, 2.0394, 1.6534, 3.2259, 4.2947, 4.1012, 4.6940], + device='cuda:0'), covar=tensor([0.1685, 0.3907, 0.3453, 0.2453, 0.0644, 0.0298, 0.0206, 0.0218], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0330, 0.0363, 0.0268, 0.0253, 0.0195, 0.0217, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 09:29:28,519 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=181047.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:29:45,289 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6991, 1.6644, 2.0181, 1.6414, 4.2424, 1.3156, 2.7877, 4.6043], + device='cuda:0'), covar=tensor([0.0430, 0.2898, 0.2526, 0.2098, 0.0713, 0.2602, 0.1343, 0.0176], + device='cuda:0'), in_proj_covar=tensor([0.0421, 0.0376, 0.0394, 0.0351, 0.0381, 0.0355, 0.0391, 0.0414], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 09:29:53,845 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.012e+02 5.177e+02 6.114e+02 8.226e+02 1.424e+03, threshold=1.223e+03, percent-clipped=6.0 +2023-04-03 09:29:55,468 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=181069.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:30:06,529 INFO [train.py:903] (0/4) Epoch 27, batch 3550, loss[loss=0.2255, simple_loss=0.3159, pruned_loss=0.06759, over 19595.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2834, pruned_loss=0.06063, over 3816229.11 frames. ], batch size: 61, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:30:27,206 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=181094.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:30:27,435 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=181094.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:31:09,199 INFO [train.py:903] (0/4) Epoch 27, batch 3600, loss[loss=0.2341, simple_loss=0.311, pruned_loss=0.07856, over 19652.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2836, pruned_loss=0.06052, over 3829821.24 frames. ], batch size: 58, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:31:30,383 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.59 vs. limit=5.0 +2023-04-03 09:32:00,753 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.317e+02 4.881e+02 5.766e+02 7.270e+02 1.755e+03, threshold=1.153e+03, percent-clipped=5.0 +2023-04-03 09:32:12,504 INFO [train.py:903] (0/4) Epoch 27, batch 3650, loss[loss=0.1717, simple_loss=0.2569, pruned_loss=0.04329, over 19849.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2843, pruned_loss=0.06054, over 3815971.45 frames. ], batch size: 52, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:32:24,257 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=181187.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:32:52,745 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=181209.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:32:54,499 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.55 vs. limit=5.0 +2023-04-03 09:33:15,867 INFO [train.py:903] (0/4) Epoch 27, batch 3700, loss[loss=0.2268, simple_loss=0.2998, pruned_loss=0.07691, over 19323.00 frames. ], tot_loss[loss=0.2047, simple_loss=0.286, pruned_loss=0.06169, over 3788549.49 frames. ], batch size: 66, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:33:19,931 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.49 vs. limit=2.0 +2023-04-03 09:33:46,734 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.22 vs. limit=2.0 +2023-04-03 09:34:06,272 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.101e+02 5.357e+02 6.375e+02 8.389e+02 2.807e+03, threshold=1.275e+03, percent-clipped=10.0 +2023-04-03 09:34:17,471 INFO [train.py:903] (0/4) Epoch 27, batch 3750, loss[loss=0.154, simple_loss=0.2374, pruned_loss=0.03531, over 19779.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2854, pruned_loss=0.06168, over 3793439.10 frames. ], batch size: 48, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:34:48,303 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=181302.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:34:49,693 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=181303.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:35:04,203 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=181315.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:35:20,118 INFO [train.py:903] (0/4) Epoch 27, batch 3800, loss[loss=0.1809, simple_loss=0.2589, pruned_loss=0.05141, over 19739.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2854, pruned_loss=0.06163, over 3807940.70 frames. ], batch size: 51, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:35:20,512 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=181328.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:35:48,544 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-03 09:36:10,635 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.827e+02 5.403e+02 6.773e+02 8.761e+02 1.536e+03, threshold=1.355e+03, percent-clipped=8.0 +2023-04-03 09:36:22,055 INFO [train.py:903] (0/4) Epoch 27, batch 3850, loss[loss=0.2134, simple_loss=0.2919, pruned_loss=0.06748, over 19762.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2851, pruned_loss=0.06131, over 3825173.89 frames. ], batch size: 54, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:37:25,666 INFO [train.py:903] (0/4) Epoch 27, batch 3900, loss[loss=0.1665, simple_loss=0.2508, pruned_loss=0.04107, over 19482.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2851, pruned_loss=0.061, over 3819973.43 frames. ], batch size: 49, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:37:49,513 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.60 vs. limit=5.0 +2023-04-03 09:37:52,692 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9879, 2.1057, 2.3088, 2.6811, 2.0639, 2.6133, 2.2921, 2.1289], + device='cuda:0'), covar=tensor([0.4455, 0.4313, 0.2050, 0.2522, 0.4425, 0.2283, 0.5090, 0.3519], + device='cuda:0'), in_proj_covar=tensor([0.0931, 0.1008, 0.0739, 0.0949, 0.0907, 0.0848, 0.0858, 0.0805], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 09:38:13,064 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=181465.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:38:17,127 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.394e+02 5.123e+02 6.226e+02 8.139e+02 1.802e+03, threshold=1.245e+03, percent-clipped=4.0 +2023-04-03 09:38:28,445 INFO [train.py:903] (0/4) Epoch 27, batch 3950, loss[loss=0.2368, simple_loss=0.3193, pruned_loss=0.07709, over 17462.00 frames. ], tot_loss[loss=0.2029, simple_loss=0.2846, pruned_loss=0.06062, over 3813737.84 frames. ], batch size: 101, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:38:30,603 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-03 09:38:43,818 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=181490.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:39:30,606 INFO [train.py:903] (0/4) Epoch 27, batch 4000, loss[loss=0.1903, simple_loss=0.2693, pruned_loss=0.05565, over 19728.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2846, pruned_loss=0.0608, over 3809563.46 frames. ], batch size: 51, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:39:55,701 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=181548.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:40:08,402 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=181558.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:40:14,677 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-03 09:40:21,509 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.200e+02 5.331e+02 6.990e+02 9.796e+02 2.756e+03, threshold=1.398e+03, percent-clipped=11.0 +2023-04-03 09:40:32,615 INFO [train.py:903] (0/4) Epoch 27, batch 4050, loss[loss=0.1726, simple_loss=0.2528, pruned_loss=0.0462, over 19605.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2841, pruned_loss=0.06056, over 3810132.12 frames. ], batch size: 50, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:40:38,777 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=181583.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:41:35,043 INFO [train.py:903] (0/4) Epoch 27, batch 4100, loss[loss=0.22, simple_loss=0.3037, pruned_loss=0.06814, over 19527.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2848, pruned_loss=0.06113, over 3811782.54 frames. ], batch size: 54, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:42:07,717 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-03 09:42:13,759 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=181659.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:42:26,972 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.417e+02 5.142e+02 6.045e+02 7.647e+02 1.406e+03, threshold=1.209e+03, percent-clipped=1.0 +2023-04-03 09:42:35,660 INFO [train.py:903] (0/4) Epoch 27, batch 4150, loss[loss=0.1798, simple_loss=0.2596, pruned_loss=0.04994, over 19388.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2844, pruned_loss=0.06084, over 3817954.63 frames. ], batch size: 47, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:43:38,706 INFO [train.py:903] (0/4) Epoch 27, batch 4200, loss[loss=0.2234, simple_loss=0.298, pruned_loss=0.07438, over 13327.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2846, pruned_loss=0.06106, over 3814949.31 frames. ], batch size: 136, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:43:42,089 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-03 09:44:19,619 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0901, 2.0868, 1.8256, 2.1683, 1.9383, 1.8627, 1.7459, 2.0335], + device='cuda:0'), covar=tensor([0.1115, 0.1476, 0.1570, 0.1104, 0.1435, 0.0576, 0.1592, 0.0784], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0360, 0.0319, 0.0258, 0.0309, 0.0257, 0.0322, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 09:44:30,941 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.172e+02 5.083e+02 6.482e+02 7.948e+02 1.533e+03, threshold=1.296e+03, percent-clipped=4.0 +2023-04-03 09:44:36,093 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9568, 4.4959, 2.7831, 3.9176, 0.7837, 4.4941, 4.3300, 4.4508], + device='cuda:0'), covar=tensor([0.0548, 0.0969, 0.1996, 0.0888, 0.4263, 0.0646, 0.0911, 0.1158], + device='cuda:0'), in_proj_covar=tensor([0.0526, 0.0426, 0.0510, 0.0357, 0.0406, 0.0452, 0.0447, 0.0476], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 09:44:36,260 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=181774.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:44:40,471 INFO [train.py:903] (0/4) Epoch 27, batch 4250, loss[loss=0.2098, simple_loss=0.279, pruned_loss=0.07035, over 19109.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2846, pruned_loss=0.06101, over 3820858.69 frames. ], batch size: 42, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:44:55,302 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-03 09:45:08,537 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-03 09:45:43,727 INFO [train.py:903] (0/4) Epoch 27, batch 4300, loss[loss=0.2099, simple_loss=0.2893, pruned_loss=0.06526, over 19715.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2844, pruned_loss=0.06105, over 3815714.87 frames. ], batch size: 63, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:45:44,232 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4106, 1.4373, 1.5540, 1.5577, 1.8211, 1.8972, 1.7872, 0.5880], + device='cuda:0'), covar=tensor([0.2414, 0.4272, 0.2586, 0.1993, 0.1626, 0.2339, 0.1431, 0.5069], + device='cuda:0'), in_proj_covar=tensor([0.0552, 0.0667, 0.0749, 0.0505, 0.0633, 0.0545, 0.0668, 0.0570], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 09:46:36,877 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.524e+02 4.917e+02 6.450e+02 8.224e+02 1.543e+03, threshold=1.290e+03, percent-clipped=3.0 +2023-04-03 09:46:40,227 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-03 09:46:47,396 INFO [train.py:903] (0/4) Epoch 27, batch 4350, loss[loss=0.2044, simple_loss=0.2876, pruned_loss=0.06058, over 19699.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2844, pruned_loss=0.0614, over 3808810.29 frames. ], batch size: 60, lr: 3.03e-03, grad_scale: 4.0 +2023-04-03 09:47:05,497 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=181892.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:47:21,788 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5708, 1.1441, 1.3948, 1.3051, 2.2007, 1.1292, 2.1506, 2.5237], + device='cuda:0'), covar=tensor([0.0732, 0.3093, 0.3053, 0.1787, 0.0955, 0.2132, 0.1075, 0.0490], + device='cuda:0'), in_proj_covar=tensor([0.0421, 0.0376, 0.0394, 0.0352, 0.0381, 0.0354, 0.0392, 0.0414], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 09:47:33,518 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-03 09:47:38,038 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6468, 1.7855, 2.1381, 1.9402, 3.2477, 2.9286, 3.7071, 1.6797], + device='cuda:0'), covar=tensor([0.2413, 0.4323, 0.2719, 0.1850, 0.1519, 0.1932, 0.1403, 0.4180], + device='cuda:0'), in_proj_covar=tensor([0.0555, 0.0671, 0.0754, 0.0507, 0.0638, 0.0548, 0.0671, 0.0573], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 09:47:49,883 INFO [train.py:903] (0/4) Epoch 27, batch 4400, loss[loss=0.1531, simple_loss=0.2287, pruned_loss=0.03879, over 19749.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2838, pruned_loss=0.0614, over 3817703.18 frames. ], batch size: 47, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:48:15,001 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-03 09:48:24,195 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-03 09:48:42,706 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.593e+02 4.919e+02 6.507e+02 9.105e+02 1.976e+03, threshold=1.301e+03, percent-clipped=10.0 +2023-04-03 09:48:52,975 INFO [train.py:903] (0/4) Epoch 27, batch 4450, loss[loss=0.2015, simple_loss=0.2861, pruned_loss=0.0584, over 19784.00 frames. ], tot_loss[loss=0.2051, simple_loss=0.2852, pruned_loss=0.06247, over 3817658.73 frames. ], batch size: 56, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:49:15,964 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=181996.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:49:20,366 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-182000.pt +2023-04-03 09:49:30,785 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=182007.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:49:39,962 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6331, 1.7463, 1.9819, 1.9590, 1.5145, 1.9438, 1.9659, 1.8670], + device='cuda:0'), covar=tensor([0.4230, 0.3782, 0.2127, 0.2392, 0.3898, 0.2272, 0.5302, 0.3541], + device='cuda:0'), in_proj_covar=tensor([0.0932, 0.1009, 0.0740, 0.0951, 0.0909, 0.0850, 0.0858, 0.0807], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 09:49:49,913 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-04-03 09:49:56,879 INFO [train.py:903] (0/4) Epoch 27, batch 4500, loss[loss=0.1861, simple_loss=0.2741, pruned_loss=0.04906, over 19665.00 frames. ], tot_loss[loss=0.2056, simple_loss=0.286, pruned_loss=0.06264, over 3822937.07 frames. ], batch size: 53, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:49:59,836 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=182030.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:50:31,958 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=182055.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:50:38,352 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.72 vs. limit=5.0 +2023-04-03 09:50:49,963 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.312e+02 5.059e+02 6.218e+02 7.785e+02 2.105e+03, threshold=1.244e+03, percent-clipped=5.0 +2023-04-03 09:51:00,203 INFO [train.py:903] (0/4) Epoch 27, batch 4550, loss[loss=0.1771, simple_loss=0.2478, pruned_loss=0.05318, over 19738.00 frames. ], tot_loss[loss=0.2042, simple_loss=0.2846, pruned_loss=0.06195, over 3827191.12 frames. ], batch size: 46, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:51:09,780 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-03 09:51:24,544 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0267, 1.4699, 1.8080, 1.7930, 4.5477, 1.2086, 2.8001, 5.0117], + device='cuda:0'), covar=tensor([0.0486, 0.2915, 0.2875, 0.1983, 0.0764, 0.2674, 0.1344, 0.0149], + device='cuda:0'), in_proj_covar=tensor([0.0423, 0.0378, 0.0396, 0.0353, 0.0384, 0.0357, 0.0394, 0.0416], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 09:51:32,201 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-03 09:52:02,921 INFO [train.py:903] (0/4) Epoch 27, batch 4600, loss[loss=0.1736, simple_loss=0.2575, pruned_loss=0.04486, over 19851.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2841, pruned_loss=0.06143, over 3822908.03 frames. ], batch size: 52, lr: 3.03e-03, grad_scale: 8.0 +2023-04-03 09:52:54,753 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.336e+02 4.830e+02 5.724e+02 7.323e+02 1.391e+03, threshold=1.145e+03, percent-clipped=2.0 +2023-04-03 09:53:05,189 INFO [train.py:903] (0/4) Epoch 27, batch 4650, loss[loss=0.1982, simple_loss=0.286, pruned_loss=0.05521, over 19515.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2838, pruned_loss=0.06079, over 3825023.00 frames. ], batch size: 54, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 09:53:22,614 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-03 09:53:34,163 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-03 09:54:07,711 INFO [train.py:903] (0/4) Epoch 27, batch 4700, loss[loss=0.1999, simple_loss=0.2783, pruned_loss=0.06073, over 19356.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2834, pruned_loss=0.06015, over 3819236.08 frames. ], batch size: 47, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 09:54:23,689 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2435, 2.0447, 2.0631, 1.9210, 1.5907, 1.8940, 0.6410, 1.3200], + device='cuda:0'), covar=tensor([0.0658, 0.0746, 0.0504, 0.0931, 0.1304, 0.0935, 0.1510, 0.1145], + device='cuda:0'), in_proj_covar=tensor([0.0365, 0.0363, 0.0369, 0.0391, 0.0469, 0.0395, 0.0347, 0.0348], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 09:54:30,872 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-03 09:54:51,531 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=182263.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:54:59,128 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.994e+02 4.730e+02 5.895e+02 7.660e+02 1.174e+03, threshold=1.179e+03, percent-clipped=2.0 +2023-04-03 09:55:06,179 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8752, 4.0256, 4.4000, 4.4019, 2.7908, 4.1023, 3.7896, 4.1694], + device='cuda:0'), covar=tensor([0.1403, 0.2668, 0.0618, 0.0707, 0.4263, 0.1255, 0.0589, 0.1015], + device='cuda:0'), in_proj_covar=tensor([0.0817, 0.0779, 0.0986, 0.0864, 0.0858, 0.0750, 0.0584, 0.0915], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 09:55:10,403 INFO [train.py:903] (0/4) Epoch 27, batch 4750, loss[loss=0.2065, simple_loss=0.2902, pruned_loss=0.06139, over 19530.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2843, pruned_loss=0.06068, over 3819021.94 frames. ], batch size: 54, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 09:55:22,528 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=182288.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:55:35,987 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=182299.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:56:12,231 INFO [train.py:903] (0/4) Epoch 27, batch 4800, loss[loss=0.1904, simple_loss=0.2852, pruned_loss=0.04782, over 19648.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2826, pruned_loss=0.0596, over 3829519.16 frames. ], batch size: 55, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 09:56:26,917 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=182340.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:57:01,655 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5547, 1.5118, 1.4571, 2.0063, 1.4166, 1.7827, 1.8525, 1.6247], + device='cuda:0'), covar=tensor([0.0872, 0.0944, 0.1069, 0.0731, 0.0934, 0.0802, 0.0860, 0.0755], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0224, 0.0227, 0.0240, 0.0226, 0.0212, 0.0187, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 09:57:03,562 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.957e+02 4.862e+02 5.780e+02 7.243e+02 1.108e+03, threshold=1.156e+03, percent-clipped=0.0 +2023-04-03 09:57:13,600 INFO [train.py:903] (0/4) Epoch 27, batch 4850, loss[loss=0.1783, simple_loss=0.2755, pruned_loss=0.04055, over 19666.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2836, pruned_loss=0.06037, over 3832954.92 frames. ], batch size: 55, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 09:57:36,904 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-03 09:57:50,863 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8416, 1.5600, 1.4826, 1.7888, 1.5336, 1.5981, 1.4901, 1.7133], + device='cuda:0'), covar=tensor([0.1108, 0.1319, 0.1533, 0.0992, 0.1301, 0.0586, 0.1512, 0.0794], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0358, 0.0317, 0.0257, 0.0307, 0.0255, 0.0320, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 09:57:58,344 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-03 09:58:03,903 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-03 09:58:03,929 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-03 09:58:13,248 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-03 09:58:14,419 INFO [train.py:903] (0/4) Epoch 27, batch 4900, loss[loss=0.2032, simple_loss=0.2828, pruned_loss=0.06187, over 19541.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2839, pruned_loss=0.06062, over 3822101.86 frames. ], batch size: 54, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 09:58:34,909 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-03 09:58:50,310 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=182455.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 09:59:07,219 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.920e+02 4.786e+02 5.871e+02 7.388e+02 1.622e+03, threshold=1.174e+03, percent-clipped=2.0 +2023-04-03 09:59:18,873 INFO [train.py:903] (0/4) Epoch 27, batch 4950, loss[loss=0.2058, simple_loss=0.2939, pruned_loss=0.05884, over 19312.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.283, pruned_loss=0.05998, over 3831013.78 frames. ], batch size: 66, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 09:59:36,556 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-03 10:00:00,867 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-03 10:00:21,846 INFO [train.py:903] (0/4) Epoch 27, batch 5000, loss[loss=0.1865, simple_loss=0.2749, pruned_loss=0.04905, over 19664.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2828, pruned_loss=0.05963, over 3844707.68 frames. ], batch size: 55, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:00:27,411 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=182532.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:00:32,564 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-03 10:00:44,441 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-03 10:01:15,165 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.165e+02 4.878e+02 5.976e+02 7.448e+02 1.686e+03, threshold=1.195e+03, percent-clipped=3.0 +2023-04-03 10:01:25,326 INFO [train.py:903] (0/4) Epoch 27, batch 5050, loss[loss=0.1993, simple_loss=0.2836, pruned_loss=0.05747, over 19539.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2827, pruned_loss=0.05956, over 3845750.41 frames. ], batch size: 56, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:02:02,775 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-03 10:02:27,489 INFO [train.py:903] (0/4) Epoch 27, batch 5100, loss[loss=0.1664, simple_loss=0.2572, pruned_loss=0.03778, over 19827.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2822, pruned_loss=0.05963, over 3849504.49 frames. ], batch size: 52, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:02:37,798 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-03 10:02:41,999 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-03 10:02:46,615 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-03 10:02:46,760 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=182643.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:03:14,678 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.59 vs. limit=5.0 +2023-04-03 10:03:16,996 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-03 10:03:19,757 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.602e+02 5.110e+02 6.408e+02 8.268e+02 2.195e+03, threshold=1.282e+03, percent-clipped=9.0 +2023-04-03 10:03:30,281 INFO [train.py:903] (0/4) Epoch 27, batch 5150, loss[loss=0.2484, simple_loss=0.319, pruned_loss=0.08886, over 17514.00 frames. ], tot_loss[loss=0.2, simple_loss=0.2813, pruned_loss=0.05933, over 3847718.71 frames. ], batch size: 101, lr: 3.02e-03, grad_scale: 4.0 +2023-04-03 10:03:44,257 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-03 10:03:45,806 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7195, 1.7316, 1.9626, 1.8846, 2.8165, 2.3678, 2.8887, 1.6112], + device='cuda:0'), covar=tensor([0.2350, 0.4090, 0.2723, 0.1954, 0.1442, 0.2178, 0.1448, 0.4315], + device='cuda:0'), in_proj_covar=tensor([0.0553, 0.0668, 0.0753, 0.0507, 0.0634, 0.0546, 0.0670, 0.0572], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 10:04:12,421 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=182711.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:04:21,110 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 10:04:25,033 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.21 vs. limit=5.0 +2023-04-03 10:04:30,866 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.05 vs. limit=5.0 +2023-04-03 10:04:34,470 INFO [train.py:903] (0/4) Epoch 27, batch 5200, loss[loss=0.2086, simple_loss=0.3008, pruned_loss=0.05825, over 19529.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2823, pruned_loss=0.05958, over 3851139.71 frames. ], batch size: 54, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:04:45,062 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=182736.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:04:50,583 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-03 10:05:06,003 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9289, 0.8873, 0.9118, 1.0246, 0.8019, 0.9883, 0.9541, 0.9536], + device='cuda:0'), covar=tensor([0.0721, 0.0742, 0.0815, 0.0512, 0.0980, 0.0657, 0.0827, 0.0620], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0225, 0.0229, 0.0241, 0.0226, 0.0214, 0.0188, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 10:05:11,577 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=182758.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:05:28,502 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.060e+02 4.726e+02 5.796e+02 7.282e+02 1.371e+03, threshold=1.159e+03, percent-clipped=1.0 +2023-04-03 10:05:35,310 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-03 10:05:37,679 INFO [train.py:903] (0/4) Epoch 27, batch 5250, loss[loss=0.1607, simple_loss=0.2479, pruned_loss=0.03679, over 19605.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.282, pruned_loss=0.0593, over 3852949.06 frames. ], batch size: 50, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:05:43,493 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.5993, 4.7541, 5.3391, 5.3213, 2.3906, 4.9778, 4.3724, 5.0441], + device='cuda:0'), covar=tensor([0.1621, 0.1611, 0.0549, 0.0637, 0.5683, 0.0997, 0.0583, 0.1088], + device='cuda:0'), in_proj_covar=tensor([0.0812, 0.0779, 0.0984, 0.0867, 0.0855, 0.0748, 0.0582, 0.0913], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 10:06:10,113 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3697, 4.0051, 2.6387, 3.5164, 0.8077, 3.9470, 3.8136, 3.9188], + device='cuda:0'), covar=tensor([0.0725, 0.1176, 0.1986, 0.0926, 0.4059, 0.0800, 0.1034, 0.1207], + device='cuda:0'), in_proj_covar=tensor([0.0530, 0.0427, 0.0515, 0.0356, 0.0410, 0.0456, 0.0451, 0.0479], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:06:31,765 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6435, 2.3690, 1.7955, 1.5829, 2.1339, 1.4186, 1.4839, 2.0337], + device='cuda:0'), covar=tensor([0.1120, 0.0842, 0.1087, 0.0975, 0.0639, 0.1409, 0.0813, 0.0525], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0320, 0.0337, 0.0273, 0.0251, 0.0345, 0.0292, 0.0279], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:06:39,519 INFO [train.py:903] (0/4) Epoch 27, batch 5300, loss[loss=0.1802, simple_loss=0.2663, pruned_loss=0.04704, over 19848.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2817, pruned_loss=0.05906, over 3833777.39 frames. ], batch size: 52, lr: 3.02e-03, grad_scale: 4.0 +2023-04-03 10:06:57,461 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-03 10:07:07,883 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6075, 1.4871, 1.5305, 2.1474, 1.5845, 1.8658, 1.9782, 1.7133], + device='cuda:0'), covar=tensor([0.0885, 0.0933, 0.1013, 0.0703, 0.0863, 0.0772, 0.0807, 0.0722], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0224, 0.0228, 0.0240, 0.0224, 0.0213, 0.0187, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 10:07:34,150 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.440e+02 4.985e+02 5.848e+02 7.587e+02 2.195e+03, threshold=1.170e+03, percent-clipped=4.0 +2023-04-03 10:07:36,641 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=182874.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:07:39,954 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=182876.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:07:42,166 INFO [train.py:903] (0/4) Epoch 27, batch 5350, loss[loss=0.1923, simple_loss=0.2718, pruned_loss=0.05637, over 19601.00 frames. ], tot_loss[loss=0.2, simple_loss=0.282, pruned_loss=0.05897, over 3838945.87 frames. ], batch size: 50, lr: 3.02e-03, grad_scale: 4.0 +2023-04-03 10:07:44,853 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=182880.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:08:15,011 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6158, 1.7286, 1.9375, 1.8678, 2.7111, 2.3203, 2.8410, 1.3885], + device='cuda:0'), covar=tensor([0.2373, 0.4120, 0.2641, 0.1847, 0.1487, 0.2143, 0.1361, 0.4294], + device='cuda:0'), in_proj_covar=tensor([0.0554, 0.0669, 0.0754, 0.0506, 0.0636, 0.0546, 0.0669, 0.0573], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 10:08:16,926 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-03 10:08:46,442 INFO [train.py:903] (0/4) Epoch 27, batch 5400, loss[loss=0.2126, simple_loss=0.2868, pruned_loss=0.06916, over 19581.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2824, pruned_loss=0.05936, over 3838276.44 frames. ], batch size: 52, lr: 3.02e-03, grad_scale: 4.0 +2023-04-03 10:08:59,335 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2676, 1.2557, 1.2799, 1.3789, 1.0133, 1.3519, 1.3208, 1.3277], + device='cuda:0'), covar=tensor([0.0931, 0.0993, 0.1042, 0.0682, 0.0892, 0.0875, 0.0892, 0.0784], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0223, 0.0226, 0.0239, 0.0224, 0.0212, 0.0186, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 10:09:41,329 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.624e+02 4.932e+02 6.009e+02 7.341e+02 1.388e+03, threshold=1.202e+03, percent-clipped=2.0 +2023-04-03 10:09:49,115 INFO [train.py:903] (0/4) Epoch 27, batch 5450, loss[loss=0.1749, simple_loss=0.2566, pruned_loss=0.04657, over 19740.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2825, pruned_loss=0.05915, over 3837601.17 frames. ], batch size: 51, lr: 3.02e-03, grad_scale: 4.0 +2023-04-03 10:10:04,470 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=182991.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:10:35,621 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=183014.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:10:41,299 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2723, 1.2534, 1.3068, 1.3962, 1.0662, 1.3185, 1.3692, 1.3202], + device='cuda:0'), covar=tensor([0.0926, 0.0975, 0.1069, 0.0670, 0.0859, 0.0900, 0.0863, 0.0821], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0223, 0.0227, 0.0240, 0.0225, 0.0213, 0.0187, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 10:10:50,992 INFO [train.py:903] (0/4) Epoch 27, batch 5500, loss[loss=0.239, simple_loss=0.3175, pruned_loss=0.08019, over 18444.00 frames. ], tot_loss[loss=0.2, simple_loss=0.2819, pruned_loss=0.05908, over 3830125.65 frames. ], batch size: 83, lr: 3.02e-03, grad_scale: 4.0 +2023-04-03 10:11:04,312 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=183039.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:11:13,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-03 10:11:45,359 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.090e+02 5.032e+02 6.184e+02 7.491e+02 1.557e+03, threshold=1.237e+03, percent-clipped=5.0 +2023-04-03 10:11:52,119 INFO [train.py:903] (0/4) Epoch 27, batch 5550, loss[loss=0.22, simple_loss=0.2986, pruned_loss=0.07069, over 19667.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2822, pruned_loss=0.05968, over 3828142.03 frames. ], batch size: 58, lr: 3.02e-03, grad_scale: 4.0 +2023-04-03 10:11:57,917 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-03 10:12:15,122 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.67 vs. limit=2.0 +2023-04-03 10:12:46,687 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-03 10:12:56,193 INFO [train.py:903] (0/4) Epoch 27, batch 5600, loss[loss=0.2198, simple_loss=0.2892, pruned_loss=0.0752, over 19780.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.281, pruned_loss=0.0594, over 3823710.67 frames. ], batch size: 46, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:13:51,540 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.195e+02 4.790e+02 6.069e+02 7.863e+02 1.689e+03, threshold=1.214e+03, percent-clipped=3.0 +2023-04-03 10:13:59,318 INFO [train.py:903] (0/4) Epoch 27, batch 5650, loss[loss=0.2155, simple_loss=0.2997, pruned_loss=0.0657, over 19388.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2814, pruned_loss=0.05967, over 3824087.43 frames. ], batch size: 70, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:14:44,839 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-03 10:14:49,504 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=183218.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:14:56,368 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=183224.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:14:58,844 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0154, 3.6690, 2.3356, 3.2279, 0.7440, 3.6377, 3.5027, 3.6051], + device='cuda:0'), covar=tensor([0.0732, 0.1049, 0.2118, 0.0977, 0.4057, 0.0772, 0.0948, 0.1190], + device='cuda:0'), in_proj_covar=tensor([0.0525, 0.0423, 0.0511, 0.0355, 0.0406, 0.0450, 0.0447, 0.0475], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:15:01,003 INFO [train.py:903] (0/4) Epoch 27, batch 5700, loss[loss=0.1744, simple_loss=0.2553, pruned_loss=0.04673, over 19406.00 frames. ], tot_loss[loss=0.2, simple_loss=0.281, pruned_loss=0.0595, over 3830581.43 frames. ], batch size: 48, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:15:25,118 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=183247.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:15:54,498 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.213e+02 5.103e+02 5.954e+02 7.593e+02 1.308e+03, threshold=1.191e+03, percent-clipped=1.0 +2023-04-03 10:15:54,885 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=183272.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:15:59,208 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-03 10:16:01,600 INFO [train.py:903] (0/4) Epoch 27, batch 5750, loss[loss=0.2153, simple_loss=0.2971, pruned_loss=0.06676, over 19582.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2817, pruned_loss=0.05943, over 3832515.05 frames. ], batch size: 61, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:16:08,302 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-03 10:16:13,727 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-03 10:16:16,169 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6291, 4.2492, 2.8159, 3.8004, 0.9777, 4.2738, 4.0772, 4.1995], + device='cuda:0'), covar=tensor([0.0660, 0.1051, 0.1787, 0.0815, 0.3913, 0.0616, 0.0880, 0.1069], + device='cuda:0'), in_proj_covar=tensor([0.0524, 0.0422, 0.0510, 0.0354, 0.0406, 0.0449, 0.0446, 0.0474], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:16:50,192 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5946, 1.5416, 2.1467, 1.8624, 3.1026, 4.7798, 4.6250, 5.1312], + device='cuda:0'), covar=tensor([0.1474, 0.3709, 0.3116, 0.2109, 0.0631, 0.0208, 0.0161, 0.0166], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0332, 0.0363, 0.0271, 0.0255, 0.0196, 0.0219, 0.0278], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 10:17:05,411 INFO [train.py:903] (0/4) Epoch 27, batch 5800, loss[loss=0.2002, simple_loss=0.28, pruned_loss=0.06024, over 19619.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2823, pruned_loss=0.05936, over 3848766.51 frames. ], batch size: 50, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:17:11,564 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=183333.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:17:19,544 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=183339.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:17:59,580 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.172e+02 5.364e+02 7.027e+02 9.052e+02 1.891e+03, threshold=1.405e+03, percent-clipped=4.0 +2023-04-03 10:18:07,742 INFO [train.py:903] (0/4) Epoch 27, batch 5850, loss[loss=0.1825, simple_loss=0.2723, pruned_loss=0.04632, over 19531.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2825, pruned_loss=0.05994, over 3843447.82 frames. ], batch size: 56, lr: 3.02e-03, grad_scale: 8.0 +2023-04-03 10:18:29,912 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9638, 1.1907, 1.5365, 1.0844, 2.1302, 2.9398, 2.7380, 3.3042], + device='cuda:0'), covar=tensor([0.2024, 0.5246, 0.4678, 0.2877, 0.0872, 0.0332, 0.0338, 0.0392], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0331, 0.0363, 0.0271, 0.0254, 0.0196, 0.0219, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 10:19:07,134 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=183426.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:19:08,994 INFO [train.py:903] (0/4) Epoch 27, batch 5900, loss[loss=0.1996, simple_loss=0.2837, pruned_loss=0.05777, over 19687.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2822, pruned_loss=0.0598, over 3836527.98 frames. ], batch size: 59, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:19:09,051 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-03 10:19:32,075 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-03 10:20:03,305 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.958e+02 4.868e+02 5.663e+02 7.388e+02 1.454e+03, threshold=1.133e+03, percent-clipped=1.0 +2023-04-03 10:20:10,247 INFO [train.py:903] (0/4) Epoch 27, batch 5950, loss[loss=0.2105, simple_loss=0.2993, pruned_loss=0.06087, over 19608.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2832, pruned_loss=0.06035, over 3828613.24 frames. ], batch size: 57, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:20:40,735 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8989, 4.5583, 3.3808, 3.8110, 1.9392, 4.5213, 4.3512, 4.4869], + device='cuda:0'), covar=tensor([0.0501, 0.0848, 0.1750, 0.0895, 0.3028, 0.0641, 0.0954, 0.1334], + device='cuda:0'), in_proj_covar=tensor([0.0528, 0.0427, 0.0515, 0.0357, 0.0410, 0.0454, 0.0450, 0.0479], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:20:58,222 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0636, 1.3758, 1.7984, 1.2271, 2.7899, 3.6158, 3.3274, 3.8037], + device='cuda:0'), covar=tensor([0.1653, 0.3705, 0.3269, 0.2572, 0.0546, 0.0186, 0.0200, 0.0257], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0331, 0.0363, 0.0271, 0.0254, 0.0196, 0.0219, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 10:21:12,623 INFO [train.py:903] (0/4) Epoch 27, batch 6000, loss[loss=0.1902, simple_loss=0.2723, pruned_loss=0.05401, over 19594.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2834, pruned_loss=0.06051, over 3814030.74 frames. ], batch size: 52, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:21:12,624 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 10:21:25,594 INFO [train.py:937] (0/4) Epoch 27, validation: loss=0.1675, simple_loss=0.2669, pruned_loss=0.03401, over 944034.00 frames. +2023-04-03 10:21:25,595 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 10:22:09,199 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1767, 1.3344, 1.7393, 1.3865, 2.8771, 3.7089, 3.4286, 3.9007], + device='cuda:0'), covar=tensor([0.1702, 0.3867, 0.3457, 0.2445, 0.0573, 0.0191, 0.0208, 0.0272], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0332, 0.0363, 0.0271, 0.0254, 0.0196, 0.0219, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 10:22:22,440 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.170e+02 5.376e+02 6.105e+02 7.773e+02 1.848e+03, threshold=1.221e+03, percent-clipped=6.0 +2023-04-03 10:22:28,428 INFO [train.py:903] (0/4) Epoch 27, batch 6050, loss[loss=0.2098, simple_loss=0.287, pruned_loss=0.06629, over 19726.00 frames. ], tot_loss[loss=0.2024, simple_loss=0.2833, pruned_loss=0.06072, over 3815990.22 frames. ], batch size: 63, lr: 3.01e-03, grad_scale: 4.0 +2023-04-03 10:22:43,106 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=183589.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:22:50,962 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=183595.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:23:14,126 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=183614.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:23:20,843 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=183620.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:23:31,579 INFO [train.py:903] (0/4) Epoch 27, batch 6100, loss[loss=0.2026, simple_loss=0.2901, pruned_loss=0.05751, over 19661.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2847, pruned_loss=0.06118, over 3809077.72 frames. ], batch size: 55, lr: 3.01e-03, grad_scale: 4.0 +2023-04-03 10:23:37,150 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=183632.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:24:27,769 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.311e+02 5.176e+02 6.035e+02 7.763e+02 1.396e+03, threshold=1.207e+03, percent-clipped=4.0 +2023-04-03 10:24:33,563 INFO [train.py:903] (0/4) Epoch 27, batch 6150, loss[loss=0.1759, simple_loss=0.2501, pruned_loss=0.05083, over 19287.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2845, pruned_loss=0.06099, over 3805304.33 frames. ], batch size: 44, lr: 3.01e-03, grad_scale: 4.0 +2023-04-03 10:25:01,302 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-03 10:25:35,001 INFO [train.py:903] (0/4) Epoch 27, batch 6200, loss[loss=0.2475, simple_loss=0.3225, pruned_loss=0.08619, over 19672.00 frames. ], tot_loss[loss=0.2044, simple_loss=0.2853, pruned_loss=0.06173, over 3791215.02 frames. ], batch size: 59, lr: 3.01e-03, grad_scale: 4.0 +2023-04-03 10:26:22,203 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.41 vs. limit=2.0 +2023-04-03 10:26:27,301 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=183770.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:26:31,934 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.388e+02 4.839e+02 5.826e+02 7.637e+02 1.855e+03, threshold=1.165e+03, percent-clipped=4.0 +2023-04-03 10:26:37,744 INFO [train.py:903] (0/4) Epoch 27, batch 6250, loss[loss=0.1809, simple_loss=0.2714, pruned_loss=0.04524, over 19790.00 frames. ], tot_loss[loss=0.2052, simple_loss=0.2861, pruned_loss=0.06208, over 3790355.40 frames. ], batch size: 56, lr: 3.01e-03, grad_scale: 4.0 +2023-04-03 10:27:08,104 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-03 10:27:40,819 INFO [train.py:903] (0/4) Epoch 27, batch 6300, loss[loss=0.1913, simple_loss=0.2684, pruned_loss=0.05713, over 19323.00 frames. ], tot_loss[loss=0.2055, simple_loss=0.2867, pruned_loss=0.0621, over 3804587.63 frames. ], batch size: 44, lr: 3.01e-03, grad_scale: 4.0 +2023-04-03 10:28:20,269 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=183860.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:28:36,676 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.559e+02 4.928e+02 5.882e+02 7.199e+02 1.705e+03, threshold=1.176e+03, percent-clipped=2.0 +2023-04-03 10:28:43,614 INFO [train.py:903] (0/4) Epoch 27, batch 6350, loss[loss=0.2401, simple_loss=0.3084, pruned_loss=0.08587, over 19646.00 frames. ], tot_loss[loss=0.2043, simple_loss=0.2857, pruned_loss=0.06147, over 3804679.03 frames. ], batch size: 58, lr: 3.01e-03, grad_scale: 4.0 +2023-04-03 10:28:51,872 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=183885.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:29:43,984 INFO [train.py:903] (0/4) Epoch 27, batch 6400, loss[loss=0.2247, simple_loss=0.3029, pruned_loss=0.07319, over 18837.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2851, pruned_loss=0.06127, over 3807978.22 frames. ], batch size: 74, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:30:39,972 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.119e+02 4.757e+02 6.035e+02 7.575e+02 1.901e+03, threshold=1.207e+03, percent-clipped=7.0 +2023-04-03 10:30:43,678 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=183976.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:30:45,010 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1826, 1.3423, 1.6876, 1.4002, 2.6797, 3.7547, 3.4338, 3.8966], + device='cuda:0'), covar=tensor([0.1735, 0.4043, 0.3717, 0.2648, 0.0697, 0.0210, 0.0220, 0.0306], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0331, 0.0363, 0.0270, 0.0253, 0.0195, 0.0219, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 10:30:45,873 INFO [train.py:903] (0/4) Epoch 27, batch 6450, loss[loss=0.2154, simple_loss=0.2973, pruned_loss=0.06676, over 19665.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2851, pruned_loss=0.06127, over 3801610.28 frames. ], batch size: 60, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:31:14,613 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-184000.pt +2023-04-03 10:31:28,486 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-03 10:31:42,559 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3563, 1.9417, 1.5193, 1.3555, 1.8070, 1.2590, 1.3153, 1.8240], + device='cuda:0'), covar=tensor([0.0907, 0.0818, 0.1062, 0.0887, 0.0581, 0.1320, 0.0695, 0.0463], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0320, 0.0337, 0.0273, 0.0251, 0.0344, 0.0293, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:31:50,539 INFO [train.py:903] (0/4) Epoch 27, batch 6500, loss[loss=0.1972, simple_loss=0.2836, pruned_loss=0.05545, over 19332.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2839, pruned_loss=0.06052, over 3805794.41 frames. ], batch size: 70, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:31:52,978 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-03 10:32:04,467 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1577, 1.2754, 1.5237, 1.5075, 2.7371, 1.2421, 2.2699, 3.1038], + device='cuda:0'), covar=tensor([0.0593, 0.2943, 0.3010, 0.1818, 0.0727, 0.2306, 0.1258, 0.0342], + device='cuda:0'), in_proj_covar=tensor([0.0424, 0.0378, 0.0397, 0.0354, 0.0385, 0.0357, 0.0395, 0.0415], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:32:11,849 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-04-03 10:32:40,201 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=184068.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:32:46,404 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.099e+02 4.542e+02 5.565e+02 7.286e+02 1.442e+03, threshold=1.113e+03, percent-clipped=3.0 +2023-04-03 10:32:53,094 INFO [train.py:903] (0/4) Epoch 27, batch 6550, loss[loss=0.1728, simple_loss=0.2545, pruned_loss=0.04553, over 19363.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2836, pruned_loss=0.06017, over 3821600.03 frames. ], batch size: 47, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:33:08,666 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=184091.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:33:24,621 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5266, 1.5685, 1.7998, 1.7223, 2.4400, 2.1856, 2.5809, 1.2078], + device='cuda:0'), covar=tensor([0.2426, 0.4266, 0.2701, 0.1980, 0.1614, 0.2192, 0.1480, 0.4740], + device='cuda:0'), in_proj_covar=tensor([0.0552, 0.0669, 0.0751, 0.0507, 0.0636, 0.0544, 0.0666, 0.0571], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 10:33:51,900 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4551, 1.4752, 1.8056, 1.6781, 3.1142, 4.4185, 4.1807, 4.7578], + device='cuda:0'), covar=tensor([0.1577, 0.3899, 0.3639, 0.2459, 0.0704, 0.0206, 0.0206, 0.0241], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0332, 0.0363, 0.0271, 0.0253, 0.0195, 0.0220, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 10:33:55,011 INFO [train.py:903] (0/4) Epoch 27, batch 6600, loss[loss=0.2068, simple_loss=0.291, pruned_loss=0.06128, over 19662.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2834, pruned_loss=0.06022, over 3809352.21 frames. ], batch size: 55, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:34:05,975 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0232, 2.1176, 2.3753, 2.6061, 2.0155, 2.5060, 2.2788, 2.1400], + device='cuda:0'), covar=tensor([0.4493, 0.4597, 0.2061, 0.2735, 0.4599, 0.2401, 0.5474, 0.3620], + device='cuda:0'), in_proj_covar=tensor([0.0933, 0.1009, 0.0740, 0.0947, 0.0909, 0.0846, 0.0859, 0.0803], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 10:34:11,860 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=184141.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:34:43,132 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=184166.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:34:46,840 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.25 vs. limit=2.0 +2023-04-03 10:34:46,856 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.17 vs. limit=2.0 +2023-04-03 10:34:50,799 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.027e+02 4.833e+02 5.798e+02 7.178e+02 1.551e+03, threshold=1.160e+03, percent-clipped=2.0 +2023-04-03 10:34:58,096 INFO [train.py:903] (0/4) Epoch 27, batch 6650, loss[loss=0.2081, simple_loss=0.281, pruned_loss=0.06761, over 19754.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2823, pruned_loss=0.05963, over 3792205.65 frames. ], batch size: 54, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:35:30,807 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=184204.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:35:59,893 INFO [train.py:903] (0/4) Epoch 27, batch 6700, loss[loss=0.1738, simple_loss=0.2555, pruned_loss=0.04604, over 19416.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2827, pruned_loss=0.06005, over 3797938.72 frames. ], batch size: 48, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:36:52,206 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.305e+02 5.122e+02 6.084e+02 8.596e+02 2.606e+03, threshold=1.217e+03, percent-clipped=9.0 +2023-04-03 10:36:57,985 INFO [train.py:903] (0/4) Epoch 27, batch 6750, loss[loss=0.202, simple_loss=0.2887, pruned_loss=0.05764, over 19324.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2824, pruned_loss=0.0596, over 3808010.53 frames. ], batch size: 66, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:37:31,840 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3612, 3.0921, 2.2781, 2.7541, 0.9345, 3.0565, 2.9193, 3.0499], + device='cuda:0'), covar=tensor([0.1079, 0.1231, 0.2046, 0.1086, 0.3657, 0.1011, 0.1208, 0.1320], + device='cuda:0'), in_proj_covar=tensor([0.0526, 0.0425, 0.0516, 0.0357, 0.0407, 0.0455, 0.0451, 0.0479], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:37:44,327 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=184319.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:37:54,077 INFO [train.py:903] (0/4) Epoch 27, batch 6800, loss[loss=0.226, simple_loss=0.2994, pruned_loss=0.07629, over 19543.00 frames. ], tot_loss[loss=0.2029, simple_loss=0.284, pruned_loss=0.06085, over 3796810.62 frames. ], batch size: 54, lr: 3.01e-03, grad_scale: 8.0 +2023-04-03 10:38:15,897 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=184347.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:38:24,982 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-27.pt +2023-04-03 10:38:39,920 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-03 10:38:40,383 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-03 10:38:43,611 INFO [train.py:903] (0/4) Epoch 28, batch 0, loss[loss=0.2302, simple_loss=0.2995, pruned_loss=0.0805, over 19835.00 frames. ], tot_loss[loss=0.2302, simple_loss=0.2995, pruned_loss=0.0805, over 19835.00 frames. ], batch size: 52, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:38:43,611 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 10:38:54,476 INFO [train.py:937] (0/4) Epoch 28, validation: loss=0.1665, simple_loss=0.2666, pruned_loss=0.03316, over 944034.00 frames. +2023-04-03 10:38:54,477 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 10:39:08,322 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-03 10:39:14,442 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=184372.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:39:15,175 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.715e+02 5.190e+02 6.304e+02 8.212e+02 1.288e+03, threshold=1.261e+03, percent-clipped=2.0 +2023-04-03 10:39:21,653 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=184377.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:39:57,733 INFO [train.py:903] (0/4) Epoch 28, batch 50, loss[loss=0.1791, simple_loss=0.2683, pruned_loss=0.04495, over 19674.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.285, pruned_loss=0.05902, over 856898.83 frames. ], batch size: 58, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:40:04,834 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=184412.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:40:25,827 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6641, 1.7877, 2.0181, 1.8714, 3.0022, 2.5583, 3.2618, 1.6646], + device='cuda:0'), covar=tensor([0.2587, 0.4418, 0.2925, 0.2044, 0.1722, 0.2282, 0.1668, 0.4669], + device='cuda:0'), in_proj_covar=tensor([0.0553, 0.0668, 0.0751, 0.0506, 0.0636, 0.0544, 0.0668, 0.0571], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 10:40:32,255 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-03 10:40:58,080 INFO [train.py:903] (0/4) Epoch 28, batch 100, loss[loss=0.2028, simple_loss=0.2905, pruned_loss=0.05754, over 18641.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2835, pruned_loss=0.05894, over 1527832.01 frames. ], batch size: 74, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:41:08,314 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-03 10:41:18,601 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.133e+02 4.572e+02 5.776e+02 7.316e+02 1.195e+03, threshold=1.155e+03, percent-clipped=0.0 +2023-04-03 10:41:56,017 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=184504.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:41:58,109 INFO [train.py:903] (0/4) Epoch 28, batch 150, loss[loss=0.1666, simple_loss=0.2453, pruned_loss=0.04396, over 19376.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2849, pruned_loss=0.06031, over 2026633.38 frames. ], batch size: 47, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:42:24,575 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=184527.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:42:57,425 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-03 10:42:58,585 INFO [train.py:903] (0/4) Epoch 28, batch 200, loss[loss=0.2132, simple_loss=0.3005, pruned_loss=0.06298, over 19326.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2822, pruned_loss=0.05885, over 2428469.46 frames. ], batch size: 66, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:43:19,482 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.222e+02 4.969e+02 6.258e+02 7.516e+02 2.266e+03, threshold=1.252e+03, percent-clipped=4.0 +2023-04-03 10:43:22,307 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=184575.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:43:24,676 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9166, 1.7128, 1.6118, 1.9195, 1.6212, 1.6111, 1.6030, 1.8178], + device='cuda:0'), covar=tensor([0.1144, 0.1513, 0.1571, 0.1004, 0.1384, 0.0654, 0.1550, 0.0861], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0360, 0.0320, 0.0258, 0.0307, 0.0258, 0.0321, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:43:51,902 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=184600.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:43:59,857 INFO [train.py:903] (0/4) Epoch 28, batch 250, loss[loss=0.2538, simple_loss=0.3095, pruned_loss=0.09912, over 19746.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2825, pruned_loss=0.05915, over 2749354.62 frames. ], batch size: 45, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:45:01,673 INFO [train.py:903] (0/4) Epoch 28, batch 300, loss[loss=0.234, simple_loss=0.3114, pruned_loss=0.07833, over 19416.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2828, pruned_loss=0.05913, over 2978788.98 frames. ], batch size: 70, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:45:22,244 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.089e+02 4.936e+02 6.396e+02 8.049e+02 1.564e+03, threshold=1.279e+03, percent-clipped=7.0 +2023-04-03 10:45:29,100 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=2.00 vs. limit=2.0 +2023-04-03 10:45:40,681 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8497, 1.1396, 1.4755, 0.6097, 2.0564, 2.4397, 2.2003, 2.6326], + device='cuda:0'), covar=tensor([0.1768, 0.4184, 0.3803, 0.3064, 0.0678, 0.0306, 0.0364, 0.0401], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0333, 0.0364, 0.0272, 0.0255, 0.0196, 0.0220, 0.0278], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 10:46:02,781 INFO [train.py:903] (0/4) Epoch 28, batch 350, loss[loss=0.1869, simple_loss=0.2745, pruned_loss=0.04959, over 19789.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2833, pruned_loss=0.05986, over 3177799.25 frames. ], batch size: 56, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:46:06,328 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 10:46:11,143 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2393, 1.4154, 2.0355, 1.6360, 3.1516, 4.6311, 4.5217, 5.0500], + device='cuda:0'), covar=tensor([0.1725, 0.4033, 0.3446, 0.2479, 0.0651, 0.0209, 0.0172, 0.0200], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0334, 0.0364, 0.0272, 0.0255, 0.0196, 0.0220, 0.0278], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 10:46:20,173 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=184721.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:47:04,871 INFO [train.py:903] (0/4) Epoch 28, batch 400, loss[loss=0.2158, simple_loss=0.2997, pruned_loss=0.06595, over 19799.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.284, pruned_loss=0.06045, over 3306032.37 frames. ], batch size: 56, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:47:07,992 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-04-03 10:47:24,994 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.223e+02 4.910e+02 5.909e+02 7.518e+02 1.907e+03, threshold=1.182e+03, percent-clipped=3.0 +2023-04-03 10:47:33,301 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6602, 1.4712, 1.4550, 1.5692, 3.2136, 1.2498, 2.4829, 3.7426], + device='cuda:0'), covar=tensor([0.0455, 0.2788, 0.3025, 0.1881, 0.0663, 0.2462, 0.1249, 0.0211], + device='cuda:0'), in_proj_covar=tensor([0.0426, 0.0380, 0.0398, 0.0354, 0.0385, 0.0359, 0.0397, 0.0417], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:47:38,193 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=184783.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:47:54,338 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9626, 1.9103, 1.6539, 2.0020, 1.7465, 1.6618, 1.7027, 1.8650], + device='cuda:0'), covar=tensor([0.1233, 0.1521, 0.1661, 0.1130, 0.1473, 0.0701, 0.1542, 0.0868], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0359, 0.0318, 0.0258, 0.0306, 0.0257, 0.0321, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:48:05,124 INFO [train.py:903] (0/4) Epoch 28, batch 450, loss[loss=0.233, simple_loss=0.3148, pruned_loss=0.07561, over 19749.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2835, pruned_loss=0.06058, over 3426538.26 frames. ], batch size: 63, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:48:07,816 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=184808.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:48:20,508 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9156, 2.9380, 2.6957, 2.8686, 2.6586, 2.4895, 2.4533, 2.9726], + device='cuda:0'), covar=tensor([0.0870, 0.1324, 0.1214, 0.1085, 0.1315, 0.0483, 0.1296, 0.0588], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0358, 0.0318, 0.0257, 0.0306, 0.0256, 0.0320, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:48:32,957 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8649, 1.5048, 1.6093, 1.7013, 3.4534, 1.2964, 2.5537, 3.9423], + device='cuda:0'), covar=tensor([0.0470, 0.2759, 0.2862, 0.1809, 0.0661, 0.2527, 0.1295, 0.0201], + device='cuda:0'), in_proj_covar=tensor([0.0424, 0.0379, 0.0398, 0.0353, 0.0384, 0.0358, 0.0396, 0.0417], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:48:38,474 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-03 10:48:39,492 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-03 10:48:43,479 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=184836.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:48:56,804 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=184848.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:49:06,863 INFO [train.py:903] (0/4) Epoch 28, batch 500, loss[loss=0.197, simple_loss=0.2764, pruned_loss=0.05879, over 19741.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2824, pruned_loss=0.05996, over 3527426.94 frames. ], batch size: 45, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:49:28,419 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.110e+02 5.109e+02 6.404e+02 8.256e+02 1.456e+03, threshold=1.281e+03, percent-clipped=5.0 +2023-04-03 10:49:33,528 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4184, 1.4229, 1.5993, 1.5692, 1.7436, 1.9092, 1.8071, 0.5281], + device='cuda:0'), covar=tensor([0.2344, 0.4266, 0.2665, 0.1903, 0.1663, 0.2226, 0.1433, 0.4999], + device='cuda:0'), in_proj_covar=tensor([0.0551, 0.0666, 0.0750, 0.0505, 0.0635, 0.0543, 0.0666, 0.0569], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 10:49:41,486 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=184884.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:50:02,657 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1148, 2.0636, 1.9279, 1.7271, 1.6072, 1.6801, 0.6507, 1.0781], + device='cuda:0'), covar=tensor([0.0721, 0.0668, 0.0512, 0.0864, 0.1305, 0.1002, 0.1394, 0.1130], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0365, 0.0371, 0.0395, 0.0472, 0.0397, 0.0347, 0.0351], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 10:50:09,150 INFO [train.py:903] (0/4) Epoch 28, batch 550, loss[loss=0.1767, simple_loss=0.2531, pruned_loss=0.05011, over 19758.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2835, pruned_loss=0.06031, over 3577063.19 frames. ], batch size: 48, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:51:08,835 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.48 vs. limit=5.0 +2023-04-03 10:51:11,428 INFO [train.py:903] (0/4) Epoch 28, batch 600, loss[loss=0.2213, simple_loss=0.2855, pruned_loss=0.07855, over 19779.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2834, pruned_loss=0.06079, over 3616634.08 frames. ], batch size: 47, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:51:12,923 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=184957.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:51:19,923 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=184963.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:51:31,268 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.601e+02 5.007e+02 6.276e+02 8.222e+02 1.849e+03, threshold=1.255e+03, percent-clipped=3.0 +2023-04-03 10:51:50,617 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-03 10:52:14,355 INFO [train.py:903] (0/4) Epoch 28, batch 650, loss[loss=0.2024, simple_loss=0.2868, pruned_loss=0.05898, over 17955.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2833, pruned_loss=0.06046, over 3675644.63 frames. ], batch size: 83, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:53:16,120 INFO [train.py:903] (0/4) Epoch 28, batch 700, loss[loss=0.2031, simple_loss=0.2827, pruned_loss=0.0618, over 19569.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2832, pruned_loss=0.06047, over 3697544.46 frames. ], batch size: 52, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:53:38,016 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.143e+02 4.589e+02 5.550e+02 7.126e+02 1.317e+03, threshold=1.110e+03, percent-clipped=1.0 +2023-04-03 10:53:45,140 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=185078.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:54:01,386 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=185092.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:54:19,786 INFO [train.py:903] (0/4) Epoch 28, batch 750, loss[loss=0.2518, simple_loss=0.3223, pruned_loss=0.09069, over 19538.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2838, pruned_loss=0.06105, over 3728504.63 frames. ], batch size: 64, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:54:34,023 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=185117.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:55:10,725 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=185147.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:55:20,945 INFO [train.py:903] (0/4) Epoch 28, batch 800, loss[loss=0.1764, simple_loss=0.2555, pruned_loss=0.04862, over 19479.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2844, pruned_loss=0.0614, over 3749375.38 frames. ], batch size: 49, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:55:30,809 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.53 vs. limit=2.0 +2023-04-03 10:55:34,897 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-03 10:55:41,840 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.136e+02 5.493e+02 6.558e+02 7.858e+02 2.224e+03, threshold=1.312e+03, percent-clipped=8.0 +2023-04-03 10:56:22,057 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8253, 4.3980, 2.8369, 3.8809, 0.9508, 4.4315, 4.2037, 4.3266], + device='cuda:0'), covar=tensor([0.0637, 0.0993, 0.1972, 0.0888, 0.4028, 0.0612, 0.1007, 0.1236], + device='cuda:0'), in_proj_covar=tensor([0.0527, 0.0426, 0.0517, 0.0359, 0.0409, 0.0456, 0.0451, 0.0480], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:56:24,245 INFO [train.py:903] (0/4) Epoch 28, batch 850, loss[loss=0.1841, simple_loss=0.2587, pruned_loss=0.05474, over 19733.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2835, pruned_loss=0.06026, over 3772259.52 frames. ], batch size: 47, lr: 2.95e-03, grad_scale: 8.0 +2023-04-03 10:56:39,480 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=185219.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:56:51,612 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=185228.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:57:11,559 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=185244.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:57:15,600 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-03 10:57:24,911 INFO [train.py:903] (0/4) Epoch 28, batch 900, loss[loss=0.1927, simple_loss=0.2702, pruned_loss=0.05765, over 19744.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2832, pruned_loss=0.06037, over 3794963.60 frames. ], batch size: 45, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 10:57:37,060 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-03 10:57:47,702 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.300e+02 4.798e+02 5.781e+02 7.336e+02 1.381e+03, threshold=1.156e+03, percent-clipped=1.0 +2023-04-03 10:58:21,541 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=185301.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:58:28,121 INFO [train.py:903] (0/4) Epoch 28, batch 950, loss[loss=0.1897, simple_loss=0.2826, pruned_loss=0.04847, over 19682.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2822, pruned_loss=0.05957, over 3810759.86 frames. ], batch size: 59, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 10:58:29,324 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-03 10:59:11,068 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2924, 1.9948, 1.5251, 1.3408, 1.8245, 1.2558, 1.1518, 1.7949], + device='cuda:0'), covar=tensor([0.0941, 0.0743, 0.1073, 0.0820, 0.0514, 0.1266, 0.0732, 0.0447], + device='cuda:0'), in_proj_covar=tensor([0.0299, 0.0317, 0.0335, 0.0271, 0.0249, 0.0341, 0.0290, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:59:14,453 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=185343.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 10:59:32,239 INFO [train.py:903] (0/4) Epoch 28, batch 1000, loss[loss=0.2024, simple_loss=0.2718, pruned_loss=0.06649, over 19764.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2817, pruned_loss=0.05939, over 3804264.63 frames. ], batch size: 45, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 10:59:49,404 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8222, 1.3847, 1.5565, 1.6631, 3.4445, 1.2969, 2.4744, 3.8958], + device='cuda:0'), covar=tensor([0.0520, 0.2884, 0.2956, 0.1907, 0.0645, 0.2501, 0.1425, 0.0224], + device='cuda:0'), in_proj_covar=tensor([0.0421, 0.0376, 0.0395, 0.0350, 0.0381, 0.0356, 0.0393, 0.0414], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 10:59:53,723 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.982e+02 4.891e+02 5.853e+02 7.878e+02 2.572e+03, threshold=1.171e+03, percent-clipped=6.0 +2023-04-03 10:59:57,745 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-03 11:00:23,372 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-03 11:00:34,020 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1238, 2.8592, 2.2307, 2.3549, 2.0920, 2.4725, 1.0997, 2.1102], + device='cuda:0'), covar=tensor([0.0782, 0.0675, 0.0785, 0.1185, 0.1132, 0.1175, 0.1517, 0.1151], + device='cuda:0'), in_proj_covar=tensor([0.0364, 0.0363, 0.0369, 0.0391, 0.0470, 0.0394, 0.0346, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 11:00:34,738 INFO [train.py:903] (0/4) Epoch 28, batch 1050, loss[loss=0.2057, simple_loss=0.2905, pruned_loss=0.06046, over 19658.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2833, pruned_loss=0.06006, over 3795217.97 frames. ], batch size: 55, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:00:47,040 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=185416.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:00:53,748 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=185422.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:01:02,617 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-03 11:01:14,704 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9789, 4.3828, 4.7347, 4.7079, 1.8278, 4.4165, 3.7577, 4.4380], + device='cuda:0'), covar=tensor([0.1742, 0.0995, 0.0608, 0.0757, 0.6108, 0.0946, 0.0752, 0.1190], + device='cuda:0'), in_proj_covar=tensor([0.0824, 0.0785, 0.0998, 0.0874, 0.0867, 0.0762, 0.0588, 0.0926], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 11:01:35,153 INFO [train.py:903] (0/4) Epoch 28, batch 1100, loss[loss=0.1813, simple_loss=0.2623, pruned_loss=0.05011, over 19779.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2853, pruned_loss=0.06126, over 3803712.43 frames. ], batch size: 54, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:01:57,231 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.219e+02 4.947e+02 6.329e+02 8.288e+02 1.903e+03, threshold=1.266e+03, percent-clipped=3.0 +2023-04-03 11:02:12,996 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=185486.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:02:18,720 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=185491.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:02:31,455 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7712, 1.2446, 1.4844, 1.4254, 3.3560, 1.1716, 2.3273, 3.7782], + device='cuda:0'), covar=tensor([0.0526, 0.3051, 0.2982, 0.2074, 0.0685, 0.2618, 0.1525, 0.0255], + device='cuda:0'), in_proj_covar=tensor([0.0422, 0.0377, 0.0396, 0.0351, 0.0381, 0.0357, 0.0393, 0.0415], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:02:35,892 INFO [train.py:903] (0/4) Epoch 28, batch 1150, loss[loss=0.1688, simple_loss=0.2461, pruned_loss=0.04575, over 19743.00 frames. ], tot_loss[loss=0.2048, simple_loss=0.2861, pruned_loss=0.06173, over 3792858.28 frames. ], batch size: 47, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:03:15,696 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=185537.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:03:28,122 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.86 vs. limit=5.0 +2023-04-03 11:03:40,791 INFO [train.py:903] (0/4) Epoch 28, batch 1200, loss[loss=0.2051, simple_loss=0.2795, pruned_loss=0.06538, over 19620.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.284, pruned_loss=0.06048, over 3808416.75 frames. ], batch size: 50, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:04:01,632 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.374e+02 5.383e+02 6.799e+02 8.653e+02 1.626e+03, threshold=1.360e+03, percent-clipped=3.0 +2023-04-03 11:04:11,766 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-03 11:04:34,004 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=185599.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:04:41,627 INFO [train.py:903] (0/4) Epoch 28, batch 1250, loss[loss=0.1933, simple_loss=0.2678, pruned_loss=0.05933, over 19399.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2833, pruned_loss=0.06007, over 3824870.18 frames. ], batch size: 48, lr: 2.94e-03, grad_scale: 16.0 +2023-04-03 11:04:42,003 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=185606.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:05:03,710 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=185624.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:05:44,892 INFO [train.py:903] (0/4) Epoch 28, batch 1300, loss[loss=0.2345, simple_loss=0.3122, pruned_loss=0.07843, over 17999.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2835, pruned_loss=0.06047, over 3811238.24 frames. ], batch size: 83, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:05:46,446 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6910, 1.4201, 1.6029, 1.6100, 3.2892, 1.2705, 2.5762, 3.6845], + device='cuda:0'), covar=tensor([0.0460, 0.2623, 0.2728, 0.1806, 0.0641, 0.2315, 0.1083, 0.0240], + device='cuda:0'), in_proj_covar=tensor([0.0421, 0.0376, 0.0396, 0.0351, 0.0381, 0.0357, 0.0392, 0.0415], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:06:04,600 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=185672.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:06:06,497 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.213e+02 4.542e+02 5.530e+02 7.592e+02 1.164e+03, threshold=1.106e+03, percent-clipped=0.0 +2023-04-03 11:06:35,852 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=185697.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:06:39,502 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0493, 3.2174, 1.9527, 2.0374, 2.9532, 1.7242, 1.6553, 2.2970], + device='cuda:0'), covar=tensor([0.1335, 0.0687, 0.1091, 0.0864, 0.0508, 0.1268, 0.0935, 0.0664], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0320, 0.0339, 0.0273, 0.0250, 0.0346, 0.0292, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:06:46,286 INFO [train.py:903] (0/4) Epoch 28, batch 1350, loss[loss=0.2076, simple_loss=0.2789, pruned_loss=0.06816, over 19391.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2826, pruned_loss=0.06005, over 3808484.81 frames. ], batch size: 48, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:07:48,956 INFO [train.py:903] (0/4) Epoch 28, batch 1400, loss[loss=0.2228, simple_loss=0.3034, pruned_loss=0.07116, over 19052.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2828, pruned_loss=0.06012, over 3812904.08 frames. ], batch size: 69, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:08:11,859 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.444e+02 4.899e+02 6.106e+02 7.699e+02 1.518e+03, threshold=1.221e+03, percent-clipped=6.0 +2023-04-03 11:08:15,718 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=185777.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:08:35,095 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=185793.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:08:44,851 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3495, 3.8325, 3.9654, 3.9665, 1.5519, 3.7749, 3.2753, 3.7487], + device='cuda:0'), covar=tensor([0.1721, 0.0914, 0.0685, 0.0791, 0.6073, 0.1037, 0.0786, 0.1145], + device='cuda:0'), in_proj_covar=tensor([0.0826, 0.0788, 0.1001, 0.0875, 0.0869, 0.0763, 0.0591, 0.0928], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 11:08:49,345 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-03 11:08:51,548 INFO [train.py:903] (0/4) Epoch 28, batch 1450, loss[loss=0.2052, simple_loss=0.2949, pruned_loss=0.05772, over 19666.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2832, pruned_loss=0.06015, over 3813167.35 frames. ], batch size: 55, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:09:05,591 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=185817.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 11:09:06,840 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=185818.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:09:20,227 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=185830.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:09:54,217 INFO [train.py:903] (0/4) Epoch 28, batch 1500, loss[loss=0.2151, simple_loss=0.3124, pruned_loss=0.05889, over 19777.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2839, pruned_loss=0.06081, over 3801159.42 frames. ], batch size: 56, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:10:01,968 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=185862.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:10:15,979 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.811e+02 5.017e+02 6.227e+02 8.666e+02 1.816e+03, threshold=1.245e+03, percent-clipped=11.0 +2023-04-03 11:10:33,538 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=185887.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:10:56,447 INFO [train.py:903] (0/4) Epoch 28, batch 1550, loss[loss=0.2001, simple_loss=0.2896, pruned_loss=0.05529, over 19800.00 frames. ], tot_loss[loss=0.2032, simple_loss=0.2845, pruned_loss=0.0609, over 3813844.92 frames. ], batch size: 56, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:11:45,477 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=185945.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:11:58,295 INFO [train.py:903] (0/4) Epoch 28, batch 1600, loss[loss=0.229, simple_loss=0.3157, pruned_loss=0.07112, over 19790.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2842, pruned_loss=0.06051, over 3825639.14 frames. ], batch size: 54, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:12:20,847 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-03 11:12:23,177 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.991e+02 4.773e+02 5.899e+02 6.974e+02 1.687e+03, threshold=1.180e+03, percent-clipped=2.0 +2023-04-03 11:12:28,398 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.88 vs. limit=5.0 +2023-04-03 11:12:54,129 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-186000.pt +2023-04-03 11:13:03,463 INFO [train.py:903] (0/4) Epoch 28, batch 1650, loss[loss=0.2658, simple_loss=0.3352, pruned_loss=0.09818, over 19112.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2837, pruned_loss=0.06011, over 3839962.85 frames. ], batch size: 69, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:13:07,279 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5358, 2.4026, 2.2754, 2.6507, 2.2185, 2.1235, 1.9612, 2.5418], + device='cuda:0'), covar=tensor([0.0991, 0.1629, 0.1410, 0.1003, 0.1504, 0.0551, 0.1601, 0.0688], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0362, 0.0320, 0.0259, 0.0309, 0.0259, 0.0323, 0.0267], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 11:13:26,163 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7866, 4.3649, 2.8110, 3.8068, 1.0726, 4.3690, 4.1904, 4.3322], + device='cuda:0'), covar=tensor([0.0556, 0.0936, 0.1927, 0.0859, 0.3909, 0.0595, 0.0896, 0.1176], + device='cuda:0'), in_proj_covar=tensor([0.0526, 0.0427, 0.0515, 0.0360, 0.0410, 0.0457, 0.0451, 0.0479], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:14:07,004 INFO [train.py:903] (0/4) Epoch 28, batch 1700, loss[loss=0.1899, simple_loss=0.2719, pruned_loss=0.05396, over 19672.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2842, pruned_loss=0.06016, over 3832061.71 frames. ], batch size: 53, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:14:29,613 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.897e+02 4.751e+02 5.670e+02 7.170e+02 1.723e+03, threshold=1.134e+03, percent-clipped=7.0 +2023-04-03 11:14:44,414 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-03 11:15:08,848 INFO [train.py:903] (0/4) Epoch 28, batch 1750, loss[loss=0.2105, simple_loss=0.2878, pruned_loss=0.06664, over 19589.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2854, pruned_loss=0.06064, over 3830020.83 frames. ], batch size: 52, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:15:24,172 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186118.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:15:28,462 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186121.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:16:11,527 INFO [train.py:903] (0/4) Epoch 28, batch 1800, loss[loss=0.1882, simple_loss=0.2782, pruned_loss=0.04913, over 19888.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2857, pruned_loss=0.06111, over 3843205.53 frames. ], batch size: 55, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:16:18,360 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186161.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 11:16:36,634 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.127e+02 5.184e+02 6.048e+02 8.514e+02 1.613e+03, threshold=1.210e+03, percent-clipped=4.0 +2023-04-03 11:16:38,124 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186176.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:16:42,852 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1438, 1.4069, 1.8739, 1.4803, 3.0007, 4.5334, 4.4508, 4.9735], + device='cuda:0'), covar=tensor([0.1798, 0.3932, 0.3557, 0.2557, 0.0732, 0.0244, 0.0174, 0.0206], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0333, 0.0366, 0.0273, 0.0257, 0.0197, 0.0220, 0.0279], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 11:17:02,769 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 11:17:08,586 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-03 11:17:09,009 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=186201.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:17:15,219 INFO [train.py:903] (0/4) Epoch 28, batch 1850, loss[loss=0.1686, simple_loss=0.246, pruned_loss=0.04559, over 19710.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2852, pruned_loss=0.06107, over 3841738.26 frames. ], batch size: 45, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:17:39,951 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=186226.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:17:46,436 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-03 11:17:51,544 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=186236.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:18:17,542 INFO [train.py:903] (0/4) Epoch 28, batch 1900, loss[loss=0.2281, simple_loss=0.3107, pruned_loss=0.0728, over 19726.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2857, pruned_loss=0.06093, over 3858866.47 frames. ], batch size: 63, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:18:33,597 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-03 11:18:38,425 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-03 11:18:39,538 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.824e+02 4.995e+02 5.845e+02 7.128e+02 1.193e+03, threshold=1.169e+03, percent-clipped=0.0 +2023-04-03 11:18:40,999 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=186276.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 11:18:42,897 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186277.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:19:02,976 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-03 11:19:19,221 INFO [train.py:903] (0/4) Epoch 28, batch 1950, loss[loss=0.1884, simple_loss=0.2834, pruned_loss=0.04665, over 17430.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.285, pruned_loss=0.0608, over 3832072.64 frames. ], batch size: 101, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:20:20,362 INFO [train.py:903] (0/4) Epoch 28, batch 2000, loss[loss=0.2231, simple_loss=0.3045, pruned_loss=0.0708, over 19543.00 frames. ], tot_loss[loss=0.2039, simple_loss=0.2855, pruned_loss=0.06114, over 3823208.84 frames. ], batch size: 54, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:20:32,046 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3833, 3.1419, 2.4740, 2.3950, 2.4733, 2.6990, 1.1910, 2.2411], + device='cuda:0'), covar=tensor([0.0759, 0.0590, 0.0762, 0.1196, 0.1009, 0.1087, 0.1467, 0.1135], + device='cuda:0'), in_proj_covar=tensor([0.0366, 0.0365, 0.0370, 0.0392, 0.0470, 0.0396, 0.0346, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 11:20:33,486 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.21 vs. limit=2.0 +2023-04-03 11:20:45,109 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.192e+02 5.060e+02 6.550e+02 8.587e+02 3.446e+03, threshold=1.310e+03, percent-clipped=8.0 +2023-04-03 11:21:10,085 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3225, 1.3205, 1.5004, 1.4748, 1.7895, 1.8313, 1.8410, 0.6010], + device='cuda:0'), covar=tensor([0.2638, 0.4618, 0.2851, 0.2113, 0.1698, 0.2482, 0.1476, 0.5246], + device='cuda:0'), in_proj_covar=tensor([0.0555, 0.0671, 0.0757, 0.0509, 0.0638, 0.0546, 0.0671, 0.0574], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 11:21:19,993 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-03 11:21:24,401 INFO [train.py:903] (0/4) Epoch 28, batch 2050, loss[loss=0.2071, simple_loss=0.2954, pruned_loss=0.05937, over 19097.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2849, pruned_loss=0.06088, over 3824240.97 frames. ], batch size: 69, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:21:40,397 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-03 11:21:42,542 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-03 11:22:02,202 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-03 11:22:04,678 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186439.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:22:26,600 INFO [train.py:903] (0/4) Epoch 28, batch 2100, loss[loss=0.1664, simple_loss=0.2471, pruned_loss=0.04279, over 19729.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2852, pruned_loss=0.06101, over 3826649.08 frames. ], batch size: 51, lr: 2.94e-03, grad_scale: 8.0 +2023-04-03 11:22:35,115 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186462.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:22:49,632 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.238e+02 5.089e+02 6.089e+02 7.390e+02 1.324e+03, threshold=1.218e+03, percent-clipped=1.0 +2023-04-03 11:22:58,700 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-03 11:23:11,820 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=186492.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:23:13,750 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186493.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:23:20,667 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-03 11:23:28,666 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.9746, 5.4236, 2.9294, 4.6544, 1.1412, 5.6166, 5.3380, 5.5914], + device='cuda:0'), covar=tensor([0.0341, 0.0805, 0.1976, 0.0774, 0.4081, 0.0483, 0.0859, 0.0960], + device='cuda:0'), in_proj_covar=tensor([0.0528, 0.0430, 0.0516, 0.0360, 0.0410, 0.0458, 0.0452, 0.0481], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:23:29,662 INFO [train.py:903] (0/4) Epoch 28, batch 2150, loss[loss=0.2045, simple_loss=0.2912, pruned_loss=0.05891, over 19517.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2853, pruned_loss=0.06111, over 3822251.10 frames. ], batch size: 64, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:23:42,668 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=186517.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:23:45,790 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186520.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:24:02,574 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=186532.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 11:24:06,044 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186535.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:24:30,567 INFO [train.py:903] (0/4) Epoch 28, batch 2200, loss[loss=0.1732, simple_loss=0.2533, pruned_loss=0.04651, over 19764.00 frames. ], tot_loss[loss=0.204, simple_loss=0.2852, pruned_loss=0.06141, over 3820489.35 frames. ], batch size: 48, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:24:32,075 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186557.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:24:32,188 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=186557.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 11:24:55,776 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.530e+02 4.864e+02 5.746e+02 7.619e+02 1.717e+03, threshold=1.149e+03, percent-clipped=3.0 +2023-04-03 11:24:58,630 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=186577.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:25:09,374 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186585.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:25:32,792 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186604.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:25:34,933 INFO [train.py:903] (0/4) Epoch 28, batch 2250, loss[loss=0.1586, simple_loss=0.2394, pruned_loss=0.03896, over 19759.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2841, pruned_loss=0.06092, over 3799340.51 frames. ], batch size: 46, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:25:54,472 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186621.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:26:09,358 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.12 vs. limit=2.0 +2023-04-03 11:26:12,596 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=186635.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:26:39,385 INFO [train.py:903] (0/4) Epoch 28, batch 2300, loss[loss=0.1648, simple_loss=0.2505, pruned_loss=0.03961, over 19862.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2838, pruned_loss=0.06084, over 3790627.34 frames. ], batch size: 52, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:26:55,478 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-03 11:27:02,271 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.323e+02 4.896e+02 5.750e+02 7.152e+02 2.246e+03, threshold=1.150e+03, percent-clipped=6.0 +2023-04-03 11:27:42,236 INFO [train.py:903] (0/4) Epoch 28, batch 2350, loss[loss=0.219, simple_loss=0.3063, pruned_loss=0.06579, over 19324.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2839, pruned_loss=0.0607, over 3792838.72 frames. ], batch size: 66, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:28:07,862 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186727.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:28:20,484 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=186736.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:28:26,081 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-03 11:28:42,185 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-03 11:28:44,646 INFO [train.py:903] (0/4) Epoch 28, batch 2400, loss[loss=0.2398, simple_loss=0.3198, pruned_loss=0.07993, over 19263.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2825, pruned_loss=0.05995, over 3808765.09 frames. ], batch size: 66, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:29:08,813 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.283e+02 4.439e+02 5.729e+02 7.466e+02 1.887e+03, threshold=1.146e+03, percent-clipped=9.0 +2023-04-03 11:29:20,254 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186783.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:29:47,722 INFO [train.py:903] (0/4) Epoch 28, batch 2450, loss[loss=0.1769, simple_loss=0.2642, pruned_loss=0.04477, over 19603.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2813, pruned_loss=0.05945, over 3808209.87 frames. ], batch size: 61, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:30:01,663 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.71 vs. limit=2.0 +2023-04-03 11:30:02,270 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2889, 2.0522, 1.8712, 2.1415, 1.9357, 1.8665, 1.7360, 2.1069], + device='cuda:0'), covar=tensor([0.1027, 0.1508, 0.1479, 0.1095, 0.1385, 0.0574, 0.1596, 0.0730], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0362, 0.0321, 0.0259, 0.0308, 0.0258, 0.0323, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 11:30:22,250 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=186833.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:30:26,673 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186837.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:30:50,780 INFO [train.py:903] (0/4) Epoch 28, batch 2500, loss[loss=0.1793, simple_loss=0.2536, pruned_loss=0.05244, over 19715.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2809, pruned_loss=0.0594, over 3808532.83 frames. ], batch size: 45, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:30:54,459 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186858.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:30:54,580 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=186858.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:31:15,070 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.341e+02 4.860e+02 5.866e+02 7.110e+02 2.029e+03, threshold=1.173e+03, percent-clipped=7.0 +2023-04-03 11:31:19,818 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186879.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:31:35,069 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=186891.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:31:45,046 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=186898.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:31:48,252 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186901.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:31:54,935 INFO [train.py:903] (0/4) Epoch 28, batch 2550, loss[loss=0.1706, simple_loss=0.2459, pruned_loss=0.04763, over 19292.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2818, pruned_loss=0.05998, over 3813259.24 frames. ], batch size: 44, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:32:06,784 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=186916.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:32:22,496 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186929.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:32:22,643 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=186929.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:32:47,690 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=186948.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:32:50,867 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-03 11:32:52,300 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=186952.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:32:56,643 INFO [train.py:903] (0/4) Epoch 28, batch 2600, loss[loss=0.1714, simple_loss=0.2647, pruned_loss=0.03905, over 19789.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2817, pruned_loss=0.05995, over 3816106.95 frames. ], batch size: 56, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:33:06,496 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5502, 1.5625, 1.7110, 1.7856, 2.2318, 2.2132, 2.3051, 0.9399], + device='cuda:0'), covar=tensor([0.2558, 0.4582, 0.2924, 0.1921, 0.1658, 0.2253, 0.1514, 0.4987], + device='cuda:0'), in_proj_covar=tensor([0.0553, 0.0669, 0.0755, 0.0507, 0.0633, 0.0545, 0.0667, 0.0571], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 11:33:20,687 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.371e+02 5.127e+02 5.898e+02 7.788e+02 1.720e+03, threshold=1.180e+03, percent-clipped=7.0 +2023-04-03 11:33:43,069 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=186992.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:33:45,238 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=186994.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:33:59,859 INFO [train.py:903] (0/4) Epoch 28, batch 2650, loss[loss=0.2424, simple_loss=0.3315, pruned_loss=0.07662, over 19738.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2827, pruned_loss=0.06046, over 3808957.23 frames. ], batch size: 63, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:34:12,821 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187016.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:34:14,013 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187017.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:34:22,437 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-03 11:34:42,879 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4191, 2.4388, 2.5715, 2.9738, 2.5126, 2.9279, 2.6182, 2.5311], + device='cuda:0'), covar=tensor([0.3489, 0.3353, 0.1619, 0.2009, 0.3241, 0.1782, 0.3685, 0.2586], + device='cuda:0'), in_proj_covar=tensor([0.0941, 0.1020, 0.0747, 0.0955, 0.0917, 0.0858, 0.0867, 0.0814], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 11:34:46,251 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=187043.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:34:47,535 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187044.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:35:02,636 INFO [train.py:903] (0/4) Epoch 28, batch 2700, loss[loss=0.1649, simple_loss=0.2417, pruned_loss=0.0441, over 19373.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2837, pruned_loss=0.06073, over 3808471.88 frames. ], batch size: 47, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:35:02,883 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=187056.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:35:13,333 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187063.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:35:22,254 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=187071.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:35:26,784 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.432e+02 4.685e+02 5.845e+02 7.450e+02 1.306e+03, threshold=1.169e+03, percent-clipped=4.0 +2023-04-03 11:36:06,491 INFO [train.py:903] (0/4) Epoch 28, batch 2750, loss[loss=0.2137, simple_loss=0.2986, pruned_loss=0.06438, over 19508.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2831, pruned_loss=0.06003, over 3799536.00 frames. ], batch size: 54, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:36:10,442 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5566, 2.2600, 1.7522, 1.5406, 2.0561, 1.4560, 1.4071, 2.0465], + device='cuda:0'), covar=tensor([0.1140, 0.0942, 0.1106, 0.0971, 0.0643, 0.1349, 0.0861, 0.0514], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0320, 0.0341, 0.0273, 0.0252, 0.0345, 0.0293, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:36:13,210 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-03 11:36:15,452 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 11:36:18,804 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6924, 1.5751, 1.6001, 2.2916, 1.6205, 2.0262, 1.9572, 1.8049], + device='cuda:0'), covar=tensor([0.0879, 0.0919, 0.0997, 0.0722, 0.0913, 0.0770, 0.0874, 0.0697], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0223, 0.0227, 0.0239, 0.0225, 0.0213, 0.0187, 0.0205], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 11:36:43,428 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2815, 1.3555, 1.7889, 1.4612, 2.8885, 3.7669, 3.4209, 3.9530], + device='cuda:0'), covar=tensor([0.1554, 0.3702, 0.3271, 0.2308, 0.0552, 0.0170, 0.0209, 0.0270], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0333, 0.0365, 0.0272, 0.0256, 0.0197, 0.0220, 0.0280], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 11:36:59,815 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.50 vs. limit=2.0 +2023-04-03 11:37:06,427 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187154.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:37:08,436 INFO [train.py:903] (0/4) Epoch 28, batch 2800, loss[loss=0.1759, simple_loss=0.2571, pruned_loss=0.04733, over 19862.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2824, pruned_loss=0.05964, over 3813014.22 frames. ], batch size: 52, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:37:31,623 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.065e+02 4.906e+02 5.640e+02 7.444e+02 1.445e+03, threshold=1.128e+03, percent-clipped=1.0 +2023-04-03 11:37:36,918 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187179.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:37:46,991 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187186.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:38:03,783 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3580, 2.3861, 2.2797, 2.5521, 2.3423, 1.9997, 2.1744, 2.3613], + device='cuda:0'), covar=tensor([0.0866, 0.1125, 0.1036, 0.0762, 0.1027, 0.0501, 0.1200, 0.0602], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0360, 0.0319, 0.0257, 0.0305, 0.0257, 0.0322, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:38:05,764 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=187202.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:38:10,176 INFO [train.py:903] (0/4) Epoch 28, batch 2850, loss[loss=0.2467, simple_loss=0.3145, pruned_loss=0.08946, over 19349.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2828, pruned_loss=0.06001, over 3806738.46 frames. ], batch size: 66, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:38:13,786 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187208.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:38:44,920 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187233.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:39:05,529 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3460, 1.0379, 1.2998, 1.9619, 1.4971, 1.2795, 1.4495, 1.2256], + device='cuda:0'), covar=tensor([0.1173, 0.1723, 0.1306, 0.0830, 0.1183, 0.1489, 0.1313, 0.1146], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0223, 0.0227, 0.0239, 0.0226, 0.0213, 0.0187, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 11:39:05,587 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187250.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:39:11,670 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-03 11:39:12,690 INFO [train.py:903] (0/4) Epoch 28, batch 2900, loss[loss=0.2263, simple_loss=0.3174, pruned_loss=0.06758, over 19668.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2825, pruned_loss=0.0596, over 3804502.01 frames. ], batch size: 60, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:39:13,294 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 11:39:29,225 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8067, 1.9178, 1.9071, 2.6783, 2.0190, 2.5796, 1.9827, 1.5914], + device='cuda:0'), covar=tensor([0.5044, 0.4871, 0.2979, 0.3205, 0.4892, 0.2628, 0.6623, 0.5470], + device='cuda:0'), in_proj_covar=tensor([0.0938, 0.1016, 0.0744, 0.0953, 0.0914, 0.0857, 0.0864, 0.0812], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 11:39:32,576 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187272.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:39:33,466 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=187273.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:39:35,532 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.364e+02 5.012e+02 6.172e+02 7.442e+02 2.226e+03, threshold=1.234e+03, percent-clipped=8.0 +2023-04-03 11:39:35,902 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187275.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:40:02,569 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187297.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:40:06,077 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187300.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:40:13,525 INFO [train.py:903] (0/4) Epoch 28, batch 2950, loss[loss=0.185, simple_loss=0.2779, pruned_loss=0.0461, over 19527.00 frames. ], tot_loss[loss=0.2029, simple_loss=0.2841, pruned_loss=0.06082, over 3805017.10 frames. ], batch size: 54, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:40:25,515 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4345, 1.8189, 1.4685, 1.4037, 1.7139, 1.3131, 1.3784, 1.6979], + device='cuda:0'), covar=tensor([0.0859, 0.0858, 0.0811, 0.0782, 0.0532, 0.1093, 0.0655, 0.0500], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0320, 0.0342, 0.0273, 0.0253, 0.0346, 0.0294, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:40:26,515 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187317.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:40:28,967 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187319.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:40:35,742 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187325.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:40:59,956 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187344.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:41:13,838 INFO [train.py:903] (0/4) Epoch 28, batch 3000, loss[loss=0.1865, simple_loss=0.2724, pruned_loss=0.05026, over 19661.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2846, pruned_loss=0.06116, over 3803195.87 frames. ], batch size: 53, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:41:13,838 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 11:41:26,714 INFO [train.py:937] (0/4) Epoch 28, validation: loss=0.1673, simple_loss=0.2667, pruned_loss=0.03394, over 944034.00 frames. +2023-04-03 11:41:26,715 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 11:41:29,148 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-03 11:41:49,227 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.693e+02 4.898e+02 6.411e+02 7.995e+02 1.373e+03, threshold=1.282e+03, percent-clipped=5.0 +2023-04-03 11:42:05,094 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=187387.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:42:06,472 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187388.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:42:08,757 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=187390.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:42:18,233 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.46 vs. limit=2.0 +2023-04-03 11:42:21,164 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=187400.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:42:27,787 INFO [train.py:903] (0/4) Epoch 28, batch 3050, loss[loss=0.1785, simple_loss=0.2633, pruned_loss=0.0468, over 19740.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2849, pruned_loss=0.06106, over 3783270.86 frames. ], batch size: 51, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:43:13,158 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187442.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:43:29,480 INFO [train.py:903] (0/4) Epoch 28, batch 3100, loss[loss=0.2147, simple_loss=0.2935, pruned_loss=0.06796, over 19533.00 frames. ], tot_loss[loss=0.2024, simple_loss=0.2836, pruned_loss=0.06059, over 3781604.95 frames. ], batch size: 54, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:43:43,911 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187467.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:43:54,449 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.395e+02 4.792e+02 5.797e+02 7.675e+02 1.223e+03, threshold=1.159e+03, percent-clipped=0.0 +2023-04-03 11:44:27,043 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187502.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:44:32,170 INFO [train.py:903] (0/4) Epoch 28, batch 3150, loss[loss=0.2235, simple_loss=0.2993, pruned_loss=0.07386, over 17049.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2823, pruned_loss=0.05982, over 3787430.34 frames. ], batch size: 100, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:44:44,911 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187515.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:44:54,656 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-03 11:45:01,635 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=187530.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:45:05,052 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6851, 4.2907, 2.8302, 3.7695, 1.0065, 4.2506, 4.0638, 4.2030], + device='cuda:0'), covar=tensor([0.0631, 0.0957, 0.1884, 0.0850, 0.3956, 0.0707, 0.0939, 0.1172], + device='cuda:0'), in_proj_covar=tensor([0.0534, 0.0435, 0.0521, 0.0364, 0.0414, 0.0462, 0.0456, 0.0487], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:45:10,634 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=187537.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:45:11,885 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2738, 1.2184, 1.2248, 1.3463, 0.9655, 1.3749, 1.2624, 1.2869], + device='cuda:0'), covar=tensor([0.0920, 0.1002, 0.1109, 0.0677, 0.0940, 0.0904, 0.0883, 0.0810], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0223, 0.0228, 0.0240, 0.0226, 0.0214, 0.0188, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 11:45:15,208 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=187541.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:45:35,289 INFO [train.py:903] (0/4) Epoch 28, batch 3200, loss[loss=0.2143, simple_loss=0.298, pruned_loss=0.06534, over 19603.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2828, pruned_loss=0.06011, over 3790567.49 frames. ], batch size: 61, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:45:45,083 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5462, 1.6308, 1.9560, 1.7943, 2.7516, 2.1579, 2.9461, 1.5902], + device='cuda:0'), covar=tensor([0.2711, 0.4460, 0.2797, 0.2169, 0.1576, 0.2509, 0.1473, 0.4604], + device='cuda:0'), in_proj_covar=tensor([0.0553, 0.0670, 0.0754, 0.0507, 0.0635, 0.0545, 0.0669, 0.0572], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 11:45:55,551 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187573.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:45:57,382 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.995e+02 4.914e+02 6.228e+02 8.001e+02 2.182e+03, threshold=1.246e+03, percent-clipped=10.0 +2023-04-03 11:46:27,891 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187598.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:46:36,797 INFO [train.py:903] (0/4) Epoch 28, batch 3250, loss[loss=0.1759, simple_loss=0.2645, pruned_loss=0.04368, over 19515.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2828, pruned_loss=0.05998, over 3790550.29 frames. ], batch size: 54, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:46:42,134 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.34 vs. limit=5.0 +2023-04-03 11:47:23,915 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187644.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:47:37,446 INFO [train.py:903] (0/4) Epoch 28, batch 3300, loss[loss=0.2145, simple_loss=0.2851, pruned_loss=0.07198, over 19613.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2826, pruned_loss=0.06008, over 3800252.03 frames. ], batch size: 50, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:47:37,470 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-03 11:47:54,272 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187669.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:48:01,982 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.234e+02 4.770e+02 6.093e+02 7.830e+02 1.620e+03, threshold=1.219e+03, percent-clipped=4.0 +2023-04-03 11:48:07,213 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.54 vs. limit=2.0 +2023-04-03 11:48:09,499 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 11:48:40,836 INFO [train.py:903] (0/4) Epoch 28, batch 3350, loss[loss=0.1737, simple_loss=0.2572, pruned_loss=0.04509, over 19481.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2834, pruned_loss=0.06052, over 3810197.94 frames. ], batch size: 49, lr: 2.93e-03, grad_scale: 8.0 +2023-04-03 11:49:14,753 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=187734.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:49:42,978 INFO [train.py:903] (0/4) Epoch 28, batch 3400, loss[loss=0.1717, simple_loss=0.2492, pruned_loss=0.04704, over 19364.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2834, pruned_loss=0.06033, over 3826172.80 frames. ], batch size: 47, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 11:49:45,544 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187758.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:50:01,974 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=187771.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:50:07,735 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.151e+02 4.801e+02 5.865e+02 7.575e+02 1.695e+03, threshold=1.173e+03, percent-clipped=1.0 +2023-04-03 11:50:16,874 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187783.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:50:32,875 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=187796.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:50:46,601 INFO [train.py:903] (0/4) Epoch 28, batch 3450, loss[loss=0.1838, simple_loss=0.258, pruned_loss=0.05481, over 19035.00 frames. ], tot_loss[loss=0.2033, simple_loss=0.2848, pruned_loss=0.06088, over 3836873.73 frames. ], batch size: 42, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 11:50:47,820 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-03 11:51:41,440 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187849.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:51:49,295 INFO [train.py:903] (0/4) Epoch 28, batch 3500, loss[loss=0.1886, simple_loss=0.2656, pruned_loss=0.05586, over 19656.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2838, pruned_loss=0.06025, over 3833164.42 frames. ], batch size: 53, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 11:52:11,978 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=187874.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:52:14,011 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.602e+02 4.936e+02 6.004e+02 7.154e+02 1.224e+03, threshold=1.201e+03, percent-clipped=1.0 +2023-04-03 11:52:20,872 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=187881.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:52:26,776 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=187885.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:52:50,775 INFO [train.py:903] (0/4) Epoch 28, batch 3550, loss[loss=0.2232, simple_loss=0.3028, pruned_loss=0.07179, over 18767.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2834, pruned_loss=0.06012, over 3835117.11 frames. ], batch size: 74, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 11:53:37,511 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9076, 1.8629, 1.8121, 1.6634, 1.4847, 1.6197, 0.5242, 0.9469], + device='cuda:0'), covar=tensor([0.0731, 0.0719, 0.0478, 0.0783, 0.1268, 0.0859, 0.1418, 0.1159], + device='cuda:0'), in_proj_covar=tensor([0.0367, 0.0366, 0.0372, 0.0393, 0.0472, 0.0397, 0.0346, 0.0350], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 11:53:52,901 INFO [train.py:903] (0/4) Epoch 28, batch 3600, loss[loss=0.1772, simple_loss=0.2586, pruned_loss=0.04783, over 19752.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2846, pruned_loss=0.06083, over 3829763.11 frames. ], batch size: 51, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 11:54:03,439 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1378, 1.7984, 1.4453, 1.2302, 1.5889, 1.1942, 1.1421, 1.6477], + device='cuda:0'), covar=tensor([0.0972, 0.0849, 0.1240, 0.0919, 0.0640, 0.1442, 0.0740, 0.0475], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0320, 0.0344, 0.0273, 0.0253, 0.0348, 0.0294, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:54:17,595 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.816e+02 4.904e+02 5.870e+02 7.567e+02 1.667e+03, threshold=1.174e+03, percent-clipped=3.0 +2023-04-03 11:54:34,389 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187989.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:54:42,580 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=187996.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:54:46,912 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-188000.pt +2023-04-03 11:54:48,307 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=188000.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:54:56,052 INFO [train.py:903] (0/4) Epoch 28, batch 3650, loss[loss=0.2126, simple_loss=0.3036, pruned_loss=0.06079, over 19671.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2837, pruned_loss=0.06073, over 3829475.99 frames. ], batch size: 59, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 11:55:51,316 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8246, 1.3526, 1.0690, 0.9814, 1.1454, 1.0208, 0.8976, 1.2092], + device='cuda:0'), covar=tensor([0.0742, 0.0925, 0.1239, 0.0879, 0.0643, 0.1448, 0.0736, 0.0567], + device='cuda:0'), in_proj_covar=tensor([0.0306, 0.0322, 0.0345, 0.0274, 0.0254, 0.0350, 0.0296, 0.0278], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 11:55:57,939 INFO [train.py:903] (0/4) Epoch 28, batch 3700, loss[loss=0.1805, simple_loss=0.2624, pruned_loss=0.04933, over 19722.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2846, pruned_loss=0.06132, over 3797744.76 frames. ], batch size: 51, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 11:56:23,997 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.791e+02 5.013e+02 6.013e+02 7.793e+02 2.143e+03, threshold=1.203e+03, percent-clipped=3.0 +2023-04-03 11:56:59,793 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=188105.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:57:00,598 INFO [train.py:903] (0/4) Epoch 28, batch 3750, loss[loss=0.2036, simple_loss=0.2864, pruned_loss=0.06041, over 13470.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2837, pruned_loss=0.06076, over 3804027.25 frames. ], batch size: 136, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 11:57:32,219 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=188130.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 11:58:03,786 INFO [train.py:903] (0/4) Epoch 28, batch 3800, loss[loss=0.2049, simple_loss=0.2873, pruned_loss=0.06126, over 17222.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2827, pruned_loss=0.05991, over 3820532.63 frames. ], batch size: 101, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 11:58:30,498 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.894e+02 5.177e+02 6.017e+02 8.386e+02 1.721e+03, threshold=1.203e+03, percent-clipped=7.0 +2023-04-03 11:58:33,991 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-03 11:59:06,276 INFO [train.py:903] (0/4) Epoch 28, batch 3850, loss[loss=0.1961, simple_loss=0.2863, pruned_loss=0.05289, over 19680.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.283, pruned_loss=0.05971, over 3829078.96 frames. ], batch size: 59, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 11:59:56,106 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=188245.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:00:05,189 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=188252.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:00:09,170 INFO [train.py:903] (0/4) Epoch 28, batch 3900, loss[loss=0.2153, simple_loss=0.2911, pruned_loss=0.06973, over 19587.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2835, pruned_loss=0.05987, over 3829764.29 frames. ], batch size: 52, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 12:00:09,596 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=188256.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:00:26,443 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=188270.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:00:34,027 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.127e+02 5.053e+02 6.264e+02 8.116e+02 1.975e+03, threshold=1.253e+03, percent-clipped=4.0 +2023-04-03 12:00:34,423 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=188277.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:00:39,133 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=188281.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:01:04,048 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=188301.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:01:04,150 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5345, 2.6467, 2.2934, 2.7585, 2.4596, 2.2186, 2.2095, 2.5542], + device='cuda:0'), covar=tensor([0.0992, 0.1430, 0.1403, 0.0957, 0.1312, 0.0556, 0.1475, 0.0672], + device='cuda:0'), in_proj_covar=tensor([0.0277, 0.0358, 0.0318, 0.0256, 0.0305, 0.0256, 0.0321, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:01:09,683 INFO [train.py:903] (0/4) Epoch 28, batch 3950, loss[loss=0.1984, simple_loss=0.2906, pruned_loss=0.05314, over 19275.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2833, pruned_loss=0.05964, over 3828691.03 frames. ], batch size: 66, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 12:01:15,495 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-03 12:02:12,176 INFO [train.py:903] (0/4) Epoch 28, batch 4000, loss[loss=0.2678, simple_loss=0.3428, pruned_loss=0.09639, over 19311.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2835, pruned_loss=0.05951, over 3832896.13 frames. ], batch size: 66, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:02:38,281 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.378e+02 4.980e+02 6.272e+02 8.090e+02 1.579e+03, threshold=1.254e+03, percent-clipped=5.0 +2023-04-03 12:02:59,863 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-03 12:03:14,916 INFO [train.py:903] (0/4) Epoch 28, batch 4050, loss[loss=0.2675, simple_loss=0.3437, pruned_loss=0.09564, over 13432.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2833, pruned_loss=0.05965, over 3832102.45 frames. ], batch size: 136, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:04:17,051 INFO [train.py:903] (0/4) Epoch 28, batch 4100, loss[loss=0.2237, simple_loss=0.3014, pruned_loss=0.07299, over 19772.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2844, pruned_loss=0.06036, over 3823482.08 frames. ], batch size: 56, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:04:43,424 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.476e+02 5.116e+02 6.170e+02 8.694e+02 1.686e+03, threshold=1.234e+03, percent-clipped=5.0 +2023-04-03 12:04:48,571 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7295, 1.7651, 1.6225, 1.4258, 1.3755, 1.4269, 0.3336, 0.7289], + device='cuda:0'), covar=tensor([0.0629, 0.0645, 0.0479, 0.0759, 0.1296, 0.0832, 0.1346, 0.1172], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0365, 0.0372, 0.0394, 0.0474, 0.0398, 0.0346, 0.0351], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 12:04:53,769 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-03 12:05:11,254 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=188499.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:05:19,164 INFO [train.py:903] (0/4) Epoch 28, batch 4150, loss[loss=0.1986, simple_loss=0.2883, pruned_loss=0.05445, over 19304.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2838, pruned_loss=0.05999, over 3810848.69 frames. ], batch size: 70, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:05:56,442 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2461, 1.2426, 1.6470, 1.0159, 2.3482, 3.0658, 2.7398, 3.2397], + device='cuda:0'), covar=tensor([0.1577, 0.4140, 0.3590, 0.2875, 0.0673, 0.0233, 0.0274, 0.0346], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0333, 0.0367, 0.0273, 0.0257, 0.0198, 0.0222, 0.0280], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 12:06:21,499 INFO [train.py:903] (0/4) Epoch 28, batch 4200, loss[loss=0.2209, simple_loss=0.3095, pruned_loss=0.06615, over 19671.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2847, pruned_loss=0.06065, over 3803678.04 frames. ], batch size: 55, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:06:24,982 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-03 12:06:46,553 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.847e+02 4.775e+02 5.885e+02 7.540e+02 1.202e+03, threshold=1.177e+03, percent-clipped=0.0 +2023-04-03 12:07:22,476 INFO [train.py:903] (0/4) Epoch 28, batch 4250, loss[loss=0.2038, simple_loss=0.284, pruned_loss=0.0618, over 19777.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.284, pruned_loss=0.06078, over 3794366.71 frames. ], batch size: 54, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 12:07:33,737 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-03 12:07:44,072 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-03 12:07:52,218 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2346, 1.3681, 1.9988, 1.7516, 3.0870, 4.5361, 4.3664, 4.9444], + device='cuda:0'), covar=tensor([0.1822, 0.4129, 0.3437, 0.2368, 0.0667, 0.0232, 0.0198, 0.0218], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0333, 0.0367, 0.0273, 0.0257, 0.0198, 0.0222, 0.0281], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 12:08:10,309 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=188645.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:08:24,256 INFO [train.py:903] (0/4) Epoch 28, batch 4300, loss[loss=0.2021, simple_loss=0.2874, pruned_loss=0.05846, over 19746.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2839, pruned_loss=0.06029, over 3811131.45 frames. ], batch size: 63, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 12:08:51,750 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.988e+02 4.883e+02 5.889e+02 8.258e+02 1.553e+03, threshold=1.178e+03, percent-clipped=6.0 +2023-04-03 12:09:15,652 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-03 12:09:26,300 INFO [train.py:903] (0/4) Epoch 28, batch 4350, loss[loss=0.2102, simple_loss=0.2975, pruned_loss=0.06151, over 19316.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2833, pruned_loss=0.05996, over 3819861.80 frames. ], batch size: 66, lr: 2.92e-03, grad_scale: 4.0 +2023-04-03 12:09:32,306 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.33 vs. limit=2.0 +2023-04-03 12:09:44,346 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.73 vs. limit=5.0 +2023-04-03 12:10:13,823 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1099, 2.0305, 1.8052, 2.1239, 1.8521, 1.8131, 1.7377, 1.9774], + device='cuda:0'), covar=tensor([0.0979, 0.1284, 0.1445, 0.0966, 0.1301, 0.0546, 0.1513, 0.0720], + device='cuda:0'), in_proj_covar=tensor([0.0275, 0.0357, 0.0317, 0.0255, 0.0302, 0.0254, 0.0319, 0.0261], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:10:25,207 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0807, 5.5171, 3.1575, 4.8457, 1.1737, 5.7219, 5.5085, 5.7332], + device='cuda:0'), covar=tensor([0.0368, 0.0819, 0.1721, 0.0741, 0.3911, 0.0482, 0.0744, 0.0837], + device='cuda:0'), in_proj_covar=tensor([0.0527, 0.0430, 0.0514, 0.0361, 0.0410, 0.0456, 0.0452, 0.0482], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:10:30,424 INFO [train.py:903] (0/4) Epoch 28, batch 4400, loss[loss=0.2093, simple_loss=0.2991, pruned_loss=0.05976, over 19272.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2826, pruned_loss=0.05976, over 3828592.01 frames. ], batch size: 66, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:10:35,431 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=188760.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:10:52,235 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-03 12:10:57,774 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.415e+02 4.797e+02 5.749e+02 7.341e+02 1.454e+03, threshold=1.150e+03, percent-clipped=2.0 +2023-04-03 12:11:02,383 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-03 12:11:31,835 INFO [train.py:903] (0/4) Epoch 28, batch 4450, loss[loss=0.2217, simple_loss=0.3038, pruned_loss=0.06979, over 19730.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2831, pruned_loss=0.05978, over 3830735.31 frames. ], batch size: 63, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:12:01,898 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.56 vs. limit=5.0 +2023-04-03 12:12:19,630 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=188843.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:12:36,103 INFO [train.py:903] (0/4) Epoch 28, batch 4500, loss[loss=0.2142, simple_loss=0.2964, pruned_loss=0.06604, over 18213.00 frames. ], tot_loss[loss=0.2, simple_loss=0.2819, pruned_loss=0.05907, over 3828191.81 frames. ], batch size: 83, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:13:04,408 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.382e+02 4.536e+02 5.549e+02 7.562e+02 1.666e+03, threshold=1.110e+03, percent-clipped=5.0 +2023-04-03 12:13:38,230 INFO [train.py:903] (0/4) Epoch 28, batch 4550, loss[loss=0.1866, simple_loss=0.2816, pruned_loss=0.04585, over 19607.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2825, pruned_loss=0.0597, over 3822574.28 frames. ], batch size: 57, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:13:44,770 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6890, 1.3406, 1.6484, 1.5193, 3.2839, 1.2695, 2.5946, 3.6971], + device='cuda:0'), covar=tensor([0.0488, 0.2844, 0.2775, 0.1923, 0.0645, 0.2470, 0.1156, 0.0252], + device='cuda:0'), in_proj_covar=tensor([0.0425, 0.0380, 0.0398, 0.0353, 0.0383, 0.0359, 0.0398, 0.0420], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:13:46,952 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-03 12:14:12,245 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-03 12:14:29,479 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=188947.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:14:31,821 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=188949.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:14:39,521 INFO [train.py:903] (0/4) Epoch 28, batch 4600, loss[loss=0.1948, simple_loss=0.2813, pruned_loss=0.05416, over 19765.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.283, pruned_loss=0.05985, over 3827540.45 frames. ], batch size: 63, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:14:43,188 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=188958.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:15:07,935 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.099e+02 4.624e+02 5.779e+02 7.629e+02 1.899e+03, threshold=1.156e+03, percent-clipped=6.0 +2023-04-03 12:15:08,228 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8689, 4.4718, 2.6929, 3.8950, 1.1144, 4.4025, 4.3126, 4.3439], + device='cuda:0'), covar=tensor([0.0584, 0.0917, 0.1905, 0.0833, 0.3796, 0.0679, 0.0882, 0.1052], + device='cuda:0'), in_proj_covar=tensor([0.0525, 0.0428, 0.0512, 0.0359, 0.0407, 0.0454, 0.0450, 0.0480], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:15:13,472 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.46 vs. limit=5.0 +2023-04-03 12:15:43,037 INFO [train.py:903] (0/4) Epoch 28, batch 4650, loss[loss=0.2067, simple_loss=0.2921, pruned_loss=0.0607, over 19531.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2832, pruned_loss=0.05988, over 3821305.18 frames. ], batch size: 54, lr: 2.92e-03, grad_scale: 8.0 +2023-04-03 12:15:56,020 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=189016.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:16:01,295 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-03 12:16:12,658 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-03 12:16:25,660 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=189041.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:16:34,726 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=189048.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:16:34,937 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6507, 1.7398, 2.0465, 2.0354, 1.5437, 2.0028, 2.0188, 1.9007], + device='cuda:0'), covar=tensor([0.4374, 0.3958, 0.2056, 0.2406, 0.4158, 0.2278, 0.5349, 0.3578], + device='cuda:0'), in_proj_covar=tensor([0.0937, 0.1014, 0.0741, 0.0951, 0.0913, 0.0854, 0.0859, 0.0806], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 12:16:44,450 INFO [train.py:903] (0/4) Epoch 28, batch 4700, loss[loss=0.2341, simple_loss=0.307, pruned_loss=0.08063, over 19644.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2823, pruned_loss=0.0598, over 3824559.73 frames. ], batch size: 60, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:17:07,720 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-03 12:17:10,870 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.010e+02 4.848e+02 5.842e+02 7.352e+02 2.015e+03, threshold=1.168e+03, percent-clipped=3.0 +2023-04-03 12:17:17,594 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=189082.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:17:36,158 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=189097.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:17:46,810 INFO [train.py:903] (0/4) Epoch 28, batch 4750, loss[loss=0.1799, simple_loss=0.2729, pruned_loss=0.04347, over 19612.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2821, pruned_loss=0.05957, over 3825152.19 frames. ], batch size: 57, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:18:47,829 INFO [train.py:903] (0/4) Epoch 28, batch 4800, loss[loss=0.2899, simple_loss=0.3456, pruned_loss=0.1171, over 12852.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2828, pruned_loss=0.05979, over 3818802.68 frames. ], batch size: 136, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:19:16,004 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.214e+02 5.029e+02 6.156e+02 7.557e+02 1.439e+03, threshold=1.231e+03, percent-clipped=4.0 +2023-04-03 12:19:43,398 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 12:19:50,819 INFO [train.py:903] (0/4) Epoch 28, batch 4850, loss[loss=0.2367, simple_loss=0.304, pruned_loss=0.08467, over 13231.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2811, pruned_loss=0.05934, over 3808752.65 frames. ], batch size: 135, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:20:01,538 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=189214.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:20:12,590 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-03 12:20:33,008 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=189239.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:20:33,799 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-03 12:20:39,539 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-03 12:20:40,693 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-03 12:20:51,046 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-03 12:20:53,319 INFO [train.py:903] (0/4) Epoch 28, batch 4900, loss[loss=0.1591, simple_loss=0.2457, pruned_loss=0.0362, over 19838.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2806, pruned_loss=0.05903, over 3808186.12 frames. ], batch size: 52, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:21:05,674 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=189266.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:21:10,810 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-03 12:21:19,662 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.182e+02 4.742e+02 6.083e+02 7.635e+02 1.565e+03, threshold=1.217e+03, percent-clipped=2.0 +2023-04-03 12:21:35,164 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=189291.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:21:38,173 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=189293.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:21:53,523 INFO [train.py:903] (0/4) Epoch 28, batch 4950, loss[loss=0.2387, simple_loss=0.3156, pruned_loss=0.08087, over 18812.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2828, pruned_loss=0.0602, over 3815711.22 frames. ], batch size: 74, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:22:10,605 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-03 12:22:34,627 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-03 12:22:41,944 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5125, 1.5570, 1.8433, 1.7377, 2.4325, 2.2393, 2.5814, 0.9050], + device='cuda:0'), covar=tensor([0.2629, 0.4640, 0.2911, 0.2133, 0.1636, 0.2332, 0.1499, 0.5339], + device='cuda:0'), in_proj_covar=tensor([0.0556, 0.0673, 0.0758, 0.0508, 0.0637, 0.0548, 0.0673, 0.0575], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 12:22:56,070 INFO [train.py:903] (0/4) Epoch 28, batch 5000, loss[loss=0.1591, simple_loss=0.2431, pruned_loss=0.03761, over 19616.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.281, pruned_loss=0.05944, over 3828684.31 frames. ], batch size: 50, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:23:04,873 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-03 12:23:08,782 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=189365.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:23:15,753 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-03 12:23:24,845 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.397e+02 5.003e+02 6.019e+02 8.012e+02 1.544e+03, threshold=1.204e+03, percent-clipped=7.0 +2023-04-03 12:23:42,149 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=189392.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:23:59,724 INFO [train.py:903] (0/4) Epoch 28, batch 5050, loss[loss=0.1584, simple_loss=0.2348, pruned_loss=0.04102, over 19733.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2811, pruned_loss=0.05924, over 3835533.00 frames. ], batch size: 45, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:24:00,199 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=189406.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:24:02,597 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=189408.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:24:24,219 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=189426.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:24:33,187 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-03 12:24:44,062 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=189441.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:25:01,213 INFO [train.py:903] (0/4) Epoch 28, batch 5100, loss[loss=0.2094, simple_loss=0.2976, pruned_loss=0.06058, over 19466.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2814, pruned_loss=0.05896, over 3844187.76 frames. ], batch size: 64, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:25:10,106 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-03 12:25:12,426 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-03 12:25:19,115 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-03 12:25:27,902 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.508e+02 5.191e+02 6.540e+02 8.236e+02 1.634e+03, threshold=1.308e+03, percent-clipped=9.0 +2023-04-03 12:26:01,392 INFO [train.py:903] (0/4) Epoch 28, batch 5150, loss[loss=0.1896, simple_loss=0.28, pruned_loss=0.04964, over 19511.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2815, pruned_loss=0.059, over 3838285.51 frames. ], batch size: 64, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:26:02,864 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=189507.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:26:11,415 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-03 12:26:18,129 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=189519.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:26:34,989 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1423, 2.0274, 1.9327, 1.7871, 1.6087, 1.7154, 0.5659, 1.0877], + device='cuda:0'), covar=tensor([0.0643, 0.0663, 0.0524, 0.0869, 0.1221, 0.0925, 0.1476, 0.1170], + device='cuda:0'), in_proj_covar=tensor([0.0366, 0.0364, 0.0369, 0.0392, 0.0471, 0.0397, 0.0346, 0.0348], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 12:26:44,870 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=189541.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:26:46,842 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 12:27:02,661 INFO [train.py:903] (0/4) Epoch 28, batch 5200, loss[loss=0.2245, simple_loss=0.3057, pruned_loss=0.07171, over 19681.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2821, pruned_loss=0.05939, over 3821886.61 frames. ], batch size: 59, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:27:03,038 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=189556.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:27:04,112 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3829, 3.1178, 2.3149, 2.8001, 0.7391, 3.1083, 2.9639, 3.0536], + device='cuda:0'), covar=tensor([0.1051, 0.1330, 0.2058, 0.1096, 0.3921, 0.0992, 0.1156, 0.1387], + device='cuda:0'), in_proj_covar=tensor([0.0529, 0.0429, 0.0517, 0.0360, 0.0410, 0.0456, 0.0451, 0.0483], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:27:17,229 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-03 12:27:30,436 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.217e+02 4.850e+02 5.941e+02 7.550e+02 1.552e+03, threshold=1.188e+03, percent-clipped=2.0 +2023-04-03 12:28:01,626 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-03 12:28:04,870 INFO [train.py:903] (0/4) Epoch 28, batch 5250, loss[loss=0.2446, simple_loss=0.3159, pruned_loss=0.08666, over 19725.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2818, pruned_loss=0.05933, over 3823566.54 frames. ], batch size: 63, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:28:09,610 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=189610.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:29:04,829 INFO [train.py:903] (0/4) Epoch 28, batch 5300, loss[loss=0.1742, simple_loss=0.2554, pruned_loss=0.04646, over 19381.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2824, pruned_loss=0.05977, over 3826023.98 frames. ], batch size: 48, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:29:08,428 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=189658.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:29:12,954 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=189662.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:29:14,979 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=189664.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:29:21,477 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-03 12:29:31,784 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.136e+02 4.811e+02 6.188e+02 7.440e+02 1.460e+03, threshold=1.238e+03, percent-clipped=2.0 +2023-04-03 12:29:44,186 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=189687.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:29:46,156 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=189689.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:30:05,519 INFO [train.py:903] (0/4) Epoch 28, batch 5350, loss[loss=0.1919, simple_loss=0.2861, pruned_loss=0.04888, over 17151.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2833, pruned_loss=0.06024, over 3816153.84 frames. ], batch size: 101, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:30:09,128 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=189709.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:30:17,203 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8618, 2.0021, 2.2539, 2.4296, 1.9032, 2.3580, 2.2153, 2.0730], + device='cuda:0'), covar=tensor([0.4141, 0.3789, 0.1859, 0.2479, 0.3949, 0.2172, 0.4776, 0.3240], + device='cuda:0'), in_proj_covar=tensor([0.0940, 0.1018, 0.0745, 0.0954, 0.0916, 0.0858, 0.0863, 0.0812], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 12:30:17,966 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4419, 2.8280, 2.9874, 3.0125, 1.3622, 2.8421, 2.4695, 2.5988], + device='cuda:0'), covar=tensor([0.2949, 0.2415, 0.1502, 0.2019, 0.7715, 0.2705, 0.1574, 0.2484], + device='cuda:0'), in_proj_covar=tensor([0.0814, 0.0782, 0.0991, 0.0870, 0.0861, 0.0756, 0.0583, 0.0922], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 12:30:25,502 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=189722.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:30:29,868 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=189725.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:30:38,363 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-03 12:30:54,233 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=189746.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:31:04,004 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0377, 3.6971, 2.4962, 3.2822, 0.7846, 3.6735, 3.5342, 3.6265], + device='cuda:0'), covar=tensor([0.0793, 0.1024, 0.2036, 0.0978, 0.3956, 0.0781, 0.1016, 0.1228], + device='cuda:0'), in_proj_covar=tensor([0.0530, 0.0430, 0.0518, 0.0361, 0.0411, 0.0457, 0.0452, 0.0483], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:31:04,996 INFO [train.py:903] (0/4) Epoch 28, batch 5400, loss[loss=0.1812, simple_loss=0.2727, pruned_loss=0.04487, over 19605.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2836, pruned_loss=0.0604, over 3820004.68 frames. ], batch size: 57, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:31:15,157 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=189763.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:31:33,364 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.324e+02 4.839e+02 5.999e+02 7.745e+02 2.007e+03, threshold=1.200e+03, percent-clipped=4.0 +2023-04-03 12:31:44,951 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=189788.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:31:56,145 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=189797.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:32:07,667 INFO [train.py:903] (0/4) Epoch 28, batch 5450, loss[loss=0.2062, simple_loss=0.2927, pruned_loss=0.05981, over 19739.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2839, pruned_loss=0.06061, over 3803633.78 frames. ], batch size: 63, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:32:15,159 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=189812.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:32:27,696 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=189822.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:32:30,083 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=189824.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:32:45,824 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=189837.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:33:08,051 INFO [train.py:903] (0/4) Epoch 28, batch 5500, loss[loss=0.1573, simple_loss=0.2368, pruned_loss=0.03886, over 19742.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2836, pruned_loss=0.06042, over 3809346.05 frames. ], batch size: 45, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:33:17,135 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=189863.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:33:30,499 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-03 12:33:35,035 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.121e+02 4.755e+02 5.876e+02 8.325e+02 1.733e+03, threshold=1.175e+03, percent-clipped=4.0 +2023-04-03 12:34:07,021 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7929, 1.9178, 2.1456, 2.2272, 1.6737, 2.1405, 2.0987, 1.9733], + device='cuda:0'), covar=tensor([0.4237, 0.3670, 0.1994, 0.2509, 0.4091, 0.2286, 0.5283, 0.3566], + device='cuda:0'), in_proj_covar=tensor([0.0940, 0.1017, 0.0745, 0.0954, 0.0917, 0.0858, 0.0862, 0.0811], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 12:34:10,042 INFO [train.py:903] (0/4) Epoch 28, batch 5550, loss[loss=0.1567, simple_loss=0.2411, pruned_loss=0.03612, over 19798.00 frames. ], tot_loss[loss=0.2031, simple_loss=0.2843, pruned_loss=0.06095, over 3819012.58 frames. ], batch size: 48, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:34:17,019 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-03 12:35:06,754 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-03 12:35:11,463 INFO [train.py:903] (0/4) Epoch 28, batch 5600, loss[loss=0.2136, simple_loss=0.2992, pruned_loss=0.06396, over 19621.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2831, pruned_loss=0.05996, over 3832652.31 frames. ], batch size: 57, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:35:40,028 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.214e+02 4.597e+02 5.831e+02 7.652e+02 2.230e+03, threshold=1.166e+03, percent-clipped=4.0 +2023-04-03 12:35:40,321 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=189978.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:35:44,628 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=189981.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:36:06,871 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-190000.pt +2023-04-03 12:36:11,923 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=190002.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:36:17,374 INFO [train.py:903] (0/4) Epoch 28, batch 5650, loss[loss=0.2039, simple_loss=0.2901, pruned_loss=0.05881, over 19743.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2826, pruned_loss=0.05944, over 3833018.77 frames. ], batch size: 63, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:36:17,791 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=190006.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:37:04,788 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-03 12:37:18,832 INFO [train.py:903] (0/4) Epoch 28, batch 5700, loss[loss=0.2177, simple_loss=0.3032, pruned_loss=0.06607, over 19504.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2827, pruned_loss=0.05974, over 3823424.39 frames. ], batch size: 64, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:37:26,614 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-04-03 12:37:27,064 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5072, 3.7264, 4.0322, 4.0510, 2.3320, 3.7703, 3.4491, 3.8217], + device='cuda:0'), covar=tensor([0.1569, 0.2885, 0.0690, 0.0803, 0.4811, 0.1521, 0.0668, 0.1094], + device='cuda:0'), in_proj_covar=tensor([0.0822, 0.0788, 0.0999, 0.0879, 0.0869, 0.0765, 0.0589, 0.0928], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 12:37:31,550 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=190066.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:37:45,569 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.971e+02 5.283e+02 6.430e+02 7.892e+02 1.365e+03, threshold=1.286e+03, percent-clipped=3.0 +2023-04-03 12:37:49,415 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=190080.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:37:54,212 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2839, 2.3813, 2.4977, 3.0898, 2.4935, 3.0077, 2.4843, 2.2809], + device='cuda:0'), covar=tensor([0.4309, 0.3901, 0.2051, 0.2481, 0.3977, 0.2153, 0.5007, 0.3530], + device='cuda:0'), in_proj_covar=tensor([0.0940, 0.1018, 0.0745, 0.0955, 0.0917, 0.0858, 0.0862, 0.0811], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 12:38:02,430 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=190090.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:38:21,051 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=190105.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:38:21,824 INFO [train.py:903] (0/4) Epoch 28, batch 5750, loss[loss=0.177, simple_loss=0.2595, pruned_loss=0.04727, over 19027.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2826, pruned_loss=0.05961, over 3816413.16 frames. ], batch size: 42, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:38:24,226 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-03 12:38:32,488 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-03 12:38:35,207 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=190117.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:38:37,078 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-03 12:38:45,177 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4469, 1.5395, 1.8927, 1.6946, 2.5806, 2.2224, 2.7745, 1.0794], + device='cuda:0'), covar=tensor([0.2557, 0.4353, 0.2773, 0.2038, 0.1505, 0.2242, 0.1400, 0.4930], + device='cuda:0'), in_proj_covar=tensor([0.0558, 0.0675, 0.0760, 0.0512, 0.0640, 0.0551, 0.0674, 0.0579], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 12:39:23,781 INFO [train.py:903] (0/4) Epoch 28, batch 5800, loss[loss=0.1877, simple_loss=0.2649, pruned_loss=0.05528, over 19481.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2831, pruned_loss=0.05979, over 3825547.30 frames. ], batch size: 49, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:39:25,231 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=190157.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 12:39:41,042 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=190169.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:39:52,006 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.261e+02 4.681e+02 5.738e+02 7.022e+02 1.341e+03, threshold=1.148e+03, percent-clipped=1.0 +2023-04-03 12:39:56,698 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=190181.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:40:01,278 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0812, 5.5078, 3.2412, 4.8426, 1.4257, 5.7180, 5.5361, 5.7079], + device='cuda:0'), covar=tensor([0.0374, 0.0931, 0.1802, 0.0770, 0.3751, 0.0516, 0.0798, 0.1045], + device='cuda:0'), in_proj_covar=tensor([0.0531, 0.0431, 0.0519, 0.0361, 0.0411, 0.0458, 0.0451, 0.0484], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:40:26,166 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=190205.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:40:27,628 INFO [train.py:903] (0/4) Epoch 28, batch 5850, loss[loss=0.175, simple_loss=0.2498, pruned_loss=0.05005, over 19742.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2835, pruned_loss=0.05975, over 3821835.05 frames. ], batch size: 47, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:41:02,540 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=190234.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:41:29,993 INFO [train.py:903] (0/4) Epoch 28, batch 5900, loss[loss=0.2088, simple_loss=0.2935, pruned_loss=0.06211, over 19385.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2831, pruned_loss=0.05938, over 3828594.77 frames. ], batch size: 70, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:41:32,305 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-03 12:41:32,753 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0182, 1.9528, 1.9126, 1.7036, 1.6455, 1.6777, 0.4992, 0.9136], + device='cuda:0'), covar=tensor([0.0673, 0.0670, 0.0445, 0.0757, 0.1210, 0.0830, 0.1418, 0.1195], + device='cuda:0'), in_proj_covar=tensor([0.0367, 0.0366, 0.0367, 0.0393, 0.0471, 0.0398, 0.0346, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 12:41:33,912 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=190259.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:41:54,172 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-03 12:41:56,488 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.705e+02 4.583e+02 5.674e+02 7.520e+02 1.844e+03, threshold=1.135e+03, percent-clipped=9.0 +2023-04-03 12:42:24,248 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.43 vs. limit=2.0 +2023-04-03 12:42:31,979 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8433, 1.9514, 2.1851, 2.3305, 1.7894, 2.2331, 2.1676, 2.0635], + device='cuda:0'), covar=tensor([0.4131, 0.3853, 0.2044, 0.2406, 0.4012, 0.2306, 0.5128, 0.3475], + device='cuda:0'), in_proj_covar=tensor([0.0938, 0.1017, 0.0746, 0.0955, 0.0916, 0.0857, 0.0862, 0.0810], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 12:42:32,653 INFO [train.py:903] (0/4) Epoch 28, batch 5950, loss[loss=0.2234, simple_loss=0.3024, pruned_loss=0.07224, over 18730.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2825, pruned_loss=0.05927, over 3842259.75 frames. ], batch size: 74, lr: 2.91e-03, grad_scale: 8.0 +2023-04-03 12:43:07,998 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=190334.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:43:13,912 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6264, 1.5104, 1.6251, 1.6313, 3.1874, 1.1721, 2.4580, 3.7049], + device='cuda:0'), covar=tensor([0.0495, 0.2669, 0.2829, 0.1788, 0.0688, 0.2524, 0.1305, 0.0234], + device='cuda:0'), in_proj_covar=tensor([0.0425, 0.0379, 0.0399, 0.0354, 0.0383, 0.0359, 0.0399, 0.0420], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:43:34,404 INFO [train.py:903] (0/4) Epoch 28, batch 6000, loss[loss=0.1956, simple_loss=0.2853, pruned_loss=0.05302, over 19534.00 frames. ], tot_loss[loss=0.2, simple_loss=0.282, pruned_loss=0.05894, over 3824820.67 frames. ], batch size: 56, lr: 2.90e-03, grad_scale: 8.0 +2023-04-03 12:43:34,404 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 12:43:48,500 INFO [train.py:937] (0/4) Epoch 28, validation: loss=0.1668, simple_loss=0.2663, pruned_loss=0.03368, over 944034.00 frames. +2023-04-03 12:43:48,501 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 12:44:08,734 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=190373.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:44:15,152 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.216e+02 4.800e+02 5.909e+02 8.119e+02 1.607e+03, threshold=1.182e+03, percent-clipped=4.0 +2023-04-03 12:44:41,353 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=190398.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:44:50,436 INFO [train.py:903] (0/4) Epoch 28, batch 6050, loss[loss=0.2895, simple_loss=0.3517, pruned_loss=0.1137, over 18603.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2824, pruned_loss=0.05933, over 3821870.68 frames. ], batch size: 84, lr: 2.90e-03, grad_scale: 8.0 +2023-04-03 12:45:04,557 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=190417.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:45:11,941 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.1333, 5.5967, 2.9027, 4.8824, 1.3330, 5.8376, 5.5603, 5.7521], + device='cuda:0'), covar=tensor([0.0382, 0.0802, 0.1992, 0.0802, 0.3611, 0.0463, 0.0770, 0.0914], + device='cuda:0'), in_proj_covar=tensor([0.0531, 0.0431, 0.0519, 0.0361, 0.0411, 0.0458, 0.0450, 0.0483], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:45:28,529 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=190437.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:45:50,732 INFO [train.py:903] (0/4) Epoch 28, batch 6100, loss[loss=0.1851, simple_loss=0.2558, pruned_loss=0.05723, over 19764.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2818, pruned_loss=0.05883, over 3833359.26 frames. ], batch size: 47, lr: 2.90e-03, grad_scale: 8.0 +2023-04-03 12:45:57,913 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=190461.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:45:58,976 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=190462.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:46:17,944 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.255e+02 4.836e+02 5.950e+02 7.277e+02 1.439e+03, threshold=1.190e+03, percent-clipped=1.0 +2023-04-03 12:46:21,259 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.62 vs. limit=5.0 +2023-04-03 12:46:28,780 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=190486.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:46:46,946 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=190501.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 12:46:53,523 INFO [train.py:903] (0/4) Epoch 28, batch 6150, loss[loss=0.1802, simple_loss=0.2545, pruned_loss=0.05294, over 19770.00 frames. ], tot_loss[loss=0.2, simple_loss=0.2817, pruned_loss=0.05909, over 3832239.97 frames. ], batch size: 46, lr: 2.90e-03, grad_scale: 8.0 +2023-04-03 12:47:02,041 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=190513.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:47:23,080 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-03 12:47:56,358 INFO [train.py:903] (0/4) Epoch 28, batch 6200, loss[loss=0.2379, simple_loss=0.3168, pruned_loss=0.07949, over 19061.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2818, pruned_loss=0.05905, over 3821871.25 frames. ], batch size: 69, lr: 2.90e-03, grad_scale: 8.0 +2023-04-03 12:48:23,170 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.156e+02 4.897e+02 5.800e+02 7.277e+02 1.783e+03, threshold=1.160e+03, percent-clipped=5.0 +2023-04-03 12:48:59,514 INFO [train.py:903] (0/4) Epoch 28, batch 6250, loss[loss=0.224, simple_loss=0.302, pruned_loss=0.073, over 17549.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2814, pruned_loss=0.05866, over 3814422.76 frames. ], batch size: 101, lr: 2.90e-03, grad_scale: 16.0 +2023-04-03 12:49:03,304 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=190609.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 12:49:11,045 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=190616.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 12:49:26,251 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=190627.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:49:27,618 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=190628.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:49:31,959 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-03 12:49:58,804 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 12:50:00,489 INFO [train.py:903] (0/4) Epoch 28, batch 6300, loss[loss=0.1881, simple_loss=0.2753, pruned_loss=0.05048, over 19650.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2818, pruned_loss=0.05884, over 3823742.01 frames. ], batch size: 55, lr: 2.90e-03, grad_scale: 8.0 +2023-04-03 12:50:28,870 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=190678.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:50:29,806 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.019e+02 4.803e+02 6.276e+02 7.910e+02 2.564e+03, threshold=1.255e+03, percent-clipped=8.0 +2023-04-03 12:50:47,619 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=190693.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:51:04,091 INFO [train.py:903] (0/4) Epoch 28, batch 6350, loss[loss=0.2119, simple_loss=0.3019, pruned_loss=0.061, over 19526.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2807, pruned_loss=0.05823, over 3819297.94 frames. ], batch size: 56, lr: 2.90e-03, grad_scale: 8.0 +2023-04-03 12:52:06,519 INFO [train.py:903] (0/4) Epoch 28, batch 6400, loss[loss=0.2081, simple_loss=0.2984, pruned_loss=0.05892, over 19531.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2805, pruned_loss=0.05787, over 3828454.93 frames. ], batch size: 54, lr: 2.90e-03, grad_scale: 8.0 +2023-04-03 12:52:13,578 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=190761.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:52:36,099 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.096e+02 4.755e+02 6.032e+02 8.092e+02 1.400e+03, threshold=1.206e+03, percent-clipped=2.0 +2023-04-03 12:52:52,684 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=190793.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:53:08,682 INFO [train.py:903] (0/4) Epoch 28, batch 6450, loss[loss=0.1832, simple_loss=0.2735, pruned_loss=0.0465, over 19653.00 frames. ], tot_loss[loss=0.1982, simple_loss=0.2804, pruned_loss=0.05804, over 3828085.94 frames. ], batch size: 55, lr: 2.90e-03, grad_scale: 4.0 +2023-04-03 12:53:09,393 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-03 12:53:55,217 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-03 12:54:10,281 INFO [train.py:903] (0/4) Epoch 28, batch 6500, loss[loss=0.1986, simple_loss=0.2852, pruned_loss=0.05601, over 17380.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2808, pruned_loss=0.05821, over 3822605.27 frames. ], batch size: 101, lr: 2.90e-03, grad_scale: 4.0 +2023-04-03 12:54:17,926 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-03 12:54:32,041 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=190872.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 12:54:36,560 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=190876.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:54:36,683 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8373, 1.9195, 2.0740, 2.3281, 1.9071, 2.2698, 2.1091, 2.0071], + device='cuda:0'), covar=tensor([0.3542, 0.3134, 0.1760, 0.2036, 0.3285, 0.1827, 0.3859, 0.2731], + device='cuda:0'), in_proj_covar=tensor([0.0939, 0.1018, 0.0746, 0.0954, 0.0915, 0.0857, 0.0861, 0.0809], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 12:54:40,960 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.047e+02 4.758e+02 5.991e+02 7.680e+02 1.999e+03, threshold=1.198e+03, percent-clipped=5.0 +2023-04-03 12:54:47,189 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=190884.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:54:49,298 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4559, 2.3785, 2.1944, 2.6245, 2.3471, 2.0194, 1.9808, 2.4491], + device='cuda:0'), covar=tensor([0.1004, 0.1623, 0.1443, 0.0988, 0.1336, 0.0583, 0.1564, 0.0687], + device='cuda:0'), in_proj_covar=tensor([0.0276, 0.0358, 0.0316, 0.0256, 0.0304, 0.0253, 0.0319, 0.0260], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:55:01,700 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=190897.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 12:55:13,652 INFO [train.py:903] (0/4) Epoch 28, batch 6550, loss[loss=0.1938, simple_loss=0.2693, pruned_loss=0.05916, over 19562.00 frames. ], tot_loss[loss=0.2, simple_loss=0.282, pruned_loss=0.05896, over 3817992.43 frames. ], batch size: 52, lr: 2.90e-03, grad_scale: 4.0 +2023-04-03 12:55:18,344 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=190909.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:55:45,215 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0122, 3.6774, 2.6401, 3.2023, 0.6932, 3.6336, 3.4811, 3.5884], + device='cuda:0'), covar=tensor([0.0811, 0.1124, 0.1919, 0.1059, 0.4193, 0.0808, 0.1028, 0.1254], + device='cuda:0'), in_proj_covar=tensor([0.0534, 0.0433, 0.0521, 0.0362, 0.0411, 0.0459, 0.0452, 0.0485], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 12:56:14,193 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=190953.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 12:56:17,101 INFO [train.py:903] (0/4) Epoch 28, batch 6600, loss[loss=0.2154, simple_loss=0.3005, pruned_loss=0.0651, over 19519.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2822, pruned_loss=0.05908, over 3832716.07 frames. ], batch size: 56, lr: 2.90e-03, grad_scale: 4.0 +2023-04-03 12:56:35,867 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=190971.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:56:48,224 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.060e+02 4.870e+02 6.050e+02 8.150e+02 2.542e+03, threshold=1.210e+03, percent-clipped=13.0 +2023-04-03 12:56:50,203 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.23 vs. limit=2.0 +2023-04-03 12:56:53,412 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2712, 1.3458, 1.9338, 1.7062, 2.8047, 4.6013, 4.4427, 5.1804], + device='cuda:0'), covar=tensor([0.1940, 0.5216, 0.4516, 0.2599, 0.0828, 0.0227, 0.0233, 0.0220], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0335, 0.0367, 0.0273, 0.0257, 0.0198, 0.0222, 0.0280], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 12:57:19,914 INFO [train.py:903] (0/4) Epoch 28, batch 6650, loss[loss=0.2339, simple_loss=0.3189, pruned_loss=0.07449, over 19540.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2824, pruned_loss=0.0593, over 3837826.00 frames. ], batch size: 56, lr: 2.90e-03, grad_scale: 2.0 +2023-04-03 12:57:54,087 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.80 vs. limit=5.0 +2023-04-03 12:57:59,368 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=191037.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:58:14,533 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=191049.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:58:22,325 INFO [train.py:903] (0/4) Epoch 28, batch 6700, loss[loss=0.1689, simple_loss=0.2476, pruned_loss=0.04506, over 19749.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2819, pruned_loss=0.05914, over 3828178.91 frames. ], batch size: 46, lr: 2.90e-03, grad_scale: 2.0 +2023-04-03 12:58:38,427 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=191068.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 12:58:45,532 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=191074.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:58:52,989 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.114e+02 5.245e+02 6.198e+02 8.145e+02 2.088e+03, threshold=1.240e+03, percent-clipped=7.0 +2023-04-03 12:58:58,999 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=191086.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:58:59,066 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=191086.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 12:59:21,821 INFO [train.py:903] (0/4) Epoch 28, batch 6750, loss[loss=0.175, simple_loss=0.2609, pruned_loss=0.04454, over 19668.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2832, pruned_loss=0.05978, over 3805924.69 frames. ], batch size: 53, lr: 2.90e-03, grad_scale: 2.0 +2023-04-03 12:59:51,524 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=191132.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:00:11,310 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=191150.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:00:13,715 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=191152.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:00:17,983 INFO [train.py:903] (0/4) Epoch 28, batch 6800, loss[loss=0.1961, simple_loss=0.2794, pruned_loss=0.05637, over 19536.00 frames. ], tot_loss[loss=0.202, simple_loss=0.2838, pruned_loss=0.06011, over 3816233.78 frames. ], batch size: 54, lr: 2.90e-03, grad_scale: 4.0 +2023-04-03 13:00:19,490 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=191157.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:00:45,263 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.070e+02 5.099e+02 6.292e+02 8.518e+02 1.501e+03, threshold=1.258e+03, percent-clipped=3.0 +2023-04-03 13:00:48,674 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-28.pt +2023-04-03 13:01:05,957 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-03 13:01:06,445 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-03 13:01:08,828 INFO [train.py:903] (0/4) Epoch 29, batch 0, loss[loss=0.2003, simple_loss=0.2883, pruned_loss=0.05617, over 18823.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2883, pruned_loss=0.05617, over 18823.00 frames. ], batch size: 74, lr: 2.85e-03, grad_scale: 8.0 +2023-04-03 13:01:08,829 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 13:01:20,496 INFO [train.py:937] (0/4) Epoch 29, validation: loss=0.1669, simple_loss=0.2669, pruned_loss=0.03339, over 944034.00 frames. +2023-04-03 13:01:20,497 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 13:01:31,763 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-03 13:01:43,687 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=191203.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:01:46,195 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6549, 1.7459, 2.0120, 1.9767, 1.5393, 1.9562, 2.0154, 1.8810], + device='cuda:0'), covar=tensor([0.4020, 0.3599, 0.2026, 0.2367, 0.3737, 0.2223, 0.5053, 0.3505], + device='cuda:0'), in_proj_covar=tensor([0.0946, 0.1023, 0.0750, 0.0960, 0.0921, 0.0863, 0.0868, 0.0815], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 13:02:19,669 INFO [train.py:903] (0/4) Epoch 29, batch 50, loss[loss=0.202, simple_loss=0.2879, pruned_loss=0.05809, over 19742.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.2811, pruned_loss=0.05866, over 872862.15 frames. ], batch size: 63, lr: 2.85e-03, grad_scale: 8.0 +2023-04-03 13:02:55,005 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-03 13:03:15,130 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.166e+02 5.210e+02 6.215e+02 8.429e+02 1.754e+03, threshold=1.243e+03, percent-clipped=6.0 +2023-04-03 13:03:18,590 INFO [train.py:903] (0/4) Epoch 29, batch 100, loss[loss=0.1833, simple_loss=0.2713, pruned_loss=0.04765, over 19540.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2822, pruned_loss=0.0588, over 1526969.76 frames. ], batch size: 56, lr: 2.85e-03, grad_scale: 8.0 +2023-04-03 13:03:33,067 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-03 13:04:07,329 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=191324.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 13:04:19,667 INFO [train.py:903] (0/4) Epoch 29, batch 150, loss[loss=0.1764, simple_loss=0.2502, pruned_loss=0.05131, over 19301.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2833, pruned_loss=0.05964, over 2013027.66 frames. ], batch size: 44, lr: 2.85e-03, grad_scale: 8.0 +2023-04-03 13:04:29,084 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=191342.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:04:36,951 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=191349.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 13:04:59,838 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=191367.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:05:15,368 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.397e+02 5.272e+02 6.303e+02 7.742e+02 1.475e+03, threshold=1.261e+03, percent-clipped=3.0 +2023-04-03 13:05:15,449 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-03 13:05:18,068 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=191383.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:05:18,900 INFO [train.py:903] (0/4) Epoch 29, batch 200, loss[loss=0.2178, simple_loss=0.2946, pruned_loss=0.07046, over 19795.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2836, pruned_loss=0.05994, over 2425215.77 frames. ], batch size: 56, lr: 2.85e-03, grad_scale: 8.0 +2023-04-03 13:05:37,012 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3029, 2.9324, 2.2858, 2.3180, 2.2971, 2.5119, 0.8850, 2.1395], + device='cuda:0'), covar=tensor([0.0709, 0.0625, 0.0744, 0.1228, 0.1022, 0.1191, 0.1610, 0.1147], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0366, 0.0367, 0.0393, 0.0471, 0.0398, 0.0346, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 13:05:48,138 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=191408.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:06:13,703 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=191430.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:06:17,621 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=191433.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:06:20,059 INFO [train.py:903] (0/4) Epoch 29, batch 250, loss[loss=0.1883, simple_loss=0.2725, pruned_loss=0.05204, over 19664.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2816, pruned_loss=0.05901, over 2748616.44 frames. ], batch size: 53, lr: 2.85e-03, grad_scale: 8.0 +2023-04-03 13:06:41,362 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.34 vs. limit=2.0 +2023-04-03 13:07:16,621 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.382e+02 5.311e+02 6.503e+02 8.385e+02 2.446e+03, threshold=1.301e+03, percent-clipped=7.0 +2023-04-03 13:07:20,125 INFO [train.py:903] (0/4) Epoch 29, batch 300, loss[loss=0.1947, simple_loss=0.2739, pruned_loss=0.05776, over 19833.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2819, pruned_loss=0.05939, over 2979696.08 frames. ], batch size: 52, lr: 2.85e-03, grad_scale: 8.0 +2023-04-03 13:07:33,147 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=191494.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:08:19,686 INFO [train.py:903] (0/4) Epoch 29, batch 350, loss[loss=0.1992, simple_loss=0.2842, pruned_loss=0.05711, over 19335.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2828, pruned_loss=0.0597, over 3170914.21 frames. ], batch size: 66, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:08:27,197 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 13:08:33,067 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=191545.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:08:35,154 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=191547.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:09:16,624 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.400e+02 4.772e+02 5.697e+02 6.938e+02 1.378e+03, threshold=1.139e+03, percent-clipped=1.0 +2023-04-03 13:09:19,962 INFO [train.py:903] (0/4) Epoch 29, batch 400, loss[loss=0.2231, simple_loss=0.3019, pruned_loss=0.07212, over 19657.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2809, pruned_loss=0.05896, over 3322485.48 frames. ], batch size: 58, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:09:50,409 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=191609.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:10:11,890 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.72 vs. limit=5.0 +2023-04-03 13:10:19,945 INFO [train.py:903] (0/4) Epoch 29, batch 450, loss[loss=0.235, simple_loss=0.3042, pruned_loss=0.08287, over 13449.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2817, pruned_loss=0.05911, over 3434051.01 frames. ], batch size: 135, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:10:55,256 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=191662.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:10:57,246 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-03 13:10:58,386 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-03 13:11:17,385 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.061e+02 4.702e+02 5.973e+02 7.083e+02 1.398e+03, threshold=1.195e+03, percent-clipped=2.0 +2023-04-03 13:11:20,347 INFO [train.py:903] (0/4) Epoch 29, batch 500, loss[loss=0.1817, simple_loss=0.2593, pruned_loss=0.0521, over 19785.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2813, pruned_loss=0.05891, over 3518217.97 frames. ], batch size: 45, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:12:13,366 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=191727.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:12:21,066 INFO [train.py:903] (0/4) Epoch 29, batch 550, loss[loss=0.2079, simple_loss=0.2914, pruned_loss=0.06215, over 18220.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2816, pruned_loss=0.05947, over 3581040.59 frames. ], batch size: 83, lr: 2.84e-03, grad_scale: 4.0 +2023-04-03 13:12:45,860 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2717, 2.0061, 1.9369, 2.1795, 1.7816, 1.8891, 1.8086, 2.1956], + device='cuda:0'), covar=tensor([0.1067, 0.1536, 0.1503, 0.1097, 0.1549, 0.0578, 0.1563, 0.0713], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0362, 0.0320, 0.0259, 0.0307, 0.0256, 0.0323, 0.0262], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 13:13:04,063 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8793, 1.1925, 1.5389, 0.6472, 1.8558, 2.2385, 1.9701, 2.3331], + device='cuda:0'), covar=tensor([0.1577, 0.3629, 0.3144, 0.2846, 0.0851, 0.0346, 0.0371, 0.0443], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0334, 0.0366, 0.0272, 0.0256, 0.0198, 0.0221, 0.0279], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 13:13:04,169 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2699, 1.3050, 1.4462, 1.4091, 1.6527, 1.7366, 1.6479, 0.6429], + device='cuda:0'), covar=tensor([0.2293, 0.3890, 0.2448, 0.1860, 0.1543, 0.2155, 0.1432, 0.4968], + device='cuda:0'), in_proj_covar=tensor([0.0555, 0.0672, 0.0759, 0.0509, 0.0635, 0.0546, 0.0671, 0.0575], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 13:13:06,427 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4467, 1.4989, 1.8737, 1.7514, 2.7986, 2.3295, 2.8950, 1.4504], + device='cuda:0'), covar=tensor([0.2599, 0.4426, 0.2761, 0.1921, 0.1372, 0.2174, 0.1400, 0.4341], + device='cuda:0'), in_proj_covar=tensor([0.0555, 0.0672, 0.0759, 0.0509, 0.0635, 0.0546, 0.0671, 0.0575], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 13:13:19,228 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.607e+02 5.419e+02 6.471e+02 8.968e+02 1.855e+03, threshold=1.294e+03, percent-clipped=10.0 +2023-04-03 13:13:22,383 INFO [train.py:903] (0/4) Epoch 29, batch 600, loss[loss=0.1732, simple_loss=0.2562, pruned_loss=0.04511, over 19752.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2818, pruned_loss=0.05938, over 3631455.58 frames. ], batch size: 51, lr: 2.84e-03, grad_scale: 4.0 +2023-04-03 13:13:41,790 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=191801.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:14:02,329 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-03 13:14:13,249 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=191826.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:14:22,136 INFO [train.py:903] (0/4) Epoch 29, batch 650, loss[loss=0.1674, simple_loss=0.2482, pruned_loss=0.04333, over 19760.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2834, pruned_loss=0.06058, over 3655996.87 frames. ], batch size: 47, lr: 2.84e-03, grad_scale: 4.0 +2023-04-03 13:14:31,527 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=191842.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:14:42,693 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.53 vs. limit=2.0 +2023-04-03 13:15:01,021 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=191865.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:15:18,381 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.813e+02 4.883e+02 6.286e+02 7.850e+02 2.365e+03, threshold=1.257e+03, percent-clipped=4.0 +2023-04-03 13:15:21,601 INFO [train.py:903] (0/4) Epoch 29, batch 700, loss[loss=0.1974, simple_loss=0.2742, pruned_loss=0.06031, over 19492.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2836, pruned_loss=0.06048, over 3702764.79 frames. ], batch size: 49, lr: 2.84e-03, grad_scale: 4.0 +2023-04-03 13:15:29,636 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=191890.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:16:03,817 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=191918.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:16:23,392 INFO [train.py:903] (0/4) Epoch 29, batch 750, loss[loss=0.2083, simple_loss=0.2925, pruned_loss=0.06211, over 19546.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2817, pruned_loss=0.05944, over 3721460.19 frames. ], batch size: 56, lr: 2.84e-03, grad_scale: 4.0 +2023-04-03 13:16:34,311 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=191943.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:17:21,355 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.970e+02 4.754e+02 5.591e+02 7.301e+02 1.189e+03, threshold=1.118e+03, percent-clipped=0.0 +2023-04-03 13:17:23,701 INFO [train.py:903] (0/4) Epoch 29, batch 800, loss[loss=0.2078, simple_loss=0.2919, pruned_loss=0.06183, over 19729.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2814, pruned_loss=0.059, over 3749317.86 frames. ], batch size: 63, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:17:24,007 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3774, 3.8577, 3.9771, 3.9827, 1.6141, 3.8097, 3.2877, 3.7413], + device='cuda:0'), covar=tensor([0.1664, 0.0882, 0.0705, 0.0849, 0.6020, 0.0972, 0.0777, 0.1200], + device='cuda:0'), in_proj_covar=tensor([0.0821, 0.0786, 0.0994, 0.0877, 0.0864, 0.0762, 0.0588, 0.0927], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 13:17:36,666 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-03 13:17:42,712 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-192000.pt +2023-04-03 13:18:16,913 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3258, 2.1386, 2.0815, 2.1813, 1.9153, 1.9373, 1.8602, 2.1940], + device='cuda:0'), covar=tensor([0.0951, 0.1391, 0.1347, 0.1096, 0.1405, 0.0558, 0.1503, 0.0701], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0362, 0.0321, 0.0260, 0.0307, 0.0257, 0.0323, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 13:18:26,430 INFO [train.py:903] (0/4) Epoch 29, batch 850, loss[loss=0.1794, simple_loss=0.2618, pruned_loss=0.04853, over 19485.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2807, pruned_loss=0.05851, over 3763863.88 frames. ], batch size: 49, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:19:17,573 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-03 13:19:23,135 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.899e+02 4.771e+02 5.573e+02 7.809e+02 2.111e+03, threshold=1.115e+03, percent-clipped=7.0 +2023-04-03 13:19:25,393 INFO [train.py:903] (0/4) Epoch 29, batch 900, loss[loss=0.1781, simple_loss=0.2655, pruned_loss=0.04541, over 19537.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2814, pruned_loss=0.05895, over 3780780.47 frames. ], batch size: 54, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:19:43,249 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=192098.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:20:07,630 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.87 vs. limit=2.0 +2023-04-03 13:20:09,363 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192120.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:20:12,946 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=192123.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:20:26,583 INFO [train.py:903] (0/4) Epoch 29, batch 950, loss[loss=0.1869, simple_loss=0.2741, pruned_loss=0.04982, over 19777.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2809, pruned_loss=0.05865, over 3781050.28 frames. ], batch size: 54, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:20:30,184 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-03 13:21:13,667 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192173.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:21:24,651 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.322e+02 5.272e+02 6.130e+02 7.632e+02 1.213e+03, threshold=1.226e+03, percent-clipped=2.0 +2023-04-03 13:21:27,729 INFO [train.py:903] (0/4) Epoch 29, batch 1000, loss[loss=0.1573, simple_loss=0.2443, pruned_loss=0.03512, over 19402.00 frames. ], tot_loss[loss=0.199, simple_loss=0.2808, pruned_loss=0.05858, over 3792073.79 frames. ], batch size: 48, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:22:23,176 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-03 13:22:27,654 INFO [train.py:903] (0/4) Epoch 29, batch 1050, loss[loss=0.1964, simple_loss=0.2865, pruned_loss=0.05322, over 19657.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2802, pruned_loss=0.05852, over 3796295.25 frames. ], batch size: 58, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:22:31,117 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192237.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:22:34,317 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192239.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:23:02,357 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-03 13:23:25,659 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.003e+02 4.755e+02 5.709e+02 7.263e+02 1.610e+03, threshold=1.142e+03, percent-clipped=3.0 +2023-04-03 13:23:28,075 INFO [train.py:903] (0/4) Epoch 29, batch 1100, loss[loss=0.1808, simple_loss=0.2625, pruned_loss=0.04959, over 19591.00 frames. ], tot_loss[loss=0.199, simple_loss=0.2806, pruned_loss=0.05871, over 3793710.27 frames. ], batch size: 52, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:24:29,310 INFO [train.py:903] (0/4) Epoch 29, batch 1150, loss[loss=0.1828, simple_loss=0.2711, pruned_loss=0.04726, over 19687.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2807, pruned_loss=0.05859, over 3804476.52 frames. ], batch size: 59, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:24:53,834 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5786, 1.5425, 1.8954, 1.8280, 2.7352, 2.4404, 2.9183, 1.4433], + device='cuda:0'), covar=tensor([0.2490, 0.4485, 0.2808, 0.1924, 0.1568, 0.2070, 0.1447, 0.4508], + device='cuda:0'), in_proj_covar=tensor([0.0555, 0.0672, 0.0760, 0.0511, 0.0635, 0.0545, 0.0670, 0.0574], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 13:25:27,745 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.011e+02 4.839e+02 5.803e+02 7.271e+02 1.538e+03, threshold=1.161e+03, percent-clipped=1.0 +2023-04-03 13:25:30,850 INFO [train.py:903] (0/4) Epoch 29, batch 1200, loss[loss=0.1972, simple_loss=0.2825, pruned_loss=0.05598, over 19690.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2811, pruned_loss=0.05907, over 3802958.49 frames. ], batch size: 60, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:26:01,109 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192410.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:26:01,911 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-03 13:26:10,325 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192417.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:26:30,220 INFO [train.py:903] (0/4) Epoch 29, batch 1250, loss[loss=0.1824, simple_loss=0.2726, pruned_loss=0.04609, over 18203.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2809, pruned_loss=0.05892, over 3794555.45 frames. ], batch size: 83, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:26:50,089 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192451.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:27:07,149 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=192464.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:27:28,110 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.043e+02 5.075e+02 6.202e+02 7.494e+02 1.680e+03, threshold=1.240e+03, percent-clipped=4.0 +2023-04-03 13:27:30,144 INFO [train.py:903] (0/4) Epoch 29, batch 1300, loss[loss=0.1875, simple_loss=0.2815, pruned_loss=0.04676, over 19512.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2818, pruned_loss=0.05916, over 3793275.64 frames. ], batch size: 54, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:28:10,143 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=192517.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:28:14,833 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2301, 2.0941, 1.9529, 1.8727, 1.6310, 1.7372, 0.5581, 1.1780], + device='cuda:0'), covar=tensor([0.0655, 0.0691, 0.0549, 0.0947, 0.1264, 0.1086, 0.1529, 0.1220], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0365, 0.0367, 0.0391, 0.0471, 0.0398, 0.0345, 0.0347], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 13:28:15,767 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3470, 3.9808, 2.5490, 3.5211, 1.0229, 3.9287, 3.8120, 3.8859], + device='cuda:0'), covar=tensor([0.0724, 0.1031, 0.2096, 0.0921, 0.3935, 0.0747, 0.1014, 0.1194], + device='cuda:0'), in_proj_covar=tensor([0.0531, 0.0431, 0.0519, 0.0361, 0.0410, 0.0457, 0.0452, 0.0482], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 13:28:30,908 INFO [train.py:903] (0/4) Epoch 29, batch 1350, loss[loss=0.1705, simple_loss=0.2528, pruned_loss=0.04407, over 19631.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2825, pruned_loss=0.0597, over 3802828.02 frames. ], batch size: 50, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:28:45,937 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6356, 1.3259, 1.4504, 2.2449, 1.6015, 1.7386, 1.7960, 1.5849], + device='cuda:0'), covar=tensor([0.0884, 0.1115, 0.1073, 0.0695, 0.0909, 0.0885, 0.0909, 0.0814], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0222, 0.0229, 0.0239, 0.0226, 0.0214, 0.0187, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 13:29:25,922 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=192579.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:29:27,901 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=192581.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:29:28,924 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.238e+02 5.533e+02 7.774e+02 1.028e+03 2.542e+03, threshold=1.555e+03, percent-clipped=13.0 +2023-04-03 13:29:30,284 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=192583.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:29:32,193 INFO [train.py:903] (0/4) Epoch 29, batch 1400, loss[loss=0.1998, simple_loss=0.276, pruned_loss=0.06187, over 19402.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2824, pruned_loss=0.05992, over 3809732.65 frames. ], batch size: 48, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:30:30,380 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=192632.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:30:32,433 INFO [train.py:903] (0/4) Epoch 29, batch 1450, loss[loss=0.1871, simple_loss=0.2704, pruned_loss=0.05194, over 19578.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2815, pruned_loss=0.05937, over 3820049.80 frames. ], batch size: 61, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:30:32,467 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-03 13:30:32,707 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192634.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:30:46,938 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2269, 1.4113, 1.9666, 1.7467, 3.0772, 4.5419, 4.3991, 4.9830], + device='cuda:0'), covar=tensor([0.1753, 0.4041, 0.3599, 0.2388, 0.0645, 0.0214, 0.0183, 0.0219], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0335, 0.0368, 0.0274, 0.0257, 0.0198, 0.0221, 0.0279], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 13:31:29,875 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.888e+02 4.834e+02 5.734e+02 7.377e+02 1.406e+03, threshold=1.147e+03, percent-clipped=0.0 +2023-04-03 13:31:32,942 INFO [train.py:903] (0/4) Epoch 29, batch 1500, loss[loss=0.181, simple_loss=0.264, pruned_loss=0.04902, over 19595.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2817, pruned_loss=0.05927, over 3829834.43 frames. ], batch size: 50, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:31:41,373 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192691.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:31:47,423 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=192696.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:31:50,540 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=192698.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:31:59,313 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7837, 1.6713, 1.6535, 2.3311, 1.7049, 1.9741, 2.0620, 1.8178], + device='cuda:0'), covar=tensor([0.0815, 0.0897, 0.0981, 0.0708, 0.0875, 0.0821, 0.0867, 0.0708], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0221, 0.0228, 0.0239, 0.0225, 0.0214, 0.0187, 0.0206], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 13:32:32,342 INFO [train.py:903] (0/4) Epoch 29, batch 1550, loss[loss=0.2319, simple_loss=0.3067, pruned_loss=0.07857, over 18875.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2826, pruned_loss=0.05964, over 3835116.02 frames. ], batch size: 74, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:32:56,574 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192753.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:32:57,533 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=192754.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:33:05,619 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=192761.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:33:31,827 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.013e+02 5.130e+02 6.287e+02 7.695e+02 1.691e+03, threshold=1.257e+03, percent-clipped=3.0 +2023-04-03 13:33:34,856 INFO [train.py:903] (0/4) Epoch 29, batch 1600, loss[loss=0.2026, simple_loss=0.2677, pruned_loss=0.06874, over 15933.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2825, pruned_loss=0.05935, over 3824285.98 frames. ], batch size: 35, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:33:47,553 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=192795.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:33:58,469 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-03 13:34:34,143 INFO [train.py:903] (0/4) Epoch 29, batch 1650, loss[loss=0.1933, simple_loss=0.2898, pruned_loss=0.04842, over 19540.00 frames. ], tot_loss[loss=0.201, simple_loss=0.283, pruned_loss=0.05955, over 3836313.72 frames. ], batch size: 56, lr: 2.84e-03, grad_scale: 8.0 +2023-04-03 13:34:35,736 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=192835.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:34:51,830 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=192849.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 13:35:03,992 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=192860.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:35:15,832 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=192869.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:35:23,804 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=192876.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:35:30,385 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.635e+02 4.944e+02 5.872e+02 7.824e+02 1.796e+03, threshold=1.174e+03, percent-clipped=2.0 +2023-04-03 13:35:32,723 INFO [train.py:903] (0/4) Epoch 29, batch 1700, loss[loss=0.2444, simple_loss=0.3151, pruned_loss=0.08686, over 19618.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2828, pruned_loss=0.05984, over 3838858.79 frames. ], batch size: 57, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:35:38,434 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=192888.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:35:39,569 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1918, 2.0855, 1.9197, 1.8207, 1.6301, 1.7570, 0.6955, 1.1784], + device='cuda:0'), covar=tensor([0.0722, 0.0675, 0.0546, 0.0953, 0.1172, 0.1066, 0.1472, 0.1223], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0364, 0.0368, 0.0393, 0.0472, 0.0398, 0.0347, 0.0348], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 13:36:04,834 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=192910.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:36:08,224 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=192913.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:36:13,393 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-03 13:36:24,706 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.0616, 5.1519, 5.8641, 5.9089, 2.2230, 5.5567, 4.6880, 5.5388], + device='cuda:0'), covar=tensor([0.1727, 0.0812, 0.0545, 0.0599, 0.6004, 0.0868, 0.0613, 0.1070], + device='cuda:0'), in_proj_covar=tensor([0.0824, 0.0788, 0.0997, 0.0877, 0.0869, 0.0765, 0.0588, 0.0930], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 13:36:32,300 INFO [train.py:903] (0/4) Epoch 29, batch 1750, loss[loss=0.1881, simple_loss=0.2792, pruned_loss=0.04853, over 17300.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2827, pruned_loss=0.05969, over 3839344.51 frames. ], batch size: 101, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:36:55,768 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=192952.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:36:57,976 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=192954.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:37:24,841 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=192977.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:37:25,776 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=192978.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:37:27,191 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=192979.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:37:31,714 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.864e+02 4.741e+02 6.117e+02 7.797e+02 1.758e+03, threshold=1.223e+03, percent-clipped=7.0 +2023-04-03 13:37:34,999 INFO [train.py:903] (0/4) Epoch 29, batch 1800, loss[loss=0.182, simple_loss=0.2741, pruned_loss=0.04494, over 18855.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.283, pruned_loss=0.06012, over 3808841.26 frames. ], batch size: 74, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:37:58,077 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.4971, 2.5002, 2.0991, 2.4707, 2.0744, 1.9550, 1.9038, 2.2893], + device='cuda:0'), covar=tensor([0.1131, 0.1716, 0.1757, 0.1297, 0.1821, 0.0775, 0.1881, 0.0879], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0362, 0.0321, 0.0259, 0.0309, 0.0258, 0.0324, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 13:38:08,538 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-03 13:38:29,367 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-03 13:38:33,986 INFO [train.py:903] (0/4) Epoch 29, batch 1850, loss[loss=0.1549, simple_loss=0.24, pruned_loss=0.03495, over 19768.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2825, pruned_loss=0.06002, over 3800156.06 frames. ], batch size: 47, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:38:34,336 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.9458, 1.4524, 1.7960, 1.5458, 4.4580, 1.0674, 2.5754, 4.8422], + device='cuda:0'), covar=tensor([0.0478, 0.2973, 0.2900, 0.2103, 0.0732, 0.2874, 0.1616, 0.0172], + device='cuda:0'), in_proj_covar=tensor([0.0426, 0.0380, 0.0401, 0.0355, 0.0384, 0.0361, 0.0400, 0.0421], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 13:38:35,288 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=193035.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:38:35,537 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7113, 1.3690, 1.3550, 1.5797, 1.2240, 1.4299, 1.2490, 1.5288], + device='cuda:0'), covar=tensor([0.1081, 0.1008, 0.1563, 0.1035, 0.1248, 0.0595, 0.1624, 0.0825], + device='cuda:0'), in_proj_covar=tensor([0.0278, 0.0360, 0.0320, 0.0258, 0.0307, 0.0257, 0.0323, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 13:39:04,665 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=193059.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:39:05,537 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-03 13:39:32,452 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.498e+02 4.794e+02 5.817e+02 7.567e+02 2.266e+03, threshold=1.163e+03, percent-clipped=4.0 +2023-04-03 13:39:34,755 INFO [train.py:903] (0/4) Epoch 29, batch 1900, loss[loss=0.2247, simple_loss=0.3065, pruned_loss=0.07146, over 18130.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2818, pruned_loss=0.0595, over 3795548.81 frames. ], batch size: 83, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:39:46,009 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=193093.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:39:49,035 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-03 13:39:50,320 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=193097.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:39:55,322 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-03 13:40:18,666 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-03 13:40:25,416 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=193125.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:40:33,523 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=193132.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:40:35,430 INFO [train.py:903] (0/4) Epoch 29, batch 1950, loss[loss=0.268, simple_loss=0.3254, pruned_loss=0.1052, over 13579.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2827, pruned_loss=0.06046, over 3769738.81 frames. ], batch size: 136, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:40:56,480 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=193150.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:40:56,520 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=193150.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:41:04,215 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=193157.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:41:15,390 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=193166.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:41:34,349 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.41 vs. limit=2.0 +2023-04-03 13:41:34,716 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.104e+02 4.788e+02 5.811e+02 7.006e+02 1.429e+03, threshold=1.162e+03, percent-clipped=3.0 +2023-04-03 13:41:37,847 INFO [train.py:903] (0/4) Epoch 29, batch 2000, loss[loss=0.2043, simple_loss=0.2888, pruned_loss=0.05986, over 19610.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2822, pruned_loss=0.05972, over 3782870.30 frames. ], batch size: 61, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:41:46,012 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=193191.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:41:48,017 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=193193.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 13:42:10,515 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=193212.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:42:23,961 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.62 vs. limit=2.0 +2023-04-03 13:42:31,353 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-03 13:42:31,583 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=193229.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:42:37,085 INFO [train.py:903] (0/4) Epoch 29, batch 2050, loss[loss=0.2023, simple_loss=0.2846, pruned_loss=0.05997, over 18189.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2839, pruned_loss=0.06077, over 3783327.60 frames. ], batch size: 83, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:42:50,738 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-03 13:42:50,767 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-03 13:43:13,504 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-03 13:43:34,585 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.211e+02 4.912e+02 6.403e+02 8.614e+02 2.001e+03, threshold=1.281e+03, percent-clipped=8.0 +2023-04-03 13:43:36,908 INFO [train.py:903] (0/4) Epoch 29, batch 2100, loss[loss=0.2161, simple_loss=0.304, pruned_loss=0.06406, over 19318.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.2833, pruned_loss=0.06089, over 3796853.31 frames. ], batch size: 66, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:43:59,250 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=193302.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:44:07,572 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-03 13:44:07,888 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=193308.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 13:44:25,613 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-03 13:44:26,804 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.9708, 2.7958, 2.3243, 2.3636, 1.9618, 2.4795, 1.1124, 2.0263], + device='cuda:0'), covar=tensor([0.0689, 0.0663, 0.0764, 0.1116, 0.1288, 0.1139, 0.1533, 0.1139], + device='cuda:0'), in_proj_covar=tensor([0.0370, 0.0366, 0.0370, 0.0393, 0.0475, 0.0399, 0.0348, 0.0350], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 13:44:32,084 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1740, 1.7774, 1.9240, 2.7934, 1.9522, 2.1802, 2.3309, 2.0772], + device='cuda:0'), covar=tensor([0.0793, 0.0937, 0.1014, 0.0741, 0.0855, 0.0802, 0.0869, 0.0679], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0223, 0.0229, 0.0240, 0.0226, 0.0214, 0.0188, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 13:44:37,257 INFO [train.py:903] (0/4) Epoch 29, batch 2150, loss[loss=0.1973, simple_loss=0.2878, pruned_loss=0.05344, over 19602.00 frames. ], tot_loss[loss=0.2027, simple_loss=0.2838, pruned_loss=0.06087, over 3792743.46 frames. ], batch size: 57, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:44:56,837 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=193349.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:45:25,810 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=193374.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:45:35,170 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.248e+02 5.242e+02 6.391e+02 8.913e+02 2.261e+03, threshold=1.278e+03, percent-clipped=9.0 +2023-04-03 13:45:37,408 INFO [train.py:903] (0/4) Epoch 29, batch 2200, loss[loss=0.2048, simple_loss=0.2904, pruned_loss=0.05958, over 19621.00 frames. ], tot_loss[loss=0.2038, simple_loss=0.2846, pruned_loss=0.06151, over 3797976.25 frames. ], batch size: 61, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:45:58,782 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5845, 1.6758, 1.9087, 1.8625, 1.4857, 1.8462, 1.9126, 1.7647], + device='cuda:0'), covar=tensor([0.3806, 0.3445, 0.1854, 0.2192, 0.3594, 0.2067, 0.4678, 0.3247], + device='cuda:0'), in_proj_covar=tensor([0.0946, 0.1026, 0.0749, 0.0959, 0.0925, 0.0862, 0.0867, 0.0814], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 13:46:00,710 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=193403.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:46:04,393 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=193406.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:46:36,021 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=193431.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:46:39,101 INFO [train.py:903] (0/4) Epoch 29, batch 2250, loss[loss=0.1999, simple_loss=0.2848, pruned_loss=0.05748, over 19668.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2845, pruned_loss=0.06122, over 3794545.90 frames. ], batch size: 55, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:46:40,757 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2842, 2.2258, 2.0799, 2.0173, 1.7769, 1.9756, 0.8947, 1.3454], + device='cuda:0'), covar=tensor([0.0620, 0.0651, 0.0524, 0.0855, 0.1181, 0.0978, 0.1357, 0.1087], + device='cuda:0'), in_proj_covar=tensor([0.0370, 0.0366, 0.0370, 0.0394, 0.0474, 0.0400, 0.0348, 0.0351], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 13:46:45,266 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=193439.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:47:07,483 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7585, 1.6333, 1.6631, 2.1758, 1.5768, 1.8309, 1.8820, 1.8280], + device='cuda:0'), covar=tensor([0.0790, 0.0857, 0.0973, 0.0720, 0.0886, 0.0806, 0.0954, 0.0686], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0222, 0.0229, 0.0239, 0.0226, 0.0214, 0.0188, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 13:47:20,664 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=193468.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:47:25,802 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=193472.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 13:47:36,426 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.451e+02 4.884e+02 5.894e+02 7.303e+02 1.661e+03, threshold=1.179e+03, percent-clipped=4.0 +2023-04-03 13:47:38,735 INFO [train.py:903] (0/4) Epoch 29, batch 2300, loss[loss=0.1779, simple_loss=0.2715, pruned_loss=0.04215, over 19665.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2837, pruned_loss=0.06033, over 3802525.81 frames. ], batch size: 53, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:47:50,440 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=193493.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:47:53,665 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-03 13:48:20,984 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=193518.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:48:39,108 INFO [train.py:903] (0/4) Epoch 29, batch 2350, loss[loss=0.2193, simple_loss=0.2915, pruned_loss=0.07352, over 19722.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2842, pruned_loss=0.06072, over 3805480.97 frames. ], batch size: 51, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:49:16,352 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=193564.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 13:49:20,464 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-03 13:49:27,312 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=193573.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:49:36,485 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-03 13:49:37,594 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.068e+02 4.546e+02 5.901e+02 7.828e+02 1.716e+03, threshold=1.180e+03, percent-clipped=11.0 +2023-04-03 13:49:40,761 INFO [train.py:903] (0/4) Epoch 29, batch 2400, loss[loss=0.2294, simple_loss=0.3167, pruned_loss=0.07101, over 18874.00 frames. ], tot_loss[loss=0.2025, simple_loss=0.284, pruned_loss=0.06052, over 3810673.03 frames. ], batch size: 74, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:49:47,435 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=193589.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 13:50:39,568 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=193633.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:50:41,249 INFO [train.py:903] (0/4) Epoch 29, batch 2450, loss[loss=0.2207, simple_loss=0.3124, pruned_loss=0.06449, over 17335.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2849, pruned_loss=0.06107, over 3808428.09 frames. ], batch size: 101, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:50:54,653 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=193646.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:51:39,133 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.050e+02 4.882e+02 5.957e+02 8.043e+02 1.393e+03, threshold=1.191e+03, percent-clipped=5.0 +2023-04-03 13:51:41,314 INFO [train.py:903] (0/4) Epoch 29, batch 2500, loss[loss=0.2232, simple_loss=0.3001, pruned_loss=0.07312, over 19664.00 frames. ], tot_loss[loss=0.203, simple_loss=0.2846, pruned_loss=0.06069, over 3816595.88 frames. ], batch size: 58, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:51:45,865 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=193688.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:51:47,045 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1586, 1.7955, 1.4435, 1.1713, 1.6100, 1.1567, 1.1613, 1.7028], + device='cuda:0'), covar=tensor([0.0867, 0.0843, 0.1015, 0.0888, 0.0600, 0.1306, 0.0666, 0.0450], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0319, 0.0341, 0.0273, 0.0253, 0.0346, 0.0293, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 13:51:54,119 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.76 vs. limit=2.0 +2023-04-03 13:52:17,908 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-03 13:52:40,715 INFO [train.py:903] (0/4) Epoch 29, batch 2550, loss[loss=0.2278, simple_loss=0.305, pruned_loss=0.07523, over 19772.00 frames. ], tot_loss[loss=0.2036, simple_loss=0.2848, pruned_loss=0.06125, over 3818976.08 frames. ], batch size: 56, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:53:06,576 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 13:53:15,302 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=193761.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:53:30,067 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=193774.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:53:35,072 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-03 13:53:38,229 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.99 vs. limit=5.0 +2023-04-03 13:53:41,002 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.098e+02 5.186e+02 6.141e+02 8.140e+02 1.584e+03, threshold=1.228e+03, percent-clipped=10.0 +2023-04-03 13:53:41,159 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=193783.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:53:42,121 INFO [train.py:903] (0/4) Epoch 29, batch 2600, loss[loss=0.1962, simple_loss=0.2882, pruned_loss=0.0521, over 19692.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2836, pruned_loss=0.06053, over 3821279.42 frames. ], batch size: 59, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:54:02,311 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=193799.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:54:22,055 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=193816.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 13:54:44,182 INFO [train.py:903] (0/4) Epoch 29, batch 2650, loss[loss=0.2436, simple_loss=0.3145, pruned_loss=0.08638, over 19426.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2827, pruned_loss=0.0598, over 3832515.92 frames. ], batch size: 70, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:54:44,584 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0920, 3.2884, 1.8752, 1.9792, 2.9284, 1.7258, 1.4518, 2.2358], + device='cuda:0'), covar=tensor([0.1363, 0.0801, 0.1238, 0.0936, 0.0641, 0.1378, 0.1087, 0.0746], + device='cuda:0'), in_proj_covar=tensor([0.0300, 0.0317, 0.0339, 0.0271, 0.0251, 0.0344, 0.0290, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 13:54:47,932 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.6345, 1.4205, 1.5825, 1.5715, 3.2207, 1.1999, 2.4260, 3.6892], + device='cuda:0'), covar=tensor([0.0462, 0.2777, 0.2833, 0.1847, 0.0629, 0.2559, 0.1372, 0.0219], + device='cuda:0'), in_proj_covar=tensor([0.0423, 0.0378, 0.0398, 0.0353, 0.0381, 0.0358, 0.0395, 0.0417], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 13:54:53,439 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.8013, 3.2990, 3.3394, 3.3604, 1.4223, 3.2214, 2.7456, 3.1196], + device='cuda:0'), covar=tensor([0.1877, 0.1122, 0.0838, 0.0987, 0.5651, 0.1176, 0.0961, 0.1301], + device='cuda:0'), in_proj_covar=tensor([0.0825, 0.0792, 0.1002, 0.0883, 0.0872, 0.0766, 0.0592, 0.0929], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 13:54:59,843 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-03 13:55:18,879 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.40 vs. limit=2.0 +2023-04-03 13:55:43,540 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.739e+02 4.600e+02 5.538e+02 7.187e+02 1.315e+03, threshold=1.108e+03, percent-clipped=1.0 +2023-04-03 13:55:44,728 INFO [train.py:903] (0/4) Epoch 29, batch 2700, loss[loss=0.2314, simple_loss=0.3116, pruned_loss=0.07561, over 19746.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2825, pruned_loss=0.05962, over 3845450.50 frames. ], batch size: 63, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:55:48,445 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6796, 1.3011, 1.3101, 1.5578, 1.1309, 1.4107, 1.2980, 1.5040], + device='cuda:0'), covar=tensor([0.1220, 0.1177, 0.1798, 0.1087, 0.1395, 0.0715, 0.1813, 0.0930], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0361, 0.0321, 0.0258, 0.0308, 0.0258, 0.0323, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 13:56:00,565 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=193898.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:56:05,901 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=193902.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:56:42,182 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=193931.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 13:56:45,202 INFO [train.py:903] (0/4) Epoch 29, batch 2750, loss[loss=0.2419, simple_loss=0.3132, pruned_loss=0.08525, over 13362.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2824, pruned_loss=0.05937, over 3840387.45 frames. ], batch size: 135, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:56:58,030 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=193944.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:57:27,657 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.3104, 5.8276, 3.2607, 5.0400, 1.4031, 5.9220, 5.7696, 5.9120], + device='cuda:0'), covar=tensor([0.0338, 0.0725, 0.1716, 0.0690, 0.3689, 0.0465, 0.0757, 0.0754], + device='cuda:0'), in_proj_covar=tensor([0.0535, 0.0432, 0.0523, 0.0359, 0.0414, 0.0460, 0.0455, 0.0487], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 13:57:27,853 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=193969.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:57:36,663 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=193977.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:57:44,120 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.356e+02 4.680e+02 5.858e+02 7.532e+02 1.982e+03, threshold=1.172e+03, percent-clipped=4.0 +2023-04-03 13:57:45,310 INFO [train.py:903] (0/4) Epoch 29, batch 2800, loss[loss=0.2014, simple_loss=0.2873, pruned_loss=0.0577, over 19534.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2833, pruned_loss=0.06015, over 3820832.44 frames. ], batch size: 64, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:58:06,194 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-194000.pt +2023-04-03 13:58:26,734 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=194017.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:58:48,029 INFO [train.py:903] (0/4) Epoch 29, batch 2850, loss[loss=0.1905, simple_loss=0.2606, pruned_loss=0.06018, over 18638.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2833, pruned_loss=0.05983, over 3818186.27 frames. ], batch size: 41, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:58:58,385 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=194042.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:59:01,780 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1347, 1.7399, 1.4367, 1.1315, 1.6854, 1.1356, 1.1157, 1.7268], + device='cuda:0'), covar=tensor([0.0889, 0.0829, 0.1069, 0.1002, 0.0553, 0.1353, 0.0742, 0.0425], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0319, 0.0342, 0.0273, 0.0253, 0.0345, 0.0292, 0.0276], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 13:59:45,132 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-03 13:59:47,408 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.156e+02 4.652e+02 5.661e+02 7.187e+02 2.087e+03, threshold=1.132e+03, percent-clipped=4.0 +2023-04-03 13:59:48,552 INFO [train.py:903] (0/4) Epoch 29, batch 2900, loss[loss=0.1942, simple_loss=0.2656, pruned_loss=0.06134, over 19794.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2833, pruned_loss=0.06006, over 3822154.98 frames. ], batch size: 48, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 13:59:55,297 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=194090.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 13:59:57,600 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=194092.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:00:47,116 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1976, 1.8721, 1.8530, 2.0284, 1.7290, 1.8461, 1.7231, 2.0748], + device='cuda:0'), covar=tensor([0.1019, 0.1483, 0.1479, 0.1140, 0.1535, 0.0596, 0.1506, 0.0706], + device='cuda:0'), in_proj_covar=tensor([0.0280, 0.0362, 0.0322, 0.0260, 0.0310, 0.0260, 0.0325, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 14:00:47,882 INFO [train.py:903] (0/4) Epoch 29, batch 2950, loss[loss=0.1656, simple_loss=0.2439, pruned_loss=0.04367, over 19736.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2834, pruned_loss=0.06043, over 3826204.39 frames. ], batch size: 45, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 14:01:13,341 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=194154.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:01:31,697 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.65 vs. limit=2.0 +2023-04-03 14:01:42,908 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=194179.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:01:46,909 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.793e+02 4.745e+02 5.839e+02 7.587e+02 1.918e+03, threshold=1.168e+03, percent-clipped=4.0 +2023-04-03 14:01:48,092 INFO [train.py:903] (0/4) Epoch 29, batch 3000, loss[loss=0.2074, simple_loss=0.2898, pruned_loss=0.06251, over 19580.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2832, pruned_loss=0.06061, over 3813477.33 frames. ], batch size: 61, lr: 2.83e-03, grad_scale: 8.0 +2023-04-03 14:01:48,093 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 14:02:01,080 INFO [train.py:937] (0/4) Epoch 29, validation: loss=0.1668, simple_loss=0.2661, pruned_loss=0.03375, over 944034.00 frames. +2023-04-03 14:02:01,081 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 14:02:02,331 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-03 14:02:05,028 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=194187.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 14:02:36,837 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=194212.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 14:03:01,745 INFO [train.py:903] (0/4) Epoch 29, batch 3050, loss[loss=0.1993, simple_loss=0.2868, pruned_loss=0.05587, over 19700.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2827, pruned_loss=0.06009, over 3827865.89 frames. ], batch size: 60, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:03:17,754 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=194246.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:04:02,884 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.301e+02 5.546e+02 6.449e+02 7.988e+02 2.570e+03, threshold=1.290e+03, percent-clipped=8.0 +2023-04-03 14:04:03,995 INFO [train.py:903] (0/4) Epoch 29, batch 3100, loss[loss=0.1915, simple_loss=0.278, pruned_loss=0.05249, over 19675.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2816, pruned_loss=0.05956, over 3825786.52 frames. ], batch size: 53, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:05:04,139 INFO [train.py:903] (0/4) Epoch 29, batch 3150, loss[loss=0.1705, simple_loss=0.251, pruned_loss=0.04497, over 19770.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2805, pruned_loss=0.05868, over 3833099.29 frames. ], batch size: 47, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:05:06,647 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=194336.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:05:19,913 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=194348.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:05:27,475 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-03 14:05:36,256 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=194361.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:05:49,252 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=194371.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:05:51,710 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=194373.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:05:59,641 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=194380.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 14:06:02,721 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.204e+02 4.902e+02 5.790e+02 7.151e+02 3.080e+03, threshold=1.158e+03, percent-clipped=4.0 +2023-04-03 14:06:03,883 INFO [train.py:903] (0/4) Epoch 29, batch 3200, loss[loss=0.2127, simple_loss=0.2943, pruned_loss=0.06555, over 19641.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2814, pruned_loss=0.05915, over 3825364.18 frames. ], batch size: 58, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:06:28,873 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.23 vs. limit=5.0 +2023-04-03 14:07:04,909 INFO [train.py:903] (0/4) Epoch 29, batch 3250, loss[loss=0.2059, simple_loss=0.2827, pruned_loss=0.06455, over 19695.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2817, pruned_loss=0.05923, over 3823795.68 frames. ], batch size: 53, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:07:05,051 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=194434.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:08:04,291 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.869e+02 4.916e+02 6.594e+02 9.076e+02 2.390e+03, threshold=1.319e+03, percent-clipped=9.0 +2023-04-03 14:08:05,470 INFO [train.py:903] (0/4) Epoch 29, batch 3300, loss[loss=0.2005, simple_loss=0.287, pruned_loss=0.05703, over 19651.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.282, pruned_loss=0.05949, over 3837752.29 frames. ], batch size: 53, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:08:09,432 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-03 14:09:07,375 INFO [train.py:903] (0/4) Epoch 29, batch 3350, loss[loss=0.2537, simple_loss=0.3233, pruned_loss=0.09209, over 19772.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2829, pruned_loss=0.05977, over 3844697.02 frames. ], batch size: 56, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:09:24,567 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=194549.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:10:06,126 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.313e+02 5.166e+02 6.518e+02 8.363e+02 1.379e+03, threshold=1.304e+03, percent-clipped=1.0 +2023-04-03 14:10:07,290 INFO [train.py:903] (0/4) Epoch 29, batch 3400, loss[loss=0.2062, simple_loss=0.2861, pruned_loss=0.06309, over 18302.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2823, pruned_loss=0.05934, over 3846091.64 frames. ], batch size: 83, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:10:48,227 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=194617.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:11:03,208 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6150, 1.7544, 2.1622, 1.9446, 3.3400, 2.6388, 3.5547, 1.9205], + device='cuda:0'), covar=tensor([0.2769, 0.4624, 0.3009, 0.2053, 0.1458, 0.2388, 0.1624, 0.4387], + device='cuda:0'), in_proj_covar=tensor([0.0559, 0.0676, 0.0765, 0.0513, 0.0638, 0.0550, 0.0674, 0.0578], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 14:11:07,042 INFO [train.py:903] (0/4) Epoch 29, batch 3450, loss[loss=0.2121, simple_loss=0.2961, pruned_loss=0.06404, over 18776.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2822, pruned_loss=0.05915, over 3841732.51 frames. ], batch size: 74, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:11:09,266 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-03 14:11:17,038 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=194642.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:11:29,980 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=194652.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:12:01,640 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=194680.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:12:05,585 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.890e+02 4.627e+02 5.712e+02 7.363e+02 1.612e+03, threshold=1.142e+03, percent-clipped=3.0 +2023-04-03 14:12:06,660 INFO [train.py:903] (0/4) Epoch 29, batch 3500, loss[loss=0.2897, simple_loss=0.3446, pruned_loss=0.1174, over 13541.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2818, pruned_loss=0.05943, over 3821317.60 frames. ], batch size: 135, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:12:43,629 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=194715.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:12:54,391 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=194724.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 14:13:06,876 INFO [train.py:903] (0/4) Epoch 29, batch 3550, loss[loss=0.1977, simple_loss=0.2788, pruned_loss=0.05832, over 19530.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.282, pruned_loss=0.05983, over 3812947.40 frames. ], batch size: 54, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:13:49,785 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.48 vs. limit=5.0 +2023-04-03 14:14:05,432 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.1600, 5.5754, 3.2094, 4.8577, 1.2235, 5.8074, 5.5185, 5.7743], + device='cuda:0'), covar=tensor([0.0407, 0.0811, 0.1779, 0.0738, 0.4054, 0.0489, 0.0782, 0.1050], + device='cuda:0'), in_proj_covar=tensor([0.0529, 0.0426, 0.0516, 0.0355, 0.0409, 0.0456, 0.0451, 0.0482], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 14:14:06,360 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.972e+02 4.806e+02 5.826e+02 6.989e+02 1.690e+03, threshold=1.165e+03, percent-clipped=2.0 +2023-04-03 14:14:07,578 INFO [train.py:903] (0/4) Epoch 29, batch 3600, loss[loss=0.2066, simple_loss=0.2897, pruned_loss=0.06178, over 19509.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2818, pruned_loss=0.05952, over 3807217.81 frames. ], batch size: 64, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:14:20,139 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=194795.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:14:33,736 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=194805.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:15:02,770 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=194830.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:15:02,812 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=194830.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:15:06,261 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 14:15:06,688 INFO [train.py:903] (0/4) Epoch 29, batch 3650, loss[loss=0.2002, simple_loss=0.2812, pruned_loss=0.05959, over 19839.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2818, pruned_loss=0.0595, over 3805302.24 frames. ], batch size: 52, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:15:12,772 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=194839.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 14:15:45,388 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=194865.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:16:05,568 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.663e+02 5.135e+02 6.261e+02 7.520e+02 2.080e+03, threshold=1.252e+03, percent-clipped=4.0 +2023-04-03 14:16:06,488 INFO [train.py:903] (0/4) Epoch 29, batch 3700, loss[loss=0.2762, simple_loss=0.3458, pruned_loss=0.1033, over 19613.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2824, pruned_loss=0.05992, over 3814202.66 frames. ], batch size: 61, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:17:08,166 INFO [train.py:903] (0/4) Epoch 29, batch 3750, loss[loss=0.1864, simple_loss=0.2748, pruned_loss=0.04899, over 19791.00 frames. ], tot_loss[loss=0.2018, simple_loss=0.2829, pruned_loss=0.06033, over 3807146.63 frames. ], batch size: 56, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:18:06,323 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.993e+02 4.741e+02 6.225e+02 7.602e+02 2.236e+03, threshold=1.245e+03, percent-clipped=10.0 +2023-04-03 14:18:07,443 INFO [train.py:903] (0/4) Epoch 29, batch 3800, loss[loss=0.1899, simple_loss=0.2854, pruned_loss=0.04726, over 19533.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2824, pruned_loss=0.05978, over 3817436.07 frames. ], batch size: 54, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:18:21,201 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=194996.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:18:34,661 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-03 14:19:01,235 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=195029.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:19:06,531 INFO [train.py:903] (0/4) Epoch 29, batch 3850, loss[loss=0.207, simple_loss=0.2882, pruned_loss=0.0629, over 19623.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2819, pruned_loss=0.0597, over 3814435.60 frames. ], batch size: 57, lr: 2.82e-03, grad_scale: 4.0 +2023-04-03 14:19:27,234 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=195051.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:19:41,487 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=195062.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 14:19:44,000 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2492, 2.2646, 2.5398, 2.9599, 2.3480, 2.8359, 2.4691, 2.2621], + device='cuda:0'), covar=tensor([0.4470, 0.4409, 0.2036, 0.2704, 0.4655, 0.2365, 0.5119, 0.3660], + device='cuda:0'), in_proj_covar=tensor([0.0951, 0.1030, 0.0754, 0.0964, 0.0928, 0.0866, 0.0869, 0.0818], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 14:19:56,985 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=195076.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:20:06,403 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.076e+02 4.743e+02 5.822e+02 7.676e+02 1.767e+03, threshold=1.164e+03, percent-clipped=5.0 +2023-04-03 14:20:06,426 INFO [train.py:903] (0/4) Epoch 29, batch 3900, loss[loss=0.1754, simple_loss=0.2461, pruned_loss=0.05235, over 19303.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.281, pruned_loss=0.05922, over 3815356.62 frames. ], batch size: 44, lr: 2.82e-03, grad_scale: 4.0 +2023-04-03 14:20:08,963 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=195086.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:20:13,187 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=195089.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:20:20,910 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=195095.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 14:20:23,858 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5753, 4.1153, 4.2805, 4.2703, 1.6926, 4.0453, 3.4744, 4.0515], + device='cuda:0'), covar=tensor([0.1729, 0.0840, 0.0643, 0.0751, 0.6075, 0.0926, 0.0811, 0.1141], + device='cuda:0'), in_proj_covar=tensor([0.0826, 0.0794, 0.1004, 0.0880, 0.0871, 0.0769, 0.0594, 0.0935], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 14:20:39,809 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=195111.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:20:39,848 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=195111.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:20:49,224 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.86 vs. limit=5.0 +2023-04-03 14:20:50,174 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=195120.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 14:21:07,592 INFO [train.py:903] (0/4) Epoch 29, batch 3950, loss[loss=0.197, simple_loss=0.2841, pruned_loss=0.05497, over 19673.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2814, pruned_loss=0.05915, over 3810095.65 frames. ], batch size: 55, lr: 2.82e-03, grad_scale: 4.0 +2023-04-03 14:21:12,680 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-03 14:21:43,413 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4184, 1.3043, 1.9435, 1.5719, 2.9936, 4.4943, 4.4345, 4.9330], + device='cuda:0'), covar=tensor([0.1631, 0.4341, 0.3640, 0.2596, 0.0666, 0.0220, 0.0178, 0.0201], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0334, 0.0367, 0.0273, 0.0258, 0.0198, 0.0221, 0.0278], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 14:22:08,284 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.119e+02 4.632e+02 5.828e+02 7.591e+02 1.503e+03, threshold=1.166e+03, percent-clipped=4.0 +2023-04-03 14:22:08,302 INFO [train.py:903] (0/4) Epoch 29, batch 4000, loss[loss=0.1579, simple_loss=0.2394, pruned_loss=0.0382, over 19306.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2814, pruned_loss=0.05922, over 3815850.50 frames. ], batch size: 44, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:22:38,710 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=195209.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:22:52,712 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-03 14:23:07,865 INFO [train.py:903] (0/4) Epoch 29, batch 4050, loss[loss=0.2083, simple_loss=0.2796, pruned_loss=0.06846, over 19381.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2828, pruned_loss=0.06032, over 3810821.54 frames. ], batch size: 48, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:23:19,843 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 14:23:24,432 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.26 vs. limit=2.0 +2023-04-03 14:24:08,388 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.197e+02 5.118e+02 6.147e+02 8.377e+02 1.783e+03, threshold=1.229e+03, percent-clipped=5.0 +2023-04-03 14:24:08,406 INFO [train.py:903] (0/4) Epoch 29, batch 4100, loss[loss=0.2106, simple_loss=0.2926, pruned_loss=0.06432, over 19656.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2823, pruned_loss=0.06013, over 3823297.25 frames. ], batch size: 58, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:24:41,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-03 14:24:56,207 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=195324.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:25:08,841 INFO [train.py:903] (0/4) Epoch 29, batch 4150, loss[loss=0.2101, simple_loss=0.2893, pruned_loss=0.06545, over 17507.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2828, pruned_loss=0.06029, over 3815516.87 frames. ], batch size: 101, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:25:49,120 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=195367.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:25:56,296 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=195373.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:26:10,052 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.652e+02 4.925e+02 6.547e+02 8.865e+02 3.200e+03, threshold=1.309e+03, percent-clipped=9.0 +2023-04-03 14:26:10,071 INFO [train.py:903] (0/4) Epoch 29, batch 4200, loss[loss=0.1975, simple_loss=0.285, pruned_loss=0.05499, over 19305.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2812, pruned_loss=0.05969, over 3818557.27 frames. ], batch size: 70, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:26:13,438 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-03 14:26:19,408 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=195392.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:26:35,643 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=195406.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 14:26:40,089 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=195410.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:27:07,654 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=195433.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:27:08,632 INFO [train.py:903] (0/4) Epoch 29, batch 4250, loss[loss=0.2146, simple_loss=0.297, pruned_loss=0.06613, over 19653.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2819, pruned_loss=0.06043, over 3818709.23 frames. ], batch size: 58, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:27:22,567 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-03 14:27:32,635 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-03 14:28:08,329 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.783e+02 4.660e+02 5.852e+02 7.750e+02 1.648e+03, threshold=1.170e+03, percent-clipped=3.0 +2023-04-03 14:28:08,348 INFO [train.py:903] (0/4) Epoch 29, batch 4300, loss[loss=0.178, simple_loss=0.2527, pruned_loss=0.05165, over 19735.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2824, pruned_loss=0.06071, over 3815522.36 frames. ], batch size: 47, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:28:14,094 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=195488.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:28:54,525 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=195521.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 14:29:01,487 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-03 14:29:09,903 INFO [train.py:903] (0/4) Epoch 29, batch 4350, loss[loss=0.194, simple_loss=0.265, pruned_loss=0.0615, over 19776.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2822, pruned_loss=0.06003, over 3820274.82 frames. ], batch size: 47, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:29:28,070 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=195548.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:29:59,932 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.5825, 1.1528, 1.3797, 1.3695, 2.2578, 1.1626, 2.1216, 2.5363], + device='cuda:0'), covar=tensor([0.0686, 0.2967, 0.3041, 0.1591, 0.0839, 0.2006, 0.1122, 0.0451], + device='cuda:0'), in_proj_covar=tensor([0.0427, 0.0383, 0.0403, 0.0356, 0.0386, 0.0361, 0.0400, 0.0421], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 14:30:05,591 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=195580.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:30:10,386 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.562e+02 4.994e+02 6.103e+02 7.630e+02 1.747e+03, threshold=1.221e+03, percent-clipped=1.0 +2023-04-03 14:30:10,405 INFO [train.py:903] (0/4) Epoch 29, batch 4400, loss[loss=0.2071, simple_loss=0.3079, pruned_loss=0.05311, over 19671.00 frames. ], tot_loss[loss=0.2024, simple_loss=0.2836, pruned_loss=0.06063, over 3806977.33 frames. ], batch size: 58, lr: 2.82e-03, grad_scale: 8.0 +2023-04-03 14:30:29,902 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-03 14:30:35,612 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=195605.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:30:38,554 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-03 14:31:09,647 INFO [train.py:903] (0/4) Epoch 29, batch 4450, loss[loss=0.1575, simple_loss=0.2397, pruned_loss=0.03764, over 16094.00 frames. ], tot_loss[loss=0.2021, simple_loss=0.2834, pruned_loss=0.06045, over 3808408.81 frames. ], batch size: 35, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:32:07,397 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.24 vs. limit=2.0 +2023-04-03 14:32:08,763 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.510e+02 5.083e+02 5.988e+02 7.824e+02 1.664e+03, threshold=1.198e+03, percent-clipped=3.0 +2023-04-03 14:32:08,782 INFO [train.py:903] (0/4) Epoch 29, batch 4500, loss[loss=0.2248, simple_loss=0.3013, pruned_loss=0.07413, over 19468.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2818, pruned_loss=0.05939, over 3813633.53 frames. ], batch size: 64, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:33:10,065 INFO [train.py:903] (0/4) Epoch 29, batch 4550, loss[loss=0.2306, simple_loss=0.3191, pruned_loss=0.07107, over 19776.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2819, pruned_loss=0.05971, over 3813140.84 frames. ], batch size: 56, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:33:15,610 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-03 14:33:21,552 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=195744.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:33:33,980 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=195754.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:33:39,345 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-03 14:33:51,816 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=195769.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:33:59,658 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.20 vs. limit=2.0 +2023-04-03 14:34:01,508 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=195777.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 14:34:08,708 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.333e+02 5.047e+02 5.905e+02 8.200e+02 1.793e+03, threshold=1.181e+03, percent-clipped=7.0 +2023-04-03 14:34:08,726 INFO [train.py:903] (0/4) Epoch 29, batch 4600, loss[loss=0.1998, simple_loss=0.2898, pruned_loss=0.05487, over 19603.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2828, pruned_loss=0.06031, over 3797355.97 frames. ], batch size: 52, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:34:31,284 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=195802.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 14:34:33,593 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=195804.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:35:05,153 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=195829.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:35:10,614 INFO [train.py:903] (0/4) Epoch 29, batch 4650, loss[loss=0.1608, simple_loss=0.247, pruned_loss=0.03728, over 19413.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2821, pruned_loss=0.05976, over 3803617.92 frames. ], batch size: 48, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:35:22,172 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-03 14:35:33,354 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-03 14:35:54,034 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=195869.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:36:00,411 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7616, 4.2452, 4.4719, 4.4489, 1.7801, 4.2190, 3.6812, 4.2306], + device='cuda:0'), covar=tensor([0.1708, 0.1017, 0.0617, 0.0752, 0.6092, 0.1030, 0.0711, 0.1098], + device='cuda:0'), in_proj_covar=tensor([0.0829, 0.0799, 0.1009, 0.0884, 0.0877, 0.0775, 0.0595, 0.0941], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 14:36:10,280 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.073e+02 5.115e+02 6.253e+02 8.443e+02 2.371e+03, threshold=1.251e+03, percent-clipped=7.0 +2023-04-03 14:36:10,298 INFO [train.py:903] (0/4) Epoch 29, batch 4700, loss[loss=0.2031, simple_loss=0.2938, pruned_loss=0.0562, over 19534.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2825, pruned_loss=0.05979, over 3810385.20 frames. ], batch size: 61, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:36:29,172 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-03 14:37:02,884 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4502, 1.4969, 1.8312, 1.7312, 2.7404, 2.2784, 2.9509, 1.2962], + device='cuda:0'), covar=tensor([0.2653, 0.4566, 0.2921, 0.2031, 0.1566, 0.2357, 0.1483, 0.4819], + device='cuda:0'), in_proj_covar=tensor([0.0560, 0.0679, 0.0768, 0.0515, 0.0641, 0.0553, 0.0676, 0.0580], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 14:37:11,420 INFO [train.py:903] (0/4) Epoch 29, batch 4750, loss[loss=0.2024, simple_loss=0.2929, pruned_loss=0.05596, over 18895.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2822, pruned_loss=0.05952, over 3804441.87 frames. ], batch size: 74, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:37:25,294 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.30 vs. limit=2.0 +2023-04-03 14:38:03,277 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 14:38:08,658 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4840, 2.2168, 1.7107, 1.6131, 2.0420, 1.4053, 1.3868, 1.9315], + device='cuda:0'), covar=tensor([0.1138, 0.0841, 0.1114, 0.0852, 0.0682, 0.1340, 0.0831, 0.0518], + device='cuda:0'), in_proj_covar=tensor([0.0306, 0.0322, 0.0344, 0.0274, 0.0255, 0.0348, 0.0292, 0.0276], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 14:38:11,673 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.651e+02 4.980e+02 6.350e+02 8.555e+02 2.103e+03, threshold=1.270e+03, percent-clipped=3.0 +2023-04-03 14:38:11,691 INFO [train.py:903] (0/4) Epoch 29, batch 4800, loss[loss=0.1729, simple_loss=0.265, pruned_loss=0.04046, over 19679.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2816, pruned_loss=0.0596, over 3808616.21 frames. ], batch size: 53, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:38:20,092 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.75 vs. limit=5.0 +2023-04-03 14:38:31,634 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-196000.pt +2023-04-03 14:38:56,380 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4806, 1.5278, 1.8184, 1.7318, 2.5341, 2.2429, 2.6812, 1.2103], + device='cuda:0'), covar=tensor([0.2620, 0.4601, 0.2914, 0.2033, 0.1699, 0.2406, 0.1602, 0.4913], + device='cuda:0'), in_proj_covar=tensor([0.0562, 0.0681, 0.0770, 0.0517, 0.0643, 0.0555, 0.0678, 0.0583], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 14:39:14,504 INFO [train.py:903] (0/4) Epoch 29, batch 4850, loss[loss=0.1755, simple_loss=0.2551, pruned_loss=0.04793, over 19585.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2817, pruned_loss=0.05959, over 3822278.33 frames. ], batch size: 52, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:39:37,046 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-03 14:39:55,383 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-03 14:40:01,746 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-03 14:40:02,677 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-03 14:40:11,420 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-03 14:40:13,822 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.943e+02 4.641e+02 5.767e+02 7.663e+02 1.512e+03, threshold=1.153e+03, percent-clipped=2.0 +2023-04-03 14:40:13,840 INFO [train.py:903] (0/4) Epoch 29, batch 4900, loss[loss=0.193, simple_loss=0.2727, pruned_loss=0.05665, over 19773.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.283, pruned_loss=0.05996, over 3812266.85 frames. ], batch size: 54, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:40:22,183 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2051, 1.2665, 1.2901, 1.1058, 1.0689, 1.1245, 0.1529, 0.4056], + device='cuda:0'), covar=tensor([0.0859, 0.0764, 0.0515, 0.0702, 0.1580, 0.0770, 0.1599, 0.1358], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0364, 0.0370, 0.0394, 0.0475, 0.0400, 0.0347, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 14:40:32,814 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-03 14:41:03,531 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=196125.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:41:14,068 INFO [train.py:903] (0/4) Epoch 29, batch 4950, loss[loss=0.156, simple_loss=0.2427, pruned_loss=0.03459, over 19384.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2826, pruned_loss=0.05982, over 3821310.92 frames. ], batch size: 48, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:41:14,818 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.09 vs. limit=2.0 +2023-04-03 14:41:16,764 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9861, 1.4809, 1.6493, 1.5936, 3.6067, 1.0955, 2.6324, 4.0688], + device='cuda:0'), covar=tensor([0.0459, 0.2784, 0.2818, 0.1968, 0.0626, 0.2661, 0.1197, 0.0193], + device='cuda:0'), in_proj_covar=tensor([0.0427, 0.0384, 0.0404, 0.0358, 0.0387, 0.0361, 0.0402, 0.0422], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 14:41:31,282 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-03 14:41:33,812 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=196150.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:41:54,207 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-03 14:41:58,580 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8071, 1.5105, 1.8983, 1.8042, 4.3552, 1.3875, 2.6222, 4.6991], + device='cuda:0'), covar=tensor([0.0498, 0.2961, 0.2768, 0.1962, 0.0732, 0.2650, 0.1604, 0.0173], + device='cuda:0'), in_proj_covar=tensor([0.0428, 0.0385, 0.0404, 0.0358, 0.0387, 0.0362, 0.0403, 0.0423], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 14:42:14,324 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.427e+02 4.864e+02 6.197e+02 7.725e+02 1.739e+03, threshold=1.239e+03, percent-clipped=10.0 +2023-04-03 14:42:14,342 INFO [train.py:903] (0/4) Epoch 29, batch 5000, loss[loss=0.2536, simple_loss=0.3233, pruned_loss=0.09193, over 19524.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2819, pruned_loss=0.05931, over 3822832.98 frames. ], batch size: 64, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:42:14,684 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0929, 1.3167, 1.7589, 1.2660, 2.6472, 3.4814, 3.1800, 3.6700], + device='cuda:0'), covar=tensor([0.1799, 0.4048, 0.3599, 0.2856, 0.0686, 0.0241, 0.0242, 0.0306], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0335, 0.0367, 0.0274, 0.0258, 0.0199, 0.0221, 0.0279], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 14:42:16,062 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.60 vs. limit=5.0 +2023-04-03 14:42:23,520 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-03 14:42:35,057 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-03 14:42:54,571 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.77 vs. limit=5.0 +2023-04-03 14:43:14,846 INFO [train.py:903] (0/4) Epoch 29, batch 5050, loss[loss=0.1857, simple_loss=0.2795, pruned_loss=0.04597, over 19368.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2818, pruned_loss=0.05882, over 3828858.22 frames. ], batch size: 70, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:43:49,704 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-03 14:44:15,609 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.659e+02 4.774e+02 5.782e+02 7.176e+02 1.455e+03, threshold=1.156e+03, percent-clipped=5.0 +2023-04-03 14:44:15,627 INFO [train.py:903] (0/4) Epoch 29, batch 5100, loss[loss=0.1699, simple_loss=0.2556, pruned_loss=0.04215, over 19834.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2806, pruned_loss=0.05829, over 3823761.66 frames. ], batch size: 52, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:44:24,665 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-03 14:44:27,971 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-03 14:44:32,504 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-03 14:45:16,433 INFO [train.py:903] (0/4) Epoch 29, batch 5150, loss[loss=0.1765, simple_loss=0.2542, pruned_loss=0.04939, over 19313.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2814, pruned_loss=0.0589, over 3814804.99 frames. ], batch size: 44, lr: 2.81e-03, grad_scale: 4.0 +2023-04-03 14:45:17,917 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.2447, 2.8851, 2.3323, 2.4071, 2.1851, 2.6590, 1.0168, 2.0709], + device='cuda:0'), covar=tensor([0.0698, 0.0686, 0.0723, 0.1154, 0.1149, 0.1091, 0.1560, 0.1247], + device='cuda:0'), in_proj_covar=tensor([0.0366, 0.0363, 0.0369, 0.0393, 0.0473, 0.0398, 0.0347, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 14:45:23,039 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=196339.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:45:26,214 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-03 14:45:59,982 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 14:46:10,236 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=196378.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:46:15,837 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8047, 1.9138, 2.1473, 2.2586, 1.7386, 2.2048, 2.0983, 1.9364], + device='cuda:0'), covar=tensor([0.4264, 0.3715, 0.2022, 0.2421, 0.3998, 0.2263, 0.5291, 0.3605], + device='cuda:0'), in_proj_covar=tensor([0.0945, 0.1023, 0.0748, 0.0958, 0.0923, 0.0862, 0.0863, 0.0812], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 14:46:17,533 INFO [train.py:903] (0/4) Epoch 29, batch 5200, loss[loss=0.2233, simple_loss=0.3021, pruned_loss=0.07225, over 18213.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.2808, pruned_loss=0.05877, over 3808809.50 frames. ], batch size: 83, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:46:18,467 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.574e+02 5.187e+02 6.337e+02 8.295e+02 1.863e+03, threshold=1.267e+03, percent-clipped=5.0 +2023-04-03 14:46:28,779 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-03 14:47:12,668 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-03 14:47:19,214 INFO [train.py:903] (0/4) Epoch 29, batch 5250, loss[loss=0.2114, simple_loss=0.2905, pruned_loss=0.06614, over 19724.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2818, pruned_loss=0.05922, over 3805682.67 frames. ], batch size: 51, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:47:43,296 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8484, 1.9274, 2.2598, 2.3135, 1.8260, 2.3184, 2.2275, 2.0213], + device='cuda:0'), covar=tensor([0.4234, 0.4055, 0.1999, 0.2532, 0.4027, 0.2263, 0.5182, 0.3715], + device='cuda:0'), in_proj_covar=tensor([0.0949, 0.1026, 0.0751, 0.0961, 0.0926, 0.0864, 0.0866, 0.0815], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 14:48:07,776 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2498, 2.2790, 2.5993, 2.8973, 2.2387, 2.7876, 2.5915, 2.3008], + device='cuda:0'), covar=tensor([0.4389, 0.4249, 0.1979, 0.2794, 0.4496, 0.2386, 0.4961, 0.3543], + device='cuda:0'), in_proj_covar=tensor([0.0948, 0.1025, 0.0750, 0.0960, 0.0925, 0.0864, 0.0865, 0.0814], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 14:48:20,445 INFO [train.py:903] (0/4) Epoch 29, batch 5300, loss[loss=0.2518, simple_loss=0.3274, pruned_loss=0.08811, over 19685.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2823, pruned_loss=0.0592, over 3813510.25 frames. ], batch size: 58, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:48:21,521 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.666e+02 4.792e+02 5.709e+02 7.130e+02 1.478e+03, threshold=1.142e+03, percent-clipped=2.0 +2023-04-03 14:48:35,852 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-03 14:49:20,266 INFO [train.py:903] (0/4) Epoch 29, batch 5350, loss[loss=0.1933, simple_loss=0.2719, pruned_loss=0.05732, over 19858.00 frames. ], tot_loss[loss=0.2, simple_loss=0.2819, pruned_loss=0.05903, over 3826433.68 frames. ], batch size: 52, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:49:51,997 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-03 14:50:09,929 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.8398, 4.3726, 2.8680, 3.8571, 1.0258, 4.3902, 4.2554, 4.3800], + device='cuda:0'), covar=tensor([0.0624, 0.1101, 0.1856, 0.0891, 0.4131, 0.0630, 0.0906, 0.1089], + device='cuda:0'), in_proj_covar=tensor([0.0534, 0.0432, 0.0518, 0.0357, 0.0409, 0.0457, 0.0452, 0.0487], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 14:50:21,626 INFO [train.py:903] (0/4) Epoch 29, batch 5400, loss[loss=0.1926, simple_loss=0.2885, pruned_loss=0.0484, over 19680.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.281, pruned_loss=0.05824, over 3828071.23 frames. ], batch size: 59, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:50:22,760 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.031e+02 4.834e+02 5.946e+02 7.840e+02 1.782e+03, threshold=1.189e+03, percent-clipped=5.0 +2023-04-03 14:50:36,245 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=196596.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:50:59,765 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3342, 2.0779, 1.6614, 1.4354, 1.8380, 1.3725, 1.3350, 1.8182], + device='cuda:0'), covar=tensor([0.0946, 0.0829, 0.1149, 0.0870, 0.0602, 0.1358, 0.0751, 0.0506], + device='cuda:0'), in_proj_covar=tensor([0.0307, 0.0324, 0.0346, 0.0277, 0.0255, 0.0350, 0.0294, 0.0278], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 14:51:22,494 INFO [train.py:903] (0/4) Epoch 29, batch 5450, loss[loss=0.2325, simple_loss=0.3127, pruned_loss=0.07617, over 19736.00 frames. ], tot_loss[loss=0.1978, simple_loss=0.2799, pruned_loss=0.05787, over 3839517.80 frames. ], batch size: 63, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:51:34,881 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0575, 4.4838, 4.8386, 4.8392, 1.8174, 4.5264, 3.9014, 4.5704], + device='cuda:0'), covar=tensor([0.1825, 0.0841, 0.0658, 0.0667, 0.6435, 0.0971, 0.0709, 0.1134], + device='cuda:0'), in_proj_covar=tensor([0.0829, 0.0802, 0.1013, 0.0886, 0.0879, 0.0774, 0.0596, 0.0938], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 14:51:37,092 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7501, 1.7599, 1.7241, 1.5012, 1.5421, 1.4902, 0.3269, 0.7205], + device='cuda:0'), covar=tensor([0.0737, 0.0667, 0.0442, 0.0704, 0.1211, 0.0820, 0.1353, 0.1219], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0365, 0.0370, 0.0394, 0.0474, 0.0400, 0.0348, 0.0350], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 14:51:41,424 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.0370, 5.5254, 2.9177, 4.7708, 1.0405, 5.6733, 5.4532, 5.5610], + device='cuda:0'), covar=tensor([0.0408, 0.0783, 0.1954, 0.0725, 0.3985, 0.0472, 0.0813, 0.1080], + device='cuda:0'), in_proj_covar=tensor([0.0537, 0.0435, 0.0522, 0.0360, 0.0413, 0.0460, 0.0456, 0.0491], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 14:52:21,536 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=196683.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:52:22,558 INFO [train.py:903] (0/4) Epoch 29, batch 5500, loss[loss=0.1871, simple_loss=0.275, pruned_loss=0.04962, over 17437.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2821, pruned_loss=0.05954, over 3825867.01 frames. ], batch size: 101, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:52:23,692 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.565e+02 4.767e+02 5.823e+02 7.066e+02 1.237e+03, threshold=1.165e+03, percent-clipped=1.0 +2023-04-03 14:52:46,212 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-03 14:53:08,296 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=196722.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:53:22,918 INFO [train.py:903] (0/4) Epoch 29, batch 5550, loss[loss=0.216, simple_loss=0.3043, pruned_loss=0.06384, over 18160.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.282, pruned_loss=0.0592, over 3834587.06 frames. ], batch size: 83, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:53:30,049 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-03 14:53:40,091 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.30 vs. limit=5.0 +2023-04-03 14:54:18,749 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-03 14:54:22,913 INFO [train.py:903] (0/4) Epoch 29, batch 5600, loss[loss=0.1978, simple_loss=0.2854, pruned_loss=0.0551, over 19533.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2822, pruned_loss=0.0593, over 3821802.94 frames. ], batch size: 56, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:54:24,076 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.131e+02 4.937e+02 5.977e+02 7.468e+02 1.263e+03, threshold=1.195e+03, percent-clipped=3.0 +2023-04-03 14:54:39,832 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=196797.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:54:40,983 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=196798.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:55:24,428 INFO [train.py:903] (0/4) Epoch 29, batch 5650, loss[loss=0.1699, simple_loss=0.2449, pruned_loss=0.04741, over 19753.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2809, pruned_loss=0.05866, over 3819993.35 frames. ], batch size: 47, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:55:28,124 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=196837.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:56:10,093 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=196872.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:56:10,963 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-03 14:56:24,244 INFO [train.py:903] (0/4) Epoch 29, batch 5700, loss[loss=0.1914, simple_loss=0.2871, pruned_loss=0.04779, over 19666.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2806, pruned_loss=0.05848, over 3814961.56 frames. ], batch size: 55, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:56:25,387 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.192e+02 5.159e+02 6.169e+02 7.777e+02 1.148e+03, threshold=1.234e+03, percent-clipped=0.0 +2023-04-03 14:56:41,796 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=196899.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:56:46,521 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.68 vs. limit=2.0 +2023-04-03 14:57:24,049 INFO [train.py:903] (0/4) Epoch 29, batch 5750, loss[loss=0.2232, simple_loss=0.2987, pruned_loss=0.07383, over 18077.00 frames. ], tot_loss[loss=0.2008, simple_loss=0.2828, pruned_loss=0.05943, over 3819547.49 frames. ], batch size: 83, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:57:25,246 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-03 14:57:30,929 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=196940.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:57:31,936 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-03 14:57:38,028 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-03 14:58:24,228 INFO [train.py:903] (0/4) Epoch 29, batch 5800, loss[loss=0.2297, simple_loss=0.3083, pruned_loss=0.07561, over 19300.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2829, pruned_loss=0.05958, over 3817733.32 frames. ], batch size: 66, lr: 2.81e-03, grad_scale: 8.0 +2023-04-03 14:58:25,407 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.058e+02 4.987e+02 6.141e+02 8.368e+02 1.662e+03, threshold=1.228e+03, percent-clipped=5.0 +2023-04-03 14:59:24,885 INFO [train.py:903] (0/4) Epoch 29, batch 5850, loss[loss=0.1639, simple_loss=0.2429, pruned_loss=0.04245, over 19370.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2822, pruned_loss=0.05937, over 3825724.68 frames. ], batch size: 48, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 14:59:27,415 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 14:59:49,575 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=197054.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 14:59:50,544 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=197055.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:00:19,099 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=197079.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:00:24,408 INFO [train.py:903] (0/4) Epoch 29, batch 5900, loss[loss=0.2275, simple_loss=0.2993, pruned_loss=0.0778, over 13613.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2824, pruned_loss=0.05953, over 3812462.56 frames. ], batch size: 136, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:00:25,177 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.76 vs. limit=5.0 +2023-04-03 15:00:25,528 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.208e+02 4.889e+02 5.905e+02 7.501e+02 2.168e+03, threshold=1.181e+03, percent-clipped=6.0 +2023-04-03 15:00:26,682 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-03 15:00:35,838 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=197093.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:00:46,566 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-03 15:00:47,958 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.1718, 5.5770, 3.0366, 4.8579, 1.1135, 5.7917, 5.6600, 5.7834], + device='cuda:0'), covar=tensor([0.0390, 0.0865, 0.1917, 0.0694, 0.4071, 0.0521, 0.0755, 0.1162], + device='cuda:0'), in_proj_covar=tensor([0.0534, 0.0433, 0.0520, 0.0358, 0.0409, 0.0459, 0.0454, 0.0487], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 15:01:06,458 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=197118.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:01:24,666 INFO [train.py:903] (0/4) Epoch 29, batch 5950, loss[loss=0.2083, simple_loss=0.2975, pruned_loss=0.05954, over 18064.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2832, pruned_loss=0.06031, over 3808040.70 frames. ], batch size: 83, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:01:32,755 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=197141.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:01:59,523 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4726, 1.2968, 1.6168, 1.6771, 3.0687, 1.4149, 2.4102, 3.4557], + device='cuda:0'), covar=tensor([0.0541, 0.3087, 0.2905, 0.1830, 0.0738, 0.2239, 0.1201, 0.0267], + device='cuda:0'), in_proj_covar=tensor([0.0427, 0.0383, 0.0402, 0.0357, 0.0385, 0.0360, 0.0400, 0.0421], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 15:02:18,285 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2127, 1.2655, 1.2409, 1.0559, 1.0692, 1.0815, 0.0969, 0.3420], + device='cuda:0'), covar=tensor([0.0802, 0.0703, 0.0507, 0.0650, 0.1303, 0.0735, 0.1536, 0.1237], + device='cuda:0'), in_proj_covar=tensor([0.0366, 0.0363, 0.0369, 0.0393, 0.0471, 0.0400, 0.0347, 0.0348], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 15:02:24,688 INFO [train.py:903] (0/4) Epoch 29, batch 6000, loss[loss=0.1926, simple_loss=0.2813, pruned_loss=0.052, over 19664.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2822, pruned_loss=0.05986, over 3822999.83 frames. ], batch size: 58, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:02:24,689 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 15:02:43,151 INFO [train.py:937] (0/4) Epoch 29, validation: loss=0.167, simple_loss=0.2662, pruned_loss=0.03392, over 944034.00 frames. +2023-04-03 15:02:43,152 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 15:02:44,361 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.479e+02 4.704e+02 5.833e+02 7.372e+02 1.660e+03, threshold=1.167e+03, percent-clipped=5.0 +2023-04-03 15:02:48,714 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.06 vs. limit=5.0 +2023-04-03 15:03:23,050 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=197216.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:03:44,117 INFO [train.py:903] (0/4) Epoch 29, batch 6050, loss[loss=0.1953, simple_loss=0.2831, pruned_loss=0.05375, over 18144.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2816, pruned_loss=0.05943, over 3823760.82 frames. ], batch size: 83, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:03:55,194 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=197243.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:04:11,936 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=197256.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:04:45,073 INFO [train.py:903] (0/4) Epoch 29, batch 6100, loss[loss=0.1936, simple_loss=0.278, pruned_loss=0.05462, over 18066.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2822, pruned_loss=0.05958, over 3813204.82 frames. ], batch size: 83, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:04:46,195 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.008e+02 5.045e+02 6.568e+02 8.070e+02 1.422e+03, threshold=1.314e+03, percent-clipped=5.0 +2023-04-03 15:05:17,438 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=197311.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:05:42,450 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=197331.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:05:45,474 INFO [train.py:903] (0/4) Epoch 29, batch 6150, loss[loss=0.1924, simple_loss=0.2761, pruned_loss=0.05436, over 19572.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2832, pruned_loss=0.05992, over 3816737.87 frames. ], batch size: 61, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:05:48,290 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=197336.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:06:13,308 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-03 15:06:14,644 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=197358.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:06:18,107 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0644, 1.0890, 1.2101, 1.1807, 1.4041, 1.4480, 1.4005, 0.6408], + device='cuda:0'), covar=tensor([0.1881, 0.3416, 0.2026, 0.1592, 0.1312, 0.1866, 0.1175, 0.4344], + device='cuda:0'), in_proj_covar=tensor([0.0561, 0.0679, 0.0770, 0.0517, 0.0644, 0.0550, 0.0677, 0.0581], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 15:06:28,788 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3098, 1.9696, 1.5533, 1.4498, 1.8377, 1.3298, 1.4071, 1.7848], + device='cuda:0'), covar=tensor([0.1008, 0.0887, 0.1106, 0.0855, 0.0584, 0.1267, 0.0661, 0.0473], + device='cuda:0'), in_proj_covar=tensor([0.0309, 0.0323, 0.0346, 0.0277, 0.0255, 0.0349, 0.0294, 0.0278], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 15:06:46,118 INFO [train.py:903] (0/4) Epoch 29, batch 6200, loss[loss=0.2248, simple_loss=0.3021, pruned_loss=0.07375, over 19493.00 frames. ], tot_loss[loss=0.2041, simple_loss=0.2852, pruned_loss=0.0615, over 3787015.04 frames. ], batch size: 64, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:06:47,111 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.451e+02 5.150e+02 5.975e+02 7.505e+02 2.002e+03, threshold=1.195e+03, percent-clipped=6.0 +2023-04-03 15:06:49,795 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7832, 1.7145, 1.3533, 1.7204, 1.6109, 1.4234, 1.4717, 1.6049], + device='cuda:0'), covar=tensor([0.1322, 0.1522, 0.2011, 0.1378, 0.1524, 0.1033, 0.1962, 0.1091], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0359, 0.0319, 0.0258, 0.0308, 0.0259, 0.0322, 0.0263], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 15:07:45,763 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=197433.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 15:07:46,599 INFO [train.py:903] (0/4) Epoch 29, batch 6250, loss[loss=0.2233, simple_loss=0.3109, pruned_loss=0.06783, over 19544.00 frames. ], tot_loss[loss=0.2034, simple_loss=0.2844, pruned_loss=0.06119, over 3797314.27 frames. ], batch size: 54, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:08:18,022 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-03 15:08:44,657 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.38 vs. limit=5.0 +2023-04-03 15:08:48,229 INFO [train.py:903] (0/4) Epoch 29, batch 6300, loss[loss=0.1777, simple_loss=0.2524, pruned_loss=0.05151, over 19755.00 frames. ], tot_loss[loss=0.2035, simple_loss=0.2847, pruned_loss=0.06117, over 3793077.12 frames. ], batch size: 47, lr: 2.80e-03, grad_scale: 4.0 +2023-04-03 15:08:50,598 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.231e+02 4.620e+02 5.575e+02 6.491e+02 1.508e+03, threshold=1.115e+03, percent-clipped=2.0 +2023-04-03 15:09:21,986 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=197512.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:09:49,201 INFO [train.py:903] (0/4) Epoch 29, batch 6350, loss[loss=0.2079, simple_loss=0.2944, pruned_loss=0.06074, over 19308.00 frames. ], tot_loss[loss=0.2037, simple_loss=0.2848, pruned_loss=0.06129, over 3800495.23 frames. ], batch size: 66, lr: 2.80e-03, grad_scale: 4.0 +2023-04-03 15:09:49,506 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=197534.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:09:52,872 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=197537.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:10:23,553 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=197563.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:10:49,097 INFO [train.py:903] (0/4) Epoch 29, batch 6400, loss[loss=0.2095, simple_loss=0.295, pruned_loss=0.06198, over 19386.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.2842, pruned_loss=0.06069, over 3809740.91 frames. ], batch size: 70, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:10:52,221 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.915e+02 4.287e+02 5.708e+02 7.339e+02 1.464e+03, threshold=1.142e+03, percent-clipped=5.0 +2023-04-03 15:10:53,753 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=197587.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:10:59,369 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2262, 1.2866, 1.2702, 1.0575, 1.0889, 1.1275, 0.1092, 0.4252], + device='cuda:0'), covar=tensor([0.0774, 0.0729, 0.0488, 0.0663, 0.1415, 0.0726, 0.1504, 0.1259], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0364, 0.0369, 0.0394, 0.0472, 0.0400, 0.0347, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 15:11:23,628 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=197612.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:11:25,916 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=197614.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:11:29,838 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4096, 1.3655, 1.3945, 1.8375, 1.4249, 1.5988, 1.5758, 1.4702], + device='cuda:0'), covar=tensor([0.0908, 0.0966, 0.1085, 0.0695, 0.0868, 0.0842, 0.0867, 0.0790], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0225, 0.0230, 0.0243, 0.0228, 0.0215, 0.0189, 0.0210], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 15:11:49,897 INFO [train.py:903] (0/4) Epoch 29, batch 6450, loss[loss=0.2009, simple_loss=0.2893, pruned_loss=0.05625, over 19746.00 frames. ], tot_loss[loss=0.2028, simple_loss=0.284, pruned_loss=0.06081, over 3801571.89 frames. ], batch size: 63, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:11:56,894 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=197639.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:12:27,630 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.7961, 4.3766, 2.6972, 3.8403, 1.2522, 4.3631, 4.2488, 4.3695], + device='cuda:0'), covar=tensor([0.0582, 0.0987, 0.1914, 0.0839, 0.3697, 0.0597, 0.0918, 0.1013], + device='cuda:0'), in_proj_covar=tensor([0.0532, 0.0431, 0.0519, 0.0356, 0.0409, 0.0458, 0.0451, 0.0485], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 15:12:37,432 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-03 15:12:37,699 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.0125, 4.4506, 4.7708, 4.7628, 1.6876, 4.4460, 3.8262, 4.4918], + device='cuda:0'), covar=tensor([0.1831, 0.0932, 0.0594, 0.0690, 0.6852, 0.1071, 0.0744, 0.1183], + device='cuda:0'), in_proj_covar=tensor([0.0824, 0.0793, 0.1001, 0.0876, 0.0868, 0.0766, 0.0590, 0.0932], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 15:12:49,913 INFO [train.py:903] (0/4) Epoch 29, batch 6500, loss[loss=0.2109, simple_loss=0.2943, pruned_loss=0.0637, over 19770.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2839, pruned_loss=0.0603, over 3803799.19 frames. ], batch size: 56, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:12:52,103 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.446e+02 4.970e+02 6.097e+02 8.194e+02 1.467e+03, threshold=1.219e+03, percent-clipped=8.0 +2023-04-03 15:12:58,637 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-03 15:13:50,290 INFO [train.py:903] (0/4) Epoch 29, batch 6550, loss[loss=0.224, simple_loss=0.3146, pruned_loss=0.06677, over 19667.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2833, pruned_loss=0.05982, over 3815505.91 frames. ], batch size: 55, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:14:24,643 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.9562, 1.2027, 1.5519, 0.6181, 1.8791, 2.3986, 2.0865, 2.5340], + device='cuda:0'), covar=tensor([0.1628, 0.4051, 0.3632, 0.3166, 0.0786, 0.0338, 0.0382, 0.0438], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0334, 0.0367, 0.0275, 0.0256, 0.0199, 0.0221, 0.0279], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 15:14:42,371 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=197777.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 15:14:50,633 INFO [train.py:903] (0/4) Epoch 29, batch 6600, loss[loss=0.178, simple_loss=0.268, pruned_loss=0.044, over 19662.00 frames. ], tot_loss[loss=0.2024, simple_loss=0.2839, pruned_loss=0.06042, over 3801837.29 frames. ], batch size: 55, lr: 2.80e-03, grad_scale: 4.0 +2023-04-03 15:14:54,068 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.141e+02 4.758e+02 5.997e+02 7.159e+02 2.116e+03, threshold=1.199e+03, percent-clipped=4.0 +2023-04-03 15:15:51,179 INFO [train.py:903] (0/4) Epoch 29, batch 6650, loss[loss=0.2047, simple_loss=0.291, pruned_loss=0.05918, over 19601.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2822, pruned_loss=0.05951, over 3813723.62 frames. ], batch size: 61, lr: 2.80e-03, grad_scale: 4.0 +2023-04-03 15:16:45,534 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=197878.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:16:52,896 INFO [train.py:903] (0/4) Epoch 29, batch 6700, loss[loss=0.2083, simple_loss=0.2952, pruned_loss=0.06069, over 18762.00 frames. ], tot_loss[loss=0.1999, simple_loss=0.2816, pruned_loss=0.05915, over 3813597.74 frames. ], batch size: 74, lr: 2.80e-03, grad_scale: 4.0 +2023-04-03 15:16:56,372 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.106e+02 5.093e+02 5.944e+02 7.634e+02 1.783e+03, threshold=1.189e+03, percent-clipped=4.0 +2023-04-03 15:17:02,443 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=197892.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 15:17:19,980 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=197907.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:17:23,570 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=197910.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:17:50,663 INFO [train.py:903] (0/4) Epoch 29, batch 6750, loss[loss=0.1983, simple_loss=0.2792, pruned_loss=0.05866, over 19762.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2813, pruned_loss=0.0587, over 3815494.96 frames. ], batch size: 54, lr: 2.80e-03, grad_scale: 4.0 +2023-04-03 15:18:45,950 INFO [train.py:903] (0/4) Epoch 29, batch 6800, loss[loss=0.238, simple_loss=0.3137, pruned_loss=0.08118, over 19402.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.281, pruned_loss=0.05859, over 3813259.75 frames. ], batch size: 70, lr: 2.80e-03, grad_scale: 8.0 +2023-04-03 15:18:49,169 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.543e+02 4.853e+02 6.089e+02 7.661e+02 1.249e+03, threshold=1.218e+03, percent-clipped=1.0 +2023-04-03 15:18:56,375 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=197993.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:19:03,965 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-198000.pt +2023-04-03 15:19:17,309 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-29.pt +2023-04-03 15:19:33,599 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9 from training. Duration: 27.47775 +2023-04-03 15:19:34,669 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9 from training. Duration: 26.6166875 +2023-04-03 15:19:37,004 INFO [train.py:903] (0/4) Epoch 30, batch 0, loss[loss=0.2249, simple_loss=0.3002, pruned_loss=0.0748, over 19771.00 frames. ], tot_loss[loss=0.2249, simple_loss=0.3002, pruned_loss=0.0748, over 19771.00 frames. ], batch size: 54, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:19:37,004 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 15:19:49,312 INFO [train.py:937] (0/4) Epoch 30, validation: loss=0.167, simple_loss=0.2667, pruned_loss=0.03362, over 944034.00 frames. +2023-04-03 15:19:49,313 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 15:20:02,508 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=198022.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:20:03,317 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425 from training. Duration: 25.775 +2023-04-03 15:20:25,396 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.09 vs. limit=2.0 +2023-04-03 15:20:51,425 INFO [train.py:903] (0/4) Epoch 30, batch 50, loss[loss=0.229, simple_loss=0.3054, pruned_loss=0.07632, over 17614.00 frames. ], tot_loss[loss=0.1938, simple_loss=0.2761, pruned_loss=0.05579, over 872107.73 frames. ], batch size: 101, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:21:20,382 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.592e+02 4.959e+02 5.838e+02 7.711e+02 1.808e+03, threshold=1.168e+03, percent-clipped=5.0 +2023-04-03 15:21:26,250 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1_rvb from training. Duration: 27.0318125 +2023-04-03 15:21:36,610 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=198099.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:21:52,097 INFO [train.py:903] (0/4) Epoch 30, batch 100, loss[loss=0.1592, simple_loss=0.234, pruned_loss=0.04222, over 19741.00 frames. ], tot_loss[loss=0.195, simple_loss=0.2775, pruned_loss=0.05627, over 1545461.15 frames. ], batch size: 45, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:22:04,498 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9 from training. Duration: 29.1166875 +2023-04-03 15:22:37,742 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=198148.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 15:22:54,433 INFO [train.py:903] (0/4) Epoch 30, batch 150, loss[loss=0.1766, simple_loss=0.2654, pruned_loss=0.0439, over 19672.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2801, pruned_loss=0.05747, over 2047784.52 frames. ], batch size: 60, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:23:07,463 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=198173.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 15:23:25,369 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.519e+02 4.863e+02 5.858e+02 7.546e+02 1.579e+03, threshold=1.172e+03, percent-clipped=2.0 +2023-04-03 15:23:52,163 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994 from training. Duration: 27.14 +2023-04-03 15:23:55,535 INFO [train.py:903] (0/4) Epoch 30, batch 200, loss[loss=0.2071, simple_loss=0.2887, pruned_loss=0.06275, over 19640.00 frames. ], tot_loss[loss=0.1961, simple_loss=0.2784, pruned_loss=0.05684, over 2453991.56 frames. ], batch size: 53, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:23:58,278 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5472, 1.6390, 1.8638, 1.8355, 2.7768, 2.3637, 2.9232, 1.3988], + device='cuda:0'), covar=tensor([0.2634, 0.4509, 0.2962, 0.1956, 0.1540, 0.2301, 0.1554, 0.4659], + device='cuda:0'), in_proj_covar=tensor([0.0561, 0.0681, 0.0770, 0.0516, 0.0642, 0.0552, 0.0676, 0.0579], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 15:24:42,391 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=198249.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:24:47,668 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=198254.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:24:57,539 INFO [train.py:903] (0/4) Epoch 30, batch 250, loss[loss=0.2272, simple_loss=0.3003, pruned_loss=0.07709, over 19470.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2792, pruned_loss=0.05787, over 2759369.44 frames. ], batch size: 49, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:25:13,978 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=198274.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:25:19,546 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=198278.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:25:29,212 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.056e+02 4.793e+02 5.720e+02 7.294e+02 1.524e+03, threshold=1.144e+03, percent-clipped=6.0 +2023-04-03 15:25:49,137 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=198303.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:26:01,281 INFO [train.py:903] (0/4) Epoch 30, batch 300, loss[loss=0.1592, simple_loss=0.2396, pruned_loss=0.03939, over 19332.00 frames. ], tot_loss[loss=0.1974, simple_loss=0.2789, pruned_loss=0.0579, over 3004329.50 frames. ], batch size: 44, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:27:03,946 INFO [train.py:903] (0/4) Epoch 30, batch 350, loss[loss=0.1795, simple_loss=0.2688, pruned_loss=0.04515, over 19863.00 frames. ], tot_loss[loss=0.1983, simple_loss=0.2792, pruned_loss=0.05873, over 3170505.65 frames. ], batch size: 52, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:27:06,264 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 15:27:12,103 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=198369.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:27:33,409 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.869e+02 4.876e+02 6.987e+02 8.997e+02 2.429e+03, threshold=1.397e+03, percent-clipped=9.0 +2023-04-03 15:27:48,237 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5972, 1.4318, 2.0364, 1.7728, 3.1165, 4.4695, 4.3289, 4.8586], + device='cuda:0'), covar=tensor([0.1563, 0.4094, 0.3557, 0.2394, 0.0680, 0.0242, 0.0193, 0.0260], + device='cuda:0'), in_proj_covar=tensor([0.0279, 0.0331, 0.0364, 0.0272, 0.0254, 0.0198, 0.0219, 0.0277], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 15:28:04,751 INFO [train.py:903] (0/4) Epoch 30, batch 400, loss[loss=0.1762, simple_loss=0.2611, pruned_loss=0.04566, over 19778.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2812, pruned_loss=0.05957, over 3315189.04 frames. ], batch size: 54, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:28:08,485 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([5.7090, 5.2824, 3.1311, 4.6619, 1.4578, 5.3080, 5.1562, 5.2584], + device='cuda:0'), covar=tensor([0.0400, 0.0727, 0.1820, 0.0635, 0.3548, 0.0550, 0.0810, 0.0918], + device='cuda:0'), in_proj_covar=tensor([0.0535, 0.0435, 0.0523, 0.0359, 0.0412, 0.0460, 0.0454, 0.0488], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 15:28:44,693 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=198443.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:28:56,374 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.4188, 4.0284, 2.6079, 3.5882, 0.8729, 4.0411, 3.8822, 3.9667], + device='cuda:0'), covar=tensor([0.0650, 0.1025, 0.2058, 0.0860, 0.3976, 0.0698, 0.0981, 0.1182], + device='cuda:0'), in_proj_covar=tensor([0.0533, 0.0434, 0.0522, 0.0358, 0.0411, 0.0459, 0.0453, 0.0487], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 15:29:06,540 INFO [train.py:903] (0/4) Epoch 30, batch 450, loss[loss=0.2052, simple_loss=0.2831, pruned_loss=0.06359, over 19769.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2805, pruned_loss=0.059, over 3424628.58 frames. ], batch size: 54, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:29:38,786 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.061e+02 5.102e+02 6.085e+02 7.779e+02 1.785e+03, threshold=1.217e+03, percent-clipped=4.0 +2023-04-03 15:29:40,854 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149 from training. Duration: 25.285 +2023-04-03 15:29:42,036 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9 from training. Duration: 25.3333125 +2023-04-03 15:30:08,498 INFO [train.py:903] (0/4) Epoch 30, batch 500, loss[loss=0.1856, simple_loss=0.271, pruned_loss=0.05004, over 19774.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2816, pruned_loss=0.05939, over 3521856.30 frames. ], batch size: 54, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:30:15,668 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=198517.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:31:06,286 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=198558.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:31:11,502 INFO [train.py:903] (0/4) Epoch 30, batch 550, loss[loss=0.1754, simple_loss=0.2593, pruned_loss=0.04572, over 19749.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.281, pruned_loss=0.05922, over 3599555.86 frames. ], batch size: 51, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:31:40,924 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.425e+02 4.928e+02 6.417e+02 8.667e+02 2.852e+03, threshold=1.283e+03, percent-clipped=10.0 +2023-04-03 15:32:13,288 INFO [train.py:903] (0/4) Epoch 30, batch 600, loss[loss=0.2101, simple_loss=0.2965, pruned_loss=0.06184, over 19769.00 frames. ], tot_loss[loss=0.1984, simple_loss=0.2799, pruned_loss=0.0584, over 3652335.21 frames. ], batch size: 54, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:32:28,418 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=198625.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:32:38,384 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=198633.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:32:52,871 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9 from training. Duration: 27.8166875 +2023-04-03 15:33:01,369 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=198650.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:33:14,980 INFO [train.py:903] (0/4) Epoch 30, batch 650, loss[loss=0.2111, simple_loss=0.295, pruned_loss=0.06357, over 19524.00 frames. ], tot_loss[loss=0.1974, simple_loss=0.2793, pruned_loss=0.05779, over 3678609.31 frames. ], batch size: 56, lr: 2.75e-03, grad_scale: 8.0 +2023-04-03 15:33:46,592 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.134e+02 5.000e+02 5.780e+02 7.067e+02 2.104e+03, threshold=1.156e+03, percent-clipped=2.0 +2023-04-03 15:34:16,543 INFO [train.py:903] (0/4) Epoch 30, batch 700, loss[loss=0.1804, simple_loss=0.2677, pruned_loss=0.04654, over 18251.00 frames. ], tot_loss[loss=0.1983, simple_loss=0.2801, pruned_loss=0.05823, over 3711478.10 frames. ], batch size: 84, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:35:18,689 INFO [train.py:903] (0/4) Epoch 30, batch 750, loss[loss=0.2246, simple_loss=0.2868, pruned_loss=0.08118, over 19777.00 frames. ], tot_loss[loss=0.199, simple_loss=0.2808, pruned_loss=0.05858, over 3741736.86 frames. ], batch size: 45, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:35:29,180 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3228, 1.4238, 1.5733, 1.4976, 1.7428, 1.8157, 1.8177, 0.7196], + device='cuda:0'), covar=tensor([0.2543, 0.4344, 0.2633, 0.2013, 0.1715, 0.2423, 0.1532, 0.4862], + device='cuda:0'), in_proj_covar=tensor([0.0559, 0.0679, 0.0766, 0.0515, 0.0640, 0.0551, 0.0674, 0.0578], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 15:35:51,148 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.284e+02 4.908e+02 5.891e+02 7.920e+02 1.978e+03, threshold=1.178e+03, percent-clipped=7.0 +2023-04-03 15:36:04,801 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.81 vs. limit=2.0 +2023-04-03 15:36:05,385 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1676, 2.1823, 2.5408, 2.8421, 2.1590, 2.6675, 2.5059, 2.3025], + device='cuda:0'), covar=tensor([0.4408, 0.4474, 0.2065, 0.2944, 0.4896, 0.2650, 0.5126, 0.3599], + device='cuda:0'), in_proj_covar=tensor([0.0946, 0.1029, 0.0750, 0.0960, 0.0926, 0.0866, 0.0866, 0.0815], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 15:36:23,528 INFO [train.py:903] (0/4) Epoch 30, batch 800, loss[loss=0.2072, simple_loss=0.2862, pruned_loss=0.06406, over 19790.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2812, pruned_loss=0.05837, over 3772742.10 frames. ], batch size: 56, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:36:26,289 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=198814.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:36:39,843 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1 from training. Duration: 0.9681875 +2023-04-03 15:36:56,797 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=198839.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:37:25,153 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=198861.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:37:26,192 INFO [train.py:903] (0/4) Epoch 30, batch 850, loss[loss=0.2004, simple_loss=0.2881, pruned_loss=0.05641, over 18883.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2817, pruned_loss=0.05875, over 3793097.01 frames. ], batch size: 74, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:37:35,492 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=198870.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:37:55,560 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.084e+02 4.755e+02 5.781e+02 7.125e+02 1.514e+03, threshold=1.156e+03, percent-clipped=6.0 +2023-04-03 15:37:59,378 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.6657, 2.6246, 2.3261, 2.7110, 2.4258, 2.2320, 2.0867, 2.4842], + device='cuda:0'), covar=tensor([0.0978, 0.1453, 0.1392, 0.1070, 0.1388, 0.0551, 0.1455, 0.0731], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0360, 0.0321, 0.0259, 0.0311, 0.0260, 0.0324, 0.0265], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 15:38:13,706 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=198901.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:38:20,330 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9_rvb from training. Duration: 25.061125 +2023-04-03 15:38:26,181 INFO [train.py:903] (0/4) Epoch 30, batch 900, loss[loss=0.1868, simple_loss=0.2737, pruned_loss=0.04997, over 19696.00 frames. ], tot_loss[loss=0.1982, simple_loss=0.2804, pruned_loss=0.05794, over 3816051.52 frames. ], batch size: 53, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:39:07,730 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=198945.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:39:27,822 INFO [train.py:903] (0/4) Epoch 30, batch 950, loss[loss=0.1904, simple_loss=0.2827, pruned_loss=0.04907, over 19764.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2803, pruned_loss=0.05787, over 3820819.08 frames. ], batch size: 56, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:39:32,359 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9_rvb from training. Duration: 26.32775 +2023-04-03 15:39:46,217 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=198976.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:39:48,040 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=198977.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:39:59,668 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.142e+02 5.052e+02 6.305e+02 7.870e+02 1.588e+03, threshold=1.261e+03, percent-clipped=4.0 +2023-04-03 15:40:30,256 INFO [train.py:903] (0/4) Epoch 30, batch 1000, loss[loss=0.1954, simple_loss=0.2747, pruned_loss=0.0581, over 19745.00 frames. ], tot_loss[loss=0.1978, simple_loss=0.2803, pruned_loss=0.05767, over 3827067.28 frames. ], batch size: 51, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:40:46,862 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.80 vs. limit=2.0 +2023-04-03 15:41:23,982 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0026-44402_sp0.9 from training. Duration: 25.061125 +2023-04-03 15:41:33,182 INFO [train.py:903] (0/4) Epoch 30, batch 1050, loss[loss=0.1912, simple_loss=0.2818, pruned_loss=0.05028, over 19327.00 frames. ], tot_loss[loss=0.1977, simple_loss=0.2797, pruned_loss=0.05785, over 3832974.96 frames. ], batch size: 66, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:42:03,435 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.270e+02 4.631e+02 5.961e+02 7.864e+02 1.953e+03, threshold=1.192e+03, percent-clipped=2.0 +2023-04-03 15:42:04,592 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1 from training. Duration: 0.7545625 +2023-04-03 15:42:07,364 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0583, 1.2081, 1.4774, 0.6231, 1.9376, 2.4832, 2.1532, 2.5899], + device='cuda:0'), covar=tensor([0.1488, 0.3943, 0.3690, 0.2991, 0.0713, 0.0290, 0.0354, 0.0390], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0333, 0.0365, 0.0273, 0.0256, 0.0199, 0.0220, 0.0279], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 15:42:10,992 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=199092.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:42:35,182 INFO [train.py:903] (0/4) Epoch 30, batch 1100, loss[loss=0.1952, simple_loss=0.2849, pruned_loss=0.05281, over 19612.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2802, pruned_loss=0.05793, over 3835497.46 frames. ], batch size: 57, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:43:38,415 INFO [train.py:903] (0/4) Epoch 30, batch 1150, loss[loss=0.1803, simple_loss=0.2599, pruned_loss=0.05036, over 19373.00 frames. ], tot_loss[loss=0.1974, simple_loss=0.2793, pruned_loss=0.05769, over 3835161.21 frames. ], batch size: 47, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:44:03,851 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.9060, 1.3963, 1.6240, 1.4640, 3.5328, 1.1311, 2.4999, 3.9724], + device='cuda:0'), covar=tensor([0.0481, 0.2982, 0.2781, 0.2024, 0.0631, 0.2689, 0.1403, 0.0201], + device='cuda:0'), in_proj_covar=tensor([0.0426, 0.0381, 0.0400, 0.0355, 0.0385, 0.0359, 0.0401, 0.0422], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 15:44:10,294 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.218e+02 5.110e+02 5.915e+02 7.639e+02 1.372e+03, threshold=1.183e+03, percent-clipped=6.0 +2023-04-03 15:44:18,764 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=199194.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:44:41,424 INFO [train.py:903] (0/4) Epoch 30, batch 1200, loss[loss=0.1981, simple_loss=0.284, pruned_loss=0.05607, over 19666.00 frames. ], tot_loss[loss=0.1986, simple_loss=0.2806, pruned_loss=0.05825, over 3847549.88 frames. ], batch size: 60, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:44:43,951 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=199214.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:45:06,864 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=199232.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:45:14,390 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983 from training. Duration: 0.83 +2023-04-03 15:45:22,181 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=199245.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:45:38,781 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=199257.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:45:44,964 INFO [train.py:903] (0/4) Epoch 30, batch 1250, loss[loss=0.2132, simple_loss=0.2967, pruned_loss=0.06484, over 19746.00 frames. ], tot_loss[loss=0.1984, simple_loss=0.2808, pruned_loss=0.05804, over 3843948.83 frames. ], batch size: 63, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:46:14,639 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.608e+02 5.004e+02 5.991e+02 7.632e+02 1.398e+03, threshold=1.198e+03, percent-clipped=2.0 +2023-04-03 15:46:17,138 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=199289.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:46:46,512 INFO [train.py:903] (0/4) Epoch 30, batch 1300, loss[loss=0.1751, simple_loss=0.2565, pruned_loss=0.04678, over 19606.00 frames. ], tot_loss[loss=0.1985, simple_loss=0.2808, pruned_loss=0.05811, over 3845229.27 frames. ], batch size: 50, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:47:07,591 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=199329.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:47:22,765 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6082, 2.3373, 1.8384, 1.5834, 2.1589, 1.5186, 1.4254, 2.0383], + device='cuda:0'), covar=tensor([0.1235, 0.0870, 0.1132, 0.1012, 0.0660, 0.1334, 0.0873, 0.0577], + device='cuda:0'), in_proj_covar=tensor([0.0305, 0.0320, 0.0344, 0.0275, 0.0253, 0.0348, 0.0290, 0.0275], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 15:47:32,054 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=199348.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:47:46,062 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=199360.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:47:47,889 INFO [train.py:903] (0/4) Epoch 30, batch 1350, loss[loss=0.2087, simple_loss=0.3004, pruned_loss=0.05856, over 19085.00 frames. ], tot_loss[loss=0.1978, simple_loss=0.2799, pruned_loss=0.05791, over 3839793.06 frames. ], batch size: 69, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 15:48:03,226 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=199373.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:48:13,520 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2389, 2.0544, 1.8703, 2.2054, 1.9018, 1.8284, 1.7198, 2.0879], + device='cuda:0'), covar=tensor([0.1015, 0.1415, 0.1491, 0.1168, 0.1484, 0.0595, 0.1594, 0.0743], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0360, 0.0321, 0.0259, 0.0310, 0.0260, 0.0323, 0.0264], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 15:48:20,885 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.443e+02 4.672e+02 5.739e+02 7.239e+02 1.592e+03, threshold=1.148e+03, percent-clipped=7.0 +2023-04-03 15:48:40,982 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=199404.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:48:50,926 INFO [train.py:903] (0/4) Epoch 30, batch 1400, loss[loss=0.198, simple_loss=0.2872, pruned_loss=0.05434, over 19677.00 frames. ], tot_loss[loss=0.1979, simple_loss=0.2801, pruned_loss=0.05788, over 3823658.17 frames. ], batch size: 60, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 15:48:58,114 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=199418.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:49:55,030 INFO [train.py:903] (0/4) Epoch 30, batch 1450, loss[loss=0.1958, simple_loss=0.2791, pruned_loss=0.05629, over 19676.00 frames. ], tot_loss[loss=0.1976, simple_loss=0.2795, pruned_loss=0.05787, over 3820438.83 frames. ], batch size: 58, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 15:49:56,195 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_rvb from training. Duration: 25.85 +2023-04-03 15:50:11,579 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=199476.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:50:25,541 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.487e+02 5.040e+02 6.121e+02 7.927e+02 1.972e+03, threshold=1.224e+03, percent-clipped=6.0 +2023-04-03 15:50:56,161 INFO [train.py:903] (0/4) Epoch 30, batch 1500, loss[loss=0.1854, simple_loss=0.2749, pruned_loss=0.04792, over 19648.00 frames. ], tot_loss[loss=0.1981, simple_loss=0.2798, pruned_loss=0.05814, over 3816021.92 frames. ], batch size: 55, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 15:50:56,473 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=199512.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 15:51:28,282 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=199538.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:51:46,243 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6889, 1.7076, 1.8569, 1.5387, 4.2440, 1.2121, 2.8212, 4.5309], + device='cuda:0'), covar=tensor([0.0553, 0.2845, 0.2860, 0.2262, 0.0815, 0.2868, 0.1479, 0.0219], + device='cuda:0'), in_proj_covar=tensor([0.0427, 0.0383, 0.0403, 0.0356, 0.0388, 0.0362, 0.0403, 0.0424], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 15:51:47,398 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=199553.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 15:51:58,248 INFO [train.py:903] (0/4) Epoch 30, batch 1550, loss[loss=0.2109, simple_loss=0.2931, pruned_loss=0.06434, over 19657.00 frames. ], tot_loss[loss=0.1975, simple_loss=0.2794, pruned_loss=0.05781, over 3826858.24 frames. ], batch size: 55, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 15:52:28,222 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=199585.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:52:31,301 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.284e+02 4.717e+02 5.804e+02 6.903e+02 1.639e+03, threshold=1.161e+03, percent-clipped=1.0 +2023-04-03 15:52:45,778 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1862, 2.8318, 2.2427, 2.3477, 2.0455, 2.4727, 1.1970, 2.0365], + device='cuda:0'), covar=tensor([0.0732, 0.0680, 0.0739, 0.1227, 0.1170, 0.1182, 0.1539, 0.1204], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0364, 0.0370, 0.0395, 0.0472, 0.0398, 0.0347, 0.0348], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 15:52:59,336 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=199610.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:53:01,355 INFO [train.py:903] (0/4) Epoch 30, batch 1600, loss[loss=0.1738, simple_loss=0.2519, pruned_loss=0.04791, over 19803.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2797, pruned_loss=0.05813, over 3824324.93 frames. ], batch size: 49, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:53:06,435 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=199616.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:53:26,404 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9_rvb from training. Duration: 30.1555625 +2023-04-03 15:53:37,680 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=199641.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:53:53,145 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=199653.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:54:01,260 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=199660.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:54:03,854 INFO [train.py:903] (0/4) Epoch 30, batch 1650, loss[loss=0.2191, simple_loss=0.3067, pruned_loss=0.06574, over 18847.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2798, pruned_loss=0.05813, over 3827715.06 frames. ], batch size: 74, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:54:32,751 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=199685.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:54:35,822 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.122e+02 4.724e+02 5.858e+02 7.730e+02 3.208e+03, threshold=1.172e+03, percent-clipped=5.0 +2023-04-03 15:55:06,080 INFO [train.py:903] (0/4) Epoch 30, batch 1700, loss[loss=0.1741, simple_loss=0.2522, pruned_loss=0.04797, over 19366.00 frames. ], tot_loss[loss=0.1974, simple_loss=0.2795, pruned_loss=0.05761, over 3819352.24 frames. ], batch size: 47, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 15:55:24,025 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6804, 1.6778, 1.6139, 1.4445, 1.3360, 1.4156, 0.3305, 0.7149], + device='cuda:0'), covar=tensor([0.0730, 0.0704, 0.0519, 0.0728, 0.1387, 0.0837, 0.1420, 0.1269], + device='cuda:0'), in_proj_covar=tensor([0.0371, 0.0366, 0.0373, 0.0398, 0.0476, 0.0402, 0.0350, 0.0351], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 15:55:40,547 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.74 vs. limit=2.0 +2023-04-03 15:55:47,583 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590 from training. Duration: 25.85 +2023-04-03 15:56:09,058 INFO [train.py:903] (0/4) Epoch 30, batch 1750, loss[loss=0.208, simple_loss=0.2789, pruned_loss=0.06859, over 19775.00 frames. ], tot_loss[loss=0.1984, simple_loss=0.2804, pruned_loss=0.0582, over 3824623.98 frames. ], batch size: 47, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 15:56:09,219 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=199762.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:56:42,934 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.076e+02 4.893e+02 6.270e+02 7.375e+02 1.627e+03, threshold=1.254e+03, percent-clipped=1.0 +2023-04-03 15:57:00,379 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.2394, 4.3147, 4.8086, 4.8246, 2.7390, 4.5060, 4.0286, 4.5826], + device='cuda:0'), covar=tensor([0.1516, 0.3121, 0.0633, 0.0695, 0.4853, 0.1216, 0.0651, 0.1052], + device='cuda:0'), in_proj_covar=tensor([0.0829, 0.0803, 0.1015, 0.0891, 0.0877, 0.0776, 0.0600, 0.0943], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 15:57:11,412 INFO [train.py:903] (0/4) Epoch 30, batch 1800, loss[loss=0.1997, simple_loss=0.2838, pruned_loss=0.05781, over 19451.00 frames. ], tot_loss[loss=0.1982, simple_loss=0.2799, pruned_loss=0.05826, over 3808683.04 frames. ], batch size: 70, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 15:57:21,128 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=199820.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:58:02,048 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3152, 1.2958, 1.8566, 1.5434, 2.9677, 4.5721, 4.4414, 5.0475], + device='cuda:0'), covar=tensor([0.1850, 0.4401, 0.3735, 0.2574, 0.0716, 0.0228, 0.0209, 0.0228], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0335, 0.0368, 0.0275, 0.0258, 0.0200, 0.0221, 0.0281], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 15:58:06,485 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=199856.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 15:58:08,434 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9_rvb from training. Duration: 25.0944375 +2023-04-03 15:58:13,037 INFO [train.py:903] (0/4) Epoch 30, batch 1850, loss[loss=0.2362, simple_loss=0.3069, pruned_loss=0.08271, over 19601.00 frames. ], tot_loss[loss=0.198, simple_loss=0.2798, pruned_loss=0.05806, over 3817486.75 frames. ], batch size: 61, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 15:58:17,124 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.27 vs. limit=2.0 +2023-04-03 15:58:30,832 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.14 vs. limit=2.0 +2023-04-03 15:58:32,555 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=199877.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:58:46,820 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.213e+02 5.043e+02 6.338e+02 8.325e+02 2.069e+03, threshold=1.268e+03, percent-clipped=7.0 +2023-04-03 15:58:46,868 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_sp0.9_rvb from training. Duration: 27.8166875 +2023-04-03 15:58:57,062 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=199897.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 15:59:12,917 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=199909.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:59:15,995 INFO [train.py:903] (0/4) Epoch 30, batch 1900, loss[loss=0.2194, simple_loss=0.2965, pruned_loss=0.07113, over 19588.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2809, pruned_loss=0.05896, over 3809785.28 frames. ], batch size: 57, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 15:59:21,954 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.85 vs. limit=5.0 +2023-04-03 15:59:33,274 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9 from training. Duration: 29.816625 +2023-04-03 15:59:33,545 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=199927.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:59:38,037 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9_rvb from training. Duration: 27.02225 +2023-04-03 15:59:42,738 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=199934.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 15:59:43,900 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=199935.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:00:03,959 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9 from training. Duration: 0.92225 +2023-04-03 16:00:17,079 INFO [train.py:903] (0/4) Epoch 30, batch 1950, loss[loss=0.2135, simple_loss=0.2905, pruned_loss=0.06824, over 19747.00 frames. ], tot_loss[loss=0.1991, simple_loss=0.2806, pruned_loss=0.05877, over 3827638.14 frames. ], batch size: 54, lr: 2.74e-03, grad_scale: 4.0 +2023-04-03 16:00:28,492 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=199971.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 16:00:51,089 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.452e+02 5.211e+02 6.525e+02 7.685e+02 1.771e+03, threshold=1.305e+03, percent-clipped=2.0 +2023-04-03 16:01:05,405 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-200000.pt +2023-04-03 16:01:21,200 INFO [train.py:903] (0/4) Epoch 30, batch 2000, loss[loss=0.1797, simple_loss=0.2592, pruned_loss=0.05007, over 19620.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2808, pruned_loss=0.05884, over 3828056.15 frames. ], batch size: 50, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 16:01:21,565 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=200012.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 16:02:21,494 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_rvb from training. Duration: 26.8349375 +2023-04-03 16:02:23,756 INFO [train.py:903] (0/4) Epoch 30, batch 2050, loss[loss=0.1963, simple_loss=0.2741, pruned_loss=0.05926, over 19747.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2812, pruned_loss=0.05902, over 3831160.84 frames. ], batch size: 51, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 16:02:43,238 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1 from training. Duration: 0.95 +2023-04-03 16:02:43,273 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp1.1 from training. Duration: 27.0318125 +2023-04-03 16:02:57,798 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.348e+02 4.986e+02 6.252e+02 8.206e+02 1.738e+03, threshold=1.250e+03, percent-clipped=4.0 +2023-04-03 16:03:03,565 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1 from training. Duration: 0.836375 +2023-04-03 16:03:21,109 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3554, 1.4220, 1.5205, 1.5187, 1.7171, 1.8160, 1.7825, 0.5660], + device='cuda:0'), covar=tensor([0.2670, 0.4468, 0.2910, 0.2107, 0.1824, 0.2586, 0.1638, 0.5408], + device='cuda:0'), in_proj_covar=tensor([0.0559, 0.0679, 0.0768, 0.0516, 0.0640, 0.0550, 0.0673, 0.0580], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 16:03:26,468 INFO [train.py:903] (0/4) Epoch 30, batch 2100, loss[loss=0.2132, simple_loss=0.2815, pruned_loss=0.07246, over 13386.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2814, pruned_loss=0.05938, over 3809446.76 frames. ], batch size: 137, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 16:03:32,776 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4820, 2.1009, 1.5691, 1.4137, 1.9124, 1.1999, 1.4767, 1.8982], + device='cuda:0'), covar=tensor([0.1034, 0.0799, 0.1199, 0.0908, 0.0675, 0.1485, 0.0729, 0.0507], + device='cuda:0'), in_proj_covar=tensor([0.0303, 0.0318, 0.0343, 0.0275, 0.0252, 0.0347, 0.0289, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:03:52,976 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=200133.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:03:57,170 WARNING [train.py:1073] (0/4) Exclude cut with ID 453-131332-0000-131866_sp0.9_rvb from training. Duration: 25.3333125 +2023-04-03 16:04:18,934 WARNING [train.py:1073] (0/4) Exclude cut with ID 7699-105389-0094-102071_sp0.9_rvb from training. Duration: 26.6166875 +2023-04-03 16:04:24,909 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=200158.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:04:29,061 INFO [train.py:903] (0/4) Epoch 30, batch 2150, loss[loss=0.2519, simple_loss=0.3188, pruned_loss=0.09252, over 13002.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2815, pruned_loss=0.05932, over 3809361.14 frames. ], batch size: 136, lr: 2.74e-03, grad_scale: 8.0 +2023-04-03 16:05:02,783 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.654e+02 4.819e+02 5.957e+02 8.119e+02 2.108e+03, threshold=1.191e+03, percent-clipped=2.0 +2023-04-03 16:05:05,506 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=200191.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:05:10,684 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=200195.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 16:05:18,887 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2991, 1.9759, 1.5909, 1.3739, 1.8153, 1.2725, 1.3695, 1.8022], + device='cuda:0'), covar=tensor([0.0945, 0.0856, 0.1167, 0.0930, 0.0696, 0.1353, 0.0651, 0.0497], + device='cuda:0'), in_proj_covar=tensor([0.0304, 0.0319, 0.0343, 0.0275, 0.0252, 0.0346, 0.0289, 0.0274], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:05:30,915 INFO [train.py:903] (0/4) Epoch 30, batch 2200, loss[loss=0.2026, simple_loss=0.2923, pruned_loss=0.05646, over 19788.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2824, pruned_loss=0.05972, over 3811871.14 frames. ], batch size: 54, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:05:37,225 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=200216.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:05:50,090 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=200227.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 16:06:11,241 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.5075, 1.4880, 1.6258, 1.5267, 3.1106, 1.2367, 2.4522, 3.5500], + device='cuda:0'), covar=tensor([0.0482, 0.2708, 0.2893, 0.1855, 0.0648, 0.2460, 0.1220, 0.0253], + device='cuda:0'), in_proj_covar=tensor([0.0428, 0.0385, 0.0406, 0.0358, 0.0389, 0.0363, 0.0405, 0.0425], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:06:19,640 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1311, 1.8454, 1.9874, 2.7415, 1.9784, 2.2116, 2.2344, 2.0826], + device='cuda:0'), covar=tensor([0.0793, 0.0876, 0.0925, 0.0716, 0.0901, 0.0782, 0.0933, 0.0663], + device='cuda:0'), in_proj_covar=tensor([0.0213, 0.0224, 0.0229, 0.0242, 0.0228, 0.0216, 0.0189, 0.0210], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 16:06:20,833 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=200252.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 16:06:33,037 INFO [train.py:903] (0/4) Epoch 30, batch 2250, loss[loss=0.1825, simple_loss=0.2638, pruned_loss=0.05061, over 19860.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2829, pruned_loss=0.05991, over 3807230.28 frames. ], batch size: 52, lr: 2.73e-03, grad_scale: 4.0 +2023-04-03 16:06:41,166 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=200268.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 16:06:44,364 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=200271.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:07:08,305 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.532e+02 5.298e+02 6.530e+02 8.599e+02 1.543e+03, threshold=1.306e+03, percent-clipped=6.0 +2023-04-03 16:07:12,065 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=200293.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 16:07:17,586 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=200297.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:07:35,304 INFO [train.py:903] (0/4) Epoch 30, batch 2300, loss[loss=0.2146, simple_loss=0.2987, pruned_loss=0.06522, over 19291.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2836, pruned_loss=0.06013, over 3812799.06 frames. ], batch size: 66, lr: 2.73e-03, grad_scale: 4.0 +2023-04-03 16:07:51,139 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_rvb from training. Duration: 26.205 +2023-04-03 16:08:37,753 INFO [train.py:903] (0/4) Epoch 30, batch 2350, loss[loss=0.1757, simple_loss=0.2564, pruned_loss=0.04752, over 19796.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2832, pruned_loss=0.05987, over 3816309.69 frames. ], batch size: 49, lr: 2.73e-03, grad_scale: 4.0 +2023-04-03 16:09:07,650 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=200386.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:09:12,873 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.379e+02 4.836e+02 5.833e+02 7.163e+02 1.475e+03, threshold=1.167e+03, percent-clipped=1.0 +2023-04-03 16:09:21,112 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_rvb from training. Duration: 25.775 +2023-04-03 16:09:38,177 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9_rvb from training. Duration: 25.45 +2023-04-03 16:09:39,769 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8016, 1.7048, 1.6899, 2.4203, 1.9718, 2.0989, 2.0900, 1.8908], + device='cuda:0'), covar=tensor([0.0770, 0.0843, 0.0933, 0.0619, 0.0773, 0.0691, 0.0812, 0.0642], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0223, 0.0227, 0.0240, 0.0227, 0.0215, 0.0188, 0.0208], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 16:09:40,547 INFO [train.py:903] (0/4) Epoch 30, batch 2400, loss[loss=0.241, simple_loss=0.3151, pruned_loss=0.08347, over 17400.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2829, pruned_loss=0.05965, over 3824367.60 frames. ], batch size: 101, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:10:09,850 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.95 vs. limit=5.0 +2023-04-03 16:10:43,362 INFO [train.py:903] (0/4) Epoch 30, batch 2450, loss[loss=0.2149, simple_loss=0.2996, pruned_loss=0.06514, over 19399.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2825, pruned_loss=0.05939, over 3824879.47 frames. ], batch size: 70, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:11:19,106 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.119e+02 5.121e+02 5.946e+02 7.814e+02 2.121e+03, threshold=1.189e+03, percent-clipped=7.0 +2023-04-03 16:11:46,457 INFO [train.py:903] (0/4) Epoch 30, batch 2500, loss[loss=0.2122, simple_loss=0.2902, pruned_loss=0.06714, over 18260.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2819, pruned_loss=0.05841, over 3837051.94 frames. ], batch size: 84, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:12:20,432 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=200539.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 16:12:48,790 INFO [train.py:903] (0/4) Epoch 30, batch 2550, loss[loss=0.1615, simple_loss=0.2452, pruned_loss=0.0389, over 19123.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2817, pruned_loss=0.05843, over 3830576.42 frames. ], batch size: 42, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:13:23,054 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.619e+02 4.923e+02 6.163e+02 7.973e+02 2.573e+03, threshold=1.233e+03, percent-clipped=12.0 +2023-04-03 16:13:47,194 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9 from training. Duration: 25.988875 +2023-04-03 16:13:51,623 INFO [train.py:903] (0/4) Epoch 30, batch 2600, loss[loss=0.1922, simple_loss=0.2826, pruned_loss=0.05088, over 18791.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2823, pruned_loss=0.05864, over 3836716.69 frames. ], batch size: 74, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:14:09,391 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([0.8892, 1.3659, 1.1150, 1.0369, 1.1870, 1.0441, 0.9228, 1.2388], + device='cuda:0'), covar=tensor([0.0748, 0.0991, 0.1229, 0.0901, 0.0693, 0.1475, 0.0712, 0.0581], + device='cuda:0'), in_proj_covar=tensor([0.0302, 0.0318, 0.0342, 0.0274, 0.0251, 0.0346, 0.0289, 0.0273], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:14:28,338 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=200641.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:14:29,742 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=200642.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:14:44,503 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=200654.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 16:14:54,171 INFO [train.py:903] (0/4) Epoch 30, batch 2650, loss[loss=0.2565, simple_loss=0.3177, pruned_loss=0.09761, over 13263.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.2813, pruned_loss=0.05853, over 3820649.54 frames. ], batch size: 136, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:14:55,575 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=200663.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:15:00,342 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=200667.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:15:16,869 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9_rvb from training. Duration: 27.25 +2023-04-03 16:15:28,264 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.276e+02 4.881e+02 6.412e+02 8.116e+02 1.737e+03, threshold=1.282e+03, percent-clipped=7.0 +2023-04-03 16:15:52,122 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2002, 1.3003, 1.2131, 1.0657, 1.0309, 1.0797, 0.1873, 0.3750], + device='cuda:0'), covar=tensor([0.0849, 0.0868, 0.0573, 0.0742, 0.1670, 0.0969, 0.1618, 0.1523], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0364, 0.0371, 0.0395, 0.0475, 0.0401, 0.0348, 0.0350], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 16:15:55,154 INFO [train.py:903] (0/4) Epoch 30, batch 2700, loss[loss=0.1906, simple_loss=0.2694, pruned_loss=0.05587, over 19487.00 frames. ], tot_loss[loss=0.1985, simple_loss=0.2804, pruned_loss=0.05828, over 3825806.49 frames. ], batch size: 49, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:16:51,764 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=200756.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:16:59,614 INFO [train.py:903] (0/4) Epoch 30, batch 2750, loss[loss=0.2005, simple_loss=0.287, pruned_loss=0.05696, over 19684.00 frames. ], tot_loss[loss=0.1977, simple_loss=0.2798, pruned_loss=0.05774, over 3818786.33 frames. ], batch size: 59, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:17:22,995 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.9401, 1.8966, 1.8537, 1.6693, 1.4737, 1.5880, 0.5401, 0.9378], + device='cuda:0'), covar=tensor([0.0681, 0.0677, 0.0455, 0.0793, 0.1332, 0.0879, 0.1373, 0.1155], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0363, 0.0370, 0.0394, 0.0474, 0.0400, 0.0347, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 16:17:34,086 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.266e+02 4.594e+02 5.767e+02 6.844e+02 1.269e+03, threshold=1.153e+03, percent-clipped=0.0 +2023-04-03 16:18:02,570 INFO [train.py:903] (0/4) Epoch 30, batch 2800, loss[loss=0.1913, simple_loss=0.2845, pruned_loss=0.04907, over 19658.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2809, pruned_loss=0.0584, over 3799620.43 frames. ], batch size: 58, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:19:05,173 INFO [train.py:903] (0/4) Epoch 30, batch 2850, loss[loss=0.1837, simple_loss=0.2634, pruned_loss=0.05199, over 19580.00 frames. ], tot_loss[loss=0.1977, simple_loss=0.2799, pruned_loss=0.05775, over 3814007.58 frames. ], batch size: 52, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:19:26,053 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.69 vs. limit=2.0 +2023-04-03 16:19:34,827 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4329, 1.4485, 1.6712, 1.6397, 3.0610, 1.2039, 2.4336, 3.4557], + device='cuda:0'), covar=tensor([0.0498, 0.2829, 0.2797, 0.1783, 0.0654, 0.2452, 0.1280, 0.0268], + device='cuda:0'), in_proj_covar=tensor([0.0428, 0.0384, 0.0405, 0.0356, 0.0388, 0.0363, 0.0404, 0.0425], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:19:39,284 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.089e+02 4.865e+02 5.683e+02 7.765e+02 1.857e+03, threshold=1.137e+03, percent-clipped=6.0 +2023-04-03 16:20:04,653 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=200910.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 16:20:06,528 INFO [train.py:903] (0/4) Epoch 30, batch 2900, loss[loss=0.2035, simple_loss=0.2923, pruned_loss=0.05733, over 19333.00 frames. ], tot_loss[loss=0.197, simple_loss=0.2793, pruned_loss=0.05735, over 3822582.25 frames. ], batch size: 70, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:20:07,764 WARNING [train.py:1073] (0/4) Exclude cut with ID 8631-249866-0030-64025_sp0.9 from training. Duration: 26.32775 +2023-04-03 16:20:36,315 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=200935.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 16:21:08,767 INFO [train.py:903] (0/4) Epoch 30, batch 2950, loss[loss=0.1993, simple_loss=0.2745, pruned_loss=0.06204, over 19460.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.281, pruned_loss=0.05838, over 3818708.56 frames. ], batch size: 49, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:21:14,435 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.3662, 5.8310, 3.8589, 5.1092, 1.7569, 5.9972, 5.7842, 6.0724], + device='cuda:0'), covar=tensor([0.0338, 0.0802, 0.1486, 0.0761, 0.3656, 0.0473, 0.0777, 0.0902], + device='cuda:0'), in_proj_covar=tensor([0.0537, 0.0434, 0.0522, 0.0361, 0.0411, 0.0461, 0.0454, 0.0491], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:21:36,039 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.6204, 4.2390, 2.9111, 3.7311, 1.4307, 4.2101, 4.0450, 4.1943], + device='cuda:0'), covar=tensor([0.0665, 0.0923, 0.1877, 0.0860, 0.3494, 0.0681, 0.0992, 0.1242], + device='cuda:0'), in_proj_covar=tensor([0.0538, 0.0435, 0.0522, 0.0361, 0.0412, 0.0461, 0.0454, 0.0491], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:21:44,100 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.449e+02 4.844e+02 6.179e+02 7.399e+02 1.416e+03, threshold=1.236e+03, percent-clipped=4.0 +2023-04-03 16:22:05,886 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=4.95 vs. limit=5.0 +2023-04-03 16:22:06,373 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=201007.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:22:11,917 INFO [train.py:903] (0/4) Epoch 30, batch 3000, loss[loss=0.2064, simple_loss=0.2909, pruned_loss=0.0609, over 18812.00 frames. ], tot_loss[loss=0.1992, simple_loss=0.2811, pruned_loss=0.0587, over 3813620.97 frames. ], batch size: 74, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:22:11,917 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 16:22:26,185 INFO [train.py:937] (0/4) Epoch 30, validation: loss=0.1666, simple_loss=0.266, pruned_loss=0.03357, over 944034.00 frames. +2023-04-03 16:22:26,186 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 16:22:26,650 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=201012.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:22:32,380 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_rvb from training. Duration: 29.735 +2023-04-03 16:22:57,393 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=201037.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:23:27,833 INFO [train.py:903] (0/4) Epoch 30, batch 3050, loss[loss=0.1988, simple_loss=0.2737, pruned_loss=0.06199, over 19752.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2821, pruned_loss=0.05923, over 3809623.78 frames. ], batch size: 47, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:24:01,432 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=201089.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:24:03,232 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.915e+02 4.896e+02 5.871e+02 7.483e+02 2.064e+03, threshold=1.174e+03, percent-clipped=5.0 +2023-04-03 16:24:32,128 INFO [train.py:903] (0/4) Epoch 30, batch 3100, loss[loss=0.1833, simple_loss=0.2633, pruned_loss=0.05164, over 19473.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2819, pruned_loss=0.05886, over 3814234.05 frames. ], batch size: 49, lr: 2.73e-03, grad_scale: 4.0 +2023-04-03 16:24:43,634 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=201122.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:25:33,510 INFO [train.py:903] (0/4) Epoch 30, batch 3150, loss[loss=0.1979, simple_loss=0.2881, pruned_loss=0.0539, over 19360.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.2816, pruned_loss=0.05872, over 3810997.60 frames. ], batch size: 66, lr: 2.73e-03, grad_scale: 4.0 +2023-04-03 16:25:55,006 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.3325, 0.9464, 1.3649, 1.3458, 2.7208, 1.0973, 2.3930, 3.1857], + device='cuda:0'), covar=tensor([0.0759, 0.4078, 0.3404, 0.2382, 0.1178, 0.2981, 0.1419, 0.0450], + device='cuda:0'), in_proj_covar=tensor([0.0429, 0.0385, 0.0404, 0.0356, 0.0388, 0.0362, 0.0405, 0.0425], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:26:02,514 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1 from training. Duration: 25.3818125 +2023-04-03 16:26:10,329 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.152e+02 4.625e+02 5.752e+02 7.402e+02 1.953e+03, threshold=1.150e+03, percent-clipped=5.0 +2023-04-03 16:26:35,927 INFO [train.py:903] (0/4) Epoch 30, batch 3200, loss[loss=0.1742, simple_loss=0.255, pruned_loss=0.04673, over 19755.00 frames. ], tot_loss[loss=0.1987, simple_loss=0.281, pruned_loss=0.05814, over 3813566.96 frames. ], batch size: 47, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:27:40,087 INFO [train.py:903] (0/4) Epoch 30, batch 3250, loss[loss=0.2124, simple_loss=0.2966, pruned_loss=0.06408, over 19523.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2814, pruned_loss=0.05808, over 3823998.58 frames. ], batch size: 54, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:27:51,501 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=201271.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:28:14,884 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.466e+02 4.729e+02 5.856e+02 7.119e+02 1.424e+03, threshold=1.171e+03, percent-clipped=4.0 +2023-04-03 16:28:16,382 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=201292.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:28:38,652 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4654, 1.5642, 1.8362, 1.7463, 2.7358, 2.2686, 2.8806, 1.3872], + device='cuda:0'), covar=tensor([0.2583, 0.4492, 0.2974, 0.2058, 0.1499, 0.2271, 0.1470, 0.4668], + device='cuda:0'), in_proj_covar=tensor([0.0555, 0.0678, 0.0765, 0.0514, 0.0636, 0.0548, 0.0671, 0.0580], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 16:28:42,757 INFO [train.py:903] (0/4) Epoch 30, batch 3300, loss[loss=0.2117, simple_loss=0.2923, pruned_loss=0.06551, over 17576.00 frames. ], tot_loss[loss=0.1989, simple_loss=0.2814, pruned_loss=0.05826, over 3814906.49 frames. ], batch size: 101, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:28:49,701 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465 from training. Duration: 26.8349375 +2023-04-03 16:29:19,893 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2571, 1.1697, 1.1788, 1.5708, 1.3681, 1.3669, 1.3795, 1.2720], + device='cuda:0'), covar=tensor([0.0718, 0.0790, 0.0896, 0.0568, 0.0883, 0.0757, 0.0816, 0.0666], + device='cuda:0'), in_proj_covar=tensor([0.0211, 0.0222, 0.0226, 0.0239, 0.0226, 0.0215, 0.0186, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 16:29:44,727 INFO [train.py:903] (0/4) Epoch 30, batch 3350, loss[loss=0.1919, simple_loss=0.2707, pruned_loss=0.05649, over 19725.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2817, pruned_loss=0.05873, over 3810889.96 frames. ], batch size: 51, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:30:05,895 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=201378.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:30:21,793 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.184e+02 4.772e+02 5.759e+02 7.012e+02 1.305e+03, threshold=1.152e+03, percent-clipped=2.0 +2023-04-03 16:30:29,192 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.2197, 2.1220, 2.0121, 1.8428, 1.7392, 1.8637, 0.6812, 1.2057], + device='cuda:0'), covar=tensor([0.0690, 0.0694, 0.0556, 0.0980, 0.1315, 0.0981, 0.1531, 0.1291], + device='cuda:0'), in_proj_covar=tensor([0.0369, 0.0366, 0.0371, 0.0395, 0.0474, 0.0401, 0.0349, 0.0351], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 16:30:37,339 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=201403.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:30:48,356 INFO [train.py:903] (0/4) Epoch 30, batch 3400, loss[loss=0.222, simple_loss=0.304, pruned_loss=0.07001, over 19581.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2824, pruned_loss=0.05911, over 3795587.18 frames. ], batch size: 61, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:30:53,485 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.3132, 1.4007, 1.9985, 1.5552, 3.0997, 4.6645, 4.5429, 5.1281], + device='cuda:0'), covar=tensor([0.1693, 0.4048, 0.3438, 0.2499, 0.0657, 0.0233, 0.0169, 0.0221], + device='cuda:0'), in_proj_covar=tensor([0.0281, 0.0334, 0.0368, 0.0275, 0.0257, 0.0200, 0.0222, 0.0281], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 16:31:15,990 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=201433.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:31:51,802 INFO [train.py:903] (0/4) Epoch 30, batch 3450, loss[loss=0.1743, simple_loss=0.2471, pruned_loss=0.05077, over 18205.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2819, pruned_loss=0.05868, over 3801291.50 frames. ], batch size: 40, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:31:57,383 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9_rvb from training. Duration: 25.2444375 +2023-04-03 16:32:27,319 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.371e+02 5.269e+02 6.567e+02 8.372e+02 2.121e+03, threshold=1.313e+03, percent-clipped=9.0 +2023-04-03 16:32:55,648 INFO [train.py:903] (0/4) Epoch 30, batch 3500, loss[loss=0.1944, simple_loss=0.2794, pruned_loss=0.0547, over 19543.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2834, pruned_loss=0.05947, over 3800380.95 frames. ], batch size: 56, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:33:41,235 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=201548.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:33:58,201 INFO [train.py:903] (0/4) Epoch 30, batch 3550, loss[loss=0.2087, simple_loss=0.2957, pruned_loss=0.06081, over 19542.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2832, pruned_loss=0.05973, over 3793512.14 frames. ], batch size: 56, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:34:21,221 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.51 vs. limit=5.0 +2023-04-03 16:34:35,157 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.339e+02 4.645e+02 6.006e+02 7.208e+02 1.141e+03, threshold=1.201e+03, percent-clipped=0.0 +2023-04-03 16:34:43,644 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.8563, 1.9476, 2.2766, 2.4151, 1.8535, 2.3267, 2.2342, 2.0596], + device='cuda:0'), covar=tensor([0.4265, 0.4061, 0.2121, 0.2503, 0.4231, 0.2374, 0.5316, 0.3611], + device='cuda:0'), in_proj_covar=tensor([0.0952, 0.1036, 0.0754, 0.0963, 0.0929, 0.0870, 0.0869, 0.0820], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 16:35:01,951 INFO [train.py:903] (0/4) Epoch 30, batch 3600, loss[loss=0.1949, simple_loss=0.2824, pruned_loss=0.05376, over 19486.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.282, pruned_loss=0.05933, over 3797600.17 frames. ], batch size: 64, lr: 2.73e-03, grad_scale: 8.0 +2023-04-03 16:35:06,951 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=201615.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:35:22,560 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.7388, 1.5388, 1.6209, 1.4174, 3.2247, 1.1943, 2.5773, 3.7900], + device='cuda:0'), covar=tensor([0.0615, 0.2989, 0.3128, 0.2385, 0.0870, 0.2826, 0.1378, 0.0272], + device='cuda:0'), in_proj_covar=tensor([0.0430, 0.0386, 0.0405, 0.0358, 0.0390, 0.0364, 0.0404, 0.0427], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:35:32,740 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=201636.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:36:04,976 INFO [train.py:903] (0/4) Epoch 30, batch 3650, loss[loss=0.2434, simple_loss=0.3306, pruned_loss=0.07809, over 19685.00 frames. ], tot_loss[loss=0.2015, simple_loss=0.2833, pruned_loss=0.05983, over 3789151.22 frames. ], batch size: 60, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:36:05,277 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=201662.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:36:27,511 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=201679.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:36:42,040 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.871e+02 4.844e+02 5.974e+02 7.556e+02 1.962e+03, threshold=1.195e+03, percent-clipped=4.0 +2023-04-03 16:37:09,559 INFO [train.py:903] (0/4) Epoch 30, batch 3700, loss[loss=0.2136, simple_loss=0.2926, pruned_loss=0.06729, over 17281.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2829, pruned_loss=0.05986, over 3781310.03 frames. ], batch size: 101, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:37:23,482 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=201724.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:37:31,395 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=201730.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:37:44,548 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.38 vs. limit=2.0 +2023-04-03 16:37:57,759 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=201751.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:38:11,312 INFO [train.py:903] (0/4) Epoch 30, batch 3750, loss[loss=0.1623, simple_loss=0.2377, pruned_loss=0.04341, over 19762.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2813, pruned_loss=0.05896, over 3793043.19 frames. ], batch size: 47, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:38:25,360 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.0030, 1.2936, 1.5552, 0.6288, 2.0346, 2.4343, 2.1296, 2.5894], + device='cuda:0'), covar=tensor([0.1658, 0.4000, 0.3737, 0.3090, 0.0708, 0.0331, 0.0393, 0.0445], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0336, 0.0369, 0.0276, 0.0259, 0.0201, 0.0223, 0.0283], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 16:38:47,453 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.241e+02 4.921e+02 6.534e+02 8.381e+02 2.079e+03, threshold=1.307e+03, percent-clipped=7.0 +2023-04-03 16:39:04,358 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=201804.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:39:14,005 INFO [train.py:903] (0/4) Epoch 30, batch 3800, loss[loss=0.2415, simple_loss=0.3231, pruned_loss=0.0799, over 19523.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.281, pruned_loss=0.05877, over 3805157.17 frames. ], batch size: 56, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:39:35,650 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=201829.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:39:41,272 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914_sp0.9_rvb from training. Duration: 29.1166875 +2023-04-03 16:39:43,446 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-04-03 16:40:15,946 INFO [train.py:903] (0/4) Epoch 30, batch 3850, loss[loss=0.21, simple_loss=0.2947, pruned_loss=0.06266, over 19520.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.2811, pruned_loss=0.05873, over 3808928.40 frames. ], batch size: 64, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:40:51,663 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.891e+02 4.977e+02 6.719e+02 8.916e+02 2.147e+03, threshold=1.344e+03, percent-clipped=8.0 +2023-04-03 16:41:18,289 INFO [train.py:903] (0/4) Epoch 30, batch 3900, loss[loss=0.2006, simple_loss=0.2929, pruned_loss=0.05418, over 18851.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2806, pruned_loss=0.05847, over 3818095.06 frames. ], batch size: 74, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:42:20,923 INFO [train.py:903] (0/4) Epoch 30, batch 3950, loss[loss=0.1916, simple_loss=0.2816, pruned_loss=0.05079, over 19349.00 frames. ], tot_loss[loss=0.1997, simple_loss=0.2813, pruned_loss=0.05901, over 3809500.49 frames. ], batch size: 66, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:42:20,941 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9 from training. Duration: 28.0944375 +2023-04-03 16:42:51,071 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=201986.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:42:56,913 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.990e+02 4.814e+02 5.725e+02 7.208e+02 1.816e+03, threshold=1.145e+03, percent-clipped=2.0 +2023-04-03 16:43:08,291 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-202000.pt +2023-04-03 16:43:11,140 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0058, 1.9457, 1.8124, 1.6715, 1.6773, 1.6468, 0.4372, 0.8853], + device='cuda:0'), covar=tensor([0.0649, 0.0668, 0.0473, 0.0776, 0.1156, 0.0842, 0.1365, 0.1209], + device='cuda:0'), in_proj_covar=tensor([0.0369, 0.0366, 0.0371, 0.0394, 0.0473, 0.0400, 0.0348, 0.0351], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 16:43:16,585 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=202006.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:43:18,097 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=202007.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:43:19,102 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=202008.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:43:22,778 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=202011.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:43:24,459 INFO [train.py:903] (0/4) Epoch 30, batch 4000, loss[loss=0.191, simple_loss=0.2724, pruned_loss=0.05487, over 19581.00 frames. ], tot_loss[loss=0.2002, simple_loss=0.2815, pruned_loss=0.05941, over 3807185.64 frames. ], batch size: 52, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:43:38,597 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=202023.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:43:49,619 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=202032.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:44:08,184 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp1.1_rvb from training. Duration: 0.7545625 +2023-04-03 16:44:17,512 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.7296, 1.7516, 1.7295, 1.5432, 1.4440, 1.5197, 0.4577, 0.7162], + device='cuda:0'), covar=tensor([0.0663, 0.0635, 0.0408, 0.0617, 0.1201, 0.0742, 0.1266, 0.1146], + device='cuda:0'), in_proj_covar=tensor([0.0370, 0.0368, 0.0372, 0.0396, 0.0474, 0.0402, 0.0350, 0.0352], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 16:44:27,670 INFO [train.py:903] (0/4) Epoch 30, batch 4050, loss[loss=0.1782, simple_loss=0.2698, pruned_loss=0.04331, over 18640.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2812, pruned_loss=0.05882, over 3805389.26 frames. ], batch size: 74, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:44:34,768 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=202068.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:45:02,985 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.018e+02 4.674e+02 6.115e+02 7.260e+02 2.667e+03, threshold=1.223e+03, percent-clipped=7.0 +2023-04-03 16:45:30,397 INFO [train.py:903] (0/4) Epoch 30, batch 4100, loss[loss=0.1937, simple_loss=0.2626, pruned_loss=0.0624, over 19739.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.2815, pruned_loss=0.05875, over 3808840.13 frames. ], batch size: 51, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:45:40,874 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=202121.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:46:03,029 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=202138.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:46:04,977 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1_rvb from training. Duration: 0.97725 +2023-04-03 16:46:32,907 INFO [train.py:903] (0/4) Epoch 30, batch 4150, loss[loss=0.1905, simple_loss=0.2671, pruned_loss=0.05692, over 19864.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2816, pruned_loss=0.05902, over 3808220.93 frames. ], batch size: 52, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:46:59,278 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=202183.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:47:08,858 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.767e+02 4.823e+02 5.653e+02 7.013e+02 1.196e+03, threshold=1.131e+03, percent-clipped=0.0 +2023-04-03 16:47:34,529 INFO [train.py:903] (0/4) Epoch 30, batch 4200, loss[loss=0.1819, simple_loss=0.2515, pruned_loss=0.05614, over 19755.00 frames. ], tot_loss[loss=0.1994, simple_loss=0.2811, pruned_loss=0.05885, over 3810851.88 frames. ], batch size: 46, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:47:37,801 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9 from training. Duration: 26.438875 +2023-04-03 16:48:22,549 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([4.3951, 4.0133, 2.6489, 3.5331, 1.0932, 4.0112, 3.8250, 3.9054], + device='cuda:0'), covar=tensor([0.0676, 0.0916, 0.1954, 0.0935, 0.3781, 0.0643, 0.0990, 0.1183], + device='cuda:0'), in_proj_covar=tensor([0.0536, 0.0435, 0.0519, 0.0360, 0.0412, 0.0460, 0.0454, 0.0490], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0002, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:48:35,838 INFO [train.py:903] (0/4) Epoch 30, batch 4250, loss[loss=0.2168, simple_loss=0.2942, pruned_loss=0.06967, over 19672.00 frames. ], tot_loss[loss=0.2019, simple_loss=0.2833, pruned_loss=0.06023, over 3807846.23 frames. ], batch size: 55, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:48:52,011 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0003-9465_sp0.9_rvb from training. Duration: 29.816625 +2023-04-03 16:49:03,275 WARNING [train.py:1073] (0/4) Exclude cut with ID 4964-30587-0040-138716_sp0.9 from training. Duration: 25.0944375 +2023-04-03 16:49:12,500 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.375e+02 5.334e+02 6.771e+02 9.277e+02 2.113e+03, threshold=1.354e+03, percent-clipped=7.0 +2023-04-03 16:49:26,578 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.15 vs. limit=2.0 +2023-04-03 16:49:38,448 INFO [train.py:903] (0/4) Epoch 30, batch 4300, loss[loss=0.191, simple_loss=0.2788, pruned_loss=0.05162, over 19763.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2828, pruned_loss=0.05986, over 3819784.87 frames. ], batch size: 54, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:49:41,191 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.78 vs. limit=5.0 +2023-04-03 16:50:27,842 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1200, 2.0319, 1.9135, 1.8190, 1.5568, 1.7266, 0.6288, 1.1917], + device='cuda:0'), covar=tensor([0.0700, 0.0735, 0.0533, 0.0871, 0.1342, 0.1108, 0.1517, 0.1220], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0366, 0.0371, 0.0393, 0.0472, 0.0400, 0.0347, 0.0349], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 16:50:28,701 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=202352.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:50:34,354 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9 from training. Duration: 33.038875 +2023-04-03 16:50:41,236 INFO [train.py:903] (0/4) Epoch 30, batch 4350, loss[loss=0.2202, simple_loss=0.3056, pruned_loss=0.06736, over 19706.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2822, pruned_loss=0.05941, over 3819530.13 frames. ], batch size: 59, lr: 2.72e-03, grad_scale: 4.0 +2023-04-03 16:50:59,658 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=202377.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:51:18,784 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.299e+02 4.712e+02 5.570e+02 7.585e+02 1.545e+03, threshold=1.114e+03, percent-clipped=3.0 +2023-04-03 16:51:21,618 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=202394.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:51:31,108 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=202402.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:51:43,048 INFO [train.py:903] (0/4) Epoch 30, batch 4400, loss[loss=0.1885, simple_loss=0.2666, pruned_loss=0.0552, over 19418.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2821, pruned_loss=0.05926, over 3820364.86 frames. ], batch size: 48, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:51:52,579 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=202419.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:52:08,197 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_rvb from training. Duration: 25.285 +2023-04-03 16:52:17,015 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0009-44385_sp0.9 from training. Duration: 27.02225 +2023-04-03 16:52:17,359 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=202439.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:52:34,058 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.45 vs. limit=5.0 +2023-04-03 16:52:46,168 INFO [train.py:903] (0/4) Epoch 30, batch 4450, loss[loss=0.2239, simple_loss=0.3019, pruned_loss=0.07295, over 18067.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2822, pruned_loss=0.05932, over 3805606.99 frames. ], batch size: 83, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:52:48,916 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=202464.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:52:52,444 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=202467.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:53:23,823 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.801e+02 4.882e+02 5.921e+02 7.342e+02 1.295e+03, threshold=1.184e+03, percent-clipped=2.0 +2023-04-03 16:53:38,941 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([6.3069, 5.7634, 3.1657, 5.0478, 1.4301, 5.8975, 5.8033, 5.8941], + device='cuda:0'), covar=tensor([0.0362, 0.0944, 0.1975, 0.0774, 0.3835, 0.0546, 0.0841, 0.1043], + device='cuda:0'), in_proj_covar=tensor([0.0541, 0.0439, 0.0524, 0.0364, 0.0416, 0.0464, 0.0458, 0.0495], + device='cuda:0'), out_proj_covar=tensor([0.0004, 0.0003, 0.0004, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 16:53:48,736 INFO [train.py:903] (0/4) Epoch 30, batch 4500, loss[loss=0.2259, simple_loss=0.3095, pruned_loss=0.07116, over 19325.00 frames. ], tot_loss[loss=0.1998, simple_loss=0.2819, pruned_loss=0.05888, over 3823662.73 frames. ], batch size: 66, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:54:50,685 INFO [train.py:903] (0/4) Epoch 30, batch 4550, loss[loss=0.227, simple_loss=0.3059, pruned_loss=0.07409, over 18212.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2824, pruned_loss=0.05952, over 3817511.63 frames. ], batch size: 84, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:55:01,220 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9_rvb from training. Duration: 28.72225 +2023-04-03 16:55:25,937 WARNING [train.py:1073] (0/4) Exclude cut with ID 4278-13270-0009-62705_sp0.9 from training. Duration: 25.45 +2023-04-03 16:55:27,104 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.113e+02 5.006e+02 6.254e+02 8.173e+02 1.576e+03, threshold=1.251e+03, percent-clipped=3.0 +2023-04-03 16:55:52,915 INFO [train.py:903] (0/4) Epoch 30, batch 4600, loss[loss=0.2734, simple_loss=0.3396, pruned_loss=0.1036, over 13090.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2824, pruned_loss=0.05988, over 3821064.79 frames. ], batch size: 136, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:56:54,631 INFO [train.py:903] (0/4) Epoch 30, batch 4650, loss[loss=0.211, simple_loss=0.2918, pruned_loss=0.06516, over 19730.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2822, pruned_loss=0.05954, over 3825258.29 frames. ], batch size: 63, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:57:12,740 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_rvb from training. Duration: 0.92 +2023-04-03 16:57:23,009 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_rvb from training. Duration: 0.83 +2023-04-03 16:57:31,779 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.354e+02 5.029e+02 5.947e+02 7.618e+02 1.686e+03, threshold=1.189e+03, percent-clipped=3.0 +2023-04-03 16:57:36,926 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.1262, 1.2592, 1.6598, 1.2039, 2.6535, 3.5206, 3.2241, 3.7234], + device='cuda:0'), covar=tensor([0.1737, 0.4057, 0.3761, 0.2732, 0.0658, 0.0182, 0.0226, 0.0309], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0336, 0.0369, 0.0275, 0.0258, 0.0200, 0.0221, 0.0283], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 16:57:55,569 INFO [train.py:903] (0/4) Epoch 30, batch 4700, loss[loss=0.173, simple_loss=0.2494, pruned_loss=0.04829, over 19728.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2822, pruned_loss=0.05984, over 3824753.29 frames. ], batch size: 46, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:58:10,497 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=202723.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:58:18,367 WARNING [train.py:1073] (0/4) Exclude cut with ID 3033-130750-0096-107983_sp0.9_rvb from training. Duration: 0.92225 +2023-04-03 16:58:41,330 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=202748.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 16:58:59,207 INFO [train.py:903] (0/4) Epoch 30, batch 4750, loss[loss=0.1968, simple_loss=0.2906, pruned_loss=0.05149, over 19554.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2827, pruned_loss=0.05971, over 3819778.85 frames. ], batch size: 56, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 16:59:35,969 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.278e+02 5.055e+02 5.959e+02 7.636e+02 1.705e+03, threshold=1.192e+03, percent-clipped=6.0 +2023-04-03 17:00:00,024 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.7352, 3.2442, 3.2806, 3.2752, 1.4513, 3.1120, 2.7521, 3.0720], + device='cuda:0'), covar=tensor([0.1933, 0.1319, 0.0889, 0.1038, 0.5655, 0.1287, 0.0895, 0.1396], + device='cuda:0'), in_proj_covar=tensor([0.0823, 0.0801, 0.1008, 0.0886, 0.0871, 0.0774, 0.0599, 0.0941], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0002, 0.0001, 0.0002], + device='cuda:0') +2023-04-03 17:00:01,983 INFO [train.py:903] (0/4) Epoch 30, batch 4800, loss[loss=0.2132, simple_loss=0.2983, pruned_loss=0.06406, over 19525.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2829, pruned_loss=0.05962, over 3818887.46 frames. ], batch size: 56, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 17:00:19,400 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2686, 1.4849, 1.9071, 1.7726, 3.0213, 4.5454, 4.4944, 5.0698], + device='cuda:0'), covar=tensor([0.1737, 0.3983, 0.3643, 0.2430, 0.0700, 0.0232, 0.0187, 0.0266], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0337, 0.0368, 0.0275, 0.0258, 0.0200, 0.0221, 0.0283], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0004, 0.0003, 0.0002, 0.0002, 0.0003], + device='cuda:0') +2023-04-03 17:00:25,692 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.83 vs. limit=2.0 +2023-04-03 17:00:39,731 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.1369, 1.7718, 1.7349, 1.9574, 1.6957, 1.8112, 1.6212, 1.9836], + device='cuda:0'), covar=tensor([0.1066, 0.1443, 0.1595, 0.1244, 0.1437, 0.0588, 0.1591, 0.0760], + device='cuda:0'), in_proj_covar=tensor([0.0283, 0.0361, 0.0323, 0.0262, 0.0311, 0.0261, 0.0326, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 17:00:47,813 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=202848.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 17:01:03,241 INFO [train.py:903] (0/4) Epoch 30, batch 4850, loss[loss=0.1843, simple_loss=0.2679, pruned_loss=0.05032, over 19483.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2825, pruned_loss=0.05966, over 3820077.19 frames. ], batch size: 64, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 17:01:27,942 WARNING [train.py:1073] (0/4) Exclude cut with ID 774-127930-0014-48411_sp1.1_rvb from training. Duration: 0.95 +2023-04-03 17:01:41,382 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.446e+02 5.041e+02 6.050e+02 7.335e+02 2.031e+03, threshold=1.210e+03, percent-clipped=4.0 +2023-04-03 17:01:47,272 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1_rvb from training. Duration: 0.9409375 +2023-04-03 17:01:53,244 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467 from training. Duration: 25.035 +2023-04-03 17:01:55,294 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9 from training. Duration: 27.511125 +2023-04-03 17:02:03,499 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp1.1_rvb from training. Duration: 25.3818125 +2023-04-03 17:02:05,916 INFO [train.py:903] (0/4) Epoch 30, batch 4900, loss[loss=0.246, simple_loss=0.3154, pruned_loss=0.08824, over 19562.00 frames. ], tot_loss[loss=0.2014, simple_loss=0.2831, pruned_loss=0.05981, over 3825334.65 frames. ], batch size: 61, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 17:02:25,017 WARNING [train.py:1073] (0/4) Exclude cut with ID 3972-170212-0014-103914 from training. Duration: 26.205 +2023-04-03 17:02:49,242 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 17:03:07,819 INFO [train.py:903] (0/4) Epoch 30, batch 4950, loss[loss=0.258, simple_loss=0.3231, pruned_loss=0.0965, over 19383.00 frames. ], tot_loss[loss=0.2022, simple_loss=0.2837, pruned_loss=0.06032, over 3830939.14 frames. ], batch size: 70, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 17:03:22,390 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0007-9590_sp0.9 from training. Duration: 28.72225 +2023-04-03 17:03:43,837 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.897e+02 4.771e+02 6.104e+02 8.250e+02 2.222e+03, threshold=1.221e+03, percent-clipped=2.0 +2023-04-03 17:03:46,184 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585 from training. Duration: 0.92 +2023-04-03 17:04:09,795 INFO [train.py:903] (0/4) Epoch 30, batch 5000, loss[loss=0.1743, simple_loss=0.2466, pruned_loss=0.05101, over 19761.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2823, pruned_loss=0.05951, over 3834232.52 frames. ], batch size: 46, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 17:04:15,808 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.8541, 1.3102, 1.4899, 1.6529, 3.3805, 1.1911, 2.5494, 3.8353], + device='cuda:0'), covar=tensor([0.0594, 0.3245, 0.3178, 0.1975, 0.0877, 0.2815, 0.1405, 0.0288], + device='cuda:0'), in_proj_covar=tensor([0.0429, 0.0385, 0.0404, 0.0358, 0.0389, 0.0363, 0.0403, 0.0426], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 17:04:16,558 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_rvb from training. Duration: 27.14 +2023-04-03 17:04:28,735 WARNING [train.py:1073] (0/4) Exclude cut with ID 5239-32139-0047-92994_sp0.9 from training. Duration: 30.1555625 +2023-04-03 17:05:02,503 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=203056.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:05:09,013 INFO [train.py:903] (0/4) Epoch 30, batch 5050, loss[loss=0.1719, simple_loss=0.2527, pruned_loss=0.04556, over 19748.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2815, pruned_loss=0.05886, over 3840231.09 frames. ], batch size: 47, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 17:05:45,269 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9_rvb from training. Duration: 26.62775 +2023-04-03 17:05:46,420 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.400e+02 4.511e+02 5.411e+02 6.959e+02 2.884e+03, threshold=1.082e+03, percent-clipped=1.0 +2023-04-03 17:05:59,223 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=203103.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:06:10,209 INFO [train.py:903] (0/4) Epoch 30, batch 5100, loss[loss=0.2406, simple_loss=0.3135, pruned_loss=0.0838, over 13285.00 frames. ], tot_loss[loss=0.1996, simple_loss=0.2818, pruned_loss=0.05873, over 3829954.11 frames. ], batch size: 136, lr: 2.72e-03, grad_scale: 8.0 +2023-04-03 17:06:23,809 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9 from training. Duration: 31.02225 +2023-04-03 17:06:27,361 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0005-9467_rvb from training. Duration: 25.035 +2023-04-03 17:06:30,785 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_rvb from training. Duration: 27.92 +2023-04-03 17:07:12,617 INFO [train.py:903] (0/4) Epoch 30, batch 5150, loss[loss=0.1951, simple_loss=0.2833, pruned_loss=0.05346, over 19494.00 frames. ], tot_loss[loss=0.1988, simple_loss=0.2814, pruned_loss=0.05811, over 3838791.96 frames. ], batch size: 64, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:07:27,927 WARNING [train.py:1073] (0/4) Exclude cut with ID 2411-132532-0017-25057_sp1.1 from training. Duration: 0.9681875 +2023-04-03 17:07:49,458 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.244e+02 5.115e+02 6.435e+02 8.085e+02 2.061e+03, threshold=1.287e+03, percent-clipped=7.0 +2023-04-03 17:07:49,619 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=203192.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 17:08:01,706 WARNING [train.py:1073] (0/4) Exclude cut with ID 4133-6541-0027-26893_sp1.1_rvb from training. Duration: 0.9681875 +2023-04-03 17:08:14,990 INFO [train.py:903] (0/4) Epoch 30, batch 5200, loss[loss=0.2052, simple_loss=0.2802, pruned_loss=0.06511, over 19767.00 frames. ], tot_loss[loss=0.1993, simple_loss=0.282, pruned_loss=0.05828, over 3834663.79 frames. ], batch size: 47, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:08:30,115 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9 from training. Duration: 28.638875 +2023-04-03 17:09:14,165 WARNING [train.py:1073] (0/4) Exclude cut with ID 6709-74022-0004-57021_sp1.1 from training. Duration: 0.9409375 +2023-04-03 17:09:16,394 INFO [train.py:903] (0/4) Epoch 30, batch 5250, loss[loss=0.2124, simple_loss=0.3081, pruned_loss=0.05839, over 19662.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.2821, pruned_loss=0.05849, over 3840136.14 frames. ], batch size: 60, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:09:53,736 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.978e+02 4.929e+02 6.276e+02 8.119e+02 2.486e+03, threshold=1.255e+03, percent-clipped=4.0 +2023-04-03 17:10:12,243 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=203307.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 17:10:17,557 INFO [train.py:903] (0/4) Epoch 30, batch 5300, loss[loss=0.2016, simple_loss=0.2932, pruned_loss=0.05495, over 19721.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2833, pruned_loss=0.05951, over 3825440.03 frames. ], batch size: 51, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:10:36,836 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390 from training. Duration: 27.92 +2023-04-03 17:10:45,012 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0273, 2.0991, 2.4380, 2.5938, 1.9942, 2.5420, 2.4372, 2.2245], + device='cuda:0'), covar=tensor([0.4435, 0.4140, 0.1928, 0.2634, 0.4341, 0.2384, 0.5048, 0.3475], + device='cuda:0'), in_proj_covar=tensor([0.0955, 0.1038, 0.0755, 0.0965, 0.0933, 0.0871, 0.0870, 0.0819], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 17:11:18,766 INFO [train.py:903] (0/4) Epoch 30, batch 5350, loss[loss=0.2075, simple_loss=0.2936, pruned_loss=0.0607, over 19616.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2835, pruned_loss=0.05956, over 3833045.16 frames. ], batch size: 57, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:11:53,885 WARNING [train.py:1073] (0/4) Exclude cut with ID 8291-282929-0024-9607_sp0.9_rvb from training. Duration: 26.438875 +2023-04-03 17:11:56,049 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.690e+02 5.524e+02 6.450e+02 8.325e+02 1.910e+03, threshold=1.290e+03, percent-clipped=5.0 +2023-04-03 17:12:05,343 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=203400.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:12:20,978 INFO [train.py:903] (0/4) Epoch 30, batch 5400, loss[loss=0.2405, simple_loss=0.3162, pruned_loss=0.08242, over 19534.00 frames. ], tot_loss[loss=0.202, simple_loss=0.284, pruned_loss=0.06002, over 3820058.75 frames. ], batch size: 54, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:13:02,933 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=203447.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:13:20,920 INFO [train.py:903] (0/4) Epoch 30, batch 5450, loss[loss=0.2169, simple_loss=0.3017, pruned_loss=0.06604, over 19763.00 frames. ], tot_loss[loss=0.2026, simple_loss=0.2844, pruned_loss=0.06044, over 3798857.13 frames. ], batch size: 51, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:13:30,385 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4992, 1.5674, 1.8339, 1.7732, 2.7615, 2.3231, 2.9728, 1.3612], + device='cuda:0'), covar=tensor([0.2663, 0.4576, 0.2931, 0.2020, 0.1523, 0.2282, 0.1460, 0.4852], + device='cuda:0'), in_proj_covar=tensor([0.0563, 0.0683, 0.0770, 0.0519, 0.0640, 0.0554, 0.0674, 0.0584], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 17:13:50,566 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=203485.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:13:59,461 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.086e+02 4.910e+02 6.531e+02 8.117e+02 1.984e+03, threshold=1.306e+03, percent-clipped=3.0 +2023-04-03 17:14:10,083 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.4689, 1.3851, 1.5914, 1.3924, 3.0961, 1.1108, 2.3585, 3.4991], + device='cuda:0'), covar=tensor([0.0531, 0.2852, 0.2918, 0.1964, 0.0700, 0.2562, 0.1279, 0.0256], + device='cuda:0'), in_proj_covar=tensor([0.0429, 0.0385, 0.0404, 0.0358, 0.0389, 0.0363, 0.0403, 0.0426], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 17:14:23,409 INFO [train.py:903] (0/4) Epoch 30, batch 5500, loss[loss=0.188, simple_loss=0.278, pruned_loss=0.04903, over 19783.00 frames. ], tot_loss[loss=0.2007, simple_loss=0.2827, pruned_loss=0.05934, over 3795541.38 frames. ], batch size: 56, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:14:26,909 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=203515.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:14:48,291 WARNING [train.py:1073] (0/4) Exclude cut with ID 3488-85273-0017-111273_sp0.9_rvb from training. Duration: 27.47775 +2023-04-03 17:15:24,340 INFO [train.py:903] (0/4) Epoch 30, batch 5550, loss[loss=0.2316, simple_loss=0.3091, pruned_loss=0.07705, over 18147.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.2813, pruned_loss=0.05883, over 3810232.83 frames. ], batch size: 83, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:15:24,675 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=203562.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:15:25,788 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=203563.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 17:15:31,874 WARNING [train.py:1073] (0/4) Exclude cut with ID 7255-291500-0009-9471_sp0.9 from training. Duration: 26.62775 +2023-04-03 17:15:56,752 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=203588.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 17:16:01,701 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.118e+02 4.952e+02 6.311e+02 7.819e+02 2.120e+03, threshold=1.262e+03, percent-clipped=2.0 +2023-04-03 17:16:09,340 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.50 vs. limit=2.0 +2023-04-03 17:16:22,138 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403 from training. Duration: 29.735 +2023-04-03 17:16:27,768 INFO [train.py:903] (0/4) Epoch 30, batch 5600, loss[loss=0.2357, simple_loss=0.3055, pruned_loss=0.08299, over 13742.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2824, pruned_loss=0.05993, over 3796603.12 frames. ], batch size: 136, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:16:29,344 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.2481, 1.2157, 1.2551, 1.3790, 1.0943, 1.3396, 1.2751, 1.3140], + device='cuda:0'), covar=tensor([0.0918, 0.0931, 0.1030, 0.0624, 0.0830, 0.0852, 0.0819, 0.0782], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0222, 0.0227, 0.0238, 0.0226, 0.0215, 0.0187, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 17:17:29,533 INFO [train.py:903] (0/4) Epoch 30, batch 5650, loss[loss=0.1824, simple_loss=0.2614, pruned_loss=0.05174, over 19787.00 frames. ], tot_loss[loss=0.2003, simple_loss=0.2818, pruned_loss=0.05938, over 3803639.62 frames. ], batch size: 48, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:18:06,339 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.203e+02 4.913e+02 5.869e+02 7.597e+02 1.607e+03, threshold=1.174e+03, percent-clipped=2.0 +2023-04-03 17:18:18,654 WARNING [train.py:1073] (0/4) Exclude cut with ID 6951-79737-0043-83149_sp0.9_rvb from training. Duration: 28.0944375 +2023-04-03 17:18:29,974 INFO [train.py:903] (0/4) Epoch 30, batch 5700, loss[loss=0.2251, simple_loss=0.2926, pruned_loss=0.07878, over 19462.00 frames. ], tot_loss[loss=0.2001, simple_loss=0.2817, pruned_loss=0.05929, over 3812060.69 frames. ], batch size: 49, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:19:31,946 INFO [train.py:903] (0/4) Epoch 30, batch 5750, loss[loss=0.2169, simple_loss=0.2971, pruned_loss=0.0683, over 19114.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2821, pruned_loss=0.05958, over 3818445.51 frames. ], batch size: 69, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:19:35,360 WARNING [train.py:1073] (0/4) Exclude cut with ID 2929-85685-0079-61403_sp0.9_rvb from training. Duration: 33.038875 +2023-04-03 17:19:35,648 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=203765.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:19:44,335 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=203771.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:19:45,202 WARNING [train.py:1073] (0/4) Exclude cut with ID 7859-102521-0017-21930_sp0.9 from training. Duration: 27.25 +2023-04-03 17:19:51,012 WARNING [train.py:1073] (0/4) Exclude cut with ID 497-129325-0061-9566_sp1.1 from training. Duration: 0.97725 +2023-04-03 17:20:00,638 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=203784.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:20:09,170 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.085e+02 5.021e+02 5.985e+02 7.897e+02 1.857e+03, threshold=1.197e+03, percent-clipped=7.0 +2023-04-03 17:20:15,313 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=203796.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:20:35,334 INFO [train.py:903] (0/4) Epoch 30, batch 5800, loss[loss=0.2436, simple_loss=0.3228, pruned_loss=0.08223, over 19684.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2827, pruned_loss=0.05991, over 3814682.66 frames. ], batch size: 59, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:20:43,781 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=203818.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:20:56,427 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=203829.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:21:13,735 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=203843.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:21:30,585 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.5652, 2.1941, 1.6325, 1.5798, 1.9913, 1.3053, 1.5160, 1.9094], + device='cuda:0'), covar=tensor([0.1087, 0.0851, 0.1104, 0.0871, 0.0607, 0.1362, 0.0727, 0.0554], + device='cuda:0'), in_proj_covar=tensor([0.0306, 0.0323, 0.0346, 0.0279, 0.0256, 0.0350, 0.0292, 0.0280], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 17:21:37,906 INFO [train.py:903] (0/4) Epoch 30, batch 5850, loss[loss=0.2249, simple_loss=0.3085, pruned_loss=0.0707, over 19659.00 frames. ], tot_loss[loss=0.2017, simple_loss=0.2834, pruned_loss=0.05998, over 3807837.05 frames. ], batch size: 58, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:21:41,678 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=203865.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:22:15,806 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.367e+02 4.919e+02 5.824e+02 7.977e+02 1.569e+03, threshold=1.165e+03, percent-clipped=4.0 +2023-04-03 17:22:20,468 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0083, 1.9498, 1.6975, 2.0428, 1.8465, 1.7322, 1.6109, 1.9285], + device='cuda:0'), covar=tensor([0.1168, 0.1437, 0.1686, 0.1105, 0.1379, 0.0614, 0.1686, 0.0853], + device='cuda:0'), in_proj_covar=tensor([0.0282, 0.0363, 0.0324, 0.0261, 0.0312, 0.0261, 0.0327, 0.0266], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0004, 0.0004, 0.0003, 0.0003, 0.0003, 0.0004, 0.0003], + device='cuda:0') +2023-04-03 17:22:39,481 INFO [train.py:903] (0/4) Epoch 30, batch 5900, loss[loss=0.172, simple_loss=0.2598, pruned_loss=0.04207, over 19668.00 frames. ], tot_loss[loss=0.2006, simple_loss=0.2828, pruned_loss=0.05924, over 3821149.58 frames. ], batch size: 58, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:22:42,676 WARNING [train.py:1073] (0/4) Exclude cut with ID 1914-133440-0024-53073_sp0.9 from training. Duration: 25.2444375 +2023-04-03 17:22:45,444 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=2.54 vs. limit=5.0 +2023-04-03 17:22:47,235 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=203918.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 17:23:06,097 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0021-44397_sp0.9_rvb from training. Duration: 27.511125 +2023-04-03 17:23:08,132 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.18 vs. limit=2.0 +2023-04-03 17:23:19,584 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=203944.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:23:21,446 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-03 17:23:40,675 INFO [train.py:903] (0/4) Epoch 30, batch 5950, loss[loss=0.2017, simple_loss=0.2893, pruned_loss=0.05708, over 19589.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2834, pruned_loss=0.05963, over 3829804.17 frames. ], batch size: 61, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:24:18,148 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.982e+02 5.092e+02 6.183e+02 7.281e+02 1.501e+03, threshold=1.237e+03, percent-clipped=4.0 +2023-04-03 17:24:28,252 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/checkpoint-204000.pt +2023-04-03 17:24:45,143 INFO [train.py:903] (0/4) Epoch 30, batch 6000, loss[loss=0.2127, simple_loss=0.2975, pruned_loss=0.06394, over 17351.00 frames. ], tot_loss[loss=0.2005, simple_loss=0.2827, pruned_loss=0.05911, over 3812607.32 frames. ], batch size: 101, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:24:45,144 INFO [train.py:928] (0/4) Computing validation loss +2023-04-03 17:24:54,740 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4881, 1.5356, 1.8355, 1.8573, 1.3966, 1.7772, 1.7675, 1.6420], + device='cuda:0'), covar=tensor([0.3936, 0.4292, 0.1894, 0.2544, 0.4383, 0.2385, 0.4573, 0.3370], + device='cuda:0'), in_proj_covar=tensor([0.0955, 0.1039, 0.0755, 0.0968, 0.0934, 0.0872, 0.0868, 0.0820], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 17:24:58,730 INFO [train.py:937] (0/4) Epoch 30, validation: loss=0.167, simple_loss=0.2658, pruned_loss=0.03407, over 944034.00 frames. +2023-04-03 17:24:58,732 INFO [train.py:938] (0/4) Maximum memory allocated so far is 18803MB +2023-04-03 17:25:11,720 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.0577, 3.3204, 2.0615, 1.6066, 3.1201, 1.5407, 1.6433, 2.3812], + device='cuda:0'), covar=tensor([0.1117, 0.0503, 0.0841, 0.0879, 0.0416, 0.1133, 0.0782, 0.0560], + device='cuda:0'), in_proj_covar=tensor([0.0308, 0.0324, 0.0347, 0.0279, 0.0257, 0.0351, 0.0294, 0.0281], + device='cuda:0'), out_proj_covar=tensor([0.0003, 0.0003, 0.0003, 0.0003, 0.0002, 0.0003, 0.0003, 0.0003], + device='cuda:0') +2023-04-03 17:25:21,218 INFO [scaling.py:679] (0/4) Whitening: num_groups=1, num_channels=256, metric=3.40 vs. limit=5.0 +2023-04-03 17:26:02,074 INFO [train.py:903] (0/4) Epoch 30, batch 6050, loss[loss=0.1883, simple_loss=0.2821, pruned_loss=0.04724, over 19666.00 frames. ], tot_loss[loss=0.1995, simple_loss=0.282, pruned_loss=0.05853, over 3820175.19 frames. ], batch size: 58, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:26:17,195 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.66 vs. limit=2.0 +2023-04-03 17:26:38,220 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.289e+02 5.054e+02 6.352e+02 7.854e+02 1.582e+03, threshold=1.270e+03, percent-clipped=1.0 +2023-04-03 17:26:42,111 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=204095.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:26:57,988 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.13 vs. limit=2.0 +2023-04-03 17:27:00,739 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=204109.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:27:04,103 INFO [train.py:903] (0/4) Epoch 30, batch 6100, loss[loss=0.2409, simple_loss=0.3132, pruned_loss=0.08434, over 19618.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2831, pruned_loss=0.05946, over 3818509.64 frames. ], batch size: 57, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:27:22,493 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=204128.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:28:04,388 INFO [train.py:903] (0/4) Epoch 30, batch 6150, loss[loss=0.1716, simple_loss=0.2555, pruned_loss=0.04389, over 19118.00 frames. ], tot_loss[loss=0.2023, simple_loss=0.2841, pruned_loss=0.06024, over 3812607.14 frames. ], batch size: 42, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:28:29,853 INFO [zipformer.py:1188] (0/4) warmup_begin=1333.3, warmup_end=2000.0, batch_count=204181.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:28:34,972 WARNING [train.py:1073] (0/4) Exclude cut with ID 7357-94126-0014-44390_sp0.9_rvb from training. Duration: 31.02225 +2023-04-03 17:28:42,858 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.206e+02 5.061e+02 6.485e+02 7.945e+02 1.594e+03, threshold=1.297e+03, percent-clipped=2.0 +2023-04-03 17:28:52,784 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=204200.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:29:03,790 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=204209.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:29:07,246 INFO [train.py:903] (0/4) Epoch 30, batch 6200, loss[loss=0.1588, simple_loss=0.2396, pruned_loss=0.03904, over 19772.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2833, pruned_loss=0.05958, over 3820228.95 frames. ], batch size: 48, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:29:24,464 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=204224.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:29:25,548 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=204225.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:29:45,904 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=204243.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:30:09,328 INFO [train.py:903] (0/4) Epoch 30, batch 6250, loss[loss=0.2136, simple_loss=0.2923, pruned_loss=0.06745, over 19121.00 frames. ], tot_loss[loss=0.2013, simple_loss=0.2832, pruned_loss=0.05966, over 3817723.36 frames. ], batch size: 69, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:30:10,639 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=204262.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 17:30:38,588 WARNING [train.py:1073] (0/4) Exclude cut with ID 6758-72288-0033-148662_sp0.9_rvb from training. Duration: 25.988875 +2023-04-03 17:30:46,054 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.091e+02 5.004e+02 5.938e+02 7.424e+02 1.100e+03, threshold=1.188e+03, percent-clipped=0.0 +2023-04-03 17:31:10,305 INFO [train.py:903] (0/4) Epoch 30, batch 6300, loss[loss=0.2081, simple_loss=0.2816, pruned_loss=0.06728, over 19859.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2826, pruned_loss=0.05909, over 3820009.07 frames. ], batch size: 52, lr: 2.71e-03, grad_scale: 4.0 +2023-04-03 17:31:12,722 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.4860, 1.4471, 1.4824, 1.8416, 1.3115, 1.6186, 1.5815, 1.5804], + device='cuda:0'), covar=tensor([0.0928, 0.0965, 0.1047, 0.0647, 0.0860, 0.0866, 0.0897, 0.0763], + device='cuda:0'), in_proj_covar=tensor([0.0212, 0.0221, 0.0228, 0.0238, 0.0226, 0.0215, 0.0187, 0.0207], + device='cuda:0'), out_proj_covar=tensor([0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0005, 0.0004, 0.0005], + device='cuda:0') +2023-04-03 17:31:24,171 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=204324.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:32:11,313 INFO [train.py:903] (0/4) Epoch 30, batch 6350, loss[loss=0.1996, simple_loss=0.2869, pruned_loss=0.05618, over 19707.00 frames. ], tot_loss[loss=0.2004, simple_loss=0.2824, pruned_loss=0.05922, over 3813078.36 frames. ], batch size: 59, lr: 2.71e-03, grad_scale: 4.0 +2023-04-03 17:32:30,530 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=204377.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 17:32:50,298 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.055e+02 4.728e+02 6.049e+02 7.442e+02 1.533e+03, threshold=1.210e+03, percent-clipped=4.0 +2023-04-03 17:33:12,659 INFO [train.py:903] (0/4) Epoch 30, batch 6400, loss[loss=0.1842, simple_loss=0.2684, pruned_loss=0.05001, over 19748.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.2832, pruned_loss=0.05955, over 3812173.32 frames. ], batch size: 51, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:33:45,675 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=204439.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:34:13,521 INFO [train.py:903] (0/4) Epoch 30, batch 6450, loss[loss=0.2117, simple_loss=0.2979, pruned_loss=0.06271, over 19665.00 frames. ], tot_loss[loss=0.2011, simple_loss=0.283, pruned_loss=0.05961, over 3812500.88 frames. ], batch size: 55, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:34:35,926 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=204480.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:34:51,302 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.580e+02 4.981e+02 6.276e+02 8.010e+02 2.155e+03, threshold=1.255e+03, percent-clipped=8.0 +2023-04-03 17:34:54,951 WARNING [train.py:1073] (0/4) Exclude cut with ID 5622-44585-0006-50425_sp0.9_rvb from training. Duration: 28.638875 +2023-04-03 17:34:59,449 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=204499.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:35:08,012 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=204505.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:35:11,893 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.65 vs. limit=2.0 +2023-04-03 17:35:16,836 INFO [train.py:903] (0/4) Epoch 30, batch 6500, loss[loss=0.2101, simple_loss=0.2991, pruned_loss=0.0605, over 19698.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2829, pruned_loss=0.05974, over 3820588.58 frames. ], batch size: 59, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:35:19,123 WARNING [train.py:1073] (0/4) Exclude cut with ID 3557-8342-0013-71585_sp1.1_rvb from training. Duration: 0.836375 +2023-04-03 17:35:30,896 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=204524.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:35:31,791 INFO [zipformer.py:1188] (0/4) warmup_begin=666.7, warmup_end=1333.3, batch_count=204525.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:36:09,128 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=204554.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:36:17,963 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=192, metric=1.78 vs. limit=2.0 +2023-04-03 17:36:18,056 INFO [train.py:903] (0/4) Epoch 30, batch 6550, loss[loss=0.231, simple_loss=0.3193, pruned_loss=0.0713, over 19552.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2824, pruned_loss=0.05971, over 3821735.51 frames. ], batch size: 54, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:36:21,768 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([2.3102, 2.1354, 1.9696, 1.8493, 1.6086, 1.8156, 0.9201, 1.2540], + device='cuda:0'), covar=tensor([0.0704, 0.0767, 0.0613, 0.1065, 0.1428, 0.1170, 0.1402, 0.1257], + device='cuda:0'), in_proj_covar=tensor([0.0368, 0.0367, 0.0371, 0.0396, 0.0475, 0.0400, 0.0350, 0.0351], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 17:36:39,390 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([3.1933, 2.8940, 2.2905, 2.2515, 2.0730, 2.6209, 1.1226, 2.0783], + device='cuda:0'), covar=tensor([0.0692, 0.0648, 0.0747, 0.1248, 0.1155, 0.1111, 0.1417, 0.1160], + device='cuda:0'), in_proj_covar=tensor([0.0367, 0.0367, 0.0371, 0.0395, 0.0474, 0.0400, 0.0349, 0.0350], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002, 0.0002, 0.0002], + device='cuda:0') +2023-04-03 17:36:40,570 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=204580.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:36:56,735 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.39 vs. limit=2.0 +2023-04-03 17:36:56,887 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.390e+02 5.288e+02 6.916e+02 9.470e+02 2.608e+03, threshold=1.383e+03, percent-clipped=11.0 +2023-04-03 17:37:11,416 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=204605.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:37:20,114 INFO [train.py:903] (0/4) Epoch 30, batch 6600, loss[loss=0.182, simple_loss=0.2644, pruned_loss=0.04975, over 19603.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2828, pruned_loss=0.0598, over 3825812.54 frames. ], batch size: 52, lr: 2.71e-03, grad_scale: 8.0 +2023-04-03 17:37:46,903 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=204633.0, num_to_drop=1, layers_to_drop={0} +2023-04-03 17:37:54,578 INFO [zipformer.py:1188] (0/4) warmup_begin=2000.0, warmup_end=2666.7, batch_count=204640.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:38:17,108 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=204658.0, num_to_drop=1, layers_to_drop={1} +2023-04-03 17:38:21,421 INFO [train.py:903] (0/4) Epoch 30, batch 6650, loss[loss=0.2293, simple_loss=0.3094, pruned_loss=0.07456, over 19722.00 frames. ], tot_loss[loss=0.2016, simple_loss=0.2832, pruned_loss=0.06, over 3819669.79 frames. ], batch size: 63, lr: 2.70e-03, grad_scale: 8.0 +2023-04-03 17:38:33,351 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.09 vs. limit=2.0 +2023-04-03 17:38:59,519 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 3.417e+02 4.923e+02 6.497e+02 8.594e+02 3.232e+03, threshold=1.299e+03, percent-clipped=5.0 +2023-04-03 17:39:24,423 INFO [train.py:903] (0/4) Epoch 30, batch 6700, loss[loss=0.2118, simple_loss=0.2912, pruned_loss=0.06615, over 19541.00 frames. ], tot_loss[loss=0.2009, simple_loss=0.2823, pruned_loss=0.05976, over 3818842.80 frames. ], batch size: 54, lr: 2.70e-03, grad_scale: 4.0 +2023-04-03 17:40:21,767 INFO [train.py:903] (0/4) Epoch 30, batch 6750, loss[loss=0.2123, simple_loss=0.2786, pruned_loss=0.07298, over 19297.00 frames. ], tot_loss[loss=0.2012, simple_loss=0.2824, pruned_loss=0.06005, over 3821451.34 frames. ], batch size: 44, lr: 2.70e-03, grad_scale: 4.0 +2023-04-03 17:40:31,887 INFO [zipformer.py:2441] (0/4) attn_weights_entropy = tensor([1.6385, 1.7560, 1.9958, 1.8657, 2.6923, 2.3478, 2.8272, 1.5101], + device='cuda:0'), covar=tensor([0.2433, 0.4124, 0.2626, 0.1948, 0.1604, 0.2275, 0.1509, 0.4473], + device='cuda:0'), in_proj_covar=tensor([0.0563, 0.0685, 0.0771, 0.0518, 0.0640, 0.0554, 0.0674, 0.0584], + device='cuda:0'), out_proj_covar=tensor([0.0002, 0.0003, 0.0003, 0.0002, 0.0002, 0.0002, 0.0003, 0.0002], + device='cuda:0') +2023-04-03 17:40:34,745 INFO [scaling.py:679] (0/4) Whitening: num_groups=8, num_channels=96, metric=1.35 vs. limit=2.0 +2023-04-03 17:40:58,541 INFO [optim.py:369] (0/4) Clipping_scale=2.0, grad-norm quartiles 2.758e+02 4.849e+02 6.262e+02 9.905e+02 2.863e+03, threshold=1.252e+03, percent-clipped=11.0 +2023-04-03 17:41:17,955 INFO [zipformer.py:1188] (0/4) warmup_begin=3333.3, warmup_end=4000.0, batch_count=204810.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:41:19,750 INFO [train.py:903] (0/4) Epoch 30, batch 6800, loss[loss=0.2134, simple_loss=0.2954, pruned_loss=0.06564, over 17354.00 frames. ], tot_loss[loss=0.201, simple_loss=0.2825, pruned_loss=0.05976, over 3826462.74 frames. ], batch size: 101, lr: 2.70e-03, grad_scale: 8.0 +2023-04-03 17:41:45,179 INFO [zipformer.py:1188] (0/4) warmup_begin=2666.7, warmup_end=3333.3, batch_count=204835.0, num_to_drop=0, layers_to_drop=set() +2023-04-03 17:41:50,448 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless7_streaming/exp/v2/epoch-30.pt +2023-04-03 17:41:51,270 INFO [train.py:1171] (0/4) Done!