decoding_results/ctc-decoding/log-decode-epoch-20-avg-1-use-averaged-model-2023-10-17-18-28-40

2023-10-17 18:28:40,557 INFO [ctc_decode.py:560] Decoding started
2023-10-17 18:28:40,557 INFO [ctc_decode.py:566] Device: cuda:0
2023-10-17 18:28:40,557 INFO [ctc_decode.py:567] {'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': '1.24.3', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': '821ebc378e7fb99b8adc81950227963332821e01', 'k2-git-date': 'Wed Jul 19 15:38:25 2023', 'lhotse-version': '1.16.0.dev+git.1db4d97a.clean', 'torch-version': '1.11.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.9', 'icefall-git-branch': 'dev_multi_zh-hans', 'icefall-git-sha1': '919793d3-dirty', 'icefall-git-date': 'Thu Sep 7 21:06:37 2023', 'icefall-path': '/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/icefall-1.0-py3.9.egg', 'k2-path': '/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/k2-1.24.3.dev20230721+cuda10.2.torch1.11.0-py3.9-linux-x86_64.egg/k2/__init__.py', 'lhotse-path': '/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/lhotse-1.16.0.dev0+git.1db4d97a.clean-py3.9.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-1-1220091118-57c4d55446-mvd6x', 'IP address': '10.177.22.19'}, 'frame_shift_ms': 10, 'search_beam': 20, 'output_beam': 8, 'min_active_states': 30, 'max_active_states': 10000, 'use_double_scores': True, 'epoch': 20, 'iter': 0, 'avg': 1, 'use_averaged_model': True, 'exp_dir': PosixPath('zipformer/exp-w-ctc'), 'bpe_model': 'data/lang_bpe_2000/bpe.model', 'lang_dir': PosixPath('data/lang_bpe_2000'), 'context_size': 2, 'decoding_method': 'ctc-decoding', 'num_paths': 100, 'nbest_scale': 1.0, 'num_encoder_layers': '2,2,3,4,3,2', 'downsampling_factor': '1,2,4,8,4,2', 'feedforward_dim': '512,768,1024,1536,1024,768', 'num_heads': '4,4,4,8,4,4', 'encoder_dim': '192,256,384,512,384,256', 'query_head_dim': '32', 'value_head_dim': '12', 'pos_head_dim': '4', 'pos_dim': 48, 'encoder_unmasked_dim': '192,192,256,256,256,192', 'cnn_module_kernel': '31,31,15,15,15,31', 'decoder_dim': 512, 'joiner_dim': 512, 'causal': False, 'chunk_size': '16,32,64,-1', 'left_context_frames': '64,128,256,-1', 'use_transducer': True, 'use_ctc': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1200, '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', 'res_dir': PosixPath('zipformer/exp-w-ctc/ctc-decoding'), 'suffix': 'epoch-20-avg-1-use-averaged-model'}
2023-10-17 18:28:42,268 INFO [lexicon.py:168] Loading pre-compiled data/lang_bpe_2000/Linv.pt
2023-10-17 18:28:48,063 INFO [ctc_decode.py:587] About to create model
2023-10-17 18:28:48,655 INFO [ctc_decode.py:654] Calculating the averaged model over epoch range from 19 (excluded) to 20
2023-10-17 18:28:53,461 INFO [ctc_decode.py:671] Number of model parameters: 69651511
2023-10-17 18:28:53,462 INFO [multi_dataset.py:221] About to get multidataset test cuts
2023-10-17 18:28:53,462 INFO [multi_dataset.py:224] Loading Aidatatang_200zh set in lazy mode
2023-10-17 18:28:53,480 INFO [multi_dataset.py:233] Loading Aishell set in lazy mode
2023-10-17 18:28:53,483 INFO [multi_dataset.py:242] Loading Aishell-2 set in lazy mode
2023-10-17 18:28:53,486 INFO [multi_dataset.py:251] Loading Aishell-4 TEST set in lazy mode
2023-10-17 18:28:53,488 INFO [multi_dataset.py:257] Loading Ali-Meeting set in lazy mode
2023-10-17 18:28:53,492 INFO [multi_dataset.py:266] Loading MagicData set in lazy mode
2023-10-17 18:28:53,495 INFO [multi_dataset.py:275] Loading KeSpeech set in lazy mode
2023-10-17 18:28:53,500 INFO [multi_dataset.py:287] Loading WeNetSpeech set in lazy mode
2023-10-17 18:28:59,880 WARNING [ctc_decode.py:683] Excluding cut with ID: TEST_NET_Y0000000004_0ub4ZzdHzBc_S00023 from decoding, num_frames: 8
2023-10-17 18:29:00,864 INFO [ctc_decode.py:695] Start decoding test set: aidatatang_test
2023-10-17 18:29:03,097 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 321
2023-10-17 18:29:55,906 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([2.2702, 2.7870, 3.1113, 2.7685], device='cuda:0')
2023-10-17 18:30:06,942 INFO [ctc_decode.py:485] batch 100/?, cuts processed until now is 35376
2023-10-17 18:30:29,144 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-aidatatang_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:30:29,895 INFO [utils.py:565] [aidatatang_test-ctc-decoding] %WER 15.26% [43139 / 282666, 7245 ins, 10406 del, 25488 sub ]
2023-10-17 18:30:31,523 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-aidatatang_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:30:31,526 INFO [ctc_decode.py:522] 
For aidatatang_test, WER of different settings are:
ctc-decoding	15.26	best for aidatatang_test

2023-10-17 18:30:31,527 INFO [ctc_decode.py:695] Start decoding test set: aidatatang_dev
2023-10-17 18:30:33,562 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 325
2023-10-17 18:31:02,140 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([4.4563, 4.2981, 3.9703, 4.6275], device='cuda:0')
2023-10-17 18:31:08,199 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([3.1070, 2.5969, 1.4553, 1.7578], device='cuda:0')
2023-10-17 18:31:08,818 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([3.7113, 3.0239, 2.1108, 2.3879], device='cuda:0')
2023-10-17 18:31:15,999 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-aidatatang_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:31:16,380 INFO [utils.py:565] [aidatatang_dev-ctc-decoding] %WER 14.57% [20711 / 142150, 3442 ins, 5435 del, 11834 sub ]
2023-10-17 18:31:17,106 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-aidatatang_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:31:17,110 INFO [ctc_decode.py:522] 
For aidatatang_dev, WER of different settings are:
ctc-decoding	14.57	best for aidatatang_dev

2023-10-17 18:31:17,110 INFO [ctc_decode.py:695] Start decoding test set: alimeeting_test
2023-10-17 18:31:19,674 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 177
2023-10-17 18:31:25,592 WARNING [ctc_decode.py:683] Excluding cut with ID: R8008_M8016-8062-123 from decoding, num_frames: 6
2023-10-17 18:31:51,823 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-alimeeting_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:31:51,924 INFO [utils.py:565] [alimeeting_test-ctc-decoding] %WER 69.70% [11401 / 16357, 1 ins, 2019 del, 9381 sub ]
2023-10-17 18:31:52,197 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-alimeeting_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:31:52,201 INFO [ctc_decode.py:522] 
For alimeeting_test, WER of different settings are:
ctc-decoding	69.7	best for alimeeting_test

2023-10-17 18:31:52,201 INFO [ctc_decode.py:695] Start decoding test set: alimeeting_eval
2023-10-17 18:31:54,746 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 140
2023-10-17 18:32:06,274 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-alimeeting_eval-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:32:06,314 INFO [utils.py:565] [alimeeting_eval-ctc-decoding] %WER 72.85% [4704 / 6457, 0 ins, 930 del, 3774 sub ]
2023-10-17 18:32:06,496 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-alimeeting_eval-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:32:06,499 INFO [ctc_decode.py:522] 
For alimeeting_eval, WER of different settings are:
ctc-decoding	72.85	best for alimeeting_eval

2023-10-17 18:32:06,499 INFO [ctc_decode.py:695] Start decoding test set: aishell_test
2023-10-17 18:32:09,216 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 190
2023-10-17 18:32:15,951 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([3.6632, 3.6327, 2.7261, 2.6312], device='cuda:0')
2023-10-17 18:32:28,588 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-aishell_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:32:28,732 INFO [utils.py:565] [aishell_test-ctc-decoding] %WER 13.76% [8867 / 64428, 866 ins, 2091 del, 5910 sub ]
2023-10-17 18:32:29,114 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-aishell_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:32:29,118 INFO [ctc_decode.py:522] 
For aishell_test, WER of different settings are:
ctc-decoding	13.76	best for aishell_test

2023-10-17 18:32:29,119 INFO [ctc_decode.py:695] Start decoding test set: aishell_dev
2023-10-17 18:32:32,157 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 215
2023-10-17 18:33:07,311 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-aishell_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:33:07,677 INFO [utils.py:565] [aishell_dev-ctc-decoding] %WER 12.87% [16438 / 127698, 1671 ins, 3693 del, 11074 sub ]
2023-10-17 18:33:08,275 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-aishell_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:33:08,279 INFO [ctc_decode.py:522] 
For aishell_dev, WER of different settings are:
ctc-decoding	12.87	best for aishell_dev

2023-10-17 18:33:08,279 INFO [ctc_decode.py:695] Start decoding test set: aishell-2_test
2023-10-17 18:33:10,740 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 334
2023-10-17 18:33:18,787 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-aishell-2_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:33:18,825 INFO [utils.py:565] [aishell-2_test-ctc-decoding] %WER 25.55% [1278 / 5002, 0 ins, 3 del, 1275 sub ]
2023-10-17 18:33:18,904 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-aishell-2_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:33:18,907 INFO [ctc_decode.py:522] 
For aishell-2_test, WER of different settings are:
ctc-decoding	25.55	best for aishell-2_test

2023-10-17 18:33:18,907 INFO [ctc_decode.py:695] Start decoding test set: aishell-2_dev
2023-10-17 18:33:20,413 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 209
2023-10-17 18:33:24,585 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-aishell-2_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:33:24,602 INFO [utils.py:565] [aishell-2_dev-ctc-decoding] %WER 23.56% [589 / 2500, 0 ins, 0 del, 589 sub ]
2023-10-17 18:33:24,638 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-aishell-2_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:33:24,642 INFO [ctc_decode.py:522] 
For aishell-2_dev, WER of different settings are:
ctc-decoding	23.56	best for aishell-2_dev

2023-10-17 18:33:24,642 INFO [ctc_decode.py:695] Start decoding test set: aishell-4
2023-10-17 18:33:28,250 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 132
2023-10-17 18:33:57,329 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-aishell-4-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:33:57,399 INFO [utils.py:565] [aishell-4-ctc-decoding] %WER 71.75% [7590 / 10579, 47 ins, 534 del, 7009 sub ]
2023-10-17 18:33:57,590 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-aishell-4-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:33:57,593 INFO [ctc_decode.py:522] 
For aishell-4, WER of different settings are:
ctc-decoding	71.75	best for aishell-4

2023-10-17 18:33:57,594 INFO [ctc_decode.py:695] Start decoding test set: magicdata_test
2023-10-17 18:34:00,583 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 230
2023-10-17 18:35:01,490 INFO [ctc_decode.py:485] batch 100/?, cuts processed until now is 23940
2023-10-17 18:35:02,615 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-magicdata_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:35:02,855 INFO [utils.py:565] [magicdata_test-ctc-decoding] %WER 19.34% [4698 / 24286, 284 ins, 7 del, 4407 sub ]
2023-10-17 18:35:03,201 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-magicdata_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:35:03,205 INFO [ctc_decode.py:522] 
For magicdata_test, WER of different settings are:
ctc-decoding	19.34	best for magicdata_test

2023-10-17 18:35:03,205 INFO [ctc_decode.py:695] Start decoding test set: magicdata_dev
2023-10-17 18:35:06,237 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 211
2023-10-17 18:35:24,163 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([2.0037, 2.2850, 2.0049, 3.6487], device='cuda:0')
2023-10-17 18:35:38,292 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-magicdata_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:35:38,366 INFO [utils.py:565] [magicdata_dev-ctc-decoding] %WER 22.35% [2653 / 11872, 22 ins, 79 del, 2552 sub ]
2023-10-17 18:35:38,552 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-magicdata_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:35:38,556 INFO [ctc_decode.py:522] 
For magicdata_dev, WER of different settings are:
ctc-decoding	22.35	best for magicdata_dev

2023-10-17 18:35:38,556 INFO [ctc_decode.py:695] Start decoding test set: kespeech-asr_test
2023-10-17 18:35:41,963 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 179
2023-10-17 18:35:48,964 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([2.2457, 2.7709, 3.0159, 2.7526, 2.8671, 2.7205, 3.0313, 2.9588],
       device='cuda:0')
2023-10-17 18:35:52,190 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([2.1268, 2.6636, 2.8915, 2.6335, 2.7311, 2.5814, 2.9027, 2.8249],
       device='cuda:0')
2023-10-17 18:36:14,760 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([5.0753, 4.3347, 4.8011, 4.4720], device='cuda:0')
2023-10-17 18:36:18,629 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([3.0111, 4.3811, 3.5554, 3.9260], device='cuda:0')
2023-10-17 18:36:19,830 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([5.0782, 4.3275, 4.7918, 4.4625], device='cuda:0')
2023-10-17 18:36:45,716 INFO [ctc_decode.py:485] batch 100/?, cuts processed until now is 19455
2023-10-17 18:36:46,846 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-kespeech-asr_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:36:46,972 INFO [utils.py:565] [kespeech-asr_test-ctc-decoding] %WER 48.71% [9608 / 19723, 0 ins, 3 del, 9605 sub ]
2023-10-17 18:36:47,307 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-kespeech-asr_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:36:47,311 INFO [ctc_decode.py:522] 
For kespeech-asr_test, WER of different settings are:
ctc-decoding	48.71	best for kespeech-asr_test

2023-10-17 18:36:47,312 INFO [ctc_decode.py:695] Start decoding test set: kespeech-asr_dev_phase1
2023-10-17 18:36:49,451 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 179
2023-10-17 18:36:56,484 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-kespeech-asr_dev_phase1-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:36:56,499 INFO [utils.py:565] [kespeech-asr_dev_phase1-ctc-decoding] %WER 42.38% [932 / 2199, 0 ins, 0 del, 932 sub ]
2023-10-17 18:36:56,534 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-kespeech-asr_dev_phase1-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:36:56,537 INFO [ctc_decode.py:522] 
For kespeech-asr_dev_phase1, WER of different settings are:
ctc-decoding	42.38	best for kespeech-asr_dev_phase1

2023-10-17 18:36:56,537 INFO [ctc_decode.py:695] Start decoding test set: kespeech-asr_dev_phase2
2023-10-17 18:36:58,442 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 175
2023-10-17 18:37:05,129 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-kespeech-asr_dev_phase2-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:37:05,145 INFO [utils.py:565] [kespeech-asr_dev_phase2-ctc-decoding] %WER 26.90% [594 / 2208, 0 ins, 0 del, 594 sub ]
2023-10-17 18:37:05,178 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-kespeech-asr_dev_phase2-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:37:05,181 INFO [ctc_decode.py:522] 
For kespeech-asr_dev_phase2, WER of different settings are:
ctc-decoding	26.9	best for kespeech-asr_dev_phase2

2023-10-17 18:37:05,182 INFO [ctc_decode.py:695] Start decoding test set: wenetspeech-meeting_test
2023-10-17 18:37:07,190 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 112
2023-10-17 18:37:33,122 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([1.8152, 2.0526, 1.9635, 2.3204, 2.4993, 2.4532, 2.2485, 1.9477],
       device='cuda:0')
2023-10-17 18:37:41,448 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-wenetspeech-meeting_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:37:41,502 INFO [utils.py:565] [wenetspeech-meeting_test-ctc-decoding] %WER 67.29% [5632 / 8370, 2 ins, 0 del, 5630 sub ]
2023-10-17 18:37:41,673 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-wenetspeech-meeting_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:37:41,676 INFO [ctc_decode.py:522] 
For wenetspeech-meeting_test, WER of different settings are:
ctc-decoding	67.29	best for wenetspeech-meeting_test

2023-10-17 18:37:41,676 INFO [ctc_decode.py:695] Start decoding test set: wenetspeech-net_test
2023-10-17 18:37:41,901 WARNING [ctc_decode.py:683] Excluding cut with ID: TEST_NET_Y0000000004_0ub4ZzdHzBc_S00023 from decoding, num_frames: 8
2023-10-17 18:37:43,917 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 175
2023-10-17 18:38:04,564 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([4.2461, 2.2640, 4.2913, 3.8631], device='cuda:0')
2023-10-17 18:38:08,841 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([3.7114, 3.6090, 3.4062, 3.7334], device='cuda:0')
2023-10-17 18:38:38,781 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-wenetspeech-net_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:38:39,042 INFO [utils.py:565] [wenetspeech-net_test-ctc-decoding] %WER 54.24% [13438 / 24773, 2 ins, 19 del, 13417 sub ]
2023-10-17 18:38:39,477 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-wenetspeech-net_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:38:39,480 INFO [ctc_decode.py:522] 
For wenetspeech-net_test, WER of different settings are:
ctc-decoding	54.24	best for wenetspeech-net_test

2023-10-17 18:38:39,481 INFO [ctc_decode.py:695] Start decoding test set: wenetspeech_dev
2023-10-17 18:38:41,576 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 157
2023-10-17 18:38:51,281 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([4.5085, 2.3355, 4.5779, 4.0812], device='cuda:0')
2023-10-17 18:39:09,310 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([2.2675, 3.6412, 3.7390, 3.8346], device='cuda:0')
2023-10-17 18:39:20,015 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([1.0711, 2.5839, 2.5771, 3.9200], device='cuda:0')
2023-10-17 18:39:24,326 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-wenetspeech_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:39:24,419 INFO [utils.py:565] [wenetspeech_dev-ctc-decoding] %WER 64.88% [8970 / 13825, 1 ins, 1 del, 8968 sub ]
2023-10-17 18:39:24,662 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-wenetspeech_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:39:24,666 INFO [ctc_decode.py:522] 
For wenetspeech_dev, WER of different settings are:
ctc-decoding	64.88	best for wenetspeech_dev

2023-10-17 18:39:24,666 INFO [ctc_decode.py:714] Done!