logs/modified_beam_search/log-decode-epoch-5-avg-1-modified_beam_search-beam-size-4-blank-penalty-2.0-use-averaged-model-2023-06-15-11-10-35

2023-06-15 11:10:35,954 INFO [decode.py:685] Decoding started
2023-06-15 11:10:35,955 INFO [decode.py:691] Device: cuda:0
2023-06-15 11:10:37,605 INFO [lexicon.py:168] Loading pre-compiled data/lang_char/Linv.pt
2023-06-15 11:10:37,849 INFO [decode.py:702] {'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': '1.24.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'c51a0b9684442a88ee37f3ce0af686a04b66855b', 'k2-git-date': 'Mon May 1 21:38:03 2023', 'lhotse-version': '1.14.0.dev+git.0f812851.dirty', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'zipformer_wenetspeech', 'icefall-git-sha1': '28d3f6d-dirty', 'icefall-git-date': 'Thu Jun 15 10:30:34 2023', 'icefall-path': '/star-kw/kangwei/code/icefall_wenetspeech', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.14.0.dev0+git.0f812851.dirty-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-10-0221105906-5745685d6b-t8zzx', 'IP address': '10.177.57.19'}, 'epoch': 5, 'iter': 0, 'avg': 1, 'use_averaged_model': True, 'exp_dir': PosixPath('zipformer/exp_L_context_2'), 'lang_dir': PosixPath('data/lang_char'), 'decoding_method': 'modified_beam_search', 'beam_size': 4, 'beam': 20.0, 'ngram_lm_scale': 0.01, 'max_contexts': 8, 'max_states': 64, 'context_size': 2, 'max_sym_per_frame': 1, 'num_paths': 200, 'nbest_scale': 0.5, 'blank_penalty': 2.0, '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', 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'training_subset': 'L', 'res_dir': PosixPath('zipformer/exp_L_context_2/modified_beam_search'), 'suffix': 'epoch-5-avg-1-modified_beam_search-beam-size-4-blank-penalty-2.0-use-averaged-model', 'blank_id': 0, 'vocab_size': 5537}
2023-06-15 11:10:37,850 INFO [decode.py:704] About to create model
2023-06-15 11:10:38,526 INFO [decode.py:771] Calculating the averaged model over epoch range from 4 (excluded) to 5
2023-06-15 11:10:49,584 INFO [decode.py:802] Number of model parameters: 75879898
2023-06-15 11:10:49,585 INFO [asr_datamodule.py:398] About to get dev cuts
2023-06-15 11:10:49,600 INFO [asr_datamodule.py:336] About to create dev dataset
2023-06-15 11:10:50,155 INFO [asr_datamodule.py:354] About to create dev dataloader
2023-06-15 11:10:50,156 INFO [asr_datamodule.py:403] About to get TEST_NET cuts
2023-06-15 11:10:50,158 INFO [asr_datamodule.py:367] About to create test dataset
2023-06-15 11:10:50,208 WARNING [decode.py:811] Exclude cut with ID TEST_NET_Y0000000004_0ub4ZzdHzBc_S00023 from decoding, num_frames : 8.
2023-06-15 11:10:50,725 INFO [asr_datamodule.py:408] About to get TEST_MEETING cuts
2023-06-15 11:10:50,733 INFO [asr_datamodule.py:367] About to create test dataset
2023-06-15 11:11:00,637 INFO [decode.py:581] batch 0/?, cuts processed until now is 130
2023-06-15 11:13:34,377 INFO [decode.py:581] batch 20/?, cuts processed until now is 3192
2023-06-15 11:14:04,088 INFO [zipformer.py:1728] name=None, attn_weights_entropy = tensor([1.3729, 2.8987, 3.9559, 4.0580], device='cuda:0')
2023-06-15 11:16:06,816 INFO [decode.py:581] batch 40/?, cuts processed until now is 6421
2023-06-15 11:18:39,438 INFO [decode.py:581] batch 60/?, cuts processed until now is 10176
2023-06-15 11:20:39,081 INFO [decode.py:581] batch 80/?, cuts processed until now is 13727
2023-06-15 11:20:46,228 INFO [decode.py:597] The transcripts are stored in zipformer/exp_L_context_2/modified_beam_search/recogs-DEV-beam_size_4_blank_penalty_2.0-epoch-5-avg-1-modified_beam_search-beam-size-4-blank-penalty-2.0-use-averaged-model.txt
2023-06-15 11:20:46,604 INFO [utils.py:562] [DEV-beam_size_4_blank_penalty_2.0] %WER 7.68% [25394 / 330498, 2700 ins, 9917 del, 12777 sub ]
2023-06-15 11:20:47,617 INFO [decode.py:610] Wrote detailed error stats to zipformer/exp_L_context_2/modified_beam_search/errs-DEV-beam_size_4_blank_penalty_2.0-epoch-5-avg-1-modified_beam_search-beam-size-4-blank-penalty-2.0-use-averaged-model.txt
2023-06-15 11:20:47,621 INFO [decode.py:626] 
For DEV, WER of different settings are:
beam_size_4_blank_penalty_2.0	7.68	best for DEV

2023-06-15 11:20:47,911 WARNING [decode.py:811] Exclude cut with ID TEST_NET_Y0000000004_0ub4ZzdHzBc_S00023 from decoding, num_frames : 8.
2023-06-15 11:20:56,952 INFO [decode.py:581] batch 0/?, cuts processed until now is 146
2023-06-15 11:23:18,668 INFO [decode.py:581] batch 20/?, cuts processed until now is 4116
2023-06-15 11:24:17,463 INFO [zipformer.py:1728] name=None, attn_weights_entropy = tensor([1.2145, 3.2727, 2.1439, 3.3618], device='cuda:0')
2023-06-15 11:25:46,620 INFO [decode.py:581] batch 40/?, cuts processed until now is 8601
2023-06-15 11:28:12,903 INFO [decode.py:581] batch 60/?, cuts processed until now is 14082
2023-06-15 11:30:36,727 INFO [decode.py:581] batch 80/?, cuts processed until now is 18750
2023-06-15 11:31:22,516 INFO [zipformer.py:1728] name=None, attn_weights_entropy = tensor([1.2437, 1.5504, 1.1034, 1.5311, 1.6075, 1.5702, 1.5321, 1.5561],
       device='cuda:0')
2023-06-15 11:31:28,577 INFO [zipformer.py:1728] name=None, attn_weights_entropy = tensor([1.5481, 1.7845, 1.9991, 2.0549], device='cuda:0')
2023-06-15 11:32:23,606 INFO [decode.py:581] batch 100/?, cuts processed until now is 24487
2023-06-15 11:32:38,675 INFO [decode.py:597] The transcripts are stored in zipformer/exp_L_context_2/modified_beam_search/recogs-TEST_NET-beam_size_4_blank_penalty_2.0-epoch-5-avg-1-modified_beam_search-beam-size-4-blank-penalty-2.0-use-averaged-model.txt
2023-06-15 11:32:39,182 INFO [utils.py:562] [TEST_NET-beam_size_4_blank_penalty_2.0] %WER 8.41% [34955 / 415746, 3767 ins, 7849 del, 23339 sub ]
2023-06-15 11:32:40,522 INFO [decode.py:610] Wrote detailed error stats to zipformer/exp_L_context_2/modified_beam_search/errs-TEST_NET-beam_size_4_blank_penalty_2.0-epoch-5-avg-1-modified_beam_search-beam-size-4-blank-penalty-2.0-use-averaged-model.txt
2023-06-15 11:32:40,526 INFO [decode.py:626] 
For TEST_NET, WER of different settings are:
beam_size_4_blank_penalty_2.0	8.41	best for TEST_NET

2023-06-15 11:32:49,972 INFO [decode.py:581] batch 0/?, cuts processed until now is 93
2023-06-15 11:34:04,541 INFO [zipformer.py:1728] name=None, attn_weights_entropy = tensor([2.2329, 2.8619, 2.2066, 2.7164, 3.0813, 2.0707, 3.0810, 2.7717],
       device='cuda:0')
2023-06-15 11:35:18,846 INFO [decode.py:581] batch 20/?, cuts processed until now is 2345
2023-06-15 11:37:08,297 INFO [zipformer.py:1728] name=None, attn_weights_entropy = tensor([2.6514, 3.2670, 3.0073, 2.9856, 3.1906, 3.0949, 3.2590, 3.3916],
       device='cuda:0')
2023-06-15 11:37:55,923 INFO [decode.py:581] batch 40/?, cuts processed until now is 4929
2023-06-15 11:39:57,394 INFO [decode.py:581] batch 60/?, cuts processed until now is 7955
2023-06-15 11:40:27,194 INFO [decode.py:597] The transcripts are stored in zipformer/exp_L_context_2/modified_beam_search/recogs-TEST_MEETING-beam_size_4_blank_penalty_2.0-epoch-5-avg-1-modified_beam_search-beam-size-4-blank-penalty-2.0-use-averaged-model.txt
2023-06-15 11:40:27,444 INFO [utils.py:562] [TEST_MEETING-beam_size_4_blank_penalty_2.0] %WER 13.27% [29256 / 220385, 3007 ins, 12590 del, 13659 sub ]
2023-06-15 11:40:28,131 INFO [decode.py:610] Wrote detailed error stats to zipformer/exp_L_context_2/modified_beam_search/errs-TEST_MEETING-beam_size_4_blank_penalty_2.0-epoch-5-avg-1-modified_beam_search-beam-size-4-blank-penalty-2.0-use-averaged-model.txt
2023-06-15 11:40:28,134 INFO [decode.py:626] 
For TEST_MEETING, WER of different settings are:
beam_size_4_blank_penalty_2.0	13.27	best for TEST_MEETING

2023-06-15 11:40:28,135 INFO [decode.py:849] Done!