logs/modified_beam_search/log-decode-epoch-6-avg-1-chunk-16-left-context-256-modified_beam_search-beam-size-4-blank-penalty-1.5-use-averaged-model-2023-06-19-12-58-39

2023-06-19 12:58:39,385 INFO [decode.py:652] Decoding started
2023-06-19 12:58:39,386 INFO [decode.py:658] Device: cuda:0
2023-06-19 12:58:41,388 INFO [lexicon.py:168] Loading pre-compiled data/lang_char/Linv.pt
2023-06-19 12:58:41,618 INFO [decode.py:669] {'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': 'a7d0588-clean', 'icefall-git-date': 'Mon Jun 19 12:12:33 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-1-1220091118-57c4d55446-mlpzc', 'IP address': '10.177.22.19'}, 'epoch': 6, 'iter': 0, 'avg': 1, 'use_averaged_model': True, 'exp_dir': PosixPath('zipformer/exp_L_causal_context_2'), 'lang_dir': PosixPath('data/lang_char'), 'decoding_method': 'modified_beam_search', 'beam_size': 4, 'beam': 20.0, 'ngram_lm_scale': 0.01, 'ilme_scale': 0.2, 'max_contexts': 8, 'max_states': 64, 'context_size': 2, 'max_sym_per_frame': 1, 'num_paths': 200, 'nbest_scale': 0.5, 'blank_penalty': 1.5, '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': True, 'chunk_size': '16', 'left_context_frames': '256', '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_causal_context_2/modified_beam_search'), 'suffix': 'epoch-6-avg-1-chunk-16-left-context-256-modified_beam_search-beam-size-4-blank-penalty-1.5-use-averaged-model', 'blank_id': 0, 'vocab_size': 5537}
2023-06-19 12:58:41,623 INFO [decode.py:671] About to create model
2023-06-19 12:58:42,367 INFO [decode.py:738] Calculating the averaged model over epoch range from 5 (excluded) to 6
2023-06-19 12:58:59,211 INFO [decode.py:769] Number of model parameters: 76441818
2023-06-19 12:58:59,211 INFO [asr_datamodule.py:398] About to get dev cuts
2023-06-19 12:58:59,217 INFO [asr_datamodule.py:336] About to create dev dataset
2023-06-19 12:58:59,801 INFO [asr_datamodule.py:354] About to create dev dataloader
2023-06-19 12:58:59,802 INFO [asr_datamodule.py:403] About to get TEST_NET cuts
2023-06-19 12:58:59,819 INFO [asr_datamodule.py:367] About to create test dataset
2023-06-19 12:58:59,885 WARNING [decode.py:778] Exclude cut with ID TEST_NET_Y0000000004_0ub4ZzdHzBc_S00023 from decoding, num_frames : 8.
2023-06-19 12:59:00,439 INFO [asr_datamodule.py:408] About to get TEST_MEETING cuts
2023-06-19 12:59:00,444 INFO [asr_datamodule.py:367] About to create test dataset
2023-06-19 12:59:01,269 WARNING [decode.py:778] Exclude cut with ID TEST_NET_Y0000000004_0ub4ZzdHzBc_S00023 from decoding, num_frames : 8.
2023-06-19 12:59:11,649 INFO [decode.py:548] batch 0/?, cuts processed until now is 146
2023-06-19 13:00:53,180 INFO [zipformer.py:1728] name=None, attn_weights_entropy = tensor([1.1208, 2.5132, 1.3942, 2.4992], device='cuda:0')
2023-06-19 13:01:55,483 INFO [decode.py:548] batch 20/?, cuts processed until now is 4116
2023-06-19 13:04:49,590 INFO [decode.py:548] batch 40/?, cuts processed until now is 8601
2023-06-19 13:07:42,318 INFO [decode.py:548] batch 60/?, cuts processed until now is 14082
2023-06-19 13:10:39,491 INFO [decode.py:548] batch 80/?, cuts processed until now is 18750
2023-06-19 13:12:54,002 INFO [decode.py:548] batch 100/?, cuts processed until now is 24487
2023-06-19 13:12:54,161 INFO [zipformer.py:1728] name=None, attn_weights_entropy = tensor([4.2272, 2.1059, 2.4529, 4.1916], device='cuda:0')
2023-06-19 13:13:10,918 INFO [decode.py:564] The transcripts are stored in zipformer/exp_L_causal_context_2/modified_beam_search/recogs-TEST_NET-beam_size_4_blank_penalty_1.5-epoch-6-avg-1-chunk-16-left-context-256-modified_beam_search-beam-size-4-blank-penalty-1.5-use-averaged-model.txt
2023-06-19 13:13:11,466 INFO [utils.py:562] [TEST_NET-beam_size_4_blank_penalty_1.5] %WER 10.48% [43589 / 415746, 4082 ins, 8362 del, 31145 sub ]
2023-06-19 13:13:12,963 INFO [decode.py:577] Wrote detailed error stats to zipformer/exp_L_causal_context_2/modified_beam_search/errs-TEST_NET-beam_size_4_blank_penalty_1.5-epoch-6-avg-1-chunk-16-left-context-256-modified_beam_search-beam-size-4-blank-penalty-1.5-use-averaged-model.txt
2023-06-19 13:13:12,966 INFO [decode.py:593] 
For TEST_NET, WER of different settings are:
beam_size_4_blank_penalty_1.5	10.48	best for TEST_NET

2023-06-19 13:13:12,967 INFO [decode.py:817] Done!