decoding_results/fast_beam_search/log-decode-epoch-24-avg-5-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model-2024-03-20-11-29-15

2024-03-20 11:29:15,295 INFO [decode_char.py:651] Decoding started
2024-03-20 11:29:15,296 INFO [decode_char.py:657] Device: cuda:0
2024-03-20 11:29:15,325 INFO [lexicon.py:168] Loading pre-compiled data/zh-HK/lang_char/Linv.pt
2024-03-20 11:29:15,335 INFO [decode_char.py:668] {'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': '1.24.4', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': '2989b0b1186fa6022932804f5b39fbb2781ebf42', 'k2-git-date': 'Fri Nov 24 11:34:10 2023', 'lhotse-version': '1.22.0.dev+git.d8ed1bbb.dirty', 'torch-version': '1.11.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.9', 'icefall-git-branch': 'dev/cv-zipformer', 'icefall-git-sha1': '6993183d-dirty', 'icefall-git-date': 'Fri Mar 15 19:31:35 2024', '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/k2-1.24.4.dev20231207+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-1.22.0.dev0+git.d8ed1bbb.dirty-py3.9.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-1-1207150822-75498b8c5f-sb8fn', 'IP address': '10.177.74.212'}, 'epoch': 24, 'iter': 0, 'avg': 5, 'use_averaged_model': True, 'exp_dir': PosixPath('zipformer/exp_val'), 'lang_dir': PosixPath('data/zh-HK/lang_char'), 'decoding_method': 'fast_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': 1, 'max_sym_per_frame': 1, 'num_paths': 200, 'nbest_scale': 0.5, 'blank_penalty': 0.0, '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': False, 'language': 'zh-HK', 'cv_manifest_dir': PosixPath('data/zh-HK/fbank'), 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 200.0, '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_val/fast_beam_search'), 'suffix': 'epoch-24-avg-5-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model', 'blank_id': 0, 'vocab_size': 3904}
2024-03-20 11:29:15,336 INFO [decode_char.py:670] About to create model
2024-03-20 11:29:15,935 INFO [decode_char.py:737] Calculating the averaged model over epoch range from 19 (excluded) to 24
2024-03-20 11:29:24,776 INFO [decode_char.py:768] Number of model parameters: 72526519
2024-03-20 11:29:24,776 INFO [asr_datamodule.py:430] About to get dev cuts
2024-03-20 11:29:24,791 INFO [asr_datamodule.py:353] About to create dev dataset
2024-03-20 11:29:25,180 INFO [asr_datamodule.py:370] About to create dev dataloader
2024-03-20 11:29:25,180 INFO [asr_datamodule.py:437] About to get test cuts
2024-03-20 11:29:26,899 INFO [decode_char.py:547] batch 0/?, cuts processed until now is 30
2024-03-20 11:29:35,546 INFO [decode_char.py:547] batch 20/?, cuts processed until now is 713
2024-03-20 11:29:43,958 INFO [decode_char.py:547] batch 40/?, cuts processed until now is 1430
2024-03-20 11:29:52,173 INFO [decode_char.py:547] batch 60/?, cuts processed until now is 2241
2024-03-20 11:30:01,101 INFO [decode_char.py:547] batch 80/?, cuts processed until now is 2888
2024-03-20 11:30:09,446 INFO [decode_char.py:547] batch 100/?, cuts processed until now is 3694
2024-03-20 11:30:17,836 INFO [decode_char.py:547] batch 120/?, cuts processed until now is 4399
2024-03-20 11:30:24,987 INFO [decode_char.py:547] batch 140/?, cuts processed until now is 5355
2024-03-20 11:30:28,158 INFO [decode_char.py:563] The transcripts are stored in zipformer/exp_val/fast_beam_search/recogs-dev-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64-epoch-24-avg-5-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model.txt
2024-03-20 11:30:28,279 INFO [utils.py:656] [dev-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64] %WER 1.08% [651 / 60042, 54 ins, 356 del, 241 sub ]
2024-03-20 11:30:28,511 INFO [decode_char.py:576] Wrote detailed error stats to zipformer/exp_val/fast_beam_search/errs-dev-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64-epoch-24-avg-5-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model.txt
2024-03-20 11:30:28,515 INFO [decode_char.py:592] 
For dev, WER of different settings are:
blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64	1.08	best for dev

2024-03-20 11:30:29,878 INFO [decode_char.py:547] batch 0/?, cuts processed until now is 28
2024-03-20 11:30:38,806 INFO [decode_char.py:547] batch 20/?, cuts processed until now is 656
2024-03-20 11:30:42,062 INFO [zipformer.py:1858] name=None, attn_weights_entropy = tensor([2.9603, 3.0786, 3.8276, 3.6708], device='cuda:0')
2024-03-20 11:30:47,495 INFO [decode_char.py:547] batch 40/?, cuts processed until now is 1316
2024-03-20 11:30:57,249 INFO [decode_char.py:547] batch 60/?, cuts processed until now is 2060
2024-03-20 11:31:08,044 INFO [zipformer.py:1858] name=None, attn_weights_entropy = tensor([4.5873, 2.6105, 3.8525, 4.0749], device='cuda:0')
2024-03-20 11:31:19,066 INFO [decode_char.py:547] batch 80/?, cuts processed until now is 2659
2024-03-20 11:31:39,079 INFO [zipformer.py:1858] name=None, attn_weights_entropy = tensor([2.9966, 2.9309, 3.3164, 3.2892], device='cuda:0')
2024-03-20 11:31:42,127 INFO [decode_char.py:547] batch 100/?, cuts processed until now is 3413
2024-03-20 11:32:10,900 INFO [decode_char.py:547] batch 120/?, cuts processed until now is 4088
2024-03-20 11:32:20,504 INFO [decode_char.py:547] batch 140/?, cuts processed until now is 4889
2024-03-20 11:32:23,258 INFO [zipformer.py:1858] name=None, attn_weights_entropy = tensor([3.2237, 2.6764, 2.9873, 2.5913, 2.8635, 2.0107, 2.9021, 2.6796],
       device='cuda:0')
2024-03-20 11:32:27,884 INFO [decode_char.py:547] batch 160/?, cuts processed until now is 5448
2024-03-20 11:32:28,112 INFO [decode_char.py:563] The transcripts are stored in zipformer/exp_val/fast_beam_search/recogs-test-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64-epoch-24-avg-5-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model.txt
2024-03-20 11:32:28,225 INFO [utils.py:656] [test-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64] %WER 1.27% [760 / 59775, 55 ins, 504 del, 201 sub ]
2024-03-20 11:32:28,448 INFO [decode_char.py:576] Wrote detailed error stats to zipformer/exp_val/fast_beam_search/errs-test-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64-epoch-24-avg-5-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model.txt
2024-03-20 11:32:28,589 INFO [decode_char.py:592] 
For test, WER of different settings are:
blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64	1.27	best for test

2024-03-20 11:32:28,589 INFO [decode_char.py:809] Done!