2024-03-11 10:30:57,383 INFO [decode.py:651] Decoding started
2024-03-11 10:30:57,383 INFO [decode.py:657] Device: cuda:0
2024-03-11 10:30:57,477 INFO [lexicon.py:168] Loading pre-compiled data/lang_char/Linv.pt
2024-03-11 10:30:57,499 INFO [decode.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/mdcc', 'icefall-git-sha1': '78b39d9a-clean', 'icefall-git-date': 'Mon Mar 11 09:38:42 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-9-0208143539-7dbf569d4f-r7nrb', 'IP address': '10.177.13.150'}, 'epoch': 45, 'iter': 0, 'avg': 35, 'use_averaged_model': True, 'exp_dir': PosixPath('zipformer/exp'), 'lang_dir': PosixPath('data/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', '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, 'res_dir': PosixPath('zipformer/exp/fast_beam_search'), 'suffix': 'epoch-45-avg-35-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model', 'blank_id': 0, 'vocab_size': 4852}
2024-03-11 10:30:57,499 INFO [decode.py:670] About to create model
2024-03-11 10:30:58,449 INFO [decode.py:737] Calculating the averaged model over epoch range from 10 (excluded) to 45
2024-03-11 10:31:20,689 INFO [decode.py:768] Number of model parameters: 74470867
2024-03-11 10:31:20,690 INFO [asr_datamodule.py:376] About to get valid cuts
2024-03-11 10:31:20,711 INFO [asr_datamodule.py:315] About to create dev dataset
2024-03-11 10:31:21,299 INFO [asr_datamodule.py:332] About to create dev dataloader
2024-03-11 10:31:21,300 INFO [asr_datamodule.py:381] About to get test cuts
2024-03-11 10:31:21,303 INFO [asr_datamodule.py:344] About to create test dataset
2024-03-11 10:31:24,990 INFO [decode.py:547] batch 0/?, cuts processed until now is 39
2024-03-11 10:31:47,106 INFO [decode.py:547] batch 20/?, cuts processed until now is 945
2024-03-11 10:32:04,981 INFO [decode.py:547] batch 40/?, cuts processed until now is 1923
2024-03-11 10:32:22,808 INFO [decode.py:547] batch 60/?, cuts processed until now is 3088
2024-03-11 10:32:42,782 INFO [zipformer.py:1858] name=None, attn_weights_entropy = tensor([2.1404, 4.6001, 4.5984, 4.1254], device='cuda:0')
2024-03-11 10:32:44,343 INFO [decode.py:547] batch 80/?, cuts processed until now is 4031
2024-03-11 10:33:05,258 INFO [decode.py:547] batch 100/?, cuts processed until now is 5361
2024-03-11 10:33:11,644 INFO [decode.py:563] The transcripts are stored in zipformer/exp/fast_beam_search/recogs-valid-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64-epoch-45-avg-35-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model.txt
2024-03-11 10:33:11,854 INFO [utils.py:656] [valid-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64] %WER 7.28% [4990 / 68548, 88 ins, 1285 del, 3617 sub ]
2024-03-11 10:33:12,323 INFO [decode.py:576] Wrote detailed error stats to zipformer/exp/fast_beam_search/errs-valid-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64-epoch-45-avg-35-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model.txt
2024-03-11 10:33:12,327 INFO [decode.py:592] 
For valid, WER of different settings are:
blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64	7.28	best for valid

2024-03-11 10:33:15,087 INFO [decode.py:547] batch 0/?, cuts processed until now is 39
2024-03-11 10:33:36,868 INFO [decode.py:547] batch 20/?, cuts processed until now is 958
2024-03-11 10:33:56,943 INFO [decode.py:547] batch 40/?, cuts processed until now is 1954
2024-03-11 10:34:16,230 INFO [decode.py:547] batch 60/?, cuts processed until now is 3138
2024-03-11 10:34:38,458 INFO [decode.py:547] batch 80/?, cuts processed until now is 3977
2024-03-11 10:34:42,374 INFO [zipformer.py:1858] name=None, attn_weights_entropy = tensor([0.8906, 2.5631, 2.7992, 2.7330], device='cuda:0')
2024-03-11 10:34:58,362 INFO [decode.py:547] batch 100/?, cuts processed until now is 5175
2024-03-11 10:35:21,242 INFO [decode.py:547] batch 120/?, cuts processed until now is 6110
2024-03-11 10:35:38,530 INFO [zipformer.py:1858] name=None, attn_weights_entropy = tensor([1.5033, 2.0741, 2.1039, 2.0672, 2.0715, 2.0543, 2.1015, 2.0753],
       device='cuda:0')
2024-03-11 10:35:41,074 INFO [decode.py:547] batch 140/?, cuts processed until now is 7375
2024-03-11 10:35:52,065 INFO [zipformer.py:1858] name=None, attn_weights_entropy = tensor([1.6016, 2.2394, 2.1761, 2.1755, 2.1933, 2.1462, 2.2294, 2.2214],
       device='cuda:0')
2024-03-11 10:35:59,419 INFO [decode.py:547] batch 160/?, cuts processed until now is 8663
2024-03-11 10:36:18,918 INFO [decode.py:547] batch 180/?, cuts processed until now is 9787
2024-03-11 10:36:41,226 INFO [decode.py:547] batch 200/?, cuts processed until now is 10707
2024-03-11 10:37:03,121 INFO [decode.py:547] batch 220/?, cuts processed until now is 11738
2024-03-11 10:37:11,984 INFO [decode.py:563] The transcripts are stored in zipformer/exp/fast_beam_search/recogs-test-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64-epoch-45-avg-35-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model.txt
2024-03-11 10:37:12,458 INFO [utils.py:656] [test-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64] %WER 7.22% [10796 / 149459, 164 ins, 2988 del, 7644 sub ]
2024-03-11 10:37:13,469 INFO [decode.py:576] Wrote detailed error stats to zipformer/exp/fast_beam_search/errs-test-blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64-epoch-45-avg-35-beam-20.0-max-contexts-8-max-states-64-blank-penalty-0.0-use-averaged-model.txt
2024-03-11 10:37:13,475 INFO [decode.py:592] 
For test, WER of different settings are:
blank_penalty_0.0_beam_20.0_max_contexts_8_max_states_64	7.22	best for test

2024-03-11 10:37:13,476 INFO [decode.py:809] Done!