log/greedy_search/log-decode-epoch-29-avg-19-context-2-max-sym-per-frame-1-2022-05-17-09-48-50

2022-05-17 09:48:50,718 INFO [decode.py:472] Decoding started
2022-05-17 09:48:50,719 INFO [decode.py:478] Device: cuda:0
2022-05-17 09:48:50,791 INFO [lexicon.py:176] Loading pre-compiled data/lang_char/Linv.pt
2022-05-17 09:48:50,801 INFO [decode.py:484] {'best_train_loss': inf, 'best_valid_loss': inf, 'best_train_epoch': -1, 'best_valid_epoch': -1, 'batch_idx_train': 10, 'log_interval': 1, 'reset_interval': 200, 'valid_interval': 400, 'feature_dim': 80, 'subsampling_factor': 4, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 200, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.2.0.dev+git.de75634.clean', 'torch-version': '1.11.0', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'pruned-rnnt2-for-aidatatang-zh', 'icefall-git-sha1': '822fe21-clean', 'icefall-git-date': 'Tue May 17 09:38:25 2022', 'icefall-path': '/ceph-meixu/luomingshuang/icefall', 'k2-path': '/ceph-ms/luomingshuang/k2_latest/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-meixu/luomingshuang/anaconda3/envs/k2-python/lib/python3.8/site-packages/lhotse-1.2.0.dev0+git.de75634.clean-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-1-0307195509-54c966b95f-rtpfq', 'IP address': '10.177.22.9'}, 'epoch': 29, 'batch': None, 'avg': 19, 'avg_last_n': 0, 'exp_dir': PosixPath('pruned_transducer_stateless2/exp'), 'lang_dir': 'data/lang_char', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 600, 'bucketing_sampler': True, 'num_buckets': 300, '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, 'res_dir': PosixPath('pruned_transducer_stateless2/exp/greedy_search'), 'suffix': 'epoch-29-avg-19-context-2-max-sym-per-frame-1', 'blank_id': 0, 'vocab_size': 3945}
2022-05-17 09:48:50,802 INFO [decode.py:486] About to create model
2022-05-17 09:48:51,281 INFO [decode.py:507] averaging ['pruned_transducer_stateless2/exp/epoch-11.pt', 'pruned_transducer_stateless2/exp/epoch-12.pt', 'pruned_transducer_stateless2/exp/epoch-13.pt', 'pruned_transducer_stateless2/exp/epoch-14.pt', 'pruned_transducer_stateless2/exp/epoch-15.pt', 'pruned_transducer_stateless2/exp/epoch-16.pt', 'pruned_transducer_stateless2/exp/epoch-17.pt', 'pruned_transducer_stateless2/exp/epoch-18.pt', 'pruned_transducer_stateless2/exp/epoch-19.pt', 'pruned_transducer_stateless2/exp/epoch-20.pt', 'pruned_transducer_stateless2/exp/epoch-21.pt', 'pruned_transducer_stateless2/exp/epoch-22.pt', 'pruned_transducer_stateless2/exp/epoch-23.pt', 'pruned_transducer_stateless2/exp/epoch-24.pt', 'pruned_transducer_stateless2/exp/epoch-25.pt', 'pruned_transducer_stateless2/exp/epoch-26.pt', 'pruned_transducer_stateless2/exp/epoch-27.pt', 'pruned_transducer_stateless2/exp/epoch-28.pt', 'pruned_transducer_stateless2/exp/epoch-29.pt']
2022-05-17 09:49:10,812 INFO [decode.py:521] Number of model parameters: 85713735
2022-05-17 09:49:10,815 INFO [asr_datamodule.py:340] About to create dev dataset
2022-05-17 09:49:13,101 INFO [asr_datamodule.py:361] About to create dev dataloader
2022-05-17 09:49:19,714 INFO [decode.py:392] batch 0/?, cuts processed until now is 173
2022-05-17 09:50:07,609 INFO [decode.py:392] batch 100/?, cuts processed until now is 19629
2022-05-17 09:50:18,025 INFO [decode.py:409] The transcripts are stored in pruned_transducer_stateless2/exp/greedy_search/recogs-dev-greedy_search-epoch-29-avg-19-context-2-max-sym-per-frame-1.txt
2022-05-17 09:50:18,310 INFO [utils.py:406] [dev-greedy_search] %WER 5.53% [12975 / 234524, 1170 ins, 1487 del, 10318 sub ]
2022-05-17 09:50:19,057 INFO [decode.py:422] Wrote detailed error stats to pruned_transducer_stateless2/exp/greedy_search/errs-dev-greedy_search-epoch-29-avg-19-context-2-max-sym-per-frame-1.txt
2022-05-17 09:50:19,057 INFO [decode.py:439] 
For dev, WER of different settings are:
greedy_search	5.53	best for dev

2022-05-17 09:50:23,412 INFO [decode.py:392] batch 0/?, cuts processed until now is 173
2022-05-17 09:51:11,589 INFO [decode.py:392] batch 100/?, cuts processed until now is 19296
2022-05-17 09:52:00,302 INFO [decode.py:392] batch 200/?, cuts processed until now is 39188
2022-05-17 09:52:21,033 INFO [decode.py:409] The transcripts are stored in pruned_transducer_stateless2/exp/greedy_search/recogs-test-greedy_search-epoch-29-avg-19-context-2-max-sym-per-frame-1.txt
2022-05-17 09:52:21,653 INFO [utils.py:406] [test-greedy_search] %WER 6.59% [30905 / 468933, 2533 ins, 3290 del, 25082 sub ]
2022-05-17 09:52:23,247 INFO [decode.py:422] Wrote detailed error stats to pruned_transducer_stateless2/exp/greedy_search/errs-test-greedy_search-epoch-29-avg-19-context-2-max-sym-per-frame-1.txt
2022-05-17 09:52:23,248 INFO [decode.py:439] 
For test, WER of different settings are:
greedy_search	6.59	best for test

2022-05-17 09:52:23,248 INFO [decode.py:596] Done!