log/log-decode-epoch-39-avg-15-context-2-max-sym-per-frame-1-2022-03-01-10-29-35

2022-03-01 10:29:35,513 INFO [decode.py:416] Decoding started
2022-03-01 10:29:35,513 INFO [decode.py:422] Device: cuda:0
2022-03-01 10:29:35,519 INFO [decode.py:431] {'feature_dim': 80, 'encoder_out_dim': 512, 'subsampling_factor': 4, 'attention_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'vgg_frontend': False, 'env_info': {'k2-version': '1.13', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f4fefe4882bc0ae59af951da3f47335d5495ef71', 'k2-git-date': 'Thu Feb 10 15:16:02 2022', 'lhotse-version': '1.0.0.dev+missing.version.file', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'multiple-datasets', 'icefall-git-sha1': 'aadd7ca-clean', 'icefall-git-date': 'Mon Feb 21 15:11:46 2022', 'icefall-path': '/ceph-fj/fangjun/open-source-2/icefall-multi-datasets', 'k2-path': '/ceph-fj/fangjun/open-source-2/k2-multi-datasets/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-fj/fangjun/open-source-2/lhotse-multi-datasets/lhotse/__init__.py', 'hostname': 'de-74279-k2-dev-8-0218134109-68b6994b68-l4qrs', 'IP address': '10.177.63.208'}, 'epoch': 39, 'avg': 15, 'exp_dir': PosixPath('transducer_stateless_multi_datasets/exp-full-2'), 'bpe_model': './data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'context_size': 2, 'max_sym_per_frame': 1, 'max_duration': 100, 'bucketing_sampler': True, 'num_buckets': 30, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'manifest_dir': PosixPath('data/fbank'), 'on_the_fly_feats': False, 'res_dir': PosixPath('transducer_stateless_multi_datasets/exp-full-2/greedy_search'), 'suffix': 'epoch-39-avg-15-context-2-max-sym-per-frame-1', 'blank_id': 0, 'vocab_size': 500}
2022-03-01 10:29:35,519 INFO [decode.py:433] About to create model
2022-03-01 10:29:36,012 INFO [decode.py:444] averaging ['transducer_stateless_multi_datasets/exp-full-2/epoch-25.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-26.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-27.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-28.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-29.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-30.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-31.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-32.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-33.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-34.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-35.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-36.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-37.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-38.pt', 'transducer_stateless_multi_datasets/exp-full-2/epoch-39.pt']
2022-03-01 10:30:20,294 INFO [decode.py:455] Number of model parameters: 84007924
2022-03-01 10:30:20,294 INFO [librispeech.py:58] About to get test-clean cuts from data/fbank/cuts_test-clean.json.gz
2022-03-01 10:30:20,427 INFO [librispeech.py:63] About to get test-other cuts from data/fbank/cuts_test-other.json.gz
2022-03-01 10:30:21,865 INFO [decode.py:341] batch 0/?, cuts processed until now is 20
2022-03-01 10:31:36,080 INFO [decode.py:341] batch 100/?, cuts processed until now is 1406
2022-03-01 10:32:59,172 INFO [decode.py:341] batch 200/?, cuts processed until now is 2563
2022-03-01 10:33:07,070 INFO [decode.py:358] The transcripts are stored in transducer_stateless_multi_datasets/exp-full-2/greedy_search/recogs-test-clean-greedy_search-epoch-39-avg-15-context-2-max-sym-per-frame-1.txt
2022-03-01 10:33:07,143 INFO [utils.py:404] [test-clean-greedy_search] %WER 2.64% [1389 / 52576, 162 ins, 111 del, 1116 sub ]
2022-03-01 10:33:07,355 INFO [decode.py:371] Wrote detailed error stats to transducer_stateless_multi_datasets/exp-full-2/greedy_search/errs-test-clean-greedy_search-epoch-39-avg-15-context-2-max-sym-per-frame-1.txt
2022-03-01 10:33:07,356 INFO [decode.py:388] 
For test-clean, WER of different settings are:
greedy_search	2.64	best for test-clean

2022-03-01 10:33:08,566 INFO [decode.py:341] batch 0/?, cuts processed until now is 23
2022-03-01 10:34:33,508 INFO [decode.py:341] batch 100/?, cuts processed until now is 1614
2022-03-01 10:35:55,650 INFO [decode.py:341] batch 200/?, cuts processed until now is 2899
2022-03-01 10:36:00,523 INFO [decode.py:358] The transcripts are stored in transducer_stateless_multi_datasets/exp-full-2/greedy_search/recogs-test-other-greedy_search-epoch-39-avg-15-context-2-max-sym-per-frame-1.txt
2022-03-01 10:36:00,599 INFO [utils.py:404] [test-other-greedy_search] %WER 6.55% [3431 / 52343, 360 ins, 321 del, 2750 sub ]
2022-03-01 10:36:00,819 INFO [decode.py:371] Wrote detailed error stats to transducer_stateless_multi_datasets/exp-full-2/greedy_search/errs-test-other-greedy_search-epoch-39-avg-15-context-2-max-sym-per-frame-1.txt
2022-03-01 10:36:00,819 INFO [decode.py:388] 
For test-other, WER of different settings are:
greedy_search	6.55	best for test-other

2022-03-01 10:36:00,820 INFO [decode.py:483] Done!