decoding_results/ctc-decoding/log-decode-epoch-20-avg-1-use-averaged-model-2023-10-17-18-23-09

2023-10-17 18:23:09,112 INFO [ctc_decode.py:560] Decoding started
2023-10-17 18:23:09,112 INFO [ctc_decode.py:566] Device: cuda:0
2023-10-17 18:23:09,112 INFO [ctc_decode.py:567] {'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': '1.24.3', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': '821ebc378e7fb99b8adc81950227963332821e01', 'k2-git-date': 'Wed Jul 19 15:38:25 2023', 'lhotse-version': '1.16.0.dev+git.1db4d97a.clean', 'torch-version': '1.11.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.9', 'icefall-git-branch': 'dev_multi_zh-hans', 'icefall-git-sha1': '919793d3-dirty', 'icefall-git-date': 'Thu Sep 7 21:06:37 2023', '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/k2-1.24.3.dev20230721+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-1.16.0.dev0+git.1db4d97a.clean-py3.9.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-1-1220091118-57c4d55446-mvd6x', 'IP address': '10.177.22.19'}, 'frame_shift_ms': 10, 'search_beam': 20, 'output_beam': 8, 'min_active_states': 30, 'max_active_states': 10000, 'use_double_scores': True, 'epoch': 20, 'iter': 0, 'avg': 1, 'use_averaged_model': True, 'exp_dir': PosixPath('zipformer/exp-w-ctc'), 'bpe_model': 'data/lang_bpe_2000/bpe.model', 'lang_dir': PosixPath('data/lang_bpe_2000'), 'context_size': 2, 'decoding_method': 'ctc-decoding', 'num_paths': 100, 'nbest_scale': 1.0, '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': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 300.0, '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-w-ctc/ctc-decoding'), 'suffix': 'epoch-20-avg-1-use-averaged-model'}
2023-10-17 18:23:10,868 INFO [lexicon.py:168] Loading pre-compiled data/lang_bpe_2000/Linv.pt
2023-10-17 18:23:16,848 INFO [ctc_decode.py:587] About to create model
2023-10-17 18:23:17,431 INFO [ctc_decode.py:654] Calculating the averaged model over epoch range from 19 (excluded) to 20
2023-10-17 18:23:22,164 INFO [ctc_decode.py:671] Number of model parameters: 69651511
2023-10-17 18:23:22,164 INFO [multi_dataset.py:221] About to get multidataset test cuts
2023-10-17 18:23:22,165 INFO [multi_dataset.py:224] Loading Aidatatang_200zh set in lazy mode
2023-10-17 18:23:22,182 INFO [multi_dataset.py:233] Loading Aishell set in lazy mode
2023-10-17 18:23:22,186 INFO [multi_dataset.py:242] Loading Aishell-2 set in lazy mode
2023-10-17 18:23:22,189 INFO [multi_dataset.py:251] Loading Aishell-4 TEST set in lazy mode
2023-10-17 18:23:22,191 INFO [multi_dataset.py:257] Loading Ali-Meeting set in lazy mode
2023-10-17 18:23:22,194 INFO [multi_dataset.py:266] Loading MagicData set in lazy mode
2023-10-17 18:23:22,197 INFO [multi_dataset.py:275] Loading KeSpeech set in lazy mode
2023-10-17 18:23:22,202 INFO [multi_dataset.py:287] Loading WeNetSpeech set in lazy mode
2023-10-17 18:23:28,526 WARNING [ctc_decode.py:683] Excluding cut with ID: TEST_NET_Y0000000004_0ub4ZzdHzBc_S00023 from decoding, num_frames: 8
2023-10-17 18:23:29,552 INFO [ctc_decode.py:695] Start decoding test set: aidatatang_test
2023-10-17 18:23:30,937 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 80
2023-10-17 18:23:31,149 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([4.3243, 4.1785, 3.8951, 4.4938], device='cuda:0')
2023-10-17 18:23:32,075 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([4.6184, 3.9753, 4.4056, 4.1082], device='cuda:0')
2023-10-17 18:23:46,072 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([1.7554, 1.8501, 1.9709, 1.4977], device='cuda:0')
2023-10-17 18:23:47,852 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([2.0318, 2.9265, 2.8671, 2.9708], device='cuda:0')
2023-10-17 18:23:49,153 INFO [ctc_decode.py:485] batch 100/?, cuts processed until now is 9084
2023-10-17 18:23:53,157 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([1.7971, 2.1446, 2.3575, 2.1475, 2.1401, 2.0719, 2.1833, 2.1629],
       device='cuda:0')
2023-10-17 18:24:07,354 INFO [ctc_decode.py:485] batch 200/?, cuts processed until now is 18516
2023-10-17 18:24:25,501 INFO [ctc_decode.py:485] batch 300/?, cuts processed until now is 28179
2023-10-17 18:24:28,319 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([3.8385, 2.2905, 3.9372, 2.3478], device='cuda:0')
2023-10-17 18:24:41,962 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([5.3786, 4.9526, 5.2852, 4.9957], device='cuda:0')
2023-10-17 18:24:43,488 INFO [ctc_decode.py:485] batch 400/?, cuts processed until now is 37667
2023-10-17 18:24:48,276 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([3.4047, 2.8120, 1.7099, 2.0791], device='cuda:0')
2023-10-17 18:24:48,972 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([3.6806, 3.0397, 2.2822, 2.4676], device='cuda:0')
2023-10-17 18:25:00,399 INFO [ctc_decode.py:485] batch 500/?, cuts processed until now is 46172
2023-10-17 18:25:06,922 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-aidatatang_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:25:07,811 INFO [utils.py:565] [aidatatang_test-ctc-decoding] %WER 15.26% [43137 / 282666, 7255 ins, 10411 del, 25471 sub ]
2023-10-17 18:25:09,250 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-aidatatang_test-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:25:09,253 INFO [ctc_decode.py:522] 
For aidatatang_test, WER of different settings are:
ctc-decoding	15.26	best for aidatatang_test

2023-10-17 18:25:09,254 INFO [ctc_decode.py:695] Start decoding test set: aidatatang_dev
2023-10-17 18:25:10,768 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 81
2023-10-17 18:25:21,290 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([1.6683, 3.5095, 3.2214, 3.0099], device='cuda:0')
2023-10-17 18:25:28,677 INFO [ctc_decode.py:485] batch 100/?, cuts processed until now is 9077
2023-10-17 18:25:45,744 INFO [zipformer.py:1853] name=None, attn_weights_entropy = tensor([5.3476, 5.0039, 5.2630, 5.0075], device='cuda:0')
2023-10-17 18:25:46,062 INFO [ctc_decode.py:485] batch 200/?, cuts processed until now is 18432
2023-10-17 18:25:58,007 INFO [ctc_decode.py:499] The transcripts are stored in zipformer/exp-w-ctc/ctc-decoding/recogs-aidatatang_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:25:58,413 INFO [utils.py:565] [aidatatang_dev-ctc-decoding] %WER 14.58% [20723 / 142150, 3444 ins, 5445 del, 11834 sub ]
2023-10-17 18:25:59,266 INFO [ctc_decode.py:508] Wrote detailed error stats to zipformer/exp-w-ctc/ctc-decoding/errs-aidatatang_dev-epoch-20-avg-1-use-averaged-model.txt
2023-10-17 18:25:59,276 INFO [ctc_decode.py:522] 
For aidatatang_dev, WER of different settings are:
ctc-decoding	14.58	best for aidatatang_dev

2023-10-17 18:25:59,277 INFO [ctc_decode.py:695] Start decoding test set: alimeeting_test
2023-10-17 18:26:01,217 INFO [ctc_decode.py:485] batch 0/?, cuts processed until now is 44
2023-10-17 18:26:10,131 WARNING [ctc_decode.py:683] Excluding cut with ID: R8008_M8016-8062-123 from decoding, num_frames: 6