modified_beam_search/log-decode-epoch-30-avg-13-modified_beam_search-beam-size-4-use-averaged-model-2023-11-30-12-21-53

2023-11-30 12:21:53,914 INFO [decode.py:836] Decoding started
2023-11-30 12:21:53,914 INFO [decode.py:842] Device: cuda:0
2023-11-30 12:21:53,936 INFO [decode.py:852] {'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': '2b2ac14b326d61d79d04e53fbd69b1ff6d630411', 'k2-git-date': 'Thu Aug 24 05:58:26 2023', 'lhotse-version': '1.16.0', 'torch-version': '2.0.1+cu117', 'torch-cuda-available': True, 'torch-cuda-version': '11.7', 'python-version': '3.1', 'icefall-git-branch': 'multi_KD', 'icefall-git-sha1': '02ab7e51-dirty', 'icefall-git-date': 'Tue Nov 28 15:55:57 2023', 'icefall-path': '/star-data/xiaoyu/icefall_multi_KD', 'k2-path': '/star-xy/softwares/k2_development/k2/k2/python/k2/__init__.py', 'lhotse-path': '/star-xy/softwares/anaconda3/envs/multi_KD/lib/python3.10/site-packages/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-1-1220091118-57c4d55446-mvd6x', 'IP address': '10.177.22.19'}, 'epoch': 30, 'iter': 0, 'avg': 13, 'use_averaged_model': True, 'exp_dir': PosixPath('multi_KD/exp_train_asr_full_libri1_do_audio_tagging1_as_unbalanced_scale1.0'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'lang_dir': PosixPath('data/lang_bpe_500'), 'decoding_method': 'modified_beam_search', 'beam_size': 4, 'beam': 20.0, 'ngram_lm_scale': 0.01, 'max_contexts': 8, 'max_states': 64, 'context_size': 2, 'max_sym_per_frame': 1, 'num_paths': 200, 'nbest_scale': 0.5, 'use_shallow_fusion': False, 'lm_type': 'rnn', 'lm_scale': 0.3, 'tokens_ngram': 2, 'backoff_id': 500, 'context_score': 2, 'context_file': '', '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, 'use_encoder_projection': False, 'encoder_projection_dim': 2560, 'freeze_encoder': False, 'freezing_encoder_layer_index': '-1', 'freeze_encoder_steps': -1, 'encoder_lr_scale': 1.0, 'full_libri': True, 'mini_libri': False, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 500, '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', 'lm_vocab_size': 500, 'lm_epoch': 7, 'lm_avg': 1, 'lm_exp_dir': None, 'rnn_lm_embedding_dim': 2048, 'rnn_lm_hidden_dim': 2048, 'rnn_lm_num_layers': 3, 'rnn_lm_tie_weights': True, 'transformer_lm_exp_dir': None, 'transformer_lm_dim_feedforward': 2048, 'transformer_lm_encoder_dim': 768, 'transformer_lm_embedding_dim': 768, 'transformer_lm_nhead': 8, 'transformer_lm_num_layers': 16, 'transformer_lm_tie_weights': True, 'res_dir': PosixPath('multi_KD/exp_train_asr_full_libri1_do_audio_tagging1_as_unbalanced_scale1.0/modified_beam_search'), 'has_contexts': False, 'suffix': 'epoch-30-avg-13-modified_beam_search-beam-size-4-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
2023-11-30 12:21:53,937 INFO [decode.py:854] About to create model
2023-11-30 12:21:54,728 INFO [decode.py:921] Calculating the averaged model over epoch range from 17 (excluded) to 30
2023-11-30 12:22:02,085 INFO [decode.py:1016] Number of model parameters: 65549011
2023-11-30 12:22:02,086 INFO [asr_datamodule.py:465] About to get test-clean cuts
2023-11-30 12:22:02,095 INFO [asr_datamodule.py:472] About to get test-other cuts
2023-11-30 12:22:10,122 INFO [decode.py:709] batch 0/?, cuts processed until now is 36
2023-11-30 12:23:35,975 INFO [decode.py:709] batch 20/?, cuts processed until now is 1037
2023-11-30 12:24:56,225 INFO [decode.py:709] batch 40/?, cuts processed until now is 2299
2023-11-30 12:25:19,220 INFO [decode.py:725] The transcripts are stored in multi_KD/exp_train_asr_full_libri1_do_audio_tagging1_as_unbalanced_scale1.0/modified_beam_search/recogs-test-clean-beam_size_4-epoch-30-avg-13-modified_beam_search-beam-size-4-use-averaged-model.txt
2023-11-30 12:25:19,324 INFO [utils.py:641] [test-clean-beam_size_4] %WER 2.53% [1329 / 52576, 149 ins, 96 del, 1084 sub ]
2023-11-30 12:25:19,535 INFO [decode.py:738] Wrote detailed error stats to multi_KD/exp_train_asr_full_libri1_do_audio_tagging1_as_unbalanced_scale1.0/modified_beam_search/errs-test-clean-beam_size_4-epoch-30-avg-13-modified_beam_search-beam-size-4-use-averaged-model.txt
2023-11-30 12:25:19,538 INFO [decode.py:754] 
For test-clean, WER of different settings are:
beam_size_4	2.53	best for test-clean

2023-11-30 12:25:24,921 INFO [decode.py:709] batch 0/?, cuts processed until now is 43
2023-11-30 12:26:45,319 INFO [decode.py:709] batch 20/?, cuts processed until now is 1195
2023-11-30 12:26:49,352 INFO [zipformer.py:1877] name=None, attn_weights_entropy = tensor([4.9405, 4.7531, 4.4783, 4.5459], device='cuda:0')
2023-11-30 12:28:02,115 INFO [decode.py:709] batch 40/?, cuts processed until now is 2641
2023-11-30 12:28:20,901 INFO [decode.py:725] The transcripts are stored in multi_KD/exp_train_asr_full_libri1_do_audio_tagging1_as_unbalanced_scale1.0/modified_beam_search/recogs-test-other-beam_size_4-epoch-30-avg-13-modified_beam_search-beam-size-4-use-averaged-model.txt
2023-11-30 12:28:21,029 INFO [utils.py:641] [test-other-beam_size_4] %WER 5.85% [3061 / 52343, 315 ins, 232 del, 2514 sub ]
2023-11-30 12:28:21,275 INFO [decode.py:738] Wrote detailed error stats to multi_KD/exp_train_asr_full_libri1_do_audio_tagging1_as_unbalanced_scale1.0/modified_beam_search/errs-test-other-beam_size_4-epoch-30-avg-13-modified_beam_search-beam-size-4-use-averaged-model.txt
2023-11-30 12:28:21,278 INFO [decode.py:754] 
For test-other, WER of different settings are:
beam_size_4	5.85	best for test-other

2023-11-30 12:28:21,279 INFO [decode.py:1051] Done!