decoding-results/log-decode-epoch-25-avg-5-modified_beam_search-beam-size-10-use-averaged-model-2024-02-19-20-37-30

2024-02-19 20:37:30,542 INFO [decode_seame.py:870] Decoding started
2024-02-19 20:37:30,542 INFO [decode_seame.py:876] Device: cuda:0
2024-02-19 20:37:30,557 INFO [decode_seame.py:886] {'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': 5000, 'env_info': {'k2-version': '1.24.3', 'k2-build-type': 'Debug', 'k2-with-cuda': True, 'k2-git-sha1': '1a76309e5c6343c4d18965b7ce134a7f311d9d3a', 'k2-git-date': 'Sun May 28 02:04:03 2023', 'lhotse-version': '1.17.0.dev+git.1c7c0cf.dirty', 'torch-version': '2.0.0+cu117', 'torch-cuda-available': True, 'torch-cuda-version': '11.7', 'python-version': '3.8', 'icefall-git-branch': None, 'icefall-git-sha1': None, 'icefall-git-date': None, 'icefall-path': '/home/hltcoe/ahussein/icefall', 'k2-path': '/home/hltcoe/ahussein/k2/k2/python/k2/__init__.py', 'lhotse-path': '/home/hltcoe/ahussein/lhotse/lhotse/__init__.py', 'hostname': 'r2n04', 'IP address': '10.1.2.4'}, 'epoch': 25, 'clean': True, 'iter': 0, 'avg': 5, 'use_averaged_model': True, 'exp_dir': PosixPath('zipformer/exp-asr-seame'), 'bpe_model': 'data_seame/lang_bpe_4000/bpe.model', 'lang_dir': PosixPath('data/lang_bpe_500'), 'decoding_method': 'modified_beam_search', 'beam_size': 10, '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': '2,2,2,2,2,2', 'downsampling_factor': '1,2,4,8,4,2', 'feedforward_dim': '512,768,1024,1024,1024,768', 'num_heads': '4,4,4,8,4,4', 'encoder_dim': '192,256,256,256,256,256', 'query_head_dim': '32', 'value_head_dim': '12', 'pos_head_dim': '4', 'pos_dim': 48, 'encoder_unmasked_dim': '192,192,192,192,192,192', 'cnn_module_kernel': '31,31,15,15,15,31', 'decoder_dim': 256, 'joiner_dim': 256, 'causal': False, 'chunk_size': '16,32,64,-1', 'left_context_frames': '64,128,256,-1', 'use_transducer': True, 'use_ctc': False, 'manifest_dir': PosixPath('data_seame/fbank'), 'max_duration': 800, 'lid': False, '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': 10, '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('zipformer/exp-asr-seame/modified_beam_search'), 'has_contexts': False, 'suffix': 'epoch-25-avg-5-modified_beam_search-beam-size-10-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 4000}
2024-02-19 20:37:30,557 INFO [decode_seame.py:888] About to create model
2024-02-19 20:37:30,978 INFO [decode_seame.py:955] Calculating the averaged model over epoch range from 20 (excluded) to 25
2024-02-19 20:37:33,399 INFO [decode_seame.py:1049] Number of model parameters: 28595315
2024-02-19 20:37:33,399 INFO [asr_datamodule_seame.py:420] About to get dev_man cuts
2024-02-19 20:37:33,402 INFO [asr_datamodule_seame.py:425] About to get dev_sge cuts
2024-02-19 20:37:47,776 INFO [decode_seame.py:743] batch 0/?, cuts processed until now is 108
2024-02-19 20:41:20,752 INFO [decode_seame.py:743] batch 20/?, cuts processed until now is 3323
2024-02-19 20:44:03,549 INFO [decode_seame.py:743] batch 40/?, cuts processed until now is 6273
2024-02-19 20:44:26,264 INFO [decode_seame.py:759] The transcripts are stored in zipformer/exp-asr-seame/modified_beam_search/recogs-dev_man-beam_size_10-epoch-25-avg-5-modified_beam_search-beam-size-10-use-averaged-model.txt
2024-02-19 20:44:26,477 INFO [utils.py:565] [dev_man-beam_size_10] %WER 21.87% [25641 / 117256, 1821 ins, 8709 del, 15111 sub ]
2024-02-19 20:44:26,918 INFO [decode_seame.py:772] Wrote detailed error stats to zipformer/exp-asr-seame/modified_beam_search/errs-dev_man-beam_size_10-epoch-25-avg-5-modified_beam_search-beam-size-10-use-averaged-model.txt
2024-02-19 20:44:26,924 INFO [decode_seame.py:788] 
For dev_man, WER of different settings are:
beam_size_10	21.87	best for dev_man

2024-02-19 20:44:39,499 INFO [decode_seame.py:743] batch 0/?, cuts processed until now is 167
2024-02-19 20:47:53,925 INFO [decode_seame.py:743] batch 20/?, cuts processed until now is 4601
2024-02-19 20:48:14,945 INFO [decode_seame.py:759] The transcripts are stored in zipformer/exp-asr-seame/modified_beam_search/recogs-dev_sge-beam_size_10-epoch-25-avg-5-modified_beam_search-beam-size-10-use-averaged-model.txt
2024-02-19 20:48:15,071 INFO [utils.py:565] [dev_sge-beam_size_10] %WER 29.04% [19329 / 66566, 1619 ins, 5920 del, 11790 sub ]
2024-02-19 20:48:15,348 INFO [decode_seame.py:772] Wrote detailed error stats to zipformer/exp-asr-seame/modified_beam_search/errs-dev_sge-beam_size_10-epoch-25-avg-5-modified_beam_search-beam-size-10-use-averaged-model.txt
2024-02-19 20:48:15,349 INFO [decode_seame.py:788] 
For dev_sge, WER of different settings are:
beam_size_10	29.04	best for dev_sge

2024-02-19 20:48:15,350 INFO [decode_seame.py:1084] Done!