icefall-asr-librispeech-zipformer-large-2023-05-16 / decoding_result /greedy_search /log-decode-epoch-30-avg-9-context-2-max-sym-per-frame-1-use-averaged-model-2023-05-16-16-17-23
Zengwei's picture
upload files
8d435a5
raw
history blame
4.9 kB
2023-05-16 16:17:23,182 INFO [decode.py:678] Decoding started
2023-05-16 16:17:23,182 INFO [decode.py:684] Device: cuda:0
2023-05-16 16:17:23,185 INFO [decode.py:694] {'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': '1.24.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'c51a0b9684442a88ee37f3ce0af686a04b66855b', 'k2-git-date': 'Mon May 1 21:38:03 2023', 'lhotse-version': '1.12.0.dev+git.891bad1.clean', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'zipformer2_from_dan', 'icefall-git-sha1': '85d51991-dirty', 'icefall-git-date': 'Tue May 16 14:39:13 2023', 'icefall-path': '/ceph-zw/workspace/zipformer/icefall_zipformer', 'k2-path': '/ceph-zw/workspace/k2/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-zw/workspace/share/lhotse/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-1-1220091118-57c4d55446-mlpzc', 'IP address': '10.177.22.19'}, 'epoch': 30, 'iter': 0, 'avg': 9, 'use_averaged_model': True, 'exp_dir': PosixPath('zipformer/exp-large'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'lang_dir': PosixPath('data/lang_bpe_500'), 'decoding_method': 'greedy_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, 'num_encoder_layers': '2,2,4,5,4,2', 'downsampling_factor': '1,2,4,8,4,2', 'feedforward_dim': '512,768,1536,2048,1536,768', 'num_heads': '4,4,4,8,4,4', 'encoder_dim': '192,256,512,768,512,256', 'query_head_dim': '32', 'value_head_dim': '12', 'pos_head_dim': '4', 'pos_dim': 48, 'encoder_unmasked_dim': '192,192,256,320,256,192', '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', 'full_libri': True, 'mini_libri': False, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 600, '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-large/greedy_search'), 'suffix': 'epoch-30-avg-9-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
2023-05-16 16:17:23,186 INFO [decode.py:696] About to create model
2023-05-16 16:17:24,155 INFO [decode.py:763] Calculating the averaged model over epoch range from 21 (excluded) to 30
2023-05-16 16:17:54,525 INFO [decode.py:799] Number of model parameters: 148439574
2023-05-16 16:17:54,525 INFO [asr_datamodule.py:465] About to get test-clean cuts
2023-05-16 16:17:54,529 INFO [asr_datamodule.py:472] About to get test-other cuts
2023-05-16 16:17:56,104 INFO [decode.py:570] batch 0/?, cuts processed until now is 43
2023-05-16 16:17:56,300 INFO [zipformer.py:1711] name=None, attn_weights_entropy = tensor([2.5593, 3.8085, 3.8116, 3.4860, 3.4736, 3.9205, 3.1368, 3.0704],
device='cuda:0')
2023-05-16 16:18:26,649 INFO [decode.py:588] The transcripts are stored in zipformer/exp-large/greedy_search/recogs-test-clean-greedy_search-epoch-30-avg-9-context-2-max-sym-per-frame-1-use-averaged-model.txt
2023-05-16 16:18:26,718 INFO [utils.py:558] [test-clean-greedy_search] %WER 2.12% [1116 / 52576, 118 ins, 89 del, 909 sub ]
2023-05-16 16:18:26,890 INFO [decode.py:601] Wrote detailed error stats to zipformer/exp-large/greedy_search/errs-test-clean-greedy_search-epoch-30-avg-9-context-2-max-sym-per-frame-1-use-averaged-model.txt
2023-05-16 16:18:26,891 INFO [decode.py:618]
For test-clean, WER of different settings are:
greedy_search 2.12 best for test-clean
2023-05-16 16:18:28,118 INFO [decode.py:570] batch 0/?, cuts processed until now is 52
2023-05-16 16:18:56,735 INFO [decode.py:588] The transcripts are stored in zipformer/exp-large/greedy_search/recogs-test-other-greedy_search-epoch-30-avg-9-context-2-max-sym-per-frame-1-use-averaged-model.txt
2023-05-16 16:18:56,812 INFO [utils.py:558] [test-other-greedy_search] %WER 4.91% [2570 / 52343, 255 ins, 221 del, 2094 sub ]
2023-05-16 16:18:56,999 INFO [decode.py:601] Wrote detailed error stats to zipformer/exp-large/greedy_search/errs-test-other-greedy_search-epoch-30-avg-9-context-2-max-sym-per-frame-1-use-averaged-model.txt
2023-05-16 16:18:57,000 INFO [decode.py:618]
For test-other, WER of different settings are:
greedy_search 4.91 best for test-other
2023-05-16 16:18:57,000 INFO [decode.py:830] Done!