add files

exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/epoch-16-avg-4.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b5871854db3a0365e6f4e7a2aa14f1b8bf21386c8f6b159c2f34f156b894a1c
+size 264116385

exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/log-decode-epoch-11-avg-1-context-2-max-sym-per-frame-1-use-averaged-model-2023-12-07-10-51-49

@@ -0,0 +1,29 @@
+2023-12-07 10:51:49,231 INFO [decode.py:836] Decoding started
+2023-12-07 10:51:49,232 INFO [decode.py:842] Device: cuda:0
+2023-12-07 10:51:49,236 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': 'e400fa3b456faf8afe0ee5bfe572946b4921a3db', 'k2-git-date': 'Sat Jul 15 04:21:50 2023', 'lhotse-version': '1.16.0', 'torch-version': '2.0.1+cu117', 'torch-cuda-available': True, 'torch-cuda-version': '11.7', 'python-version': '3.9', 'icefall-git-branch': 'multi_KD', 'icefall-git-sha1': 'a77761c2-dirty', 'icefall-git-date': 'Tue Nov 28 15:54:58 2023', 'icefall-path': '/xy/mnt/yangxiaoyu/workspace/icefall_multi_KD', 'k2-path': '/root/anaconda3/lib/python3.9/site-packages/k2/__init__.py', 'lhotse-path': '/root/anaconda3/lib/python3.9/site-packages/lhotse/__init__.py', 'hostname': 'NGK_xiaoyu'}, 'epoch': 11, 'iter': 0, 'avg': 1, 'use_averaged_model': True, 'exp_dir': PosixPath('multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0'), '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, '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_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search'), 'has_contexts': False, 'suffix': 'epoch-11-avg-1-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2023-12-07 10:51:49,236 INFO [decode.py:854] About to create model
+2023-12-07 10:51:49,590 INFO [decode.py:921] Calculating the averaged model over epoch range from 10 (excluded) to 11
+2023-12-07 10:52:00,924 INFO [decode.py:1016] Number of model parameters: 65549011
+2023-12-07 10:52:00,925 INFO [asr_datamodule.py:465] About to get test-clean cuts
+2023-12-07 10:52:00,966 INFO [asr_datamodule.py:472] About to get test-other cuts
+2023-12-07 10:52:06,639 INFO [decode.py:709] batch 0/?, cuts processed until now is 36
+2023-12-07 10:52:48,711 INFO [decode.py:709] batch 50/?, cuts processed until now is 2610
+2023-12-07 10:52:49,648 INFO [decode.py:725] The transcripts are stored in multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/recogs-test-clean-greedy_search-epoch-11-avg-1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2023-12-07 10:52:49,720 INFO [utils.py:641] [test-clean-greedy_search] %WER 2.99% [1573 / 52576, 165 ins, 124 del, 1284 sub ]
+2023-12-07 10:52:49,848 INFO [decode.py:738] Wrote detailed error stats to multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/errs-test-clean-greedy_search-epoch-11-avg-1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2023-12-07 10:52:49,851 INFO [decode.py:754] 
+For test-clean, WER of different settings are:
+greedy_search	2.99	best for test-clean
+
+2023-12-07 10:52:51,069 INFO [decode.py:709] batch 0/?, cuts processed until now is 43
+2023-12-07 10:52:54,626 INFO [zipformer.py:1877] name=None, attn_weights_entropy = tensor([3.1463, 4.6075, 2.6228, 4.8510], device='cuda:0')
+2023-12-07 10:53:06,434 INFO [zipformer.py:1877] name=None, attn_weights_entropy = tensor([4.3109, 5.3362, 3.8469, 5.5580], device='cuda:0')
+2023-12-07 10:53:31,261 INFO [decode.py:709] batch 50/?, cuts processed until now is 2939
+2023-12-07 10:53:31,357 INFO [decode.py:725] The transcripts are stored in multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/recogs-test-other-greedy_search-epoch-11-avg-1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2023-12-07 10:53:31,424 INFO [utils.py:641] [test-other-greedy_search] %WER 7.42% [3883 / 52343, 383 ins, 344 del, 3156 sub ]
+2023-12-07 10:53:31,554 INFO [decode.py:738] Wrote detailed error stats to multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/errs-test-other-greedy_search-epoch-11-avg-1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2023-12-07 10:53:31,558 INFO [decode.py:754] 
+For test-other, WER of different settings are:
+greedy_search	7.42	best for test-other
+
+2023-12-07 10:53:31,558 INFO [decode.py:1051] Done!

exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/log-decode-epoch-11-avg-2-context-2-max-sym-per-frame-1-use-averaged-model-2023-12-07-10-45-45

@@ -0,0 +1,29 @@
+2023-12-07 10:45:45,792 INFO [decode.py:836] Decoding started
+2023-12-07 10:45:45,792 INFO [decode.py:842] Device: cuda:0
+2023-12-07 10:45:45,799 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': 'e400fa3b456faf8afe0ee5bfe572946b4921a3db', 'k2-git-date': 'Sat Jul 15 04:21:50 2023', 'lhotse-version': '1.16.0', 'torch-version': '2.0.1+cu117', 'torch-cuda-available': True, 'torch-cuda-version': '11.7', 'python-version': '3.9', 'icefall-git-branch': 'multi_KD', 'icefall-git-sha1': 'a77761c2-dirty', 'icefall-git-date': 'Tue Nov 28 15:54:58 2023', 'icefall-path': '/xy/mnt/yangxiaoyu/workspace/icefall_multi_KD', 'k2-path': '/root/anaconda3/lib/python3.9/site-packages/k2/__init__.py', 'lhotse-path': '/root/anaconda3/lib/python3.9/site-packages/lhotse/__init__.py', 'hostname': 'NGK_xiaoyu'}, 'epoch': 11, 'iter': 0, 'avg': 2, 'use_averaged_model': True, 'exp_dir': PosixPath('multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0'), '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, '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_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search'), 'has_contexts': False, 'suffix': 'epoch-11-avg-2-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2023-12-07 10:45:45,800 INFO [decode.py:854] About to create model
+2023-12-07 10:45:46,188 INFO [decode.py:921] Calculating the averaged model over epoch range from 9 (excluded) to 11
+2023-12-07 10:46:03,594 INFO [decode.py:1016] Number of model parameters: 65549011
+2023-12-07 10:46:03,594 INFO [asr_datamodule.py:465] About to get test-clean cuts
+2023-12-07 10:46:03,744 INFO [asr_datamodule.py:472] About to get test-other cuts
+2023-12-07 10:46:11,374 INFO [decode.py:709] batch 0/?, cuts processed until now is 36
+2023-12-07 10:46:35,478 INFO [zipformer.py:1877] name=None, attn_weights_entropy = tensor([4.8468, 3.6469, 3.5498, 4.2244], device='cuda:0')
+2023-12-07 10:46:52,652 INFO [decode.py:709] batch 50/?, cuts processed until now is 2610
+2023-12-07 10:46:53,524 INFO [decode.py:725] The transcripts are stored in multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/recogs-test-clean-greedy_search-epoch-11-avg-2-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2023-12-07 10:46:53,591 INFO [utils.py:641] [test-clean-greedy_search] %WER 3.01% [1585 / 52576, 165 ins, 129 del, 1291 sub ]
+2023-12-07 10:46:53,718 INFO [decode.py:738] Wrote detailed error stats to multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/errs-test-clean-greedy_search-epoch-11-avg-2-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2023-12-07 10:46:53,722 INFO [decode.py:754] 
+For test-clean, WER of different settings are:
+greedy_search	3.01	best for test-clean
+
+2023-12-07 10:46:54,978 INFO [decode.py:709] batch 0/?, cuts processed until now is 43
+2023-12-07 10:47:19,687 INFO [zipformer.py:1877] name=None, attn_weights_entropy = tensor([3.5610, 2.8195, 1.5632, 2.6141], device='cuda:0')
+2023-12-07 10:47:32,676 INFO [decode.py:709] batch 50/?, cuts processed until now is 2939
+2023-12-07 10:47:32,758 INFO [decode.py:725] The transcripts are stored in multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/recogs-test-other-greedy_search-epoch-11-avg-2-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2023-12-07 10:47:32,828 INFO [utils.py:641] [test-other-greedy_search] %WER 7.45% [3899 / 52343, 405 ins, 334 del, 3160 sub ]
+2023-12-07 10:47:32,960 INFO [decode.py:738] Wrote detailed error stats to multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/errs-test-other-greedy_search-epoch-11-avg-2-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2023-12-07 10:47:32,963 INFO [decode.py:754] 
+For test-other, WER of different settings are:
+greedy_search	7.45	best for test-other
+
+2023-12-07 10:47:32,963 INFO [decode.py:1051] Done!

exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search/log-decode-epoch-11-avg-2-context-2-max-sym-per-frame-1-use-averaged-model-2023-12-07-10-51-25

@@ -0,0 +1,5 @@
+2023-12-07 10:51:25,613 INFO [decode.py:836] Decoding started
+2023-12-07 10:51:25,613 INFO [decode.py:842] Device: cuda:0
+2023-12-07 10:51:25,616 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': 'e400fa3b456faf8afe0ee5bfe572946b4921a3db', 'k2-git-date': 'Sat Jul 15 04:21:50 2023', 'lhotse-version': '1.16.0', 'torch-version': '2.0.1+cu117', 'torch-cuda-available': True, 'torch-cuda-version': '11.7', 'python-version': '3.9', 'icefall-git-branch': 'multi_KD', 'icefall-git-sha1': 'a77761c2-dirty', 'icefall-git-date': 'Tue Nov 28 15:54:58 2023', 'icefall-path': '/xy/mnt/yangxiaoyu/workspace/icefall_multi_KD', 'k2-path': '/root/anaconda3/lib/python3.9/site-packages/k2/__init__.py', 'lhotse-path': '/root/anaconda3/lib/python3.9/site-packages/lhotse/__init__.py', 'hostname': 'NGK_xiaoyu'}, 'epoch': 11, 'iter': 0, 'avg': 2, 'use_averaged_model': True, 'exp_dir': PosixPath('multi_KD/exp_train_asr_full_libri1_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0'), '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, '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_6-fold_do_AT1_as_unbalanced_scale1.0_do_SV1_only_vox2_scale10.0_musan0/greedy_search'), 'has_contexts': False, 'suffix': 'epoch-11-avg-2-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2023-12-07 10:51:25,616 INFO [decode.py:854] About to create model
+2023-12-07 10:51:25,963 INFO [decode.py:921] Calculating the averaged model over epoch range from 9 (excluded) to 11