Add training decoding logs and pretrained model

data/lang_bpe_500/bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c53433de083c4a6ad12d034550ef22de68cec62c4f58932a7b6b8b2f1e743fa5
+size 244865

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-1-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-32-38

@@ -0,0 +1,25 @@
+2022-05-30 11:32:38,010 INFO [decode.py:595] Decoding started
+2022-05-30 11:32:38,010 INFO [decode.py:601] Device: cuda:0
+2022-05-30 11:32:38,012 INFO [decode.py:616] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 1, 'left_context': -1, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-1-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:32:38,013 INFO [decode.py:618] About to create model
+2022-05-30 11:32:38,479 INFO [decode.py:685] Calculating the averaged model over epoch range from 23 (excluded) to 29
+2022-05-30 11:32:44,993 INFO [decode.py:707] Number of model parameters: 78648040
+2022-05-30 11:32:44,994 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-30 11:32:45,214 INFO [asr_datamodule.py:427] About to get test-other cuts
+2022-05-30 11:32:46,570 INFO [decode.py:502] batch 0/?, cuts processed until now is 205
+2022-05-30 11:33:02,021 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:33:02,089 INFO [utils.py:405] [test-clean-greedy_search] %WER 5.49% [2885 / 52576, 226 ins, 509 del, 2150 sub ]
+2022-05-30 11:33:02,322 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:33:02,323 INFO [decode.py:549] 
+For test-clean, WER of different settings are:
+greedy_search	5.49	best for test-clean
+
+2022-05-30 11:33:03,528 INFO [decode.py:502] batch 0/?, cuts processed until now is 231
+2022-05-30 11:33:17,923 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:33:18,023 INFO [utils.py:405] [test-other-greedy_search] %WER 13.44% [7036 / 52343, 497 ins, 1541 del, 4998 sub ]
+2022-05-30 11:33:18,373 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:33:18,374 INFO [decode.py:549] 
+For test-other, WER of different settings are:
+greedy_search	13.44	best for test-other
+
+2022-05-30 11:33:18,374 INFO [decode.py:735] Done!

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-1-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-30-01

@@ -0,0 +1,8 @@
+2022-05-30 11:30:01,921 INFO [decode.py:594] Decoding started
+2022-05-30 11:30:01,921 INFO [decode.py:600] Device: cuda:0
+2022-05-30 11:30:01,925 INFO [decode.py:615] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 1, 'left_context': 32, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-1-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:30:01,925 INFO [decode.py:617] About to create model
+2022-05-30 11:30:02,343 INFO [decode.py:684] Calculating the averaged model over epoch range from 23 (excluded) to 29
+2022-05-30 11:30:24,928 INFO [decode.py:706] Number of model parameters: 78648040
+2022-05-30 11:30:24,928 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-30 11:30:25,145 INFO [asr_datamodule.py:427] About to get test-other cuts

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-1-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-31-06

@@ -0,0 +1,25 @@
+2022-05-30 11:31:06,111 INFO [decode.py:595] Decoding started
+2022-05-30 11:31:06,112 INFO [decode.py:601] Device: cuda:0
+2022-05-30 11:31:06,114 INFO [decode.py:616] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 1, 'left_context': 32, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-1-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:31:06,114 INFO [decode.py:618] About to create model
+2022-05-30 11:31:06,580 INFO [decode.py:685] Calculating the averaged model over epoch range from 23 (excluded) to 29
+2022-05-30 11:31:13,017 INFO [decode.py:707] Number of model parameters: 78648040
+2022-05-30 11:31:13,017 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-30 11:31:13,214 INFO [asr_datamodule.py:427] About to get test-other cuts
+2022-05-30 11:31:14,679 INFO [decode.py:502] batch 0/?, cuts processed until now is 205
+2022-05-30 11:31:31,053 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:31:31,121 INFO [utils.py:405] [test-clean-greedy_search] %WER 5.42% [2852 / 52576, 225 ins, 474 del, 2153 sub ]
+2022-05-30 11:31:31,307 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:31:31,308 INFO [decode.py:549] 
+For test-clean, WER of different settings are:
+greedy_search	5.42	best for test-clean
+
+2022-05-30 11:31:33,179 INFO [decode.py:502] batch 0/?, cuts processed until now is 231
+2022-05-30 11:31:48,441 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:31:48,544 INFO [utils.py:405] [test-other-greedy_search] %WER 13.88% [7267 / 52343, 495 ins, 1530 del, 5242 sub ]
+2022-05-30 11:31:48,934 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:31:48,935 INFO [decode.py:549] 
+For test-other, WER of different settings are:
+greedy_search	13.88	best for test-other
+
+2022-05-30 11:31:48,935 INFO [decode.py:735] Done!

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-1-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-30-30

@@ -0,0 +1,5 @@
+2022-05-30 11:30:30,211 INFO [decode.py:594] Decoding started
+2022-05-30 11:30:30,212 INFO [decode.py:600] Device: cuda:0
+2022-05-30 11:30:30,214 INFO [decode.py:615] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 1, 'left_context': 64, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-1-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:30:30,214 INFO [decode.py:617] About to create model
+2022-05-30 11:30:30,643 INFO [decode.py:684] Calculating the averaged model over epoch range from 23 (excluded) to 29

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-1-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-31-51

@@ -0,0 +1,25 @@
+2022-05-30 11:31:51,944 INFO [decode.py:595] Decoding started
+2022-05-30 11:31:51,944 INFO [decode.py:601] Device: cuda:0
+2022-05-30 11:31:51,946 INFO [decode.py:616] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 1, 'left_context': 64, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-1-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:31:51,946 INFO [decode.py:618] About to create model
+2022-05-30 11:31:52,364 INFO [decode.py:685] Calculating the averaged model over epoch range from 23 (excluded) to 29
+2022-05-30 11:31:58,890 INFO [decode.py:707] Number of model parameters: 78648040
+2022-05-30 11:31:58,891 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-30 11:31:59,074 INFO [asr_datamodule.py:427] About to get test-other cuts
+2022-05-30 11:32:00,496 INFO [decode.py:502] batch 0/?, cuts processed until now is 205
+2022-05-30 11:32:17,000 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:32:17,069 INFO [utils.py:405] [test-clean-greedy_search] %WER 5.43% [2856 / 52576, 243 ins, 481 del, 2132 sub ]
+2022-05-30 11:32:17,256 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:32:17,257 INFO [decode.py:549] 
+For test-clean, WER of different settings are:
+greedy_search	5.43	best for test-clean
+
+2022-05-30 11:32:18,554 INFO [decode.py:502] batch 0/?, cuts processed until now is 231
+2022-05-30 11:32:34,311 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:32:34,391 INFO [utils.py:405] [test-other-greedy_search] %WER 13.45% [7041 / 52343, 491 ins, 1479 del, 5071 sub ]
+2022-05-30 11:32:34,761 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-1-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:32:34,762 INFO [decode.py:549] 
+For test-other, WER of different settings are:
+greedy_search	13.45	best for test-other
+
+2022-05-30 11:32:34,762 INFO [decode.py:735] Done!

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-16-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-41-50

@@ -0,0 +1,25 @@
+2022-05-30 11:41:50,931 INFO [decode.py:595] Decoding started
+2022-05-30 11:41:50,931 INFO [decode.py:601] Device: cuda:0
+2022-05-30 11:41:50,933 INFO [decode.py:616] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 16, 'left_context': -1, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-16-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:41:50,933 INFO [decode.py:618] About to create model
+2022-05-30 11:41:51,411 INFO [decode.py:685] Calculating the averaged model over epoch range from 23 (excluded) to 29
+2022-05-30 11:41:58,335 INFO [decode.py:707] Number of model parameters: 78648040
+2022-05-30 11:41:58,335 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-30 11:41:58,540 INFO [asr_datamodule.py:427] About to get test-other cuts
+2022-05-30 11:42:00,058 INFO [decode.py:502] batch 0/?, cuts processed until now is 205
+2022-05-30 11:42:16,167 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:42:16,259 INFO [utils.py:405] [test-clean-greedy_search] %WER 3.51% [1847 / 52576, 193 ins, 162 del, 1492 sub ]
+2022-05-30 11:42:16,502 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:42:16,503 INFO [decode.py:549] 
+For test-clean, WER of different settings are:
+greedy_search	3.51	best for test-clean
+
+2022-05-30 11:42:17,734 INFO [decode.py:502] batch 0/?, cuts processed until now is 231
+2022-05-30 11:42:32,761 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:42:32,857 INFO [utils.py:405] [test-other-greedy_search] %WER 8.68% [4543 / 52343, 461 ins, 478 del, 3604 sub ]
+2022-05-30 11:42:33,110 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:42:33,110 INFO [decode.py:549] 
+For test-other, WER of different settings are:
+greedy_search	8.68	best for test-other
+
+2022-05-30 11:42:33,111 INFO [decode.py:735] Done!

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-16-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-40-17

@@ -0,0 +1,25 @@
+2022-05-30 11:40:17,275 INFO [decode.py:595] Decoding started
+2022-05-30 11:40:17,276 INFO [decode.py:601] Device: cuda:0
+2022-05-30 11:40:17,278 INFO [decode.py:616] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 16, 'left_context': 32, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-16-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:40:17,278 INFO [decode.py:618] About to create model
+2022-05-30 11:40:17,728 INFO [decode.py:685] Calculating the averaged model over epoch range from 23 (excluded) to 29
+2022-05-30 11:40:24,733 INFO [decode.py:707] Number of model parameters: 78648040
+2022-05-30 11:40:24,733 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-30 11:40:24,970 INFO [asr_datamodule.py:427] About to get test-other cuts
+2022-05-30 11:40:26,491 INFO [decode.py:502] batch 0/?, cuts processed until now is 205
+2022-05-30 11:40:42,800 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:40:42,869 INFO [utils.py:405] [test-clean-greedy_search] %WER 3.41% [1795 / 52576, 188 ins, 150 del, 1457 sub ]
+2022-05-30 11:40:43,057 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:40:43,058 INFO [decode.py:549] 
+For test-clean, WER of different settings are:
+greedy_search	3.41	best for test-clean
+
+2022-05-30 11:40:44,309 INFO [decode.py:502] batch 0/?, cuts processed until now is 231
+2022-05-30 11:41:00,010 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:41:00,108 INFO [utils.py:405] [test-other-greedy_search] %WER 8.86% [4640 / 52343, 466 ins, 483 del, 3691 sub ]
+2022-05-30 11:41:00,345 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:41:00,346 INFO [decode.py:549] 
+For test-other, WER of different settings are:
+greedy_search	8.86	best for test-other
+
+2022-05-30 11:41:00,346 INFO [decode.py:735] Done!

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-2022-05-29-18-24-55

@@ -0,0 +1,25 @@
+2022-05-29 18:24:55,449 INFO [decode.py:592] Decoding started
+2022-05-29 18:24:55,449 INFO [decode.py:598] Device: cuda:0
+2022-05-29 18:24:55,452 INFO [decode.py:613] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-7-0309102938-68688b4cbd-xhtcg', 'IP address': '10.48.32.137'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': False, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 16, 'left_context': 64, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-29 18:24:55,452 INFO [decode.py:615] About to create model
+2022-05-29 18:24:55,859 INFO [decode.py:644] averaging ['streaming_pruned_transducer_stateless4/exp/epoch-24.pt', 'streaming_pruned_transducer_stateless4/exp/epoch-25.pt', 'streaming_pruned_transducer_stateless4/exp/epoch-26.pt', 'streaming_pruned_transducer_stateless4/exp/epoch-27.pt', 'streaming_pruned_transducer_stateless4/exp/epoch-28.pt', 'streaming_pruned_transducer_stateless4/exp/epoch-29.pt']
+2022-05-29 18:25:05,191 INFO [decode.py:704] Number of model parameters: 78648040
+2022-05-29 18:25:05,191 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-29 18:25:05,310 INFO [asr_datamodule.py:427] About to get test-other cuts
+2022-05-29 18:25:07,080 INFO [decode.py:499] batch 0/?, cuts processed until now is 205
+2022-05-29 18:25:23,183 INFO [decode.py:516] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1.txt
+2022-05-29 18:25:23,253 INFO [utils.py:405] [test-clean-greedy_search] %WER 3.40% [1790 / 52576, 184 ins, 153 del, 1453 sub ]
+2022-05-29 18:25:23,440 INFO [decode.py:529] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1.txt
+2022-05-29 18:25:23,448 INFO [decode.py:546] 
+For test-clean, WER of different settings are:
+greedy_search	3.4	best for test-clean
+
+2022-05-29 18:25:24,678 INFO [decode.py:499] batch 0/?, cuts processed until now is 231
+2022-05-29 18:25:40,369 INFO [decode.py:516] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1.txt
+2022-05-29 18:25:40,443 INFO [utils.py:405] [test-other-greedy_search] %WER 8.77% [4592 / 52343, 482 ins, 448 del, 3662 sub ]
+2022-05-29 18:25:40,641 INFO [decode.py:529] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1.txt
+2022-05-29 18:25:40,649 INFO [decode.py:546] 
+For test-other, WER of different settings are:
+greedy_search	8.77	best for test-other
+
+2022-05-29 18:25:40,649 INFO [decode.py:732] Done!

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-41-03

@@ -0,0 +1,25 @@
+2022-05-30 11:41:03,602 INFO [decode.py:595] Decoding started
+2022-05-30 11:41:03,602 INFO [decode.py:601] Device: cuda:0
+2022-05-30 11:41:03,604 INFO [decode.py:616] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 16, 'left_context': 64, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:41:03,604 INFO [decode.py:618] About to create model
+2022-05-30 11:41:04,044 INFO [decode.py:685] Calculating the averaged model over epoch range from 23 (excluded) to 29
+2022-05-30 11:41:11,215 INFO [decode.py:707] Number of model parameters: 78648040
+2022-05-30 11:41:11,215 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-30 11:41:11,440 INFO [asr_datamodule.py:427] About to get test-other cuts
+2022-05-30 11:41:12,944 INFO [decode.py:502] batch 0/?, cuts processed until now is 205
+2022-05-30 11:41:30,041 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:41:30,134 INFO [utils.py:405] [test-clean-greedy_search] %WER 3.35% [1763 / 52576, 182 ins, 158 del, 1423 sub ]
+2022-05-30 11:41:30,381 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:41:30,382 INFO [decode.py:549] 
+For test-clean, WER of different settings are:
+greedy_search	3.35	best for test-clean
+
+2022-05-30 11:41:31,731 INFO [decode.py:502] batch 0/?, cuts processed until now is 231
+2022-05-30 11:41:47,728 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:41:47,803 INFO [utils.py:405] [test-other-greedy_search] %WER 8.62% [4514 / 52343, 456 ins, 465 del, 3593 sub ]
+2022-05-30 11:41:48,030 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:41:48,031 INFO [decode.py:549] 
+For test-other, WER of different settings are:
+greedy_search	8.62	best for test-other
+
+2022-05-30 11:41:48,031 INFO [decode.py:735] Done!

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-2-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-34-54

@@ -0,0 +1,25 @@
+2022-05-30 11:34:54,451 INFO [decode.py:595] Decoding started
+2022-05-30 11:34:54,451 INFO [decode.py:601] Device: cuda:0
+2022-05-30 11:34:54,453 INFO [decode.py:616] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 2, 'left_context': -1, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-2-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:34:54,453 INFO [decode.py:618] About to create model
+2022-05-30 11:34:54,952 INFO [decode.py:685] Calculating the averaged model over epoch range from 23 (excluded) to 29
+2022-05-30 11:35:02,174 INFO [decode.py:707] Number of model parameters: 78648040
+2022-05-30 11:35:02,175 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-30 11:35:02,377 INFO [asr_datamodule.py:427] About to get test-other cuts
+2022-05-30 11:35:03,929 INFO [decode.py:502] batch 0/?, cuts processed until now is 205
+2022-05-30 11:35:19,792 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-2-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:35:19,863 INFO [utils.py:405] [test-clean-greedy_search] %WER 4.61% [2424 / 52576, 226 ins, 247 del, 1951 sub ]
+2022-05-30 11:35:20,112 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-2-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:35:20,112 INFO [decode.py:549] 
+For test-clean, WER of different settings are:
+greedy_search	4.61	best for test-clean
+
+2022-05-30 11:35:21,316 INFO [decode.py:502] batch 0/?, cuts processed until now is 231
+2022-05-30 11:35:36,235 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-2-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:35:36,339 INFO [utils.py:405] [test-other-greedy_search] %WER 10.84% [5673 / 52343, 497 ins, 771 del, 4405 sub ]
+2022-05-30 11:35:36,536 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-2-left-context--1-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:35:36,536 INFO [decode.py:549] 
+For test-other, WER of different settings are:
+greedy_search	10.84	best for test-other
+
+2022-05-30 11:35:36,536 INFO [decode.py:735] Done!

decoding_results/greedy_search/log-decode-epoch-29-avg-6-streaming-chunk-size-2-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-33-21

@@ -0,0 +1,25 @@
+2022-05-30 11:33:21,303 INFO [decode.py:595] Decoding started
+2022-05-30 11:33:21,303 INFO [decode.py:601] Device: cuda:0
+2022-05-30 11:33:21,306 INFO [decode.py:616] {'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, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 2, 'left_context': 32, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-2-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500}
+2022-05-30 11:33:21,306 INFO [decode.py:618] About to create model
+2022-05-30 11:33:21,754 INFO [decode.py:685] Calculating the averaged model over epoch range from 23 (excluded) to 29
+2022-05-30 11:33:28,457 INFO [decode.py:707] Number of model parameters: 78648040
+2022-05-30 11:33:28,457 INFO [asr_datamodule.py:422] About to get test-clean cuts
+2022-05-30 11:33:28,680 INFO [asr_datamodule.py:427] About to get test-other cuts
+2022-05-30 11:33:30,346 INFO [decode.py:502] batch 0/?, cuts processed until now is 205
+2022-05-30 11:33:46,473 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-2-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:33:46,541 INFO [utils.py:405] [test-clean-greedy_search] %WER 4.25% [2236 / 52576, 225 ins, 213 del, 1798 sub ]
+2022-05-30 11:33:46,743 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-2-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:33:46,745 INFO [decode.py:549] 
+For test-clean, WER of different settings are:
+greedy_search	4.25	best for test-clean
+
+2022-05-30 11:33:48,183 INFO [decode.py:502] batch 0/?, cuts processed until now is 231
+2022-05-30 11:34:03,899 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-2-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:34:03,979 INFO [utils.py:405] [test-other-greedy_search] %WER 10.85% [5679 / 52343, 525 ins, 708 del, 4446 sub ]
+2022-05-30 11:34:04,175 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-2-left-context-32-context-2-max-sym-per-frame-1-use-averaged-model.txt
+2022-05-30 11:34:04,176 INFO [decode.py:549] 
+For test-other, WER of different settings are:
+greedy_search	10.85	best for test-other
+
+2022-05-30 11:34:04,176 INFO [decode.py:735] Done!