2022-05-27 13:57:22,671 INFO [train.py:887] (0/4) Training started 2022-05-27 13:57:22,674 INFO [train.py:897] (0/4) Device: cuda:0 2022-05-27 13:57:22,677 INFO [train.py:906] (0/4) {'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': 1600, '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': '1.13', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f4fefe4882bc0ae59af951da3f47335d5495ef71', 'k2-git-date': 'Thu Feb 10 15:16:02 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': 'stateless6', 'icefall-git-sha1': '50641cd-dirty', 'icefall-git-date': 'Fri May 27 13:49:39 2022', 'icefall-path': '/ceph-data2/ly/open_source/vq2_icefall', 'k2-path': '/ceph-jb/yaozengwei/workspace/rnnt/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-ly/open-source/hubert/lhotse/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-9-0425111216-65f66bdf4-bkrql', 'IP address': '10.177.77.9'}, 'enable_distiallation': False, 'distillation_layer': 5, 'num_codebooks': 16, 'world_size': 4, 'master_port': 12359, 'tensorboard': True, 'num_epochs': 50, 'start_epoch': 1, 'start_batch': 0, 'exp_dir': PosixPath('pruned_transducer_stateless6/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'initial_lr': 0.003, 'lr_batches': 5000, 'lr_epochs': 6, 'context_size': 2, 'prune_range': 5, 'lm_scale': 0.25, 'am_scale': 0.0, 'simple_loss_scale': 0.5, 'codebook_loss_scale': 0.1, 'seed': 42, 'print_diagnostics': False, 'save_every_n': 8000, 'keep_last_k': 20, 'average_period': 100, 'use_fp16': False, 'full_libri': False, 'manifest_dir': PosixPath('data/vq_fbank'), 'max_duration': 300, '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': -1, 'enable_musan': True, 'input_strategy': 'PrecomputedFeatures', 'blank_id': 0, 'vocab_size': 500} 2022-05-27 13:57:22,677 INFO [train.py:908] (0/4) About to create model 2022-05-27 13:57:23,146 INFO [train.py:912] (0/4) Number of model parameters: 78648040 2022-05-27 13:57:28,269 INFO [train.py:927] (0/4) Using DDP 2022-05-27 13:57:29,320 INFO [asr_datamodule.py:408] (0/4) About to get train-clean-100 cuts 2022-05-27 13:57:38,657 INFO [asr_datamodule.py:225] (0/4) Enable MUSAN 2022-05-27 13:57:38,657 INFO [asr_datamodule.py:226] (0/4) About to get Musan cuts 2022-05-27 13:57:42,323 INFO [asr_datamodule.py:254] (0/4) Enable SpecAugment 2022-05-27 13:57:42,324 INFO [asr_datamodule.py:255] (0/4) Time warp factor: -1 2022-05-27 13:57:42,324 INFO [asr_datamodule.py:267] (0/4) Num frame mask: 10 2022-05-27 13:57:42,324 INFO [asr_datamodule.py:280] (0/4) About to create train dataset 2022-05-27 13:57:42,324 INFO [asr_datamodule.py:309] (0/4) Using BucketingSampler. 2022-05-27 13:57:42,661 INFO [asr_datamodule.py:325] (0/4) About to create train dataloader 2022-05-27 13:57:42,663 INFO [asr_datamodule.py:429] (0/4) About to get dev-clean cuts 2022-05-27 13:57:42,820 INFO [asr_datamodule.py:434] (0/4) About to get dev-other cuts 2022-05-27 13:57:42,980 INFO [asr_datamodule.py:356] (0/4) About to create dev dataset 2022-05-27 13:57:42,992 INFO [asr_datamodule.py:375] (0/4) About to create dev dataloader 2022-05-27 13:57:42,992 INFO [train.py:1054] (0/4) Sanity check -- see if any of the batches in epoch 1 would cause OOM. 2022-05-27 13:57:47,474 INFO [distributed.py:874] (0/4) Reducer buckets have been rebuilt in this iteration. 2022-05-27 13:58:01,531 INFO [train.py:823] (0/4) Epoch 1, batch 0, loss[loss=0.9294, simple_loss=1.859, pruned_loss=6.839, over 7282.00 frames.], tot_loss[loss=0.9294, simple_loss=1.859, pruned_loss=6.839, over 7282.00 frames.], batch size: 21, lr: 3.00e-03 2022-05-27 13:58:40,707 INFO [train.py:823] (0/4) Epoch 1, batch 50, loss[loss=0.5262, simple_loss=1.052, pruned_loss=7.058, over 7130.00 frames.], tot_loss[loss=0.5726, simple_loss=1.145, pruned_loss=7.154, over 321419.36 frames.], batch size: 23, lr: 3.00e-03 2022-05-27 13:59:20,058 INFO [train.py:823] (0/4) Epoch 1, batch 100, loss[loss=0.4676, simple_loss=0.9351, pruned_loss=7.028, over 7192.00 frames.], tot_loss[loss=0.5138, simple_loss=1.028, pruned_loss=7.071, over 563050.92 frames.], batch size: 20, lr: 3.00e-03 2022-05-27 13:59:59,629 INFO [train.py:823] (0/4) Epoch 1, batch 150, loss[loss=0.4618, simple_loss=0.9235, pruned_loss=6.913, over 7347.00 frames.], tot_loss[loss=0.4793, simple_loss=0.9585, pruned_loss=6.975, over 753492.84 frames.], batch size: 23, lr: 3.00e-03 2022-05-27 14:00:39,038 INFO [train.py:823] (0/4) Epoch 1, batch 200, loss[loss=0.397, simple_loss=0.794, pruned_loss=6.703, over 7289.00 frames.], tot_loss[loss=0.4579, simple_loss=0.9159, pruned_loss=6.901, over 903320.03 frames.], batch size: 19, lr: 3.00e-03 2022-05-27 14:01:18,153 INFO [train.py:823] (0/4) Epoch 1, batch 250, loss[loss=0.3687, simple_loss=0.7373, pruned_loss=6.596, over 7314.00 frames.], tot_loss[loss=0.443, simple_loss=0.886, pruned_loss=6.832, over 1015539.07 frames.], batch size: 17, lr: 3.00e-03 2022-05-27 14:01:57,476 INFO [train.py:823] (0/4) Epoch 1, batch 300, loss[loss=0.4031, simple_loss=0.8062, pruned_loss=6.76, over 7236.00 frames.], tot_loss[loss=0.4299, simple_loss=0.8597, pruned_loss=6.781, over 1107411.98 frames.], batch size: 24, lr: 3.00e-03 2022-05-27 14:02:36,753 INFO [train.py:823] (0/4) Epoch 1, batch 350, loss[loss=0.4124, simple_loss=0.8248, pruned_loss=6.577, over 6281.00 frames.], tot_loss[loss=0.4194, simple_loss=0.8389, pruned_loss=6.746, over 1177683.71 frames.], batch size: 34, lr: 3.00e-03 2022-05-27 14:03:16,146 INFO [train.py:823] (0/4) Epoch 1, batch 400, loss[loss=0.3645, simple_loss=0.729, pruned_loss=6.543, over 4486.00 frames.], tot_loss[loss=0.4098, simple_loss=0.8196, pruned_loss=6.718, over 1228099.51 frames.], batch size: 46, lr: 3.00e-03 2022-05-27 14:03:55,417 INFO [train.py:823] (0/4) Epoch 1, batch 450, loss[loss=0.3272, simple_loss=0.6544, pruned_loss=6.502, over 7204.00 frames.], tot_loss[loss=0.3969, simple_loss=0.7938, pruned_loss=6.7, over 1273735.05 frames.], batch size: 19, lr: 2.99e-03 2022-05-27 14:04:34,528 INFO [train.py:823] (0/4) Epoch 1, batch 500, loss[loss=0.3215, simple_loss=0.643, pruned_loss=6.599, over 7383.00 frames.], tot_loss[loss=0.3824, simple_loss=0.7648, pruned_loss=6.695, over 1308413.27 frames.], batch size: 19, lr: 2.99e-03 2022-05-27 14:05:13,689 INFO [train.py:823] (0/4) Epoch 1, batch 550, loss[loss=0.3086, simple_loss=0.6171, pruned_loss=6.823, over 7205.00 frames.], tot_loss[loss=0.3653, simple_loss=0.7306, pruned_loss=6.691, over 1329823.24 frames.], batch size: 25, lr: 2.99e-03 2022-05-27 14:05:53,147 INFO [train.py:823] (0/4) Epoch 1, batch 600, loss[loss=0.2655, simple_loss=0.5309, pruned_loss=6.652, over 7297.00 frames.], tot_loss[loss=0.3478, simple_loss=0.6956, pruned_loss=6.685, over 1347320.99 frames.], batch size: 19, lr: 2.99e-03 2022-05-27 14:06:31,944 INFO [train.py:823] (0/4) Epoch 1, batch 650, loss[loss=0.2681, simple_loss=0.5362, pruned_loss=6.723, over 7096.00 frames.], tot_loss[loss=0.3323, simple_loss=0.6647, pruned_loss=6.684, over 1361729.54 frames.], batch size: 19, lr: 2.99e-03 2022-05-27 14:07:11,384 INFO [train.py:823] (0/4) Epoch 1, batch 700, loss[loss=0.2399, simple_loss=0.4798, pruned_loss=6.591, over 7158.00 frames.], tot_loss[loss=0.3167, simple_loss=0.6333, pruned_loss=6.684, over 1372893.15 frames.], batch size: 17, lr: 2.99e-03 2022-05-27 14:07:50,412 INFO [train.py:823] (0/4) Epoch 1, batch 750, loss[loss=0.2288, simple_loss=0.4577, pruned_loss=6.561, over 6811.00 frames.], tot_loss[loss=0.3038, simple_loss=0.6075, pruned_loss=6.69, over 1385894.45 frames.], batch size: 15, lr: 2.98e-03 2022-05-27 14:08:29,953 INFO [train.py:823] (0/4) Epoch 1, batch 800, loss[loss=0.2548, simple_loss=0.5095, pruned_loss=6.748, over 7146.00 frames.], tot_loss[loss=0.2931, simple_loss=0.5862, pruned_loss=6.696, over 1391450.48 frames.], batch size: 23, lr: 2.98e-03 2022-05-27 14:09:08,828 INFO [train.py:823] (0/4) Epoch 1, batch 850, loss[loss=0.2248, simple_loss=0.4496, pruned_loss=6.67, over 7014.00 frames.], tot_loss[loss=0.2824, simple_loss=0.5649, pruned_loss=6.703, over 1399222.13 frames.], batch size: 16, lr: 2.98e-03 2022-05-27 14:09:47,659 INFO [train.py:823] (0/4) Epoch 1, batch 900, loss[loss=0.2193, simple_loss=0.4386, pruned_loss=6.709, over 7293.00 frames.], tot_loss[loss=0.2727, simple_loss=0.5454, pruned_loss=6.712, over 1402692.29 frames.], batch size: 17, lr: 2.98e-03 2022-05-27 14:10:26,106 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-1.pt 2022-05-27 14:10:41,066 INFO [train.py:823] (0/4) Epoch 2, batch 0, loss[loss=0.2314, simple_loss=0.4628, pruned_loss=6.728, over 7099.00 frames.], tot_loss[loss=0.2314, simple_loss=0.4628, pruned_loss=6.728, over 7099.00 frames.], batch size: 19, lr: 2.95e-03 2022-05-27 14:11:20,869 INFO [train.py:823] (0/4) Epoch 2, batch 50, loss[loss=0.2426, simple_loss=0.4851, pruned_loss=6.801, over 7380.00 frames.], tot_loss[loss=0.2286, simple_loss=0.4572, pruned_loss=6.724, over 322119.63 frames.], batch size: 21, lr: 2.95e-03 2022-05-27 14:11:59,981 INFO [train.py:823] (0/4) Epoch 2, batch 100, loss[loss=0.231, simple_loss=0.4619, pruned_loss=6.745, over 7042.00 frames.], tot_loss[loss=0.2273, simple_loss=0.4545, pruned_loss=6.73, over 564395.34 frames.], batch size: 26, lr: 2.95e-03 2022-05-27 14:12:39,654 INFO [train.py:823] (0/4) Epoch 2, batch 150, loss[loss=0.1997, simple_loss=0.3993, pruned_loss=6.744, over 7285.00 frames.], tot_loss[loss=0.2237, simple_loss=0.4475, pruned_loss=6.734, over 757905.99 frames.], batch size: 17, lr: 2.94e-03 2022-05-27 14:13:18,810 INFO [train.py:823] (0/4) Epoch 2, batch 200, loss[loss=0.2164, simple_loss=0.4327, pruned_loss=6.8, over 7093.00 frames.], tot_loss[loss=0.2217, simple_loss=0.4434, pruned_loss=6.739, over 905530.69 frames.], batch size: 18, lr: 2.94e-03 2022-05-27 14:13:58,349 INFO [train.py:823] (0/4) Epoch 2, batch 250, loss[loss=0.2057, simple_loss=0.4114, pruned_loss=6.637, over 7157.00 frames.], tot_loss[loss=0.2197, simple_loss=0.4393, pruned_loss=6.733, over 1016440.45 frames.], batch size: 17, lr: 2.94e-03 2022-05-27 14:14:37,463 INFO [train.py:823] (0/4) Epoch 2, batch 300, loss[loss=0.2001, simple_loss=0.4002, pruned_loss=6.651, over 7013.00 frames.], tot_loss[loss=0.2181, simple_loss=0.4362, pruned_loss=6.742, over 1106875.89 frames.], batch size: 16, lr: 2.93e-03 2022-05-27 14:15:20,808 INFO [train.py:823] (0/4) Epoch 2, batch 350, loss[loss=0.2226, simple_loss=0.4451, pruned_loss=6.77, over 7173.00 frames.], tot_loss[loss=0.2166, simple_loss=0.4332, pruned_loss=6.744, over 1174517.39 frames.], batch size: 23, lr: 2.93e-03 2022-05-27 14:15:59,768 INFO [train.py:823] (0/4) Epoch 2, batch 400, loss[loss=0.2286, simple_loss=0.4571, pruned_loss=6.837, over 7091.00 frames.], tot_loss[loss=0.2144, simple_loss=0.4288, pruned_loss=6.743, over 1224865.02 frames.], batch size: 18, lr: 2.93e-03 2022-05-27 14:16:39,198 INFO [train.py:823] (0/4) Epoch 2, batch 450, loss[loss=0.2202, simple_loss=0.4404, pruned_loss=6.832, over 7272.00 frames.], tot_loss[loss=0.2131, simple_loss=0.4261, pruned_loss=6.749, over 1265251.22 frames.], batch size: 21, lr: 2.92e-03 2022-05-27 14:17:18,175 INFO [train.py:823] (0/4) Epoch 2, batch 500, loss[loss=0.2202, simple_loss=0.4404, pruned_loss=6.796, over 6857.00 frames.], tot_loss[loss=0.2118, simple_loss=0.4236, pruned_loss=6.756, over 1301605.69 frames.], batch size: 29, lr: 2.92e-03 2022-05-27 14:17:57,425 INFO [train.py:823] (0/4) Epoch 2, batch 550, loss[loss=0.2221, simple_loss=0.4442, pruned_loss=6.765, over 4835.00 frames.], tot_loss[loss=0.2095, simple_loss=0.4189, pruned_loss=6.754, over 1322781.31 frames.], batch size: 46, lr: 2.92e-03 2022-05-27 14:18:36,778 INFO [train.py:823] (0/4) Epoch 2, batch 600, loss[loss=0.2075, simple_loss=0.4149, pruned_loss=6.795, over 7285.00 frames.], tot_loss[loss=0.2084, simple_loss=0.4167, pruned_loss=6.755, over 1340084.42 frames.], batch size: 21, lr: 2.91e-03 2022-05-27 14:19:16,678 INFO [train.py:823] (0/4) Epoch 2, batch 650, loss[loss=0.2154, simple_loss=0.4308, pruned_loss=6.856, over 7298.00 frames.], tot_loss[loss=0.2068, simple_loss=0.4137, pruned_loss=6.76, over 1358595.98 frames.], batch size: 22, lr: 2.91e-03 2022-05-27 14:19:56,885 INFO [train.py:823] (0/4) Epoch 2, batch 700, loss[loss=0.1686, simple_loss=0.3373, pruned_loss=6.608, over 7008.00 frames.], tot_loss[loss=0.2049, simple_loss=0.4099, pruned_loss=6.763, over 1374114.72 frames.], batch size: 17, lr: 2.90e-03 2022-05-27 14:20:36,958 INFO [train.py:823] (0/4) Epoch 2, batch 750, loss[loss=0.2012, simple_loss=0.4024, pruned_loss=6.827, over 7110.00 frames.], tot_loss[loss=0.2034, simple_loss=0.4067, pruned_loss=6.765, over 1382025.28 frames.], batch size: 20, lr: 2.90e-03 2022-05-27 14:21:16,497 INFO [train.py:823] (0/4) Epoch 2, batch 800, loss[loss=0.1955, simple_loss=0.391, pruned_loss=6.803, over 4442.00 frames.], tot_loss[loss=0.2016, simple_loss=0.4031, pruned_loss=6.767, over 1386964.03 frames.], batch size: 47, lr: 2.89e-03 2022-05-27 14:21:57,518 INFO [train.py:823] (0/4) Epoch 2, batch 850, loss[loss=0.1913, simple_loss=0.3826, pruned_loss=6.79, over 7200.00 frames.], tot_loss[loss=0.2, simple_loss=0.4001, pruned_loss=6.766, over 1390488.35 frames.], batch size: 20, lr: 2.89e-03 2022-05-27 14:22:36,474 INFO [train.py:823] (0/4) Epoch 2, batch 900, loss[loss=0.1688, simple_loss=0.3376, pruned_loss=6.604, over 7311.00 frames.], tot_loss[loss=0.1987, simple_loss=0.3975, pruned_loss=6.77, over 1394268.29 frames.], batch size: 18, lr: 2.89e-03 2022-05-27 14:23:15,074 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-2.pt 2022-05-27 14:23:29,985 INFO [train.py:823] (0/4) Epoch 3, batch 0, loss[loss=0.1785, simple_loss=0.357, pruned_loss=6.719, over 7289.00 frames.], tot_loss[loss=0.1785, simple_loss=0.357, pruned_loss=6.719, over 7289.00 frames.], batch size: 17, lr: 2.83e-03 2022-05-27 14:24:09,550 INFO [train.py:823] (0/4) Epoch 3, batch 50, loss[loss=0.2248, simple_loss=0.4496, pruned_loss=6.825, over 4797.00 frames.], tot_loss[loss=0.1897, simple_loss=0.3795, pruned_loss=6.752, over 319171.48 frames.], batch size: 46, lr: 2.82e-03 2022-05-27 14:24:48,967 INFO [train.py:823] (0/4) Epoch 3, batch 100, loss[loss=0.2049, simple_loss=0.4098, pruned_loss=6.867, over 6967.00 frames.], tot_loss[loss=0.1907, simple_loss=0.3813, pruned_loss=6.763, over 564980.65 frames.], batch size: 26, lr: 2.82e-03 2022-05-27 14:25:28,610 INFO [train.py:823] (0/4) Epoch 3, batch 150, loss[loss=0.2203, simple_loss=0.4407, pruned_loss=6.873, over 7367.00 frames.], tot_loss[loss=0.1904, simple_loss=0.3807, pruned_loss=6.774, over 755529.16 frames.], batch size: 20, lr: 2.81e-03 2022-05-27 14:26:08,170 INFO [train.py:823] (0/4) Epoch 3, batch 200, loss[loss=0.1718, simple_loss=0.3436, pruned_loss=6.725, over 7107.00 frames.], tot_loss[loss=0.1892, simple_loss=0.3785, pruned_loss=6.773, over 906194.11 frames.], batch size: 20, lr: 2.81e-03 2022-05-27 14:26:47,673 INFO [train.py:823] (0/4) Epoch 3, batch 250, loss[loss=0.1882, simple_loss=0.3764, pruned_loss=6.773, over 6953.00 frames.], tot_loss[loss=0.1886, simple_loss=0.3772, pruned_loss=6.778, over 1024131.91 frames.], batch size: 26, lr: 2.80e-03 2022-05-27 14:27:26,952 INFO [train.py:823] (0/4) Epoch 3, batch 300, loss[loss=0.1812, simple_loss=0.3624, pruned_loss=6.811, over 7396.00 frames.], tot_loss[loss=0.1875, simple_loss=0.375, pruned_loss=6.782, over 1114169.65 frames.], batch size: 19, lr: 2.80e-03 2022-05-27 14:28:06,710 INFO [train.py:823] (0/4) Epoch 3, batch 350, loss[loss=0.2093, simple_loss=0.4186, pruned_loss=6.873, over 7337.00 frames.], tot_loss[loss=0.1877, simple_loss=0.3755, pruned_loss=6.791, over 1186161.71 frames.], batch size: 23, lr: 2.79e-03 2022-05-27 14:28:45,741 INFO [train.py:823] (0/4) Epoch 3, batch 400, loss[loss=0.1582, simple_loss=0.3163, pruned_loss=6.676, over 7316.00 frames.], tot_loss[loss=0.1876, simple_loss=0.3751, pruned_loss=6.791, over 1240032.58 frames.], batch size: 18, lr: 2.79e-03 2022-05-27 14:29:24,485 INFO [train.py:823] (0/4) Epoch 3, batch 450, loss[loss=0.1767, simple_loss=0.3534, pruned_loss=6.779, over 7203.00 frames.], tot_loss[loss=0.1872, simple_loss=0.3744, pruned_loss=6.791, over 1273998.30 frames.], batch size: 18, lr: 2.78e-03 2022-05-27 14:30:03,734 INFO [train.py:823] (0/4) Epoch 3, batch 500, loss[loss=0.1633, simple_loss=0.3266, pruned_loss=6.667, over 7307.00 frames.], tot_loss[loss=0.1863, simple_loss=0.3725, pruned_loss=6.795, over 1305396.37 frames.], batch size: 18, lr: 2.77e-03 2022-05-27 14:30:42,937 INFO [train.py:823] (0/4) Epoch 3, batch 550, loss[loss=0.2121, simple_loss=0.4242, pruned_loss=6.907, over 7176.00 frames.], tot_loss[loss=0.186, simple_loss=0.372, pruned_loss=6.797, over 1333504.39 frames.], batch size: 21, lr: 2.77e-03 2022-05-27 14:31:21,884 INFO [train.py:823] (0/4) Epoch 3, batch 600, loss[loss=0.2052, simple_loss=0.4104, pruned_loss=6.905, over 7377.00 frames.], tot_loss[loss=0.1854, simple_loss=0.3709, pruned_loss=6.79, over 1346846.61 frames.], batch size: 20, lr: 2.76e-03 2022-05-27 14:32:01,057 INFO [train.py:823] (0/4) Epoch 3, batch 650, loss[loss=0.1842, simple_loss=0.3685, pruned_loss=6.803, over 4591.00 frames.], tot_loss[loss=0.1853, simple_loss=0.3707, pruned_loss=6.797, over 1363202.46 frames.], batch size: 47, lr: 2.76e-03 2022-05-27 14:32:40,557 INFO [train.py:823] (0/4) Epoch 3, batch 700, loss[loss=0.1765, simple_loss=0.353, pruned_loss=6.847, over 7306.00 frames.], tot_loss[loss=0.1846, simple_loss=0.3692, pruned_loss=6.797, over 1376018.44 frames.], batch size: 22, lr: 2.75e-03 2022-05-27 14:33:19,791 INFO [train.py:823] (0/4) Epoch 3, batch 750, loss[loss=0.1747, simple_loss=0.3495, pruned_loss=6.757, over 7177.00 frames.], tot_loss[loss=0.1832, simple_loss=0.3664, pruned_loss=6.797, over 1384610.56 frames.], batch size: 19, lr: 2.75e-03 2022-05-27 14:33:58,484 INFO [train.py:823] (0/4) Epoch 3, batch 800, loss[loss=0.1958, simple_loss=0.3916, pruned_loss=6.837, over 7425.00 frames.], tot_loss[loss=0.1828, simple_loss=0.3656, pruned_loss=6.794, over 1394736.72 frames.], batch size: 22, lr: 2.74e-03 2022-05-27 14:34:38,086 INFO [train.py:823] (0/4) Epoch 3, batch 850, loss[loss=0.1872, simple_loss=0.3745, pruned_loss=6.794, over 7098.00 frames.], tot_loss[loss=0.1827, simple_loss=0.3654, pruned_loss=6.796, over 1397192.95 frames.], batch size: 19, lr: 2.74e-03 2022-05-27 14:35:16,877 INFO [train.py:823] (0/4) Epoch 3, batch 900, loss[loss=0.1899, simple_loss=0.3799, pruned_loss=6.867, over 4963.00 frames.], tot_loss[loss=0.1829, simple_loss=0.3657, pruned_loss=6.8, over 1392641.94 frames.], batch size: 47, lr: 2.73e-03 2022-05-27 14:35:55,757 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-3.pt 2022-05-27 14:36:10,828 INFO [train.py:823] (0/4) Epoch 4, batch 0, loss[loss=0.1783, simple_loss=0.3567, pruned_loss=6.828, over 7089.00 frames.], tot_loss[loss=0.1783, simple_loss=0.3567, pruned_loss=6.828, over 7089.00 frames.], batch size: 19, lr: 2.64e-03 2022-05-27 14:36:49,882 INFO [train.py:823] (0/4) Epoch 4, batch 50, loss[loss=0.1657, simple_loss=0.3313, pruned_loss=6.733, over 7020.00 frames.], tot_loss[loss=0.1707, simple_loss=0.3413, pruned_loss=6.783, over 319743.39 frames.], batch size: 17, lr: 2.64e-03 2022-05-27 14:37:30,276 INFO [train.py:823] (0/4) Epoch 4, batch 100, loss[loss=0.1806, simple_loss=0.3612, pruned_loss=6.803, over 7368.00 frames.], tot_loss[loss=0.1738, simple_loss=0.3477, pruned_loss=6.795, over 564735.01 frames.], batch size: 21, lr: 2.63e-03 2022-05-27 14:38:10,757 INFO [train.py:823] (0/4) Epoch 4, batch 150, loss[loss=0.171, simple_loss=0.3421, pruned_loss=6.759, over 7154.00 frames.], tot_loss[loss=0.1751, simple_loss=0.3501, pruned_loss=6.801, over 751645.67 frames.], batch size: 17, lr: 2.63e-03 2022-05-27 14:38:51,380 INFO [train.py:823] (0/4) Epoch 4, batch 200, loss[loss=0.2676, simple_loss=0.3325, pruned_loss=1.014, over 7171.00 frames.], tot_loss[loss=0.2719, simple_loss=0.3626, pruned_loss=4.92, over 904176.39 frames.], batch size: 18, lr: 2.62e-03 2022-05-27 14:39:31,800 INFO [train.py:823] (0/4) Epoch 4, batch 250, loss[loss=0.2622, simple_loss=0.3718, pruned_loss=0.7624, over 7381.00 frames.], tot_loss[loss=0.2697, simple_loss=0.3597, pruned_loss=3.658, over 1023856.21 frames.], batch size: 21, lr: 2.62e-03 2022-05-27 14:40:11,100 INFO [train.py:823] (0/4) Epoch 4, batch 300, loss[loss=0.2223, simple_loss=0.3528, pruned_loss=0.4588, over 7193.00 frames.], tot_loss[loss=0.2594, simple_loss=0.3576, pruned_loss=2.789, over 1108770.09 frames.], batch size: 20, lr: 2.61e-03 2022-05-27 14:40:49,899 INFO [train.py:823] (0/4) Epoch 4, batch 350, loss[loss=0.2331, simple_loss=0.393, pruned_loss=0.3662, over 7143.00 frames.], tot_loss[loss=0.2475, simple_loss=0.3563, pruned_loss=2.15, over 1174664.22 frames.], batch size: 23, lr: 2.60e-03 2022-05-27 14:41:29,698 INFO [train.py:823] (0/4) Epoch 4, batch 400, loss[loss=0.2215, simple_loss=0.3845, pruned_loss=0.2926, over 7196.00 frames.], tot_loss[loss=0.236, simple_loss=0.3545, pruned_loss=1.672, over 1227816.59 frames.], batch size: 25, lr: 2.60e-03 2022-05-27 14:42:08,255 INFO [train.py:823] (0/4) Epoch 4, batch 450, loss[loss=0.1971, simple_loss=0.3448, pruned_loss=0.2468, over 7153.00 frames.], tot_loss[loss=0.2272, simple_loss=0.3541, pruned_loss=1.318, over 1269905.00 frames.], batch size: 17, lr: 2.59e-03 2022-05-27 14:42:47,478 INFO [train.py:823] (0/4) Epoch 4, batch 500, loss[loss=0.1825, simple_loss=0.3308, pruned_loss=0.1709, over 7197.00 frames.], tot_loss[loss=0.2196, simple_loss=0.353, pruned_loss=1.048, over 1306807.02 frames.], batch size: 25, lr: 2.59e-03 2022-05-27 14:43:26,637 INFO [train.py:823] (0/4) Epoch 4, batch 550, loss[loss=0.1733, simple_loss=0.3145, pruned_loss=0.1603, over 7393.00 frames.], tot_loss[loss=0.2123, simple_loss=0.3499, pruned_loss=0.8441, over 1333739.50 frames.], batch size: 19, lr: 2.58e-03 2022-05-27 14:44:06,020 INFO [train.py:823] (0/4) Epoch 4, batch 600, loss[loss=0.2177, simple_loss=0.3885, pruned_loss=0.2342, over 7182.00 frames.], tot_loss[loss=0.2078, simple_loss=0.3496, pruned_loss=0.6901, over 1355588.96 frames.], batch size: 21, lr: 2.57e-03 2022-05-27 14:44:44,732 INFO [train.py:823] (0/4) Epoch 4, batch 650, loss[loss=0.1919, simple_loss=0.3475, pruned_loss=0.1811, over 7378.00 frames.], tot_loss[loss=0.204, simple_loss=0.349, pruned_loss=0.5719, over 1371038.35 frames.], batch size: 20, lr: 2.57e-03 2022-05-27 14:45:23,897 INFO [train.py:823] (0/4) Epoch 4, batch 700, loss[loss=0.2006, simple_loss=0.3638, pruned_loss=0.1869, over 5308.00 frames.], tot_loss[loss=0.2037, simple_loss=0.3529, pruned_loss=0.4871, over 1376945.37 frames.], batch size: 47, lr: 2.56e-03 2022-05-27 14:46:02,608 INFO [train.py:823] (0/4) Epoch 4, batch 750, loss[loss=0.1876, simple_loss=0.3426, pruned_loss=0.1634, over 7095.00 frames.], tot_loss[loss=0.2011, simple_loss=0.3523, pruned_loss=0.4162, over 1385200.14 frames.], batch size: 19, lr: 2.56e-03 2022-05-27 14:46:41,995 INFO [train.py:823] (0/4) Epoch 4, batch 800, loss[loss=0.1606, simple_loss=0.295, pruned_loss=0.1316, over 7011.00 frames.], tot_loss[loss=0.1982, simple_loss=0.3502, pruned_loss=0.3602, over 1386689.69 frames.], batch size: 17, lr: 2.55e-03 2022-05-27 14:47:21,026 INFO [train.py:823] (0/4) Epoch 4, batch 850, loss[loss=0.2026, simple_loss=0.3708, pruned_loss=0.1719, over 7296.00 frames.], tot_loss[loss=0.1957, simple_loss=0.3486, pruned_loss=0.3146, over 1392078.99 frames.], batch size: 22, lr: 2.54e-03 2022-05-27 14:47:59,964 INFO [train.py:823] (0/4) Epoch 4, batch 900, loss[loss=0.2022, simple_loss=0.3672, pruned_loss=0.1863, over 7195.00 frames.], tot_loss[loss=0.1938, simple_loss=0.3471, pruned_loss=0.2806, over 1388559.96 frames.], batch size: 18, lr: 2.54e-03 2022-05-27 14:48:39,135 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-4.pt 2022-05-27 14:48:51,282 INFO [train.py:823] (0/4) Epoch 5, batch 0, loss[loss=0.1905, simple_loss=0.3506, pruned_loss=0.1525, over 7321.00 frames.], tot_loss[loss=0.1905, simple_loss=0.3506, pruned_loss=0.1525, over 7321.00 frames.], batch size: 23, lr: 2.44e-03 2022-05-27 14:49:30,534 INFO [train.py:823] (0/4) Epoch 5, batch 50, loss[loss=0.2024, simple_loss=0.3682, pruned_loss=0.1825, over 6978.00 frames.], tot_loss[loss=0.183, simple_loss=0.3369, pruned_loss=0.1457, over 326001.34 frames.], batch size: 26, lr: 2.44e-03 2022-05-27 14:50:10,131 INFO [train.py:823] (0/4) Epoch 5, batch 100, loss[loss=0.2063, simple_loss=0.3787, pruned_loss=0.1695, over 7112.00 frames.], tot_loss[loss=0.1826, simple_loss=0.3365, pruned_loss=0.1437, over 570810.90 frames.], batch size: 20, lr: 2.43e-03 2022-05-27 14:50:49,492 INFO [train.py:823] (0/4) Epoch 5, batch 150, loss[loss=0.1988, simple_loss=0.3642, pruned_loss=0.1667, over 7375.00 frames.], tot_loss[loss=0.1825, simple_loss=0.3362, pruned_loss=0.1436, over 759066.29 frames.], batch size: 20, lr: 2.42e-03 2022-05-27 14:51:28,462 INFO [train.py:823] (0/4) Epoch 5, batch 200, loss[loss=0.2027, simple_loss=0.3718, pruned_loss=0.1681, over 7182.00 frames.], tot_loss[loss=0.1829, simple_loss=0.3371, pruned_loss=0.144, over 905889.99 frames.], batch size: 22, lr: 2.42e-03 2022-05-27 14:52:07,828 INFO [train.py:823] (0/4) Epoch 5, batch 250, loss[loss=0.1849, simple_loss=0.3413, pruned_loss=0.1424, over 5280.00 frames.], tot_loss[loss=0.1823, simple_loss=0.336, pruned_loss=0.143, over 1014645.04 frames.], batch size: 46, lr: 2.41e-03 2022-05-27 14:52:46,744 INFO [train.py:823] (0/4) Epoch 5, batch 300, loss[loss=0.1908, simple_loss=0.3552, pruned_loss=0.1314, over 7141.00 frames.], tot_loss[loss=0.183, simple_loss=0.3374, pruned_loss=0.1432, over 1105555.23 frames.], batch size: 23, lr: 2.41e-03 2022-05-27 14:53:26,226 INFO [train.py:823] (0/4) Epoch 5, batch 350, loss[loss=0.1956, simple_loss=0.3616, pruned_loss=0.1485, over 7246.00 frames.], tot_loss[loss=0.1832, simple_loss=0.3379, pruned_loss=0.1425, over 1174961.11 frames.], batch size: 24, lr: 2.40e-03 2022-05-27 14:54:05,561 INFO [train.py:823] (0/4) Epoch 5, batch 400, loss[loss=0.174, simple_loss=0.3196, pruned_loss=0.1424, over 7017.00 frames.], tot_loss[loss=0.1824, simple_loss=0.3367, pruned_loss=0.1406, over 1234678.67 frames.], batch size: 17, lr: 2.39e-03 2022-05-27 14:54:45,108 INFO [train.py:823] (0/4) Epoch 5, batch 450, loss[loss=0.202, simple_loss=0.3718, pruned_loss=0.1605, over 7022.00 frames.], tot_loss[loss=0.1814, simple_loss=0.3351, pruned_loss=0.1381, over 1271520.51 frames.], batch size: 26, lr: 2.39e-03 2022-05-27 14:55:24,632 INFO [train.py:823] (0/4) Epoch 5, batch 500, loss[loss=0.1633, simple_loss=0.306, pruned_loss=0.1032, over 7194.00 frames.], tot_loss[loss=0.1811, simple_loss=0.3349, pruned_loss=0.1365, over 1306123.29 frames.], batch size: 19, lr: 2.38e-03 2022-05-27 14:56:03,723 INFO [train.py:823] (0/4) Epoch 5, batch 550, loss[loss=0.1833, simple_loss=0.3402, pruned_loss=0.1315, over 6897.00 frames.], tot_loss[loss=0.1809, simple_loss=0.3347, pruned_loss=0.1355, over 1331245.78 frames.], batch size: 29, lr: 2.38e-03 2022-05-27 14:56:42,868 INFO [train.py:823] (0/4) Epoch 5, batch 600, loss[loss=0.214, simple_loss=0.3926, pruned_loss=0.1772, over 6616.00 frames.], tot_loss[loss=0.1812, simple_loss=0.3353, pruned_loss=0.1355, over 1349282.70 frames.], batch size: 34, lr: 2.37e-03 2022-05-27 14:57:22,157 INFO [train.py:823] (0/4) Epoch 5, batch 650, loss[loss=0.1921, simple_loss=0.3571, pruned_loss=0.1359, over 7284.00 frames.], tot_loss[loss=0.1804, simple_loss=0.3341, pruned_loss=0.1334, over 1364778.27 frames.], batch size: 21, lr: 2.37e-03 2022-05-27 14:58:00,829 INFO [train.py:823] (0/4) Epoch 5, batch 700, loss[loss=0.2095, simple_loss=0.3834, pruned_loss=0.1778, over 6989.00 frames.], tot_loss[loss=0.1801, simple_loss=0.3337, pruned_loss=0.1326, over 1374104.70 frames.], batch size: 26, lr: 2.36e-03 2022-05-27 14:58:39,874 INFO [train.py:823] (0/4) Epoch 5, batch 750, loss[loss=0.1948, simple_loss=0.3592, pruned_loss=0.1517, over 7129.00 frames.], tot_loss[loss=0.1808, simple_loss=0.335, pruned_loss=0.1327, over 1381650.66 frames.], batch size: 23, lr: 2.35e-03 2022-05-27 14:59:18,662 INFO [train.py:823] (0/4) Epoch 5, batch 800, loss[loss=0.1986, simple_loss=0.3596, pruned_loss=0.1879, over 4870.00 frames.], tot_loss[loss=0.1801, simple_loss=0.3338, pruned_loss=0.1315, over 1391075.86 frames.], batch size: 46, lr: 2.35e-03 2022-05-27 14:59:59,052 INFO [train.py:823] (0/4) Epoch 5, batch 850, loss[loss=0.1824, simple_loss=0.3382, pruned_loss=0.1332, over 7155.00 frames.], tot_loss[loss=0.1789, simple_loss=0.332, pruned_loss=0.129, over 1397986.52 frames.], batch size: 17, lr: 2.34e-03 2022-05-27 15:00:37,904 INFO [train.py:823] (0/4) Epoch 5, batch 900, loss[loss=0.1898, simple_loss=0.3528, pruned_loss=0.134, over 6846.00 frames.], tot_loss[loss=0.1786, simple_loss=0.3317, pruned_loss=0.1277, over 1400068.50 frames.], batch size: 29, lr: 2.34e-03 2022-05-27 15:01:19,551 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-5.pt 2022-05-27 15:01:33,747 INFO [train.py:823] (0/4) Epoch 6, batch 0, loss[loss=0.1982, simple_loss=0.3685, pruned_loss=0.1395, over 7163.00 frames.], tot_loss[loss=0.1982, simple_loss=0.3685, pruned_loss=0.1395, over 7163.00 frames.], batch size: 22, lr: 2.24e-03 2022-05-27 15:02:12,492 INFO [train.py:823] (0/4) Epoch 6, batch 50, loss[loss=0.1803, simple_loss=0.3385, pruned_loss=0.1106, over 7205.00 frames.], tot_loss[loss=0.1764, simple_loss=0.3289, pruned_loss=0.1195, over 319721.85 frames.], batch size: 21, lr: 2.23e-03 2022-05-27 15:02:52,426 INFO [train.py:823] (0/4) Epoch 6, batch 100, loss[loss=0.1776, simple_loss=0.3318, pruned_loss=0.1173, over 7241.00 frames.], tot_loss[loss=0.1725, simple_loss=0.322, pruned_loss=0.1151, over 565386.28 frames.], batch size: 24, lr: 2.23e-03 2022-05-27 15:03:32,859 INFO [train.py:823] (0/4) Epoch 6, batch 150, loss[loss=0.1829, simple_loss=0.3396, pruned_loss=0.1306, over 7294.00 frames.], tot_loss[loss=0.1723, simple_loss=0.3217, pruned_loss=0.1143, over 754866.26 frames.], batch size: 19, lr: 2.22e-03 2022-05-27 15:04:12,199 INFO [train.py:823] (0/4) Epoch 6, batch 200, loss[loss=0.1791, simple_loss=0.3353, pruned_loss=0.1146, over 7191.00 frames.], tot_loss[loss=0.1731, simple_loss=0.3231, pruned_loss=0.1153, over 901340.08 frames.], batch size: 25, lr: 2.22e-03 2022-05-27 15:04:50,842 INFO [train.py:823] (0/4) Epoch 6, batch 250, loss[loss=0.1991, simple_loss=0.3672, pruned_loss=0.1547, over 6523.00 frames.], tot_loss[loss=0.1739, simple_loss=0.3246, pruned_loss=0.1157, over 1017073.62 frames.], batch size: 34, lr: 2.21e-03 2022-05-27 15:05:29,766 INFO [train.py:823] (0/4) Epoch 6, batch 300, loss[loss=0.1825, simple_loss=0.3373, pruned_loss=0.1385, over 7197.00 frames.], tot_loss[loss=0.1745, simple_loss=0.3257, pruned_loss=0.1162, over 1107695.46 frames.], batch size: 20, lr: 2.21e-03 2022-05-27 15:06:08,806 INFO [train.py:823] (0/4) Epoch 6, batch 350, loss[loss=0.1555, simple_loss=0.2931, pruned_loss=0.08907, over 7103.00 frames.], tot_loss[loss=0.1739, simple_loss=0.3248, pruned_loss=0.115, over 1179330.01 frames.], batch size: 18, lr: 2.20e-03 2022-05-27 15:06:48,039 INFO [train.py:823] (0/4) Epoch 6, batch 400, loss[loss=0.1731, simple_loss=0.3247, pruned_loss=0.1077, over 7174.00 frames.], tot_loss[loss=0.1724, simple_loss=0.3222, pruned_loss=0.1134, over 1234780.64 frames.], batch size: 22, lr: 2.19e-03 2022-05-27 15:07:26,428 INFO [train.py:823] (0/4) Epoch 6, batch 450, loss[loss=0.1816, simple_loss=0.3404, pruned_loss=0.1139, over 6679.00 frames.], tot_loss[loss=0.1728, simple_loss=0.3228, pruned_loss=0.1138, over 1268213.95 frames.], batch size: 34, lr: 2.19e-03 2022-05-27 15:08:05,524 INFO [train.py:823] (0/4) Epoch 6, batch 500, loss[loss=0.1877, simple_loss=0.3478, pruned_loss=0.1377, over 7139.00 frames.], tot_loss[loss=0.1727, simple_loss=0.3227, pruned_loss=0.1135, over 1297666.35 frames.], batch size: 23, lr: 2.18e-03 2022-05-27 15:08:44,650 INFO [train.py:823] (0/4) Epoch 6, batch 550, loss[loss=0.1544, simple_loss=0.2907, pruned_loss=0.09042, over 7094.00 frames.], tot_loss[loss=0.1728, simple_loss=0.3228, pruned_loss=0.1137, over 1325109.64 frames.], batch size: 18, lr: 2.18e-03 2022-05-27 15:09:24,175 INFO [train.py:823] (0/4) Epoch 6, batch 600, loss[loss=0.1615, simple_loss=0.3021, pruned_loss=0.1043, over 7082.00 frames.], tot_loss[loss=0.1723, simple_loss=0.322, pruned_loss=0.1134, over 1342810.58 frames.], batch size: 18, lr: 2.17e-03 2022-05-27 15:10:02,616 INFO [train.py:823] (0/4) Epoch 6, batch 650, loss[loss=0.1602, simple_loss=0.3005, pruned_loss=0.09937, over 7405.00 frames.], tot_loss[loss=0.1715, simple_loss=0.3205, pruned_loss=0.1121, over 1359334.44 frames.], batch size: 19, lr: 2.17e-03 2022-05-27 15:10:41,867 INFO [train.py:823] (0/4) Epoch 6, batch 700, loss[loss=0.1505, simple_loss=0.2852, pruned_loss=0.07861, over 7204.00 frames.], tot_loss[loss=0.1717, simple_loss=0.321, pruned_loss=0.1119, over 1374894.53 frames.], batch size: 19, lr: 2.16e-03 2022-05-27 15:11:20,981 INFO [train.py:823] (0/4) Epoch 6, batch 750, loss[loss=0.1646, simple_loss=0.3078, pruned_loss=0.1075, over 7103.00 frames.], tot_loss[loss=0.172, simple_loss=0.3217, pruned_loss=0.1118, over 1383355.77 frames.], batch size: 19, lr: 2.16e-03 2022-05-27 15:12:00,593 INFO [train.py:823] (0/4) Epoch 6, batch 800, loss[loss=0.1508, simple_loss=0.2866, pruned_loss=0.07521, over 7017.00 frames.], tot_loss[loss=0.1708, simple_loss=0.3195, pruned_loss=0.1101, over 1390210.15 frames.], batch size: 16, lr: 2.15e-03 2022-05-27 15:12:39,769 INFO [train.py:823] (0/4) Epoch 6, batch 850, loss[loss=0.1612, simple_loss=0.2998, pruned_loss=0.1134, over 6804.00 frames.], tot_loss[loss=0.171, simple_loss=0.3199, pruned_loss=0.1103, over 1394704.25 frames.], batch size: 15, lr: 2.15e-03 2022-05-27 15:13:19,342 INFO [train.py:823] (0/4) Epoch 6, batch 900, loss[loss=0.1611, simple_loss=0.3013, pruned_loss=0.1047, over 7290.00 frames.], tot_loss[loss=0.1701, simple_loss=0.3184, pruned_loss=0.1092, over 1397222.65 frames.], batch size: 17, lr: 2.14e-03 2022-05-27 15:13:58,414 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-6.pt 2022-05-27 15:14:12,732 INFO [train.py:823] (0/4) Epoch 7, batch 0, loss[loss=0.1551, simple_loss=0.2917, pruned_loss=0.09202, over 7099.00 frames.], tot_loss[loss=0.1551, simple_loss=0.2917, pruned_loss=0.09202, over 7099.00 frames.], batch size: 19, lr: 2.05e-03 2022-05-27 15:14:52,608 INFO [train.py:823] (0/4) Epoch 7, batch 50, loss[loss=0.1517, simple_loss=0.285, pruned_loss=0.09211, over 7271.00 frames.], tot_loss[loss=0.1632, simple_loss=0.3068, pruned_loss=0.09816, over 322497.52 frames.], batch size: 16, lr: 2.04e-03 2022-05-27 15:15:31,797 INFO [train.py:823] (0/4) Epoch 7, batch 100, loss[loss=0.1696, simple_loss=0.3193, pruned_loss=0.09997, over 7106.00 frames.], tot_loss[loss=0.1641, simple_loss=0.3082, pruned_loss=0.09997, over 562126.43 frames.], batch size: 20, lr: 2.04e-03 2022-05-27 15:16:10,848 INFO [train.py:823] (0/4) Epoch 7, batch 150, loss[loss=0.1775, simple_loss=0.3319, pruned_loss=0.116, over 7366.00 frames.], tot_loss[loss=0.165, simple_loss=0.3099, pruned_loss=0.1001, over 752329.94 frames.], batch size: 21, lr: 2.03e-03 2022-05-27 15:16:50,090 INFO [train.py:823] (0/4) Epoch 7, batch 200, loss[loss=0.177, simple_loss=0.3321, pruned_loss=0.1096, over 7056.00 frames.], tot_loss[loss=0.1664, simple_loss=0.3124, pruned_loss=0.1025, over 904106.01 frames.], batch size: 26, lr: 2.03e-03 2022-05-27 15:17:29,152 INFO [train.py:823] (0/4) Epoch 7, batch 250, loss[loss=0.185, simple_loss=0.3452, pruned_loss=0.1234, over 7305.00 frames.], tot_loss[loss=0.1661, simple_loss=0.3118, pruned_loss=0.1019, over 1019288.64 frames.], batch size: 22, lr: 2.02e-03 2022-05-27 15:18:07,887 INFO [train.py:823] (0/4) Epoch 7, batch 300, loss[loss=0.1448, simple_loss=0.2735, pruned_loss=0.08065, over 7167.00 frames.], tot_loss[loss=0.1661, simple_loss=0.3119, pruned_loss=0.1015, over 1108352.01 frames.], batch size: 17, lr: 2.02e-03 2022-05-27 15:18:47,543 INFO [train.py:823] (0/4) Epoch 7, batch 350, loss[loss=0.3083, simple_loss=0.3493, pruned_loss=0.1336, over 7298.00 frames.], tot_loss[loss=0.1961, simple_loss=0.3149, pruned_loss=0.1057, over 1176384.51 frames.], batch size: 19, lr: 2.01e-03 2022-05-27 15:19:26,414 INFO [train.py:823] (0/4) Epoch 7, batch 400, loss[loss=0.3036, simple_loss=0.3523, pruned_loss=0.1275, over 7329.00 frames.], tot_loss[loss=0.2153, simple_loss=0.3168, pruned_loss=0.1067, over 1231323.26 frames.], batch size: 23, lr: 2.01e-03 2022-05-27 15:20:05,959 INFO [train.py:823] (0/4) Epoch 7, batch 450, loss[loss=0.274, simple_loss=0.3348, pruned_loss=0.1066, over 7170.00 frames.], tot_loss[loss=0.2278, simple_loss=0.3175, pruned_loss=0.1067, over 1268117.33 frames.], batch size: 22, lr: 2.00e-03 2022-05-27 15:20:45,033 INFO [train.py:823] (0/4) Epoch 7, batch 500, loss[loss=0.2555, simple_loss=0.3192, pruned_loss=0.09589, over 6981.00 frames.], tot_loss[loss=0.2346, simple_loss=0.3173, pruned_loss=0.1045, over 1301981.99 frames.], batch size: 26, lr: 2.00e-03 2022-05-27 15:21:24,293 INFO [train.py:823] (0/4) Epoch 7, batch 550, loss[loss=0.2894, simple_loss=0.3468, pruned_loss=0.1161, over 6613.00 frames.], tot_loss[loss=0.2384, simple_loss=0.3157, pruned_loss=0.1024, over 1326044.12 frames.], batch size: 34, lr: 1.99e-03 2022-05-27 15:22:03,239 INFO [train.py:823] (0/4) Epoch 7, batch 600, loss[loss=0.3098, simple_loss=0.3529, pruned_loss=0.1333, over 7377.00 frames.], tot_loss[loss=0.2438, simple_loss=0.3163, pruned_loss=0.1023, over 1344122.94 frames.], batch size: 21, lr: 1.99e-03 2022-05-27 15:22:42,687 INFO [train.py:823] (0/4) Epoch 7, batch 650, loss[loss=0.2849, simple_loss=0.3357, pruned_loss=0.1171, over 7111.00 frames.], tot_loss[loss=0.2474, simple_loss=0.317, pruned_loss=0.1018, over 1360469.10 frames.], batch size: 20, lr: 1.98e-03 2022-05-27 15:23:24,559 INFO [train.py:823] (0/4) Epoch 7, batch 700, loss[loss=0.2111, simple_loss=0.2869, pruned_loss=0.06758, over 7097.00 frames.], tot_loss[loss=0.2492, simple_loss=0.3162, pruned_loss=0.101, over 1369613.37 frames.], batch size: 18, lr: 1.98e-03 2022-05-27 15:24:03,785 INFO [train.py:823] (0/4) Epoch 7, batch 750, loss[loss=0.2469, simple_loss=0.315, pruned_loss=0.08942, over 7053.00 frames.], tot_loss[loss=0.2505, simple_loss=0.3158, pruned_loss=0.1002, over 1378307.53 frames.], batch size: 26, lr: 1.97e-03 2022-05-27 15:24:43,876 INFO [train.py:823] (0/4) Epoch 7, batch 800, loss[loss=0.2524, simple_loss=0.3128, pruned_loss=0.09597, over 7183.00 frames.], tot_loss[loss=0.2516, simple_loss=0.3164, pruned_loss=0.09935, over 1389336.64 frames.], batch size: 19, lr: 1.97e-03 2022-05-27 15:25:23,410 INFO [train.py:823] (0/4) Epoch 7, batch 850, loss[loss=0.2839, simple_loss=0.344, pruned_loss=0.1119, over 7375.00 frames.], tot_loss[loss=0.2532, simple_loss=0.317, pruned_loss=0.09926, over 1389686.29 frames.], batch size: 21, lr: 1.97e-03 2022-05-27 15:26:02,042 INFO [train.py:823] (0/4) Epoch 7, batch 900, loss[loss=0.2652, simple_loss=0.3317, pruned_loss=0.09932, over 6950.00 frames.], tot_loss[loss=0.2555, simple_loss=0.3183, pruned_loss=0.09997, over 1392275.27 frames.], batch size: 29, lr: 1.96e-03 2022-05-27 15:26:41,748 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-7.pt 2022-05-27 15:26:53,906 INFO [train.py:823] (0/4) Epoch 8, batch 0, loss[loss=0.2559, simple_loss=0.318, pruned_loss=0.09691, over 7418.00 frames.], tot_loss[loss=0.2559, simple_loss=0.318, pruned_loss=0.09691, over 7418.00 frames.], batch size: 22, lr: 1.88e-03 2022-05-27 15:27:33,930 INFO [train.py:823] (0/4) Epoch 8, batch 50, loss[loss=0.2288, simple_loss=0.313, pruned_loss=0.07226, over 7210.00 frames.], tot_loss[loss=0.2485, simple_loss=0.3151, pruned_loss=0.091, over 320597.49 frames.], batch size: 24, lr: 1.87e-03 2022-05-27 15:28:13,342 INFO [train.py:823] (0/4) Epoch 8, batch 100, loss[loss=0.232, simple_loss=0.2899, pruned_loss=0.08704, over 7034.00 frames.], tot_loss[loss=0.252, simple_loss=0.3174, pruned_loss=0.09326, over 564097.10 frames.], batch size: 17, lr: 1.87e-03 2022-05-27 15:28:52,008 INFO [train.py:823] (0/4) Epoch 8, batch 150, loss[loss=0.2281, simple_loss=0.298, pruned_loss=0.07909, over 7278.00 frames.], tot_loss[loss=0.2481, simple_loss=0.3137, pruned_loss=0.0913, over 753242.75 frames.], batch size: 20, lr: 1.86e-03 2022-05-27 15:29:31,405 INFO [train.py:823] (0/4) Epoch 8, batch 200, loss[loss=0.264, simple_loss=0.3157, pruned_loss=0.1062, over 7015.00 frames.], tot_loss[loss=0.2459, simple_loss=0.3121, pruned_loss=0.08982, over 898578.19 frames.], batch size: 16, lr: 1.86e-03 2022-05-27 15:30:10,538 INFO [train.py:823] (0/4) Epoch 8, batch 250, loss[loss=0.2424, simple_loss=0.3133, pruned_loss=0.08569, over 7149.00 frames.], tot_loss[loss=0.2435, simple_loss=0.3098, pruned_loss=0.08866, over 1012516.25 frames.], batch size: 23, lr: 1.85e-03 2022-05-27 15:30:49,824 INFO [train.py:823] (0/4) Epoch 8, batch 300, loss[loss=0.2698, simple_loss=0.336, pruned_loss=0.1018, over 7387.00 frames.], tot_loss[loss=0.245, simple_loss=0.3109, pruned_loss=0.08949, over 1105751.87 frames.], batch size: 19, lr: 1.85e-03 2022-05-27 15:31:28,735 INFO [train.py:823] (0/4) Epoch 8, batch 350, loss[loss=0.1984, simple_loss=0.2628, pruned_loss=0.06696, over 7010.00 frames.], tot_loss[loss=0.2444, simple_loss=0.3101, pruned_loss=0.08933, over 1166246.19 frames.], batch size: 16, lr: 1.85e-03 2022-05-27 15:32:08,143 INFO [train.py:823] (0/4) Epoch 8, batch 400, loss[loss=0.2636, simple_loss=0.3329, pruned_loss=0.09711, over 7173.00 frames.], tot_loss[loss=0.2453, simple_loss=0.3112, pruned_loss=0.08968, over 1221526.97 frames.], batch size: 22, lr: 1.84e-03 2022-05-27 15:32:47,379 INFO [train.py:823] (0/4) Epoch 8, batch 450, loss[loss=0.269, simple_loss=0.3336, pruned_loss=0.1022, over 6560.00 frames.], tot_loss[loss=0.2453, simple_loss=0.3114, pruned_loss=0.08959, over 1264366.01 frames.], batch size: 34, lr: 1.84e-03 2022-05-27 15:33:26,784 INFO [train.py:823] (0/4) Epoch 8, batch 500, loss[loss=0.1909, simple_loss=0.257, pruned_loss=0.0624, over 7296.00 frames.], tot_loss[loss=0.2448, simple_loss=0.3106, pruned_loss=0.08947, over 1300633.97 frames.], batch size: 17, lr: 1.83e-03 2022-05-27 15:34:05,412 INFO [train.py:823] (0/4) Epoch 8, batch 550, loss[loss=0.302, simple_loss=0.3475, pruned_loss=0.1283, over 7182.00 frames.], tot_loss[loss=0.245, simple_loss=0.3113, pruned_loss=0.0893, over 1325280.64 frames.], batch size: 22, lr: 1.83e-03 2022-05-27 15:34:44,893 INFO [train.py:823] (0/4) Epoch 8, batch 600, loss[loss=0.2435, simple_loss=0.3085, pruned_loss=0.08925, over 7045.00 frames.], tot_loss[loss=0.2444, simple_loss=0.3111, pruned_loss=0.08892, over 1343094.23 frames.], batch size: 17, lr: 1.82e-03 2022-05-27 15:35:24,015 INFO [train.py:823] (0/4) Epoch 8, batch 650, loss[loss=0.2606, simple_loss=0.3369, pruned_loss=0.09215, over 6992.00 frames.], tot_loss[loss=0.2442, simple_loss=0.3112, pruned_loss=0.08858, over 1360455.81 frames.], batch size: 26, lr: 1.82e-03 2022-05-27 15:36:03,185 INFO [train.py:823] (0/4) Epoch 8, batch 700, loss[loss=0.2321, simple_loss=0.3026, pruned_loss=0.0808, over 7278.00 frames.], tot_loss[loss=0.2426, simple_loss=0.3098, pruned_loss=0.08766, over 1379138.72 frames.], batch size: 19, lr: 1.82e-03 2022-05-27 15:36:42,105 INFO [train.py:823] (0/4) Epoch 8, batch 750, loss[loss=0.2052, simple_loss=0.2822, pruned_loss=0.06415, over 7097.00 frames.], tot_loss[loss=0.2416, simple_loss=0.3091, pruned_loss=0.08705, over 1386181.96 frames.], batch size: 18, lr: 1.81e-03 2022-05-27 15:37:21,562 INFO [train.py:823] (0/4) Epoch 8, batch 800, loss[loss=0.2552, simple_loss=0.3193, pruned_loss=0.09554, over 4802.00 frames.], tot_loss[loss=0.2429, simple_loss=0.3104, pruned_loss=0.08773, over 1387521.81 frames.], batch size: 47, lr: 1.81e-03 2022-05-27 15:38:00,601 INFO [train.py:823] (0/4) Epoch 8, batch 850, loss[loss=0.2672, simple_loss=0.339, pruned_loss=0.09769, over 7197.00 frames.], tot_loss[loss=0.2412, simple_loss=0.3089, pruned_loss=0.08675, over 1391114.08 frames.], batch size: 20, lr: 1.80e-03 2022-05-27 15:38:39,693 INFO [train.py:823] (0/4) Epoch 8, batch 900, loss[loss=0.2144, simple_loss=0.2883, pruned_loss=0.07024, over 7096.00 frames.], tot_loss[loss=0.2435, simple_loss=0.3113, pruned_loss=0.08787, over 1394806.59 frames.], batch size: 18, lr: 1.80e-03 2022-05-27 15:39:19,450 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-8.pt 2022-05-27 15:39:31,039 INFO [train.py:823] (0/4) Epoch 9, batch 0, loss[loss=0.2915, simple_loss=0.3597, pruned_loss=0.1116, over 7191.00 frames.], tot_loss[loss=0.2915, simple_loss=0.3597, pruned_loss=0.1116, over 7191.00 frames.], batch size: 21, lr: 1.72e-03 2022-05-27 15:40:10,099 INFO [train.py:823] (0/4) Epoch 9, batch 50, loss[loss=0.2006, simple_loss=0.2822, pruned_loss=0.05953, over 7392.00 frames.], tot_loss[loss=0.2315, simple_loss=0.3022, pruned_loss=0.08047, over 319727.14 frames.], batch size: 19, lr: 1.72e-03 2022-05-27 15:40:49,179 INFO [train.py:823] (0/4) Epoch 9, batch 100, loss[loss=0.2485, simple_loss=0.3173, pruned_loss=0.08988, over 7286.00 frames.], tot_loss[loss=0.2339, simple_loss=0.3042, pruned_loss=0.08184, over 562944.57 frames.], batch size: 19, lr: 1.71e-03 2022-05-27 15:41:28,180 INFO [train.py:823] (0/4) Epoch 9, batch 150, loss[loss=0.2019, simple_loss=0.2801, pruned_loss=0.06188, over 7102.00 frames.], tot_loss[loss=0.234, simple_loss=0.305, pruned_loss=0.08151, over 753056.35 frames.], batch size: 19, lr: 1.71e-03 2022-05-27 15:42:06,935 INFO [train.py:823] (0/4) Epoch 9, batch 200, loss[loss=0.2038, simple_loss=0.2953, pruned_loss=0.05611, over 7288.00 frames.], tot_loss[loss=0.2345, simple_loss=0.3059, pruned_loss=0.08158, over 895064.84 frames.], batch size: 20, lr: 1.71e-03 2022-05-27 15:42:46,304 INFO [train.py:823] (0/4) Epoch 9, batch 250, loss[loss=0.2403, simple_loss=0.3126, pruned_loss=0.08401, over 7201.00 frames.], tot_loss[loss=0.233, simple_loss=0.3047, pruned_loss=0.08068, over 1011561.61 frames.], batch size: 20, lr: 1.70e-03 2022-05-27 15:43:24,862 INFO [train.py:823] (0/4) Epoch 9, batch 300, loss[loss=0.1971, simple_loss=0.2768, pruned_loss=0.05868, over 7198.00 frames.], tot_loss[loss=0.2322, simple_loss=0.3043, pruned_loss=0.08008, over 1104404.54 frames.], batch size: 18, lr: 1.70e-03 2022-05-27 15:44:04,283 INFO [train.py:823] (0/4) Epoch 9, batch 350, loss[loss=0.23, simple_loss=0.2876, pruned_loss=0.0862, over 7283.00 frames.], tot_loss[loss=0.2313, simple_loss=0.3034, pruned_loss=0.07958, over 1173807.02 frames.], batch size: 17, lr: 1.70e-03 2022-05-27 15:44:43,689 INFO [train.py:823] (0/4) Epoch 9, batch 400, loss[loss=0.2104, simple_loss=0.2977, pruned_loss=0.06158, over 7309.00 frames.], tot_loss[loss=0.2305, simple_loss=0.3026, pruned_loss=0.0792, over 1229658.50 frames.], batch size: 22, lr: 1.69e-03 2022-05-27 15:44:44,801 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/checkpoint-8000.pt 2022-05-27 15:45:28,869 INFO [train.py:823] (0/4) Epoch 9, batch 450, loss[loss=0.2152, simple_loss=0.2897, pruned_loss=0.07029, over 7205.00 frames.], tot_loss[loss=0.2317, simple_loss=0.3039, pruned_loss=0.07978, over 1270999.59 frames.], batch size: 19, lr: 1.69e-03 2022-05-27 15:46:08,962 INFO [train.py:823] (0/4) Epoch 9, batch 500, loss[loss=0.2084, simple_loss=0.2916, pruned_loss=0.06262, over 7239.00 frames.], tot_loss[loss=0.2331, simple_loss=0.3052, pruned_loss=0.08054, over 1304334.40 frames.], batch size: 24, lr: 1.68e-03 2022-05-27 15:46:48,233 INFO [train.py:823] (0/4) Epoch 9, batch 550, loss[loss=0.28, simple_loss=0.3361, pruned_loss=0.112, over 7188.00 frames.], tot_loss[loss=0.2347, simple_loss=0.3059, pruned_loss=0.08178, over 1333764.13 frames.], batch size: 19, lr: 1.68e-03 2022-05-27 15:47:28,233 INFO [train.py:823] (0/4) Epoch 9, batch 600, loss[loss=0.2384, simple_loss=0.3003, pruned_loss=0.08829, over 7145.00 frames.], tot_loss[loss=0.2343, simple_loss=0.3053, pruned_loss=0.08167, over 1354292.90 frames.], batch size: 17, lr: 1.68e-03 2022-05-27 15:48:07,782 INFO [train.py:823] (0/4) Epoch 9, batch 650, loss[loss=0.2435, simple_loss=0.319, pruned_loss=0.08405, over 6781.00 frames.], tot_loss[loss=0.2343, simple_loss=0.3054, pruned_loss=0.08157, over 1366920.89 frames.], batch size: 29, lr: 1.67e-03 2022-05-27 15:48:46,356 INFO [train.py:823] (0/4) Epoch 9, batch 700, loss[loss=0.2094, simple_loss=0.293, pruned_loss=0.06287, over 7300.00 frames.], tot_loss[loss=0.2339, simple_loss=0.3055, pruned_loss=0.08116, over 1374739.74 frames.], batch size: 22, lr: 1.67e-03 2022-05-27 15:49:25,419 INFO [train.py:823] (0/4) Epoch 9, batch 750, loss[loss=0.1996, simple_loss=0.2753, pruned_loss=0.06198, over 7199.00 frames.], tot_loss[loss=0.2348, simple_loss=0.3066, pruned_loss=0.08151, over 1385500.11 frames.], batch size: 18, lr: 1.67e-03 2022-05-27 15:50:03,860 INFO [train.py:823] (0/4) Epoch 9, batch 800, loss[loss=0.2612, simple_loss=0.3211, pruned_loss=0.1007, over 7099.00 frames.], tot_loss[loss=0.2343, simple_loss=0.3062, pruned_loss=0.08127, over 1386877.50 frames.], batch size: 19, lr: 1.66e-03 2022-05-27 15:50:43,429 INFO [train.py:823] (0/4) Epoch 9, batch 850, loss[loss=0.2364, simple_loss=0.2858, pruned_loss=0.09352, over 7228.00 frames.], tot_loss[loss=0.2325, simple_loss=0.3048, pruned_loss=0.08011, over 1397227.45 frames.], batch size: 16, lr: 1.66e-03 2022-05-27 15:51:23,326 INFO [train.py:823] (0/4) Epoch 9, batch 900, loss[loss=0.2016, simple_loss=0.2706, pruned_loss=0.06629, over 6798.00 frames.], tot_loss[loss=0.2331, simple_loss=0.3051, pruned_loss=0.08056, over 1400112.47 frames.], batch size: 15, lr: 1.65e-03 2022-05-27 15:52:02,719 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-9.pt 2022-05-27 15:52:14,259 INFO [train.py:823] (0/4) Epoch 10, batch 0, loss[loss=0.2253, simple_loss=0.3024, pruned_loss=0.07413, over 7105.00 frames.], tot_loss[loss=0.2253, simple_loss=0.3024, pruned_loss=0.07413, over 7105.00 frames.], batch size: 20, lr: 1.59e-03 2022-05-27 15:52:52,860 INFO [train.py:823] (0/4) Epoch 10, batch 50, loss[loss=0.1746, simple_loss=0.2536, pruned_loss=0.04778, over 7026.00 frames.], tot_loss[loss=0.2286, simple_loss=0.3008, pruned_loss=0.07826, over 319276.30 frames.], batch size: 17, lr: 1.58e-03 2022-05-27 15:53:32,110 INFO [train.py:823] (0/4) Epoch 10, batch 100, loss[loss=0.2027, simple_loss=0.2857, pruned_loss=0.05981, over 7378.00 frames.], tot_loss[loss=0.2249, simple_loss=0.2995, pruned_loss=0.07513, over 559832.53 frames.], batch size: 20, lr: 1.58e-03 2022-05-27 15:54:10,802 INFO [train.py:823] (0/4) Epoch 10, batch 150, loss[loss=0.298, simple_loss=0.3385, pruned_loss=0.1288, over 7277.00 frames.], tot_loss[loss=0.2255, simple_loss=0.3007, pruned_loss=0.07516, over 749113.03 frames.], batch size: 20, lr: 1.58e-03 2022-05-27 15:54:50,484 INFO [train.py:823] (0/4) Epoch 10, batch 200, loss[loss=0.2525, simple_loss=0.3289, pruned_loss=0.08807, over 7274.00 frames.], tot_loss[loss=0.2247, simple_loss=0.2997, pruned_loss=0.07482, over 900515.57 frames.], batch size: 21, lr: 1.57e-03 2022-05-27 15:55:29,320 INFO [train.py:823] (0/4) Epoch 10, batch 250, loss[loss=0.235, simple_loss=0.3018, pruned_loss=0.08407, over 7381.00 frames.], tot_loss[loss=0.2254, simple_loss=0.3009, pruned_loss=0.0749, over 1016211.49 frames.], batch size: 21, lr: 1.57e-03 2022-05-27 15:56:08,350 INFO [train.py:823] (0/4) Epoch 10, batch 300, loss[loss=0.2554, simple_loss=0.3269, pruned_loss=0.09196, over 7022.00 frames.], tot_loss[loss=0.2257, simple_loss=0.301, pruned_loss=0.07517, over 1107388.77 frames.], batch size: 26, lr: 1.57e-03 2022-05-27 15:56:47,610 INFO [train.py:823] (0/4) Epoch 10, batch 350, loss[loss=0.2258, simple_loss=0.2899, pruned_loss=0.08088, over 7218.00 frames.], tot_loss[loss=0.2248, simple_loss=0.3003, pruned_loss=0.07459, over 1174159.74 frames.], batch size: 16, lr: 1.56e-03 2022-05-27 15:57:26,972 INFO [train.py:823] (0/4) Epoch 10, batch 400, loss[loss=0.2336, simple_loss=0.3133, pruned_loss=0.07698, over 7105.00 frames.], tot_loss[loss=0.2254, simple_loss=0.3013, pruned_loss=0.0747, over 1223698.95 frames.], batch size: 19, lr: 1.56e-03 2022-05-27 15:58:05,987 INFO [train.py:823] (0/4) Epoch 10, batch 450, loss[loss=0.2362, simple_loss=0.313, pruned_loss=0.07968, over 7274.00 frames.], tot_loss[loss=0.2258, simple_loss=0.3011, pruned_loss=0.07528, over 1265108.42 frames.], batch size: 20, lr: 1.56e-03 2022-05-27 15:58:45,240 INFO [train.py:823] (0/4) Epoch 10, batch 500, loss[loss=0.217, simple_loss=0.305, pruned_loss=0.0645, over 7280.00 frames.], tot_loss[loss=0.2247, simple_loss=0.3001, pruned_loss=0.07466, over 1298102.09 frames.], batch size: 20, lr: 1.55e-03 2022-05-27 15:59:23,992 INFO [train.py:823] (0/4) Epoch 10, batch 550, loss[loss=0.2028, simple_loss=0.2906, pruned_loss=0.05753, over 7099.00 frames.], tot_loss[loss=0.2239, simple_loss=0.2993, pruned_loss=0.0742, over 1327992.01 frames.], batch size: 18, lr: 1.55e-03 2022-05-27 16:00:03,421 INFO [train.py:823] (0/4) Epoch 10, batch 600, loss[loss=0.2052, simple_loss=0.294, pruned_loss=0.05824, over 7288.00 frames.], tot_loss[loss=0.2262, simple_loss=0.3012, pruned_loss=0.07567, over 1352973.78 frames.], batch size: 19, lr: 1.55e-03 2022-05-27 16:00:42,540 INFO [train.py:823] (0/4) Epoch 10, batch 650, loss[loss=0.2241, simple_loss=0.3056, pruned_loss=0.07135, over 7184.00 frames.], tot_loss[loss=0.2249, simple_loss=0.2999, pruned_loss=0.07493, over 1370457.37 frames.], batch size: 21, lr: 1.54e-03 2022-05-27 16:01:22,299 INFO [train.py:823] (0/4) Epoch 10, batch 700, loss[loss=0.2513, simple_loss=0.3099, pruned_loss=0.09635, over 7002.00 frames.], tot_loss[loss=0.2249, simple_loss=0.3002, pruned_loss=0.07481, over 1384586.48 frames.], batch size: 16, lr: 1.54e-03 2022-05-27 16:02:01,167 INFO [train.py:823] (0/4) Epoch 10, batch 750, loss[loss=0.2176, simple_loss=0.2956, pruned_loss=0.06986, over 7190.00 frames.], tot_loss[loss=0.2241, simple_loss=0.2994, pruned_loss=0.07438, over 1392103.72 frames.], batch size: 18, lr: 1.54e-03 2022-05-27 16:02:40,249 INFO [train.py:823] (0/4) Epoch 10, batch 800, loss[loss=0.2508, simple_loss=0.3236, pruned_loss=0.08898, over 7243.00 frames.], tot_loss[loss=0.225, simple_loss=0.3003, pruned_loss=0.07483, over 1398892.82 frames.], batch size: 25, lr: 1.53e-03 2022-05-27 16:03:19,615 INFO [train.py:823] (0/4) Epoch 10, batch 850, loss[loss=0.2236, simple_loss=0.3122, pruned_loss=0.0675, over 7163.00 frames.], tot_loss[loss=0.2233, simple_loss=0.2988, pruned_loss=0.07392, over 1405487.34 frames.], batch size: 22, lr: 1.53e-03 2022-05-27 16:03:59,156 INFO [train.py:823] (0/4) Epoch 10, batch 900, loss[loss=0.1988, simple_loss=0.2669, pruned_loss=0.06534, over 7234.00 frames.], tot_loss[loss=0.223, simple_loss=0.2985, pruned_loss=0.07371, over 1406851.46 frames.], batch size: 16, lr: 1.53e-03 2022-05-27 16:04:42,055 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-10.pt 2022-05-27 16:04:53,503 INFO [train.py:823] (0/4) Epoch 11, batch 0, loss[loss=0.2183, simple_loss=0.3036, pruned_loss=0.06646, over 7098.00 frames.], tot_loss[loss=0.2183, simple_loss=0.3036, pruned_loss=0.06646, over 7098.00 frames.], batch size: 19, lr: 1.47e-03 2022-05-27 16:05:32,581 INFO [train.py:823] (0/4) Epoch 11, batch 50, loss[loss=0.2013, simple_loss=0.289, pruned_loss=0.05681, over 6362.00 frames.], tot_loss[loss=0.222, simple_loss=0.2973, pruned_loss=0.07342, over 323612.88 frames.], batch size: 34, lr: 1.47e-03 2022-05-27 16:06:11,614 INFO [train.py:823] (0/4) Epoch 11, batch 100, loss[loss=0.2046, simple_loss=0.2889, pruned_loss=0.0601, over 7166.00 frames.], tot_loss[loss=0.216, simple_loss=0.2928, pruned_loss=0.06961, over 570541.92 frames.], batch size: 17, lr: 1.46e-03 2022-05-27 16:06:50,764 INFO [train.py:823] (0/4) Epoch 11, batch 150, loss[loss=0.2356, simple_loss=0.3182, pruned_loss=0.07654, over 7238.00 frames.], tot_loss[loss=0.217, simple_loss=0.2942, pruned_loss=0.06984, over 762032.97 frames.], batch size: 24, lr: 1.46e-03 2022-05-27 16:07:29,199 INFO [train.py:823] (0/4) Epoch 11, batch 200, loss[loss=0.1989, simple_loss=0.2814, pruned_loss=0.05817, over 7101.00 frames.], tot_loss[loss=0.2194, simple_loss=0.297, pruned_loss=0.07083, over 902108.06 frames.], batch size: 19, lr: 1.46e-03 2022-05-27 16:08:08,810 INFO [train.py:823] (0/4) Epoch 11, batch 250, loss[loss=0.2556, simple_loss=0.3167, pruned_loss=0.09724, over 7102.00 frames.], tot_loss[loss=0.2186, simple_loss=0.2963, pruned_loss=0.07045, over 1015204.12 frames.], batch size: 18, lr: 1.45e-03 2022-05-27 16:08:50,592 INFO [train.py:823] (0/4) Epoch 11, batch 300, loss[loss=0.2557, simple_loss=0.329, pruned_loss=0.09122, over 7185.00 frames.], tot_loss[loss=0.2202, simple_loss=0.2969, pruned_loss=0.07177, over 1105979.83 frames.], batch size: 25, lr: 1.45e-03 2022-05-27 16:09:29,764 INFO [train.py:823] (0/4) Epoch 11, batch 350, loss[loss=0.221, simple_loss=0.3106, pruned_loss=0.06566, over 7227.00 frames.], tot_loss[loss=0.2196, simple_loss=0.2965, pruned_loss=0.07137, over 1177153.20 frames.], batch size: 25, lr: 1.45e-03 2022-05-27 16:10:08,819 INFO [train.py:823] (0/4) Epoch 11, batch 400, loss[loss=0.2453, simple_loss=0.3301, pruned_loss=0.08025, over 7095.00 frames.], tot_loss[loss=0.2179, simple_loss=0.2956, pruned_loss=0.07008, over 1231888.14 frames.], batch size: 19, lr: 1.44e-03 2022-05-27 16:10:47,981 INFO [train.py:823] (0/4) Epoch 11, batch 450, loss[loss=0.1921, simple_loss=0.2847, pruned_loss=0.0498, over 7322.00 frames.], tot_loss[loss=0.2173, simple_loss=0.2951, pruned_loss=0.06979, over 1271004.69 frames.], batch size: 22, lr: 1.44e-03 2022-05-27 16:11:28,138 INFO [train.py:823] (0/4) Epoch 11, batch 500, loss[loss=0.2132, simple_loss=0.3101, pruned_loss=0.05813, over 6444.00 frames.], tot_loss[loss=0.2167, simple_loss=0.2952, pruned_loss=0.06911, over 1304210.43 frames.], batch size: 34, lr: 1.44e-03 2022-05-27 16:12:07,642 INFO [train.py:823] (0/4) Epoch 11, batch 550, loss[loss=0.2368, simple_loss=0.311, pruned_loss=0.08134, over 7409.00 frames.], tot_loss[loss=0.2178, simple_loss=0.2967, pruned_loss=0.06945, over 1332592.29 frames.], batch size: 22, lr: 1.44e-03 2022-05-27 16:12:46,618 INFO [train.py:823] (0/4) Epoch 11, batch 600, loss[loss=0.1851, simple_loss=0.2709, pruned_loss=0.04963, over 7389.00 frames.], tot_loss[loss=0.2177, simple_loss=0.2964, pruned_loss=0.06945, over 1351233.92 frames.], batch size: 19, lr: 1.43e-03 2022-05-27 16:13:26,092 INFO [train.py:823] (0/4) Epoch 11, batch 650, loss[loss=0.1741, simple_loss=0.2532, pruned_loss=0.04749, over 7310.00 frames.], tot_loss[loss=0.2173, simple_loss=0.2957, pruned_loss=0.06951, over 1368291.67 frames.], batch size: 18, lr: 1.43e-03 2022-05-27 16:14:04,486 INFO [train.py:823] (0/4) Epoch 11, batch 700, loss[loss=0.2666, simple_loss=0.3241, pruned_loss=0.1045, over 7159.00 frames.], tot_loss[loss=0.2176, simple_loss=0.2959, pruned_loss=0.06966, over 1382418.52 frames.], batch size: 17, lr: 1.43e-03 2022-05-27 16:14:45,012 INFO [train.py:823] (0/4) Epoch 11, batch 750, loss[loss=0.1728, simple_loss=0.2502, pruned_loss=0.04774, over 7297.00 frames.], tot_loss[loss=0.216, simple_loss=0.2943, pruned_loss=0.0688, over 1390779.15 frames.], batch size: 17, lr: 1.42e-03 2022-05-27 16:15:23,508 INFO [train.py:823] (0/4) Epoch 11, batch 800, loss[loss=0.2135, simple_loss=0.2947, pruned_loss=0.06618, over 7216.00 frames.], tot_loss[loss=0.216, simple_loss=0.2944, pruned_loss=0.06877, over 1395711.64 frames.], batch size: 19, lr: 1.42e-03 2022-05-27 16:16:03,290 INFO [train.py:823] (0/4) Epoch 11, batch 850, loss[loss=0.2387, simple_loss=0.3248, pruned_loss=0.0763, over 7124.00 frames.], tot_loss[loss=0.217, simple_loss=0.2953, pruned_loss=0.06939, over 1398401.12 frames.], batch size: 20, lr: 1.42e-03 2022-05-27 16:16:42,236 INFO [train.py:823] (0/4) Epoch 11, batch 900, loss[loss=0.2081, simple_loss=0.2706, pruned_loss=0.07279, over 7255.00 frames.], tot_loss[loss=0.2174, simple_loss=0.296, pruned_loss=0.06938, over 1398612.58 frames.], batch size: 16, lr: 1.42e-03 2022-05-27 16:17:21,035 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-11.pt 2022-05-27 16:17:32,986 INFO [train.py:823] (0/4) Epoch 12, batch 0, loss[loss=0.1751, simple_loss=0.2486, pruned_loss=0.05082, over 7295.00 frames.], tot_loss[loss=0.1751, simple_loss=0.2486, pruned_loss=0.05082, over 7295.00 frames.], batch size: 17, lr: 1.36e-03 2022-05-27 16:18:12,388 INFO [train.py:823] (0/4) Epoch 12, batch 50, loss[loss=0.24, simple_loss=0.3126, pruned_loss=0.08369, over 7256.00 frames.], tot_loss[loss=0.2139, simple_loss=0.2912, pruned_loss=0.06834, over 318337.27 frames.], batch size: 24, lr: 1.36e-03 2022-05-27 16:18:51,682 INFO [train.py:823] (0/4) Epoch 12, batch 100, loss[loss=0.2342, simple_loss=0.3102, pruned_loss=0.0791, over 7141.00 frames.], tot_loss[loss=0.2107, simple_loss=0.2898, pruned_loss=0.06578, over 562440.04 frames.], batch size: 23, lr: 1.36e-03 2022-05-27 16:19:30,771 INFO [train.py:823] (0/4) Epoch 12, batch 150, loss[loss=0.1945, simple_loss=0.2835, pruned_loss=0.0527, over 7277.00 frames.], tot_loss[loss=0.2105, simple_loss=0.2901, pruned_loss=0.06546, over 752990.12 frames.], batch size: 20, lr: 1.36e-03 2022-05-27 16:20:10,254 INFO [train.py:823] (0/4) Epoch 12, batch 200, loss[loss=0.2001, simple_loss=0.2774, pruned_loss=0.06145, over 6853.00 frames.], tot_loss[loss=0.211, simple_loss=0.2907, pruned_loss=0.06563, over 899931.32 frames.], batch size: 15, lr: 1.35e-03 2022-05-27 16:20:49,258 INFO [train.py:823] (0/4) Epoch 12, batch 250, loss[loss=0.2439, simple_loss=0.3208, pruned_loss=0.08352, over 7001.00 frames.], tot_loss[loss=0.2115, simple_loss=0.2909, pruned_loss=0.06605, over 1017017.68 frames.], batch size: 26, lr: 1.35e-03 2022-05-27 16:21:28,535 INFO [train.py:823] (0/4) Epoch 12, batch 300, loss[loss=0.2186, simple_loss=0.2855, pruned_loss=0.0759, over 7198.00 frames.], tot_loss[loss=0.2126, simple_loss=0.2922, pruned_loss=0.06654, over 1103632.00 frames.], batch size: 19, lr: 1.35e-03 2022-05-27 16:22:07,688 INFO [train.py:823] (0/4) Epoch 12, batch 350, loss[loss=0.2128, simple_loss=0.2926, pruned_loss=0.06653, over 7333.00 frames.], tot_loss[loss=0.2124, simple_loss=0.2924, pruned_loss=0.06625, over 1177241.99 frames.], batch size: 23, lr: 1.35e-03 2022-05-27 16:22:46,794 INFO [train.py:823] (0/4) Epoch 12, batch 400, loss[loss=0.2096, simple_loss=0.2991, pruned_loss=0.06004, over 6912.00 frames.], tot_loss[loss=0.2116, simple_loss=0.2918, pruned_loss=0.06566, over 1231385.95 frames.], batch size: 29, lr: 1.34e-03 2022-05-27 16:23:26,112 INFO [train.py:823] (0/4) Epoch 12, batch 450, loss[loss=0.1887, simple_loss=0.277, pruned_loss=0.05015, over 7368.00 frames.], tot_loss[loss=0.2107, simple_loss=0.2915, pruned_loss=0.06498, over 1272878.51 frames.], batch size: 20, lr: 1.34e-03 2022-05-27 16:24:05,952 INFO [train.py:823] (0/4) Epoch 12, batch 500, loss[loss=0.1869, simple_loss=0.2868, pruned_loss=0.04345, over 7286.00 frames.], tot_loss[loss=0.21, simple_loss=0.2909, pruned_loss=0.06454, over 1310392.15 frames.], batch size: 20, lr: 1.34e-03 2022-05-27 16:24:44,890 INFO [train.py:823] (0/4) Epoch 12, batch 550, loss[loss=0.1972, simple_loss=0.2742, pruned_loss=0.06011, over 7423.00 frames.], tot_loss[loss=0.2111, simple_loss=0.2919, pruned_loss=0.06521, over 1338250.71 frames.], batch size: 18, lr: 1.34e-03 2022-05-27 16:25:24,213 INFO [train.py:823] (0/4) Epoch 12, batch 600, loss[loss=0.197, simple_loss=0.2741, pruned_loss=0.05995, over 7207.00 frames.], tot_loss[loss=0.2103, simple_loss=0.2911, pruned_loss=0.06478, over 1358922.83 frames.], batch size: 16, lr: 1.33e-03 2022-05-27 16:26:03,103 INFO [train.py:823] (0/4) Epoch 12, batch 650, loss[loss=0.2112, simple_loss=0.311, pruned_loss=0.05568, over 7283.00 frames.], tot_loss[loss=0.2117, simple_loss=0.2921, pruned_loss=0.0656, over 1371291.09 frames.], batch size: 21, lr: 1.33e-03 2022-05-27 16:26:41,903 INFO [train.py:823] (0/4) Epoch 12, batch 700, loss[loss=0.2313, simple_loss=0.2934, pruned_loss=0.08462, over 7283.00 frames.], tot_loss[loss=0.212, simple_loss=0.2924, pruned_loss=0.06575, over 1382273.00 frames.], batch size: 20, lr: 1.33e-03 2022-05-27 16:27:21,103 INFO [train.py:823] (0/4) Epoch 12, batch 750, loss[loss=0.2421, simple_loss=0.3118, pruned_loss=0.08615, over 7310.00 frames.], tot_loss[loss=0.2124, simple_loss=0.2928, pruned_loss=0.06595, over 1388624.78 frames.], batch size: 22, lr: 1.33e-03 2022-05-27 16:28:00,217 INFO [train.py:823] (0/4) Epoch 12, batch 800, loss[loss=0.2524, simple_loss=0.3384, pruned_loss=0.08313, over 7299.00 frames.], tot_loss[loss=0.2124, simple_loss=0.2926, pruned_loss=0.06616, over 1395505.19 frames.], batch size: 22, lr: 1.32e-03 2022-05-27 16:28:38,842 INFO [train.py:823] (0/4) Epoch 12, batch 850, loss[loss=0.2007, simple_loss=0.2853, pruned_loss=0.05806, over 7187.00 frames.], tot_loss[loss=0.2122, simple_loss=0.2925, pruned_loss=0.06595, over 1400624.21 frames.], batch size: 18, lr: 1.32e-03 2022-05-27 16:29:17,826 INFO [train.py:823] (0/4) Epoch 12, batch 900, loss[loss=0.2413, simple_loss=0.3097, pruned_loss=0.08645, over 7096.00 frames.], tot_loss[loss=0.2118, simple_loss=0.2917, pruned_loss=0.06593, over 1397287.80 frames.], batch size: 19, lr: 1.32e-03 2022-05-27 16:29:56,680 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-12.pt 2022-05-27 16:30:08,405 INFO [train.py:823] (0/4) Epoch 13, batch 0, loss[loss=0.2053, simple_loss=0.2896, pruned_loss=0.06051, over 7163.00 frames.], tot_loss[loss=0.2053, simple_loss=0.2896, pruned_loss=0.06051, over 7163.00 frames.], batch size: 22, lr: 1.27e-03 2022-05-27 16:30:48,280 INFO [train.py:823] (0/4) Epoch 13, batch 50, loss[loss=0.2003, simple_loss=0.2882, pruned_loss=0.05618, over 7294.00 frames.], tot_loss[loss=0.2094, simple_loss=0.2903, pruned_loss=0.06423, over 317840.62 frames.], batch size: 19, lr: 1.27e-03 2022-05-27 16:31:28,616 INFO [train.py:823] (0/4) Epoch 13, batch 100, loss[loss=0.1778, simple_loss=0.2545, pruned_loss=0.05051, over 7317.00 frames.], tot_loss[loss=0.2095, simple_loss=0.2899, pruned_loss=0.06457, over 561442.36 frames.], batch size: 18, lr: 1.27e-03 2022-05-27 16:32:11,890 INFO [train.py:823] (0/4) Epoch 13, batch 150, loss[loss=0.1687, simple_loss=0.2493, pruned_loss=0.04409, over 7394.00 frames.], tot_loss[loss=0.2062, simple_loss=0.288, pruned_loss=0.06226, over 752219.54 frames.], batch size: 19, lr: 1.26e-03 2022-05-27 16:32:57,571 INFO [train.py:823] (0/4) Epoch 13, batch 200, loss[loss=0.1733, simple_loss=0.2585, pruned_loss=0.04401, over 7020.00 frames.], tot_loss[loss=0.2064, simple_loss=0.2877, pruned_loss=0.06253, over 903397.09 frames.], batch size: 17, lr: 1.26e-03 2022-05-27 16:33:37,070 INFO [train.py:823] (0/4) Epoch 13, batch 250, loss[loss=0.2123, simple_loss=0.2955, pruned_loss=0.0645, over 7157.00 frames.], tot_loss[loss=0.2056, simple_loss=0.2867, pruned_loss=0.06226, over 1016485.07 frames.], batch size: 22, lr: 1.26e-03 2022-05-27 16:34:18,168 INFO [train.py:823] (0/4) Epoch 13, batch 300, loss[loss=0.1741, simple_loss=0.2527, pruned_loss=0.04773, over 7295.00 frames.], tot_loss[loss=0.2064, simple_loss=0.2871, pruned_loss=0.06289, over 1111018.63 frames.], batch size: 17, lr: 1.26e-03 2022-05-27 16:34:57,075 INFO [train.py:823] (0/4) Epoch 13, batch 350, loss[loss=0.2321, simple_loss=0.3162, pruned_loss=0.07404, over 6480.00 frames.], tot_loss[loss=0.2072, simple_loss=0.2882, pruned_loss=0.06306, over 1177334.45 frames.], batch size: 34, lr: 1.26e-03 2022-05-27 16:35:36,334 INFO [train.py:823] (0/4) Epoch 13, batch 400, loss[loss=0.2227, simple_loss=0.3048, pruned_loss=0.07029, over 7012.00 frames.], tot_loss[loss=0.2075, simple_loss=0.2885, pruned_loss=0.06322, over 1230347.48 frames.], batch size: 26, lr: 1.25e-03 2022-05-27 16:36:15,517 INFO [train.py:823] (0/4) Epoch 13, batch 450, loss[loss=0.2051, simple_loss=0.2921, pruned_loss=0.05908, over 6904.00 frames.], tot_loss[loss=0.2069, simple_loss=0.2876, pruned_loss=0.06314, over 1267737.77 frames.], batch size: 29, lr: 1.25e-03 2022-05-27 16:36:55,091 INFO [train.py:823] (0/4) Epoch 13, batch 500, loss[loss=0.2193, simple_loss=0.2979, pruned_loss=0.07034, over 6918.00 frames.], tot_loss[loss=0.2061, simple_loss=0.2868, pruned_loss=0.06269, over 1300764.90 frames.], batch size: 29, lr: 1.25e-03 2022-05-27 16:37:34,235 INFO [train.py:823] (0/4) Epoch 13, batch 550, loss[loss=0.2279, simple_loss=0.3056, pruned_loss=0.07512, over 7286.00 frames.], tot_loss[loss=0.2084, simple_loss=0.288, pruned_loss=0.0644, over 1322395.18 frames.], batch size: 19, lr: 1.25e-03 2022-05-27 16:38:13,496 INFO [train.py:823] (0/4) Epoch 13, batch 600, loss[loss=0.2287, simple_loss=0.3016, pruned_loss=0.07786, over 7266.00 frames.], tot_loss[loss=0.2093, simple_loss=0.2894, pruned_loss=0.0646, over 1345047.16 frames.], batch size: 20, lr: 1.24e-03 2022-05-27 16:38:53,771 INFO [train.py:823] (0/4) Epoch 13, batch 650, loss[loss=0.2005, simple_loss=0.2797, pruned_loss=0.06063, over 7197.00 frames.], tot_loss[loss=0.2093, simple_loss=0.2898, pruned_loss=0.06436, over 1362263.19 frames.], batch size: 19, lr: 1.24e-03 2022-05-27 16:39:33,147 INFO [train.py:823] (0/4) Epoch 13, batch 700, loss[loss=0.1694, simple_loss=0.2513, pruned_loss=0.04376, over 7027.00 frames.], tot_loss[loss=0.2076, simple_loss=0.2886, pruned_loss=0.06329, over 1372604.05 frames.], batch size: 17, lr: 1.24e-03 2022-05-27 16:40:12,571 INFO [train.py:823] (0/4) Epoch 13, batch 750, loss[loss=0.2039, simple_loss=0.2876, pruned_loss=0.06013, over 6908.00 frames.], tot_loss[loss=0.207, simple_loss=0.2885, pruned_loss=0.06269, over 1380001.38 frames.], batch size: 29, lr: 1.24e-03 2022-05-27 16:40:51,203 INFO [train.py:823] (0/4) Epoch 13, batch 800, loss[loss=0.2298, simple_loss=0.3092, pruned_loss=0.07519, over 7138.00 frames.], tot_loss[loss=0.207, simple_loss=0.2885, pruned_loss=0.0627, over 1388194.11 frames.], batch size: 23, lr: 1.24e-03 2022-05-27 16:41:30,432 INFO [train.py:823] (0/4) Epoch 13, batch 850, loss[loss=0.2117, simple_loss=0.2938, pruned_loss=0.06476, over 7288.00 frames.], tot_loss[loss=0.2053, simple_loss=0.2874, pruned_loss=0.06156, over 1397660.16 frames.], batch size: 20, lr: 1.23e-03 2022-05-27 16:42:09,481 INFO [train.py:823] (0/4) Epoch 13, batch 900, loss[loss=0.187, simple_loss=0.2618, pruned_loss=0.05609, over 7299.00 frames.], tot_loss[loss=0.2058, simple_loss=0.2882, pruned_loss=0.06165, over 1396600.47 frames.], batch size: 19, lr: 1.23e-03 2022-05-27 16:42:48,623 INFO [train.py:823] (0/4) Epoch 13, batch 950, loss[loss=0.1781, simple_loss=0.2539, pruned_loss=0.05121, over 6994.00 frames.], tot_loss[loss=0.206, simple_loss=0.2883, pruned_loss=0.06187, over 1394877.55 frames.], batch size: 16, lr: 1.23e-03 2022-05-27 16:42:49,900 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-13.pt 2022-05-27 16:43:01,790 INFO [train.py:823] (0/4) Epoch 14, batch 0, loss[loss=0.1848, simple_loss=0.2764, pruned_loss=0.04656, over 7317.00 frames.], tot_loss[loss=0.1848, simple_loss=0.2764, pruned_loss=0.04656, over 7317.00 frames.], batch size: 22, lr: 1.19e-03 2022-05-27 16:43:41,480 INFO [train.py:823] (0/4) Epoch 14, batch 50, loss[loss=0.2429, simple_loss=0.3242, pruned_loss=0.0808, over 7207.00 frames.], tot_loss[loss=0.2006, simple_loss=0.2831, pruned_loss=0.05909, over 324291.78 frames.], batch size: 25, lr: 1.19e-03 2022-05-27 16:44:20,917 INFO [train.py:823] (0/4) Epoch 14, batch 100, loss[loss=0.2034, simple_loss=0.2907, pruned_loss=0.05806, over 7251.00 frames.], tot_loss[loss=0.2007, simple_loss=0.2841, pruned_loss=0.05861, over 570898.26 frames.], batch size: 24, lr: 1.19e-03 2022-05-27 16:44:59,969 INFO [train.py:823] (0/4) Epoch 14, batch 150, loss[loss=0.2609, simple_loss=0.3257, pruned_loss=0.09804, over 7279.00 frames.], tot_loss[loss=0.2039, simple_loss=0.2866, pruned_loss=0.06058, over 755846.64 frames.], batch size: 21, lr: 1.18e-03 2022-05-27 16:45:39,650 INFO [train.py:823] (0/4) Epoch 14, batch 200, loss[loss=0.222, simple_loss=0.2974, pruned_loss=0.07337, over 7372.00 frames.], tot_loss[loss=0.202, simple_loss=0.2851, pruned_loss=0.05945, over 901321.54 frames.], batch size: 21, lr: 1.18e-03 2022-05-27 16:46:18,469 INFO [train.py:823] (0/4) Epoch 14, batch 250, loss[loss=0.2103, simple_loss=0.299, pruned_loss=0.06082, over 7288.00 frames.], tot_loss[loss=0.1994, simple_loss=0.2827, pruned_loss=0.05804, over 1019531.62 frames.], batch size: 19, lr: 1.18e-03 2022-05-27 16:46:57,752 INFO [train.py:823] (0/4) Epoch 14, batch 300, loss[loss=0.2483, simple_loss=0.3294, pruned_loss=0.08362, over 6394.00 frames.], tot_loss[loss=0.2002, simple_loss=0.2833, pruned_loss=0.05852, over 1099565.31 frames.], batch size: 34, lr: 1.18e-03 2022-05-27 16:47:36,971 INFO [train.py:823] (0/4) Epoch 14, batch 350, loss[loss=0.1769, simple_loss=0.2605, pruned_loss=0.04666, over 7287.00 frames.], tot_loss[loss=0.2015, simple_loss=0.2845, pruned_loss=0.05925, over 1176075.43 frames.], batch size: 19, lr: 1.18e-03 2022-05-27 16:48:16,005 INFO [train.py:823] (0/4) Epoch 14, batch 400, loss[loss=0.1809, simple_loss=0.2724, pruned_loss=0.04467, over 7294.00 frames.], tot_loss[loss=0.2013, simple_loss=0.2847, pruned_loss=0.05897, over 1230812.99 frames.], batch size: 19, lr: 1.17e-03 2022-05-27 16:48:54,598 INFO [train.py:823] (0/4) Epoch 14, batch 450, loss[loss=0.1934, simple_loss=0.266, pruned_loss=0.06039, over 7093.00 frames.], tot_loss[loss=0.2014, simple_loss=0.2848, pruned_loss=0.05898, over 1269522.44 frames.], batch size: 18, lr: 1.17e-03 2022-05-27 16:49:33,653 INFO [train.py:823] (0/4) Epoch 14, batch 500, loss[loss=0.2477, simple_loss=0.3163, pruned_loss=0.08962, over 7186.00 frames.], tot_loss[loss=0.2026, simple_loss=0.2854, pruned_loss=0.05986, over 1304955.11 frames.], batch size: 21, lr: 1.17e-03 2022-05-27 16:50:12,834 INFO [train.py:823] (0/4) Epoch 14, batch 550, loss[loss=0.2217, simple_loss=0.3142, pruned_loss=0.06455, over 7194.00 frames.], tot_loss[loss=0.2022, simple_loss=0.2853, pruned_loss=0.05955, over 1335054.22 frames.], batch size: 25, lr: 1.17e-03 2022-05-27 16:50:52,404 INFO [train.py:823] (0/4) Epoch 14, batch 600, loss[loss=0.1766, simple_loss=0.2721, pruned_loss=0.04055, over 7406.00 frames.], tot_loss[loss=0.2008, simple_loss=0.2837, pruned_loss=0.05898, over 1355930.05 frames.], batch size: 19, lr: 1.17e-03 2022-05-27 16:51:31,480 INFO [train.py:823] (0/4) Epoch 14, batch 650, loss[loss=0.1813, simple_loss=0.2659, pruned_loss=0.04834, over 7323.00 frames.], tot_loss[loss=0.2012, simple_loss=0.2837, pruned_loss=0.05931, over 1369720.52 frames.], batch size: 17, lr: 1.16e-03 2022-05-27 16:52:10,491 INFO [train.py:823] (0/4) Epoch 14, batch 700, loss[loss=0.1841, simple_loss=0.277, pruned_loss=0.04562, over 7284.00 frames.], tot_loss[loss=0.2005, simple_loss=0.283, pruned_loss=0.05899, over 1378217.45 frames.], batch size: 21, lr: 1.16e-03 2022-05-27 16:52:49,098 INFO [train.py:823] (0/4) Epoch 14, batch 750, loss[loss=0.2054, simple_loss=0.2973, pruned_loss=0.0568, over 7108.00 frames.], tot_loss[loss=0.2005, simple_loss=0.2832, pruned_loss=0.05891, over 1388494.93 frames.], batch size: 20, lr: 1.16e-03 2022-05-27 16:53:28,611 INFO [train.py:823] (0/4) Epoch 14, batch 800, loss[loss=0.1922, simple_loss=0.2762, pruned_loss=0.05411, over 7190.00 frames.], tot_loss[loss=0.2008, simple_loss=0.2832, pruned_loss=0.05914, over 1394670.82 frames.], batch size: 19, lr: 1.16e-03 2022-05-27 16:54:09,128 INFO [train.py:823] (0/4) Epoch 14, batch 850, loss[loss=0.1787, simple_loss=0.2722, pruned_loss=0.04254, over 7284.00 frames.], tot_loss[loss=0.2001, simple_loss=0.2826, pruned_loss=0.05884, over 1397294.38 frames.], batch size: 20, lr: 1.16e-03 2022-05-27 16:54:48,403 INFO [train.py:823] (0/4) Epoch 14, batch 900, loss[loss=0.1972, simple_loss=0.2717, pruned_loss=0.06132, over 7022.00 frames.], tot_loss[loss=0.2008, simple_loss=0.2833, pruned_loss=0.05915, over 1401028.46 frames.], batch size: 17, lr: 1.15e-03 2022-05-27 16:55:28,613 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-14.pt 2022-05-27 16:55:39,852 INFO [train.py:823] (0/4) Epoch 15, batch 0, loss[loss=0.1774, simple_loss=0.2647, pruned_loss=0.04509, over 7204.00 frames.], tot_loss[loss=0.1774, simple_loss=0.2647, pruned_loss=0.04509, over 7204.00 frames.], batch size: 19, lr: 1.12e-03 2022-05-27 16:56:18,949 INFO [train.py:823] (0/4) Epoch 15, batch 50, loss[loss=0.215, simple_loss=0.2849, pruned_loss=0.07255, over 7189.00 frames.], tot_loss[loss=0.2034, simple_loss=0.2859, pruned_loss=0.06039, over 319766.04 frames.], batch size: 18, lr: 1.12e-03 2022-05-27 16:56:57,709 INFO [train.py:823] (0/4) Epoch 15, batch 100, loss[loss=0.1909, simple_loss=0.2871, pruned_loss=0.04728, over 7419.00 frames.], tot_loss[loss=0.1983, simple_loss=0.2809, pruned_loss=0.0578, over 559784.42 frames.], batch size: 22, lr: 1.11e-03 2022-05-27 16:57:38,750 INFO [train.py:823] (0/4) Epoch 15, batch 150, loss[loss=0.1945, simple_loss=0.2575, pruned_loss=0.06574, over 7288.00 frames.], tot_loss[loss=0.1968, simple_loss=0.2784, pruned_loss=0.05757, over 751348.21 frames.], batch size: 17, lr: 1.11e-03 2022-05-27 16:58:17,781 INFO [train.py:823] (0/4) Epoch 15, batch 200, loss[loss=0.2039, simple_loss=0.2935, pruned_loss=0.05721, over 7124.00 frames.], tot_loss[loss=0.1984, simple_loss=0.2807, pruned_loss=0.05807, over 898231.68 frames.], batch size: 23, lr: 1.11e-03 2022-05-27 16:58:57,124 INFO [train.py:823] (0/4) Epoch 15, batch 250, loss[loss=0.1999, simple_loss=0.2929, pruned_loss=0.05346, over 6380.00 frames.], tot_loss[loss=0.1974, simple_loss=0.2806, pruned_loss=0.05707, over 1013890.90 frames.], batch size: 34, lr: 1.11e-03 2022-05-27 16:59:36,256 INFO [train.py:823] (0/4) Epoch 15, batch 300, loss[loss=0.2367, simple_loss=0.3109, pruned_loss=0.08129, over 7199.00 frames.], tot_loss[loss=0.1963, simple_loss=0.2793, pruned_loss=0.05667, over 1104130.74 frames.], batch size: 18, lr: 1.11e-03 2022-05-27 17:00:15,461 INFO [train.py:823] (0/4) Epoch 15, batch 350, loss[loss=0.2274, simple_loss=0.3034, pruned_loss=0.07565, over 7370.00 frames.], tot_loss[loss=0.1957, simple_loss=0.2792, pruned_loss=0.05605, over 1176752.87 frames.], batch size: 20, lr: 1.10e-03 2022-05-27 17:00:54,485 INFO [train.py:823] (0/4) Epoch 15, batch 400, loss[loss=0.2309, simple_loss=0.3097, pruned_loss=0.07607, over 7095.00 frames.], tot_loss[loss=0.1963, simple_loss=0.2796, pruned_loss=0.05652, over 1228438.38 frames.], batch size: 19, lr: 1.10e-03 2022-05-27 17:01:34,118 INFO [train.py:823] (0/4) Epoch 15, batch 450, loss[loss=0.1946, simple_loss=0.2869, pruned_loss=0.05119, over 7229.00 frames.], tot_loss[loss=0.1958, simple_loss=0.2795, pruned_loss=0.05601, over 1276095.57 frames.], batch size: 24, lr: 1.10e-03 2022-05-27 17:02:13,040 INFO [train.py:823] (0/4) Epoch 15, batch 500, loss[loss=0.1894, simple_loss=0.2782, pruned_loss=0.05032, over 7106.00 frames.], tot_loss[loss=0.1972, simple_loss=0.2808, pruned_loss=0.0568, over 1311688.65 frames.], batch size: 20, lr: 1.10e-03 2022-05-27 17:02:53,423 INFO [train.py:823] (0/4) Epoch 15, batch 550, loss[loss=0.1717, simple_loss=0.2474, pruned_loss=0.048, over 7436.00 frames.], tot_loss[loss=0.1969, simple_loss=0.2805, pruned_loss=0.05662, over 1332014.79 frames.], batch size: 18, lr: 1.10e-03 2022-05-27 17:03:32,319 INFO [train.py:823] (0/4) Epoch 15, batch 600, loss[loss=0.1825, simple_loss=0.2761, pruned_loss=0.04448, over 7299.00 frames.], tot_loss[loss=0.1974, simple_loss=0.2813, pruned_loss=0.0567, over 1356244.50 frames.], batch size: 19, lr: 1.10e-03 2022-05-27 17:04:11,956 INFO [train.py:823] (0/4) Epoch 15, batch 650, loss[loss=0.2506, simple_loss=0.3246, pruned_loss=0.08829, over 7174.00 frames.], tot_loss[loss=0.196, simple_loss=0.2803, pruned_loss=0.05588, over 1367230.40 frames.], batch size: 22, lr: 1.09e-03 2022-05-27 17:04:51,137 INFO [train.py:823] (0/4) Epoch 15, batch 700, loss[loss=0.2173, simple_loss=0.3051, pruned_loss=0.06473, over 6971.00 frames.], tot_loss[loss=0.196, simple_loss=0.2803, pruned_loss=0.05583, over 1381901.29 frames.], batch size: 29, lr: 1.09e-03 2022-05-27 17:05:30,529 INFO [train.py:823] (0/4) Epoch 15, batch 750, loss[loss=0.172, simple_loss=0.2591, pruned_loss=0.04241, over 4728.00 frames.], tot_loss[loss=0.195, simple_loss=0.2796, pruned_loss=0.05524, over 1384849.99 frames.], batch size: 46, lr: 1.09e-03 2022-05-27 17:06:09,171 INFO [train.py:823] (0/4) Epoch 15, batch 800, loss[loss=0.2328, simple_loss=0.3019, pruned_loss=0.08181, over 7195.00 frames.], tot_loss[loss=0.1954, simple_loss=0.2802, pruned_loss=0.05534, over 1390123.25 frames.], batch size: 19, lr: 1.09e-03 2022-05-27 17:06:48,508 INFO [train.py:823] (0/4) Epoch 15, batch 850, loss[loss=0.2262, simple_loss=0.3094, pruned_loss=0.07156, over 7237.00 frames.], tot_loss[loss=0.1958, simple_loss=0.2804, pruned_loss=0.05563, over 1394603.10 frames.], batch size: 25, lr: 1.09e-03 2022-05-27 17:07:27,483 INFO [train.py:823] (0/4) Epoch 15, batch 900, loss[loss=0.2072, simple_loss=0.2919, pruned_loss=0.06124, over 7110.00 frames.], tot_loss[loss=0.1967, simple_loss=0.2813, pruned_loss=0.05602, over 1399998.41 frames.], batch size: 19, lr: 1.09e-03 2022-05-27 17:08:06,669 INFO [train.py:823] (0/4) Epoch 15, batch 950, loss[loss=0.2094, simple_loss=0.2781, pruned_loss=0.07038, over 4862.00 frames.], tot_loss[loss=0.1978, simple_loss=0.2818, pruned_loss=0.05691, over 1381684.76 frames.], batch size: 47, lr: 1.08e-03 2022-05-27 17:08:07,873 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-15.pt 2022-05-27 17:08:19,729 INFO [train.py:823] (0/4) Epoch 16, batch 0, loss[loss=0.2108, simple_loss=0.2831, pruned_loss=0.06929, over 5095.00 frames.], tot_loss[loss=0.2108, simple_loss=0.2831, pruned_loss=0.06929, over 5095.00 frames.], batch size: 47, lr: 1.05e-03 2022-05-27 17:08:59,132 INFO [train.py:823] (0/4) Epoch 16, batch 50, loss[loss=0.1816, simple_loss=0.2542, pruned_loss=0.05451, over 7001.00 frames.], tot_loss[loss=0.1923, simple_loss=0.2763, pruned_loss=0.05418, over 318526.77 frames.], batch size: 16, lr: 1.05e-03 2022-05-27 17:09:38,398 INFO [train.py:823] (0/4) Epoch 16, batch 100, loss[loss=0.1715, simple_loss=0.2663, pruned_loss=0.03835, over 7198.00 frames.], tot_loss[loss=0.1923, simple_loss=0.2764, pruned_loss=0.05409, over 560996.02 frames.], batch size: 19, lr: 1.05e-03 2022-05-27 17:10:18,248 INFO [train.py:823] (0/4) Epoch 16, batch 150, loss[loss=0.1982, simple_loss=0.2891, pruned_loss=0.05359, over 7381.00 frames.], tot_loss[loss=0.1919, simple_loss=0.2771, pruned_loss=0.05333, over 756152.27 frames.], batch size: 19, lr: 1.05e-03 2022-05-27 17:10:57,590 INFO [train.py:823] (0/4) Epoch 16, batch 200, loss[loss=0.2389, simple_loss=0.3203, pruned_loss=0.07881, over 7167.00 frames.], tot_loss[loss=0.1946, simple_loss=0.2788, pruned_loss=0.05521, over 903967.41 frames.], batch size: 23, lr: 1.05e-03 2022-05-27 17:11:36,665 INFO [train.py:823] (0/4) Epoch 16, batch 250, loss[loss=0.1943, simple_loss=0.2771, pruned_loss=0.05574, over 7198.00 frames.], tot_loss[loss=0.1953, simple_loss=0.2794, pruned_loss=0.05563, over 1013375.97 frames.], batch size: 25, lr: 1.04e-03 2022-05-27 17:12:15,986 INFO [train.py:823] (0/4) Epoch 16, batch 300, loss[loss=0.2107, simple_loss=0.2935, pruned_loss=0.06398, over 7268.00 frames.], tot_loss[loss=0.1935, simple_loss=0.2772, pruned_loss=0.05485, over 1105792.01 frames.], batch size: 24, lr: 1.04e-03 2022-05-27 17:12:55,722 INFO [train.py:823] (0/4) Epoch 16, batch 350, loss[loss=0.1867, simple_loss=0.2837, pruned_loss=0.0449, over 7352.00 frames.], tot_loss[loss=0.1945, simple_loss=0.2786, pruned_loss=0.05515, over 1173641.50 frames.], batch size: 23, lr: 1.04e-03 2022-05-27 17:13:34,563 INFO [train.py:823] (0/4) Epoch 16, batch 400, loss[loss=0.1642, simple_loss=0.2567, pruned_loss=0.03585, over 7298.00 frames.], tot_loss[loss=0.1938, simple_loss=0.2781, pruned_loss=0.05471, over 1228199.73 frames.], batch size: 19, lr: 1.04e-03 2022-05-27 17:14:13,818 INFO [train.py:823] (0/4) Epoch 16, batch 450, loss[loss=0.2044, simple_loss=0.299, pruned_loss=0.05489, over 7420.00 frames.], tot_loss[loss=0.1944, simple_loss=0.2786, pruned_loss=0.05514, over 1275513.44 frames.], batch size: 22, lr: 1.04e-03 2022-05-27 17:14:53,406 INFO [train.py:823] (0/4) Epoch 16, batch 500, loss[loss=0.2072, simple_loss=0.2958, pruned_loss=0.0593, over 6933.00 frames.], tot_loss[loss=0.1943, simple_loss=0.2786, pruned_loss=0.05502, over 1311835.38 frames.], batch size: 29, lr: 1.04e-03 2022-05-27 17:15:32,892 INFO [train.py:823] (0/4) Epoch 16, batch 550, loss[loss=0.2473, simple_loss=0.3356, pruned_loss=0.07953, over 7374.00 frames.], tot_loss[loss=0.1956, simple_loss=0.2802, pruned_loss=0.05552, over 1330599.95 frames.], batch size: 21, lr: 1.03e-03 2022-05-27 17:16:11,581 INFO [train.py:823] (0/4) Epoch 16, batch 600, loss[loss=0.1769, simple_loss=0.2614, pruned_loss=0.04618, over 7102.00 frames.], tot_loss[loss=0.1952, simple_loss=0.2799, pruned_loss=0.05525, over 1345800.52 frames.], batch size: 19, lr: 1.03e-03 2022-05-27 17:16:52,591 INFO [train.py:823] (0/4) Epoch 16, batch 650, loss[loss=0.1912, simple_loss=0.2772, pruned_loss=0.05263, over 7213.00 frames.], tot_loss[loss=0.197, simple_loss=0.2814, pruned_loss=0.05632, over 1362052.58 frames.], batch size: 16, lr: 1.03e-03 2022-05-27 17:17:31,504 INFO [train.py:823] (0/4) Epoch 16, batch 700, loss[loss=0.1915, simple_loss=0.2807, pruned_loss=0.05115, over 7287.00 frames.], tot_loss[loss=0.1959, simple_loss=0.2803, pruned_loss=0.05572, over 1371167.51 frames.], batch size: 19, lr: 1.03e-03 2022-05-27 17:18:12,140 INFO [train.py:823] (0/4) Epoch 16, batch 750, loss[loss=0.1641, simple_loss=0.2554, pruned_loss=0.03647, over 7192.00 frames.], tot_loss[loss=0.1962, simple_loss=0.2806, pruned_loss=0.05584, over 1384248.13 frames.], batch size: 18, lr: 1.03e-03 2022-05-27 17:18:51,381 INFO [train.py:823] (0/4) Epoch 16, batch 800, loss[loss=0.1739, simple_loss=0.2622, pruned_loss=0.04278, over 7374.00 frames.], tot_loss[loss=0.1951, simple_loss=0.2798, pruned_loss=0.05522, over 1393902.45 frames.], batch size: 20, lr: 1.03e-03 2022-05-27 17:19:30,853 INFO [train.py:823] (0/4) Epoch 16, batch 850, loss[loss=0.1873, simple_loss=0.2731, pruned_loss=0.05077, over 7191.00 frames.], tot_loss[loss=0.1944, simple_loss=0.2793, pruned_loss=0.05476, over 1400752.81 frames.], batch size: 21, lr: 1.03e-03 2022-05-27 17:20:11,122 INFO [train.py:823] (0/4) Epoch 16, batch 900, loss[loss=0.1853, simple_loss=0.2722, pruned_loss=0.04914, over 7427.00 frames.], tot_loss[loss=0.1937, simple_loss=0.2786, pruned_loss=0.05444, over 1401128.73 frames.], batch size: 18, lr: 1.02e-03 2022-05-27 17:20:50,524 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-16.pt 2022-05-27 17:21:02,865 INFO [train.py:823] (0/4) Epoch 17, batch 0, loss[loss=0.2127, simple_loss=0.3002, pruned_loss=0.06263, over 7194.00 frames.], tot_loss[loss=0.2127, simple_loss=0.3002, pruned_loss=0.06263, over 7194.00 frames.], batch size: 21, lr: 9.94e-04 2022-05-27 17:21:41,997 INFO [train.py:823] (0/4) Epoch 17, batch 50, loss[loss=0.177, simple_loss=0.2615, pruned_loss=0.0462, over 7009.00 frames.], tot_loss[loss=0.1992, simple_loss=0.2825, pruned_loss=0.05796, over 315674.40 frames.], batch size: 26, lr: 9.92e-04 2022-05-27 17:22:21,439 INFO [train.py:823] (0/4) Epoch 17, batch 100, loss[loss=0.206, simple_loss=0.2902, pruned_loss=0.06086, over 7028.00 frames.], tot_loss[loss=0.195, simple_loss=0.2798, pruned_loss=0.05507, over 561127.59 frames.], batch size: 26, lr: 9.91e-04 2022-05-27 17:23:00,055 INFO [train.py:823] (0/4) Epoch 17, batch 150, loss[loss=0.1743, simple_loss=0.2586, pruned_loss=0.04502, over 7180.00 frames.], tot_loss[loss=0.1937, simple_loss=0.2797, pruned_loss=0.05384, over 748034.30 frames.], batch size: 18, lr: 9.89e-04 2022-05-27 17:23:39,043 INFO [train.py:823] (0/4) Epoch 17, batch 200, loss[loss=0.1847, simple_loss=0.2752, pruned_loss=0.04706, over 6954.00 frames.], tot_loss[loss=0.1931, simple_loss=0.2793, pruned_loss=0.05349, over 897084.83 frames.], batch size: 29, lr: 9.88e-04 2022-05-27 17:24:17,995 INFO [train.py:823] (0/4) Epoch 17, batch 250, loss[loss=0.2185, simple_loss=0.3047, pruned_loss=0.06616, over 7335.00 frames.], tot_loss[loss=0.1919, simple_loss=0.2783, pruned_loss=0.05279, over 1017589.17 frames.], batch size: 23, lr: 9.86e-04 2022-05-27 17:24:57,786 INFO [train.py:823] (0/4) Epoch 17, batch 300, loss[loss=0.2111, simple_loss=0.2853, pruned_loss=0.06846, over 7311.00 frames.], tot_loss[loss=0.1922, simple_loss=0.2784, pruned_loss=0.05305, over 1103993.07 frames.], batch size: 18, lr: 9.85e-04 2022-05-27 17:25:36,775 INFO [train.py:823] (0/4) Epoch 17, batch 350, loss[loss=0.1952, simple_loss=0.2648, pruned_loss=0.06276, over 7386.00 frames.], tot_loss[loss=0.1914, simple_loss=0.277, pruned_loss=0.05293, over 1170499.78 frames.], batch size: 19, lr: 9.84e-04 2022-05-27 17:26:17,708 INFO [train.py:823] (0/4) Epoch 17, batch 400, loss[loss=0.2196, simple_loss=0.294, pruned_loss=0.07264, over 7096.00 frames.], tot_loss[loss=0.19, simple_loss=0.2753, pruned_loss=0.05235, over 1226564.35 frames.], batch size: 19, lr: 9.82e-04 2022-05-27 17:26:56,055 INFO [train.py:823] (0/4) Epoch 17, batch 450, loss[loss=0.1839, simple_loss=0.2804, pruned_loss=0.04372, over 5111.00 frames.], tot_loss[loss=0.1897, simple_loss=0.2752, pruned_loss=0.05209, over 1261236.01 frames.], batch size: 47, lr: 9.81e-04 2022-05-27 17:27:35,515 INFO [train.py:823] (0/4) Epoch 17, batch 500, loss[loss=0.1736, simple_loss=0.2495, pruned_loss=0.04887, over 7024.00 frames.], tot_loss[loss=0.1897, simple_loss=0.2754, pruned_loss=0.05198, over 1296723.05 frames.], batch size: 16, lr: 9.79e-04 2022-05-27 17:28:14,713 INFO [train.py:823] (0/4) Epoch 17, batch 550, loss[loss=0.2041, simple_loss=0.2989, pruned_loss=0.05465, over 7120.00 frames.], tot_loss[loss=0.1903, simple_loss=0.2755, pruned_loss=0.05257, over 1326053.24 frames.], batch size: 20, lr: 9.78e-04 2022-05-27 17:28:54,204 INFO [train.py:823] (0/4) Epoch 17, batch 600, loss[loss=0.1815, simple_loss=0.2693, pruned_loss=0.04683, over 7290.00 frames.], tot_loss[loss=0.1906, simple_loss=0.2755, pruned_loss=0.05279, over 1348646.57 frames.], batch size: 22, lr: 9.76e-04 2022-05-27 17:29:33,209 INFO [train.py:823] (0/4) Epoch 17, batch 650, loss[loss=0.158, simple_loss=0.2431, pruned_loss=0.0364, over 7025.00 frames.], tot_loss[loss=0.1896, simple_loss=0.2751, pruned_loss=0.05205, over 1361831.91 frames.], batch size: 16, lr: 9.75e-04 2022-05-27 17:30:12,620 INFO [train.py:823] (0/4) Epoch 17, batch 700, loss[loss=0.1869, simple_loss=0.2622, pruned_loss=0.05578, over 7218.00 frames.], tot_loss[loss=0.1899, simple_loss=0.2751, pruned_loss=0.05231, over 1373935.38 frames.], batch size: 16, lr: 9.74e-04 2022-05-27 17:30:51,270 INFO [train.py:823] (0/4) Epoch 17, batch 750, loss[loss=0.1561, simple_loss=0.2405, pruned_loss=0.03586, over 7146.00 frames.], tot_loss[loss=0.1898, simple_loss=0.2751, pruned_loss=0.05227, over 1385769.99 frames.], batch size: 17, lr: 9.72e-04 2022-05-27 17:31:30,947 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/checkpoint-16000.pt 2022-05-27 17:31:34,698 INFO [train.py:823] (0/4) Epoch 17, batch 800, loss[loss=0.1734, simple_loss=0.2589, pruned_loss=0.04402, over 7006.00 frames.], tot_loss[loss=0.1899, simple_loss=0.2754, pruned_loss=0.05221, over 1388684.50 frames.], batch size: 16, lr: 9.71e-04 2022-05-27 17:32:13,777 INFO [train.py:823] (0/4) Epoch 17, batch 850, loss[loss=0.1924, simple_loss=0.2876, pruned_loss=0.04864, over 7412.00 frames.], tot_loss[loss=0.1892, simple_loss=0.2746, pruned_loss=0.05193, over 1394909.89 frames.], batch size: 22, lr: 9.69e-04 2022-05-27 17:32:53,193 INFO [train.py:823] (0/4) Epoch 17, batch 900, loss[loss=0.1617, simple_loss=0.2399, pruned_loss=0.04179, over 7281.00 frames.], tot_loss[loss=0.1897, simple_loss=0.2747, pruned_loss=0.05234, over 1401149.05 frames.], batch size: 17, lr: 9.68e-04 2022-05-27 17:33:32,085 INFO [train.py:823] (0/4) Epoch 17, batch 950, loss[loss=0.1942, simple_loss=0.2824, pruned_loss=0.05303, over 4736.00 frames.], tot_loss[loss=0.1898, simple_loss=0.2748, pruned_loss=0.05243, over 1396143.38 frames.], batch size: 48, lr: 9.67e-04 2022-05-27 17:33:33,406 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-17.pt 2022-05-27 17:33:44,910 INFO [train.py:823] (0/4) Epoch 18, batch 0, loss[loss=0.1989, simple_loss=0.2888, pruned_loss=0.05453, over 7377.00 frames.], tot_loss[loss=0.1989, simple_loss=0.2888, pruned_loss=0.05453, over 7377.00 frames.], batch size: 21, lr: 9.41e-04 2022-05-27 17:34:24,249 INFO [train.py:823] (0/4) Epoch 18, batch 50, loss[loss=0.1873, simple_loss=0.2729, pruned_loss=0.05089, over 7346.00 frames.], tot_loss[loss=0.1854, simple_loss=0.2718, pruned_loss=0.04948, over 321569.21 frames.], batch size: 23, lr: 9.40e-04 2022-05-27 17:35:03,330 INFO [train.py:823] (0/4) Epoch 18, batch 100, loss[loss=0.2028, simple_loss=0.2835, pruned_loss=0.06108, over 7280.00 frames.], tot_loss[loss=0.1868, simple_loss=0.2739, pruned_loss=0.0498, over 562834.09 frames.], batch size: 20, lr: 9.39e-04 2022-05-27 17:35:42,635 INFO [train.py:823] (0/4) Epoch 18, batch 150, loss[loss=0.2173, simple_loss=0.2999, pruned_loss=0.06735, over 7191.00 frames.], tot_loss[loss=0.1877, simple_loss=0.2741, pruned_loss=0.05071, over 757500.62 frames.], batch size: 20, lr: 9.37e-04 2022-05-27 17:36:21,772 INFO [train.py:823] (0/4) Epoch 18, batch 200, loss[loss=0.1867, simple_loss=0.2773, pruned_loss=0.04803, over 7287.00 frames.], tot_loss[loss=0.1879, simple_loss=0.2738, pruned_loss=0.05102, over 907976.04 frames.], batch size: 21, lr: 9.36e-04 2022-05-27 17:37:01,141 INFO [train.py:823] (0/4) Epoch 18, batch 250, loss[loss=0.198, simple_loss=0.2815, pruned_loss=0.05724, over 7314.00 frames.], tot_loss[loss=0.1909, simple_loss=0.2764, pruned_loss=0.05273, over 1016133.17 frames.], batch size: 22, lr: 9.35e-04 2022-05-27 17:37:40,124 INFO [train.py:823] (0/4) Epoch 18, batch 300, loss[loss=0.1722, simple_loss=0.25, pruned_loss=0.04719, over 7032.00 frames.], tot_loss[loss=0.1894, simple_loss=0.2747, pruned_loss=0.05208, over 1105769.61 frames.], batch size: 17, lr: 9.33e-04 2022-05-27 17:38:19,148 INFO [train.py:823] (0/4) Epoch 18, batch 350, loss[loss=0.1819, simple_loss=0.2707, pruned_loss=0.04654, over 7296.00 frames.], tot_loss[loss=0.1888, simple_loss=0.2744, pruned_loss=0.05155, over 1176114.49 frames.], batch size: 20, lr: 9.32e-04 2022-05-27 17:38:58,304 INFO [train.py:823] (0/4) Epoch 18, batch 400, loss[loss=0.181, simple_loss=0.2633, pruned_loss=0.0493, over 7390.00 frames.], tot_loss[loss=0.1898, simple_loss=0.2757, pruned_loss=0.0519, over 1227518.10 frames.], batch size: 19, lr: 9.31e-04 2022-05-27 17:39:39,178 INFO [train.py:823] (0/4) Epoch 18, batch 450, loss[loss=0.1967, simple_loss=0.273, pruned_loss=0.06021, over 7144.00 frames.], tot_loss[loss=0.1894, simple_loss=0.2756, pruned_loss=0.0516, over 1269528.99 frames.], batch size: 23, lr: 9.29e-04 2022-05-27 17:40:18,334 INFO [train.py:823] (0/4) Epoch 18, batch 500, loss[loss=0.2028, simple_loss=0.2962, pruned_loss=0.0547, over 7431.00 frames.], tot_loss[loss=0.189, simple_loss=0.2751, pruned_loss=0.05148, over 1307298.45 frames.], batch size: 22, lr: 9.28e-04 2022-05-27 17:40:58,983 INFO [train.py:823] (0/4) Epoch 18, batch 550, loss[loss=0.1869, simple_loss=0.274, pruned_loss=0.04993, over 7362.00 frames.], tot_loss[loss=0.1873, simple_loss=0.2732, pruned_loss=0.05067, over 1335421.71 frames.], batch size: 23, lr: 9.27e-04 2022-05-27 17:41:37,890 INFO [train.py:823] (0/4) Epoch 18, batch 600, loss[loss=0.1773, simple_loss=0.2625, pruned_loss=0.04607, over 7296.00 frames.], tot_loss[loss=0.1871, simple_loss=0.2728, pruned_loss=0.05067, over 1357597.31 frames.], batch size: 19, lr: 9.26e-04 2022-05-27 17:42:17,207 INFO [train.py:823] (0/4) Epoch 18, batch 650, loss[loss=0.1736, simple_loss=0.2663, pruned_loss=0.04045, over 7095.00 frames.], tot_loss[loss=0.1863, simple_loss=0.2727, pruned_loss=0.04996, over 1371667.89 frames.], batch size: 19, lr: 9.24e-04 2022-05-27 17:42:57,774 INFO [train.py:823] (0/4) Epoch 18, batch 700, loss[loss=0.1853, simple_loss=0.2692, pruned_loss=0.05065, over 7191.00 frames.], tot_loss[loss=0.1863, simple_loss=0.2724, pruned_loss=0.05007, over 1375644.99 frames.], batch size: 19, lr: 9.23e-04 2022-05-27 17:43:37,096 INFO [train.py:823] (0/4) Epoch 18, batch 750, loss[loss=0.1772, simple_loss=0.2556, pruned_loss=0.04937, over 7103.00 frames.], tot_loss[loss=0.1875, simple_loss=0.2736, pruned_loss=0.05069, over 1387610.18 frames.], batch size: 18, lr: 9.22e-04 2022-05-27 17:44:16,117 INFO [train.py:823] (0/4) Epoch 18, batch 800, loss[loss=0.1718, simple_loss=0.2553, pruned_loss=0.04422, over 7199.00 frames.], tot_loss[loss=0.1872, simple_loss=0.2731, pruned_loss=0.05066, over 1391007.15 frames.], batch size: 20, lr: 9.21e-04 2022-05-27 17:44:55,760 INFO [train.py:823] (0/4) Epoch 18, batch 850, loss[loss=0.1809, simple_loss=0.2794, pruned_loss=0.04117, over 7183.00 frames.], tot_loss[loss=0.186, simple_loss=0.272, pruned_loss=0.04997, over 1395040.56 frames.], batch size: 21, lr: 9.19e-04 2022-05-27 17:45:34,706 INFO [train.py:823] (0/4) Epoch 18, batch 900, loss[loss=0.1558, simple_loss=0.2467, pruned_loss=0.03245, over 7143.00 frames.], tot_loss[loss=0.1865, simple_loss=0.2727, pruned_loss=0.0502, over 1401683.63 frames.], batch size: 17, lr: 9.18e-04 2022-05-27 17:46:13,678 INFO [train.py:823] (0/4) Epoch 18, batch 950, loss[loss=0.1807, simple_loss=0.26, pruned_loss=0.05067, over 4612.00 frames.], tot_loss[loss=0.1868, simple_loss=0.2727, pruned_loss=0.05044, over 1373629.33 frames.], batch size: 46, lr: 9.17e-04 2022-05-27 17:46:14,982 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-18.pt 2022-05-27 17:46:27,023 INFO [train.py:823] (0/4) Epoch 19, batch 0, loss[loss=0.1897, simple_loss=0.2885, pruned_loss=0.04543, over 7014.00 frames.], tot_loss[loss=0.1897, simple_loss=0.2885, pruned_loss=0.04543, over 7014.00 frames.], batch size: 26, lr: 8.94e-04 2022-05-27 17:47:05,747 INFO [train.py:823] (0/4) Epoch 19, batch 50, loss[loss=0.1676, simple_loss=0.2529, pruned_loss=0.04109, over 7190.00 frames.], tot_loss[loss=0.1861, simple_loss=0.2711, pruned_loss=0.05058, over 325407.41 frames.], batch size: 19, lr: 8.92e-04 2022-05-27 17:47:44,971 INFO [train.py:823] (0/4) Epoch 19, batch 100, loss[loss=0.1861, simple_loss=0.2682, pruned_loss=0.05203, over 6528.00 frames.], tot_loss[loss=0.186, simple_loss=0.2713, pruned_loss=0.05038, over 565915.06 frames.], batch size: 34, lr: 8.91e-04 2022-05-27 17:48:24,263 INFO [train.py:823] (0/4) Epoch 19, batch 150, loss[loss=0.1596, simple_loss=0.2491, pruned_loss=0.03506, over 7088.00 frames.], tot_loss[loss=0.183, simple_loss=0.2681, pruned_loss=0.04895, over 758742.98 frames.], batch size: 18, lr: 8.90e-04 2022-05-27 17:49:03,194 INFO [train.py:823] (0/4) Epoch 19, batch 200, loss[loss=0.2035, simple_loss=0.292, pruned_loss=0.05749, over 7153.00 frames.], tot_loss[loss=0.1835, simple_loss=0.269, pruned_loss=0.04906, over 901426.56 frames.], batch size: 22, lr: 8.89e-04 2022-05-27 17:49:41,875 INFO [train.py:823] (0/4) Epoch 19, batch 250, loss[loss=0.1553, simple_loss=0.2545, pruned_loss=0.02805, over 7091.00 frames.], tot_loss[loss=0.1843, simple_loss=0.2704, pruned_loss=0.04908, over 1017612.28 frames.], batch size: 19, lr: 8.88e-04 2022-05-27 17:50:22,321 INFO [train.py:823] (0/4) Epoch 19, batch 300, loss[loss=0.168, simple_loss=0.2496, pruned_loss=0.04317, over 7010.00 frames.], tot_loss[loss=0.186, simple_loss=0.2714, pruned_loss=0.05032, over 1109230.37 frames.], batch size: 16, lr: 8.87e-04 2022-05-27 17:51:01,421 INFO [train.py:823] (0/4) Epoch 19, batch 350, loss[loss=0.2842, simple_loss=0.3478, pruned_loss=0.1103, over 7311.00 frames.], tot_loss[loss=0.1866, simple_loss=0.272, pruned_loss=0.05055, over 1176559.91 frames.], batch size: 18, lr: 8.85e-04 2022-05-27 17:51:40,814 INFO [train.py:823] (0/4) Epoch 19, batch 400, loss[loss=0.149, simple_loss=0.2318, pruned_loss=0.03307, over 7024.00 frames.], tot_loss[loss=0.1859, simple_loss=0.272, pruned_loss=0.04985, over 1234519.25 frames.], batch size: 16, lr: 8.84e-04 2022-05-27 17:52:20,011 INFO [train.py:823] (0/4) Epoch 19, batch 450, loss[loss=0.1926, simple_loss=0.2807, pruned_loss=0.05228, over 7143.00 frames.], tot_loss[loss=0.1866, simple_loss=0.2728, pruned_loss=0.05018, over 1276543.88 frames.], batch size: 23, lr: 8.83e-04 2022-05-27 17:52:59,875 INFO [train.py:823] (0/4) Epoch 19, batch 500, loss[loss=0.1864, simple_loss=0.2848, pruned_loss=0.04393, over 6471.00 frames.], tot_loss[loss=0.1862, simple_loss=0.2727, pruned_loss=0.04983, over 1309654.32 frames.], batch size: 34, lr: 8.82e-04 2022-05-27 17:53:39,210 INFO [train.py:823] (0/4) Epoch 19, batch 550, loss[loss=0.1772, simple_loss=0.2569, pruned_loss=0.04877, over 7024.00 frames.], tot_loss[loss=0.185, simple_loss=0.2719, pruned_loss=0.04907, over 1331356.52 frames.], batch size: 17, lr: 8.81e-04 2022-05-27 17:54:18,342 INFO [train.py:823] (0/4) Epoch 19, batch 600, loss[loss=0.1799, simple_loss=0.2646, pruned_loss=0.04766, over 7095.00 frames.], tot_loss[loss=0.1853, simple_loss=0.2726, pruned_loss=0.04896, over 1351567.21 frames.], batch size: 19, lr: 8.80e-04 2022-05-27 17:54:57,232 INFO [train.py:823] (0/4) Epoch 19, batch 650, loss[loss=0.1604, simple_loss=0.2441, pruned_loss=0.0383, over 7018.00 frames.], tot_loss[loss=0.1851, simple_loss=0.2719, pruned_loss=0.04914, over 1365541.40 frames.], batch size: 17, lr: 8.78e-04 2022-05-27 17:55:36,372 INFO [train.py:823] (0/4) Epoch 19, batch 700, loss[loss=0.179, simple_loss=0.2602, pruned_loss=0.04895, over 7038.00 frames.], tot_loss[loss=0.1837, simple_loss=0.2706, pruned_loss=0.04844, over 1377840.23 frames.], batch size: 26, lr: 8.77e-04 2022-05-27 17:56:14,895 INFO [train.py:823] (0/4) Epoch 19, batch 750, loss[loss=0.195, simple_loss=0.2882, pruned_loss=0.05093, over 7370.00 frames.], tot_loss[loss=0.1846, simple_loss=0.2713, pruned_loss=0.04894, over 1388028.33 frames.], batch size: 21, lr: 8.76e-04 2022-05-27 17:56:54,504 INFO [train.py:823] (0/4) Epoch 19, batch 800, loss[loss=0.2171, simple_loss=0.2938, pruned_loss=0.07019, over 7301.00 frames.], tot_loss[loss=0.1844, simple_loss=0.271, pruned_loss=0.04892, over 1397424.01 frames.], batch size: 22, lr: 8.75e-04 2022-05-27 17:57:33,430 INFO [train.py:823] (0/4) Epoch 19, batch 850, loss[loss=0.2287, simple_loss=0.3079, pruned_loss=0.07477, over 7369.00 frames.], tot_loss[loss=0.1849, simple_loss=0.2716, pruned_loss=0.04908, over 1402557.55 frames.], batch size: 21, lr: 8.74e-04 2022-05-27 17:58:12,600 INFO [train.py:823] (0/4) Epoch 19, batch 900, loss[loss=0.2069, simple_loss=0.3035, pruned_loss=0.05518, over 6989.00 frames.], tot_loss[loss=0.1849, simple_loss=0.2714, pruned_loss=0.04916, over 1394423.79 frames.], batch size: 26, lr: 8.73e-04 2022-05-27 17:58:50,874 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-19.pt 2022-05-27 17:59:02,775 INFO [train.py:823] (0/4) Epoch 20, batch 0, loss[loss=0.1504, simple_loss=0.2382, pruned_loss=0.03131, over 6403.00 frames.], tot_loss[loss=0.1504, simple_loss=0.2382, pruned_loss=0.03131, over 6403.00 frames.], batch size: 34, lr: 8.51e-04 2022-05-27 17:59:42,294 INFO [train.py:823] (0/4) Epoch 20, batch 50, loss[loss=0.1629, simple_loss=0.2459, pruned_loss=0.03997, over 7316.00 frames.], tot_loss[loss=0.1793, simple_loss=0.2671, pruned_loss=0.04572, over 322231.09 frames.], batch size: 18, lr: 8.49e-04 2022-05-27 18:00:21,133 INFO [train.py:823] (0/4) Epoch 20, batch 100, loss[loss=0.2006, simple_loss=0.2758, pruned_loss=0.06273, over 5098.00 frames.], tot_loss[loss=0.1818, simple_loss=0.2687, pruned_loss=0.04744, over 562810.51 frames.], batch size: 47, lr: 8.48e-04 2022-05-27 18:01:00,387 INFO [train.py:823] (0/4) Epoch 20, batch 150, loss[loss=0.1878, simple_loss=0.2504, pruned_loss=0.06257, over 7290.00 frames.], tot_loss[loss=0.1826, simple_loss=0.2689, pruned_loss=0.04816, over 750807.12 frames.], batch size: 17, lr: 8.47e-04 2022-05-27 18:01:41,347 INFO [train.py:823] (0/4) Epoch 20, batch 200, loss[loss=0.1874, simple_loss=0.2688, pruned_loss=0.05304, over 7002.00 frames.], tot_loss[loss=0.1831, simple_loss=0.2694, pruned_loss=0.04841, over 902540.12 frames.], batch size: 16, lr: 8.46e-04 2022-05-27 18:02:20,205 INFO [train.py:823] (0/4) Epoch 20, batch 250, loss[loss=0.1619, simple_loss=0.242, pruned_loss=0.04089, over 7313.00 frames.], tot_loss[loss=0.1839, simple_loss=0.2695, pruned_loss=0.04919, over 1017391.23 frames.], batch size: 18, lr: 8.45e-04 2022-05-27 18:02:59,635 INFO [train.py:823] (0/4) Epoch 20, batch 300, loss[loss=0.1752, simple_loss=0.2679, pruned_loss=0.04123, over 7301.00 frames.], tot_loss[loss=0.1834, simple_loss=0.2696, pruned_loss=0.04856, over 1107033.18 frames.], batch size: 22, lr: 8.44e-04 2022-05-27 18:03:38,764 INFO [train.py:823] (0/4) Epoch 20, batch 350, loss[loss=0.1846, simple_loss=0.2794, pruned_loss=0.04489, over 7193.00 frames.], tot_loss[loss=0.1833, simple_loss=0.2696, pruned_loss=0.04852, over 1176627.07 frames.], batch size: 20, lr: 8.43e-04 2022-05-27 18:04:18,919 INFO [train.py:823] (0/4) Epoch 20, batch 400, loss[loss=0.2062, simple_loss=0.298, pruned_loss=0.05717, over 7167.00 frames.], tot_loss[loss=0.183, simple_loss=0.2693, pruned_loss=0.04829, over 1231527.08 frames.], batch size: 23, lr: 8.42e-04 2022-05-27 18:04:57,927 INFO [train.py:823] (0/4) Epoch 20, batch 450, loss[loss=0.1699, simple_loss=0.2442, pruned_loss=0.04783, over 7178.00 frames.], tot_loss[loss=0.1833, simple_loss=0.27, pruned_loss=0.04827, over 1270008.98 frames.], batch size: 17, lr: 8.41e-04 2022-05-27 18:05:38,460 INFO [train.py:823] (0/4) Epoch 20, batch 500, loss[loss=0.1786, simple_loss=0.2623, pruned_loss=0.04742, over 7430.00 frames.], tot_loss[loss=0.1829, simple_loss=0.2699, pruned_loss=0.04792, over 1305734.98 frames.], batch size: 18, lr: 8.40e-04 2022-05-27 18:06:18,318 INFO [train.py:823] (0/4) Epoch 20, batch 550, loss[loss=0.1966, simple_loss=0.2884, pruned_loss=0.05235, over 7153.00 frames.], tot_loss[loss=0.1825, simple_loss=0.269, pruned_loss=0.04801, over 1333804.95 frames.], batch size: 23, lr: 8.39e-04 2022-05-27 18:06:57,346 INFO [train.py:823] (0/4) Epoch 20, batch 600, loss[loss=0.1743, simple_loss=0.257, pruned_loss=0.04579, over 7095.00 frames.], tot_loss[loss=0.1826, simple_loss=0.2691, pruned_loss=0.0481, over 1348871.07 frames.], batch size: 18, lr: 8.38e-04 2022-05-27 18:07:37,165 INFO [train.py:823] (0/4) Epoch 20, batch 650, loss[loss=0.166, simple_loss=0.2619, pruned_loss=0.03506, over 6906.00 frames.], tot_loss[loss=0.1827, simple_loss=0.2692, pruned_loss=0.04812, over 1364727.51 frames.], batch size: 29, lr: 8.37e-04 2022-05-27 18:08:16,118 INFO [train.py:823] (0/4) Epoch 20, batch 700, loss[loss=0.1577, simple_loss=0.2406, pruned_loss=0.03745, over 7088.00 frames.], tot_loss[loss=0.1815, simple_loss=0.2684, pruned_loss=0.04733, over 1378908.50 frames.], batch size: 18, lr: 8.36e-04 2022-05-27 18:08:55,659 INFO [train.py:823] (0/4) Epoch 20, batch 750, loss[loss=0.1795, simple_loss=0.2649, pruned_loss=0.04703, over 7290.00 frames.], tot_loss[loss=0.181, simple_loss=0.2676, pruned_loss=0.04724, over 1389389.22 frames.], batch size: 21, lr: 8.35e-04 2022-05-27 18:09:34,501 INFO [train.py:823] (0/4) Epoch 20, batch 800, loss[loss=0.1486, simple_loss=0.222, pruned_loss=0.03763, over 7019.00 frames.], tot_loss[loss=0.1815, simple_loss=0.2683, pruned_loss=0.04731, over 1397525.28 frames.], batch size: 17, lr: 8.34e-04 2022-05-27 18:10:13,635 INFO [train.py:823] (0/4) Epoch 20, batch 850, loss[loss=0.1796, simple_loss=0.2709, pruned_loss=0.04414, over 6998.00 frames.], tot_loss[loss=0.1812, simple_loss=0.2676, pruned_loss=0.04735, over 1400422.08 frames.], batch size: 26, lr: 8.33e-04 2022-05-27 18:10:52,728 INFO [train.py:823] (0/4) Epoch 20, batch 900, loss[loss=0.1611, simple_loss=0.2403, pruned_loss=0.04095, over 7201.00 frames.], tot_loss[loss=0.1812, simple_loss=0.2675, pruned_loss=0.04741, over 1397818.99 frames.], batch size: 16, lr: 8.31e-04 2022-05-27 18:11:30,992 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-20.pt 2022-05-27 18:11:42,878 INFO [train.py:823] (0/4) Epoch 21, batch 0, loss[loss=0.1642, simple_loss=0.2487, pruned_loss=0.03989, over 7187.00 frames.], tot_loss[loss=0.1642, simple_loss=0.2487, pruned_loss=0.03989, over 7187.00 frames.], batch size: 18, lr: 8.11e-04 2022-05-27 18:12:21,434 INFO [train.py:823] (0/4) Epoch 21, batch 50, loss[loss=0.194, simple_loss=0.2831, pruned_loss=0.05246, over 7203.00 frames.], tot_loss[loss=0.1805, simple_loss=0.2656, pruned_loss=0.04772, over 317930.57 frames.], batch size: 25, lr: 8.10e-04 2022-05-27 18:13:00,845 INFO [train.py:823] (0/4) Epoch 21, batch 100, loss[loss=0.1782, simple_loss=0.2685, pruned_loss=0.04392, over 6585.00 frames.], tot_loss[loss=0.1779, simple_loss=0.2643, pruned_loss=0.04578, over 562256.16 frames.], batch size: 34, lr: 8.09e-04 2022-05-27 18:13:40,038 INFO [train.py:823] (0/4) Epoch 21, batch 150, loss[loss=0.1935, simple_loss=0.2688, pruned_loss=0.05912, over 7277.00 frames.], tot_loss[loss=0.179, simple_loss=0.2653, pruned_loss=0.04633, over 755931.62 frames.], batch size: 20, lr: 8.08e-04 2022-05-27 18:14:20,277 INFO [train.py:823] (0/4) Epoch 21, batch 200, loss[loss=0.1514, simple_loss=0.225, pruned_loss=0.03891, over 7306.00 frames.], tot_loss[loss=0.1786, simple_loss=0.2655, pruned_loss=0.04585, over 903699.51 frames.], batch size: 18, lr: 8.07e-04 2022-05-27 18:14:59,158 INFO [train.py:823] (0/4) Epoch 21, batch 250, loss[loss=0.1775, simple_loss=0.2637, pruned_loss=0.04566, over 7284.00 frames.], tot_loss[loss=0.1785, simple_loss=0.2654, pruned_loss=0.04583, over 1011194.91 frames.], batch size: 20, lr: 8.06e-04 2022-05-27 18:15:38,177 INFO [train.py:823] (0/4) Epoch 21, batch 300, loss[loss=0.1898, simple_loss=0.2781, pruned_loss=0.05073, over 6567.00 frames.], tot_loss[loss=0.1806, simple_loss=0.2673, pruned_loss=0.04698, over 1100172.24 frames.], batch size: 34, lr: 8.05e-04 2022-05-27 18:16:17,448 INFO [train.py:823] (0/4) Epoch 21, batch 350, loss[loss=0.2143, simple_loss=0.2897, pruned_loss=0.06946, over 7415.00 frames.], tot_loss[loss=0.1809, simple_loss=0.2677, pruned_loss=0.04705, over 1171574.29 frames.], batch size: 22, lr: 8.04e-04 2022-05-27 18:16:56,870 INFO [train.py:823] (0/4) Epoch 21, batch 400, loss[loss=0.1419, simple_loss=0.2174, pruned_loss=0.03319, over 7297.00 frames.], tot_loss[loss=0.1815, simple_loss=0.2685, pruned_loss=0.04721, over 1226977.76 frames.], batch size: 17, lr: 8.03e-04 2022-05-27 18:17:36,042 INFO [train.py:823] (0/4) Epoch 21, batch 450, loss[loss=0.1769, simple_loss=0.273, pruned_loss=0.04039, over 7188.00 frames.], tot_loss[loss=0.1812, simple_loss=0.2685, pruned_loss=0.04692, over 1270534.63 frames.], batch size: 21, lr: 8.02e-04 2022-05-27 18:18:15,648 INFO [train.py:823] (0/4) Epoch 21, batch 500, loss[loss=0.1707, simple_loss=0.2546, pruned_loss=0.0434, over 7181.00 frames.], tot_loss[loss=0.1808, simple_loss=0.2683, pruned_loss=0.04669, over 1303721.79 frames.], batch size: 18, lr: 8.01e-04 2022-05-27 18:18:54,647 INFO [train.py:823] (0/4) Epoch 21, batch 550, loss[loss=0.1891, simple_loss=0.2671, pruned_loss=0.05558, over 7380.00 frames.], tot_loss[loss=0.1796, simple_loss=0.2672, pruned_loss=0.04601, over 1335295.93 frames.], batch size: 21, lr: 8.00e-04 2022-05-27 18:19:34,106 INFO [train.py:823] (0/4) Epoch 21, batch 600, loss[loss=0.2058, simple_loss=0.2935, pruned_loss=0.05907, over 6405.00 frames.], tot_loss[loss=0.1805, simple_loss=0.2677, pruned_loss=0.04663, over 1354161.81 frames.], batch size: 34, lr: 8.00e-04 2022-05-27 18:20:13,153 INFO [train.py:823] (0/4) Epoch 21, batch 650, loss[loss=0.1678, simple_loss=0.2704, pruned_loss=0.0326, over 7291.00 frames.], tot_loss[loss=0.1807, simple_loss=0.268, pruned_loss=0.04667, over 1369486.73 frames.], batch size: 22, lr: 7.99e-04 2022-05-27 18:20:52,858 INFO [train.py:823] (0/4) Epoch 21, batch 700, loss[loss=0.1659, simple_loss=0.2566, pruned_loss=0.03757, over 7194.00 frames.], tot_loss[loss=0.1811, simple_loss=0.2682, pruned_loss=0.04701, over 1380593.03 frames.], batch size: 20, lr: 7.98e-04 2022-05-27 18:21:31,547 INFO [train.py:823] (0/4) Epoch 21, batch 750, loss[loss=0.186, simple_loss=0.2813, pruned_loss=0.0454, over 7212.00 frames.], tot_loss[loss=0.1799, simple_loss=0.2668, pruned_loss=0.04647, over 1379377.52 frames.], batch size: 25, lr: 7.97e-04 2022-05-27 18:22:11,043 INFO [train.py:823] (0/4) Epoch 21, batch 800, loss[loss=0.2187, simple_loss=0.3063, pruned_loss=0.06552, over 7337.00 frames.], tot_loss[loss=0.1795, simple_loss=0.2668, pruned_loss=0.04612, over 1384859.83 frames.], batch size: 23, lr: 7.96e-04 2022-05-27 18:22:50,060 INFO [train.py:823] (0/4) Epoch 21, batch 850, loss[loss=0.181, simple_loss=0.2754, pruned_loss=0.04327, over 7197.00 frames.], tot_loss[loss=0.1802, simple_loss=0.267, pruned_loss=0.04667, over 1389671.90 frames.], batch size: 20, lr: 7.95e-04 2022-05-27 18:23:29,436 INFO [train.py:823] (0/4) Epoch 21, batch 900, loss[loss=0.1592, simple_loss=0.2446, pruned_loss=0.03693, over 7383.00 frames.], tot_loss[loss=0.1802, simple_loss=0.267, pruned_loss=0.04674, over 1388439.81 frames.], batch size: 20, lr: 7.94e-04 2022-05-27 18:24:08,159 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-21.pt 2022-05-27 18:24:19,604 INFO [train.py:823] (0/4) Epoch 22, batch 0, loss[loss=0.1631, simple_loss=0.2629, pruned_loss=0.03162, over 7372.00 frames.], tot_loss[loss=0.1631, simple_loss=0.2629, pruned_loss=0.03162, over 7372.00 frames.], batch size: 21, lr: 7.75e-04 2022-05-27 18:25:00,022 INFO [train.py:823] (0/4) Epoch 22, batch 50, loss[loss=0.2027, simple_loss=0.2888, pruned_loss=0.05835, over 7175.00 frames.], tot_loss[loss=0.1788, simple_loss=0.2659, pruned_loss=0.0459, over 322024.20 frames.], batch size: 22, lr: 7.74e-04 2022-05-27 18:25:39,865 INFO [train.py:823] (0/4) Epoch 22, batch 100, loss[loss=0.1921, simple_loss=0.2768, pruned_loss=0.05374, over 7110.00 frames.], tot_loss[loss=0.1783, simple_loss=0.2659, pruned_loss=0.04538, over 567534.92 frames.], batch size: 20, lr: 7.73e-04 2022-05-27 18:26:18,846 INFO [train.py:823] (0/4) Epoch 22, batch 150, loss[loss=0.1932, simple_loss=0.2696, pruned_loss=0.05837, over 5241.00 frames.], tot_loss[loss=0.1805, simple_loss=0.2675, pruned_loss=0.04678, over 755485.70 frames.], batch size: 47, lr: 7.73e-04 2022-05-27 18:26:59,552 INFO [train.py:823] (0/4) Epoch 22, batch 200, loss[loss=0.1798, simple_loss=0.2631, pruned_loss=0.04823, over 7118.00 frames.], tot_loss[loss=0.1795, simple_loss=0.2664, pruned_loss=0.04629, over 899416.67 frames.], batch size: 20, lr: 7.72e-04 2022-05-27 18:27:38,546 INFO [train.py:823] (0/4) Epoch 22, batch 250, loss[loss=0.1661, simple_loss=0.2564, pruned_loss=0.03793, over 7098.00 frames.], tot_loss[loss=0.1782, simple_loss=0.2657, pruned_loss=0.04537, over 1016438.08 frames.], batch size: 18, lr: 7.71e-04 2022-05-27 18:28:18,073 INFO [train.py:823] (0/4) Epoch 22, batch 300, loss[loss=0.18, simple_loss=0.2695, pruned_loss=0.04523, over 7193.00 frames.], tot_loss[loss=0.179, simple_loss=0.2665, pruned_loss=0.04578, over 1103534.23 frames.], batch size: 18, lr: 7.70e-04 2022-05-27 18:28:58,662 INFO [train.py:823] (0/4) Epoch 22, batch 350, loss[loss=0.2136, simple_loss=0.2979, pruned_loss=0.06471, over 6958.00 frames.], tot_loss[loss=0.1791, simple_loss=0.2664, pruned_loss=0.04595, over 1174556.41 frames.], batch size: 29, lr: 7.69e-04 2022-05-27 18:29:37,850 INFO [train.py:823] (0/4) Epoch 22, batch 400, loss[loss=0.1862, simple_loss=0.2698, pruned_loss=0.05131, over 7184.00 frames.], tot_loss[loss=0.1778, simple_loss=0.2646, pruned_loss=0.04549, over 1230866.66 frames.], batch size: 21, lr: 7.68e-04 2022-05-27 18:30:17,485 INFO [train.py:823] (0/4) Epoch 22, batch 450, loss[loss=0.1711, simple_loss=0.2434, pruned_loss=0.04936, over 7237.00 frames.], tot_loss[loss=0.178, simple_loss=0.265, pruned_loss=0.04554, over 1276971.09 frames.], batch size: 16, lr: 7.67e-04 2022-05-27 18:30:56,596 INFO [train.py:823] (0/4) Epoch 22, batch 500, loss[loss=0.1716, simple_loss=0.2689, pruned_loss=0.03721, over 6512.00 frames.], tot_loss[loss=0.1767, simple_loss=0.264, pruned_loss=0.04473, over 1302990.77 frames.], batch size: 34, lr: 7.66e-04 2022-05-27 18:31:36,000 INFO [train.py:823] (0/4) Epoch 22, batch 550, loss[loss=0.1652, simple_loss=0.2564, pruned_loss=0.03703, over 6949.00 frames.], tot_loss[loss=0.1762, simple_loss=0.2633, pruned_loss=0.04457, over 1329987.47 frames.], batch size: 29, lr: 7.65e-04 2022-05-27 18:32:15,231 INFO [train.py:823] (0/4) Epoch 22, batch 600, loss[loss=0.1983, simple_loss=0.2731, pruned_loss=0.06172, over 7038.00 frames.], tot_loss[loss=0.1757, simple_loss=0.2629, pruned_loss=0.04431, over 1349114.58 frames.], batch size: 17, lr: 7.65e-04 2022-05-27 18:32:54,264 INFO [train.py:823] (0/4) Epoch 22, batch 650, loss[loss=0.1908, simple_loss=0.2836, pruned_loss=0.04905, over 7108.00 frames.], tot_loss[loss=0.1767, simple_loss=0.2635, pruned_loss=0.04494, over 1358675.83 frames.], batch size: 20, lr: 7.64e-04 2022-05-27 18:33:33,446 INFO [train.py:823] (0/4) Epoch 22, batch 700, loss[loss=0.1813, simple_loss=0.2678, pruned_loss=0.04746, over 7097.00 frames.], tot_loss[loss=0.1769, simple_loss=0.2638, pruned_loss=0.04497, over 1371207.16 frames.], batch size: 19, lr: 7.63e-04 2022-05-27 18:34:12,332 INFO [train.py:823] (0/4) Epoch 22, batch 750, loss[loss=0.1872, simple_loss=0.2644, pruned_loss=0.055, over 7012.00 frames.], tot_loss[loss=0.1764, simple_loss=0.2636, pruned_loss=0.04457, over 1380743.17 frames.], batch size: 16, lr: 7.62e-04 2022-05-27 18:34:51,824 INFO [train.py:823] (0/4) Epoch 22, batch 800, loss[loss=0.1683, simple_loss=0.2477, pruned_loss=0.04443, over 7381.00 frames.], tot_loss[loss=0.1771, simple_loss=0.2645, pruned_loss=0.04488, over 1390137.49 frames.], batch size: 20, lr: 7.61e-04 2022-05-27 18:35:30,969 INFO [train.py:823] (0/4) Epoch 22, batch 850, loss[loss=0.1769, simple_loss=0.2739, pruned_loss=0.03999, over 6534.00 frames.], tot_loss[loss=0.1776, simple_loss=0.2649, pruned_loss=0.04513, over 1398696.02 frames.], batch size: 34, lr: 7.60e-04 2022-05-27 18:36:10,309 INFO [train.py:823] (0/4) Epoch 22, batch 900, loss[loss=0.2004, simple_loss=0.2878, pruned_loss=0.05653, over 7151.00 frames.], tot_loss[loss=0.1786, simple_loss=0.2661, pruned_loss=0.04553, over 1403639.89 frames.], batch size: 23, lr: 7.59e-04 2022-05-27 18:36:49,548 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-22.pt 2022-05-27 18:37:01,207 INFO [train.py:823] (0/4) Epoch 23, batch 0, loss[loss=0.1538, simple_loss=0.2341, pruned_loss=0.03674, over 7217.00 frames.], tot_loss[loss=0.1538, simple_loss=0.2341, pruned_loss=0.03674, over 7217.00 frames.], batch size: 16, lr: 7.42e-04 2022-05-27 18:37:41,611 INFO [train.py:823] (0/4) Epoch 23, batch 50, loss[loss=0.1792, simple_loss=0.2794, pruned_loss=0.03949, over 7369.00 frames.], tot_loss[loss=0.1759, simple_loss=0.2658, pruned_loss=0.04296, over 320355.90 frames.], batch size: 21, lr: 7.41e-04 2022-05-27 18:38:20,829 INFO [train.py:823] (0/4) Epoch 23, batch 100, loss[loss=0.1667, simple_loss=0.2569, pruned_loss=0.03827, over 7380.00 frames.], tot_loss[loss=0.1749, simple_loss=0.2642, pruned_loss=0.04277, over 561612.76 frames.], batch size: 20, lr: 7.41e-04 2022-05-27 18:39:00,151 INFO [train.py:823] (0/4) Epoch 23, batch 150, loss[loss=0.1628, simple_loss=0.2396, pruned_loss=0.04298, over 7306.00 frames.], tot_loss[loss=0.1766, simple_loss=0.2651, pruned_loss=0.04407, over 752438.03 frames.], batch size: 18, lr: 7.40e-04 2022-05-27 18:39:39,814 INFO [train.py:823] (0/4) Epoch 23, batch 200, loss[loss=0.1967, simple_loss=0.2882, pruned_loss=0.05255, over 5231.00 frames.], tot_loss[loss=0.1769, simple_loss=0.2648, pruned_loss=0.04452, over 899493.95 frames.], batch size: 47, lr: 7.39e-04 2022-05-27 18:40:19,147 INFO [train.py:823] (0/4) Epoch 23, batch 250, loss[loss=0.1636, simple_loss=0.249, pruned_loss=0.03911, over 7094.00 frames.], tot_loss[loss=0.177, simple_loss=0.2648, pruned_loss=0.0446, over 1018799.34 frames.], batch size: 18, lr: 7.38e-04 2022-05-27 18:40:58,277 INFO [train.py:823] (0/4) Epoch 23, batch 300, loss[loss=0.1541, simple_loss=0.2525, pruned_loss=0.0278, over 7291.00 frames.], tot_loss[loss=0.1776, simple_loss=0.2659, pruned_loss=0.04469, over 1112054.42 frames.], batch size: 22, lr: 7.37e-04 2022-05-27 18:41:37,499 INFO [train.py:823] (0/4) Epoch 23, batch 350, loss[loss=0.1805, simple_loss=0.2724, pruned_loss=0.04435, over 7273.00 frames.], tot_loss[loss=0.1771, simple_loss=0.2649, pruned_loss=0.04468, over 1182108.73 frames.], batch size: 20, lr: 7.36e-04 2022-05-27 18:42:16,433 INFO [train.py:823] (0/4) Epoch 23, batch 400, loss[loss=0.1437, simple_loss=0.2297, pruned_loss=0.02886, over 7294.00 frames.], tot_loss[loss=0.1756, simple_loss=0.263, pruned_loss=0.04405, over 1234502.28 frames.], batch size: 17, lr: 7.36e-04 2022-05-27 18:42:55,384 INFO [train.py:823] (0/4) Epoch 23, batch 450, loss[loss=0.1729, simple_loss=0.256, pruned_loss=0.04491, over 4806.00 frames.], tot_loss[loss=0.1749, simple_loss=0.2624, pruned_loss=0.04374, over 1273934.66 frames.], batch size: 47, lr: 7.35e-04 2022-05-27 18:43:34,453 INFO [train.py:823] (0/4) Epoch 23, batch 500, loss[loss=0.1791, simple_loss=0.2766, pruned_loss=0.04078, over 6532.00 frames.], tot_loss[loss=0.1754, simple_loss=0.263, pruned_loss=0.0439, over 1302600.23 frames.], batch size: 34, lr: 7.34e-04 2022-05-27 18:44:13,788 INFO [train.py:823] (0/4) Epoch 23, batch 550, loss[loss=0.1649, simple_loss=0.262, pruned_loss=0.03387, over 7237.00 frames.], tot_loss[loss=0.1759, simple_loss=0.2642, pruned_loss=0.04387, over 1333231.73 frames.], batch size: 24, lr: 7.33e-04 2022-05-27 18:44:52,835 INFO [train.py:823] (0/4) Epoch 23, batch 600, loss[loss=0.1987, simple_loss=0.2789, pruned_loss=0.05923, over 4998.00 frames.], tot_loss[loss=0.1765, simple_loss=0.2642, pruned_loss=0.04438, over 1348882.31 frames.], batch size: 46, lr: 7.32e-04 2022-05-27 18:45:32,269 INFO [train.py:823] (0/4) Epoch 23, batch 650, loss[loss=0.1443, simple_loss=0.2392, pruned_loss=0.02465, over 7084.00 frames.], tot_loss[loss=0.1754, simple_loss=0.2633, pruned_loss=0.04375, over 1363145.61 frames.], batch size: 19, lr: 7.32e-04 2022-05-27 18:46:11,138 INFO [train.py:823] (0/4) Epoch 23, batch 700, loss[loss=0.1744, simple_loss=0.2654, pruned_loss=0.0417, over 7000.00 frames.], tot_loss[loss=0.1758, simple_loss=0.2633, pruned_loss=0.04412, over 1369896.53 frames.], batch size: 16, lr: 7.31e-04 2022-05-27 18:46:50,489 INFO [train.py:823] (0/4) Epoch 23, batch 750, loss[loss=0.1995, simple_loss=0.2871, pruned_loss=0.05596, over 5008.00 frames.], tot_loss[loss=0.1756, simple_loss=0.2636, pruned_loss=0.0438, over 1375698.31 frames.], batch size: 46, lr: 7.30e-04 2022-05-27 18:47:30,856 INFO [train.py:823] (0/4) Epoch 23, batch 800, loss[loss=0.1892, simple_loss=0.2762, pruned_loss=0.0511, over 7185.00 frames.], tot_loss[loss=0.1754, simple_loss=0.263, pruned_loss=0.0439, over 1388002.29 frames.], batch size: 18, lr: 7.29e-04 2022-05-27 18:48:10,090 INFO [train.py:823] (0/4) Epoch 23, batch 850, loss[loss=0.1726, simple_loss=0.269, pruned_loss=0.03809, over 7148.00 frames.], tot_loss[loss=0.1758, simple_loss=0.2636, pruned_loss=0.04401, over 1395267.33 frames.], batch size: 23, lr: 7.28e-04 2022-05-27 18:48:48,827 INFO [train.py:823] (0/4) Epoch 23, batch 900, loss[loss=0.1562, simple_loss=0.2341, pruned_loss=0.0392, over 7021.00 frames.], tot_loss[loss=0.1746, simple_loss=0.2622, pruned_loss=0.04347, over 1399933.30 frames.], batch size: 17, lr: 7.28e-04 2022-05-27 18:49:27,866 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-23.pt 2022-05-27 18:49:41,152 INFO [train.py:823] (0/4) Epoch 24, batch 0, loss[loss=0.1508, simple_loss=0.227, pruned_loss=0.03732, over 7298.00 frames.], tot_loss[loss=0.1508, simple_loss=0.227, pruned_loss=0.03732, over 7298.00 frames.], batch size: 18, lr: 7.12e-04 2022-05-27 18:50:19,957 INFO [train.py:823] (0/4) Epoch 24, batch 50, loss[loss=0.1513, simple_loss=0.2396, pruned_loss=0.03154, over 7160.00 frames.], tot_loss[loss=0.1738, simple_loss=0.2635, pruned_loss=0.04203, over 318944.66 frames.], batch size: 17, lr: 7.11e-04 2022-05-27 18:51:00,451 INFO [train.py:823] (0/4) Epoch 24, batch 100, loss[loss=0.1675, simple_loss=0.264, pruned_loss=0.03545, over 6529.00 frames.], tot_loss[loss=0.1746, simple_loss=0.2633, pruned_loss=0.04295, over 560042.34 frames.], batch size: 34, lr: 7.10e-04 2022-05-27 18:51:39,697 INFO [train.py:823] (0/4) Epoch 24, batch 150, loss[loss=0.1827, simple_loss=0.2737, pruned_loss=0.04587, over 6935.00 frames.], tot_loss[loss=0.1746, simple_loss=0.2632, pruned_loss=0.04305, over 750756.44 frames.], batch size: 29, lr: 7.10e-04 2022-05-27 18:52:18,808 INFO [train.py:823] (0/4) Epoch 24, batch 200, loss[loss=0.1752, simple_loss=0.2628, pruned_loss=0.04379, over 7278.00 frames.], tot_loss[loss=0.1728, simple_loss=0.2613, pruned_loss=0.0422, over 899419.48 frames.], batch size: 21, lr: 7.09e-04 2022-05-27 18:52:58,053 INFO [train.py:823] (0/4) Epoch 24, batch 250, loss[loss=0.1552, simple_loss=0.236, pruned_loss=0.03719, over 7301.00 frames.], tot_loss[loss=0.1734, simple_loss=0.2618, pruned_loss=0.04253, over 1014984.18 frames.], batch size: 17, lr: 7.08e-04 2022-05-27 18:53:37,188 INFO [train.py:823] (0/4) Epoch 24, batch 300, loss[loss=0.1627, simple_loss=0.2568, pruned_loss=0.03424, over 7345.00 frames.], tot_loss[loss=0.1746, simple_loss=0.2622, pruned_loss=0.04348, over 1099976.33 frames.], batch size: 23, lr: 7.07e-04 2022-05-27 18:54:15,992 INFO [train.py:823] (0/4) Epoch 24, batch 350, loss[loss=0.144, simple_loss=0.2296, pruned_loss=0.02918, over 7284.00 frames.], tot_loss[loss=0.1743, simple_loss=0.262, pruned_loss=0.04325, over 1175030.11 frames.], batch size: 17, lr: 7.07e-04 2022-05-27 18:54:55,393 INFO [train.py:823] (0/4) Epoch 24, batch 400, loss[loss=0.186, simple_loss=0.2853, pruned_loss=0.04339, over 7342.00 frames.], tot_loss[loss=0.1735, simple_loss=0.2611, pruned_loss=0.04291, over 1227640.77 frames.], batch size: 23, lr: 7.06e-04 2022-05-27 18:55:34,351 INFO [train.py:823] (0/4) Epoch 24, batch 450, loss[loss=0.178, simple_loss=0.2611, pruned_loss=0.04742, over 7175.00 frames.], tot_loss[loss=0.1733, simple_loss=0.2608, pruned_loss=0.0429, over 1269126.81 frames.], batch size: 18, lr: 7.05e-04 2022-05-27 18:56:13,749 INFO [train.py:823] (0/4) Epoch 24, batch 500, loss[loss=0.162, simple_loss=0.2494, pruned_loss=0.03732, over 7278.00 frames.], tot_loss[loss=0.1734, simple_loss=0.2613, pruned_loss=0.04277, over 1304389.75 frames.], batch size: 21, lr: 7.04e-04 2022-05-27 18:56:52,900 INFO [train.py:823] (0/4) Epoch 24, batch 550, loss[loss=0.1631, simple_loss=0.269, pruned_loss=0.0286, over 6526.00 frames.], tot_loss[loss=0.1732, simple_loss=0.2613, pruned_loss=0.04258, over 1327589.75 frames.], batch size: 34, lr: 7.04e-04 2022-05-27 18:57:32,018 INFO [train.py:823] (0/4) Epoch 24, batch 600, loss[loss=0.1846, simple_loss=0.267, pruned_loss=0.0511, over 7145.00 frames.], tot_loss[loss=0.1736, simple_loss=0.2616, pruned_loss=0.04282, over 1345283.08 frames.], batch size: 23, lr: 7.03e-04 2022-05-27 18:58:10,817 INFO [train.py:823] (0/4) Epoch 24, batch 650, loss[loss=0.1784, simple_loss=0.2616, pruned_loss=0.04764, over 7106.00 frames.], tot_loss[loss=0.1741, simple_loss=0.262, pruned_loss=0.0431, over 1357212.65 frames.], batch size: 19, lr: 7.02e-04 2022-05-27 18:58:49,813 INFO [train.py:823] (0/4) Epoch 24, batch 700, loss[loss=0.1491, simple_loss=0.2422, pruned_loss=0.02796, over 7179.00 frames.], tot_loss[loss=0.1746, simple_loss=0.2628, pruned_loss=0.04316, over 1371079.36 frames.], batch size: 22, lr: 7.01e-04 2022-05-27 18:59:29,043 INFO [train.py:823] (0/4) Epoch 24, batch 750, loss[loss=0.1582, simple_loss=0.2448, pruned_loss=0.03581, over 7098.00 frames.], tot_loss[loss=0.1744, simple_loss=0.2628, pruned_loss=0.04298, over 1384734.85 frames.], batch size: 20, lr: 7.01e-04 2022-05-27 19:00:08,809 INFO [train.py:823] (0/4) Epoch 24, batch 800, loss[loss=0.1427, simple_loss=0.2212, pruned_loss=0.03205, over 6869.00 frames.], tot_loss[loss=0.1738, simple_loss=0.2619, pruned_loss=0.04283, over 1392357.04 frames.], batch size: 15, lr: 7.00e-04 2022-05-27 19:00:47,534 INFO [train.py:823] (0/4) Epoch 24, batch 850, loss[loss=0.1733, simple_loss=0.2678, pruned_loss=0.03937, over 7112.00 frames.], tot_loss[loss=0.1736, simple_loss=0.2616, pruned_loss=0.04276, over 1395103.40 frames.], batch size: 20, lr: 6.99e-04 2022-05-27 19:01:28,129 INFO [train.py:823] (0/4) Epoch 24, batch 900, loss[loss=0.182, simple_loss=0.2676, pruned_loss=0.04817, over 6492.00 frames.], tot_loss[loss=0.1736, simple_loss=0.2618, pruned_loss=0.04274, over 1398601.38 frames.], batch size: 35, lr: 6.98e-04 2022-05-27 19:02:07,087 INFO [train.py:823] (0/4) Epoch 24, batch 950, loss[loss=0.1795, simple_loss=0.2564, pruned_loss=0.05131, over 7093.00 frames.], tot_loss[loss=0.1746, simple_loss=0.2625, pruned_loss=0.0433, over 1393116.50 frames.], batch size: 18, lr: 6.98e-04 2022-05-27 19:02:08,193 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-24.pt 2022-05-27 19:02:19,950 INFO [train.py:823] (0/4) Epoch 25, batch 0, loss[loss=0.1814, simple_loss=0.2744, pruned_loss=0.0442, over 7276.00 frames.], tot_loss[loss=0.1814, simple_loss=0.2744, pruned_loss=0.0442, over 7276.00 frames.], batch size: 21, lr: 6.84e-04 2022-05-27 19:02:58,868 INFO [train.py:823] (0/4) Epoch 25, batch 50, loss[loss=0.1581, simple_loss=0.2375, pruned_loss=0.03937, over 7312.00 frames.], tot_loss[loss=0.1738, simple_loss=0.2625, pruned_loss=0.04259, over 324407.99 frames.], batch size: 17, lr: 6.83e-04 2022-05-27 19:03:38,200 INFO [train.py:823] (0/4) Epoch 25, batch 100, loss[loss=0.1707, simple_loss=0.2515, pruned_loss=0.045, over 6853.00 frames.], tot_loss[loss=0.1722, simple_loss=0.2607, pruned_loss=0.04186, over 564700.38 frames.], batch size: 15, lr: 6.82e-04 2022-05-27 19:04:16,997 INFO [train.py:823] (0/4) Epoch 25, batch 150, loss[loss=0.18, simple_loss=0.2843, pruned_loss=0.03785, over 7303.00 frames.], tot_loss[loss=0.1723, simple_loss=0.2601, pruned_loss=0.04224, over 758738.38 frames.], batch size: 22, lr: 6.82e-04 2022-05-27 19:04:56,720 INFO [train.py:823] (0/4) Epoch 25, batch 200, loss[loss=0.1825, simple_loss=0.2751, pruned_loss=0.04496, over 7285.00 frames.], tot_loss[loss=0.1727, simple_loss=0.2605, pruned_loss=0.04241, over 911267.01 frames.], batch size: 21, lr: 6.81e-04 2022-05-27 19:05:35,258 INFO [train.py:823] (0/4) Epoch 25, batch 250, loss[loss=0.1669, simple_loss=0.245, pruned_loss=0.04441, over 7295.00 frames.], tot_loss[loss=0.1725, simple_loss=0.2603, pruned_loss=0.04233, over 1022579.21 frames.], batch size: 17, lr: 6.80e-04 2022-05-27 19:06:14,520 INFO [train.py:823] (0/4) Epoch 25, batch 300, loss[loss=0.1756, simple_loss=0.2637, pruned_loss=0.0438, over 7282.00 frames.], tot_loss[loss=0.173, simple_loss=0.2609, pruned_loss=0.04249, over 1116799.39 frames.], batch size: 21, lr: 6.80e-04 2022-05-27 19:06:53,437 INFO [train.py:823] (0/4) Epoch 25, batch 350, loss[loss=0.217, simple_loss=0.3025, pruned_loss=0.0658, over 7125.00 frames.], tot_loss[loss=0.173, simple_loss=0.2609, pruned_loss=0.04251, over 1182651.37 frames.], batch size: 23, lr: 6.79e-04 2022-05-27 19:07:32,936 INFO [train.py:823] (0/4) Epoch 25, batch 400, loss[loss=0.1826, simple_loss=0.2697, pruned_loss=0.04772, over 7172.00 frames.], tot_loss[loss=0.1741, simple_loss=0.2628, pruned_loss=0.04267, over 1238642.61 frames.], batch size: 25, lr: 6.78e-04 2022-05-27 19:08:11,835 INFO [train.py:823] (0/4) Epoch 25, batch 450, loss[loss=0.122, simple_loss=0.2049, pruned_loss=0.0195, over 7215.00 frames.], tot_loss[loss=0.1746, simple_loss=0.2634, pruned_loss=0.04289, over 1269974.31 frames.], batch size: 16, lr: 6.77e-04 2022-05-27 19:08:50,900 INFO [train.py:823] (0/4) Epoch 25, batch 500, loss[loss=0.154, simple_loss=0.2344, pruned_loss=0.03678, over 7007.00 frames.], tot_loss[loss=0.1744, simple_loss=0.263, pruned_loss=0.04294, over 1302845.67 frames.], batch size: 16, lr: 6.77e-04 2022-05-27 19:09:29,777 INFO [train.py:823] (0/4) Epoch 25, batch 550, loss[loss=0.1725, simple_loss=0.266, pruned_loss=0.03954, over 7181.00 frames.], tot_loss[loss=0.1728, simple_loss=0.2612, pruned_loss=0.04216, over 1330646.00 frames.], batch size: 21, lr: 6.76e-04 2022-05-27 19:10:09,339 INFO [train.py:823] (0/4) Epoch 25, batch 600, loss[loss=0.164, simple_loss=0.2592, pruned_loss=0.03439, over 7288.00 frames.], tot_loss[loss=0.172, simple_loss=0.2605, pruned_loss=0.04178, over 1343376.85 frames.], batch size: 21, lr: 6.75e-04 2022-05-27 19:10:49,696 INFO [train.py:823] (0/4) Epoch 25, batch 650, loss[loss=0.1651, simple_loss=0.2592, pruned_loss=0.03554, over 7271.00 frames.], tot_loss[loss=0.1727, simple_loss=0.2614, pruned_loss=0.042, over 1357928.56 frames.], batch size: 20, lr: 6.75e-04 2022-05-27 19:11:29,303 INFO [train.py:823] (0/4) Epoch 25, batch 700, loss[loss=0.1404, simple_loss=0.2246, pruned_loss=0.02811, over 7144.00 frames.], tot_loss[loss=0.1721, simple_loss=0.2609, pruned_loss=0.04161, over 1369519.92 frames.], batch size: 17, lr: 6.74e-04 2022-05-27 19:12:08,298 INFO [train.py:823] (0/4) Epoch 25, batch 750, loss[loss=0.1685, simple_loss=0.2576, pruned_loss=0.03967, over 7373.00 frames.], tot_loss[loss=0.171, simple_loss=0.2595, pruned_loss=0.04126, over 1377975.32 frames.], batch size: 20, lr: 6.73e-04 2022-05-27 19:12:47,845 INFO [train.py:823] (0/4) Epoch 25, batch 800, loss[loss=0.1794, simple_loss=0.2687, pruned_loss=0.04504, over 7187.00 frames.], tot_loss[loss=0.1713, simple_loss=0.2596, pruned_loss=0.04147, over 1389944.32 frames.], batch size: 21, lr: 6.73e-04 2022-05-27 19:13:27,820 INFO [train.py:823] (0/4) Epoch 25, batch 850, loss[loss=0.168, simple_loss=0.261, pruned_loss=0.03746, over 7203.00 frames.], tot_loss[loss=0.1717, simple_loss=0.2603, pruned_loss=0.04153, over 1396236.11 frames.], batch size: 18, lr: 6.72e-04 2022-05-27 19:14:09,104 INFO [train.py:823] (0/4) Epoch 25, batch 900, loss[loss=0.1699, simple_loss=0.2675, pruned_loss=0.03621, over 6336.00 frames.], tot_loss[loss=0.1716, simple_loss=0.2601, pruned_loss=0.04153, over 1395536.40 frames.], batch size: 34, lr: 6.71e-04 2022-05-27 19:14:48,170 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-25.pt 2022-05-27 19:14:59,603 INFO [train.py:823] (0/4) Epoch 26, batch 0, loss[loss=0.1449, simple_loss=0.2177, pruned_loss=0.03602, over 7317.00 frames.], tot_loss[loss=0.1449, simple_loss=0.2177, pruned_loss=0.03602, over 7317.00 frames.], batch size: 18, lr: 6.58e-04 2022-05-27 19:15:39,050 INFO [train.py:823] (0/4) Epoch 26, batch 50, loss[loss=0.1608, simple_loss=0.2491, pruned_loss=0.03625, over 7382.00 frames.], tot_loss[loss=0.1681, simple_loss=0.2556, pruned_loss=0.04025, over 323424.57 frames.], batch size: 20, lr: 6.57e-04 2022-05-27 19:16:17,946 INFO [train.py:823] (0/4) Epoch 26, batch 100, loss[loss=0.1968, simple_loss=0.2772, pruned_loss=0.05819, over 7250.00 frames.], tot_loss[loss=0.1705, simple_loss=0.2583, pruned_loss=0.04141, over 567408.38 frames.], batch size: 25, lr: 6.56e-04 2022-05-27 19:16:57,369 INFO [train.py:823] (0/4) Epoch 26, batch 150, loss[loss=0.1841, simple_loss=0.2669, pruned_loss=0.05069, over 7192.00 frames.], tot_loss[loss=0.1705, simple_loss=0.2572, pruned_loss=0.04186, over 754237.90 frames.], batch size: 25, lr: 6.56e-04 2022-05-27 19:17:36,085 INFO [train.py:823] (0/4) Epoch 26, batch 200, loss[loss=0.1427, simple_loss=0.2293, pruned_loss=0.028, over 7103.00 frames.], tot_loss[loss=0.1705, simple_loss=0.2577, pruned_loss=0.04167, over 899819.24 frames.], batch size: 18, lr: 6.55e-04 2022-05-27 19:18:15,796 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/checkpoint-24000.pt 2022-05-27 19:18:20,381 INFO [train.py:823] (0/4) Epoch 26, batch 250, loss[loss=0.1507, simple_loss=0.2575, pruned_loss=0.02195, over 7444.00 frames.], tot_loss[loss=0.1695, simple_loss=0.2575, pruned_loss=0.0408, over 1015963.08 frames.], batch size: 22, lr: 6.55e-04 2022-05-27 19:18:59,358 INFO [train.py:823] (0/4) Epoch 26, batch 300, loss[loss=0.1434, simple_loss=0.2427, pruned_loss=0.02205, over 7112.00 frames.], tot_loss[loss=0.1688, simple_loss=0.2576, pruned_loss=0.03998, over 1106723.00 frames.], batch size: 20, lr: 6.54e-04 2022-05-27 19:19:38,449 INFO [train.py:823] (0/4) Epoch 26, batch 350, loss[loss=0.1841, simple_loss=0.279, pruned_loss=0.04458, over 6578.00 frames.], tot_loss[loss=0.1691, simple_loss=0.2581, pruned_loss=0.04005, over 1179176.84 frames.], batch size: 34, lr: 6.53e-04 2022-05-27 19:20:17,203 INFO [train.py:823] (0/4) Epoch 26, batch 400, loss[loss=0.1812, simple_loss=0.2708, pruned_loss=0.0458, over 7161.00 frames.], tot_loss[loss=0.1692, simple_loss=0.2579, pruned_loss=0.04024, over 1235298.68 frames.], batch size: 23, lr: 6.53e-04 2022-05-27 19:20:56,261 INFO [train.py:823] (0/4) Epoch 26, batch 450, loss[loss=0.1843, simple_loss=0.2727, pruned_loss=0.04796, over 7186.00 frames.], tot_loss[loss=0.1692, simple_loss=0.2578, pruned_loss=0.04031, over 1274916.24 frames.], batch size: 21, lr: 6.52e-04 2022-05-27 19:21:34,731 INFO [train.py:823] (0/4) Epoch 26, batch 500, loss[loss=0.1802, simple_loss=0.2849, pruned_loss=0.03777, over 7053.00 frames.], tot_loss[loss=0.1705, simple_loss=0.2592, pruned_loss=0.04091, over 1305317.20 frames.], batch size: 26, lr: 6.51e-04 2022-05-27 19:22:14,126 INFO [train.py:823] (0/4) Epoch 26, batch 550, loss[loss=0.1582, simple_loss=0.2345, pruned_loss=0.04093, over 7008.00 frames.], tot_loss[loss=0.1709, simple_loss=0.2591, pruned_loss=0.04135, over 1329279.99 frames.], batch size: 16, lr: 6.51e-04 2022-05-27 19:22:52,195 INFO [train.py:823] (0/4) Epoch 26, batch 600, loss[loss=0.1614, simple_loss=0.2599, pruned_loss=0.0315, over 7308.00 frames.], tot_loss[loss=0.1713, simple_loss=0.2595, pruned_loss=0.04153, over 1348421.09 frames.], batch size: 22, lr: 6.50e-04 2022-05-27 19:23:31,317 INFO [train.py:823] (0/4) Epoch 26, batch 650, loss[loss=0.1747, simple_loss=0.262, pruned_loss=0.04373, over 7347.00 frames.], tot_loss[loss=0.1701, simple_loss=0.2579, pruned_loss=0.04111, over 1358421.28 frames.], batch size: 23, lr: 6.49e-04 2022-05-27 19:24:10,339 INFO [train.py:823] (0/4) Epoch 26, batch 700, loss[loss=0.1834, simple_loss=0.2819, pruned_loss=0.04247, over 7072.00 frames.], tot_loss[loss=0.172, simple_loss=0.2602, pruned_loss=0.04192, over 1371558.62 frames.], batch size: 26, lr: 6.49e-04 2022-05-27 19:24:49,620 INFO [train.py:823] (0/4) Epoch 26, batch 750, loss[loss=0.1914, simple_loss=0.2786, pruned_loss=0.05214, over 7289.00 frames.], tot_loss[loss=0.1725, simple_loss=0.2606, pruned_loss=0.04217, over 1375501.51 frames.], batch size: 19, lr: 6.48e-04 2022-05-27 19:25:29,459 INFO [train.py:823] (0/4) Epoch 26, batch 800, loss[loss=0.1257, simple_loss=0.2155, pruned_loss=0.01791, over 6828.00 frames.], tot_loss[loss=0.1719, simple_loss=0.26, pruned_loss=0.04191, over 1383220.07 frames.], batch size: 15, lr: 6.47e-04 2022-05-27 19:26:08,816 INFO [train.py:823] (0/4) Epoch 26, batch 850, loss[loss=0.1658, simple_loss=0.2457, pruned_loss=0.04291, over 7221.00 frames.], tot_loss[loss=0.1711, simple_loss=0.2591, pruned_loss=0.04157, over 1395019.81 frames.], batch size: 16, lr: 6.47e-04 2022-05-27 19:26:47,853 INFO [train.py:823] (0/4) Epoch 26, batch 900, loss[loss=0.1479, simple_loss=0.2326, pruned_loss=0.03166, over 7012.00 frames.], tot_loss[loss=0.1713, simple_loss=0.2597, pruned_loss=0.04144, over 1395743.60 frames.], batch size: 17, lr: 6.46e-04 2022-05-27 19:27:27,084 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-26.pt 2022-05-27 19:27:39,569 INFO [train.py:823] (0/4) Epoch 27, batch 0, loss[loss=0.1686, simple_loss=0.2519, pruned_loss=0.04269, over 7192.00 frames.], tot_loss[loss=0.1686, simple_loss=0.2519, pruned_loss=0.04269, over 7192.00 frames.], batch size: 18, lr: 6.34e-04 2022-05-27 19:28:18,588 INFO [train.py:823] (0/4) Epoch 27, batch 50, loss[loss=0.136, simple_loss=0.2246, pruned_loss=0.02366, over 7204.00 frames.], tot_loss[loss=0.1696, simple_loss=0.2574, pruned_loss=0.04084, over 321783.46 frames.], batch size: 18, lr: 6.33e-04 2022-05-27 19:28:57,916 INFO [train.py:823] (0/4) Epoch 27, batch 100, loss[loss=0.171, simple_loss=0.2605, pruned_loss=0.04074, over 7183.00 frames.], tot_loss[loss=0.1696, simple_loss=0.2588, pruned_loss=0.04017, over 564393.39 frames.], batch size: 25, lr: 6.32e-04 2022-05-27 19:29:36,490 INFO [train.py:823] (0/4) Epoch 27, batch 150, loss[loss=0.1728, simple_loss=0.2518, pruned_loss=0.04688, over 7303.00 frames.], tot_loss[loss=0.1702, simple_loss=0.2589, pruned_loss=0.04073, over 753814.56 frames.], batch size: 18, lr: 6.32e-04 2022-05-27 19:30:15,925 INFO [train.py:823] (0/4) Epoch 27, batch 200, loss[loss=0.1532, simple_loss=0.2508, pruned_loss=0.0278, over 7418.00 frames.], tot_loss[loss=0.1715, simple_loss=0.2602, pruned_loss=0.04139, over 901278.48 frames.], batch size: 22, lr: 6.31e-04 2022-05-27 19:30:54,937 INFO [train.py:823] (0/4) Epoch 27, batch 250, loss[loss=0.1552, simple_loss=0.246, pruned_loss=0.03222, over 7016.00 frames.], tot_loss[loss=0.171, simple_loss=0.2604, pruned_loss=0.04076, over 1013290.05 frames.], batch size: 17, lr: 6.31e-04 2022-05-27 19:31:34,355 INFO [train.py:823] (0/4) Epoch 27, batch 300, loss[loss=0.1593, simple_loss=0.2484, pruned_loss=0.03505, over 7372.00 frames.], tot_loss[loss=0.1698, simple_loss=0.2589, pruned_loss=0.04036, over 1107020.83 frames.], batch size: 21, lr: 6.30e-04 2022-05-27 19:32:13,806 INFO [train.py:823] (0/4) Epoch 27, batch 350, loss[loss=0.1485, simple_loss=0.2371, pruned_loss=0.02997, over 7290.00 frames.], tot_loss[loss=0.1697, simple_loss=0.2585, pruned_loss=0.04049, over 1177813.66 frames.], batch size: 19, lr: 6.29e-04 2022-05-27 19:32:52,942 INFO [train.py:823] (0/4) Epoch 27, batch 400, loss[loss=0.164, simple_loss=0.2548, pruned_loss=0.03656, over 7275.00 frames.], tot_loss[loss=0.1701, simple_loss=0.2586, pruned_loss=0.04076, over 1232280.29 frames.], batch size: 20, lr: 6.29e-04 2022-05-27 19:33:33,772 INFO [train.py:823] (0/4) Epoch 27, batch 450, loss[loss=0.1525, simple_loss=0.2467, pruned_loss=0.02912, over 4852.00 frames.], tot_loss[loss=0.1694, simple_loss=0.2575, pruned_loss=0.04063, over 1275684.13 frames.], batch size: 46, lr: 6.28e-04 2022-05-27 19:34:12,765 INFO [train.py:823] (0/4) Epoch 27, batch 500, loss[loss=0.2035, simple_loss=0.2878, pruned_loss=0.05955, over 7159.00 frames.], tot_loss[loss=0.1695, simple_loss=0.2575, pruned_loss=0.04074, over 1300678.73 frames.], batch size: 23, lr: 6.28e-04 2022-05-27 19:34:52,011 INFO [train.py:823] (0/4) Epoch 27, batch 550, loss[loss=0.1648, simple_loss=0.2559, pruned_loss=0.03679, over 7294.00 frames.], tot_loss[loss=0.17, simple_loss=0.2585, pruned_loss=0.04077, over 1329735.25 frames.], batch size: 20, lr: 6.27e-04 2022-05-27 19:35:31,026 INFO [train.py:823] (0/4) Epoch 27, batch 600, loss[loss=0.1485, simple_loss=0.231, pruned_loss=0.03298, over 7308.00 frames.], tot_loss[loss=0.1707, simple_loss=0.259, pruned_loss=0.0412, over 1356063.87 frames.], batch size: 18, lr: 6.26e-04 2022-05-27 19:36:10,982 INFO [train.py:823] (0/4) Epoch 27, batch 650, loss[loss=0.1846, simple_loss=0.2749, pruned_loss=0.04719, over 7198.00 frames.], tot_loss[loss=0.1706, simple_loss=0.2591, pruned_loss=0.04105, over 1374137.74 frames.], batch size: 19, lr: 6.26e-04 2022-05-27 19:36:51,841 INFO [train.py:823] (0/4) Epoch 27, batch 700, loss[loss=0.1907, simple_loss=0.2837, pruned_loss=0.04885, over 7375.00 frames.], tot_loss[loss=0.1702, simple_loss=0.2589, pruned_loss=0.0408, over 1384426.58 frames.], batch size: 21, lr: 6.25e-04 2022-05-27 19:37:31,036 INFO [train.py:823] (0/4) Epoch 27, batch 750, loss[loss=0.1638, simple_loss=0.254, pruned_loss=0.03682, over 7199.00 frames.], tot_loss[loss=0.1691, simple_loss=0.258, pruned_loss=0.04013, over 1391860.74 frames.], batch size: 19, lr: 6.25e-04 2022-05-27 19:38:10,210 INFO [train.py:823] (0/4) Epoch 27, batch 800, loss[loss=0.2168, simple_loss=0.2996, pruned_loss=0.06703, over 7159.00 frames.], tot_loss[loss=0.1707, simple_loss=0.2594, pruned_loss=0.04103, over 1393794.27 frames.], batch size: 23, lr: 6.24e-04 2022-05-27 19:38:49,292 INFO [train.py:823] (0/4) Epoch 27, batch 850, loss[loss=0.1622, simple_loss=0.2513, pruned_loss=0.03656, over 7113.00 frames.], tot_loss[loss=0.1701, simple_loss=0.2586, pruned_loss=0.04087, over 1397120.42 frames.], batch size: 20, lr: 6.23e-04 2022-05-27 19:39:28,902 INFO [train.py:823] (0/4) Epoch 27, batch 900, loss[loss=0.1401, simple_loss=0.2201, pruned_loss=0.03007, over 7295.00 frames.], tot_loss[loss=0.1692, simple_loss=0.2578, pruned_loss=0.04027, over 1398718.14 frames.], batch size: 17, lr: 6.23e-04 2022-05-27 19:40:08,121 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-27.pt 2022-05-27 19:40:22,693 INFO [train.py:823] (0/4) Epoch 28, batch 0, loss[loss=0.1718, simple_loss=0.2587, pruned_loss=0.04244, over 7193.00 frames.], tot_loss[loss=0.1718, simple_loss=0.2587, pruned_loss=0.04244, over 7193.00 frames.], batch size: 20, lr: 6.11e-04 2022-05-27 19:41:02,548 INFO [train.py:823] (0/4) Epoch 28, batch 50, loss[loss=0.1677, simple_loss=0.26, pruned_loss=0.03769, over 7109.00 frames.], tot_loss[loss=0.1665, simple_loss=0.2558, pruned_loss=0.03864, over 315948.05 frames.], batch size: 20, lr: 6.11e-04 2022-05-27 19:41:41,864 INFO [train.py:823] (0/4) Epoch 28, batch 100, loss[loss=0.1701, simple_loss=0.2619, pruned_loss=0.03916, over 7031.00 frames.], tot_loss[loss=0.168, simple_loss=0.2566, pruned_loss=0.03968, over 561115.17 frames.], batch size: 26, lr: 6.10e-04 2022-05-27 19:42:21,450 INFO [train.py:823] (0/4) Epoch 28, batch 150, loss[loss=0.1661, simple_loss=0.2434, pruned_loss=0.04437, over 4702.00 frames.], tot_loss[loss=0.1685, simple_loss=0.2568, pruned_loss=0.04003, over 749097.80 frames.], batch size: 46, lr: 6.09e-04 2022-05-27 19:43:00,526 INFO [train.py:823] (0/4) Epoch 28, batch 200, loss[loss=0.1806, simple_loss=0.2535, pruned_loss=0.05386, over 7200.00 frames.], tot_loss[loss=0.1673, simple_loss=0.2559, pruned_loss=0.03935, over 900055.03 frames.], batch size: 20, lr: 6.09e-04 2022-05-27 19:43:40,065 INFO [train.py:823] (0/4) Epoch 28, batch 250, loss[loss=0.1984, simple_loss=0.2819, pruned_loss=0.05749, over 7335.00 frames.], tot_loss[loss=0.168, simple_loss=0.2565, pruned_loss=0.03971, over 1015074.57 frames.], batch size: 23, lr: 6.08e-04 2022-05-27 19:44:19,075 INFO [train.py:823] (0/4) Epoch 28, batch 300, loss[loss=0.1762, simple_loss=0.2666, pruned_loss=0.04289, over 6951.00 frames.], tot_loss[loss=0.1689, simple_loss=0.2574, pruned_loss=0.04016, over 1102984.22 frames.], batch size: 29, lr: 6.08e-04 2022-05-27 19:44:58,832 INFO [train.py:823] (0/4) Epoch 28, batch 350, loss[loss=0.1676, simple_loss=0.2648, pruned_loss=0.03521, over 7347.00 frames.], tot_loss[loss=0.168, simple_loss=0.2567, pruned_loss=0.03971, over 1173709.74 frames.], batch size: 23, lr: 6.07e-04 2022-05-27 19:45:37,851 INFO [train.py:823] (0/4) Epoch 28, batch 400, loss[loss=0.2092, simple_loss=0.2926, pruned_loss=0.06293, over 7291.00 frames.], tot_loss[loss=0.1686, simple_loss=0.2568, pruned_loss=0.04016, over 1228175.41 frames.], batch size: 21, lr: 6.07e-04 2022-05-27 19:46:17,180 INFO [train.py:823] (0/4) Epoch 28, batch 450, loss[loss=0.1735, simple_loss=0.2592, pruned_loss=0.04385, over 6877.00 frames.], tot_loss[loss=0.1683, simple_loss=0.2565, pruned_loss=0.03999, over 1268629.79 frames.], batch size: 29, lr: 6.06e-04 2022-05-27 19:46:56,126 INFO [train.py:823] (0/4) Epoch 28, batch 500, loss[loss=0.1833, simple_loss=0.2793, pruned_loss=0.04362, over 6908.00 frames.], tot_loss[loss=0.1676, simple_loss=0.2562, pruned_loss=0.03949, over 1305951.96 frames.], batch size: 29, lr: 6.06e-04 2022-05-27 19:47:35,632 INFO [train.py:823] (0/4) Epoch 28, batch 550, loss[loss=0.1548, simple_loss=0.2619, pruned_loss=0.02388, over 7111.00 frames.], tot_loss[loss=0.1671, simple_loss=0.2562, pruned_loss=0.03899, over 1330227.64 frames.], batch size: 20, lr: 6.05e-04 2022-05-27 19:48:14,479 INFO [train.py:823] (0/4) Epoch 28, batch 600, loss[loss=0.1599, simple_loss=0.2544, pruned_loss=0.03267, over 7199.00 frames.], tot_loss[loss=0.1672, simple_loss=0.2565, pruned_loss=0.03892, over 1348680.01 frames.], batch size: 19, lr: 6.04e-04 2022-05-27 19:48:53,932 INFO [train.py:823] (0/4) Epoch 28, batch 650, loss[loss=0.1589, simple_loss=0.2456, pruned_loss=0.03607, over 7291.00 frames.], tot_loss[loss=0.1679, simple_loss=0.2567, pruned_loss=0.03952, over 1367169.26 frames.], batch size: 19, lr: 6.04e-04 2022-05-27 19:49:34,149 INFO [train.py:823] (0/4) Epoch 28, batch 700, loss[loss=0.1768, simple_loss=0.2567, pruned_loss=0.04841, over 7306.00 frames.], tot_loss[loss=0.168, simple_loss=0.2568, pruned_loss=0.03956, over 1376999.72 frames.], batch size: 18, lr: 6.03e-04 2022-05-27 19:50:13,562 INFO [train.py:823] (0/4) Epoch 28, batch 750, loss[loss=0.1863, simple_loss=0.28, pruned_loss=0.0463, over 4891.00 frames.], tot_loss[loss=0.1683, simple_loss=0.2571, pruned_loss=0.03976, over 1383023.62 frames.], batch size: 47, lr: 6.03e-04 2022-05-27 19:50:52,551 INFO [train.py:823] (0/4) Epoch 28, batch 800, loss[loss=0.1365, simple_loss=0.2129, pruned_loss=0.03003, over 6995.00 frames.], tot_loss[loss=0.1685, simple_loss=0.2574, pruned_loss=0.03986, over 1395526.74 frames.], batch size: 16, lr: 6.02e-04 2022-05-27 19:51:31,709 INFO [train.py:823] (0/4) Epoch 28, batch 850, loss[loss=0.1566, simple_loss=0.2565, pruned_loss=0.0284, over 7380.00 frames.], tot_loss[loss=0.1683, simple_loss=0.2575, pruned_loss=0.03952, over 1399713.80 frames.], batch size: 21, lr: 6.02e-04 2022-05-27 19:52:10,736 INFO [train.py:823] (0/4) Epoch 28, batch 900, loss[loss=0.1654, simple_loss=0.2592, pruned_loss=0.03577, over 7375.00 frames.], tot_loss[loss=0.1682, simple_loss=0.2575, pruned_loss=0.03947, over 1402062.24 frames.], batch size: 21, lr: 6.01e-04 2022-05-27 19:52:49,587 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-28.pt 2022-05-27 19:53:03,346 INFO [train.py:823] (0/4) Epoch 29, batch 0, loss[loss=0.1908, simple_loss=0.2811, pruned_loss=0.05025, over 7040.00 frames.], tot_loss[loss=0.1908, simple_loss=0.2811, pruned_loss=0.05025, over 7040.00 frames.], batch size: 26, lr: 5.90e-04 2022-05-27 19:53:42,702 INFO [train.py:823] (0/4) Epoch 29, batch 50, loss[loss=0.1618, simple_loss=0.2594, pruned_loss=0.03216, over 7294.00 frames.], tot_loss[loss=0.1685, simple_loss=0.2563, pruned_loss=0.04035, over 321469.51 frames.], batch size: 21, lr: 5.90e-04 2022-05-27 19:54:22,196 INFO [train.py:823] (0/4) Epoch 29, batch 100, loss[loss=0.1917, simple_loss=0.2737, pruned_loss=0.05483, over 7248.00 frames.], tot_loss[loss=0.1698, simple_loss=0.2578, pruned_loss=0.0409, over 570051.05 frames.], batch size: 24, lr: 5.89e-04 2022-05-27 19:55:01,923 INFO [train.py:823] (0/4) Epoch 29, batch 150, loss[loss=0.1864, simple_loss=0.2649, pruned_loss=0.05393, over 7292.00 frames.], tot_loss[loss=0.1682, simple_loss=0.2552, pruned_loss=0.04058, over 760073.50 frames.], batch size: 19, lr: 5.89e-04 2022-05-27 19:55:40,949 INFO [train.py:823] (0/4) Epoch 29, batch 200, loss[loss=0.1841, simple_loss=0.2683, pruned_loss=0.04995, over 7333.00 frames.], tot_loss[loss=0.1688, simple_loss=0.257, pruned_loss=0.04031, over 900566.63 frames.], batch size: 23, lr: 5.88e-04 2022-05-27 19:56:21,509 INFO [train.py:823] (0/4) Epoch 29, batch 250, loss[loss=0.1594, simple_loss=0.2464, pruned_loss=0.0362, over 7385.00 frames.], tot_loss[loss=0.1673, simple_loss=0.2555, pruned_loss=0.03951, over 1017449.36 frames.], batch size: 19, lr: 5.88e-04 2022-05-27 19:57:00,563 INFO [train.py:823] (0/4) Epoch 29, batch 300, loss[loss=0.1666, simple_loss=0.252, pruned_loss=0.04066, over 7290.00 frames.], tot_loss[loss=0.1683, simple_loss=0.2561, pruned_loss=0.04023, over 1106994.08 frames.], batch size: 20, lr: 5.87e-04 2022-05-27 19:57:40,047 INFO [train.py:823] (0/4) Epoch 29, batch 350, loss[loss=0.1529, simple_loss=0.2335, pruned_loss=0.0361, over 6814.00 frames.], tot_loss[loss=0.1684, simple_loss=0.2568, pruned_loss=0.04, over 1176117.79 frames.], batch size: 15, lr: 5.87e-04 2022-05-27 19:58:19,097 INFO [train.py:823] (0/4) Epoch 29, batch 400, loss[loss=0.1381, simple_loss=0.2202, pruned_loss=0.02802, over 7297.00 frames.], tot_loss[loss=0.1689, simple_loss=0.2578, pruned_loss=0.03996, over 1231069.44 frames.], batch size: 17, lr: 5.86e-04 2022-05-27 19:58:59,806 INFO [train.py:823] (0/4) Epoch 29, batch 450, loss[loss=0.1502, simple_loss=0.2373, pruned_loss=0.03158, over 7101.00 frames.], tot_loss[loss=0.169, simple_loss=0.2577, pruned_loss=0.04015, over 1270954.49 frames.], batch size: 18, lr: 5.85e-04 2022-05-27 19:59:40,115 INFO [train.py:823] (0/4) Epoch 29, batch 500, loss[loss=0.1655, simple_loss=0.2571, pruned_loss=0.03695, over 7103.00 frames.], tot_loss[loss=0.1678, simple_loss=0.2566, pruned_loss=0.0395, over 1298016.06 frames.], batch size: 20, lr: 5.85e-04 2022-05-27 20:00:19,254 INFO [train.py:823] (0/4) Epoch 29, batch 550, loss[loss=0.1695, simple_loss=0.268, pruned_loss=0.03548, over 6585.00 frames.], tot_loss[loss=0.1675, simple_loss=0.2567, pruned_loss=0.03916, over 1327741.47 frames.], batch size: 34, lr: 5.84e-04 2022-05-27 20:00:58,208 INFO [train.py:823] (0/4) Epoch 29, batch 600, loss[loss=0.1912, simple_loss=0.2864, pruned_loss=0.048, over 6528.00 frames.], tot_loss[loss=0.1688, simple_loss=0.258, pruned_loss=0.03975, over 1347958.75 frames.], batch size: 34, lr: 5.84e-04 2022-05-27 20:01:37,808 INFO [train.py:823] (0/4) Epoch 29, batch 650, loss[loss=0.1999, simple_loss=0.29, pruned_loss=0.0549, over 7367.00 frames.], tot_loss[loss=0.1691, simple_loss=0.2582, pruned_loss=0.04005, over 1364917.35 frames.], batch size: 20, lr: 5.83e-04 2022-05-27 20:02:16,316 INFO [train.py:823] (0/4) Epoch 29, batch 700, loss[loss=0.1777, simple_loss=0.2624, pruned_loss=0.04648, over 7196.00 frames.], tot_loss[loss=0.1678, simple_loss=0.2569, pruned_loss=0.03935, over 1371729.99 frames.], batch size: 19, lr: 5.83e-04 2022-05-27 20:02:55,464 INFO [train.py:823] (0/4) Epoch 29, batch 750, loss[loss=0.1481, simple_loss=0.2445, pruned_loss=0.0259, over 4947.00 frames.], tot_loss[loss=0.1675, simple_loss=0.2569, pruned_loss=0.03899, over 1379485.32 frames.], batch size: 47, lr: 5.82e-04 2022-05-27 20:03:34,140 INFO [train.py:823] (0/4) Epoch 29, batch 800, loss[loss=0.1424, simple_loss=0.2314, pruned_loss=0.02673, over 7205.00 frames.], tot_loss[loss=0.1673, simple_loss=0.257, pruned_loss=0.03887, over 1387828.67 frames.], batch size: 18, lr: 5.82e-04 2022-05-27 20:04:13,231 INFO [train.py:823] (0/4) Epoch 29, batch 850, loss[loss=0.1858, simple_loss=0.2777, pruned_loss=0.0469, over 7230.00 frames.], tot_loss[loss=0.1679, simple_loss=0.2577, pruned_loss=0.039, over 1398123.25 frames.], batch size: 24, lr: 5.81e-04 2022-05-27 20:04:52,052 INFO [train.py:823] (0/4) Epoch 29, batch 900, loss[loss=0.2055, simple_loss=0.3, pruned_loss=0.05555, over 7174.00 frames.], tot_loss[loss=0.169, simple_loss=0.2586, pruned_loss=0.03973, over 1396261.12 frames.], batch size: 22, lr: 5.81e-04 2022-05-27 20:05:30,821 INFO [train.py:823] (0/4) Epoch 29, batch 950, loss[loss=0.2067, simple_loss=0.2813, pruned_loss=0.06611, over 4651.00 frames.], tot_loss[loss=0.1688, simple_loss=0.258, pruned_loss=0.0398, over 1389420.44 frames.], batch size: 46, lr: 5.80e-04 2022-05-27 20:05:32,007 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-29.pt 2022-05-27 20:05:46,337 INFO [train.py:823] (0/4) Epoch 30, batch 0, loss[loss=0.1857, simple_loss=0.2733, pruned_loss=0.04899, over 7372.00 frames.], tot_loss[loss=0.1857, simple_loss=0.2733, pruned_loss=0.04899, over 7372.00 frames.], batch size: 20, lr: 5.71e-04 2022-05-27 20:06:25,778 INFO [train.py:823] (0/4) Epoch 30, batch 50, loss[loss=0.1847, simple_loss=0.2634, pruned_loss=0.053, over 7106.00 frames.], tot_loss[loss=0.1672, simple_loss=0.2557, pruned_loss=0.03933, over 314588.25 frames.], batch size: 19, lr: 5.70e-04 2022-05-27 20:07:04,843 INFO [train.py:823] (0/4) Epoch 30, batch 100, loss[loss=0.144, simple_loss=0.2264, pruned_loss=0.03077, over 7310.00 frames.], tot_loss[loss=0.1644, simple_loss=0.2537, pruned_loss=0.03761, over 561815.73 frames.], batch size: 17, lr: 5.70e-04 2022-05-27 20:07:43,985 INFO [train.py:823] (0/4) Epoch 30, batch 150, loss[loss=0.1665, simple_loss=0.2651, pruned_loss=0.03391, over 7155.00 frames.], tot_loss[loss=0.167, simple_loss=0.2561, pruned_loss=0.03898, over 754615.73 frames.], batch size: 23, lr: 5.69e-04 2022-05-27 20:08:23,066 INFO [train.py:823] (0/4) Epoch 30, batch 200, loss[loss=0.2197, simple_loss=0.3006, pruned_loss=0.06945, over 7137.00 frames.], tot_loss[loss=0.1671, simple_loss=0.2564, pruned_loss=0.0389, over 901545.41 frames.], batch size: 23, lr: 5.69e-04 2022-05-27 20:09:02,538 INFO [train.py:823] (0/4) Epoch 30, batch 250, loss[loss=0.144, simple_loss=0.2361, pruned_loss=0.02599, over 7101.00 frames.], tot_loss[loss=0.1675, simple_loss=0.2571, pruned_loss=0.03898, over 1013704.23 frames.], batch size: 19, lr: 5.68e-04 2022-05-27 20:09:41,419 INFO [train.py:823] (0/4) Epoch 30, batch 300, loss[loss=0.1423, simple_loss=0.2309, pruned_loss=0.02686, over 7157.00 frames.], tot_loss[loss=0.1691, simple_loss=0.2584, pruned_loss=0.03987, over 1107099.45 frames.], batch size: 17, lr: 5.68e-04 2022-05-27 20:10:20,676 INFO [train.py:823] (0/4) Epoch 30, batch 350, loss[loss=0.1834, simple_loss=0.2718, pruned_loss=0.04751, over 7245.00 frames.], tot_loss[loss=0.1678, simple_loss=0.2568, pruned_loss=0.03935, over 1177615.16 frames.], batch size: 24, lr: 5.67e-04 2022-05-27 20:10:59,229 INFO [train.py:823] (0/4) Epoch 30, batch 400, loss[loss=0.1708, simple_loss=0.2657, pruned_loss=0.038, over 7022.00 frames.], tot_loss[loss=0.1674, simple_loss=0.2567, pruned_loss=0.03905, over 1231683.91 frames.], batch size: 26, lr: 5.67e-04 2022-05-27 20:11:38,657 INFO [train.py:823] (0/4) Epoch 30, batch 450, loss[loss=0.2267, simple_loss=0.2983, pruned_loss=0.0775, over 6904.00 frames.], tot_loss[loss=0.1676, simple_loss=0.2566, pruned_loss=0.03929, over 1270821.01 frames.], batch size: 29, lr: 5.66e-04 2022-05-27 20:12:17,441 INFO [train.py:823] (0/4) Epoch 30, batch 500, loss[loss=0.1467, simple_loss=0.2447, pruned_loss=0.0243, over 7100.00 frames.], tot_loss[loss=0.1678, simple_loss=0.2567, pruned_loss=0.03945, over 1303192.22 frames.], batch size: 19, lr: 5.66e-04 2022-05-27 20:12:57,039 INFO [train.py:823] (0/4) Epoch 30, batch 550, loss[loss=0.1571, simple_loss=0.2534, pruned_loss=0.03042, over 7416.00 frames.], tot_loss[loss=0.1672, simple_loss=0.2562, pruned_loss=0.03906, over 1327785.83 frames.], batch size: 22, lr: 5.65e-04 2022-05-27 20:13:37,209 INFO [train.py:823] (0/4) Epoch 30, batch 600, loss[loss=0.1946, simple_loss=0.2787, pruned_loss=0.05526, over 7186.00 frames.], tot_loss[loss=0.1676, simple_loss=0.257, pruned_loss=0.03917, over 1344922.30 frames.], batch size: 19, lr: 5.65e-04 2022-05-27 20:14:16,692 INFO [train.py:823] (0/4) Epoch 30, batch 650, loss[loss=0.1914, simple_loss=0.2826, pruned_loss=0.05016, over 7415.00 frames.], tot_loss[loss=0.1668, simple_loss=0.2559, pruned_loss=0.03892, over 1358690.52 frames.], batch size: 22, lr: 5.64e-04 2022-05-27 20:14:55,804 INFO [train.py:823] (0/4) Epoch 30, batch 700, loss[loss=0.1497, simple_loss=0.2353, pruned_loss=0.03205, over 7285.00 frames.], tot_loss[loss=0.1651, simple_loss=0.2541, pruned_loss=0.03803, over 1376904.99 frames.], batch size: 19, lr: 5.64e-04 2022-05-27 20:15:35,119 INFO [train.py:823] (0/4) Epoch 30, batch 750, loss[loss=0.1329, simple_loss=0.2266, pruned_loss=0.01958, over 7095.00 frames.], tot_loss[loss=0.1659, simple_loss=0.2551, pruned_loss=0.0384, over 1382787.94 frames.], batch size: 18, lr: 5.63e-04 2022-05-27 20:16:13,855 INFO [train.py:823] (0/4) Epoch 30, batch 800, loss[loss=0.1639, simple_loss=0.2551, pruned_loss=0.03637, over 6965.00 frames.], tot_loss[loss=0.1656, simple_loss=0.2549, pruned_loss=0.03814, over 1392624.51 frames.], batch size: 26, lr: 5.63e-04 2022-05-27 20:16:53,199 INFO [train.py:823] (0/4) Epoch 30, batch 850, loss[loss=0.1758, simple_loss=0.247, pruned_loss=0.05225, over 7190.00 frames.], tot_loss[loss=0.1659, simple_loss=0.2548, pruned_loss=0.03849, over 1391376.60 frames.], batch size: 18, lr: 5.62e-04 2022-05-27 20:17:32,089 INFO [train.py:823] (0/4) Epoch 30, batch 900, loss[loss=0.1644, simple_loss=0.2459, pruned_loss=0.0414, over 7288.00 frames.], tot_loss[loss=0.1652, simple_loss=0.2545, pruned_loss=0.038, over 1395655.64 frames.], batch size: 19, lr: 5.62e-04 2022-05-27 20:18:10,477 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-30.pt 2022-05-27 20:18:24,265 INFO [train.py:823] (0/4) Epoch 31, batch 0, loss[loss=0.1629, simple_loss=0.2517, pruned_loss=0.03698, over 7383.00 frames.], tot_loss[loss=0.1629, simple_loss=0.2517, pruned_loss=0.03698, over 7383.00 frames.], batch size: 20, lr: 5.52e-04 2022-05-27 20:19:03,838 INFO [train.py:823] (0/4) Epoch 31, batch 50, loss[loss=0.1521, simple_loss=0.2424, pruned_loss=0.03093, over 7187.00 frames.], tot_loss[loss=0.1648, simple_loss=0.253, pruned_loss=0.03832, over 324952.09 frames.], batch size: 18, lr: 5.52e-04 2022-05-27 20:19:44,320 INFO [train.py:823] (0/4) Epoch 31, batch 100, loss[loss=0.1582, simple_loss=0.2369, pruned_loss=0.03972, over 6820.00 frames.], tot_loss[loss=0.1636, simple_loss=0.2526, pruned_loss=0.03732, over 565025.96 frames.], batch size: 15, lr: 5.51e-04 2022-05-27 20:20:23,586 INFO [train.py:823] (0/4) Epoch 31, batch 150, loss[loss=0.1648, simple_loss=0.2602, pruned_loss=0.03476, over 7212.00 frames.], tot_loss[loss=0.1643, simple_loss=0.2536, pruned_loss=0.03753, over 753637.92 frames.], batch size: 25, lr: 5.51e-04 2022-05-27 20:21:02,308 INFO [train.py:823] (0/4) Epoch 31, batch 200, loss[loss=0.1548, simple_loss=0.2455, pruned_loss=0.03199, over 7091.00 frames.], tot_loss[loss=0.1649, simple_loss=0.2547, pruned_loss=0.03755, over 898792.18 frames.], batch size: 18, lr: 5.50e-04 2022-05-27 20:21:41,472 INFO [train.py:823] (0/4) Epoch 31, batch 250, loss[loss=0.1661, simple_loss=0.2456, pruned_loss=0.04332, over 7146.00 frames.], tot_loss[loss=0.1649, simple_loss=0.254, pruned_loss=0.03788, over 1006530.97 frames.], batch size: 17, lr: 5.50e-04 2022-05-27 20:22:21,749 INFO [train.py:823] (0/4) Epoch 31, batch 300, loss[loss=0.1591, simple_loss=0.2533, pruned_loss=0.03241, over 7296.00 frames.], tot_loss[loss=0.1658, simple_loss=0.2549, pruned_loss=0.03829, over 1099320.29 frames.], batch size: 22, lr: 5.49e-04 2022-05-27 20:23:00,952 INFO [train.py:823] (0/4) Epoch 31, batch 350, loss[loss=0.1393, simple_loss=0.2215, pruned_loss=0.02856, over 7152.00 frames.], tot_loss[loss=0.1657, simple_loss=0.2547, pruned_loss=0.03835, over 1165276.44 frames.], batch size: 17, lr: 5.49e-04 2022-05-27 20:23:41,373 INFO [train.py:823] (0/4) Epoch 31, batch 400, loss[loss=0.1662, simple_loss=0.2486, pruned_loss=0.04192, over 7396.00 frames.], tot_loss[loss=0.1658, simple_loss=0.255, pruned_loss=0.03828, over 1226818.06 frames.], batch size: 19, lr: 5.49e-04 2022-05-27 20:24:20,604 INFO [train.py:823] (0/4) Epoch 31, batch 450, loss[loss=0.1509, simple_loss=0.2411, pruned_loss=0.03034, over 7307.00 frames.], tot_loss[loss=0.166, simple_loss=0.2552, pruned_loss=0.03843, over 1270660.13 frames.], batch size: 18, lr: 5.48e-04 2022-05-27 20:24:59,818 INFO [train.py:823] (0/4) Epoch 31, batch 500, loss[loss=0.1472, simple_loss=0.2337, pruned_loss=0.0303, over 7095.00 frames.], tot_loss[loss=0.1655, simple_loss=0.2544, pruned_loss=0.03826, over 1302951.76 frames.], batch size: 18, lr: 5.48e-04 2022-05-27 20:25:39,397 INFO [train.py:823] (0/4) Epoch 31, batch 550, loss[loss=0.1536, simple_loss=0.2481, pruned_loss=0.02958, over 7387.00 frames.], tot_loss[loss=0.1645, simple_loss=0.2532, pruned_loss=0.03792, over 1327195.68 frames.], batch size: 19, lr: 5.47e-04 2022-05-27 20:26:18,511 INFO [train.py:823] (0/4) Epoch 31, batch 600, loss[loss=0.162, simple_loss=0.2294, pruned_loss=0.04725, over 7194.00 frames.], tot_loss[loss=0.1651, simple_loss=0.2538, pruned_loss=0.03823, over 1346922.06 frames.], batch size: 16, lr: 5.47e-04 2022-05-27 20:26:57,579 INFO [train.py:823] (0/4) Epoch 31, batch 650, loss[loss=0.2094, simple_loss=0.2928, pruned_loss=0.06298, over 7163.00 frames.], tot_loss[loss=0.165, simple_loss=0.2539, pruned_loss=0.03801, over 1362299.99 frames.], batch size: 22, lr: 5.46e-04 2022-05-27 20:27:36,410 INFO [train.py:823] (0/4) Epoch 31, batch 700, loss[loss=0.1475, simple_loss=0.2349, pruned_loss=0.03002, over 7295.00 frames.], tot_loss[loss=0.1655, simple_loss=0.2547, pruned_loss=0.03813, over 1371187.81 frames.], batch size: 17, lr: 5.46e-04 2022-05-27 20:28:15,515 INFO [train.py:823] (0/4) Epoch 31, batch 750, loss[loss=0.1431, simple_loss=0.236, pruned_loss=0.02516, over 7307.00 frames.], tot_loss[loss=0.1649, simple_loss=0.2544, pruned_loss=0.03771, over 1382807.24 frames.], batch size: 18, lr: 5.45e-04 2022-05-27 20:28:54,268 INFO [train.py:823] (0/4) Epoch 31, batch 800, loss[loss=0.1594, simple_loss=0.2338, pruned_loss=0.04251, over 6809.00 frames.], tot_loss[loss=0.1654, simple_loss=0.2547, pruned_loss=0.03806, over 1393016.38 frames.], batch size: 15, lr: 5.45e-04 2022-05-27 20:29:32,789 INFO [train.py:823] (0/4) Epoch 31, batch 850, loss[loss=0.1598, simple_loss=0.2503, pruned_loss=0.03471, over 7004.00 frames.], tot_loss[loss=0.1658, simple_loss=0.2551, pruned_loss=0.03831, over 1391702.94 frames.], batch size: 26, lr: 5.44e-04 2022-05-27 20:30:11,757 INFO [train.py:823] (0/4) Epoch 31, batch 900, loss[loss=0.1755, simple_loss=0.2553, pruned_loss=0.04782, over 7095.00 frames.], tot_loss[loss=0.166, simple_loss=0.2555, pruned_loss=0.03826, over 1397333.30 frames.], batch size: 19, lr: 5.44e-04 2022-05-27 20:30:51,863 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-31.pt 2022-05-27 20:31:03,432 INFO [train.py:823] (0/4) Epoch 32, batch 0, loss[loss=0.16, simple_loss=0.2479, pruned_loss=0.03603, over 4702.00 frames.], tot_loss[loss=0.16, simple_loss=0.2479, pruned_loss=0.03603, over 4702.00 frames.], batch size: 47, lr: 5.35e-04 2022-05-27 20:31:42,673 INFO [train.py:823] (0/4) Epoch 32, batch 50, loss[loss=0.1541, simple_loss=0.2354, pruned_loss=0.0364, over 7293.00 frames.], tot_loss[loss=0.1655, simple_loss=0.2554, pruned_loss=0.03776, over 319621.29 frames.], batch size: 17, lr: 5.35e-04 2022-05-27 20:32:21,581 INFO [train.py:823] (0/4) Epoch 32, batch 100, loss[loss=0.1785, simple_loss=0.2709, pruned_loss=0.04301, over 7162.00 frames.], tot_loss[loss=0.1666, simple_loss=0.2557, pruned_loss=0.03875, over 565057.22 frames.], batch size: 22, lr: 5.34e-04 2022-05-27 20:33:00,033 INFO [train.py:823] (0/4) Epoch 32, batch 150, loss[loss=0.1724, simple_loss=0.2583, pruned_loss=0.04322, over 7204.00 frames.], tot_loss[loss=0.1663, simple_loss=0.2549, pruned_loss=0.03885, over 758396.09 frames.], batch size: 19, lr: 5.34e-04 2022-05-27 20:33:39,218 INFO [train.py:823] (0/4) Epoch 32, batch 200, loss[loss=0.1678, simple_loss=0.2646, pruned_loss=0.03544, over 7189.00 frames.], tot_loss[loss=0.1664, simple_loss=0.256, pruned_loss=0.03837, over 904844.43 frames.], batch size: 19, lr: 5.33e-04 2022-05-27 20:34:18,160 INFO [train.py:823] (0/4) Epoch 32, batch 250, loss[loss=0.1708, simple_loss=0.2688, pruned_loss=0.03645, over 7188.00 frames.], tot_loss[loss=0.1655, simple_loss=0.255, pruned_loss=0.03803, over 1021309.88 frames.], batch size: 19, lr: 5.33e-04 2022-05-27 20:34:57,817 INFO [train.py:823] (0/4) Epoch 32, batch 300, loss[loss=0.1591, simple_loss=0.2378, pruned_loss=0.04023, over 7294.00 frames.], tot_loss[loss=0.164, simple_loss=0.2532, pruned_loss=0.03743, over 1105912.18 frames.], batch size: 19, lr: 5.32e-04 2022-05-27 20:35:36,756 INFO [train.py:823] (0/4) Epoch 32, batch 350, loss[loss=0.1658, simple_loss=0.244, pruned_loss=0.04375, over 7010.00 frames.], tot_loss[loss=0.1647, simple_loss=0.2538, pruned_loss=0.03777, over 1175931.53 frames.], batch size: 16, lr: 5.32e-04 2022-05-27 20:36:15,987 INFO [train.py:823] (0/4) Epoch 32, batch 400, loss[loss=0.1623, simple_loss=0.2514, pruned_loss=0.03657, over 6414.00 frames.], tot_loss[loss=0.1653, simple_loss=0.2551, pruned_loss=0.03782, over 1225924.56 frames.], batch size: 34, lr: 5.32e-04 2022-05-27 20:36:54,880 INFO [train.py:823] (0/4) Epoch 32, batch 450, loss[loss=0.1751, simple_loss=0.2679, pruned_loss=0.04109, over 7159.00 frames.], tot_loss[loss=0.1653, simple_loss=0.2547, pruned_loss=0.03789, over 1266801.13 frames.], batch size: 23, lr: 5.31e-04 2022-05-27 20:37:35,404 INFO [train.py:823] (0/4) Epoch 32, batch 500, loss[loss=0.1425, simple_loss=0.2349, pruned_loss=0.02502, over 7206.00 frames.], tot_loss[loss=0.1638, simple_loss=0.2532, pruned_loss=0.03724, over 1300938.48 frames.], batch size: 20, lr: 5.31e-04 2022-05-27 20:38:14,345 INFO [train.py:823] (0/4) Epoch 32, batch 550, loss[loss=0.1938, simple_loss=0.2855, pruned_loss=0.05109, over 7195.00 frames.], tot_loss[loss=0.1641, simple_loss=0.2539, pruned_loss=0.03717, over 1329452.44 frames.], batch size: 25, lr: 5.30e-04 2022-05-27 20:38:53,725 INFO [train.py:823] (0/4) Epoch 32, batch 600, loss[loss=0.1418, simple_loss=0.2203, pruned_loss=0.03171, over 7299.00 frames.], tot_loss[loss=0.1647, simple_loss=0.2544, pruned_loss=0.03745, over 1350792.51 frames.], batch size: 17, lr: 5.30e-04 2022-05-27 20:39:32,598 INFO [train.py:823] (0/4) Epoch 32, batch 650, loss[loss=0.155, simple_loss=0.2493, pruned_loss=0.0304, over 7002.00 frames.], tot_loss[loss=0.1644, simple_loss=0.2544, pruned_loss=0.03723, over 1362551.42 frames.], batch size: 26, lr: 5.29e-04 2022-05-27 20:40:11,817 INFO [train.py:823] (0/4) Epoch 32, batch 700, loss[loss=0.18, simple_loss=0.2613, pruned_loss=0.04934, over 7099.00 frames.], tot_loss[loss=0.1639, simple_loss=0.2534, pruned_loss=0.03722, over 1378483.11 frames.], batch size: 20, lr: 5.29e-04 2022-05-27 20:40:50,474 INFO [train.py:823] (0/4) Epoch 32, batch 750, loss[loss=0.1522, simple_loss=0.2368, pruned_loss=0.03385, over 7390.00 frames.], tot_loss[loss=0.1633, simple_loss=0.2527, pruned_loss=0.03699, over 1389326.82 frames.], batch size: 19, lr: 5.29e-04 2022-05-27 20:41:30,222 INFO [train.py:823] (0/4) Epoch 32, batch 800, loss[loss=0.161, simple_loss=0.2506, pruned_loss=0.03572, over 7147.00 frames.], tot_loss[loss=0.1632, simple_loss=0.2524, pruned_loss=0.03697, over 1397322.69 frames.], batch size: 17, lr: 5.28e-04 2022-05-27 20:42:10,587 INFO [train.py:823] (0/4) Epoch 32, batch 850, loss[loss=0.1244, simple_loss=0.2144, pruned_loss=0.01721, over 7020.00 frames.], tot_loss[loss=0.1635, simple_loss=0.2526, pruned_loss=0.03718, over 1400670.05 frames.], batch size: 17, lr: 5.28e-04 2022-05-27 20:42:49,937 INFO [train.py:823] (0/4) Epoch 32, batch 900, loss[loss=0.1456, simple_loss=0.2376, pruned_loss=0.02681, over 7024.00 frames.], tot_loss[loss=0.1645, simple_loss=0.2542, pruned_loss=0.03745, over 1405692.27 frames.], batch size: 17, lr: 5.27e-04 2022-05-27 20:43:28,972 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-32.pt 2022-05-27 20:43:43,935 INFO [train.py:823] (0/4) Epoch 33, batch 0, loss[loss=0.1664, simple_loss=0.2564, pruned_loss=0.03824, over 6970.00 frames.], tot_loss[loss=0.1664, simple_loss=0.2564, pruned_loss=0.03824, over 6970.00 frames.], batch size: 29, lr: 5.19e-04 2022-05-27 20:44:22,788 INFO [train.py:823] (0/4) Epoch 33, batch 50, loss[loss=0.1532, simple_loss=0.2328, pruned_loss=0.03676, over 7160.00 frames.], tot_loss[loss=0.1649, simple_loss=0.253, pruned_loss=0.03839, over 317199.37 frames.], batch size: 17, lr: 5.18e-04 2022-05-27 20:45:02,615 INFO [train.py:823] (0/4) Epoch 33, batch 100, loss[loss=0.1409, simple_loss=0.2279, pruned_loss=0.027, over 6829.00 frames.], tot_loss[loss=0.1624, simple_loss=0.2496, pruned_loss=0.03757, over 561428.41 frames.], batch size: 15, lr: 5.18e-04 2022-05-27 20:45:41,686 INFO [train.py:823] (0/4) Epoch 33, batch 150, loss[loss=0.1585, simple_loss=0.265, pruned_loss=0.02599, over 7183.00 frames.], tot_loss[loss=0.1638, simple_loss=0.2522, pruned_loss=0.03775, over 750234.80 frames.], batch size: 21, lr: 5.18e-04 2022-05-27 20:46:21,820 INFO [train.py:823] (0/4) Epoch 33, batch 200, loss[loss=0.1783, simple_loss=0.2787, pruned_loss=0.03897, over 7114.00 frames.], tot_loss[loss=0.1634, simple_loss=0.2522, pruned_loss=0.03731, over 892690.36 frames.], batch size: 20, lr: 5.17e-04 2022-05-27 20:47:00,794 INFO [train.py:823] (0/4) Epoch 33, batch 250, loss[loss=0.1801, simple_loss=0.2695, pruned_loss=0.04531, over 7126.00 frames.], tot_loss[loss=0.1631, simple_loss=0.2517, pruned_loss=0.03723, over 1013255.94 frames.], batch size: 23, lr: 5.17e-04 2022-05-27 20:47:39,838 INFO [train.py:823] (0/4) Epoch 33, batch 300, loss[loss=0.1555, simple_loss=0.2337, pruned_loss=0.03861, over 7168.00 frames.], tot_loss[loss=0.163, simple_loss=0.2516, pruned_loss=0.03723, over 1107317.93 frames.], batch size: 17, lr: 5.16e-04 2022-05-27 20:48:18,978 INFO [train.py:823] (0/4) Epoch 33, batch 350, loss[loss=0.1637, simple_loss=0.2486, pruned_loss=0.03933, over 7366.00 frames.], tot_loss[loss=0.1628, simple_loss=0.2518, pruned_loss=0.03693, over 1176926.25 frames.], batch size: 23, lr: 5.16e-04 2022-05-27 20:48:57,917 INFO [train.py:823] (0/4) Epoch 33, batch 400, loss[loss=0.1758, simple_loss=0.2592, pruned_loss=0.04617, over 7414.00 frames.], tot_loss[loss=0.1636, simple_loss=0.2528, pruned_loss=0.03725, over 1230503.01 frames.], batch size: 22, lr: 5.16e-04 2022-05-27 20:49:37,003 INFO [train.py:823] (0/4) Epoch 33, batch 450, loss[loss=0.1262, simple_loss=0.2179, pruned_loss=0.01722, over 7294.00 frames.], tot_loss[loss=0.1638, simple_loss=0.2529, pruned_loss=0.0374, over 1272074.91 frames.], batch size: 19, lr: 5.15e-04 2022-05-27 20:50:15,608 INFO [train.py:823] (0/4) Epoch 33, batch 500, loss[loss=0.1416, simple_loss=0.2322, pruned_loss=0.02551, over 6920.00 frames.], tot_loss[loss=0.1646, simple_loss=0.2538, pruned_loss=0.03769, over 1306611.18 frames.], batch size: 29, lr: 5.15e-04 2022-05-27 20:50:54,726 INFO [train.py:823] (0/4) Epoch 33, batch 550, loss[loss=0.1479, simple_loss=0.2373, pruned_loss=0.02923, over 7394.00 frames.], tot_loss[loss=0.1649, simple_loss=0.254, pruned_loss=0.03793, over 1335300.26 frames.], batch size: 19, lr: 5.14e-04 2022-05-27 20:51:33,948 INFO [train.py:823] (0/4) Epoch 33, batch 600, loss[loss=0.1631, simple_loss=0.2608, pruned_loss=0.03266, over 7416.00 frames.], tot_loss[loss=0.1641, simple_loss=0.253, pruned_loss=0.03758, over 1355329.16 frames.], batch size: 22, lr: 5.14e-04 2022-05-27 20:52:12,813 INFO [train.py:823] (0/4) Epoch 33, batch 650, loss[loss=0.1651, simple_loss=0.2443, pruned_loss=0.04299, over 7158.00 frames.], tot_loss[loss=0.1626, simple_loss=0.2515, pruned_loss=0.03683, over 1374823.00 frames.], batch size: 17, lr: 5.14e-04 2022-05-27 20:52:51,742 INFO [train.py:823] (0/4) Epoch 33, batch 700, loss[loss=0.1769, simple_loss=0.2659, pruned_loss=0.04398, over 6432.00 frames.], tot_loss[loss=0.1633, simple_loss=0.2527, pruned_loss=0.03696, over 1385985.10 frames.], batch size: 34, lr: 5.13e-04 2022-05-27 20:53:30,716 INFO [train.py:823] (0/4) Epoch 33, batch 750, loss[loss=0.171, simple_loss=0.2604, pruned_loss=0.0408, over 7189.00 frames.], tot_loss[loss=0.1645, simple_loss=0.2539, pruned_loss=0.03754, over 1392102.57 frames.], batch size: 25, lr: 5.13e-04 2022-05-27 20:54:09,380 INFO [train.py:823] (0/4) Epoch 33, batch 800, loss[loss=0.1649, simple_loss=0.2507, pruned_loss=0.03951, over 7163.00 frames.], tot_loss[loss=0.1652, simple_loss=0.2546, pruned_loss=0.03786, over 1391299.93 frames.], batch size: 22, lr: 5.12e-04 2022-05-27 20:54:48,126 INFO [train.py:823] (0/4) Epoch 33, batch 850, loss[loss=0.1643, simple_loss=0.2506, pruned_loss=0.03901, over 7084.00 frames.], tot_loss[loss=0.1647, simple_loss=0.254, pruned_loss=0.03777, over 1400000.61 frames.], batch size: 18, lr: 5.12e-04 2022-05-27 20:55:26,892 INFO [train.py:823] (0/4) Epoch 33, batch 900, loss[loss=0.1413, simple_loss=0.2291, pruned_loss=0.0267, over 7003.00 frames.], tot_loss[loss=0.1645, simple_loss=0.2536, pruned_loss=0.03775, over 1401501.18 frames.], batch size: 16, lr: 5.12e-04 2022-05-27 20:56:06,133 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-33.pt 2022-05-27 20:56:18,083 INFO [train.py:823] (0/4) Epoch 34, batch 0, loss[loss=0.1962, simple_loss=0.2871, pruned_loss=0.05263, over 7244.00 frames.], tot_loss[loss=0.1962, simple_loss=0.2871, pruned_loss=0.05263, over 7244.00 frames.], batch size: 24, lr: 5.04e-04 2022-05-27 20:56:56,730 INFO [train.py:823] (0/4) Epoch 34, batch 50, loss[loss=0.1376, simple_loss=0.2194, pruned_loss=0.02789, over 6845.00 frames.], tot_loss[loss=0.1619, simple_loss=0.2507, pruned_loss=0.03658, over 320262.50 frames.], batch size: 15, lr: 5.03e-04 2022-05-27 20:57:36,509 INFO [train.py:823] (0/4) Epoch 34, batch 100, loss[loss=0.155, simple_loss=0.2425, pruned_loss=0.03377, over 7283.00 frames.], tot_loss[loss=0.164, simple_loss=0.2532, pruned_loss=0.03741, over 560635.13 frames.], batch size: 21, lr: 5.03e-04 2022-05-27 20:58:15,739 INFO [train.py:823] (0/4) Epoch 34, batch 150, loss[loss=0.1828, simple_loss=0.283, pruned_loss=0.04124, over 7314.00 frames.], tot_loss[loss=0.1645, simple_loss=0.2541, pruned_loss=0.03739, over 754088.45 frames.], batch size: 22, lr: 5.02e-04 2022-05-27 20:58:54,846 INFO [train.py:823] (0/4) Epoch 34, batch 200, loss[loss=0.1622, simple_loss=0.2575, pruned_loss=0.03339, over 7021.00 frames.], tot_loss[loss=0.1629, simple_loss=0.2531, pruned_loss=0.03638, over 901929.11 frames.], batch size: 26, lr: 5.02e-04 2022-05-27 20:59:34,242 INFO [train.py:823] (0/4) Epoch 34, batch 250, loss[loss=0.187, simple_loss=0.2776, pruned_loss=0.04813, over 7027.00 frames.], tot_loss[loss=0.1629, simple_loss=0.2531, pruned_loss=0.03631, over 1011771.95 frames.], batch size: 26, lr: 5.02e-04 2022-05-27 21:00:13,163 INFO [train.py:823] (0/4) Epoch 34, batch 300, loss[loss=0.1618, simple_loss=0.2548, pruned_loss=0.03441, over 7371.00 frames.], tot_loss[loss=0.1624, simple_loss=0.2528, pruned_loss=0.03595, over 1102177.77 frames.], batch size: 21, lr: 5.01e-04 2022-05-27 21:00:53,312 INFO [train.py:823] (0/4) Epoch 34, batch 350, loss[loss=0.1747, simple_loss=0.2562, pruned_loss=0.04656, over 7084.00 frames.], tot_loss[loss=0.163, simple_loss=0.2528, pruned_loss=0.03659, over 1169077.79 frames.], batch size: 19, lr: 5.01e-04 2022-05-27 21:01:32,637 INFO [train.py:823] (0/4) Epoch 34, batch 400, loss[loss=0.1653, simple_loss=0.2668, pruned_loss=0.0319, over 7282.00 frames.], tot_loss[loss=0.1635, simple_loss=0.2534, pruned_loss=0.03686, over 1224470.91 frames.], batch size: 21, lr: 5.00e-04 2022-05-27 21:02:11,966 INFO [train.py:823] (0/4) Epoch 34, batch 450, loss[loss=0.1901, simple_loss=0.2896, pruned_loss=0.04531, over 7275.00 frames.], tot_loss[loss=0.1636, simple_loss=0.2534, pruned_loss=0.03691, over 1270468.98 frames.], batch size: 20, lr: 5.00e-04 2022-05-27 21:02:51,473 INFO [train.py:823] (0/4) Epoch 34, batch 500, loss[loss=0.162, simple_loss=0.2511, pruned_loss=0.03639, over 7160.00 frames.], tot_loss[loss=0.1631, simple_loss=0.2527, pruned_loss=0.03675, over 1302683.62 frames.], batch size: 23, lr: 5.00e-04 2022-05-27 21:03:31,211 INFO [train.py:823] (0/4) Epoch 34, batch 550, loss[loss=0.1988, simple_loss=0.3011, pruned_loss=0.04827, over 7210.00 frames.], tot_loss[loss=0.1624, simple_loss=0.2525, pruned_loss=0.03613, over 1334541.17 frames.], batch size: 25, lr: 4.99e-04 2022-05-27 21:04:10,490 INFO [train.py:823] (0/4) Epoch 34, batch 600, loss[loss=0.1652, simple_loss=0.2417, pruned_loss=0.04431, over 7310.00 frames.], tot_loss[loss=0.1632, simple_loss=0.2534, pruned_loss=0.03648, over 1352715.69 frames.], batch size: 17, lr: 4.99e-04 2022-05-27 21:04:51,416 INFO [train.py:823] (0/4) Epoch 34, batch 650, loss[loss=0.1509, simple_loss=0.2494, pruned_loss=0.02621, over 6964.00 frames.], tot_loss[loss=0.1622, simple_loss=0.2519, pruned_loss=0.03627, over 1367416.29 frames.], batch size: 29, lr: 4.99e-04 2022-05-27 21:04:52,544 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/checkpoint-32000.pt 2022-05-27 21:05:35,673 INFO [train.py:823] (0/4) Epoch 34, batch 700, loss[loss=0.157, simple_loss=0.2488, pruned_loss=0.03256, over 7378.00 frames.], tot_loss[loss=0.1625, simple_loss=0.2519, pruned_loss=0.03657, over 1377483.94 frames.], batch size: 20, lr: 4.98e-04 2022-05-27 21:06:14,466 INFO [train.py:823] (0/4) Epoch 34, batch 750, loss[loss=0.1325, simple_loss=0.2189, pruned_loss=0.02304, over 7001.00 frames.], tot_loss[loss=0.1631, simple_loss=0.2521, pruned_loss=0.0371, over 1388964.78 frames.], batch size: 16, lr: 4.98e-04 2022-05-27 21:06:53,488 INFO [train.py:823] (0/4) Epoch 34, batch 800, loss[loss=0.1541, simple_loss=0.2473, pruned_loss=0.03041, over 7202.00 frames.], tot_loss[loss=0.1625, simple_loss=0.2513, pruned_loss=0.03681, over 1396582.02 frames.], batch size: 19, lr: 4.97e-04 2022-05-27 21:07:32,182 INFO [train.py:823] (0/4) Epoch 34, batch 850, loss[loss=0.1558, simple_loss=0.2448, pruned_loss=0.0334, over 7373.00 frames.], tot_loss[loss=0.1627, simple_loss=0.252, pruned_loss=0.03668, over 1396571.22 frames.], batch size: 21, lr: 4.97e-04 2022-05-27 21:08:12,983 INFO [train.py:823] (0/4) Epoch 34, batch 900, loss[loss=0.157, simple_loss=0.2358, pruned_loss=0.03915, over 7102.00 frames.], tot_loss[loss=0.1613, simple_loss=0.2509, pruned_loss=0.03583, over 1401153.96 frames.], batch size: 18, lr: 4.97e-04 2022-05-27 21:08:51,509 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-34.pt 2022-05-27 21:09:07,189 INFO [train.py:823] (0/4) Epoch 35, batch 0, loss[loss=0.1887, simple_loss=0.28, pruned_loss=0.04868, over 7175.00 frames.], tot_loss[loss=0.1887, simple_loss=0.28, pruned_loss=0.04868, over 7175.00 frames.], batch size: 21, lr: 4.89e-04 2022-05-27 21:09:48,002 INFO [train.py:823] (0/4) Epoch 35, batch 50, loss[loss=0.1433, simple_loss=0.2188, pruned_loss=0.03386, over 7190.00 frames.], tot_loss[loss=0.1628, simple_loss=0.2525, pruned_loss=0.03659, over 323989.39 frames.], batch size: 18, lr: 4.89e-04 2022-05-27 21:10:26,966 INFO [train.py:823] (0/4) Epoch 35, batch 100, loss[loss=0.1503, simple_loss=0.2423, pruned_loss=0.02915, over 6398.00 frames.], tot_loss[loss=0.163, simple_loss=0.2524, pruned_loss=0.03678, over 568876.82 frames.], batch size: 34, lr: 4.88e-04 2022-05-27 21:11:06,201 INFO [train.py:823] (0/4) Epoch 35, batch 150, loss[loss=0.1804, simple_loss=0.266, pruned_loss=0.04744, over 7192.00 frames.], tot_loss[loss=0.1618, simple_loss=0.2507, pruned_loss=0.03641, over 754838.26 frames.], batch size: 25, lr: 4.88e-04 2022-05-27 21:11:44,970 INFO [train.py:823] (0/4) Epoch 35, batch 200, loss[loss=0.1647, simple_loss=0.2691, pruned_loss=0.03011, over 6880.00 frames.], tot_loss[loss=0.1601, simple_loss=0.2498, pruned_loss=0.03517, over 904194.95 frames.], batch size: 29, lr: 4.88e-04 2022-05-27 21:12:24,214 INFO [train.py:823] (0/4) Epoch 35, batch 250, loss[loss=0.1582, simple_loss=0.2441, pruned_loss=0.03612, over 7245.00 frames.], tot_loss[loss=0.1599, simple_loss=0.2497, pruned_loss=0.03503, over 1014115.61 frames.], batch size: 24, lr: 4.87e-04 2022-05-27 21:13:03,249 INFO [train.py:823] (0/4) Epoch 35, batch 300, loss[loss=0.1752, simple_loss=0.2696, pruned_loss=0.04043, over 7285.00 frames.], tot_loss[loss=0.1605, simple_loss=0.2502, pruned_loss=0.03547, over 1105833.11 frames.], batch size: 21, lr: 4.87e-04 2022-05-27 21:13:42,447 INFO [train.py:823] (0/4) Epoch 35, batch 350, loss[loss=0.1501, simple_loss=0.2287, pruned_loss=0.03573, over 7092.00 frames.], tot_loss[loss=0.1604, simple_loss=0.2502, pruned_loss=0.03532, over 1172053.64 frames.], batch size: 18, lr: 4.87e-04 2022-05-27 21:14:21,159 INFO [train.py:823] (0/4) Epoch 35, batch 400, loss[loss=0.1418, simple_loss=0.2344, pruned_loss=0.02463, over 7157.00 frames.], tot_loss[loss=0.1606, simple_loss=0.2505, pruned_loss=0.03532, over 1222162.01 frames.], batch size: 22, lr: 4.86e-04 2022-05-27 21:15:00,064 INFO [train.py:823] (0/4) Epoch 35, batch 450, loss[loss=0.1472, simple_loss=0.2309, pruned_loss=0.0318, over 7277.00 frames.], tot_loss[loss=0.161, simple_loss=0.2512, pruned_loss=0.03543, over 1270070.17 frames.], batch size: 17, lr: 4.86e-04 2022-05-27 21:15:38,761 INFO [train.py:823] (0/4) Epoch 35, batch 500, loss[loss=0.1339, simple_loss=0.2151, pruned_loss=0.02635, over 7022.00 frames.], tot_loss[loss=0.1597, simple_loss=0.2498, pruned_loss=0.03481, over 1304531.32 frames.], batch size: 17, lr: 4.86e-04 2022-05-27 21:16:17,687 INFO [train.py:823] (0/4) Epoch 35, batch 550, loss[loss=0.1494, simple_loss=0.2342, pruned_loss=0.03232, over 7021.00 frames.], tot_loss[loss=0.1595, simple_loss=0.2495, pruned_loss=0.03477, over 1328332.34 frames.], batch size: 17, lr: 4.85e-04 2022-05-27 21:16:57,046 INFO [train.py:823] (0/4) Epoch 35, batch 600, loss[loss=0.2118, simple_loss=0.2873, pruned_loss=0.06818, over 7283.00 frames.], tot_loss[loss=0.1602, simple_loss=0.2501, pruned_loss=0.03515, over 1349035.48 frames.], batch size: 20, lr: 4.85e-04 2022-05-27 21:17:36,295 INFO [train.py:823] (0/4) Epoch 35, batch 650, loss[loss=0.1619, simple_loss=0.2506, pruned_loss=0.03657, over 7070.00 frames.], tot_loss[loss=0.1598, simple_loss=0.2495, pruned_loss=0.03505, over 1367443.08 frames.], batch size: 26, lr: 4.84e-04 2022-05-27 21:18:15,571 INFO [train.py:823] (0/4) Epoch 35, batch 700, loss[loss=0.1508, simple_loss=0.249, pruned_loss=0.02636, over 7281.00 frames.], tot_loss[loss=0.1602, simple_loss=0.2496, pruned_loss=0.03536, over 1377191.58 frames.], batch size: 20, lr: 4.84e-04 2022-05-27 21:18:54,742 INFO [train.py:823] (0/4) Epoch 35, batch 750, loss[loss=0.1621, simple_loss=0.2485, pruned_loss=0.03782, over 7091.00 frames.], tot_loss[loss=0.1609, simple_loss=0.2505, pruned_loss=0.03566, over 1389443.53 frames.], batch size: 19, lr: 4.84e-04 2022-05-27 21:19:32,971 INFO [train.py:823] (0/4) Epoch 35, batch 800, loss[loss=0.1627, simple_loss=0.2437, pruned_loss=0.04082, over 7309.00 frames.], tot_loss[loss=0.161, simple_loss=0.2506, pruned_loss=0.0357, over 1394033.84 frames.], batch size: 18, lr: 4.83e-04 2022-05-27 21:20:12,107 INFO [train.py:823] (0/4) Epoch 35, batch 850, loss[loss=0.1574, simple_loss=0.2469, pruned_loss=0.03401, over 7422.00 frames.], tot_loss[loss=0.1616, simple_loss=0.2511, pruned_loss=0.03605, over 1401846.74 frames.], batch size: 22, lr: 4.83e-04 2022-05-27 21:20:50,657 INFO [train.py:823] (0/4) Epoch 35, batch 900, loss[loss=0.1597, simple_loss=0.2551, pruned_loss=0.03217, over 6559.00 frames.], tot_loss[loss=0.1611, simple_loss=0.2506, pruned_loss=0.03586, over 1400056.92 frames.], batch size: 35, lr: 4.83e-04 2022-05-27 21:21:29,569 INFO [train.py:823] (0/4) Epoch 35, batch 950, loss[loss=0.1939, simple_loss=0.2867, pruned_loss=0.05058, over 5315.00 frames.], tot_loss[loss=0.1624, simple_loss=0.2519, pruned_loss=0.0364, over 1379631.45 frames.], batch size: 48, lr: 4.82e-04 2022-05-27 21:21:30,802 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-35.pt 2022-05-27 21:21:42,972 INFO [train.py:823] (0/4) Epoch 36, batch 0, loss[loss=0.1775, simple_loss=0.2704, pruned_loss=0.04227, over 7422.00 frames.], tot_loss[loss=0.1775, simple_loss=0.2704, pruned_loss=0.04227, over 7422.00 frames.], batch size: 22, lr: 4.76e-04 2022-05-27 21:22:22,326 INFO [train.py:823] (0/4) Epoch 36, batch 50, loss[loss=0.1397, simple_loss=0.2241, pruned_loss=0.02761, over 7148.00 frames.], tot_loss[loss=0.1584, simple_loss=0.2475, pruned_loss=0.03459, over 319378.23 frames.], batch size: 17, lr: 4.75e-04 2022-05-27 21:23:01,909 INFO [train.py:823] (0/4) Epoch 36, batch 100, loss[loss=0.1519, simple_loss=0.2556, pruned_loss=0.02413, over 6697.00 frames.], tot_loss[loss=0.1597, simple_loss=0.249, pruned_loss=0.03519, over 565123.46 frames.], batch size: 34, lr: 4.75e-04 2022-05-27 21:23:40,583 INFO [train.py:823] (0/4) Epoch 36, batch 150, loss[loss=0.1414, simple_loss=0.2451, pruned_loss=0.01885, over 7173.00 frames.], tot_loss[loss=0.162, simple_loss=0.2516, pruned_loss=0.03626, over 752012.27 frames.], batch size: 25, lr: 4.74e-04 2022-05-27 21:24:21,414 INFO [train.py:823] (0/4) Epoch 36, batch 200, loss[loss=0.1392, simple_loss=0.2194, pruned_loss=0.02948, over 7307.00 frames.], tot_loss[loss=0.1614, simple_loss=0.2505, pruned_loss=0.03613, over 899231.73 frames.], batch size: 17, lr: 4.74e-04 2022-05-27 21:24:59,991 INFO [train.py:823] (0/4) Epoch 36, batch 250, loss[loss=0.1544, simple_loss=0.2447, pruned_loss=0.03201, over 7389.00 frames.], tot_loss[loss=0.1617, simple_loss=0.2511, pruned_loss=0.03611, over 1012557.00 frames.], batch size: 19, lr: 4.74e-04 2022-05-27 21:25:39,294 INFO [train.py:823] (0/4) Epoch 36, batch 300, loss[loss=0.1674, simple_loss=0.258, pruned_loss=0.03839, over 7331.00 frames.], tot_loss[loss=0.1616, simple_loss=0.2506, pruned_loss=0.03629, over 1101537.66 frames.], batch size: 23, lr: 4.73e-04 2022-05-27 21:26:18,971 INFO [train.py:823] (0/4) Epoch 36, batch 350, loss[loss=0.1712, simple_loss=0.2551, pruned_loss=0.0437, over 7377.00 frames.], tot_loss[loss=0.1614, simple_loss=0.2507, pruned_loss=0.03606, over 1172013.22 frames.], batch size: 20, lr: 4.73e-04 2022-05-27 21:26:58,351 INFO [train.py:823] (0/4) Epoch 36, batch 400, loss[loss=0.1592, simple_loss=0.2464, pruned_loss=0.03604, over 7086.00 frames.], tot_loss[loss=0.1615, simple_loss=0.2511, pruned_loss=0.0359, over 1227459.55 frames.], batch size: 18, lr: 4.73e-04 2022-05-27 21:27:39,298 INFO [train.py:823] (0/4) Epoch 36, batch 450, loss[loss=0.1664, simple_loss=0.2604, pruned_loss=0.03615, over 7050.00 frames.], tot_loss[loss=0.1613, simple_loss=0.2505, pruned_loss=0.03601, over 1269986.73 frames.], batch size: 26, lr: 4.72e-04 2022-05-27 21:28:18,454 INFO [train.py:823] (0/4) Epoch 36, batch 500, loss[loss=0.1674, simple_loss=0.2496, pruned_loss=0.04259, over 7222.00 frames.], tot_loss[loss=0.1602, simple_loss=0.2492, pruned_loss=0.03556, over 1301818.23 frames.], batch size: 24, lr: 4.72e-04 2022-05-27 21:28:57,546 INFO [train.py:823] (0/4) Epoch 36, batch 550, loss[loss=0.1457, simple_loss=0.2248, pruned_loss=0.03333, over 7301.00 frames.], tot_loss[loss=0.1596, simple_loss=0.2487, pruned_loss=0.03527, over 1328292.89 frames.], batch size: 17, lr: 4.72e-04 2022-05-27 21:29:37,086 INFO [train.py:823] (0/4) Epoch 36, batch 600, loss[loss=0.1286, simple_loss=0.2135, pruned_loss=0.02187, over 7296.00 frames.], tot_loss[loss=0.1597, simple_loss=0.2484, pruned_loss=0.03553, over 1347300.39 frames.], batch size: 17, lr: 4.71e-04 2022-05-27 21:30:16,485 INFO [train.py:823] (0/4) Epoch 36, batch 650, loss[loss=0.1786, simple_loss=0.2704, pruned_loss=0.04343, over 7378.00 frames.], tot_loss[loss=0.161, simple_loss=0.2503, pruned_loss=0.03588, over 1362700.66 frames.], batch size: 21, lr: 4.71e-04 2022-05-27 21:30:56,835 INFO [train.py:823] (0/4) Epoch 36, batch 700, loss[loss=0.1322, simple_loss=0.2176, pruned_loss=0.02338, over 7304.00 frames.], tot_loss[loss=0.1603, simple_loss=0.2497, pruned_loss=0.03541, over 1379204.29 frames.], batch size: 17, lr: 4.71e-04 2022-05-27 21:31:36,036 INFO [train.py:823] (0/4) Epoch 36, batch 750, loss[loss=0.1579, simple_loss=0.2499, pruned_loss=0.03293, over 7284.00 frames.], tot_loss[loss=0.1597, simple_loss=0.249, pruned_loss=0.0352, over 1387413.03 frames.], batch size: 21, lr: 4.70e-04 2022-05-27 21:32:16,417 INFO [train.py:823] (0/4) Epoch 36, batch 800, loss[loss=0.1504, simple_loss=0.2397, pruned_loss=0.03055, over 7372.00 frames.], tot_loss[loss=0.1597, simple_loss=0.2492, pruned_loss=0.03512, over 1388803.54 frames.], batch size: 21, lr: 4.70e-04 2022-05-27 21:32:55,433 INFO [train.py:823] (0/4) Epoch 36, batch 850, loss[loss=0.1841, simple_loss=0.2753, pruned_loss=0.04643, over 7353.00 frames.], tot_loss[loss=0.1598, simple_loss=0.249, pruned_loss=0.03528, over 1389254.64 frames.], batch size: 23, lr: 4.70e-04 2022-05-27 21:33:34,404 INFO [train.py:823] (0/4) Epoch 36, batch 900, loss[loss=0.1621, simple_loss=0.2594, pruned_loss=0.03239, over 7424.00 frames.], tot_loss[loss=0.1598, simple_loss=0.2494, pruned_loss=0.03509, over 1396881.66 frames.], batch size: 22, lr: 4.69e-04 2022-05-27 21:34:13,053 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-36.pt 2022-05-27 21:34:27,383 INFO [train.py:823] (0/4) Epoch 37, batch 0, loss[loss=0.1843, simple_loss=0.2887, pruned_loss=0.04, over 6635.00 frames.], tot_loss[loss=0.1843, simple_loss=0.2887, pruned_loss=0.04, over 6635.00 frames.], batch size: 34, lr: 4.63e-04 2022-05-27 21:35:06,541 INFO [train.py:823] (0/4) Epoch 37, batch 50, loss[loss=0.1726, simple_loss=0.2647, pruned_loss=0.04023, over 7307.00 frames.], tot_loss[loss=0.1637, simple_loss=0.2546, pruned_loss=0.0364, over 318525.74 frames.], batch size: 22, lr: 4.62e-04 2022-05-27 21:35:45,355 INFO [train.py:823] (0/4) Epoch 37, batch 100, loss[loss=0.1712, simple_loss=0.2651, pruned_loss=0.03866, over 7236.00 frames.], tot_loss[loss=0.1613, simple_loss=0.2521, pruned_loss=0.03529, over 561276.73 frames.], batch size: 24, lr: 4.62e-04 2022-05-27 21:36:24,662 INFO [train.py:823] (0/4) Epoch 37, batch 150, loss[loss=0.1387, simple_loss=0.2335, pruned_loss=0.02198, over 7195.00 frames.], tot_loss[loss=0.1615, simple_loss=0.2517, pruned_loss=0.03568, over 749955.91 frames.], batch size: 21, lr: 4.62e-04 2022-05-27 21:37:04,060 INFO [train.py:823] (0/4) Epoch 37, batch 200, loss[loss=0.2038, simple_loss=0.2914, pruned_loss=0.05807, over 7234.00 frames.], tot_loss[loss=0.1599, simple_loss=0.2491, pruned_loss=0.0354, over 902402.59 frames.], batch size: 24, lr: 4.61e-04 2022-05-27 21:37:43,522 INFO [train.py:823] (0/4) Epoch 37, batch 250, loss[loss=0.1565, simple_loss=0.2524, pruned_loss=0.0303, over 7040.00 frames.], tot_loss[loss=0.1606, simple_loss=0.2505, pruned_loss=0.03536, over 1018745.61 frames.], batch size: 26, lr: 4.61e-04 2022-05-27 21:38:22,753 INFO [train.py:823] (0/4) Epoch 37, batch 300, loss[loss=0.132, simple_loss=0.217, pruned_loss=0.02347, over 7017.00 frames.], tot_loss[loss=0.1604, simple_loss=0.2497, pruned_loss=0.03559, over 1104991.59 frames.], batch size: 16, lr: 4.61e-04 2022-05-27 21:39:02,453 INFO [train.py:823] (0/4) Epoch 37, batch 350, loss[loss=0.2112, simple_loss=0.295, pruned_loss=0.06369, over 7227.00 frames.], tot_loss[loss=0.1612, simple_loss=0.2508, pruned_loss=0.03577, over 1172141.29 frames.], batch size: 25, lr: 4.60e-04 2022-05-27 21:39:41,212 INFO [train.py:823] (0/4) Epoch 37, batch 400, loss[loss=0.1324, simple_loss=0.217, pruned_loss=0.02387, over 7302.00 frames.], tot_loss[loss=0.1613, simple_loss=0.2513, pruned_loss=0.03567, over 1228452.71 frames.], batch size: 17, lr: 4.60e-04 2022-05-27 21:40:19,851 INFO [train.py:823] (0/4) Epoch 37, batch 450, loss[loss=0.1609, simple_loss=0.2437, pruned_loss=0.039, over 7204.00 frames.], tot_loss[loss=0.1612, simple_loss=0.2513, pruned_loss=0.03549, over 1268148.21 frames.], batch size: 19, lr: 4.60e-04 2022-05-27 21:40:58,931 INFO [train.py:823] (0/4) Epoch 37, batch 500, loss[loss=0.1572, simple_loss=0.2399, pruned_loss=0.03727, over 7017.00 frames.], tot_loss[loss=0.1615, simple_loss=0.2516, pruned_loss=0.03576, over 1303576.46 frames.], batch size: 16, lr: 4.59e-04 2022-05-27 21:41:38,548 INFO [train.py:823] (0/4) Epoch 37, batch 550, loss[loss=0.1266, simple_loss=0.2168, pruned_loss=0.01818, over 7009.00 frames.], tot_loss[loss=0.1607, simple_loss=0.2507, pruned_loss=0.03535, over 1329231.87 frames.], batch size: 16, lr: 4.59e-04 2022-05-27 21:42:17,368 INFO [train.py:823] (0/4) Epoch 37, batch 600, loss[loss=0.1629, simple_loss=0.2556, pruned_loss=0.03508, over 7333.00 frames.], tot_loss[loss=0.1607, simple_loss=0.2509, pruned_loss=0.03525, over 1349029.63 frames.], batch size: 23, lr: 4.59e-04 2022-05-27 21:42:55,880 INFO [train.py:823] (0/4) Epoch 37, batch 650, loss[loss=0.1672, simple_loss=0.2444, pruned_loss=0.04498, over 7163.00 frames.], tot_loss[loss=0.1604, simple_loss=0.2502, pruned_loss=0.03529, over 1364174.92 frames.], batch size: 17, lr: 4.58e-04 2022-05-27 21:43:34,820 INFO [train.py:823] (0/4) Epoch 37, batch 700, loss[loss=0.1499, simple_loss=0.2456, pruned_loss=0.02714, over 7424.00 frames.], tot_loss[loss=0.1602, simple_loss=0.2503, pruned_loss=0.03506, over 1371300.87 frames.], batch size: 22, lr: 4.58e-04 2022-05-27 21:44:14,148 INFO [train.py:823] (0/4) Epoch 37, batch 750, loss[loss=0.1791, simple_loss=0.2585, pruned_loss=0.04979, over 5292.00 frames.], tot_loss[loss=0.1601, simple_loss=0.2505, pruned_loss=0.03488, over 1378771.64 frames.], batch size: 46, lr: 4.58e-04 2022-05-27 21:44:53,002 INFO [train.py:823] (0/4) Epoch 37, batch 800, loss[loss=0.146, simple_loss=0.2424, pruned_loss=0.02477, over 7279.00 frames.], tot_loss[loss=0.16, simple_loss=0.2504, pruned_loss=0.0348, over 1383103.50 frames.], batch size: 21, lr: 4.57e-04 2022-05-27 21:45:31,954 INFO [train.py:823] (0/4) Epoch 37, batch 850, loss[loss=0.1262, simple_loss=0.215, pruned_loss=0.01874, over 7230.00 frames.], tot_loss[loss=0.1589, simple_loss=0.2491, pruned_loss=0.03436, over 1385080.33 frames.], batch size: 16, lr: 4.57e-04 2022-05-27 21:46:10,767 INFO [train.py:823] (0/4) Epoch 37, batch 900, loss[loss=0.1862, simple_loss=0.2854, pruned_loss=0.04352, over 7140.00 frames.], tot_loss[loss=0.1587, simple_loss=0.2487, pruned_loss=0.03431, over 1392564.63 frames.], batch size: 23, lr: 4.57e-04 2022-05-27 21:46:50,835 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-37.pt 2022-05-27 21:47:05,041 INFO [train.py:823] (0/4) Epoch 38, batch 0, loss[loss=0.1659, simple_loss=0.2594, pruned_loss=0.03621, over 7403.00 frames.], tot_loss[loss=0.1659, simple_loss=0.2594, pruned_loss=0.03621, over 7403.00 frames.], batch size: 19, lr: 4.50e-04 2022-05-27 21:47:43,993 INFO [train.py:823] (0/4) Epoch 38, batch 50, loss[loss=0.1603, simple_loss=0.2523, pruned_loss=0.03419, over 7104.00 frames.], tot_loss[loss=0.1612, simple_loss=0.2505, pruned_loss=0.03595, over 321676.42 frames.], batch size: 19, lr: 4.50e-04 2022-05-27 21:48:24,772 INFO [train.py:823] (0/4) Epoch 38, batch 100, loss[loss=0.1648, simple_loss=0.2586, pruned_loss=0.03551, over 7336.00 frames.], tot_loss[loss=0.159, simple_loss=0.2475, pruned_loss=0.03524, over 564678.45 frames.], batch size: 23, lr: 4.50e-04 2022-05-27 21:49:03,850 INFO [train.py:823] (0/4) Epoch 38, batch 150, loss[loss=0.1586, simple_loss=0.2489, pruned_loss=0.03415, over 7009.00 frames.], tot_loss[loss=0.1577, simple_loss=0.2469, pruned_loss=0.03422, over 753178.30 frames.], batch size: 26, lr: 4.50e-04 2022-05-27 21:49:43,211 INFO [train.py:823] (0/4) Epoch 38, batch 200, loss[loss=0.1769, simple_loss=0.2697, pruned_loss=0.04202, over 6459.00 frames.], tot_loss[loss=0.1583, simple_loss=0.2481, pruned_loss=0.03425, over 901142.64 frames.], batch size: 34, lr: 4.49e-04 2022-05-27 21:50:22,158 INFO [train.py:823] (0/4) Epoch 38, batch 250, loss[loss=0.1566, simple_loss=0.2424, pruned_loss=0.03537, over 7109.00 frames.], tot_loss[loss=0.1586, simple_loss=0.248, pruned_loss=0.03459, over 1019923.08 frames.], batch size: 20, lr: 4.49e-04 2022-05-27 21:51:03,012 INFO [train.py:823] (0/4) Epoch 38, batch 300, loss[loss=0.136, simple_loss=0.22, pruned_loss=0.02601, over 7280.00 frames.], tot_loss[loss=0.159, simple_loss=0.2485, pruned_loss=0.03474, over 1107291.37 frames.], batch size: 21, lr: 4.49e-04 2022-05-27 21:51:42,035 INFO [train.py:823] (0/4) Epoch 38, batch 350, loss[loss=0.1464, simple_loss=0.2224, pruned_loss=0.03524, over 6834.00 frames.], tot_loss[loss=0.1598, simple_loss=0.2492, pruned_loss=0.03521, over 1181692.81 frames.], batch size: 15, lr: 4.48e-04 2022-05-27 21:52:21,193 INFO [train.py:823] (0/4) Epoch 38, batch 400, loss[loss=0.1837, simple_loss=0.2651, pruned_loss=0.05116, over 4590.00 frames.], tot_loss[loss=0.1598, simple_loss=0.2495, pruned_loss=0.03502, over 1235126.87 frames.], batch size: 46, lr: 4.48e-04 2022-05-27 21:53:00,081 INFO [train.py:823] (0/4) Epoch 38, batch 450, loss[loss=0.165, simple_loss=0.257, pruned_loss=0.03649, over 7189.00 frames.], tot_loss[loss=0.1589, simple_loss=0.2484, pruned_loss=0.03467, over 1279513.47 frames.], batch size: 20, lr: 4.48e-04 2022-05-27 21:53:39,581 INFO [train.py:823] (0/4) Epoch 38, batch 500, loss[loss=0.2045, simple_loss=0.2976, pruned_loss=0.05565, over 7273.00 frames.], tot_loss[loss=0.1583, simple_loss=0.2476, pruned_loss=0.03448, over 1314383.28 frames.], batch size: 21, lr: 4.47e-04 2022-05-27 21:54:19,854 INFO [train.py:823] (0/4) Epoch 38, batch 550, loss[loss=0.1758, simple_loss=0.2649, pruned_loss=0.04339, over 7197.00 frames.], tot_loss[loss=0.1587, simple_loss=0.2481, pruned_loss=0.03459, over 1332505.75 frames.], batch size: 20, lr: 4.47e-04 2022-05-27 21:54:59,341 INFO [train.py:823] (0/4) Epoch 38, batch 600, loss[loss=0.1387, simple_loss=0.2338, pruned_loss=0.02185, over 6527.00 frames.], tot_loss[loss=0.1597, simple_loss=0.2491, pruned_loss=0.03514, over 1351308.73 frames.], batch size: 34, lr: 4.47e-04 2022-05-27 21:55:39,782 INFO [train.py:823] (0/4) Epoch 38, batch 650, loss[loss=0.1292, simple_loss=0.2197, pruned_loss=0.01932, over 7285.00 frames.], tot_loss[loss=0.1594, simple_loss=0.249, pruned_loss=0.0349, over 1367305.62 frames.], batch size: 20, lr: 4.46e-04 2022-05-27 21:56:18,840 INFO [train.py:823] (0/4) Epoch 38, batch 700, loss[loss=0.1834, simple_loss=0.269, pruned_loss=0.04896, over 7162.00 frames.], tot_loss[loss=0.1604, simple_loss=0.2501, pruned_loss=0.03537, over 1377496.31 frames.], batch size: 22, lr: 4.46e-04 2022-05-27 21:56:57,007 INFO [train.py:823] (0/4) Epoch 38, batch 750, loss[loss=0.1728, simple_loss=0.2666, pruned_loss=0.03952, over 7241.00 frames.], tot_loss[loss=0.1595, simple_loss=0.2495, pruned_loss=0.0348, over 1382585.16 frames.], batch size: 24, lr: 4.46e-04 2022-05-27 21:57:36,103 INFO [train.py:823] (0/4) Epoch 38, batch 800, loss[loss=0.1662, simple_loss=0.2624, pruned_loss=0.03502, over 7376.00 frames.], tot_loss[loss=0.1593, simple_loss=0.2493, pruned_loss=0.03468, over 1385576.74 frames.], batch size: 21, lr: 4.45e-04 2022-05-27 21:58:14,995 INFO [train.py:823] (0/4) Epoch 38, batch 850, loss[loss=0.1793, simple_loss=0.2785, pruned_loss=0.04001, over 6921.00 frames.], tot_loss[loss=0.1594, simple_loss=0.2494, pruned_loss=0.03467, over 1395205.21 frames.], batch size: 29, lr: 4.45e-04 2022-05-27 21:58:54,466 INFO [train.py:823] (0/4) Epoch 38, batch 900, loss[loss=0.1535, simple_loss=0.2447, pruned_loss=0.03109, over 7020.00 frames.], tot_loss[loss=0.16, simple_loss=0.25, pruned_loss=0.03506, over 1398915.83 frames.], batch size: 16, lr: 4.45e-04 2022-05-27 21:59:32,661 INFO [train.py:823] (0/4) Epoch 38, batch 950, loss[loss=0.152, simple_loss=0.2462, pruned_loss=0.02897, over 5243.00 frames.], tot_loss[loss=0.1608, simple_loss=0.2504, pruned_loss=0.03559, over 1375342.06 frames.], batch size: 46, lr: 4.45e-04 2022-05-27 21:59:34,031 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-38.pt 2022-05-27 21:59:45,981 INFO [train.py:823] (0/4) Epoch 39, batch 0, loss[loss=0.1528, simple_loss=0.2411, pruned_loss=0.03229, over 7298.00 frames.], tot_loss[loss=0.1528, simple_loss=0.2411, pruned_loss=0.03229, over 7298.00 frames.], batch size: 19, lr: 4.39e-04 2022-05-27 22:00:25,286 INFO [train.py:823] (0/4) Epoch 39, batch 50, loss[loss=0.1653, simple_loss=0.2629, pruned_loss=0.03381, over 7437.00 frames.], tot_loss[loss=0.1595, simple_loss=0.25, pruned_loss=0.03445, over 322043.20 frames.], batch size: 22, lr: 4.39e-04 2022-05-27 22:01:04,267 INFO [train.py:823] (0/4) Epoch 39, batch 100, loss[loss=0.1322, simple_loss=0.2191, pruned_loss=0.02263, over 7317.00 frames.], tot_loss[loss=0.1563, simple_loss=0.246, pruned_loss=0.03333, over 566699.54 frames.], batch size: 18, lr: 4.38e-04 2022-05-27 22:01:43,940 INFO [train.py:823] (0/4) Epoch 39, batch 150, loss[loss=0.1801, simple_loss=0.2708, pruned_loss=0.04469, over 7194.00 frames.], tot_loss[loss=0.1558, simple_loss=0.2448, pruned_loss=0.03342, over 755141.61 frames.], batch size: 25, lr: 4.38e-04 2022-05-27 22:02:23,462 INFO [train.py:823] (0/4) Epoch 39, batch 200, loss[loss=0.1603, simple_loss=0.2508, pruned_loss=0.03492, over 7386.00 frames.], tot_loss[loss=0.1557, simple_loss=0.2448, pruned_loss=0.03328, over 907062.08 frames.], batch size: 19, lr: 4.38e-04 2022-05-27 22:03:03,268 INFO [train.py:823] (0/4) Epoch 39, batch 250, loss[loss=0.158, simple_loss=0.2494, pruned_loss=0.03333, over 7285.00 frames.], tot_loss[loss=0.1567, simple_loss=0.2466, pruned_loss=0.03341, over 1021280.90 frames.], batch size: 19, lr: 4.37e-04 2022-05-27 22:03:42,556 INFO [train.py:823] (0/4) Epoch 39, batch 300, loss[loss=0.1434, simple_loss=0.2418, pruned_loss=0.02248, over 7295.00 frames.], tot_loss[loss=0.1577, simple_loss=0.2475, pruned_loss=0.03395, over 1113672.30 frames.], batch size: 19, lr: 4.37e-04 2022-05-27 22:04:21,964 INFO [train.py:823] (0/4) Epoch 39, batch 350, loss[loss=0.1628, simple_loss=0.2588, pruned_loss=0.03339, over 7369.00 frames.], tot_loss[loss=0.1582, simple_loss=0.248, pruned_loss=0.03414, over 1185595.53 frames.], batch size: 20, lr: 4.37e-04 2022-05-27 22:05:01,301 INFO [train.py:823] (0/4) Epoch 39, batch 400, loss[loss=0.1561, simple_loss=0.2405, pruned_loss=0.03585, over 7020.00 frames.], tot_loss[loss=0.1584, simple_loss=0.2484, pruned_loss=0.0342, over 1242631.43 frames.], batch size: 17, lr: 4.36e-04 2022-05-27 22:05:40,476 INFO [train.py:823] (0/4) Epoch 39, batch 450, loss[loss=0.1814, simple_loss=0.2732, pruned_loss=0.04476, over 7070.00 frames.], tot_loss[loss=0.1581, simple_loss=0.2484, pruned_loss=0.03389, over 1281286.47 frames.], batch size: 26, lr: 4.36e-04 2022-05-27 22:06:19,096 INFO [train.py:823] (0/4) Epoch 39, batch 500, loss[loss=0.1764, simple_loss=0.2577, pruned_loss=0.04756, over 5084.00 frames.], tot_loss[loss=0.1578, simple_loss=0.2478, pruned_loss=0.03386, over 1311321.67 frames.], batch size: 48, lr: 4.36e-04 2022-05-27 22:06:58,252 INFO [train.py:823] (0/4) Epoch 39, batch 550, loss[loss=0.1645, simple_loss=0.266, pruned_loss=0.03152, over 7216.00 frames.], tot_loss[loss=0.1583, simple_loss=0.2484, pruned_loss=0.03411, over 1332058.43 frames.], batch size: 25, lr: 4.36e-04 2022-05-27 22:07:37,595 INFO [train.py:823] (0/4) Epoch 39, batch 600, loss[loss=0.1349, simple_loss=0.221, pruned_loss=0.02444, over 7435.00 frames.], tot_loss[loss=0.1583, simple_loss=0.2481, pruned_loss=0.03426, over 1354644.76 frames.], batch size: 18, lr: 4.35e-04 2022-05-27 22:08:17,285 INFO [train.py:823] (0/4) Epoch 39, batch 650, loss[loss=0.1731, simple_loss=0.2569, pruned_loss=0.04469, over 7403.00 frames.], tot_loss[loss=0.1583, simple_loss=0.2479, pruned_loss=0.03429, over 1373520.51 frames.], batch size: 19, lr: 4.35e-04 2022-05-27 22:08:55,706 INFO [train.py:823] (0/4) Epoch 39, batch 700, loss[loss=0.1752, simple_loss=0.2606, pruned_loss=0.04488, over 7210.00 frames.], tot_loss[loss=0.1591, simple_loss=0.2488, pruned_loss=0.03468, over 1383565.62 frames.], batch size: 24, lr: 4.35e-04 2022-05-27 22:09:34,839 INFO [train.py:823] (0/4) Epoch 39, batch 750, loss[loss=0.1544, simple_loss=0.2428, pruned_loss=0.03299, over 7367.00 frames.], tot_loss[loss=0.1581, simple_loss=0.2478, pruned_loss=0.03421, over 1390250.33 frames.], batch size: 20, lr: 4.34e-04 2022-05-27 22:10:14,105 INFO [train.py:823] (0/4) Epoch 39, batch 800, loss[loss=0.1665, simple_loss=0.2438, pruned_loss=0.04463, over 7178.00 frames.], tot_loss[loss=0.1587, simple_loss=0.2482, pruned_loss=0.03458, over 1398989.09 frames.], batch size: 18, lr: 4.34e-04 2022-05-27 22:10:52,836 INFO [train.py:823] (0/4) Epoch 39, batch 850, loss[loss=0.1624, simple_loss=0.2615, pruned_loss=0.03167, over 7332.00 frames.], tot_loss[loss=0.1592, simple_loss=0.2491, pruned_loss=0.03465, over 1398956.95 frames.], batch size: 23, lr: 4.34e-04 2022-05-27 22:11:31,872 INFO [train.py:823] (0/4) Epoch 39, batch 900, loss[loss=0.1706, simple_loss=0.2639, pruned_loss=0.03864, over 6910.00 frames.], tot_loss[loss=0.1602, simple_loss=0.2504, pruned_loss=0.03498, over 1391846.10 frames.], batch size: 29, lr: 4.34e-04 2022-05-27 22:12:11,432 INFO [train.py:823] (0/4) Epoch 39, batch 950, loss[loss=0.1622, simple_loss=0.2501, pruned_loss=0.03713, over 4951.00 frames.], tot_loss[loss=0.1606, simple_loss=0.2503, pruned_loss=0.03539, over 1365918.50 frames.], batch size: 47, lr: 4.33e-04 2022-05-27 22:12:12,651 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-39.pt 2022-05-27 22:12:24,465 INFO [train.py:823] (0/4) Epoch 40, batch 0, loss[loss=0.1717, simple_loss=0.2649, pruned_loss=0.03923, over 7124.00 frames.], tot_loss[loss=0.1717, simple_loss=0.2649, pruned_loss=0.03923, over 7124.00 frames.], batch size: 23, lr: 4.28e-04 2022-05-27 22:13:03,455 INFO [train.py:823] (0/4) Epoch 40, batch 50, loss[loss=0.1504, simple_loss=0.2451, pruned_loss=0.02781, over 7108.00 frames.], tot_loss[loss=0.1556, simple_loss=0.2456, pruned_loss=0.03285, over 318936.26 frames.], batch size: 20, lr: 4.28e-04 2022-05-27 22:13:43,942 INFO [train.py:823] (0/4) Epoch 40, batch 100, loss[loss=0.1498, simple_loss=0.2375, pruned_loss=0.03108, over 6791.00 frames.], tot_loss[loss=0.157, simple_loss=0.2466, pruned_loss=0.03364, over 559988.89 frames.], batch size: 15, lr: 4.27e-04 2022-05-27 22:14:23,282 INFO [train.py:823] (0/4) Epoch 40, batch 150, loss[loss=0.1553, simple_loss=0.244, pruned_loss=0.03328, over 6875.00 frames.], tot_loss[loss=0.156, simple_loss=0.2459, pruned_loss=0.03308, over 745706.84 frames.], batch size: 29, lr: 4.27e-04 2022-05-27 22:15:02,181 INFO [train.py:823] (0/4) Epoch 40, batch 200, loss[loss=0.1751, simple_loss=0.2657, pruned_loss=0.0423, over 7177.00 frames.], tot_loss[loss=0.1581, simple_loss=0.2477, pruned_loss=0.03428, over 896891.11 frames.], batch size: 21, lr: 4.27e-04 2022-05-27 22:15:42,863 INFO [train.py:823] (0/4) Epoch 40, batch 250, loss[loss=0.1641, simple_loss=0.2388, pruned_loss=0.04468, over 6859.00 frames.], tot_loss[loss=0.1582, simple_loss=0.2479, pruned_loss=0.03428, over 1013886.90 frames.], batch size: 15, lr: 4.26e-04 2022-05-27 22:16:22,013 INFO [train.py:823] (0/4) Epoch 40, batch 300, loss[loss=0.1591, simple_loss=0.2502, pruned_loss=0.03402, over 7380.00 frames.], tot_loss[loss=0.1578, simple_loss=0.2477, pruned_loss=0.03398, over 1104174.19 frames.], batch size: 20, lr: 4.26e-04 2022-05-27 22:17:01,411 INFO [train.py:823] (0/4) Epoch 40, batch 350, loss[loss=0.1677, simple_loss=0.2622, pruned_loss=0.03657, over 6491.00 frames.], tot_loss[loss=0.1576, simple_loss=0.2475, pruned_loss=0.03385, over 1177283.89 frames.], batch size: 34, lr: 4.26e-04 2022-05-27 22:17:42,247 INFO [train.py:823] (0/4) Epoch 40, batch 400, loss[loss=0.1332, simple_loss=0.2233, pruned_loss=0.02154, over 6986.00 frames.], tot_loss[loss=0.1569, simple_loss=0.2466, pruned_loss=0.03359, over 1236329.16 frames.], batch size: 16, lr: 4.26e-04 2022-05-27 22:18:21,400 INFO [train.py:823] (0/4) Epoch 40, batch 450, loss[loss=0.1501, simple_loss=0.2245, pruned_loss=0.03785, over 7226.00 frames.], tot_loss[loss=0.1568, simple_loss=0.2466, pruned_loss=0.03352, over 1276872.52 frames.], batch size: 16, lr: 4.25e-04 2022-05-27 22:19:00,407 INFO [train.py:823] (0/4) Epoch 40, batch 500, loss[loss=0.1467, simple_loss=0.2304, pruned_loss=0.03148, over 7374.00 frames.], tot_loss[loss=0.1572, simple_loss=0.2471, pruned_loss=0.03361, over 1309649.41 frames.], batch size: 20, lr: 4.25e-04 2022-05-27 22:19:39,952 INFO [train.py:823] (0/4) Epoch 40, batch 550, loss[loss=0.1688, simple_loss=0.273, pruned_loss=0.03235, over 7297.00 frames.], tot_loss[loss=0.1574, simple_loss=0.247, pruned_loss=0.03388, over 1337346.11 frames.], batch size: 22, lr: 4.25e-04 2022-05-27 22:20:18,674 INFO [train.py:823] (0/4) Epoch 40, batch 600, loss[loss=0.1663, simple_loss=0.2611, pruned_loss=0.03575, over 7306.00 frames.], tot_loss[loss=0.1578, simple_loss=0.2475, pruned_loss=0.03401, over 1356685.62 frames.], batch size: 22, lr: 4.24e-04 2022-05-27 22:20:58,182 INFO [train.py:823] (0/4) Epoch 40, batch 650, loss[loss=0.1433, simple_loss=0.2293, pruned_loss=0.02862, over 7202.00 frames.], tot_loss[loss=0.1578, simple_loss=0.2476, pruned_loss=0.03403, over 1365442.55 frames.], batch size: 19, lr: 4.24e-04 2022-05-27 22:21:37,011 INFO [train.py:823] (0/4) Epoch 40, batch 700, loss[loss=0.2028, simple_loss=0.3027, pruned_loss=0.05141, over 7197.00 frames.], tot_loss[loss=0.1586, simple_loss=0.2489, pruned_loss=0.03415, over 1377756.37 frames.], batch size: 20, lr: 4.24e-04 2022-05-27 22:22:16,277 INFO [train.py:823] (0/4) Epoch 40, batch 750, loss[loss=0.1722, simple_loss=0.2532, pruned_loss=0.0456, over 4809.00 frames.], tot_loss[loss=0.1588, simple_loss=0.249, pruned_loss=0.03431, over 1388436.92 frames.], batch size: 48, lr: 4.24e-04 2022-05-27 22:22:55,270 INFO [train.py:823] (0/4) Epoch 40, batch 800, loss[loss=0.1578, simple_loss=0.2466, pruned_loss=0.03453, over 7190.00 frames.], tot_loss[loss=0.1586, simple_loss=0.2489, pruned_loss=0.03415, over 1388847.88 frames.], batch size: 21, lr: 4.23e-04 2022-05-27 22:23:34,569 INFO [train.py:823] (0/4) Epoch 40, batch 850, loss[loss=0.166, simple_loss=0.2534, pruned_loss=0.03927, over 7179.00 frames.], tot_loss[loss=0.1585, simple_loss=0.2489, pruned_loss=0.03405, over 1397844.88 frames.], batch size: 22, lr: 4.23e-04 2022-05-27 22:24:12,886 INFO [train.py:823] (0/4) Epoch 40, batch 900, loss[loss=0.1374, simple_loss=0.2193, pruned_loss=0.02774, over 7373.00 frames.], tot_loss[loss=0.1578, simple_loss=0.2479, pruned_loss=0.03389, over 1390177.13 frames.], batch size: 20, lr: 4.23e-04 2022-05-27 22:24:52,039 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-40.pt 2022-05-27 22:25:04,145 INFO [train.py:823] (0/4) Epoch 41, batch 0, loss[loss=0.1407, simple_loss=0.2311, pruned_loss=0.02514, over 7108.00 frames.], tot_loss[loss=0.1407, simple_loss=0.2311, pruned_loss=0.02514, over 7108.00 frames.], batch size: 19, lr: 4.17e-04 2022-05-27 22:25:43,068 INFO [train.py:823] (0/4) Epoch 41, batch 50, loss[loss=0.1757, simple_loss=0.2595, pruned_loss=0.04595, over 7375.00 frames.], tot_loss[loss=0.1578, simple_loss=0.2479, pruned_loss=0.03389, over 321353.79 frames.], batch size: 20, lr: 4.17e-04 2022-05-27 22:26:22,265 INFO [train.py:823] (0/4) Epoch 41, batch 100, loss[loss=0.1308, simple_loss=0.2217, pruned_loss=0.01993, over 7095.00 frames.], tot_loss[loss=0.1592, simple_loss=0.2489, pruned_loss=0.03472, over 560363.24 frames.], batch size: 18, lr: 4.17e-04 2022-05-27 22:27:00,994 INFO [train.py:823] (0/4) Epoch 41, batch 150, loss[loss=0.1571, simple_loss=0.2494, pruned_loss=0.03241, over 7004.00 frames.], tot_loss[loss=0.1583, simple_loss=0.2485, pruned_loss=0.03408, over 752043.31 frames.], batch size: 26, lr: 4.17e-04 2022-05-27 22:27:40,404 INFO [train.py:823] (0/4) Epoch 41, batch 200, loss[loss=0.1607, simple_loss=0.2478, pruned_loss=0.03681, over 7388.00 frames.], tot_loss[loss=0.1597, simple_loss=0.2499, pruned_loss=0.0347, over 904693.63 frames.], batch size: 19, lr: 4.16e-04 2022-05-27 22:28:19,428 INFO [train.py:823] (0/4) Epoch 41, batch 250, loss[loss=0.1663, simple_loss=0.2533, pruned_loss=0.03964, over 7106.00 frames.], tot_loss[loss=0.158, simple_loss=0.2482, pruned_loss=0.03392, over 1015772.12 frames.], batch size: 19, lr: 4.16e-04 2022-05-27 22:28:58,365 INFO [train.py:823] (0/4) Epoch 41, batch 300, loss[loss=0.1503, simple_loss=0.2408, pruned_loss=0.02988, over 7384.00 frames.], tot_loss[loss=0.1582, simple_loss=0.2482, pruned_loss=0.03416, over 1106454.32 frames.], batch size: 20, lr: 4.16e-04 2022-05-27 22:29:36,947 INFO [train.py:823] (0/4) Epoch 41, batch 350, loss[loss=0.1683, simple_loss=0.2503, pruned_loss=0.04316, over 7171.00 frames.], tot_loss[loss=0.1587, simple_loss=0.2486, pruned_loss=0.03437, over 1174310.19 frames.], batch size: 22, lr: 4.15e-04 2022-05-27 22:30:15,648 INFO [train.py:823] (0/4) Epoch 41, batch 400, loss[loss=0.1777, simple_loss=0.2681, pruned_loss=0.04363, over 7175.00 frames.], tot_loss[loss=0.1585, simple_loss=0.2484, pruned_loss=0.03428, over 1221267.96 frames.], batch size: 23, lr: 4.15e-04 2022-05-27 22:30:54,578 INFO [train.py:823] (0/4) Epoch 41, batch 450, loss[loss=0.1281, simple_loss=0.2168, pruned_loss=0.01966, over 7096.00 frames.], tot_loss[loss=0.1585, simple_loss=0.2484, pruned_loss=0.03426, over 1263337.49 frames.], batch size: 18, lr: 4.15e-04 2022-05-27 22:31:34,134 INFO [train.py:823] (0/4) Epoch 41, batch 500, loss[loss=0.1595, simple_loss=0.2549, pruned_loss=0.03207, over 7302.00 frames.], tot_loss[loss=0.158, simple_loss=0.2485, pruned_loss=0.0338, over 1298713.71 frames.], batch size: 22, lr: 4.15e-04 2022-05-27 22:32:12,791 INFO [train.py:823] (0/4) Epoch 41, batch 550, loss[loss=0.1624, simple_loss=0.2482, pruned_loss=0.03835, over 7203.00 frames.], tot_loss[loss=0.1572, simple_loss=0.2477, pruned_loss=0.03342, over 1321892.44 frames.], batch size: 19, lr: 4.14e-04 2022-05-27 22:32:52,120 INFO [train.py:823] (0/4) Epoch 41, batch 600, loss[loss=0.204, simple_loss=0.2812, pruned_loss=0.06335, over 7180.00 frames.], tot_loss[loss=0.1564, simple_loss=0.2467, pruned_loss=0.03308, over 1339357.68 frames.], batch size: 21, lr: 4.14e-04 2022-05-27 22:33:31,328 INFO [train.py:823] (0/4) Epoch 41, batch 650, loss[loss=0.1609, simple_loss=0.2507, pruned_loss=0.03555, over 7192.00 frames.], tot_loss[loss=0.1559, simple_loss=0.2462, pruned_loss=0.03282, over 1358150.57 frames.], batch size: 21, lr: 4.14e-04 2022-05-27 22:34:10,419 INFO [train.py:823] (0/4) Epoch 41, batch 700, loss[loss=0.1388, simple_loss=0.231, pruned_loss=0.02327, over 6792.00 frames.], tot_loss[loss=0.1563, simple_loss=0.2464, pruned_loss=0.03308, over 1371297.74 frames.], batch size: 15, lr: 4.14e-04 2022-05-27 22:34:50,542 INFO [train.py:823] (0/4) Epoch 41, batch 750, loss[loss=0.1296, simple_loss=0.2139, pruned_loss=0.02268, over 7187.00 frames.], tot_loss[loss=0.156, simple_loss=0.2462, pruned_loss=0.03292, over 1379866.81 frames.], batch size: 18, lr: 4.13e-04 2022-05-27 22:35:29,536 INFO [train.py:823] (0/4) Epoch 41, batch 800, loss[loss=0.1558, simple_loss=0.2333, pruned_loss=0.03917, over 7297.00 frames.], tot_loss[loss=0.1566, simple_loss=0.2468, pruned_loss=0.03323, over 1381507.86 frames.], batch size: 17, lr: 4.13e-04 2022-05-27 22:36:09,506 INFO [train.py:823] (0/4) Epoch 41, batch 850, loss[loss=0.1547, simple_loss=0.2508, pruned_loss=0.0293, over 7291.00 frames.], tot_loss[loss=0.1563, simple_loss=0.2466, pruned_loss=0.03304, over 1393888.11 frames.], batch size: 19, lr: 4.13e-04 2022-05-27 22:36:49,002 INFO [train.py:823] (0/4) Epoch 41, batch 900, loss[loss=0.1429, simple_loss=0.2192, pruned_loss=0.03332, over 7316.00 frames.], tot_loss[loss=0.157, simple_loss=0.2467, pruned_loss=0.03364, over 1399540.93 frames.], batch size: 17, lr: 4.13e-04 2022-05-27 22:37:27,889 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-41.pt 2022-05-27 22:37:42,352 INFO [train.py:823] (0/4) Epoch 42, batch 0, loss[loss=0.1844, simple_loss=0.2795, pruned_loss=0.04466, over 7284.00 frames.], tot_loss[loss=0.1844, simple_loss=0.2795, pruned_loss=0.04466, over 7284.00 frames.], batch size: 21, lr: 4.07e-04 2022-05-27 22:38:21,693 INFO [train.py:823] (0/4) Epoch 42, batch 50, loss[loss=0.1651, simple_loss=0.2469, pruned_loss=0.04159, over 7394.00 frames.], tot_loss[loss=0.1562, simple_loss=0.2443, pruned_loss=0.034, over 323246.46 frames.], batch size: 19, lr: 4.07e-04 2022-05-27 22:39:02,301 INFO [train.py:823] (0/4) Epoch 42, batch 100, loss[loss=0.15, simple_loss=0.2262, pruned_loss=0.03689, over 6795.00 frames.], tot_loss[loss=0.1548, simple_loss=0.244, pruned_loss=0.03274, over 565981.69 frames.], batch size: 15, lr: 4.07e-04 2022-05-27 22:39:41,236 INFO [train.py:823] (0/4) Epoch 42, batch 150, loss[loss=0.1766, simple_loss=0.2594, pruned_loss=0.04685, over 7170.00 frames.], tot_loss[loss=0.1543, simple_loss=0.2436, pruned_loss=0.03254, over 755012.00 frames.], batch size: 22, lr: 4.07e-04 2022-05-27 22:40:22,040 INFO [train.py:823] (0/4) Epoch 42, batch 200, loss[loss=0.1634, simple_loss=0.2644, pruned_loss=0.03119, over 7183.00 frames.], tot_loss[loss=0.1552, simple_loss=0.2448, pruned_loss=0.03279, over 900917.02 frames.], batch size: 24, lr: 4.06e-04 2022-05-27 22:41:01,104 INFO [train.py:823] (0/4) Epoch 42, batch 250, loss[loss=0.1464, simple_loss=0.2191, pruned_loss=0.03684, over 7142.00 frames.], tot_loss[loss=0.1564, simple_loss=0.2463, pruned_loss=0.03324, over 1016916.08 frames.], batch size: 17, lr: 4.06e-04 2022-05-27 22:41:40,208 INFO [train.py:823] (0/4) Epoch 42, batch 300, loss[loss=0.1553, simple_loss=0.2587, pruned_loss=0.02594, over 7189.00 frames.], tot_loss[loss=0.157, simple_loss=0.2467, pruned_loss=0.03364, over 1101041.61 frames.], batch size: 21, lr: 4.06e-04 2022-05-27 22:42:18,668 INFO [train.py:823] (0/4) Epoch 42, batch 350, loss[loss=0.1416, simple_loss=0.2247, pruned_loss=0.02926, over 7171.00 frames.], tot_loss[loss=0.1568, simple_loss=0.2462, pruned_loss=0.03368, over 1167936.30 frames.], batch size: 17, lr: 4.06e-04 2022-05-27 22:42:57,698 INFO [train.py:823] (0/4) Epoch 42, batch 400, loss[loss=0.1356, simple_loss=0.2197, pruned_loss=0.02577, over 7287.00 frames.], tot_loss[loss=0.1568, simple_loss=0.2465, pruned_loss=0.0335, over 1217745.55 frames.], batch size: 17, lr: 4.05e-04 2022-05-27 22:43:36,610 INFO [train.py:823] (0/4) Epoch 42, batch 450, loss[loss=0.1787, simple_loss=0.2676, pruned_loss=0.04495, over 7237.00 frames.], tot_loss[loss=0.1587, simple_loss=0.2487, pruned_loss=0.03439, over 1267986.30 frames.], batch size: 25, lr: 4.05e-04 2022-05-27 22:44:16,174 INFO [train.py:823] (0/4) Epoch 42, batch 500, loss[loss=0.144, simple_loss=0.2196, pruned_loss=0.03423, over 7151.00 frames.], tot_loss[loss=0.1572, simple_loss=0.2467, pruned_loss=0.03384, over 1301995.06 frames.], batch size: 17, lr: 4.05e-04 2022-05-27 22:44:54,533 INFO [train.py:823] (0/4) Epoch 42, batch 550, loss[loss=0.1481, simple_loss=0.2388, pruned_loss=0.02865, over 7182.00 frames.], tot_loss[loss=0.157, simple_loss=0.2466, pruned_loss=0.03372, over 1321894.58 frames.], batch size: 18, lr: 4.05e-04 2022-05-27 22:45:33,820 INFO [train.py:823] (0/4) Epoch 42, batch 600, loss[loss=0.1444, simple_loss=0.2373, pruned_loss=0.02577, over 7190.00 frames.], tot_loss[loss=0.1574, simple_loss=0.2476, pruned_loss=0.03356, over 1343220.34 frames.], batch size: 20, lr: 4.04e-04 2022-05-27 22:46:12,674 INFO [train.py:823] (0/4) Epoch 42, batch 650, loss[loss=0.1623, simple_loss=0.2542, pruned_loss=0.03522, over 7165.00 frames.], tot_loss[loss=0.1578, simple_loss=0.248, pruned_loss=0.0338, over 1364515.32 frames.], batch size: 23, lr: 4.04e-04 2022-05-27 22:46:51,882 INFO [train.py:823] (0/4) Epoch 42, batch 700, loss[loss=0.1862, simple_loss=0.2839, pruned_loss=0.04427, over 7014.00 frames.], tot_loss[loss=0.1577, simple_loss=0.248, pruned_loss=0.03365, over 1371100.53 frames.], batch size: 29, lr: 4.04e-04 2022-05-27 22:47:31,147 INFO [train.py:823] (0/4) Epoch 42, batch 750, loss[loss=0.1627, simple_loss=0.2563, pruned_loss=0.0345, over 7377.00 frames.], tot_loss[loss=0.1574, simple_loss=0.2479, pruned_loss=0.03347, over 1385142.02 frames.], batch size: 21, lr: 4.04e-04 2022-05-27 22:48:10,574 INFO [train.py:823] (0/4) Epoch 42, batch 800, loss[loss=0.1604, simple_loss=0.2532, pruned_loss=0.03379, over 6573.00 frames.], tot_loss[loss=0.1574, simple_loss=0.248, pruned_loss=0.03342, over 1391991.95 frames.], batch size: 34, lr: 4.03e-04 2022-05-27 22:48:49,651 INFO [train.py:823] (0/4) Epoch 42, batch 850, loss[loss=0.1524, simple_loss=0.2345, pruned_loss=0.03515, over 7033.00 frames.], tot_loss[loss=0.1569, simple_loss=0.2473, pruned_loss=0.03328, over 1397955.22 frames.], batch size: 17, lr: 4.03e-04 2022-05-27 22:49:29,032 INFO [train.py:823] (0/4) Epoch 42, batch 900, loss[loss=0.1718, simple_loss=0.2559, pruned_loss=0.04388, over 4856.00 frames.], tot_loss[loss=0.1566, simple_loss=0.2469, pruned_loss=0.03314, over 1397417.77 frames.], batch size: 47, lr: 4.03e-04 2022-05-27 22:50:07,787 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-42.pt 2022-05-27 22:50:20,076 INFO [train.py:823] (0/4) Epoch 43, batch 0, loss[loss=0.1438, simple_loss=0.2385, pruned_loss=0.02454, over 7297.00 frames.], tot_loss[loss=0.1438, simple_loss=0.2385, pruned_loss=0.02454, over 7297.00 frames.], batch size: 19, lr: 3.98e-04 2022-05-27 22:50:59,517 INFO [train.py:823] (0/4) Epoch 43, batch 50, loss[loss=0.1497, simple_loss=0.2432, pruned_loss=0.02811, over 7375.00 frames.], tot_loss[loss=0.1543, simple_loss=0.2457, pruned_loss=0.03143, over 321901.53 frames.], batch size: 20, lr: 3.98e-04 2022-05-27 22:51:38,798 INFO [train.py:823] (0/4) Epoch 43, batch 100, loss[loss=0.176, simple_loss=0.2622, pruned_loss=0.04487, over 7149.00 frames.], tot_loss[loss=0.1534, simple_loss=0.2438, pruned_loss=0.03154, over 565716.87 frames.], batch size: 23, lr: 3.97e-04 2022-05-27 22:51:39,925 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/checkpoint-40000.pt 2022-05-27 22:52:22,698 INFO [train.py:823] (0/4) Epoch 43, batch 150, loss[loss=0.1432, simple_loss=0.2422, pruned_loss=0.02213, over 6578.00 frames.], tot_loss[loss=0.1553, simple_loss=0.2455, pruned_loss=0.03257, over 753548.29 frames.], batch size: 34, lr: 3.97e-04 2022-05-27 22:53:01,458 INFO [train.py:823] (0/4) Epoch 43, batch 200, loss[loss=0.1885, simple_loss=0.2739, pruned_loss=0.05153, over 7335.00 frames.], tot_loss[loss=0.1561, simple_loss=0.2463, pruned_loss=0.03296, over 905090.65 frames.], batch size: 23, lr: 3.97e-04 2022-05-27 22:53:40,828 INFO [train.py:823] (0/4) Epoch 43, batch 250, loss[loss=0.1399, simple_loss=0.2211, pruned_loss=0.02939, over 7314.00 frames.], tot_loss[loss=0.156, simple_loss=0.2461, pruned_loss=0.03294, over 1021880.98 frames.], batch size: 18, lr: 3.97e-04 2022-05-27 22:54:19,530 INFO [train.py:823] (0/4) Epoch 43, batch 300, loss[loss=0.1522, simple_loss=0.2385, pruned_loss=0.03296, over 7096.00 frames.], tot_loss[loss=0.1554, simple_loss=0.2453, pruned_loss=0.03278, over 1101874.69 frames.], batch size: 18, lr: 3.96e-04 2022-05-27 22:54:59,043 INFO [train.py:823] (0/4) Epoch 43, batch 350, loss[loss=0.1522, simple_loss=0.2465, pruned_loss=0.02894, over 7339.00 frames.], tot_loss[loss=0.1562, simple_loss=0.2461, pruned_loss=0.03314, over 1175233.88 frames.], batch size: 23, lr: 3.96e-04 2022-05-27 22:55:37,590 INFO [train.py:823] (0/4) Epoch 43, batch 400, loss[loss=0.1587, simple_loss=0.2542, pruned_loss=0.03165, over 7197.00 frames.], tot_loss[loss=0.1565, simple_loss=0.2464, pruned_loss=0.03335, over 1230577.76 frames.], batch size: 20, lr: 3.96e-04 2022-05-27 22:56:17,085 INFO [train.py:823] (0/4) Epoch 43, batch 450, loss[loss=0.1747, simple_loss=0.2611, pruned_loss=0.04416, over 7189.00 frames.], tot_loss[loss=0.1565, simple_loss=0.2463, pruned_loss=0.03337, over 1276892.97 frames.], batch size: 21, lr: 3.96e-04 2022-05-27 22:56:56,288 INFO [train.py:823] (0/4) Epoch 43, batch 500, loss[loss=0.1237, simple_loss=0.1986, pruned_loss=0.02442, over 7443.00 frames.], tot_loss[loss=0.1562, simple_loss=0.2461, pruned_loss=0.03313, over 1308828.09 frames.], batch size: 18, lr: 3.95e-04 2022-05-27 22:57:35,569 INFO [train.py:823] (0/4) Epoch 43, batch 550, loss[loss=0.1864, simple_loss=0.2797, pruned_loss=0.04652, over 7274.00 frames.], tot_loss[loss=0.1569, simple_loss=0.2473, pruned_loss=0.03326, over 1338344.39 frames.], batch size: 21, lr: 3.95e-04 2022-05-27 22:58:14,106 INFO [train.py:823] (0/4) Epoch 43, batch 600, loss[loss=0.1668, simple_loss=0.2616, pruned_loss=0.03595, over 7163.00 frames.], tot_loss[loss=0.1572, simple_loss=0.2477, pruned_loss=0.03333, over 1357557.30 frames.], batch size: 22, lr: 3.95e-04 2022-05-27 22:58:54,264 INFO [train.py:823] (0/4) Epoch 43, batch 650, loss[loss=0.1491, simple_loss=0.25, pruned_loss=0.02412, over 7203.00 frames.], tot_loss[loss=0.1565, simple_loss=0.2471, pruned_loss=0.03292, over 1374506.25 frames.], batch size: 20, lr: 3.95e-04 2022-05-27 22:59:34,016 INFO [train.py:823] (0/4) Epoch 43, batch 700, loss[loss=0.1291, simple_loss=0.2196, pruned_loss=0.01926, over 7424.00 frames.], tot_loss[loss=0.1555, simple_loss=0.2461, pruned_loss=0.03242, over 1384408.40 frames.], batch size: 18, lr: 3.94e-04 2022-05-27 23:00:13,135 INFO [train.py:823] (0/4) Epoch 43, batch 750, loss[loss=0.1577, simple_loss=0.2554, pruned_loss=0.03, over 7194.00 frames.], tot_loss[loss=0.1556, simple_loss=0.2462, pruned_loss=0.03245, over 1394304.15 frames.], batch size: 21, lr: 3.94e-04 2022-05-27 23:00:51,806 INFO [train.py:823] (0/4) Epoch 43, batch 800, loss[loss=0.1614, simple_loss=0.2576, pruned_loss=0.03253, over 7302.00 frames.], tot_loss[loss=0.1556, simple_loss=0.2459, pruned_loss=0.03268, over 1402924.71 frames.], batch size: 22, lr: 3.94e-04 2022-05-27 23:01:30,941 INFO [train.py:823] (0/4) Epoch 43, batch 850, loss[loss=0.1827, simple_loss=0.2778, pruned_loss=0.04383, over 7179.00 frames.], tot_loss[loss=0.1558, simple_loss=0.2459, pruned_loss=0.03283, over 1405254.16 frames.], batch size: 22, lr: 3.94e-04 2022-05-27 23:02:11,063 INFO [train.py:823] (0/4) Epoch 43, batch 900, loss[loss=0.1364, simple_loss=0.2172, pruned_loss=0.02783, over 6846.00 frames.], tot_loss[loss=0.1553, simple_loss=0.2452, pruned_loss=0.03268, over 1402777.37 frames.], batch size: 15, lr: 3.93e-04 2022-05-27 23:02:49,004 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-43.pt 2022-05-27 23:03:00,570 INFO [train.py:823] (0/4) Epoch 44, batch 0, loss[loss=0.1457, simple_loss=0.2382, pruned_loss=0.02656, over 7295.00 frames.], tot_loss[loss=0.1457, simple_loss=0.2382, pruned_loss=0.02656, over 7295.00 frames.], batch size: 22, lr: 3.89e-04 2022-05-27 23:03:41,208 INFO [train.py:823] (0/4) Epoch 44, batch 50, loss[loss=0.1436, simple_loss=0.2334, pruned_loss=0.02686, over 7023.00 frames.], tot_loss[loss=0.1531, simple_loss=0.2425, pruned_loss=0.03182, over 322105.05 frames.], batch size: 17, lr: 3.89e-04 2022-05-27 23:04:20,544 INFO [train.py:823] (0/4) Epoch 44, batch 100, loss[loss=0.1485, simple_loss=0.2442, pruned_loss=0.02635, over 7285.00 frames.], tot_loss[loss=0.1561, simple_loss=0.246, pruned_loss=0.03305, over 566839.97 frames.], batch size: 20, lr: 3.88e-04 2022-05-27 23:04:59,671 INFO [train.py:823] (0/4) Epoch 44, batch 150, loss[loss=0.1412, simple_loss=0.2352, pruned_loss=0.02354, over 7284.00 frames.], tot_loss[loss=0.1544, simple_loss=0.2441, pruned_loss=0.03237, over 757939.06 frames.], batch size: 20, lr: 3.88e-04 2022-05-27 23:05:38,963 INFO [train.py:823] (0/4) Epoch 44, batch 200, loss[loss=0.1596, simple_loss=0.2561, pruned_loss=0.03158, over 7244.00 frames.], tot_loss[loss=0.1546, simple_loss=0.2448, pruned_loss=0.03222, over 904131.59 frames.], batch size: 24, lr: 3.88e-04 2022-05-27 23:06:18,013 INFO [train.py:823] (0/4) Epoch 44, batch 250, loss[loss=0.152, simple_loss=0.2382, pruned_loss=0.0329, over 7133.00 frames.], tot_loss[loss=0.1544, simple_loss=0.2445, pruned_loss=0.03221, over 1021489.12 frames.], batch size: 23, lr: 3.88e-04 2022-05-27 23:06:56,884 INFO [train.py:823] (0/4) Epoch 44, batch 300, loss[loss=0.1562, simple_loss=0.2496, pruned_loss=0.03142, over 7288.00 frames.], tot_loss[loss=0.1556, simple_loss=0.2453, pruned_loss=0.03292, over 1108121.61 frames.], batch size: 21, lr: 3.87e-04 2022-05-27 23:07:35,756 INFO [train.py:823] (0/4) Epoch 44, batch 350, loss[loss=0.127, simple_loss=0.2028, pruned_loss=0.02556, over 7022.00 frames.], tot_loss[loss=0.1549, simple_loss=0.2449, pruned_loss=0.03242, over 1170955.88 frames.], batch size: 16, lr: 3.87e-04 2022-05-27 23:08:14,767 INFO [train.py:823] (0/4) Epoch 44, batch 400, loss[loss=0.1839, simple_loss=0.2616, pruned_loss=0.05307, over 5143.00 frames.], tot_loss[loss=0.1548, simple_loss=0.2448, pruned_loss=0.03239, over 1221155.96 frames.], batch size: 47, lr: 3.87e-04 2022-05-27 23:08:53,875 INFO [train.py:823] (0/4) Epoch 44, batch 450, loss[loss=0.1667, simple_loss=0.2637, pruned_loss=0.03482, over 7181.00 frames.], tot_loss[loss=0.1558, simple_loss=0.2461, pruned_loss=0.03281, over 1264806.40 frames.], batch size: 25, lr: 3.87e-04 2022-05-27 23:09:33,427 INFO [train.py:823] (0/4) Epoch 44, batch 500, loss[loss=0.1928, simple_loss=0.2827, pruned_loss=0.05147, over 7146.00 frames.], tot_loss[loss=0.1558, simple_loss=0.2458, pruned_loss=0.03287, over 1302052.08 frames.], batch size: 17, lr: 3.86e-04 2022-05-27 23:10:12,713 INFO [train.py:823] (0/4) Epoch 44, batch 550, loss[loss=0.1678, simple_loss=0.2575, pruned_loss=0.03903, over 7220.00 frames.], tot_loss[loss=0.1551, simple_loss=0.2459, pruned_loss=0.03214, over 1330235.52 frames.], batch size: 24, lr: 3.86e-04 2022-05-27 23:10:52,016 INFO [train.py:823] (0/4) Epoch 44, batch 600, loss[loss=0.1328, simple_loss=0.2217, pruned_loss=0.02197, over 7390.00 frames.], tot_loss[loss=0.1544, simple_loss=0.2452, pruned_loss=0.03173, over 1352655.45 frames.], batch size: 19, lr: 3.86e-04 2022-05-27 23:11:30,682 INFO [train.py:823] (0/4) Epoch 44, batch 650, loss[loss=0.1452, simple_loss=0.2478, pruned_loss=0.02128, over 7422.00 frames.], tot_loss[loss=0.1542, simple_loss=0.2449, pruned_loss=0.03178, over 1367288.87 frames.], batch size: 22, lr: 3.86e-04 2022-05-27 23:12:09,870 INFO [train.py:823] (0/4) Epoch 44, batch 700, loss[loss=0.1546, simple_loss=0.245, pruned_loss=0.03205, over 7148.00 frames.], tot_loss[loss=0.1545, simple_loss=0.2447, pruned_loss=0.03218, over 1379223.05 frames.], batch size: 23, lr: 3.85e-04 2022-05-27 23:12:48,549 INFO [train.py:823] (0/4) Epoch 44, batch 750, loss[loss=0.1605, simple_loss=0.2505, pruned_loss=0.03529, over 7136.00 frames.], tot_loss[loss=0.1554, simple_loss=0.2455, pruned_loss=0.03267, over 1390767.04 frames.], batch size: 17, lr: 3.85e-04 2022-05-27 23:13:27,414 INFO [train.py:823] (0/4) Epoch 44, batch 800, loss[loss=0.1877, simple_loss=0.2849, pruned_loss=0.04529, over 7202.00 frames.], tot_loss[loss=0.156, simple_loss=0.2462, pruned_loss=0.03293, over 1398016.34 frames.], batch size: 25, lr: 3.85e-04 2022-05-27 23:14:06,936 INFO [train.py:823] (0/4) Epoch 44, batch 850, loss[loss=0.1374, simple_loss=0.229, pruned_loss=0.02293, over 6816.00 frames.], tot_loss[loss=0.1558, simple_loss=0.2461, pruned_loss=0.03277, over 1403199.70 frames.], batch size: 15, lr: 3.85e-04 2022-05-27 23:14:46,123 INFO [train.py:823] (0/4) Epoch 44, batch 900, loss[loss=0.1322, simple_loss=0.2168, pruned_loss=0.02376, over 7281.00 frames.], tot_loss[loss=0.1558, simple_loss=0.246, pruned_loss=0.0328, over 1400957.21 frames.], batch size: 17, lr: 3.85e-04 2022-05-27 23:15:24,433 INFO [train.py:823] (0/4) Epoch 44, batch 950, loss[loss=0.15, simple_loss=0.2353, pruned_loss=0.0323, over 4513.00 frames.], tot_loss[loss=0.1563, simple_loss=0.2461, pruned_loss=0.03324, over 1377270.62 frames.], batch size: 47, lr: 3.84e-04 2022-05-27 23:15:25,611 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-44.pt 2022-05-27 23:15:37,740 INFO [train.py:823] (0/4) Epoch 45, batch 0, loss[loss=0.1562, simple_loss=0.2538, pruned_loss=0.02926, over 7271.00 frames.], tot_loss[loss=0.1562, simple_loss=0.2538, pruned_loss=0.02926, over 7271.00 frames.], batch size: 20, lr: 3.80e-04 2022-05-27 23:16:17,158 INFO [train.py:823] (0/4) Epoch 45, batch 50, loss[loss=0.1577, simple_loss=0.2626, pruned_loss=0.02636, over 7293.00 frames.], tot_loss[loss=0.1558, simple_loss=0.2464, pruned_loss=0.03257, over 323849.10 frames.], batch size: 21, lr: 3.80e-04 2022-05-27 23:16:56,291 INFO [train.py:823] (0/4) Epoch 45, batch 100, loss[loss=0.1887, simple_loss=0.2998, pruned_loss=0.03883, over 7373.00 frames.], tot_loss[loss=0.1562, simple_loss=0.2469, pruned_loss=0.03275, over 567744.56 frames.], batch size: 21, lr: 3.80e-04 2022-05-27 23:17:35,592 INFO [train.py:823] (0/4) Epoch 45, batch 150, loss[loss=0.1254, simple_loss=0.2095, pruned_loss=0.02067, over 7207.00 frames.], tot_loss[loss=0.155, simple_loss=0.2451, pruned_loss=0.03243, over 752216.81 frames.], batch size: 16, lr: 3.79e-04 2022-05-27 23:18:14,556 INFO [train.py:823] (0/4) Epoch 45, batch 200, loss[loss=0.1606, simple_loss=0.2444, pruned_loss=0.03844, over 4647.00 frames.], tot_loss[loss=0.1551, simple_loss=0.2452, pruned_loss=0.0325, over 896394.51 frames.], batch size: 46, lr: 3.79e-04 2022-05-27 23:18:53,797 INFO [train.py:823] (0/4) Epoch 45, batch 250, loss[loss=0.147, simple_loss=0.2423, pruned_loss=0.02583, over 6487.00 frames.], tot_loss[loss=0.1547, simple_loss=0.245, pruned_loss=0.03224, over 1009429.22 frames.], batch size: 34, lr: 3.79e-04 2022-05-27 23:19:32,688 INFO [train.py:823] (0/4) Epoch 45, batch 300, loss[loss=0.1594, simple_loss=0.241, pruned_loss=0.03888, over 7148.00 frames.], tot_loss[loss=0.1532, simple_loss=0.2431, pruned_loss=0.03163, over 1098910.06 frames.], batch size: 23, lr: 3.79e-04 2022-05-27 23:20:11,911 INFO [train.py:823] (0/4) Epoch 45, batch 350, loss[loss=0.16, simple_loss=0.2547, pruned_loss=0.03263, over 7418.00 frames.], tot_loss[loss=0.1542, simple_loss=0.244, pruned_loss=0.03216, over 1171370.75 frames.], batch size: 22, lr: 3.78e-04 2022-05-27 23:20:50,908 INFO [train.py:823] (0/4) Epoch 45, batch 400, loss[loss=0.1654, simple_loss=0.2555, pruned_loss=0.0376, over 7378.00 frames.], tot_loss[loss=0.1545, simple_loss=0.2443, pruned_loss=0.03232, over 1228676.52 frames.], batch size: 20, lr: 3.78e-04 2022-05-27 23:21:30,379 INFO [train.py:823] (0/4) Epoch 45, batch 450, loss[loss=0.1505, simple_loss=0.2365, pruned_loss=0.03224, over 7201.00 frames.], tot_loss[loss=0.1534, simple_loss=0.2434, pruned_loss=0.0317, over 1269799.76 frames.], batch size: 18, lr: 3.78e-04 2022-05-27 23:22:12,117 INFO [train.py:823] (0/4) Epoch 45, batch 500, loss[loss=0.1777, simple_loss=0.2726, pruned_loss=0.04142, over 7233.00 frames.], tot_loss[loss=0.154, simple_loss=0.2442, pruned_loss=0.03192, over 1308184.62 frames.], batch size: 24, lr: 3.78e-04 2022-05-27 23:22:51,883 INFO [train.py:823] (0/4) Epoch 45, batch 550, loss[loss=0.1206, simple_loss=0.2065, pruned_loss=0.01737, over 7195.00 frames.], tot_loss[loss=0.1524, simple_loss=0.2425, pruned_loss=0.03116, over 1334160.52 frames.], batch size: 18, lr: 3.78e-04 2022-05-27 23:23:31,255 INFO [train.py:823] (0/4) Epoch 45, batch 600, loss[loss=0.1606, simple_loss=0.255, pruned_loss=0.03308, over 6360.00 frames.], tot_loss[loss=0.1529, simple_loss=0.2433, pruned_loss=0.03129, over 1346956.15 frames.], batch size: 34, lr: 3.77e-04 2022-05-27 23:24:10,233 INFO [train.py:823] (0/4) Epoch 45, batch 650, loss[loss=0.1768, simple_loss=0.2732, pruned_loss=0.04022, over 7154.00 frames.], tot_loss[loss=0.1536, simple_loss=0.2437, pruned_loss=0.03169, over 1362094.04 frames.], batch size: 23, lr: 3.77e-04 2022-05-27 23:24:50,422 INFO [train.py:823] (0/4) Epoch 45, batch 700, loss[loss=0.1543, simple_loss=0.2489, pruned_loss=0.02986, over 7298.00 frames.], tot_loss[loss=0.1537, simple_loss=0.2439, pruned_loss=0.03173, over 1376186.33 frames.], batch size: 22, lr: 3.77e-04 2022-05-27 23:25:30,013 INFO [train.py:823] (0/4) Epoch 45, batch 750, loss[loss=0.1733, simple_loss=0.2671, pruned_loss=0.03975, over 6892.00 frames.], tot_loss[loss=0.1549, simple_loss=0.2453, pruned_loss=0.03226, over 1385771.33 frames.], batch size: 29, lr: 3.77e-04 2022-05-27 23:26:08,790 INFO [train.py:823] (0/4) Epoch 45, batch 800, loss[loss=0.1873, simple_loss=0.2835, pruned_loss=0.04554, over 7322.00 frames.], tot_loss[loss=0.1543, simple_loss=0.2448, pruned_loss=0.03188, over 1394755.45 frames.], batch size: 23, lr: 3.77e-04 2022-05-27 23:26:49,334 INFO [train.py:823] (0/4) Epoch 45, batch 850, loss[loss=0.1644, simple_loss=0.2576, pruned_loss=0.03564, over 7175.00 frames.], tot_loss[loss=0.155, simple_loss=0.2457, pruned_loss=0.03214, over 1397062.14 frames.], batch size: 21, lr: 3.76e-04 2022-05-27 23:27:28,222 INFO [train.py:823] (0/4) Epoch 45, batch 900, loss[loss=0.1305, simple_loss=0.2154, pruned_loss=0.02282, over 7023.00 frames.], tot_loss[loss=0.1552, simple_loss=0.2461, pruned_loss=0.03214, over 1399288.82 frames.], batch size: 17, lr: 3.76e-04 2022-05-27 23:28:08,042 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-45.pt 2022-05-27 23:28:22,763 INFO [train.py:823] (0/4) Epoch 46, batch 0, loss[loss=0.1603, simple_loss=0.2547, pruned_loss=0.03292, over 7166.00 frames.], tot_loss[loss=0.1603, simple_loss=0.2547, pruned_loss=0.03292, over 7166.00 frames.], batch size: 22, lr: 3.72e-04 2022-05-27 23:29:02,095 INFO [train.py:823] (0/4) Epoch 46, batch 50, loss[loss=0.1474, simple_loss=0.2422, pruned_loss=0.02634, over 7269.00 frames.], tot_loss[loss=0.1558, simple_loss=0.2459, pruned_loss=0.03285, over 315986.12 frames.], batch size: 20, lr: 3.72e-04 2022-05-27 23:29:41,267 INFO [train.py:823] (0/4) Epoch 46, batch 100, loss[loss=0.1324, simple_loss=0.2098, pruned_loss=0.02748, over 7011.00 frames.], tot_loss[loss=0.1532, simple_loss=0.2428, pruned_loss=0.03182, over 562500.83 frames.], batch size: 16, lr: 3.71e-04 2022-05-27 23:30:20,232 INFO [train.py:823] (0/4) Epoch 46, batch 150, loss[loss=0.1499, simple_loss=0.2509, pruned_loss=0.02449, over 7117.00 frames.], tot_loss[loss=0.1544, simple_loss=0.2441, pruned_loss=0.03236, over 754880.18 frames.], batch size: 20, lr: 3.71e-04 2022-05-27 23:30:59,849 INFO [train.py:823] (0/4) Epoch 46, batch 200, loss[loss=0.195, simple_loss=0.2788, pruned_loss=0.05563, over 7342.00 frames.], tot_loss[loss=0.1539, simple_loss=0.2436, pruned_loss=0.0321, over 907270.97 frames.], batch size: 23, lr: 3.71e-04 2022-05-27 23:31:39,088 INFO [train.py:823] (0/4) Epoch 46, batch 250, loss[loss=0.1643, simple_loss=0.2543, pruned_loss=0.03715, over 7172.00 frames.], tot_loss[loss=0.1547, simple_loss=0.2446, pruned_loss=0.03237, over 1021120.81 frames.], batch size: 23, lr: 3.71e-04 2022-05-27 23:32:18,143 INFO [train.py:823] (0/4) Epoch 46, batch 300, loss[loss=0.1676, simple_loss=0.2679, pruned_loss=0.03363, over 6894.00 frames.], tot_loss[loss=0.1551, simple_loss=0.2457, pruned_loss=0.03225, over 1107492.87 frames.], batch size: 29, lr: 3.70e-04 2022-05-27 23:32:56,766 INFO [train.py:823] (0/4) Epoch 46, batch 350, loss[loss=0.1551, simple_loss=0.2642, pruned_loss=0.023, over 6487.00 frames.], tot_loss[loss=0.1557, simple_loss=0.2467, pruned_loss=0.03242, over 1179539.16 frames.], batch size: 34, lr: 3.70e-04 2022-05-27 23:33:36,294 INFO [train.py:823] (0/4) Epoch 46, batch 400, loss[loss=0.1628, simple_loss=0.2646, pruned_loss=0.03049, over 7140.00 frames.], tot_loss[loss=0.1557, simple_loss=0.2467, pruned_loss=0.03232, over 1236144.05 frames.], batch size: 23, lr: 3.70e-04 2022-05-27 23:34:15,505 INFO [train.py:823] (0/4) Epoch 46, batch 450, loss[loss=0.1498, simple_loss=0.2428, pruned_loss=0.02839, over 7269.00 frames.], tot_loss[loss=0.156, simple_loss=0.2465, pruned_loss=0.03274, over 1279024.00 frames.], batch size: 20, lr: 3.70e-04 2022-05-27 23:34:54,741 INFO [train.py:823] (0/4) Epoch 46, batch 500, loss[loss=0.1444, simple_loss=0.2197, pruned_loss=0.03457, over 6805.00 frames.], tot_loss[loss=0.1558, simple_loss=0.2467, pruned_loss=0.03243, over 1303752.55 frames.], batch size: 15, lr: 3.70e-04 2022-05-27 23:35:33,974 INFO [train.py:823] (0/4) Epoch 46, batch 550, loss[loss=0.1763, simple_loss=0.2625, pruned_loss=0.04504, over 7317.00 frames.], tot_loss[loss=0.1556, simple_loss=0.2464, pruned_loss=0.03234, over 1333692.64 frames.], batch size: 22, lr: 3.69e-04 2022-05-27 23:36:13,164 INFO [train.py:823] (0/4) Epoch 46, batch 600, loss[loss=0.1235, simple_loss=0.2076, pruned_loss=0.01968, over 7433.00 frames.], tot_loss[loss=0.155, simple_loss=0.2461, pruned_loss=0.03194, over 1352113.16 frames.], batch size: 18, lr: 3.69e-04 2022-05-27 23:36:52,194 INFO [train.py:823] (0/4) Epoch 46, batch 650, loss[loss=0.1672, simple_loss=0.2595, pruned_loss=0.03738, over 7142.00 frames.], tot_loss[loss=0.1546, simple_loss=0.2453, pruned_loss=0.03192, over 1365948.65 frames.], batch size: 23, lr: 3.69e-04 2022-05-27 23:37:31,598 INFO [train.py:823] (0/4) Epoch 46, batch 700, loss[loss=0.15, simple_loss=0.2345, pruned_loss=0.0327, over 7163.00 frames.], tot_loss[loss=0.1542, simple_loss=0.2447, pruned_loss=0.03188, over 1374444.14 frames.], batch size: 17, lr: 3.69e-04 2022-05-27 23:38:10,485 INFO [train.py:823] (0/4) Epoch 46, batch 750, loss[loss=0.1531, simple_loss=0.2597, pruned_loss=0.0232, over 6501.00 frames.], tot_loss[loss=0.1539, simple_loss=0.2442, pruned_loss=0.03181, over 1383499.07 frames.], batch size: 34, lr: 3.69e-04 2022-05-27 23:38:49,154 INFO [train.py:823] (0/4) Epoch 46, batch 800, loss[loss=0.1611, simple_loss=0.2545, pruned_loss=0.0338, over 7196.00 frames.], tot_loss[loss=0.1544, simple_loss=0.2448, pruned_loss=0.03199, over 1386470.32 frames.], batch size: 20, lr: 3.68e-04 2022-05-27 23:39:28,295 INFO [train.py:823] (0/4) Epoch 46, batch 850, loss[loss=0.1548, simple_loss=0.2518, pruned_loss=0.02893, over 7354.00 frames.], tot_loss[loss=0.1541, simple_loss=0.2441, pruned_loss=0.03206, over 1389691.39 frames.], batch size: 23, lr: 3.68e-04 2022-05-27 23:40:07,672 INFO [train.py:823] (0/4) Epoch 46, batch 900, loss[loss=0.1529, simple_loss=0.247, pruned_loss=0.02941, over 7091.00 frames.], tot_loss[loss=0.1549, simple_loss=0.2448, pruned_loss=0.03248, over 1396323.31 frames.], batch size: 18, lr: 3.68e-04 2022-05-27 23:40:46,198 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-46.pt 2022-05-27 23:41:01,392 INFO [train.py:823] (0/4) Epoch 47, batch 0, loss[loss=0.1418, simple_loss=0.2313, pruned_loss=0.02614, over 7009.00 frames.], tot_loss[loss=0.1418, simple_loss=0.2313, pruned_loss=0.02614, over 7009.00 frames.], batch size: 16, lr: 3.64e-04 2022-05-27 23:41:40,305 INFO [train.py:823] (0/4) Epoch 47, batch 50, loss[loss=0.1505, simple_loss=0.2203, pruned_loss=0.0404, over 7311.00 frames.], tot_loss[loss=0.1511, simple_loss=0.2401, pruned_loss=0.03103, over 321726.93 frames.], batch size: 17, lr: 3.64e-04 2022-05-27 23:42:19,468 INFO [train.py:823] (0/4) Epoch 47, batch 100, loss[loss=0.1217, simple_loss=0.2073, pruned_loss=0.01805, over 7301.00 frames.], tot_loss[loss=0.1521, simple_loss=0.2411, pruned_loss=0.03154, over 565110.53 frames.], batch size: 18, lr: 3.63e-04 2022-05-27 23:42:58,952 INFO [train.py:823] (0/4) Epoch 47, batch 150, loss[loss=0.1632, simple_loss=0.2575, pruned_loss=0.03448, over 7287.00 frames.], tot_loss[loss=0.1533, simple_loss=0.2423, pruned_loss=0.03212, over 756881.20 frames.], batch size: 22, lr: 3.63e-04 2022-05-27 23:43:37,715 INFO [train.py:823] (0/4) Epoch 47, batch 200, loss[loss=0.1739, simple_loss=0.248, pruned_loss=0.04995, over 7098.00 frames.], tot_loss[loss=0.1536, simple_loss=0.2427, pruned_loss=0.03219, over 901050.54 frames.], batch size: 18, lr: 3.63e-04 2022-05-27 23:44:17,099 INFO [train.py:823] (0/4) Epoch 47, batch 250, loss[loss=0.1365, simple_loss=0.2269, pruned_loss=0.02309, over 7392.00 frames.], tot_loss[loss=0.1528, simple_loss=0.243, pruned_loss=0.03127, over 1022412.92 frames.], batch size: 19, lr: 3.63e-04 2022-05-27 23:44:56,261 INFO [train.py:823] (0/4) Epoch 47, batch 300, loss[loss=0.1507, simple_loss=0.2368, pruned_loss=0.03232, over 7198.00 frames.], tot_loss[loss=0.1538, simple_loss=0.2442, pruned_loss=0.03168, over 1111511.40 frames.], batch size: 18, lr: 3.63e-04 2022-05-27 23:45:37,069 INFO [train.py:823] (0/4) Epoch 47, batch 350, loss[loss=0.1573, simple_loss=0.2486, pruned_loss=0.03295, over 7272.00 frames.], tot_loss[loss=0.1532, simple_loss=0.2439, pruned_loss=0.03129, over 1179178.14 frames.], batch size: 20, lr: 3.62e-04 2022-05-27 23:46:17,196 INFO [train.py:823] (0/4) Epoch 47, batch 400, loss[loss=0.1582, simple_loss=0.2399, pruned_loss=0.03821, over 7289.00 frames.], tot_loss[loss=0.1531, simple_loss=0.2439, pruned_loss=0.0312, over 1233121.41 frames.], batch size: 20, lr: 3.62e-04 2022-05-27 23:46:56,141 INFO [train.py:823] (0/4) Epoch 47, batch 450, loss[loss=0.13, simple_loss=0.2148, pruned_loss=0.0226, over 7154.00 frames.], tot_loss[loss=0.1533, simple_loss=0.2441, pruned_loss=0.03128, over 1273718.03 frames.], batch size: 17, lr: 3.62e-04 2022-05-27 23:47:36,406 INFO [train.py:823] (0/4) Epoch 47, batch 500, loss[loss=0.1589, simple_loss=0.2537, pruned_loss=0.03202, over 7100.00 frames.], tot_loss[loss=0.1539, simple_loss=0.2446, pruned_loss=0.03155, over 1302667.36 frames.], batch size: 19, lr: 3.62e-04 2022-05-27 23:48:15,807 INFO [train.py:823] (0/4) Epoch 47, batch 550, loss[loss=0.1429, simple_loss=0.2337, pruned_loss=0.02604, over 7390.00 frames.], tot_loss[loss=0.1531, simple_loss=0.2436, pruned_loss=0.03133, over 1328655.86 frames.], batch size: 19, lr: 3.62e-04 2022-05-27 23:48:54,638 INFO [train.py:823] (0/4) Epoch 47, batch 600, loss[loss=0.1559, simple_loss=0.244, pruned_loss=0.03389, over 6996.00 frames.], tot_loss[loss=0.1532, simple_loss=0.2436, pruned_loss=0.03138, over 1347842.55 frames.], batch size: 26, lr: 3.61e-04 2022-05-27 23:49:34,979 INFO [train.py:823] (0/4) Epoch 47, batch 650, loss[loss=0.1367, simple_loss=0.2124, pruned_loss=0.03052, over 7281.00 frames.], tot_loss[loss=0.1536, simple_loss=0.2442, pruned_loss=0.03146, over 1364788.15 frames.], batch size: 17, lr: 3.61e-04 2022-05-27 23:50:14,039 INFO [train.py:823] (0/4) Epoch 47, batch 700, loss[loss=0.1893, simple_loss=0.2797, pruned_loss=0.04951, over 7322.00 frames.], tot_loss[loss=0.1535, simple_loss=0.2441, pruned_loss=0.03148, over 1372043.92 frames.], batch size: 23, lr: 3.61e-04 2022-05-27 23:50:53,644 INFO [train.py:823] (0/4) Epoch 47, batch 750, loss[loss=0.1468, simple_loss=0.2312, pruned_loss=0.03123, over 7280.00 frames.], tot_loss[loss=0.1536, simple_loss=0.2439, pruned_loss=0.03168, over 1384108.75 frames.], batch size: 19, lr: 3.61e-04 2022-05-27 23:51:32,289 INFO [train.py:823] (0/4) Epoch 47, batch 800, loss[loss=0.1549, simple_loss=0.254, pruned_loss=0.02788, over 7036.00 frames.], tot_loss[loss=0.1529, simple_loss=0.2433, pruned_loss=0.03127, over 1390691.27 frames.], batch size: 26, lr: 3.61e-04 2022-05-27 23:52:11,476 INFO [train.py:823] (0/4) Epoch 47, batch 850, loss[loss=0.1499, simple_loss=0.2435, pruned_loss=0.02812, over 7193.00 frames.], tot_loss[loss=0.1533, simple_loss=0.2437, pruned_loss=0.03148, over 1392890.41 frames.], batch size: 18, lr: 3.60e-04 2022-05-27 23:52:50,576 INFO [train.py:823] (0/4) Epoch 47, batch 900, loss[loss=0.1407, simple_loss=0.2411, pruned_loss=0.02015, over 7292.00 frames.], tot_loss[loss=0.1533, simple_loss=0.244, pruned_loss=0.03125, over 1398562.43 frames.], batch size: 22, lr: 3.60e-04 2022-05-27 23:53:28,851 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-47.pt 2022-05-27 23:53:43,372 INFO [train.py:823] (0/4) Epoch 48, batch 0, loss[loss=0.165, simple_loss=0.2595, pruned_loss=0.03528, over 7195.00 frames.], tot_loss[loss=0.165, simple_loss=0.2595, pruned_loss=0.03528, over 7195.00 frames.], batch size: 21, lr: 3.56e-04 2022-05-27 23:54:22,613 INFO [train.py:823] (0/4) Epoch 48, batch 50, loss[loss=0.1542, simple_loss=0.244, pruned_loss=0.03216, over 7159.00 frames.], tot_loss[loss=0.1544, simple_loss=0.2439, pruned_loss=0.03247, over 320173.00 frames.], batch size: 17, lr: 3.56e-04 2022-05-27 23:55:01,983 INFO [train.py:823] (0/4) Epoch 48, batch 100, loss[loss=0.1577, simple_loss=0.2497, pruned_loss=0.03288, over 7196.00 frames.], tot_loss[loss=0.1536, simple_loss=0.2427, pruned_loss=0.03222, over 564495.67 frames.], batch size: 25, lr: 3.56e-04 2022-05-27 23:55:41,028 INFO [train.py:823] (0/4) Epoch 48, batch 150, loss[loss=0.1361, simple_loss=0.2231, pruned_loss=0.02458, over 7305.00 frames.], tot_loss[loss=0.1536, simple_loss=0.2439, pruned_loss=0.03162, over 759341.20 frames.], batch size: 17, lr: 3.56e-04 2022-05-27 23:56:20,137 INFO [train.py:823] (0/4) Epoch 48, batch 200, loss[loss=0.1892, simple_loss=0.2883, pruned_loss=0.04502, over 7291.00 frames.], tot_loss[loss=0.1534, simple_loss=0.2437, pruned_loss=0.03157, over 907416.47 frames.], batch size: 22, lr: 3.55e-04 2022-05-27 23:56:59,338 INFO [train.py:823] (0/4) Epoch 48, batch 250, loss[loss=0.1617, simple_loss=0.2538, pruned_loss=0.03483, over 7199.00 frames.], tot_loss[loss=0.1529, simple_loss=0.2427, pruned_loss=0.03159, over 1023620.85 frames.], batch size: 19, lr: 3.55e-04 2022-05-27 23:57:38,896 INFO [train.py:823] (0/4) Epoch 48, batch 300, loss[loss=0.1664, simple_loss=0.2566, pruned_loss=0.03808, over 7019.00 frames.], tot_loss[loss=0.1529, simple_loss=0.2425, pruned_loss=0.03168, over 1115623.53 frames.], batch size: 26, lr: 3.55e-04 2022-05-27 23:58:17,890 INFO [train.py:823] (0/4) Epoch 48, batch 350, loss[loss=0.1641, simple_loss=0.2431, pruned_loss=0.0426, over 4911.00 frames.], tot_loss[loss=0.1532, simple_loss=0.2425, pruned_loss=0.03192, over 1182163.82 frames.], batch size: 46, lr: 3.55e-04 2022-05-27 23:58:57,342 INFO [train.py:823] (0/4) Epoch 48, batch 400, loss[loss=0.1523, simple_loss=0.2527, pruned_loss=0.0259, over 6644.00 frames.], tot_loss[loss=0.1534, simple_loss=0.2425, pruned_loss=0.03215, over 1237222.39 frames.], batch size: 34, lr: 3.55e-04 2022-05-27 23:59:36,496 INFO [train.py:823] (0/4) Epoch 48, batch 450, loss[loss=0.1487, simple_loss=0.2295, pruned_loss=0.034, over 7307.00 frames.], tot_loss[loss=0.154, simple_loss=0.2434, pruned_loss=0.03231, over 1280069.58 frames.], batch size: 17, lr: 3.54e-04 2022-05-28 00:00:15,794 INFO [train.py:823] (0/4) Epoch 48, batch 500, loss[loss=0.1513, simple_loss=0.2411, pruned_loss=0.03081, over 7198.00 frames.], tot_loss[loss=0.1544, simple_loss=0.2445, pruned_loss=0.03218, over 1310316.33 frames.], batch size: 20, lr: 3.54e-04 2022-05-28 00:00:54,491 INFO [train.py:823] (0/4) Epoch 48, batch 550, loss[loss=0.1486, simple_loss=0.2462, pruned_loss=0.0255, over 7425.00 frames.], tot_loss[loss=0.1541, simple_loss=0.2443, pruned_loss=0.03198, over 1329993.25 frames.], batch size: 22, lr: 3.54e-04 2022-05-28 00:01:33,244 INFO [train.py:823] (0/4) Epoch 48, batch 600, loss[loss=0.1706, simple_loss=0.2658, pruned_loss=0.03771, over 7283.00 frames.], tot_loss[loss=0.1545, simple_loss=0.2451, pruned_loss=0.03199, over 1349384.58 frames.], batch size: 20, lr: 3.54e-04 2022-05-28 00:02:11,702 INFO [train.py:823] (0/4) Epoch 48, batch 650, loss[loss=0.1534, simple_loss=0.2466, pruned_loss=0.03004, over 7379.00 frames.], tot_loss[loss=0.1546, simple_loss=0.2453, pruned_loss=0.03195, over 1363512.20 frames.], batch size: 21, lr: 3.54e-04 2022-05-28 00:02:51,183 INFO [train.py:823] (0/4) Epoch 48, batch 700, loss[loss=0.1675, simple_loss=0.2608, pruned_loss=0.03707, over 7166.00 frames.], tot_loss[loss=0.1539, simple_loss=0.245, pruned_loss=0.03136, over 1370530.19 frames.], batch size: 22, lr: 3.53e-04 2022-05-28 00:03:30,259 INFO [train.py:823] (0/4) Epoch 48, batch 750, loss[loss=0.1454, simple_loss=0.2375, pruned_loss=0.0267, over 7094.00 frames.], tot_loss[loss=0.1537, simple_loss=0.2445, pruned_loss=0.03139, over 1383833.58 frames.], batch size: 19, lr: 3.53e-04 2022-05-28 00:04:09,524 INFO [train.py:823] (0/4) Epoch 48, batch 800, loss[loss=0.1767, simple_loss=0.2692, pruned_loss=0.04213, over 7328.00 frames.], tot_loss[loss=0.1535, simple_loss=0.244, pruned_loss=0.03152, over 1391569.18 frames.], batch size: 23, lr: 3.53e-04 2022-05-28 00:04:48,525 INFO [train.py:823] (0/4) Epoch 48, batch 850, loss[loss=0.1341, simple_loss=0.2182, pruned_loss=0.02498, over 7287.00 frames.], tot_loss[loss=0.1534, simple_loss=0.2437, pruned_loss=0.0315, over 1392636.42 frames.], batch size: 17, lr: 3.53e-04 2022-05-28 00:05:27,346 INFO [train.py:823] (0/4) Epoch 48, batch 900, loss[loss=0.1628, simple_loss=0.2517, pruned_loss=0.03696, over 7310.00 frames.], tot_loss[loss=0.1528, simple_loss=0.2435, pruned_loss=0.03107, over 1394687.01 frames.], batch size: 19, lr: 3.53e-04 2022-05-28 00:06:06,446 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-48.pt 2022-05-28 00:06:18,123 INFO [train.py:823] (0/4) Epoch 49, batch 0, loss[loss=0.1471, simple_loss=0.2308, pruned_loss=0.03164, over 7371.00 frames.], tot_loss[loss=0.1471, simple_loss=0.2308, pruned_loss=0.03164, over 7371.00 frames.], batch size: 20, lr: 3.49e-04 2022-05-28 00:06:57,231 INFO [train.py:823] (0/4) Epoch 49, batch 50, loss[loss=0.1453, simple_loss=0.2443, pruned_loss=0.02319, over 7281.00 frames.], tot_loss[loss=0.1532, simple_loss=0.244, pruned_loss=0.0312, over 318102.18 frames.], batch size: 21, lr: 3.49e-04 2022-05-28 00:07:37,569 INFO [train.py:823] (0/4) Epoch 49, batch 100, loss[loss=0.1331, simple_loss=0.2252, pruned_loss=0.02054, over 7194.00 frames.], tot_loss[loss=0.1547, simple_loss=0.2454, pruned_loss=0.03205, over 560277.15 frames.], batch size: 18, lr: 3.48e-04 2022-05-28 00:08:16,664 INFO [train.py:823] (0/4) Epoch 49, batch 150, loss[loss=0.1539, simple_loss=0.2487, pruned_loss=0.02955, over 5186.00 frames.], tot_loss[loss=0.1524, simple_loss=0.243, pruned_loss=0.03091, over 751106.87 frames.], batch size: 46, lr: 3.48e-04 2022-05-28 00:08:56,073 INFO [train.py:823] (0/4) Epoch 49, batch 200, loss[loss=0.156, simple_loss=0.2487, pruned_loss=0.03165, over 7160.00 frames.], tot_loss[loss=0.1523, simple_loss=0.2431, pruned_loss=0.03074, over 901609.76 frames.], batch size: 23, lr: 3.48e-04 2022-05-28 00:09:37,922 INFO [train.py:823] (0/4) Epoch 49, batch 250, loss[loss=0.1678, simple_loss=0.2702, pruned_loss=0.03272, over 7187.00 frames.], tot_loss[loss=0.1534, simple_loss=0.2443, pruned_loss=0.0312, over 1021960.72 frames.], batch size: 20, lr: 3.48e-04 2022-05-28 00:10:16,994 INFO [train.py:823] (0/4) Epoch 49, batch 300, loss[loss=0.1364, simple_loss=0.22, pruned_loss=0.02636, over 7313.00 frames.], tot_loss[loss=0.1522, simple_loss=0.2432, pruned_loss=0.03062, over 1114446.61 frames.], batch size: 18, lr: 3.48e-04 2022-05-28 00:10:56,079 INFO [train.py:823] (0/4) Epoch 49, batch 350, loss[loss=0.1461, simple_loss=0.2463, pruned_loss=0.02296, over 7189.00 frames.], tot_loss[loss=0.1521, simple_loss=0.2431, pruned_loss=0.03052, over 1177291.43 frames.], batch size: 25, lr: 3.48e-04 2022-05-28 00:11:35,210 INFO [train.py:823] (0/4) Epoch 49, batch 400, loss[loss=0.1304, simple_loss=0.2186, pruned_loss=0.02107, over 7014.00 frames.], tot_loss[loss=0.1527, simple_loss=0.2436, pruned_loss=0.03089, over 1228015.69 frames.], batch size: 16, lr: 3.47e-04 2022-05-28 00:12:14,617 INFO [train.py:823] (0/4) Epoch 49, batch 450, loss[loss=0.1694, simple_loss=0.2565, pruned_loss=0.04118, over 7249.00 frames.], tot_loss[loss=0.1527, simple_loss=0.2434, pruned_loss=0.03105, over 1273180.16 frames.], batch size: 24, lr: 3.47e-04 2022-05-28 00:12:54,453 INFO [train.py:823] (0/4) Epoch 49, batch 500, loss[loss=0.1506, simple_loss=0.2413, pruned_loss=0.02997, over 6412.00 frames.], tot_loss[loss=0.1524, simple_loss=0.2429, pruned_loss=0.0309, over 1304634.03 frames.], batch size: 34, lr: 3.47e-04 2022-05-28 00:13:33,855 INFO [train.py:823] (0/4) Epoch 49, batch 550, loss[loss=0.1583, simple_loss=0.2425, pruned_loss=0.0371, over 7315.00 frames.], tot_loss[loss=0.1517, simple_loss=0.2417, pruned_loss=0.0309, over 1331619.58 frames.], batch size: 17, lr: 3.47e-04 2022-05-28 00:14:12,722 INFO [train.py:823] (0/4) Epoch 49, batch 600, loss[loss=0.1664, simple_loss=0.2595, pruned_loss=0.03665, over 7228.00 frames.], tot_loss[loss=0.1524, simple_loss=0.2423, pruned_loss=0.03126, over 1351105.31 frames.], batch size: 24, lr: 3.47e-04 2022-05-28 00:14:52,409 INFO [train.py:823] (0/4) Epoch 49, batch 650, loss[loss=0.1338, simple_loss=0.2124, pruned_loss=0.02766, over 7160.00 frames.], tot_loss[loss=0.1525, simple_loss=0.2423, pruned_loss=0.03134, over 1366843.44 frames.], batch size: 17, lr: 3.46e-04 2022-05-28 00:15:31,430 INFO [train.py:823] (0/4) Epoch 49, batch 700, loss[loss=0.1706, simple_loss=0.2582, pruned_loss=0.04151, over 7406.00 frames.], tot_loss[loss=0.1535, simple_loss=0.2435, pruned_loss=0.03179, over 1369828.60 frames.], batch size: 22, lr: 3.46e-04 2022-05-28 00:16:10,993 INFO [train.py:823] (0/4) Epoch 49, batch 750, loss[loss=0.1593, simple_loss=0.2462, pruned_loss=0.0362, over 7288.00 frames.], tot_loss[loss=0.1537, simple_loss=0.2441, pruned_loss=0.03167, over 1381073.50 frames.], batch size: 19, lr: 3.46e-04 2022-05-28 00:16:49,954 INFO [train.py:823] (0/4) Epoch 49, batch 800, loss[loss=0.1419, simple_loss=0.2272, pruned_loss=0.02829, over 7157.00 frames.], tot_loss[loss=0.1542, simple_loss=0.2445, pruned_loss=0.03196, over 1384229.71 frames.], batch size: 17, lr: 3.46e-04 2022-05-28 00:17:29,757 INFO [train.py:823] (0/4) Epoch 49, batch 850, loss[loss=0.1194, simple_loss=0.2059, pruned_loss=0.01644, over 7096.00 frames.], tot_loss[loss=0.1532, simple_loss=0.2434, pruned_loss=0.03146, over 1390722.24 frames.], batch size: 18, lr: 3.46e-04 2022-05-28 00:18:08,748 INFO [train.py:823] (0/4) Epoch 49, batch 900, loss[loss=0.1503, simple_loss=0.2476, pruned_loss=0.02649, over 6506.00 frames.], tot_loss[loss=0.1529, simple_loss=0.2435, pruned_loss=0.03117, over 1394224.16 frames.], batch size: 34, lr: 3.45e-04 2022-05-28 00:18:48,717 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-49.pt 2022-05-28 00:19:00,802 INFO [train.py:823] (0/4) Epoch 50, batch 0, loss[loss=0.1811, simple_loss=0.2606, pruned_loss=0.05079, over 6911.00 frames.], tot_loss[loss=0.1811, simple_loss=0.2606, pruned_loss=0.05079, over 6911.00 frames.], batch size: 29, lr: 3.42e-04 2022-05-28 00:19:39,997 INFO [train.py:823] (0/4) Epoch 50, batch 50, loss[loss=0.1501, simple_loss=0.249, pruned_loss=0.02563, over 7279.00 frames.], tot_loss[loss=0.1521, simple_loss=0.2402, pruned_loss=0.03199, over 322045.65 frames.], batch size: 20, lr: 3.42e-04 2022-05-28 00:20:19,160 INFO [train.py:823] (0/4) Epoch 50, batch 100, loss[loss=0.1688, simple_loss=0.2541, pruned_loss=0.04175, over 7148.00 frames.], tot_loss[loss=0.1514, simple_loss=0.2404, pruned_loss=0.03122, over 564000.75 frames.], batch size: 23, lr: 3.41e-04 2022-05-28 00:20:58,269 INFO [train.py:823] (0/4) Epoch 50, batch 150, loss[loss=0.1702, simple_loss=0.2639, pruned_loss=0.03823, over 7383.00 frames.], tot_loss[loss=0.1535, simple_loss=0.243, pruned_loss=0.03201, over 753075.97 frames.], batch size: 21, lr: 3.41e-04 2022-05-28 00:21:37,477 INFO [train.py:823] (0/4) Epoch 50, batch 200, loss[loss=0.1377, simple_loss=0.2211, pruned_loss=0.02712, over 7099.00 frames.], tot_loss[loss=0.1531, simple_loss=0.2431, pruned_loss=0.0316, over 902266.19 frames.], batch size: 18, lr: 3.41e-04 2022-05-28 00:22:16,558 INFO [train.py:823] (0/4) Epoch 50, batch 250, loss[loss=0.1661, simple_loss=0.2535, pruned_loss=0.03933, over 7157.00 frames.], tot_loss[loss=0.1527, simple_loss=0.2427, pruned_loss=0.03133, over 1019600.62 frames.], batch size: 22, lr: 3.41e-04 2022-05-28 00:22:55,490 INFO [train.py:823] (0/4) Epoch 50, batch 300, loss[loss=0.1513, simple_loss=0.245, pruned_loss=0.02878, over 7201.00 frames.], tot_loss[loss=0.1525, simple_loss=0.243, pruned_loss=0.03095, over 1109959.62 frames.], batch size: 20, lr: 3.41e-04 2022-05-28 00:23:34,711 INFO [train.py:823] (0/4) Epoch 50, batch 350, loss[loss=0.1906, simple_loss=0.2782, pruned_loss=0.05149, over 7421.00 frames.], tot_loss[loss=0.1525, simple_loss=0.2429, pruned_loss=0.03106, over 1178242.87 frames.], batch size: 22, lr: 3.41e-04 2022-05-28 00:24:13,715 INFO [train.py:823] (0/4) Epoch 50, batch 400, loss[loss=0.1646, simple_loss=0.2531, pruned_loss=0.03808, over 6992.00 frames.], tot_loss[loss=0.1521, simple_loss=0.2425, pruned_loss=0.03088, over 1232439.76 frames.], batch size: 26, lr: 3.40e-04 2022-05-28 00:24:52,165 INFO [train.py:823] (0/4) Epoch 50, batch 450, loss[loss=0.1429, simple_loss=0.2336, pruned_loss=0.0261, over 6375.00 frames.], tot_loss[loss=0.1526, simple_loss=0.2428, pruned_loss=0.03117, over 1272771.14 frames.], batch size: 34, lr: 3.40e-04 2022-05-28 00:25:31,603 INFO [train.py:823] (0/4) Epoch 50, batch 500, loss[loss=0.142, simple_loss=0.2323, pruned_loss=0.02585, over 7302.00 frames.], tot_loss[loss=0.153, simple_loss=0.2435, pruned_loss=0.03121, over 1306073.45 frames.], batch size: 19, lr: 3.40e-04 2022-05-28 00:26:10,808 INFO [train.py:823] (0/4) Epoch 50, batch 550, loss[loss=0.1593, simple_loss=0.2496, pruned_loss=0.03449, over 7227.00 frames.], tot_loss[loss=0.1529, simple_loss=0.2435, pruned_loss=0.03113, over 1334295.84 frames.], batch size: 24, lr: 3.40e-04 2022-05-28 00:26:49,676 INFO [train.py:823] (0/4) Epoch 50, batch 600, loss[loss=0.1444, simple_loss=0.2294, pruned_loss=0.02966, over 6985.00 frames.], tot_loss[loss=0.1531, simple_loss=0.2436, pruned_loss=0.03127, over 1353157.84 frames.], batch size: 16, lr: 3.40e-04 2022-05-28 00:27:28,492 INFO [train.py:823] (0/4) Epoch 50, batch 650, loss[loss=0.1354, simple_loss=0.2274, pruned_loss=0.02169, over 7444.00 frames.], tot_loss[loss=0.1536, simple_loss=0.2446, pruned_loss=0.03136, over 1364313.78 frames.], batch size: 17, lr: 3.39e-04 2022-05-28 00:28:07,666 INFO [train.py:823] (0/4) Epoch 50, batch 700, loss[loss=0.1205, simple_loss=0.2064, pruned_loss=0.01726, over 7026.00 frames.], tot_loss[loss=0.1523, simple_loss=0.2431, pruned_loss=0.03077, over 1376534.20 frames.], batch size: 16, lr: 3.39e-04 2022-05-28 00:28:46,794 INFO [train.py:823] (0/4) Epoch 50, batch 750, loss[loss=0.1421, simple_loss=0.2423, pruned_loss=0.02099, over 7313.00 frames.], tot_loss[loss=0.1527, simple_loss=0.2437, pruned_loss=0.03085, over 1383040.10 frames.], batch size: 22, lr: 3.39e-04 2022-05-28 00:29:26,154 INFO [train.py:823] (0/4) Epoch 50, batch 800, loss[loss=0.1393, simple_loss=0.2309, pruned_loss=0.02385, over 7099.00 frames.], tot_loss[loss=0.1519, simple_loss=0.2423, pruned_loss=0.0308, over 1390710.21 frames.], batch size: 19, lr: 3.39e-04 2022-05-28 00:30:05,641 INFO [train.py:823] (0/4) Epoch 50, batch 850, loss[loss=0.1534, simple_loss=0.2456, pruned_loss=0.03055, over 4569.00 frames.], tot_loss[loss=0.1525, simple_loss=0.2427, pruned_loss=0.03114, over 1396382.68 frames.], batch size: 46, lr: 3.39e-04 2022-05-28 00:30:45,790 INFO [train.py:823] (0/4) Epoch 50, batch 900, loss[loss=0.1619, simple_loss=0.2613, pruned_loss=0.03128, over 6435.00 frames.], tot_loss[loss=0.1527, simple_loss=0.2432, pruned_loss=0.03111, over 1398511.42 frames.], batch size: 34, lr: 3.39e-04 2022-05-28 00:31:24,702 INFO [checkpoint.py:75] (0/4) Saving checkpoint to pruned_transducer_stateless6/exp/epoch-50.pt 2022-05-28 00:31:31,224 INFO [train.py:1038] (0/4) Done!