0%|▊ | 100/27918 [03:49<3:54:30, 1.98it/s] 1%|█▋ | 200/27918 [07:39<3:52:46, 1.98it/s] 1%|██▌ | 300/27918 [11:28<4:02:38, 1.90it/s] 1%|███▍ | 400/27918 [15:20<3:53:23, 1.97it/s] 2%|████▎ | 500/27918 [19:13<3:58:47, 1.91it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.1739, 'learning_rate': 6.074933814946524e-05, 'epoch': 0.32} 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 340/340 [06:12<00:00, 1.11s/it] Configuration saved in ./checkpoint-500/config.json Model weights saved in ./checkpoint-500/pytorch_model.bin Configuration saved in ./checkpoint-500/preprocessor_config.json Configuration saved in ./preprocessor_config.json 02/07/2022 10:27:27 - WARNING - huggingface_hub.repository - Several commits (2) will be pushed upstream. Several commits (2) will be pushed upstream. Deleting older checkpoint [checkpoint-18000] due to args.save_total_limit 2%|█████ | 598/27918 [31:02<4:28:46, 1.69it/s] 3%|█████▉ | 700/27918 [34:56<3:57:31, 1.91it/s] 3%|██████▊ | 796/27918 [38:44<5:20:52, 1.41it/s] 3%|███████▋ | 900/27918 [42:37<3:41:49, 2.03it/s] 4%|████████▌ | 1000/27918 [46:28<3:49:28, 1.95it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.1504, 'learning_rate': 5.964150136068661e-05, 'epoch': 0.64} 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 340/340 [06:11<00:00, 1.12s/it] Configuration saved in ./checkpoint-1000/config.json Model weights saved in ./checkpoint-1000/pytorch_model.bin Configuration saved in ./checkpoint-1000/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-18500] due to args.save_total_limit 4%|█████████▎ | 1099/27918 [58:10<4:27:27, 1.67it/s] 4%|██████████▏ | 1200/27918 [1:02:02<3:44:02, 1.99it/s] 5%|██████████▉ | 1300/27918 [1:05:53<3:38:39, 2.03it/s] 5%|███████████▊ | 1400/27918 [1:09:46<3:35:50, 2.05it/s] 5%|████████████▋ | 1500/27918 [1:13:38<3:46:38, 1.94it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.1524, 'learning_rate': 5.8535880245485525e-05, 'epoch': 0.97} Configuration saved in ./checkpoint-1500/config.json Model weights saved in ./checkpoint-1500/pytorch_model.bin Configuration saved in ./checkpoint-1500/preprocessor_config.json {'eval_loss': 0.1544802337884903, 'eval_wer': 0.11355368821516468, 'eval_runtime': 377.7839, 'eval_samples_per_second': 14.392, 'eval_steps_per_second': 0.9, 'epoch': 0.97} Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-19000] due to args.save_total_limit 6%|█████████████▌ | 1601/27918 [1:25:21<3:45:22, 1.95it/s] 6%|██████████████▎ | 1698/27918 [1:29:06<4:34:58, 1.59it/s] 6%|███████████████▏ | 1801/27918 [1:32:55<3:38:15, 1.99it/s] 7%|████████████████ | 1898/27918 [1:36:41<4:27:38, 1.62it/s] 7%|████████████████▉ | 2000/27918 [1:40:34<3:57:15, 1.82it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.1318, 'learning_rate': 5.743025913028445e-05, 'epoch': 1.29} Configuration saved in ./checkpoint-2000/config.json {'eval_loss': 0.1547880917787552, 'eval_wer': 0.11122986262955083, 'eval_runtime': 378.6976, 'eval_samples_per_second': 14.357, 'eval_steps_per_second': 0.898, 'epoch': 1.29} Model weights saved in ./checkpoint-2000/pytorch_model.bin Configuration saved in ./checkpoint-2000/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-500] due to args.save_total_limit 8%|█████████████████▊ | 2101/27918 [1:52:16<3:39:29, 1.96it/s] 8%|██████████████████▌ | 2200/27918 [1:56:08<4:07:50, 1.73it/s] 8%|███████████████████▍ | 2300/27918 [1:59:57<4:00:25, 1.78it/s] 9%|████████████████████▎ | 2400/27918 [2:03:46<3:41:47, 1.92it/s] 9%|█████████████████████▏ | 2500/27918 [2:07:37<3:53:51, 1.81it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.1311, 'learning_rate': 5.632242234150582e-05, 'epoch': 1.61} Configuration saved in ./checkpoint-2500/config.json {'eval_loss': 0.15229424834251404, 'eval_wer': 0.11064155235471188, 'eval_runtime': 377.8113, 'eval_samples_per_second': 14.391, 'eval_steps_per_second': 0.9, 'epoch': 1.61} Model weights saved in ./checkpoint-2500/pytorch_model.bin Configuration saved in ./checkpoint-2500/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-1000] due to args.save_total_limit 9%|█████████████████████▉ | 2599/27918 [2:19:21<4:18:13, 1.63it/s] 10%|██████████████████████▊ | 2700/27918 [2:23:09<3:52:19, 1.81it/s] 10%|███████████████████████▋ | 2798/27918 [2:26:56<4:33:52, 1.53it/s] 10%|████████████████████████▌ | 2899/27918 [2:30:47<4:07:49, 1.68it/s] 11%|█████████████████████████▎ | 3000/27918 [2:34:42<3:52:17, 1.79it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.127, 'learning_rate': 5.521458555272718e-05, 'epoch': 1.93} 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 340/340 [06:07<00:00, 1.10s/it] Configuration saved in ./checkpoint-3000/config.json Model weights saved in ./checkpoint-3000/pytorch_model.bin Configuration saved in ./checkpoint-3000/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-1500] due to args.save_total_limit 11%|██████████████████████████▏ | 3098/27918 [2:46:17<4:38:02, 1.49it/s] 11%|███████████████████████████ | 3202/27918 [2:50:10<3:21:31, 2.04it/s] 12%|███████████████████████████▉ | 3301/27918 [2:54:04<4:00:02, 1.71it/s] 12%|████████████████████████████▋ | 3396/27918 [2:57:50<5:30:54, 1.24it/s] 13%|█████████████████████████████▌ | 3500/27918 [3:01:45<4:04:46, 1.66it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.1171, 'learning_rate': 5.410674876394855e-05, 'epoch': 2.26} 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 340/340 [06:05<00:00, 1.09s/it] Configuration saved in ./checkpoint-3500/config.json Model weights saved in ./checkpoint-3500/pytorch_model.bin Configuration saved in ./checkpoint-3500/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-2000] due to args.save_total_limit 13%|██████████████████████████████▍ | 3601/27918 [3:13:21<3:52:13, 1.75it/s] 13%|███████████████████████████████▎ | 3700/27918 [3:17:10<3:57:37, 1.70it/s] 14%|████████████████████████████████ | 3800/27918 [3:20:59<3:53:44, 1.72it/s] 14%|████████████████████████████████▉ | 3896/27918 [3:24:50<5:51:55, 1.14it/s] 14%|█████████████████████████████████▊ | 3999/27918 [3:28:36<4:16:04, 1.56it/s] 14%|█████████████████████████████████▊ | 4000/27918 [3:28:36<3:54:55, 1.70it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 340/340 [06:07<00:00, 1.10s/it] Configuration saved in ./checkpoint-4000/config.json Model weights saved in ./checkpoint-4000/pytorch_model.bin Configuration saved in ./checkpoint-4000/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-2500] due to args.save_total_limit 15%|██████████████████████████████████▋ | 4099/27918 [3:40:13<4:14:51, 1.56it/s] 15%|███████████████████████████████████▍ | 4198/27918 [3:44:02<4:45:45, 1.38it/s] 15%|████████████████████████████████████▎ | 4300/27918 [3:47:52<3:46:09, 1.74it/s] 16%|█████████████████████████████████████▏ | 4402/27918 [3:51:40<3:16:00, 2.00it/s] 16%|██████████████████████████████████████ | 4500/27918 [3:55:29<3:47:50, 1.71it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.1183, 'learning_rate': 5.189329085996883e-05, 'epoch': 2.9} 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 340/340 [06:06<00:00, 1.09s/it] Configuration saved in ./checkpoint-4500/config.json Model weights saved in ./checkpoint-4500/pytorch_model.bin Configuration saved in ./checkpoint-4500/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-3000] due to args.save_total_limit 16%|██████████████████████████████████████▉ | 4599/27918 [4:07:09<4:26:52, 1.46it/s] 17%|███████████████████████████████████████▋ | 4698/27918 [4:10:57<4:44:52, 1.36it/s] 17%|████████████████████████████████████████▌ | 4802/27918 [4:14:50<3:29:12, 1.84it/s] 18%|█████████████████████████████████████████▍ | 4900/27918 [4:18:33<4:03:47, 1.57it/s] 18%|██████████████████████████████████████████▎ | 4999/27918 [4:22:25<4:39:36, 1.37it/s] 18%|██████████████████████████████████████████▎ | 5000/27918 [4:22:26<4:13:47, 1.51it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 340/340 [06:07<00:00, 1.10s/it] Configuration saved in ./checkpoint-5000/config.json Model weights saved in ./checkpoint-5000/pytorch_model.bin Configuration saved in ./checkpoint-5000/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-3500] due to args.save_total_limit 18%|███████████████████████████████████████████ | 5100/27918 [4:33:59<4:13:30, 1.50it/s] 19%|███████████████████████████████████████████▉ | 5198/27918 [4:37:49<4:45:06, 1.33it/s] 19%|████████████████████████████████████████████▊ | 5299/27918 [4:41:40<4:32:07, 1.39it/s] 19%|█████████████████████████████████████████████▋ | 5398/27918 [4:45:27<4:37:06, 1.35it/s] 20%|██████████████████████████████████████████████▍ | 5500/27918 [4:49:19<4:17:54, 1.45it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.1153, 'learning_rate': 4.967761728241156e-05, 'epoch': 3.55} 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 340/340 [06:08<00:00, 1.09s/it] Configuration saved in ./checkpoint-5500/config.json Model weights saved in ./checkpoint-5500/pytorch_model.bin Configuration saved in ./checkpoint-5500/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-4000] due to args.save_total_limit 20%|███████████████████████████████████████████████▎ | 5599/27918 [5:00:57<4:25:23, 1.40it/s] 20%|████████████████████████████████████████████████▏ | 5701/27918 [5:04:44<3:43:07, 1.66it/s] 21%|█████████████████████████████████████████████████ | 5798/27918 [5:08:27<4:38:43, 1.32it/s] 21%|█████████████████████████████████████████████████▊ | 5897/27918 [5:12:15<5:24:05, 1.13it/s] 21%|██████████████████████████████████████████████████▋ | 6000/27918 [5:16:14<4:12:12, 1.45it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 {'loss': 0.1104, 'learning_rate': 4.857199616721048e-05, 'epoch': 3.87} 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 340/340 [06:07<00:00, 1.10s/it] Configuration saved in ./checkpoint-6000/config.json Model weights saved in ./checkpoint-6000/pytorch_model.bin Configuration saved in ./checkpoint-6000/preprocessor_config.json Configuration saved in ./preprocessor_config.json Deleting older checkpoint [checkpoint-4500] due to args.save_total_limit 22%|███████████████████████████████████████████████████▌ | 6101/27918 [5:27:58<3:34:09, 1.70it/s] 22%|████████████████████████████████████████████████████▍ | 6200/27918 [5:31:42<4:01:28, 1.50it/s] 23%|█████████████████████████████████████████████████████▎ | 6302/27918 [5:35:39<3:25:53, 1.75it/s] 23%|██████████████████████████████████████████████████████ | 6401/27918 [5:39:33<3:46:30, 1.58it/s] 23%|██████████████████████████████████████████████████████▉ | 6498/27918 [5:43:22<4:49:08, 1.23it/s] 23%|██████████████████████████████████████████████████████▉ | 6500/27918 [5:43:23<4:06:40, 1.45it/s]The following columns in the evaluation set don't have a corresponding argument in `Wav2Vec2ForCTC.forward` and have been ignored: input_length. If input_length are not expected by `Wav2Vec2ForCTC.forward`, you can safely ignore this message. ***** Running Evaluation ***** Num examples = 5437 Batch size = 16 1%|██▏ | 3/340 [00:01<03:21, 1.68it/s] 0%| | 0/340 [00:00