1%|▎ | 100/18250 [02:02<4:08:51, 1.22it/s] 1%|▌ | 200/18250 [04:03<4:10:12, 1.20it/s] 2%|▊ | 300/18250 [06:04<4:06:59, 1.21it/s] 2%|█▏ | 399/18250 [08:07<5:07:58, 1.04s/it] 3%|█▍ | 500/18250 [10:09<5:23:38, 1.09s/it]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 = 4843 Batch size = 8 {'loss': 3.3224, 'learning_rate': 1.8712499999999997e-05, 'epoch': 1.37} Configuration saved in ./checkpoint-500/config.json {'eval_loss': 3.335383415222168, 'eval_wer': 1.0, 'eval_runtime': 191.0801, 'eval_samples_per_second': 25.345, 'eval_steps_per_second': 3.171, 'epoch': 1.37} Model weights saved in ./checkpoint-500/pytorch_model.bin Configuration saved in ./checkpoint-500/preprocessor_config.json Configuration saved in ./preprocessor_config.json 3%|█▋ | 598/18250 [15:32<5:14:11, 1.07s/it] 4%|█▉ | 699/18250 [17:33<5:09:45, 1.06s/it] 4%|██▎ | 799/18250 [19:36<6:38:28, 1.37s/it] 5%|██▌ | 899/18250 [21:37<6:30:30, 1.35s/it] 5%|██▊ | 1000/18250 [23:38<6:23:33, 1.33s/it]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 = 4843 Batch size = 8 0%| | 0/606 [00:00