Training in progress, step 1500

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d85ffcde3dc196127ff875ccdf30a5e4b75c15ff1b1e16da1e554e52e7a9791
+oid sha256:d0862aa2d3971991ce599ac21449fce85407fae2ec83b0114af97a7dc8c839f1
 size 1262075377

runs/Feb02_22-30-13_brutasse/events.out.tfevents.1643841025.brutasse.21084.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e75f5271956fb42b97c6e862045d50b84b9f7f712b36b55ae92516756a06a0b
-size 6884
+oid sha256:b70f929b28e8e9c51ccf6a550567a21ae0e42c065facfe1cf393d91a881ab426
+size 7987

wandb/run-20220202_223026-1j0459xm/files/output.log

@@ -807,3 +807,407 @@ Configuration saved in ./checkpoint-1000/config.json
 {'eval_loss': 2.93613600730896, 'eval_wer': 1.0000283848992335, 'eval_runtime': 191.7956, 'eval_samples_per_second': 25.251, 'eval_steps_per_second': 3.16, 'epoch': 2.74}
 Model weights saved in ./checkpoint-1000/pytorch_model.bin
 Configuration saved in ./checkpoint-1000/preprocessor_config.json
+Configuration saved in ./preprocessor_config.json
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+  6%|███                                                | 1100/18250 [29:49<8:02:27,  1.69s/it]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+  7%|███▎                                               | 1200/18250 [31:49<7:43:42,  1.63s/it]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+  7%|███▋                                               | 1299/18250 [33:47<7:44:57,  1.65s/it]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+  8%|███▉                                               | 1399/18250 [35:48<7:36:23,  1.63s/it]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+  8%|████▏                                              | 1498/18250 [37:50<4:51:04,  1.04s/it]
+  8%|████▏                                              | 1500/18250 [37:52<4:35:15,  1.01it/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 = 4843
+  Batch size = 8
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Configuration saved in ./checkpoint-1500/config.json
+{'eval_loss': 1.1156758069992065, 'eval_wer': 0.8358501277320466, 'eval_runtime': 192.2259, 'eval_samples_per_second': 25.194, 'eval_steps_per_second': 3.153, 'epoch': 4.11}
+Model weights saved in ./checkpoint-1500/pytorch_model.bin
+Configuration saved in ./checkpoint-1500/preprocessor_config.json