diff --git "a/wandb/run-20220303_004520-25bnjrx1/files/output.log" "b/wandb/run-20220303_004520-25bnjrx1/files/output.log" --- "a/wandb/run-20220303_004520-25bnjrx1/files/output.log" +++ "b/wandb/run-20220303_004520-25bnjrx1/files/output.log" @@ -5670,3 +5670,358 @@ Upload file wandb/run-20220303_004520-25bnjrx1/run-25bnjrx1.wandb: 100%|██ Upload file wandb/run-20220303_004520-25bnjrx1/run-25bnjrx1.wandb: 100%|███████████| 48.1M/48.1M [00:16<00:00, 16.4MB/s][INFO|trainer.py:1492] 2022-03-03 03:10:31,663 >> 1,165 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. 03/03/2022 03:18:09 - WARNING - huggingface_hub.repository - To https://huggingface.co/sanchit-gandhi/wav2vec2-gpt2-wandb-grid-search Upload file wandb/run-20220303_004520-25bnjrx1/run-25bnjrx1.wandb: 100%|███████████| 48.1M/48.1M [05:22<00:00, 88.9kB/s][INFO|trainer.py:1492] 2022-03-03 03:10:31,663 >> 1,165 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +[INFO|modelcard.py:460] 2022-03-03 03:18:12,068 >> Dropping the following result as it does not have all the necessary fields:trainer.py:1492] 2022-03-03 03:10:31,663 >> 1,165 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +Upload file wandb/run-20220303_004520-25bnjrx1/run-25bnjrx1.wandb: 22%|██▍ | 10.5M/48.2M [00:01<00:03, 11.0MB/s]ields:trainer.py:1492] 2022-03-03 03:10:31,663 >> 1,165 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +Upload file wandb/run-20220303_004520-25bnjrx1/run-25bnjrx1.wandb: 22%|██▍ | 10.5M/48.2M [00:01<00:03, 11.0MB/s]ields:trainer.py:1492] 2022-03-03 03:10:31,663 >> 1,165 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +03/03/2022 03:18:33 - WARNING - huggingface_hub.repository - To https://huggingface.co/sanchit-gandhi/wav2vec2-gpt2-wandb-grid-search +Upload file wandb/run-20220303_004520-25bnjrx1/run-25bnjrx1.wandb: 64%|███████ | 30.7M/48.2M [00:02<00:01, 16.9MB/s]To https://huggingface.co/sanchit-gandhi/wav2vec2-gpt2-wandb-grid-searchimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +[INFO|trainer.py:2366] 2022-03-03 03:18:36,671 >> Num examples = 2642in the evaluation set don't have a corresponding argument in `SpeechEncoderDecoderModel.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message.ut_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +[INFO|trainer.py:2366] 2022-03-03 03:18:36,671 >> Num examples = 2642in the evaluation set don't have a corresponding argument in `SpeechEncoderDecoderModel.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message.ut_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +***** train metrics ***** + epoch = 1.0 + train_loss = 7.0053 + train_runtime = 2:25:11.59 + train_samples = 28538 + train_samples_per_second = 3.276 + train_steps_per_second = 0.205 + 0%| | 0/331 [00:00> Saving model checkpoint to ./ | 3/331 [00:05<10:34, 1.94s/it] argument in `SpeechEncoderDecoderModel.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message.ut_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +[INFO|trainer.py:2114] 2022-03-03 03:37:15,407 >> Saving model checkpoint to ./ | 3/331 [00:05<10:34, 1.94s/it] argument in `SpeechEncoderDecoderModel.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message.ut_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +03/03/2022 03:37:15 - INFO - datasets.metric - Removing /home/sanchit_huggingface_co/.cache/huggingface/metrics/wer/default/default_experiment-1-0.arrow +***** eval metrics ***** + epoch = 1.0 + eval_loss = 5.0542 + eval_runtime = 0:18:38.73 + eval_samples = 2642 + eval_samples_per_second = 2.362 + eval_steps_per_second = 0.296 +[INFO|modeling_utils.py:1081] 2022-03-03 03:37:31,865 >> Model weights saved in ./pytorch_model.bin:05<10:34, 1.94s/it] argument in `SpeechEncoderDecoderModel.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message.ut_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message.