diff --git "a/wandb/run-20220302_094439-2kys49al/files/output.log" "b/wandb/run-20220302_094439-2kys49al/files/output.log" --- "a/wandb/run-20220302_094439-2kys49al/files/output.log" +++ "b/wandb/run-20220302_094439-2kys49al/files/output.log" @@ -7713,3 +7713,1314 @@ 03/02/2022 12:54:38 - INFO - datasets.metric - Removing /home/sanchit_huggingface_co/.cache/huggingface/metrics/wer/default/default_experiment-1-0.arrow {'eval_loss': 6.7275872230529785, 'eval_wer': 1.4950096235887056, 'eval_runtime': 1512.5692, 'eval_samples_per_second': 1.747, 'eval_steps_per_second': 0.437, 'epoch': 0.7} 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8606, 'learning_rate': 2.4115282815410253e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7386, 'learning_rate': 2.4112341927262033e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7672, 'learning_rate': 2.4109401039113816e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9722, 'learning_rate': 2.4106460150965592e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9818, 'learning_rate': 2.4103519262817372e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.0692, 'learning_rate': 2.410057837466915e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.5583, 'learning_rate': 2.4097637486520928e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.77, 'learning_rate': 2.409469659837271e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7405, 'learning_rate': 2.4091755710224487e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.895, 'learning_rate': 2.4088814822076267e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.0186, 'learning_rate': 2.4085873933928047e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7219, 'learning_rate': 2.4082933045779823e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9566, 'learning_rate': 2.4079992157631606e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8832, 'learning_rate': 2.4077051269483386e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.0101, 'learning_rate': 2.4074110381335162e-05, 'epoch': 0.7} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.1824, 'learning_rate': 2.4071169493186942e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.6849, 'learning_rate': 2.4068228605038722e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.072, 'learning_rate': 2.40652877168905e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7548, 'learning_rate': 2.406234682874228e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9246, 'learning_rate': 2.405940594059406e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.864, 'learning_rate': 2.4056465052445837e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.4563, 'learning_rate': 2.405352416429762e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.6535, 'learning_rate': 2.40505832761494e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8977, 'learning_rate': 2.4047642388001177e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9476, 'learning_rate': 2.4044701499852956e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8524, 'learning_rate': 2.4041760611704733e-05, 'epoch': 0.71} +{'loss': 7.0802, 'learning_rate': 2.4038819723556516e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9204, 'learning_rate': 2.4035878835408295e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.0114, 'learning_rate': 2.4032937947260072e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7017, 'learning_rate': 2.402999705911185e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9901, 'learning_rate': 2.402705617096363e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.6207, 'learning_rate': 2.402411528281541e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7629, 'learning_rate': 2.402117439466719e-05, 'epoch': 0.71} +{'loss': 6.6962, 'learning_rate': 2.401823350651897e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.6469, 'learning_rate': 2.4015292618370747e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9103, 'learning_rate': 2.401235173022253e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.0212, 'learning_rate': 2.400941084207431e-05, 'epoch': 0.71} +{'loss': 6.7428, 'learning_rate': 2.4006469953926086e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7249, 'learning_rate': 2.4003529065777866e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.929, 'learning_rate': 2.4000588177629642e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.6335, 'learning_rate': 2.3997647289481425e-05, 'epoch': 0.71} +{'loss': 6.8545, 'learning_rate': 2.3994706401333205e-05, 'epoch': 0.71} + 23%|█████████████████▎ | 2500/10701 [2:44:45<3:07:46, 1.37s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▌ | 2544/10701 [3:13:33<4:03:27, 1.79s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▌ | 2544/10701 [3:13:33<4:03:27, 1.79s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.2894, 'learning_rate': 2.398882462503676e-05, 'epoch': 0.71} + 24%|█████████████████▌ | 2544/10701 [3:13:33<4:03:27, 1.79s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.5203, 'learning_rate': 2.398588373688854e-05, 'epoch': 0.71} +{'loss': 6.6533, 'learning_rate': 2.398294284874032e-05, 'epoch': 0.71} + 24%|█████████████████▌ | 2544/10701 [3:13:33<4:03:27, 1.79s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.4511, 'learning_rate': 2.397706107244388e-05, 'epoch': 0.71} +{'loss': 6.3122, 'learning_rate': 2.3974120184295656e-05, 'epoch': 0.71} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.612, 'learning_rate': 2.3971179296147436e-05, 'epoch': 0.71} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.1896, 'learning_rate': 2.396823840799922e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.861, 'learning_rate': 2.3965297519850995e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9684, 'learning_rate': 2.3962356631702775e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.1394, 'learning_rate': 2.395941574355455e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8962, 'learning_rate': 2.3956474855406334e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8713, 'learning_rate': 2.3953533967258114e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8345, 'learning_rate': 2.395059307910989e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9139, 'learning_rate': 2.394765219096167e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9511, 'learning_rate': 2.394471130281345e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.0635, 'learning_rate': 2.394177041466523e-05, 'epoch': 0.72} + 24%|█████████████████▌ | 2548/10701 [3:13:39<3:23:50, 1.50s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.881, 'learning_rate': 2.393588863836879e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8575, 'learning_rate': 2.3932947750220565e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7461, 'learning_rate': 2.3930006862072345e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7338, 'learning_rate': 2.392706597392413e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8163, 'learning_rate': 2.3924125085775905e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8898, 'learning_rate': 2.3921184197627684e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.6906, 'learning_rate': 2.3918243309479464e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.6961, 'learning_rate': 2.391530242133124e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.5963, 'learning_rate': 2.3912361533183024e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8234, 'learning_rate': 2.39094206450348e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.841, 'learning_rate': 2.390647975688658e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.5371, 'learning_rate': 2.390353886873836e-05, 'epoch': 0.72} +{'loss': 7.0885, 'learning_rate': 2.390059798059014e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8453, 'learning_rate': 2.389765709244192e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8613, 'learning_rate': 2.38947162042937e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8336, 'learning_rate': 2.3891775316145475e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.2375, 'learning_rate': 2.3888834427997255e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.4281, 'learning_rate': 2.3885893539849038e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.0128, 'learning_rate': 2.3882952651700814e-05, 'epoch': 0.72} +{'loss': 6.9641, 'learning_rate': 2.3880011763552594e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9362, 'learning_rate': 2.3877070875404373e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9083, 'learning_rate': 2.387412998725615e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8365, 'learning_rate': 2.3871189099107933e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.8831, 'learning_rate': 2.386824821095971e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 7.0533, 'learning_rate': 2.386530732281149e-05, 'epoch': 0.72} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.6844, 'learning_rate': 2.386236643466327e-05, 'epoch': 0.73} + 24%|█████████████████▋ | 2562/10701 [3:14:10<5:12:44, 2.31s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.7105, 'learning_rate': 2.385942554651505e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2589/10701 [3:15:06<4:09:53, 1.85s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▉ | 2589/10701 [3:15:06<4:09:53, 1.85s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.768, 'learning_rate': 2.3853543770218608e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2589/10701 [3:15:06<4:09:53, 1.85s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.9582, 'learning_rate': 2.3850602882070384e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2589/10701 [3:15:06<4:09:53, 1.85s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.6794, 'learning_rate': 2.3847661993922164e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2589/10701 [3:15:06<4:09:53, 1.85s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +{'loss': 6.5575, 'learning_rate': 2.3844721105773947e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2594/10701 [3:15:13<3:28:15, 1.54s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. + 24%|█████████████████▉ | 2594/10701 [3:15:13<3:28:15, 1.54s/it][INFO|trainer.py:560] 2022-03-02 12:29:26,129 >> The following columns in 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. +[WARNING|modeling_utils.py:388] 2022-03-02 12:59:57,303 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed26,129 >> The following columns in 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. +[WARNING|modeling_utils.py:388] 2022-03-02 12:59:57,303 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed26,129 >> The following columns in 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. +{'loss': 6.4152, 'learning_rate': 2.3835898441329283e-05, 'epoch': 0.73} +{'loss': 6.4186, 'learning_rate': 2.383295755318106e-05, 'epoch': 0.73} +[WARNING|modeling_utils.py:388] 2022-03-02 12:59:57,303 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed26,129 >> The following columns in 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. +{'loss': 6.5704, 'learning_rate': 2.3830016665032842e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.0848, 'learning_rate': 2.38241348887364e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 7.0668, 'learning_rate': 2.3821194000588178e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 7.2609, 'learning_rate': 2.3818253112439954e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.967, 'learning_rate': 2.3815312224291738e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.8208, 'learning_rate': 2.3812371336143517e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.6882, 'learning_rate': 2.3809430447995294e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.9377, 'learning_rate': 2.3806489559847073e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.8347, 'learning_rate': 2.3803548671698856e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.7557, 'learning_rate': 2.3800607783550633e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.8517, 'learning_rate': 2.3797666895402413e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.7304, 'learning_rate': 2.3794726007254192e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.8754, 'learning_rate': 2.379178511910597e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.774, 'learning_rate': 2.3788844230957752e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 7.0194, 'learning_rate': 2.378590334280953e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.7312, 'learning_rate': 2.3782962454661308e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 7.0986, 'learning_rate': 2.3780021566513088e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.5186, 'learning_rate': 2.3777080678364864e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.8469, 'learning_rate': 2.3774139790216647e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 7.0811, 'learning_rate': 2.3771198902068427e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.5319, 'learning_rate': 2.3768258013920203e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.6521, 'learning_rate': 2.3765317125771983e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.6511, 'learning_rate': 2.3762376237623762e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.885, 'learning_rate': 2.3759435349475542e-05, 'epoch': 0.73} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.6292, 'learning_rate': 2.3756494461327322e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 7.2157, 'learning_rate': 2.37535535731791e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.9484, 'learning_rate': 2.3750612685030878e-05, 'epoch': 0.74} +{'loss': 6.6367, 'learning_rate': 2.374767179688266e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.6773, 'learning_rate': 2.374473090873444e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.7081, 'learning_rate': 2.3741790020586217e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 7.027, 'learning_rate': 2.3738849132437997e-05, 'epoch': 0.74} +{'loss': 6.7326, 'learning_rate': 2.3735908244289773e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.8859, 'learning_rate': 2.3732967356141556e-05, 'epoch': 0.74} +{'loss': 7.0325, 'learning_rate': 2.3730026467993336e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.9581, 'learning_rate': 2.3727085579845112e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.7137, 'learning_rate': 2.3724144691696892e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.7129, 'learning_rate': 2.3721203803548672e-05, 'epoch': 0.74} +{'loss': 6.7425, 'learning_rate': 2.371826291540045e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.7668, 'learning_rate': 2.371532202725223e-05, 'epoch': 0.74} + 24%|█████████████████▉ | 2599/10701 [3:15:19<2:42:05, 1.20s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.868, 'learning_rate': 2.371238113910401e-05, 'epoch': 0.74} + 25%|██████████████████▏ | 2639/10701 [3:16:43<4:05:56, 1.83s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 25%|██████████████████▏ | 2639/10701 [3:16:43<4:05:56, 1.83s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.9415, 'learning_rate': 2.370649936280757e-05, 'epoch': 0.74} + 25%|██████████████████▏ | 2639/10701 [3:16:43<4:05:56, 1.83s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.7656, 'learning_rate': 2.370355847465935e-05, 'epoch': 0.74} + 25%|██████████████████▏ | 2639/10701 [3:16:43<4:05:56, 1.83s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 25%|███���██████████████▎ | 2643/10701 [3:16:49<3:36:52, 1.61s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 25%|██████████████████▎ | 2643/10701 [3:16:49<3:36:52, 1.61s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.5923, 'learning_rate': 2.3697676698362906e-05, 'epoch': 0.74} +{'loss': 6.5312, 'learning_rate': 2.3694735810214683e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2643/10701 [3:16:49<3:36:52, 1.61s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 25%|██████████████████▎ | 2646/10701 [3:16:53<3:12:16, 1.43s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. + 25%|██████████████████▎ | 2646/10701 [3:16:53<3:12:16, 1.43s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.57, 'learning_rate': 2.3688854033918245e-05, 'epoch': 0.74} +{'loss': 6.9269, 'learning_rate': 2.3685913145770022e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2646/10701 [3:16:53<3:12:16, 1.43s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:00:01,204 >> 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. +{'loss': 6.1079, 'learning_rate': 2.36829722576218e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2649/10701 [3:16:57<2:47:06, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▎ | 2649/10701 [3:16:57<2:47:06, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.0438, 'learning_rate': 2.367709048132536e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2649/10701 [3:16:57<2:47:06, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.7649, 'learning_rate': 2.367414959317714e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2649/10701 [3:16:57<2:47:06, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▎ | 2649/10701 [3:16:57<2:47:06, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.7911, 'learning_rate': 2.3668267816880697e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.0483, 'learning_rate': 2.3665326928732476e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.0664, 'learning_rate': 2.366238604058426e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.0949, 'learning_rate': 2.3659445152436036e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.0437, 'learning_rate': 2.3656504264287816e-05, 'epoch': 0.74} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.772, 'learning_rate': 2.3653563376139595e-05, 'epoch': 0.75} +{'loss': 6.8837, 'learning_rate': 2.3650622487991375e-05, 'epoch': 0.75} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.8825, 'learning_rate': 2.3647681599843155e-05, 'epoch': 0.75} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.917, 'learning_rate': 2.364474071169493e-05, 'epoch': 0.75} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.9344, 'learning_rate': 2.364179982354671e-05, 'epoch': 0.75} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.6294, 'learning_rate': 2.363885893539849e-05, 'epoch': 0.75} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.6454, 'learning_rate': 2.363591804725027e-05, 'epoch': 0.75} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.2488, 'learning_rate': 2.363297715910205e-05, 'epoch': 0.75} + 25%|██████████████████▎ | 2653/10701 [3:17:05<4:32:29, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.4929, 'learning_rate': 2.3627095382805606e-05, 'epoch': 0.75} + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.9594, 'learning_rate': 2.3624154494657386e-05, 'epoch': 0.75} + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.8647, 'learning_rate': 2.362121360650917e-05, 'epoch': 0.75} + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.7428, 'learning_rate': 2.3618272718360945e-05, 'epoch': 0.75} + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.6256, 'learning_rate': 2.3615331830212725e-05, 'epoch': 0.75} + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▍ | 2673/10701 [3:17:49<4:44:13, 2.12s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▍ | 2673/10701 [3:17:49<4:44:13, 2.12s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.0625, 'learning_rate': 2.360945005391628e-05, 'epoch': 0.75} + 25%|██████████████████▍ | 2673/10701 [3:17:49<4:44:13, 2.12s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.8355, 'learning_rate': 2.3606509165768064e-05, 'epoch': 0.75} + 25%|██████████████████▍ | 2673/10701 [3:17:49<4:44:13, 2.12s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.8038, 'learning_rate': 2.360356827761984e-05, 'epoch': 0.75} + 25%|██████████████████▍ | 2673/10701 [3:17:49<4:44:13, 2.12s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.4689, 'learning_rate': 2.35976865013234e-05, 'epoch': 0.75} + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.0118, 'learning_rate': 2.359474561317518e-05, 'epoch': 0.75} + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.803, 'learning_rate': 2.359180472502696e-05, 'epoch': 0.75} + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.7267, 'learning_rate': 2.358886383687874e-05, 'epoch': 0.75} + [WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.8486, 'learning_rate': 2.3582982060582295e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.6274, 'learning_rate': 2.358004117243408e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.1554, 'learning_rate': 2.3577100284285855e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.8586, 'learning_rate': 2.3574159396137634e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.5775, 'learning_rate': 2.3571218507989414e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.9805, 'learning_rate': 2.356827761984119e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.1246, 'learning_rate': 2.3565336731692974e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.7246, 'learning_rate': 2.356239584354475e-05, 'epoch': 0.75} +{'loss': 7.0059, 'learning_rate': 2.355945495539653e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.0876, 'learning_rate': 2.355651406724831e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 7.1619, 'learning_rate': 2.355357317910009e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.7, 'learning_rate': 2.355063229095187e-05, 'epoch': 0.75} + 25%|██████████████████▌ | 2682/10701 [3:18:08<4:30:56, 2.03s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.925, 'learning_rate': 2.354769140280365e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2696/10701 [3:18:32<3:14:25, 1.46s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▋ | 2696/10701 [3:18:32<3:14:25, 1.46s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. +{'loss': 6.7932, 'learning_rate': 2.3541809626507205e-05, 'epoch': 0.76} +{'loss': 6.8392, 'learning_rate': 2.3538868738358988e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2696/10701 [3:18:32<3:14:25, 1.46s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:01:38,655 >> 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. + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.311, 'learning_rate': 2.3532986962062544e-05, 'epoch': 0.76} +{'loss': 6.0855, 'learning_rate': 2.3530046073914324e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9138, 'learning_rate': 2.35271051857661e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.1443, 'learning_rate': 2.3524164297617883e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.7155, 'learning_rate': 2.352122340946966e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.1171, 'learning_rate': 2.351828252132144e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8267, 'learning_rate': 2.351534163317322e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.0675, 'learning_rate': 2.3512400745024995e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.0236, 'learning_rate': 2.3509459856876778e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.5944, 'learning_rate': 2.3506518968728558e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9538, 'learning_rate': 2.3503578080580334e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8372, 'learning_rate': 2.3500637192432114e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.1368, 'learning_rate': 2.3497696304283897e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8877, 'learning_rate': 2.3494755416135673e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.918, 'learning_rate': 2.3491814527987453e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.6072, 'learning_rate': 2.3488873639839233e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8274, 'learning_rate': 2.348593275169101e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8231, 'learning_rate': 2.3482991863542792e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9972, 'learning_rate': 2.3480050975394572e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.5907, 'learning_rate': 2.347711008724635e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8729, 'learning_rate': 2.3474169199098128e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9611, 'learning_rate': 2.3471228310949905e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8115, 'learning_rate': 2.3468287422801688e-05, 'epoch': 0.76} +{'loss': 7.1729, 'learning_rate': 2.3465346534653467e-05, 'epoch': 0.76} + 25%|████████████████���█▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.5809, 'learning_rate': 2.3462405646505244e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9671, 'learning_rate': 2.3459464758357023e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8137, 'learning_rate': 2.3456523870208803e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.6775, 'learning_rate': 2.3453582982060583e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.2042, 'learning_rate': 2.3450642093912363e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8522, 'learning_rate': 2.3447701205764142e-05, 'epoch': 0.76} +{'loss': 6.8993, 'learning_rate': 2.344476031761592e-05, 'epoch': 0.76} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.6326, 'learning_rate': 2.3441819429467702e-05, 'epoch': 0.77} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.0945, 'learning_rate': 2.343887854131948e-05, 'epoch': 0.77} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.5864, 'learning_rate': 2.3435937653171258e-05, 'epoch': 0.77} +{'loss': 6.8055, 'learning_rate': 2.3432996765023038e-05, 'epoch': 0.77} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.883, 'learning_rate': 2.3430055876874814e-05, 'epoch': 0.77} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8956, 'learning_rate': 2.3427114988726597e-05, 'epoch': 0.77} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9648, 'learning_rate': 2.3424174100578377e-05, 'epoch': 0.77} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9897, 'learning_rate': 2.3421233212430153e-05, 'epoch': 0.77} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.7348, 'learning_rate': 2.3418292324281933e-05, 'epoch': 0.77} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.0785, 'learning_rate': 2.3415351436133713e-05, 'epoch': 0.77} + 25%|██████████████████▋ | 2699/10701 [3:18:35<2:46:43, 1.25s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|██████████████████▉ | 2741/10701 [3:20:05<4:02:47, 1.83s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|██████████████████▉ | 2741/10701 [3:20:05<4:02:47, 1.83s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.7194, 'learning_rate': 2.3409469659837272e-05, 'epoch': 0.77} + 26%|██████████████████▉ | 2741/10701 [3:20:05<4:02:47, 1.83s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.6547, 'learning_rate': 2.3406528771689052e-05, 'epoch': 0.77} + 26%|██████████████████▉ | 2741/10701 [3:20:05<4:02:47, 1.83s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.7459, 'learning_rate': 2.3403587883540828e-05, 'epoch': 0.77} + 26%|██████████████████▉ | 2745/10701 [3:20:11<3:26:51, 1.56s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|██████████████████▉ | 2745/10701 [3:20:11<3:26:51, 1.56s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.7476, 'learning_rate': 2.339770610724439e-05, 'epoch': 0.77} + 26%|██████████████████▉ | 2745/10701 [3:20:11<3:26:51, 1.56s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8123, 'learning_rate': 2.3394765219096167e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.5942, 'learning_rate': 2.3388883442799723e-05, 'epoch': 0.77} +{'loss': 6.51, 'learning_rate': 2.3385942554651506e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.3627, 'learning_rate': 2.3383001666503286e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.2297, 'learning_rate': 2.3380060778355062e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.1871, 'learning_rate': 2.3377119890206842e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.7022, 'learning_rate': 2.3374179002058622e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.5546, 'learning_rate': 2.33712381139104e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.0547, 'learning_rate': 2.336829722576218e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.7496, 'learning_rate': 2.336535633761396e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9656, 'learning_rate': 2.3362415449465737e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.2435, 'learning_rate': 2.3359474561317517e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9227, 'learning_rate': 2.33565336731693e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.7228, 'learning_rate': 2.3353592785021077e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.609, 'learning_rate': 2.3350651896872856e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9451, 'learning_rate': 2.3347711008724636e-05, 'epoch': 0.77} +{'loss': 6.7429, 'learning_rate': 2.3344770120576416e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.2265, 'learning_rate': 2.3341829232428196e-05, 'epoch': 0.77} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.5575, 'learning_rate': 2.3338888344279972e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.0071, 'learning_rate': 2.333594745613175e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.469, 'learning_rate': 2.333300656798353e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8776, 'learning_rate': 2.333006567983531e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9118, 'learning_rate': 2.332712479168709e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.732, 'learning_rate': 2.332418390353887e-05, 'epoch': 0.78} +{'loss': 6.8946, 'learning_rate': 2.3321243015390647e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8659, 'learning_rate': 2.3318302127242427e-05, 'epoch': 0.78} +{'loss': 6.8897, 'learning_rate': 2.331536123909421e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 7.0182, 'learning_rate': 2.3312420350945986e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.7662, 'learning_rate': 2.3309479462797766e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.6283, 'learning_rate': 2.3306538574649545e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8099, 'learning_rate': 2.3303597686501322e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.6977, 'learning_rate': 2.3300656798353105e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.5022, 'learning_rate': 2.329771591020488e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8182, 'learning_rate': 2.329477502205666e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8634, 'learning_rate': 2.329183413390844e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8621, 'learning_rate': 2.328889324576022e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.9743, 'learning_rate': 2.3285952357612e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.6791, 'learning_rate': 2.328301146946378e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8845, 'learning_rate': 2.3280070581315556e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8778, 'learning_rate': 2.3277129693167336e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.6454, 'learning_rate': 2.327418880501912e-05, 'epoch': 0.78} + 26%|███████████████████ | 2748/10701 [3:20:14<2:56:43, 1.33s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8615, 'learning_rate': 2.3271247916870895e-05, 'epoch': 0.78} + 26%|███████████████████▎ | 2790/10701 [3:21:42<4:04:07, 1.85s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████▎ | 2790/10701 [3:21:42<4:04:07, 1.85s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.757, 'learning_rate': 2.3265366140574455e-05, 'epoch': 0.78} + 26%|███████████████████▎ | 2790/10701 [3:21:42<4:04:07, 1.85s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.4839, 'learning_rate': 2.326242525242623e-05, 'epoch': 0.78} + 26%|███████████████████▎ | 2790/10701 [3:21:42<4:04:07, 1.85s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8202, 'learning_rate': 2.3259484364278014e-05, 'epoch': 0.78} + 26%|███████████████████▎ | 2790/10701 [3:21:42<4:04:07, 1.85s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████▎ | 2795/10701 [3:21:50<3:22:43, 1.54s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. + 26%|███████████████████▎ | 2795/10701 [3:21:50<3:22:43, 1.54s/it][WARNING|modeling_utils.py:388] 2022-03-02 13:03:17,353 >> 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. +{'loss': 6.8465, 'learning_rate': 2.325360258798157e-05, 'epoch': 0.78} +[WARNING|modeling_utils.py:388] 2022-03-02 13:06:33,652 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +[WARNING|modeling_utils.py:388] 2022-03-02 13:06:33,652 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7396, 'learning_rate': 2.3247720811685126e-05, 'epoch': 0.78} +{'loss': 6.4441, 'learning_rate': 2.324477992353691e-05, 'epoch': 0.78} +[WARNING|modeling_utils.py:388] 2022-03-02 13:06:33,652 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.3132, 'learning_rate': 2.3238898147240466e-05, 'epoch': 0.78} +{'loss': 6.4199, 'learning_rate': 2.3235957259092245e-05, 'epoch': 0.78} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8663, 'learning_rate': 2.323301637094403e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7951, 'learning_rate': 2.3230075482795805e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7987, 'learning_rate': 2.3227134594647584e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9305, 'learning_rate': 2.3224193706499364e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7359, 'learning_rate': 2.322125281835114e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9297, 'learning_rate': 2.3218311930202924e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.192, 'learning_rate': 2.3215371042054703e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8315, 'learning_rate': 2.321243015390648e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8395, 'learning_rate': 2.320948926575826e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9153, 'learning_rate': 2.3206548377610036e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.918, 'learning_rate': 2.320360748946182e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6716, 'learning_rate': 2.32006666013136e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8052, 'learning_rate': 2.3197725713165375e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7442, 'learning_rate': 2.3194784825017155e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.009, 'learning_rate': 2.3191843936868938e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.1373, 'learning_rate': 2.3188903048720714e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9977, 'learning_rate': 2.3185962160572494e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7622, 'learning_rate': 2.3183021272424274e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7717, 'learning_rate': 2.318008038427605e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9384, 'learning_rate': 2.3177139496127833e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0186, 'learning_rate': 2.3174198607979613e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0688, 'learning_rate': 2.317125771983139e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7214, 'learning_rate': 2.316831683168317e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0195, 'learning_rate': 2.3165375943534945e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8692, 'learning_rate': 2.3162435055386728e-05, 'epoch': 0.79} + 26%|███████████████████▎ | 2800/10701 [3:21:56<2:59:56, 1.37s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8334, 'learning_rate': 2.3156553279090284e-05, 'epoch': 0.79} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7098, 'learning_rate': 2.3153612390942064e-05, 'epoch': 0.79} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7565, 'learning_rate': 2.3150671502793844e-05, 'epoch': 0.79} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0283, 'learning_rate': 2.3147730614645624e-05, 'epoch': 0.79} + 26%|█████���█████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9885, 'learning_rate': 2.3144789726497403e-05, 'epoch': 0.79} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9871, 'learning_rate': 2.3141848838349183e-05, 'epoch': 0.79} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8199, 'learning_rate': 2.313890795020096e-05, 'epoch': 0.79} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8542, 'learning_rate': 2.3135967062052742e-05, 'epoch': 0.79} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.831, 'learning_rate': 2.3133026173904522e-05, 'epoch': 0.79} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.1539, 'learning_rate': 2.31300852857563e-05, 'epoch': 0.79} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.774, 'learning_rate': 2.3127144397608078e-05, 'epoch': 0.8} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9036, 'learning_rate': 2.3124203509459855e-05, 'epoch': 0.8} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8835, 'learning_rate': 2.3121262621311638e-05, 'epoch': 0.8} +{'loss': 6.7455, 'learning_rate': 2.3118321733163417e-05, 'epoch': 0.8} + 26%|███████████████████▌ | 2827/10701 [3:22:56<4:34:33, 2.09s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|███████████████████▋ | 2842/10701 [3:23:24<3:45:57, 1.73s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|███████████████████▋ | 2842/10701 [3:23:24<3:45:57, 1.73s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7612, 'learning_rate': 2.3112439956866973e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2842/10701 [3:23:24<3:45:57, 1.73s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7518, 'learning_rate': 2.3109499068718753e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2842/10701 [3:23:24<3:45:57, 1.73s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.5052, 'learning_rate': 2.3106558180570533e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2846/10701 [3:23:30<3:12:55, 1.47s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|███████████████████▋ | 2846/10701 [3:23:30<3:12:55, 1.47s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +[WARNING|modeling_utils.py:388] 2022-03-02 13:08:13,545 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +[WARNING|modeling_utils.py:388] 2022-03-02 13:08:13,545 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.2193, 'learning_rate': 2.309773551612587e-05, 'epoch': 0.8} +{'loss': 6.1175, 'learning_rate': 2.309479462797765e-05, 'epoch': 0.8} +[WARNING|modeling_utils.py:388] 2022-03-02 13:08:13,545 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.2758, 'learning_rate': 2.309185373982943e-05, 'epoch': 0.8} +{'loss': 6.309, 'learning_rate': 2.3088912851681208e-05, 'epoch': 0.8} +[WARNING|modeling_utils.py:388] 2022-03-02 13:08:13,545 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0421, 'learning_rate': 2.3085971963532988e-05, 'epoch': 0.8} +[WARNING|modeling_utils.py:388] 2022-03-02 13:08:13,545 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +[WARNING|modeling_utils.py:388] 2022-03-02 13:08:13,545 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8171, 'learning_rate': 2.3083031075384767e-05, 'epoch': 0.8} +[WARNING|modeling_utils.py:388] 2022-03-02 13:08:13,545 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7008, 'learning_rate': 2.3080090187236547e-05, 'epoch': 0.8} +[WARNING|modeling_utils.py:388] 2022-03-02 13:08:13,545 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9021, 'learning_rate': 2.3077149299088327e-05, 'epoch': 0.8} +[WARNING|modeling_utils.py:388] 2022-03-02 13:08:13,545 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9849, 'learning_rate': 2.3071267522791883e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0281, 'learning_rate': 2.3068326634643663e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6281, 'learning_rate': 2.3065385746495442e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8101, 'learning_rate': 2.3062444858347222e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6716, 'learning_rate': 2.3059503970199002e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7729, 'learning_rate': 2.3056563082050778e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.5825, 'learning_rate': 2.3053622193902558e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9806, 'learning_rate': 2.305068130575434e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9067, 'learning_rate': 2.3047740417606117e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.772, 'learning_rate': 2.3044799529457897e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9875, 'learning_rate': 2.3041858641309677e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.1841, 'learning_rate': 2.3038917753161456e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9167, 'learning_rate': 2.3035976865013236e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8204, 'learning_rate': 2.3033035976865013e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6498, 'learning_rate': 2.3030095088716792e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7776, 'learning_rate': 2.3027154200568572e-05, 'epoch': 0.8} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.024, 'learning_rate': 2.302421331242035e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7506, 'learning_rate': 2.302127242427213e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8563, 'learning_rate': 2.301833153612391e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6354, 'learning_rate': 2.3015390647975687e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.133, 'learning_rate': 2.3012449759827467e-05, 'epoch': 0.81} + 27%|██████���████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8071, 'learning_rate': 2.300950887167925e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8487, 'learning_rate': 2.3006567983531027e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.968, 'learning_rate': 2.3003627095382806e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0751, 'learning_rate': 2.3000686207234586e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0885, 'learning_rate': 2.2997745319086362e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.209, 'learning_rate': 2.2994804430938146e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7331, 'learning_rate': 2.2991863542789922e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8305, 'learning_rate': 2.29889226546417e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.5465, 'learning_rate': 2.298598176649348e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0401, 'learning_rate': 2.298304087834526e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7231, 'learning_rate': 2.298009999019704e-05, 'epoch': 0.81} +{'loss': 6.787, 'learning_rate': 2.297715910204882e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6321, 'learning_rate': 2.2974218213900597e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6814, 'learning_rate': 2.2971277325752377e-05, 'epoch': 0.81} + 27%|███████████████████▋ | 2856/10701 [3:23:49<4:43:16, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|███████████████████▉ | 2892/10701 [3:25:02<3:40:05, 1.69s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|███████████████████▉ | 2892/10701 [3:25:02<3:40:05, 1.69s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0659, 'learning_rate': 2.2965395549455936e-05, 'epoch': 0.81} +{'loss': 7.0175, 'learning_rate': 2.2962454661307716e-05, 'epoch': 0.81} + 27%|███████████████████▉ | 2892/10701 [3:25:02<3:40:05, 1.69s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +[WARNING|modeling_utils.py:388] 2022-03-02 13:09:47,275 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +[WARNING|modeling_utils.py:388] 2022-03-02 13:09:47,275 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0346, 'learning_rate': 2.2956572885011272e-05, 'epoch': 0.81} +{'loss': 7.1346, 'learning_rate': 2.2953631996863055e-05, 'epoch': 0.81} +[WARNING|modeling_utils.py:388] 2022-03-02 13:09:47,275 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.4166, 'learning_rate': 2.294775022056661e-05, 'epoch': 0.81} +{'loss': 6.6833, 'learning_rate': 2.294480933241839e-05, 'epoch': 0.81} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.3395, 'learning_rate': 2.2941868444270167e-05, 'epoch': 0.81} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9685, 'learning_rate': 2.293892755612195e-05, 'epoch': 0.81} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8702, 'learning_rate': 2.293598666797373e-05, 'epoch': 0.81} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8163, 'learning_rate': 2.2933045779825506e-05, 'epoch': 0.81} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7801, 'learning_rate': 2.2930104891677286e-05, 'epoch': 0.81} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8939, 'learning_rate': 2.292716400352907e-05, 'epoch': 0.81} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7216, 'learning_rate': 2.2924223115380845e-05, 'epoch': 0.81} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.225, 'learning_rate': 2.2921282227232625e-05, 'epoch': 0.81} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7786, 'learning_rate': 2.2918341339084405e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.3385, 'learning_rate': 2.291540045093618e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8075, 'learning_rate': 2.2912459562787964e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9385, 'learning_rate': 2.2909518674639744e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7651, 'learning_rate': 2.290657778649152e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7632, 'learning_rate': 2.29036368983433e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7655, 'learning_rate': 2.2900696010195076e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7538, 'learning_rate': 2.289775512204686e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8935, 'learning_rate': 2.289481423389864e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7903, 'learning_rate': 2.2891873345750416e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9233, 'learning_rate': 2.2888932457602195e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9007, 'learning_rate': 2.2885991569453975e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8274, 'learning_rate': 2.2883050681305755e-05, 'epoch': 0.82} + 27%|████████████████████ | 2898/10701 [3:25:10<2:52:26, 1.33s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|████████████████████▏ | 2922/10701 [3:26:02<4:41:24, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|████████████████████▏ | 2922/10701 [3:26:02<4:41:24, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9414, 'learning_rate': 2.2877168905009314e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2922/10701 [3:26:02<4:41:24, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7455, 'learning_rate': 2.287422801686109e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2922/10701 [3:26:02<4:41:24, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6622, 'learning_rate': 2.2871287128712874e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2922/10701 [3:26:02<4:41:24, 2.17s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9901, 'learning_rate': 2.286540535241643e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0405, 'learning_rate': 2.286246446426821e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.5265, 'learning_rate': 2.2859523576119986e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6398, 'learning_rate': 2.285658268797177e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7795, 'learning_rate': 2.285364179982355e-05, 'epoch': 0.82} + 27%|████████████████���███▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.563, 'learning_rate': 2.2850700911675325e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9061, 'learning_rate': 2.2847760023527105e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0885, 'learning_rate': 2.2844819135378884e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9024, 'learning_rate': 2.2841878247230664e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7052, 'learning_rate': 2.2838937359082444e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8712, 'learning_rate': 2.2835996470934224e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9554, 'learning_rate': 2.2833055582786e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8205, 'learning_rate': 2.2830114694637783e-05, 'epoch': 0.82} + 27%|████████████████████▏ | 2926/10701 [3:26:10<4:36:49, 2.14s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.821, 'learning_rate': 2.2827173806489563e-05, 'epoch': 0.82} +[WARNING|modeling_utils.py:388] 2022-03-02 13:11:20,361 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +[WARNING|modeling_utils.py:388] 2022-03-02 13:11:20,361 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.64, 'learning_rate': 2.282129203019312e-05, 'epoch': 0.82} +[WARNING|modeling_utils.py:388] 2022-03-02 13:11:20,361 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8709, 'learning_rate': 2.2818351142044895e-05, 'epoch': 0.82} +[WARNING|modeling_utils.py:388] 2022-03-02 13:11:20,361 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▎ | 2945/10701 [3:26:45<3:18:47, 1.54s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▎ | 2945/10701 [3:26:45<3:18:47, 1.54s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8367, 'learning_rate': 2.2812469365748458e-05, 'epoch': 0.83} +{'loss': 6.7833, 'learning_rate': 2.2809528477600234e-05, 'epoch': 0.83} + 28%|████████████████████▎ | 2945/10701 [3:26:45<3:18:47, 1.54s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.1066, 'learning_rate': 2.2806587589452014e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.4657, 'learning_rate': 2.2800705813155574e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.5634, 'learning_rate': 2.2797764925007353e-05, 'epoch': 0.83} +{'loss': 6.0922, 'learning_rate': 2.2794824036859133e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.1393, 'learning_rate': 2.279188314871091e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.906, 'learning_rate': 2.278894226056269e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8192, 'learning_rate': 2.2786001372414472e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9494, 'learning_rate': 2.278306048426625e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6233, 'learning_rate': 2.2780119596118028e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2948/10701 [3:26:49<2:49:32, 1.31s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6977, 'learning_rate': 2.2774237819821588e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7606, 'learning_rate': 2.2771296931673367e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7873, 'learning_rate': 2.2768356043525144e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7821, 'learning_rate': 2.2765415155376924e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0452, 'learning_rate': 2.2762474267228703e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9136, 'learning_rate': 2.2759533379080483e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6759, 'learning_rate': 2.2756592490932263e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8175, 'learning_rate': 2.2753651602784042e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8422, 'learning_rate': 2.275071071463582e-05, 'epoch': 0.83} +{'loss': 7.1262, 'learning_rate': 2.27477698264876e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.705, 'learning_rate': 2.274482893833938e-05, 'epoch': 0.83} +{'loss': 6.3581, 'learning_rate': 2.2741888050191158e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.5943, 'learning_rate': 2.2738947162042938e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.0113, 'learning_rate': 2.2736006273894717e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6784, 'learning_rate': 2.2733065385746494e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7618, 'learning_rate': 2.2730124497598277e-05, 'epoch': 0.83} +{'loss': 6.5495, 'learning_rate': 2.2727183609450053e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8643, 'learning_rate': 2.2724242721301833e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.1452, 'learning_rate': 2.2721301833153613e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 7.1207, 'learning_rate': 2.2718360945005392e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7594, 'learning_rate': 2.2715420056857172e-05, 'epoch': 0.83} +{'loss': 6.8055, 'learning_rate': 2.2712479168708952e-05, 'epoch': 0.83} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9028, 'learning_rate': 2.2709538280560728e-05, 'epoch': 0.84} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9214, 'learning_rate': 2.2706597392412508e-05, 'epoch': 0.84} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7777, 'learning_rate': 2.270365650426429e-05, 'epoch': 0.84} +{'loss': 6.7552, 'learning_rate': 2.2700715616116067e-05, 'epoch': 0.84} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.8938, 'learning_rate': 2.2697774727967847e-05, 'epoch': 0.84} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6218, 'learning_rate': 2.2694833839819627e-05, 'epoch': 0.84} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9991, 'learning_rate': 2.2691892951671403e-05, 'epoch': 0.84} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9063, 'learning_rate': 2.2688952063523186e-05, 'epoch': 0.84} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.687, 'learning_rate': 2.2686011175374963e-05, 'epoch': 0.84} + 28%|████████████████████▍ | 2957/10701 [3:27:08<4:44:19, 2.20s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▋ | 2989/10701 [3:28:14<3:56:32, 1.84s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. + 28%|████████████████████▋ | 2989/10701 [3:28:14<3:56:32, 1.84s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9142, 'learning_rate': 2.2680129399078522e-05, 'epoch': 0.84} + 28%|████████████████████▋ | 2989/10701 [3:28:14<3:56:32, 1.84s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.669, 'learning_rate': 2.2677188510930302e-05, 'epoch': 0.84} + 28%|████████████████████▋ | 2989/10701 [3:28:14<3:56:32, 1.84s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.9802, 'learning_rate': 2.267424762278208e-05, 'epoch': 0.84} +{'loss': 7.0606, 'learning_rate': 2.267130673463386e-05, 'epoch': 0.84} + 28%|████████████████████▋ | 2989/10701 [3:28:14<3:56:32, 1.84s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7634, 'learning_rate': 2.2668365846485638e-05, 'epoch': 0.84} + 28%|████████████████████▋ | 2989/10701 [3:28:14<3:56:32, 1.84s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6353, 'learning_rate': 2.2665424958337417e-05, 'epoch': 0.84} + 28%|████████████████████▋ | 2989/10701 [3:28:14<3:56:32, 1.84s/it]g-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.7694, 'learning_rate': 2.26624840701892e-05, 'epoch': 0.84} +[WARNING|modeling_utils.py:388] 2022-03-02 13:13:07,334 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +[WARNING|modeling_utils.py:388] 2022-03-02 13:13:07,334 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.6915, 'learning_rate': 2.2656602293892756e-05, 'epoch': 0.84} +[WARNING|modeling_utils.py:388] 2022-03-02 13:13:07,334 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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. +{'loss': 6.3028, 'learning_rate': 2.2653661405744536e-05, 'epoch': 0.84} +{'loss': 6.3266, 'learning_rate': 2.2650720517596312e-05, 'epoch': 0.84} +[WARNING|modeling_utils.py:388] 2022-03-02 13:13:07,334 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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|trainer.py:2369] 2022-03-02 13:13:11,567 >> Batch size = 4ot estimate the number of tokens of the input, floating-point operations will not be computed-02 13:03:17,353 >> 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.