diff --git "a/wandb/run-20220301_171456-1jxdkzs7/files/output.log" "b/wandb/run-20220301_171456-1jxdkzs7/files/output.log" --- "a/wandb/run-20220301_171456-1jxdkzs7/files/output.log" +++ "b/wandb/run-20220301_171456-1jxdkzs7/files/output.log" @@ -4734,3 +4734,704 @@ [INFO|configuration_utils.py:438] 2022-03-01 19:20:34,430 >> Configuration saved in ./checkpoint-1500/config.json [INFO|trainer.py:560] 2022-03-01 19:04:51,340 >> 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. [INFO|feature_extraction_utils.py:324] 2022-03-01 19:20:39,544 >> Configuration saved in ./checkpoint-1500/preprocessor_config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. [INFO|feature_extraction_utils.py:324] 2022-03-01 19:20:39,544 >> Configuration saved in ./checkpoint-1500/preprocessor_config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. +[INFO|feature_extraction_utils.py:324] 2022-03-01 19:20:39,544 >> Configuration saved in ./checkpoint-1500/preprocessor_config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. +[INFO|feature_extraction_utils.py:324] 2022-03-01 19:20:39,544 >> Configuration saved in ./checkpoint-1500/preprocessor_config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. +[INFO|feature_extraction_utils.py:324] 2022-03-01 19:20:39,544 >> Configuration saved in ./checkpoint-1500/preprocessor_config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|█████████████████████████████████████████████████████████████▍ | 1501/1784 [2:07:43<25:19:09, 322.08s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|█████████████████████████████████████████████████████████████▍ | 1501/1784 [2:07:43<25:19:09, 322.08s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|█████████████████████████████████████████████████████████████▍ | 1501/1784 [2:07:43<25:19:09, 322.08s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|█████████████████████████████████████████████████████████████▍ | 1502/1784 [2:07:47<17:44:58, 226.59s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|█████████████████████████████████████████████████████████████▍ | 1502/1784 [2:07:47<17:44:58, 226.59s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|█████████████████████████████████████████████████████████████▍ | 1502/1784 [2:07:47<17:44:58, 226.59s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|█████████████████████████████████████████████████████████████▌ | 1503/1784 [2:07:50<12:28:07, 159.74s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|█████████████████████████████████████████████████████████████▌ | 1503/1784 [2:07:50<12:28:07, 159.74s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|█████████████████████████████████████████████████████████████▌ | 1503/1784 [2:07:50<12:28:07, 159.74s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|██████████████████████████████████████████████████████████████▍ | 1504/1784 [2:07:54<8:47:07, 112.95s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|██████████████████████████████████████████████████████████████▍ | 1504/1784 [2:07:54<8:47:07, 112.95s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|██████████████████████████████████████████████████████████████▍ | 1504/1784 [2:07:54<8:47:07, 112.95s/it]config.jsoner.py:560] 2022-03-01 19:04:51,340 >> 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. + 84%|███████████████████████████████████████████████████████████████▎ | 1505/1784 [2:07:58<6:12:49, 80.18s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 84%|███████████████████████████████████████████████████████████████▎ | 1506/1784 [2:08:02<4:25:09, 57.23s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 84%|███████████████████████████████████████████████████████████████▎ | 1506/1784 [2:08:02<4:25:09, 57.23s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1096, 'learning_rate': 2.196261682242991e-05, 'epoch': 0.84} + 84%|███████████████████████████████████████████████████████████████▎ | 1506/1784 [2:08:02<4:25:09, 57.23s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 84%|███████████████████████████████████████████████████████████████▎ | 1507/1784 [2:08:05<3:10:02, 41.17s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 84%|███████████████████████████████████████████████████████████████▎ | 1507/1784 [2:08:05<3:10:02, 41.17s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 84%|███████████████████████████████████████████████████████████████▎ | 1507/1784 [2:08:05<3:10:02, 41.17s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|███████████████████████████████████████████████████████████████▍ | 1508/1784 [2:08:09<2:17:33, 29.91s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|███████████████████████████████████████████████████████████████▍ | 1508/1784 [2:08:09<2:17:33, 29.91s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|███████████████████████████████████████████████████████████████▍ | 1508/1784 [2:08:09<2:17:33, 29.91s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:22:57,857 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|███████████████████████████████████████████████████████████████▍ | 1509/1784 [2:08:12<1:40:52, 22.01s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|███████████████████████████████████████████████████████████████▍ | 1510/1784 [2:08:16<1:15:19, 16.50s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|███████████████████████████████████████████████████████████████▍ | 1510/1784 [2:08:16<1:15:19, 16.50s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1499, 'learning_rate': 2.1651090342679128e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▏ | 1511/1784 [2:08:20<57:20, 12.60s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▏ | 1511/1784 [2:08:20<57:20, 12.60s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.9418, 'learning_rate': 2.1573208722741434e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▏ | 1511/1784 [2:08:20<57:20, 12.60s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▎ | 1512/1784 [2:08:23<44:44, 9.87s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:23:24,660 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:23:24,660 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.9088, 'learning_rate': 2.1417445482866045e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▎ | 1514/1784 [2:08:30<29:41, 6.60s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▎ | 1514/1784 [2:08:30<29:41, 6.60s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0311, 'learning_rate': 2.133956386292835e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▎ | 1514/1784 [2:08:30<29:41, 6.60s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▍ | 1515/1784 [2:08:33<25:16, 5.64s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:23:34,881 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:23:34,881 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.4708, 'learning_rate': 2.118380062305296e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▍ | 1517/1784 [2:08:40<20:05, 4.52s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▍ | 1517/1784 [2:08:40<20:05, 4.52s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0567, 'learning_rate': 2.1105919003115264e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▍ | 1517/1784 [2:08:40<20:05, 4.52s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▌ | 1518/1784 [2:08:44<18:36, 4.20s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▌ | 1518/1784 [2:08:44<18:36, 4.20s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▌ | 1518/1784 [2:08:44<18:36, 4.20s/it]g-point operations will not be computed-01 19:23:12,421 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▌ | 1519/1784 [2:08:47<17:29, 3.96s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▌ | 1520/1784 [2:08:50<16:35, 3.77s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▌ | 1520/1784 [2:08:50<16:35, 3.77s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0794, 'learning_rate': 2.087227414330218e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▌ | 1520/1784 [2:08:50<16:35, 3.77s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▋ | 1521/1784 [2:08:54<15:57, 3.64s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:23:55,194 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:23:55,194 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.241, 'learning_rate': 2.0716510903426793e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▋ | 1523/1784 [2:09:00<15:04, 3.46s/it]g-point operations will not be computed-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▋ | 1523/1784 [2:09:00<15:04, 3.46s/it]g-point operations will not be computed-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.6959, 'learning_rate': 2.06386292834891e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▋ | 1523/1784 [2:09:00<15:04, 3.46s/it]g-point operations will not be computed-01 19:23:46,993 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▊ | 1524/1784 [2:09:04<14:52, 3.43s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:03,506 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▊ | 1525/1784 [2:09:07<14:30, 3.36s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:03,506 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 85%|█████████████████████████████████████████████████████████████████▊ | 1525/1784 [2:09:07<14:30, 3.36s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:03,506 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0553, 'learning_rate': 2.0482866043613706e-05, 'epoch': 0.85} + 85%|█████████████████████████████████████████████████████████████████▊ | 1525/1784 [2:09:07<14:30, 3.36s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:03,506 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|█████████████████████████████████████████████████████████████████▊ | 1526/1784 [2:09:10<14:12, 3.30s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:09,899 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|█████████████████████████████████████████████████████████████████▉ | 1527/1784 [2:09:13<14:01, 3.27s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:09,899 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|█████████████████████████████████████████████████████████████████▉ | 1527/1784 [2:09:13<14:01, 3.27s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:09,899 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0764, 'learning_rate': 2.0327102803738318e-05, 'epoch': 0.86} + 86%|█████████████████████████████████████████████████████████████████▉ | 1527/1784 [2:09:13<14:01, 3.27s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:09,899 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|█████████████████████████████████████████████████████████████████▉ | 1528/1784 [2:09:17<13:53, 3.25s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:09,899 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|█████████████████████████████████████████████████████████████████▉ | 1528/1784 [2:09:17<13:53, 3.25s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:09,899 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|█████████████████████████████████████████████████████████████████▉ | 1528/1784 [2:09:17<13:53, 3.25s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:09,899 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|█████████████████████████████████████████████████████████████████▉ | 1529/1784 [2:09:20<13:43, 3.23s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:19,441 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|█████████████████████████████████████████████████████████████████▉ | 1529/1784 [2:09:20<13:43, 3.23s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:19,441 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████ | 1530/1784 [2:09:23<13:31, 3.19s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:19,441 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████ | 1530/1784 [2:09:23<13:31, 3.19s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:19,441 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████ | 1530/1784 [2:09:23<13:31, 3.19s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:19,441 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████ | 1531/1784 [2:09:26<13:30, 3.20s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:25,802 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████ | 1532/1784 [2:09:29<13:26, 3.20s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:25,802 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████ | 1532/1784 [2:09:29<13:26, 3.20s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:25,802 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0636, 'learning_rate': 1.9937694704049846e-05, 'epoch': 0.86} + 86%|██████████████████████████████████████████████████████████████████ | 1532/1784 [2:09:29<13:26, 3.20s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:25,802 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▏ | 1533/1784 [2:09:32<13:16, 3.17s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:32,044 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▏ | 1534/1784 [2:09:35<13:02, 3.13s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:32,044 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▏ | 1534/1784 [2:09:35<13:02, 3.13s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:32,044 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1802, 'learning_rate': 1.9781931464174454e-05, 'epoch': 0.86} + 86%|███████████████████████████��██████████████████████████████████████▏ | 1534/1784 [2:09:35<13:02, 3.13s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:32,044 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▎ | 1535/1784 [2:09:38<12:50, 3.09s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:38,031 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▎ | 1536/1784 [2:09:41<12:35, 3.05s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:38,031 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▎ | 1536/1784 [2:09:41<12:35, 3.05s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:38,031 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:24:42,345 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:38,031 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:24:42,345 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:38,031 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1737, 'learning_rate': 1.954828660436137e-05, 'epoch': 0.86} +[WARNING|modeling_utils.py:388] 2022-03-01 19:24:42,345 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:38,031 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▍ | 1538/1784 [2:09:47<12:04, 2.94s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▍ | 1539/1784 [2:09:50<11:48, 2.89s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▍ | 1539/1784 [2:09:50<11:48, 2.89s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:24:50,587 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:24:50,587 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3418, 'learning_rate': 1.9314641744548288e-05, 'epoch': 0.86} +[WARNING|modeling_utils.py:388] 2022-03-01 19:24:50,587 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 86%|██████████████████████████████████████████████████████████████████▌ | 1541/1784 [2:09:55<11:11, 2.76s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:24:55,694 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:24:55,694 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:24:58,011 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:24:58,011 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:00,130 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:00,130 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:02,049 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:02,049 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:03,870 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:03,870 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.8938, 'learning_rate': 1.884735202492212e-05, 'epoch': 0.87} +[WARNING|modeling_utils.py:388] 2022-03-01 19:25:05,444 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:05,444 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:08,181 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:08,181 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:09,927 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:09,927 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.897, 'learning_rate': 1.8535825545171338e-05, 'epoch': 0.87} +[WARNING|modeling_utils.py:388] 2022-03-01 19:25:13,969 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:13,969 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.4322, 'learning_rate': 1.8457943925233644e-05, 'epoch': 0.87} + 87%|██████��███████████████████████████████████████████████████████████▉ | 1552/1784 [2:10:20<10:52, 2.81s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|██████████████████████████████████████████████████████████████████▉ | 1552/1784 [2:10:20<10:52, 2.81s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.294, 'learning_rate': 1.838006230529595e-05, 'epoch': 0.87} + 87%|██████████████████████████████████████████████████████████████████▉ | 1552/1784 [2:10:20<10:52, 2.81s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████ | 1553/1784 [2:10:23<11:53, 3.09s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████ | 1553/1784 [2:10:23<11:53, 3.09s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████ | 1553/1784 [2:10:23<11:53, 3.09s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████ | 1554/1784 [2:10:27<12:33, 3.28s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:28,824 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:25:28,824 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3733, 'learning_rate': 1.8146417445482866e-05, 'epoch': 0.87} + 87%|███████████████████████████████████████████████████████████████████▏ | 1556/1784 [2:10:34<13:07, 3.45s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████▏ | 1556/1784 [2:10:34<13:07, 3.45s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.5383, 'learning_rate': 1.8068535825545172e-05, 'epoch': 0.87} + 87%|███████████████████████████████████████████████████████████████████▏ | 1557/1784 [2:10:38<13:12, 3.49s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████▏ | 1557/1784 [2:10:38<13:12, 3.49s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2512, 'learning_rate': 1.7990654205607478e-05, 'epoch': 0.87} + 87%|███████████████████████████████████████████████████████████████████▏ | 1557/1784 [2:10:38<13:12, 3.49s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████▏ | 1558/1784 [2:10:41<13:16, 3.52s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████▏ | 1558/1784 [2:10:41<13:16, 3.52s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████▏ | 1558/1784 [2:10:41<13:16, 3.52s/it]g-point operations will not be computed-01 19:24:46,591 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████▎ | 1559/1784 [2:10:45<13:15, 3.53s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:45,083 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████▎ | 1560/1784 [2:10:49<13:19, 3.57s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:45,083 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 87%|███████████████████████████████████████████████████████████████████▎ | 1560/1784 [2:10:49<13:19, 3.57s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:45,083 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3184, 'learning_rate': 1.775700934579439e-05, 'epoch': 0.87} + 88%|███████████████████████████████████████████████████████████████████▍ | 1561/1784 [2:10:52<13:16, 3.57s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:45,083 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▍ | 1561/1784 [2:10:52<13:16, 3.57s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:45,083 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3355, 'learning_rate': 1.7679127725856697e-05, 'epoch': 0.88} + 88%|███████████████████████████████████████████████████████████████████▍ | 1561/1784 [2:10:52<13:16, 3.57s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:45,083 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▍ | 1562/1784 [2:10:56<13:12, 3.57s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:45,083 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▍ | 1562/1784 [2:10:56<13:12, 3.57s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:45,083 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▍ | 1562/1784 [2:10:56<13:12, 3.57s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:45,083 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▍ | 1563/1784 [2:10:59<13:02, 3.54s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▌ | 1564/1784 [2:11:03<12:58, 3.54s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▌ | 1564/1784 [2:11:03<12:58, 3.54s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0244, 'learning_rate': 1.7445482866043614e-05, 'epoch': 0.88} + 88%|███████████████████████████████████████████████████████████████████▌ | 1564/1784 [2:11:03<12:58, 3.54s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▌ | 1565/1784 [2:11:06<12:53, 3.53s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:26:07,992 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:26:07,992 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.986, 'learning_rate': 1.7289719626168225e-05, 'epoch': 0.88} + 88%|███████████████████████████████████████████████████████████████████▋ | 1567/1784 [2:11:13<12:41, 3.51s/it]g-point operations will not be computed-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▋ | 1567/1784 [2:11:13<12:41, 3.51s/it]g-point operations will not be computed-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.8743, 'learning_rate': 1.721183800623053e-05, 'epoch': 0.88} + 88%|███████████████████████████████████████████████████████████████████▋ | 1567/1784 [2:11:13<12:41, 3.51s/it]g-point operations will not be computed-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▋ | 1568/1784 [2:11:17<12:32, 3.48s/it]g-point operations will not be computed-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▋ | 1568/1784 [2:11:17<12:32, 3.48s/it]g-point operations will not be computed-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▋ | 1568/1784 [2:11:17<12:32, 3.48s/it]g-point operations will not be computed-01 19:25:59,246 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▋ | 1569/1784 [2:11:20<12:23, 3.46s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▊ | 1570/1784 [2:11:24<12:16, 3.44s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▊ | 1570/1784 [2:11:24<12:16, 3.44s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0877, 'learning_rate': 1.6978193146417445e-05, 'epoch': 0.88} + 88%|███████████████████████████████████████████████████████████████████▊ | 1571/1784 [2:11:27<12:06, 3.41s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▊ | 1571/1784 [2:11:27<12:06, 3.41s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:26:28,385 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:26:28,385 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3854, 'learning_rate': 1.6822429906542056e-05, 'epoch': 0.88} + 88%|███████████████████████████████████████████████████████████████████▉ | 1573/1784 [2:11:34<11:51, 3.37s/it]g-point operations will not be computed-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▉ | 1573/1784 [2:11:34<11:51, 3.37s/it]g-point operations will not be computed-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2533, 'learning_rate': 1.6744548286604362e-05, 'epoch': 0.88} + 88%|███████████████████████████████████████████████████████████████████▉ | 1573/1784 [2:11:34<11:51, 3.37s/it]g-point operations will not be computed-01 19:26:20,071 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|███████████████████████████████████████████████████████████████████▉ | 1574/1784 [2:11:37<11:41, 3.34s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:36,679 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:26:36,679 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:26:36,679 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.8658, 'learning_rate': 1.6588785046728973e-05, 'epoch': 0.88} + [WARNING|modeling_utils.py:388] 2022-03-01 19:26:36,679 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|████████████████████████████████████████████████████████████████████ | 1576/1784 [2:11:43<11:24, 3.29s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:36,679 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|██████████████████████████���█████████████████████████████████████████ | 1576/1784 [2:11:43<11:24, 3.29s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:36,679 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|████████████████████████████████████████████████████████████████████ | 1576/1784 [2:11:43<11:24, 3.29s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:36,679 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|████████████████████████████████████████████████████████████████████ | 1577/1784 [2:11:47<11:13, 3.25s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:46,279 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|████████████████████████████████████████████████████████████████████ | 1578/1784 [2:11:50<11:05, 3.23s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:46,279 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 88%|████████████████████████████████████████████████████████████████████ | 1578/1784 [2:11:50<11:05, 3.23s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:46,279 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2041, 'learning_rate': 1.6355140186915887e-05, 'epoch': 0.88} + 88%|████████████████████████████████████████████████████████████████████ | 1578/1784 [2:11:50<11:05, 3.23s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:46,279 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▏ | 1579/1784 [2:11:53<10:54, 3.20s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:52,494 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▏ | 1580/1784 [2:11:56<10:41, 3.15s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:52,494 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▏ | 1580/1784 [2:11:56<10:41, 3.15s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:52,494 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3845, 'learning_rate': 1.6199376947040498e-05, 'epoch': 0.89} + 89%|████████████████████████████████████████████████████████████████████▏ | 1580/1784 [2:11:56<10:41, 3.15s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:52,494 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▏ | 1581/1784 [2:11:59<10:35, 3.13s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▎ | 1582/1784 [2:12:02<10:24, 3.09s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▎ | 1582/1784 [2:12:02<10:24, 3.09s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:03,067 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:03,067 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0986, 'learning_rate': 1.5965732087227415e-05, 'epoch': 0.89} + 89%|████████████████████████████████████████████████████████████████████▎ | 1584/1784 [2:12:08<10:05, 3.03s/it]g-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▎ | 1584/1784 [2:12:08<10:05, 3.03s/it]g-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:08,900 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:08,900 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0008, 'learning_rate': 1.5809968847352023e-05, 'epoch': 0.89} +[WARNING|modeling_utils.py:388] 2022-03-01 19:27:08,900 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▍ | 1586/1784 [2:12:14<09:43, 2.95s/it]g-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:14,561 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:14,561 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:17,259 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:17,259 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2799, 'learning_rate': 1.557632398753894e-05, 'epoch': 0.89} +[WARNING|modeling_utils.py:388] 2022-03-01 19:27:17,259 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:26:58,607 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▌ | 1589/1784 [2:12:22<09:05, 2.80s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:21,291 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▋ | 1590/1784 [2:12:24<08:48, 2.73s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:21,291 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▋ | 1590/1784 [2:12:24<08:48, 2.73s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:21,291 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:25,031 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:21,291 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:25,031 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:21,291 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:27,405 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:21,291 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:27:27,405 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:21,291 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1098, 'learning_rate': 1.5264797507788163e-05, 'epoch': 0.89} +[WARNING|modeling_utils.py:388] 2022-03-01 19:27:27,405 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:21,291 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▊ | 1593/1784 [2:12:32<07:55, 2.49s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:30,811 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▊ | 1593/1784 [2:12:32<07:55, 2.49s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:30,811 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▊ | 1594/1784 [2:12:34<07:31, 2.38s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:32,799 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▊ | 1594/1784 [2:12:34<07:31, 2.38s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:32,799 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▉ | 1596/1784 [2:12:37<06:28, 2.07s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:34,620 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 89%|████████████████████████████████████████████████████████████████████▉ | 1596/1784 [2:12:37<06:28, 2.07s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:34,620 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|████████████████████████████████████████████████████████████████████▉ | 1597/1784 [2:12:39<05:54, 1.90s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:37,642 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|████████████████████████████████████████████████████████████████████▉ | 1597/1784 [2:12:39<05:54, 1.90s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:37,642 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2481, 'learning_rate': 1.4875389408099688e-05, 'epoch': 0.9} + 90%|█████████████████████████████████████████████████████████████████████ | 1599/1784 [2:12:41<04:52, 1.58s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:38,932 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████ | 1599/1784 [2:12:41<04:52, 1.58s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:38,932 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████ | 1600/1784 [2:12:43<04:53, 1.60s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:40,082 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████ | 1600/1784 [2:12:43<04:53, 1.60s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:40,082 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████ | 1600/1784 [2:12:43<04:53, 1.60s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:43,296 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████ | 1601/1784 [2:12:47<07:05, 2.33s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:43,296 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████ | 1601/1784 [2:12:47<07:05, 2.33s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:43,296 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████ | 1601/1784 [2:12:47<07:05, 2.33s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:47,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▏ | 1602/1784 [2:12:51<08:18, 2.74s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:47,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▏ | 1602/1784 [2:12:51<08:18, 2.74s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:47,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.9512, 'learning_rate': 1.4485981308411214e-05, 'epoch': 0.9} + 90%|█████████████████████████████████████████████████████████████████████▏ | 1602/1784 [2:12:51<08:18, 2.74s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:47,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▏ | 1603/1784 [2:12:54<09:03, 3.00s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:47,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▏ | 1603/1784 [2:12:54<09:03, 3.00s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:47,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▏ | 1603/1784 [2:12:54<09:03, 3.00s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:47,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|██████████████████████████��██████████████████████████████████████████▏ | 1604/1784 [2:12:58<09:35, 3.20s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▎ | 1605/1784 [2:13:02<09:50, 3.30s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▎ | 1605/1784 [2:13:02<09:50, 3.30s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.9243, 'learning_rate': 1.425233644859813e-05, 'epoch': 0.9} + 90%|█████████████████████████████████████████████████████████████████████▎ | 1606/1784 [2:13:05<10:01, 3.38s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▎ | 1606/1784 [2:13:05<10:01, 3.38s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2362, 'learning_rate': 1.4174454828660435e-05, 'epoch': 0.9} + 90%|█████████████████████████████████████████████████████████████████████▎ | 1606/1784 [2:13:05<10:01, 3.38s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▎ | 1607/1784 [2:13:09<10:04, 3.42s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:28:10,191 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:28:10,191 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1772, 'learning_rate': 1.4018691588785047e-05, 'epoch': 0.9} + 90%|█████████████████████████████████████████████████████████████████████▍ | 1609/1784 [2:13:16<10:01, 3.44s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▍ | 1609/1784 [2:13:16<10:01, 3.44s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3123, 'learning_rate': 1.394080996884735e-05, 'epoch': 0.9} + 90%|█████████████████████████████████████████████████████████████████████▍ | 1609/1784 [2:13:16<10:01, 3.44s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▍ | 1610/1784 [2:13:19<09:59, 3.45s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:28:20,523 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:28:20,523 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3045, 'learning_rate': 1.3785046728971962e-05, 'epoch': 0.9} + 90%|█████████████████████████████████████████████████████████████████████▌ | 1612/1784 [2:13:26<09:49, 3.42s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|█████████████████████████████████████████████████████████████████████▌ | 1612/1784 [2:13:26<09:49, 3.42s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.095, 'learning_rate': 1.3707165109034268e-05, 'epoch': 0.9} + 90%|█████████████████████████████████████████████████████████████████████▌ | 1612/1784 [2:13:26<09:49, 3.42s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 90%|███████████████████��█████████████████████████████████████████████████▌ | 1613/1784 [2:13:29<09:45, 3.42s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:28:30,754 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:28:30,754 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.9485, 'learning_rate': 1.3551401869158877e-05, 'epoch': 0.9} + 91%|█████████████████████████████████████████████████████████████████████▋ | 1615/1784 [2:13:36<09:32, 3.39s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|█████████████████████████████████████████████████████████████████████▋ | 1615/1784 [2:13:36<09:32, 3.39s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1645, 'learning_rate': 1.3473520249221183e-05, 'epoch': 0.91} + 91%|█████████████████████████████████████████████████████████████████████▋ | 1615/1784 [2:13:36<09:32, 3.39s/it]g-point operations will not be computed-01 19:27:57,922 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|█████████████████████████████████████████████████████████████████████▋ | 1616/1784 [2:13:39<09:26, 3.37s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|█████████████████████████████████████████████████████████████████████▊ | 1617/1784 [2:13:43<09:20, 3.36s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|█████████████████████████████████████████████████████████████████████▊ | 1617/1784 [2:13:43<09:20, 3.36s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2829, 'learning_rate': 1.3317757009345794e-05, 'epoch': 0.91} + 91%|█████████████████████████████████████████████████████████████████████▊ | 1618/1784 [2:13:46<09:14, 3.34s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|█████████████████████████████████████████████████████████████████████▊ | 1618/1784 [2:13:46<09:14, 3.34s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:28:47,283 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:28:47,283 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1241, 'learning_rate': 1.3161993769470404e-05, 'epoch': 0.91} + 91%|█████████████████████████████████████████████████████████████████████▉ | 1620/1784 [2:13:52<08:59, 3.29s/it]g-point operations will not be computed-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|█████████████████████████████████████████████████████████████████████▉ | 1620/1784 [2:13:52<08:59, 3.29s/it]g-point operations will not be computed-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1715, 'learning_rate': 1.308411214953271e-05, 'epoch': 0.91} + 91%|█████████████████████████████████████████████████████████████████████▉ | 1620/1784 [2:13:52<08:59, 3.29s/it]g-point operations will not be computed-01 19:28:39,120 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|█████████████████████████████████████████████████████████████████████▉ | 1621/1784 [2:13:56<08:52, 3.27s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:28:55,390 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████ | 1622/1784 [2:13:59<08:47, 3.26s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:28:55,390 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████ | 1622/1784 [2:13:59<08:47, 3.26s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:28:55,390 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.869, 'learning_rate': 1.292834890965732e-05, 'epoch': 0.91} + 91%|██████████████████████████████████████████████████████████████████████ | 1622/1784 [2:13:59<08:47, 3.26s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:28:55,390 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████ | 1623/1784 [2:14:02<08:41, 3.24s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:01,751 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████ | 1624/1784 [2:14:05<08:31, 3.20s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:01,751 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████ | 1624/1784 [2:14:05<08:31, 3.20s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:01,751 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.9047, 'learning_rate': 1.277258566978193e-05, 'epoch': 0.91} + 91%|██████████████████████████████████████████████████████████████████████ | 1624/1784 [2:14:05<08:31, 3.20s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:01,751 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▏ | 1625/1784 [2:14:08<08:26, 3.18s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:08,034 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▏ | 1626/1784 [2:14:11<08:21, 3.18s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:08,034 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▏ | 1626/1784 [2:14:11<08:21, 3.18s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:08,034 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.4782, 'learning_rate': 1.2616822429906542e-05, 'epoch': 0.91} + 91%|██████████████████████████████████████████████████████████████████████▏ | 1627/1784 [2:14:15<08:16, 3.16s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:08,034 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▏ | 1627/1784 [2:14:15<08:16, 3.16s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:08,034 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:29:15,760 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:08,034 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:29:15,760 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:08,034 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0248, 'learning_rate': 1.2461059190031153e-05, 'epoch': 0.91} + 91%|██████████████████████████████████████████████████████████████████████▎ | 1629/1784 [2:14:21<08:03, 3.12s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:20,403 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▎ | 1629/1784 [2:14:21<08:03, 3.12s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:20,403 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▎ | 1630/1784 [2:14:24<07:56, 3.09s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:20,403 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▎ | 1630/1784 [2:14:24<07:56, 3.09s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:20,403 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1175, 'learning_rate': 1.2305295950155763e-05, 'epoch': 0.91} + 91%|██████████████████████████████████████████████████████████████████████▍ | 1631/1784 [2:14:27<07:52, 3.09s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:26,485 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▍ | 1631/1784 [2:14:27<07:52, 3.09s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:26,485 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▍ | 1632/1784 [2:14:30<07:44, 3.05s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:26,485 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 91%|██████████████████████████████████████████████████████████████████████▍ | 1632/1784 [2:14:30<07:44, 3.05s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:26,485 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1843, 'learning_rate': 1.2149532710280374e-05, 'epoch': 0.91} + 92%|██████████████████████████████████████████████████████████████████████▍ | 1633/1784 [2:14:33<07:36, 3.03s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:32,374 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▍ | 1633/1784 [2:14:33<07:36, 3.03s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:32,374 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▌ | 1634/1784 [2:14:36<07:27, 2.99s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:32,374 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▌ | 1634/1784 [2:14:36<07:27, 2.99s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:32,374 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0794, 'learning_rate': 1.1993769470404986e-05, 'epoch': 0.92} + 92%|██████████████████████████████████████████████████████████████████████▌ | 1634/1784 [2:14:36<07:27, 2.99s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:32,374 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▌ | 1635/1784 [2:14:38<07:16, 2.93s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:38,019 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▌ | 1636/1784 [2:14:41<07:05, 2.87s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:40,724 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▌ | 1636/1784 [2:14:41<07:05, 2.87s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:40,724 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▋ | 1637/1784 [2:14:44<06:54, 2.82s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:40,724 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▋ | 1637/1784 [2:14:44<06:54, 2.82s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:40,724 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:29:44,615 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:40,724 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:29:44,615 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:40,724 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2488, 'learning_rate': 1.1682242990654207e-05, 'epoch': 0.92} + 92%|██████████████████████████████████████████████████████████████████████▋ | 1639/1784 [2:14:49<06:30, 2.69s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:48,442 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▋ | 1639/1784 [2:14:49<06:30, 2.69s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:48,442 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▊ | 1640/1784 [2:14:51<06:15, 2.61s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:50,774 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▊ | 1640/1784 [2:14:51<06:15, 2.61s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:50,774 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▊ | 1641/1784 [2:14:54<05:59, 2.51s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▊ | 1641/1784 [2:14:54<05:59, 2.51s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▊ | 1642/1784 [2:14:56<05:41, 2.40s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 92%|██████████████████████████████████████████████████████████████████████▊ | 1642/1784 [2:14:56<05:41, 2.40s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:29:56,011 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:29:56,011 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:29:57,866 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:29:57,866 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:00,984 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:00,984 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:02,348 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:02,348 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.5888, 'learning_rate': 1.1059190031152649e-05, 'epoch': 0.92} +{'loss': 3.8661, 'learning_rate': 1.0981308411214954e-05, 'epoch': 0.92} +{'loss': 5.3569, 'learning_rate': 1.0903426791277258e-05, 'epoch': 0.92} +[WARNING|modeling_utils.py:388] 2022-03-01 19:30:04,775 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:04,775 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:06,410 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:06,410 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:06,410 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:10,257 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:10,257 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:10,257 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:13,951 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:13,951 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:13,951 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:29:53,015 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▎ | 1653/1784 [2:15:19<06:26, 2.95s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▍ | 1654/1784 [2:15:23<06:48, 3.14s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▍ | 1654/1784 [2:15:23<06:48, 3.14s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0707, 'learning_rate': 1.043613707165109e-05, 'epoch': 0.93} + 93%|███████████████████████████████████████████████████████████████████████▍ | 1654/1784 [2:15:23<06:48, 3.14s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▍ | 1655/1784 [2:15:27<07:01, 3.26s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▍ | 1655/1784 [2:15:27<07:01, 3.26s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|█████████████████████████████████████��█████████████████████████████████▍ | 1655/1784 [2:15:27<07:01, 3.26s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▍ | 1656/1784 [2:15:30<07:07, 3.34s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▍ | 1656/1784 [2:15:30<07:07, 3.34s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▍ | 1656/1784 [2:15:30<07:07, 3.34s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▌ | 1657/1784 [2:15:34<07:12, 3.40s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:35,331 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:35,331 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2365, 'learning_rate': 1.0124610591900312e-05, 'epoch': 0.93} +[WARNING|modeling_utils.py:388] 2022-03-01 19:30:35,331 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▌ | 1659/1784 [2:15:41<07:12, 3.46s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▌ | 1659/1784 [2:15:41<07:12, 3.46s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▌ | 1659/1784 [2:15:41<07:12, 3.46s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▋ | 1660/1784 [2:15:44<07:06, 3.44s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:45,657 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:45,657 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2778, 'learning_rate': 9.890965732087227e-06, 'epoch': 0.93} +[WARNING|modeling_utils.py:388] 2022-03-01 19:30:45,657 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▋ | 1662/1784 [2:15:51<06:59, 3.44s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▋ | 1662/1784 [2:15:51<06:59, 3.44s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▋ | 1662/1784 [2:15:51<06:59, 3.44s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▊ | 1663/1784 [2:15:54<06:55, 3.43s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:55,891 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:30:55,891 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2197, 'learning_rate': 9.657320872274144e-06, 'epoch': 0.93} +[WARNING|modeling_utils.py:388] 2022-03-01 19:30:55,891 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▊ | 1665/1784 [2:16:01<06:43, 3.39s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▊ | 1665/1784 [2:16:01<06:43, 3.39s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▊ | 1665/1784 [2:16:01<06:43, 3.39s/it]g-point operations will not be computed-01 19:30:19,473 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▉ | 1666/1784 [2:16:04<06:39, 3.39s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▉ | 1667/1784 [2:16:08<06:33, 3.36s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▉ | 1667/1784 [2:16:08<06:33, 3.36s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0015, 'learning_rate': 9.42367601246106e-06, 'epoch': 0.93} + 93%|█████████████████████████████████████████████████████████████████████��█▉ | 1667/1784 [2:16:08<06:33, 3.36s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 93%|███████████████████████████████████████████████████████████████████████▉ | 1668/1784 [2:16:11<06:28, 3.35s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:12,520 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:12,520 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0309, 'learning_rate': 9.267912772585669e-06, 'epoch': 0.94} +[WARNING|modeling_utils.py:388] 2022-03-01 19:31:12,520 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████ | 1670/1784 [2:16:18<06:17, 3.31s/it]g-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:19,044 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:19,044 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2054, 'learning_rate': 9.11214953271028e-06, 'epoch': 0.94} +[WARNING|modeling_utils.py:388] 2022-03-01 19:31:19,044 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▏ | 1672/1784 [2:16:24<06:06, 3.27s/it]g-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▏ | 1672/1784 [2:16:24<06:06, 3.27s/it]g-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▏ | 1672/1784 [2:16:24<06:06, 3.27s/it]g-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▏ | 1673/1784 [2:16:27<06:01, 3.25s/it]g-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:28,625 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:28,625 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3602, 'learning_rate': 8.878504672897196e-06, 'epoch': 0.94} +[WARNING|modeling_utils.py:388] 2022-03-01 19:31:28,625 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▎ | 1675/1784 [2:16:34<05:48, 3.19s/it]g-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:34,891 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:34,891 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0521, 'learning_rate': 8.722741433021807e-06, 'epoch': 0.94} +[WARNING|modeling_utils.py:388] 2022-03-01 19:31:34,891 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▍ | 1677/1784 [2:16:40<05:38, 3.17s/it]g-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:41,150 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:41,150 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1146, 'learning_rate': 8.566978193146417e-06, 'epoch': 0.94} +[WARNING|modeling_utils.py:388] 2022-03-01 19:31:41,150 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▍ | 1679/1784 [2:16:46<05:30, 3.14s/it]g-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:47,290 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:47,290 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.4493, 'learning_rate': 8.411214953271028e-06, 'epoch': 0.94} +[WARNING|modeling_utils.py:388] 2022-03-01 19:31:47,290 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▌ | 1681/1784 [2:16:52<05:17, 3.08s/it]g-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:53,320 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:31:53,320 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.8759, 'learning_rate': 8.255451713395638e-06, 'epoch': 0.94} +[WARNING|modeling_utils.py:388] 2022-03-01 19:31:53,320 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:04,308 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▋ | 1683/1784 [2:16:58<05:05, 3.03s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▋ | 1684/1784 [2:17:01<04:58, 2.99s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 94%|████████████████████████████████████████████████████████████████████████▋ | 1684/1784 [2:17:01<04:58, 2.99s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:02,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:02,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2389, 'learning_rate': 8.021806853582555e-06, 'epoch': 0.94} +[WARNING|modeling_utils.py:388] 2022-03-01 19:32:02,039 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 95%|████████████████████████████████████████████████████████████████████████▊ | 1686/1784 [2:17:07<04:45, 2.92s/it]g-point operations will not be computed-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:07,640 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:07,640 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0019, 'learning_rate': 7.866043613707164e-06, 'epoch': 0.95} +[WARNING|modeling_utils.py:388] 2022-03-01 19:32:07,640 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:31:57,783 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 95%|████████████████████████████████████████████████████████████████████████▊ | 1688/1784 [2:17:12<04:29, 2.81s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:11,692 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 95%|████████████████████████████████████████████████████████████████████████▊ | 1688/1784 [2:17:12<04:29, 2.81s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:11,692 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 95%|████████████████████████████████████████████████████████████████████████▉ | 1689/1784 [2:17:15<04:21, 2.75s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 95%|████████████████████████████████████████████████████████████████████████▉ | 1690/1784 [2:17:17<04:10, 2.67s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 95%|████████████████████████████████████████████████████████████████████████▉ | 1690/1784 [2:17:17<04:10, 2.67s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:17,663 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:17,663 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:19,821 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:19,821 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:21,842 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:21,842 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:23,721 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:23,721 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:25,429 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:25,429 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.4614, 'learning_rate': 7.242990654205607e-06, 'epoch': 0.95} +[WARNING|modeling_utils.py:388] 2022-03-01 19:32:28,446 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:28,446 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:29,750 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:29,750 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.5423, 'learning_rate': 7.009345794392523e-06, 'epoch': 0.95} +[WARNING|modeling_utils.py:388] 2022-03-01 19:32:32,599 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:32,599 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.4201, 'learning_rate': 6.853582554517134e-06, 'epoch': 0.95} +[WARNING|modeling_utils.py:388] 2022-03-01 19:32:36,433 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:36,433 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.5935, 'learning_rate': 6.775700934579439e-06, 'epoch': 0.95} +[WARNING|modeling_utils.py:388] 2022-03-01 19:32:40,135 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:32:40,135 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1363, 'learning_rate': 6.697819314641744e-06, 'epoch': 0.95} + 95%|█████████████████████████████████████████████████████████████████████████▌ | 1703/1784 [2:17:46<04:01, 2.98s/it]g-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 95%|█████████████████████████████████████████████████████████████████████████▌ | 1703/1784 [2:17:46<04:01, 2.98s/it]g-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0374, 'learning_rate': 6.619937694704049e-06, 'epoch': 0.95} + 96%|█████████████████████████████████████████████████████████████████████████▌ | 1704/1784 [2:17:49<04:14, 3.18s/it]g-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|█████████████████████████████████████████████████████████████████████████▌ | 1704/1784 [2:17:49<04:14, 3.18s/it]g-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1211, 'learning_rate': 6.542056074766355e-06, 'epoch': 0.96} + 96%|█████████████████████████████████████████████████████████████████████████▌ | 1704/1784 [2:17:49<04:14, 3.18s/it]g-point operations will not be computed-01 19:32:14,243 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|█████████████████████████████████████████████████████████████████████████▌ | 1705/1784 [2:17:53<04:22, 3.32s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|█████████████████████████████████████████████████████████████████████████▋ | 1706/1784 [2:17:57<04:25, 3.41s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|█████████████████████████████████████████████████████████████████████████▋ | 1706/1784 [2:17:57<04:25, 3.41s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.005, 'learning_rate': 6.386292834890965e-06, 'epoch': 0.96} + 96%|█████████████████████████████████████████████████████████████████████████▋ | 1707/1784 [2:18:00<04:27, 3.47s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|█████████████████████████████████████████████████████████████████████████▋ | 1707/1784 [2:18:00<04:27, 3.47s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0011, 'learning_rate': 6.308411214953271e-06, 'epoch': 0.96} + 96%|█████████████████████████████████████████████████████████████████████████▋ | 1708/1784 [2:18:04<04:24, 3.48s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|█████████████████████████████████████████████████████████████████████████▋ | 1708/1784 [2:18:04<04:24, 3.48s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:05,420 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:05,420 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.5252, 'learning_rate': 6.1526479750778815e-06, 'epoch': 0.96} + 96%|█████████████████████████████████████████████████████████████████████████▊ | 1710/1784 [2:18:11<04:19, 3.50s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|█████████████████████████████████████████████████████████████████████████▊ | 1710/1784 [2:18:11<04:19, 3.50s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3139, 'learning_rate': 6.074766355140187e-06, 'epoch': 0.96} + 96%|█████████████████████████████████████████████████████████████████████████▊ | 1711/1784 [2:18:14<04:15, 3.49s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|█████████████████████████████████████████████████████████████████████████▊ | 1711/1784 [2:18:14<04:15, 3.49s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:15,887 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:15,887 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:15,887 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:19,325 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:19,325 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.4532, 'learning_rate': 5.841121495327103e-06, 'epoch': 0.96} + 96%|████████████████████████████████████████████████���████████████████████████▉ | 1714/1784 [2:18:25<04:02, 3.47s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|█████████████████████████████████████████████████████████████████████████▉ | 1714/1784 [2:18:25<04:02, 3.47s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.054, 'learning_rate': 5.763239875389408e-06, 'epoch': 0.96} + 96%|██████████████████████████████████████████████████████████████████████████ | 1715/1784 [2:18:28<03:58, 3.46s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|██████████████████████████████████████████████████████████████████████████ | 1715/1784 [2:18:28<03:58, 3.46s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:29,552 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:29,552 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.36, 'learning_rate': 5.607476635514019e-06, 'epoch': 0.96} + 96%|██████████████████████████████████████████████████████████████████████████ | 1717/1784 [2:18:35<03:47, 3.40s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|██████████████████████████████████████████████████████████████████████████ | 1717/1784 [2:18:35<03:47, 3.40s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.5952, 'learning_rate': 5.529595015576324e-06, 'epoch': 0.96} + 96%|██████████████████████████████████████████████████████████████████████████ | 1717/1784 [2:18:35<03:47, 3.40s/it]g-point operations will not be computed-01 19:32:52,977 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|██████████████████████████��███████████████████████████████████████████████▏ | 1718/1784 [2:18:38<03:43, 3.39s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|██████████████████████████████████████████████████████████████████████████▏ | 1718/1784 [2:18:38<03:43, 3.39s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|██████████████████████████████████████████████████████████████████████████▏ | 1719/1784 [2:18:41<03:39, 3.37s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|██████████████████████████████████████████████████████████████████████████▏ | 1719/1784 [2:18:41<03:39, 3.37s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|██████████████████████████████████████████████████████████████████████████▏ | 1719/1784 [2:18:41<03:39, 3.37s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 96%|██████████████████████████████████████████████████████████████████████████▏ | 1720/1784 [2:18:45<03:34, 3.36s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:46,228 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:33:46,228 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2021, 'learning_rate': 5.218068535825545e-06, 'epoch': 0.96} + 97%|██████████████████████████████████████████████████████████████████████████▎ | 1722/1784 [2:18:51<03:25, 3.32s/it]g-point operations will not be computed-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▎ | 1722/1784 [2:18:51<03:25, 3.32s/it]g-point operations will not be computed-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2977, 'learning_rate': 5.14018691588785e-06, 'epoch': 0.97} + 97%|██████████████████████████████████████████████████████████████████████████▎ | 1722/1784 [2:18:51<03:25, 3.32s/it]g-point operations will not be computed-01 19:33:37,981 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▎ | 1723/1784 [2:18:55<03:22, 3.32s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▎ | 1723/1784 [2:18:55<03:22, 3.32s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▍ | 1724/1784 [2:18:58<03:17, 3.30s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▍ | 1724/1784 [2:18:58<03:17, 3.30s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▍ | 1724/1784 [2:18:58<03:17, 3.30s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▍ | 1725/1784 [2:19:01<03:13, 3.28s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:02,446 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:02,446 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1123, 'learning_rate': 4.828660436137072e-06, 'epoch': 0.97} + 97%|██████████████████████████████████████████████████████████████████████████▌ | 1727/1784 [2:19:07<03:03, 3.22s/it]g-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▌ | 1727/1784 [2:19:07<03:03, 3.22s/it]g-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:08,742 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:08,742 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2999, 'learning_rate': 4.6728971962616825e-06, 'epoch': 0.97} +[WARNING|modeling_utils.py:388] 2022-03-01 19:34:08,742 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▋ | 1729/1784 [2:19:14<02:53, 3.16s/it]g-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:14,950 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:14,950 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.216, 'learning_rate': 4.517133956386293e-06, 'epoch': 0.97} +[WARNING|modeling_utils.py:388] 2022-03-01 19:34:14,950 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████��███████████████████████████▋ | 1731/1784 [2:19:20<02:45, 3.13s/it]g-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▋ | 1731/1784 [2:19:20<02:45, 3.13s/it]g-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:21,100 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:21,100 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:21,100 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▊ | 1733/1784 [2:19:26<02:37, 3.08s/it]g-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:27,134 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:27,134 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2556, 'learning_rate': 4.205607476635514e-06, 'epoch': 0.97} +[WARNING|modeling_utils.py:388] 2022-03-01 19:34:27,134 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:33:54,489 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▉ | 1735/1784 [2:19:32<02:27, 3.02s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:31,534 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|████████████████████████████████████████████████████████████████████████���█▉ | 1735/1784 [2:19:32<02:27, 3.02s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:31,534 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▉ | 1736/1784 [2:19:35<02:22, 2.97s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:31,534 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▉ | 1736/1784 [2:19:35<02:22, 2.97s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:31,534 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▉ | 1736/1784 [2:19:35<02:22, 2.97s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:31,534 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▉ | 1737/1784 [2:19:38<02:17, 2.93s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:37,148 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|██████████████████████████████████████████████████████████████████████████▉ | 1737/1784 [2:19:38<02:17, 2.93s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:37,148 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 97%|███████████████████████████████████████████████████████████████████████████ | 1738/1784 [2:19:40<02:11, 2.86s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:37,148 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:41,084 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:34:37,148 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:41,084 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:34:37,148 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:43,622 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:34:37,148 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:34:43,622 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:34:37,148 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0528, 'learning_rate': 3.7383177570093455e-06, 'epoch': 0.98} +[WARNING|modeling_utils.py:388] 2022-03-01 19:34:43,622 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:34:37,148 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▏ | 1741/1784 [2:19:48<01:52, 2.62s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:47,207 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▏ | 1741/1784 [2:19:48<01:52, 2.62s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:47,207 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▏ | 1742/1784 [2:19:50<01:45, 2.51s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:49,416 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▏ | 1742/1784 [2:19:50<01:45, 2.51s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:49,416 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▏ | 1743/1784 [2:19:52<01:38, 2.41s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:51,526 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▏ | 1743/1784 [2:19:52<01:38, 2.41s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:51,526 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▎ | 1744/1784 [2:19:54<01:32, 2.31s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:53,522 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▎ | 1744/1784 [2:19:54<01:32, 2.31s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:53,522 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▎ | 1745/1784 [2:19:56<01:25, 2.18s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:55,387 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▎ | 1745/1784 [2:19:56<01:25, 2.18s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:55,387 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▍ | 1747/1784 [2:20:00<01:11, 1.94s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:57,094 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▍ | 1747/1784 [2:20:00<01:11, 1.94s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:57,094 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2075, 'learning_rate': 3.1931464174454827e-06, 'epoch': 0.98} + 98%|███████████████████████████████████████████████████████████████████████████▍ | 1748/1784 [2:20:01<01:04, 1.80s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:58,647 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▍ | 1748/1784 [2:20:01<01:04, 1.80s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:34:58,647 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▍ | 1749/1784 [2:20:02<00:57, 1.64s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:01,229 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▍ | 1749/1784 [2:20:02<00:57, 1.64s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:01,229 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▌ | 1750/1784 [2:20:04<00:56, 1.65s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:01,229 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▌ | 1750/1784 [2:20:04<00:56, 1.65s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:04,339 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▌ | 1750/1784 [2:20:04<00:56, 1.65s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:04,339 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▌ | 1751/1784 [2:20:08<01:16, 2.32s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▌ | 1752/1784 [2:20:12<01:27, 2.74s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▌ | 1752/1784 [2:20:12<01:27, 2.74s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.062, 'learning_rate': 2.8037383177570094e-06, 'epoch': 0.98} + 98%|███████████████████████████████████████████████████████████████████████████▌ | 1752/1784 [2:20:12<01:27, 2.74s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▋ | 1753/1784 [2:20:15<01:33, 3.02s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████���███████████████████████████████████████████████████████████▋ | 1753/1784 [2:20:15<01:33, 3.02s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▋ | 1753/1784 [2:20:15<01:33, 3.02s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▋ | 1754/1784 [2:20:19<01:35, 3.18s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:35:20,707 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:35:20,707 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.2439, 'learning_rate': 2.570093457943925e-06, 'epoch': 0.98} +[WARNING|modeling_utils.py:388] 2022-03-01 19:35:24,301 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:35:24,301 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0405, 'learning_rate': 2.4922118380062308e-06, 'epoch': 0.98} + 98%|███████████████████████████████████████████████████████████████████████████▊ | 1757/1784 [2:20:30<01:32, 3.43s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 98%|███████████████████████████████████████████████████████████████████████████▊ | 1757/1784 [2:20:30<01:32, 3.43s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0753, 'learning_rate': 2.414330218068536e-06, 'epoch': 0.98} + 98%|███████████████████████████████████████████████████████████████████████████▊ | 1757/1784 [2:20:30<01:32, 3.43s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|███████████████████████████████████████████████████████████████████████████▉ | 1758/1784 [2:20:33<01:28, 3.42s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:35:34,596 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:35:34,596 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.3627, 'learning_rate': 2.2585669781931465e-06, 'epoch': 0.99} + 99%|███████████████████████████████████████████████████████████████████████████▉ | 1760/1784 [2:20:40<01:21, 3.39s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|███████████████████████████████████████████████████████████████████████████▉ | 1760/1784 [2:20:40<01:21, 3.39s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:35:41,297 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:35:41,297 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.412, 'learning_rate': 2.102803738317757e-06, 'epoch': 0.99} + 99%|████████████████████████████████████████████████████████████████████████████ | 1762/1784 [2:20:47<01:14, 3.37s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████ | 1762/1784 [2:20:47<01:14, 3.37s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.9167, 'learning_rate': 2.0249221183800623e-06, 'epoch': 0.99} + 99%|████████████████████████████████████████████████████████████████████████████ | 1762/1784 [2:20:47<01:14, 3.37s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████ | 1763/1784 [2:20:50<01:10, 3.36s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████ | 1763/1784 [2:20:50<01:10, 3.36s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████ | 1763/1784 [2:20:50<01:10, 3.36s/it]g-point operations will not be computed-01 19:35:08,058 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▏| 1764/1784 [2:20:53<01:06, 3.34s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▏| 1765/1784 [2:20:56<01:03, 3.32s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▏| 1765/1784 [2:20:56<01:03, 3.32s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.9177, 'learning_rate': 1.7912772585669782e-06, 'epoch': 0.99} + 99%|████████████████████████████████████████████████████████████████████████████▏| 1765/1784 [2:20:56<01:03, 3.32s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|█████████████████████████████████████████████████████████████████��██████████▏| 1766/1784 [2:21:00<00:59, 3.30s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:36:01,011 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:36:01,011 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0202, 'learning_rate': 1.6355140186915887e-06, 'epoch': 0.99} +[WARNING|modeling_utils.py:388] 2022-03-01 19:36:01,011 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▎| 1768/1784 [2:21:06<00:51, 3.24s/it]g-point operations will not be computed-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:36:07,284 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:36:07,284 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.0511, 'learning_rate': 1.4797507788161994e-06, 'epoch': 0.99} +[WARNING|modeling_utils.py:388] 2022-03-01 19:36:07,284 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:35:52,961 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▍| 1770/1784 [2:21:12<00:44, 3.17s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:11,964 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▍| 1771/1784 [2:21:15<00:40, 3.15s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:11,964 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▍| 1771/1784 [2:21:15<00:40, 3.15s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:11,964 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1855, 'learning_rate': 1.32398753894081e-06, 'epoch': 0.99} + 99%|████████████████████████████████████████████████████████████████████████████▍| 1771/1784 [2:21:15<00:40, 3.15s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:11,964 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▍| 1772/1784 [2:21:18<00:37, 3.11s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:11,964 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:36:19,413 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:36:11,964 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:36:19,413 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:36:11,964 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1041, 'learning_rate': 1.1682242990654206e-06, 'epoch': 0.99} +[WARNING|modeling_utils.py:388] 2022-03-01 19:36:19,413 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:36:11,964 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▌| 1774/1784 [2:21:24<00:30, 3.01s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:23,814 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▌| 1775/1784 [2:21:27<00:26, 2.96s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:23,814 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. + 99%|████████████████████████████████████████████████████████████████████████████▌| 1775/1784 [2:21:27<00:26, 2.96s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:23,814 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:36:27,852 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:36:23,814 >> Could not estimate the number of tokens of the input, floating-point operations will not 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] 2022-03-01 19:36:27,852 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:36:23,814 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 4.1502, 'learning_rate': 9.345794392523364e-07, 'epoch': 1.0} +[WARNING|modeling_utils.py:388] 2022-03-01 19:36:27,852 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed-01 19:36:23,814 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▋| 1777/1784 [2:21:32<00:19, 2.78s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:31,680 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▋| 1777/1784 [2:21:32<00:19, 2.78s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:31,680 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▋| 1778/1784 [2:21:35<00:15, 2.66s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:33,917 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▋| 1778/1784 [2:21:35<00:15, 2.66s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:33,917 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▊| 1779/1784 [2:21:37<00:12, 2.50s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:35,891 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▊| 1779/1784 [2:21:37<00:12, 2.50s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:35,891 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▊| 1781/1784 [2:21:40<00:06, 2.09s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:37,603 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▊| 1781/1784 [2:21:40<00:06, 2.09s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:37,603 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▉| 1782/1784 [2:21:42<00:03, 1.88s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:40,396 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +100%|████████████████████████████████████████████████████████████████████████████▉| 1782/1784 [2:21:42<00:03, 1.88s/it][WARNING|modeling_utils.py:388] 2022-03-01 19:36:40,396 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.9512, 'learning_rate': 4.672897196261682e-07, 'epoch': 1.0} +[INFO|trainer.py:2114] 2022-03-01 19:36:42,071 >> Saving model checkpoint to ./=)█| 1784/1784 [2:21:44<00:00, 1.48s/it][INFO|trainer.py:1492] 2022-03-01 19:36:42,069 >> 1,590 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +[INFO|trainer.py:2114] 2022-03-01 19:36:42,071 >> Saving model checkpoint to ./=)█| 1784/1784 [2:21:44<00:00, 1.48s/it][INFO|trainer.py:1492] 2022-03-01 19:36:42,069 >> 1,590 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +{'loss': 3.4805, 'learning_rate': 3.1152647975077885e-07, 'epoch': 1.0} +[INFO|trainer.py:2114] 2022-03-01 19:36:58,542 >> Saving model checkpoint to ./ ./pytorch_model.bin:44<00:00, 1.48s/it][INFO|trainer.py:1492] 2022-03-01 19:36:42,069 >> 1,590 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message. +[INFO|modeling_utils.py:1081] 2022-03-01 19:37:14,780 >> Model weights saved in ./pytorch_model.bin:44<00:00, 1.48s/it][INFO|trainer.py:1492] 2022-03-01 19:36:42,069 >> 1,590 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed.forward` and have been ignored: input_length. If input_length are not expected by `SpeechEncoderDecoderModel.forward`, you can safely ignore this message.