/home/tony/Work/glockenspiel/s3prl/s3prl/upstream/byol_s/byol_a/common.py:20: UserWarning: torchaudio._backend.set_audio_backend has been deprecated. With dispatcher enabled, this function is no-op. You can remove the function call. torchaudio.set_audio_backend("sox_io") /home/tony/Work/glockenspiel/s3prl/s3prl/run_downstream.py:157: UserWarning: torchaudio._backend.set_audio_backend has been deprecated. With dispatcher enabled, this function is no-op. You can remove the function call. torchaudio.set_audio_backend('sox_io') Some weights of the model checkpoint at m-a-p/MERT-v1-330M were not used when initializing MERTModel: ['encoder.pos_conv_embed.conv.weight_v', 'encoder.pos_conv_embed.conv.weight_g'] - This IS expected if you are initializing MERTModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing MERTModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of MERTModel were not initialized from the model checkpoint at m-a-p/MERT-v1-330M and are newly initialized: ['encoder.pos_conv_embed.conv.parametrizations.weight.original1', 'encoder.pos_conv_embed.conv.parametrizations.weight.original0'] You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. [Featurizer] - The selected feature hidden_states's downsample rate is 320 [Runner] - Start a new experiment m-a-p/MERT-v1-330M {'model_config': None, 'refresh': False} overall: 0%| | 0/4000 [00:00 phonemes: 0%| | 0/3515 [00:00 phonemes: 13%|█▎ | 452/3515 [00:00<00:00, 4517.68it/s] word -> phonemes: 26%|██▌ | 904/3515 [00:00<00:00, 4468.11it/s] word -> phonemes: 40%|███▉ | 1397/3515 [00:00<00:00, 4675.09it/s] word -> phonemes: 53%|█████▎ | 1880/3515 [00:00<00:00, 4732.32it/s] word -> phonemes: 67%|██████▋ | 2354/3515 [00:00<00:00, 4582.34it/s] word -> phonemes: 80%|████████ | 2814/3515 [00:00<00:00, 4568.41it/s] word -> phonemes: 93%|█████████▎| 3275/3515 [00:00<00:00, 4577.38it/s] word -> phonemes: 100%|██████████| 3515/3515 [00:00<00:00, 4583.20it/s] train: 0%| | 0/3515 [00:00 phonemes: 0%| | 0/2703 [00:00 phonemes: 15%|█▍ | 395/2703 [00:00<00:00, 3947.04it/s] word -> phonemes: 36%|███▌ | 966/2703 [00:00<00:00, 4979.62it/s] word -> phonemes: 58%|█████▊ | 1558/2703 [00:00<00:00, 5407.39it/s] word -> phonemes: 80%|███████▉ | 2154/2703 [00:00<00:00, 5620.01it/s] word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 5417.88it/s] dev: 0%| | 0/2703 [00:00 phonemes: 0%| | 0/2703 [00:00 phonemes: 20%|█▉ | 535/2703 [00:00<00:00, 5339.46it/s] word -> phonemes: 40%|███▉ | 1069/2703 [00:00<00:00, 4003.84it/s] word -> phonemes: 59%|█████▉ | 1599/2703 [00:00<00:00, 4502.86it/s] word -> phonemes: 81%|████████ | 2186/2703 [00:00<00:00, 4986.43it/s] word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 4896.80it/s] dev: 0%| | 0/2703 [00:00 phonemes: 0%| | 0/2703 [00:00 phonemes: 21%|██ | 567/2703 [00:00<00:00, 5667.24it/s] word -> phonemes: 43%|████▎ | 1172/2703 [00:00<00:00, 5886.70it/s] word -> phonemes: 66%|██████▌ | 1777/2703 [00:00<00:00, 5958.01it/s] word -> phonemes: 88%|████████▊ | 2373/2703 [00:00<00:00, 5882.95it/s] word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 5875.08it/s] dev: 0%| | 0/2703 [00:00 phonemes: 0%| | 0/3515 [00:00 phonemes: 2%|▏ | 67/3515 [00:00<00:11, 297.35it/s] word -> phonemes: 4%|▍ | 139/3515 [00:00<00:10, 311.78it/s] word -> phonemes: 15%|█▍ | 521/3515 [00:00<00:02, 1266.97it/s] word -> phonemes: 28%|██▊ | 982/3515 [00:00<00:01, 2202.25it/s] word -> phonemes: 42%|████▏ | 1463/3515 [00:00<00:00, 2947.00it/s] word -> phonemes: 55%|█████▍ | 1917/3515 [00:00<00:00, 3407.98it/s] word -> phonemes: 68%|██████▊ | 2384/3515 [00:00<00:00, 3776.15it/s] word -> phonemes: 81%|████████ | 2853/3515 [00:01<00:00, 4043.23it/s] word -> phonemes: 94%|█████████▍| 3297/3515 [00:01<00:00, 4159.09it/s] word -> phonemes: 100%|██████████| 3515/3515 [00:01<00:00, 2926.76it/s] train: 0%| | 0/3515 [00:00 phonemes: 0%| | 0/2703 [00:00 phonemes: 22%|██▏ | 585/2703 [00:00<00:00, 5846.98it/s] word -> phonemes: 44%|████▍ | 1194/2703 [00:00<00:00, 5986.69it/s] word -> phonemes: 67%|██████▋ | 1800/2703 [00:00<00:00, 6016.38it/s] word -> phonemes: 89%|████████▉ | 2402/2703 [00:00<00:00, 6000.78it/s] word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 5942.39it/s] dev: 0%| | 0/2703 [00:00