/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") 2023-10-10 17:26:42 | WARNING | root | Pytorch pre-release version 2.1.0.dev20230831+cu118 - assuming intent to test it /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') [Featurizer] - The selected feature hidden_states's downsample rate is 960 [Runner] - Start a new experiment mert_test_8x_400k_4196_encode {'model_config': None, 'refresh': False} using suno's k-means clusters: mert_test_8x_400k_4196_encode load local model succeed load local feat extractor Wav2Vec2FeatureExtractor { "do_normalize": false, "feature_extractor_type": "Wav2Vec2FeatureExtractor", "feature_size": 1, "padding_side": "right", "padding_value": 0, "return_attention_mask": true, "sampling_rate": 24000 } overall: 0%| | 0/4000 [00:00 main() File "/home/tony/Work/glockenspiel/s3prl/s3prl/run_downstream.py", line 215, in main eval(f'runner.{args.mode}')() File "/home/tony/Work/glockenspiel/s3prl/s3prl/downstream/runner.py", line 292, in train features = self.upstream.model(wavs) File "/home/tony/anaconda3/envs/suno_env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1518, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/tony/anaconda3/envs/suno_env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1527, in _call_impl return forward_call(*args, **kwargs) File "/home/tony/Work/glockenspiel/s3prl/s3prl/upstream/mert/expert.py", line 185, in forward output_values = self.model(**input_values, output_hidden_states=True) File "/home/tony/anaconda3/envs/suno_env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1518, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/tony/anaconda3/envs/suno_env/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1527, in _call_impl return forward_call(*args, **kwargs) File "/home/tony/Work/glockenspiel/suno_utils/suno_utils/models/mert/modeling_MERT.py", line 173, in forward attention_mask = self._get_feature_vector_attention_mask( File "/home/tony/anaconda3/envs/suno_env/lib/python3.10/site-packages/transformers/models/hubert/modeling_hubert.py", line 926, in _get_feature_vector_attention_mask attention_mask = attention_mask.flip([-1]).cumsum(-1).flip([-1]).bool() KeyboardInterrupt overall: 57%|█████▋ | 2285/4000 [26:03<19:33, 1.46it/s]