/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") 2023-10-11 20:56:50 | 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_encode {'model_config': None, 'refresh': False} Loading model from Local mert_test_8x_400k_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 } using suno's k-means clusters: mert_test_8x_400k_encode overall: 0%| | 0/4000 [00:00 phonemes: 0%| | 0/3515 [00:00 phonemes: 10%|▉ | 339/3515 [00:00<00:00, 3368.22it/s] word -> phonemes: 20%|██ | 720/3515 [00:00<00:00, 3623.56it/s] word -> phonemes: 31%|███ | 1083/3515 [00:00<00:00, 3456.77it/s] word -> phonemes: 42%|████▏ | 1463/3515 [00:00<00:00, 3587.56it/s] word -> phonemes: 52%|█████▏ | 1823/3515 [00:00<00:00, 3417.21it/s] word -> phonemes: 63%|██████▎ | 2209/3515 [00:00<00:00, 3560.56it/s] word -> phonemes: 73%|███████▎ | 2571/3515 [00:00<00:00, 3577.88it/s] word -> phonemes: 83%|████████▎ | 2931/3515 [00:00<00:00, 3542.67it/s] word -> phonemes: 95%|█████████▍| 3327/3515 [00:00<00:00, 3669.00it/s] word -> phonemes: 100%|██████████| 3515/3515 [00:00<00:00, 3570.72it/s] train: 0%| | 0/3515 [00:00 phonemes: 0%| | 0/2703 [00:00 phonemes: 15%|█▌ | 412/2703 [00:00<00:00, 4116.05it/s] word -> phonemes: 31%|███ | 838/2703 [00:00<00:00, 4199.05it/s] word -> phonemes: 47%|████▋ | 1258/2703 [00:00<00:00, 3997.93it/s] word -> phonemes: 62%|██████▏ | 1666/2703 [00:00<00:00, 4028.23it/s] word -> phonemes: 78%|███████▊ | 2097/2703 [00:00<00:00, 4119.65it/s] word -> phonemes: 93%|█████████▎| 2510/2703 [00:00<00:00, 4053.55it/s] word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 4069.45it/s] dev: 0%| | 0/2703 [00:00 phonemes: 0%| | 0/2703 [00:00 phonemes: 18%|█▊ | 494/2703 [00:00<00:00, 4937.39it/s] word -> phonemes: 37%|███▋ | 988/2703 [00:00<00:00, 2032.44it/s] word -> phonemes: 59%|█████▉ | 1592/2703 [00:00<00:00, 3061.73it/s] word -> phonemes: 81%|████████ | 2192/2703 [00:00<00:00, 3850.27it/s] word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 3688.52it/s] dev: 0%| | 0/2703 [00:00 phonemes: 0%| | 0/2703 [00:00 phonemes: 22%|██▏ | 588/2703 [00:00<00:00, 5876.12it/s] word -> phonemes: 44%|████▍ | 1190/2703 [00:00<00:00, 5958.31it/s] word -> phonemes: 67%|██████▋ | 1809/2703 [00:00<00:00, 6063.91it/s] word -> phonemes: 89%|████████▉ | 2416/2703 [00:00<00:00, 6000.49it/s] word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 5961.35it/s] dev: 0%| | 0/2703 [00:00 phonemes: 0%| | 0/3515 [00:00 phonemes: 12%|█▏ | 413/3515 [00:00<00:00, 4123.98it/s] word -> phonemes: 23%|██▎ | 826/3515 [00:00<00:00, 3960.58it/s] word -> phonemes: 35%|███▍ | 1223/3515 [00:00<00:01, 2145.90it/s] word -> phonemes: 49%|████▊ | 1706/3515 [00:00<00:00, 2832.88it/s] word -> phonemes: 63%|██████▎ | 2202/3515 [00:00<00:00, 3401.92it/s] word -> phonemes: 76%|███████▌ | 2658/3515 [00:00<00:00, 3722.47it/s] word -> phonemes: 89%|████████▉ | 3121/3515 [00:00<00:00, 3979.03it/s] word -> phonemes: 100%|██████████| 3515/3515 [00:00<00:00, 3567.65it/s] train: 0%| | 0/3515 [00:00 phonemes: 0%| | 0/2703 [00:00 phonemes: 21%|██ | 555/2703 [00:00<00:00, 5546.29it/s] word -> phonemes: 43%|████▎ | 1158/2703 [00:00<00:00, 5830.67it/s] word -> phonemes: 65%|██████▌ | 1760/2703 [00:00<00:00, 5915.72it/s] word -> phonemes: 87%|████████▋ | 2352/2703 [00:00<00:00, 5907.13it/s] word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 5835.76it/s] dev: 0%| | 0/2703 [00:00