I’m using the GTZAN dataset and am trying to make a CNN music genre classifier. My .wav files are stores in a subdirectory called ‘genres_original’, which is within the root directory, ‘Data’. I’m having two issues: my .wav files aren’t being recognized although I’m certain they are in the right format for PyTorch (.wav) and I have a tensor mismatch issue which I think is related to the wav file problem as well?