Music Segment Sequencing with Triplet Network
- Python 3.7
- Pytorch 1.2.0
- Numpy 1.16.4
- Librosa 0.6.3
- Pydub 0.23.1
- Madmom 0.16.1
- Visdom 0.1.8.8
- You can replace the music segments in
audio_datawith your propose audio data. - Run the code
python train.pyto train the model. - Pre-trained model has been saved in
runsfolder.