This is an implementation in TensorFlow to implement the state-of-the-art model presented in the following paper:

https://program.ismir2020.net/poster_4-14.html

This GitHub currently does not include the code I used to set up all the data preparation and augmentation, as this code actually includes automatically downloading and unpacking all datasets I used. Some of these datasets have their own licenses, which I do not want to breach. In addition to this, the code used for training is used at my company, and therefore I cannot make it public. Feel free to message me, and I will give you pointers on how I did it.