search

bzamecnik / ml

Machine learning projects, often on audio datasets
MIT License
93 stars 39 forks source link

Why are you using binary cross entropy #46

Closed aribornstein closed 6 years ago

aribornstein commented 6 years ago

Hey this is really cool work. Is there are a reason you are using binary cross entropy loss over categorical cross entropy loss given that this seems to be a multiclass classification problem in this sample https://github.com/bzamecnik/ml/blob/master/chord-recognition/lstm_chord_classification_training.py

bzamecnik commented 6 years ago

Hi, it is multi-label classification, ie. N independent binary classifiers, each with sigmoid non-linearity and binary_crossentropy. Softmax would make it dependent and normalize it, so that there's a single class.