MusicTaggerCRNN model is added in keras.application for tagging or feature extract

My convolutional recurrent neural network-based music tagger, that is part of music-auto_tagging-keras is added as keras.multiplication.

This means you can easily build the model and load pre-trained weights as below.


Music-tagging is explained in my papers (1, 2). Let me explain about music feature extractor. I copy&paste from my repo.

Which is the better feature extractor?

By setting include_top=False, you can get 256-dim (MusicTaggerCNN) or 32-dim (MusicTaggerCRNN) feature representation.

In general, I would recommend to use MusicTaggerCRNN and 32-dim feature as for predicting 50 tags, 256 features actually sound bit too large. I haven’t looked into 256-dim feature but only 32-dim features. I thought of using PCA to reduce the dimension more, but ended up not applying it because mean(abs(recovered - original) / original) are .12(dim: 32->16), .05 (dim: 32->24) – which don’t seem good enough.

Probably the 256-dim features are redundant (which then you can reduce them down effectively with PCA), or they just include more information than 32-dim ones (e.g., features in different hierarchical levels). If the dimension size would not matter, it’s worth choosing 256-dim ones.

tl;dr. setting include_top=False and get a feature extractor that outputs the second last node activation of the network.

Details are in keras documentation and music-auto_tagging-keras.





One thought on “MusicTaggerCRNN model is added in keras.application for tagging or feature extract

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s