How to obtain 2-d feature for kde diagram?

Hi there,

Thanks much for sharing the code and great work!
Quick question on the features used for kde - which method was used to convert the high-dimensional features after glocal average pooling for the gaussian kde? Is this TSNE or some other method? 

Thanks again!
YS