/MMD-Variational-Autoencoder-Pytorch-InfoVAE

Implementation of the MMD VAE paper (InfoVAE: Information Maximizing Variational Autoencoders) in pytorch

Primary LanguageJupyter Notebook

Pytorch Implementation of MMD Variational Autoencoder


Implementation of the paper InfoVAE: Information Maximizing Variational Autoencoders

The Code has been converted from the TensorFlow implementation by Shengjia Zhao

The results shown are generated by the given pytorch code.

"Details and motivation are described in this paper or tutorial"

Sample generated after 27 epochs of training on MNIST

mnist

Scatter Plot on the distribution of labels in the feature space (for 2 dimensional z)

Before Training

mnist

After Training

mnist