Convolutional Variational Autoencoder

This repository contains a convolutional implementation of the described in Auto-Encoding Variational Bayes. The implemented model uses the MNIST dataset for classification in addition to the ADAM optimizer, batch normalization, weight decay, and ReLU non-linearities.

Dependencies

• Python 3.5 or greater
• Tensorflow 0.12.0 or greater