A minimal implementation of an Variational Auto-Encoder
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
LICENCE
README.md
vae_mnist.py

README.md

Minimal Variational Auto-Encoder

This is a minimal implementation of an Variational Auto-Encoder in Tensorflow applied to MNIST.

Some example generated numbers:

VAE results

How to run

Simply clone the directory and run the file vae_mnist.py. Results will be displayed in real time, while full training takes a few minutes.

Implementation details

The implementation follows Auto-Encoding Variational Bayes. Both the generator and discriminator uses 3 convolutional layers with 5x5 convolutions, with obvious room for improvements.