Sliced-Wasserstein Autoencoder - Pytorch

Implementation of "Sliced-Wasserstein Autoencoder: An Embarrassingly Simple Generative Model" using PyTorch.

Quick Start

This repo requires Python 3.x.

To quickly get started training with the Sliced Wasserstein Autoencoder and run the MNIST example install the swae python package and example dependencies.

1. Pull down this repo and run pip install swae-pytorch/
2. Change directory into the base of the repo and run pip install -r requirements.txt

References

Based on the original Keras implementation by skolouri.