Home

Awesome

hvae-nips

This Code to reproduce the experiments in the Hamiltonian Variational Auto-Encoder (HVAE) paper (available here).

This paper appeared in the proceedings of the 2018 conference on Neural Information Processing Systems (NeurIPS).

Structure

The code is structured into two separate directories:

  1. gaussian for the Gaussian example
  2. mnist for the MNIST example

Requirements

All tests work with Python 3.5.x and 3.6.x (Python 2 functionality not guaranteed). The tests use tensorflow, and have been tested with versions 1.3 and 1.8. Other main requirements include numpy, matplotlib, and scipy.