Skip to content
Code for reproducing key results in the paper "InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets"
Python
Branch: master
Clone or download
Pull request Compare This branch is 5 commits ahead, 2 commits behind openai:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
infogan
launchers
tests
.gitignore
README.md
requirements.txt

README.md

InfoGAN

Code for reproducing key results in the paper InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets by Xi Chen, Yan Duan, Rein Houthooft, John Schulman, Ilya Sutskever, Pieter Abbeel.

Dependencies

This fork of the original changes part of the code to accomodate the new TensorFlow API. I am using version 1.3.0. Since this project also used prettytensor, which has it's own problems, you'll need to make a fix to this library as well (instructions below).

In addition, please pip install the following packages:

  • prettytensor
  • progressbar2
  • python-dateutil

Modifying PrettyTensor

You'll likely get an error like ValueError: too many values to unpack. You can fix this by either pulling the relevant commit from the pull request or merely making this trivial change.

Running Experiment

We provide the source code to run the MNIST example:

PYTHONPATH='.' python launchers/run_mnist_exp.py

You can launch TensorBoard to view the generated images:

tensorboard --logdir logs/mnist
You can’t perform that action at this time.