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.

Downloading Datasets

Line or Rectangles datasets are available on s3.

# Single Line. Foreground Noise. (14.5 mb)
curl http://whiskey-ginger-analytics-public.s3.amazonaws.com/datasets/BASICPROP.zip | tar -xf- -C ./

# Pair of Rectangles. (579 kb)
curl http://whiskey-ginger-analytics-public.s3.amazonaws.com/datasets/BASICPROP-angle.zip | tar -xf- -C ./

# Pair of Rectangles. Foreground Noise. (53.7 mb)
curl http://whiskey-ginger-analytics-public.s3.amazonaws.com/datasets/BASICPROP-angle-noise.zip | tar -xf- -C ./

# Pair of Rectangles. FG + Background Noise. (76.1 mb)
curl http://whiskey-ginger-analytics-public.s3.amazonaws.com/datasets/BASICPROP-angle-noise-bg.zip | tar -xf- -C ./

Dependencies

This project currently requires an antiquated version of tensorflow. For Mac:

pip install -U https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0-py2-none-any.whl

In addition, please pip install the following packages:

prettytensor
progressbar
python-dateutil

Running in Docker

$ git clone git@github.com:openai/InfoGAN.git
$ docker run -v $(pwd)/InfoGAN:/InfoGAN -w /InfoGAN -it -p 8888:8888 gcr.io/tensorflow/tensorflow:r0.9rc0-devel
root@X:/InfoGAN# pip install -r requirements.txt
root@X:/InfoGAN# python launchers/run_mnist_exp.py

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

License

MIT

kevtimova / InfoGAN