Repository files navigation

Code for the paperBanach Wasserstein GAN.

TraditionalWGAN uses an approximation of the Wasserstein metric to opimize the generator. This Wasserstein metric in turn depends upon an underlying metric onimages which is taken to be the norm

The article extends the theory ofWGAN-GP to anyBanach space, while this code can be used to train WGAN over anySobolev space with norm

The parametersp can be used to control the focus on outliers, with highp indicating a strong focus on the worst offenders.s can be used to control focus on small/large scale behaviour, where negatives indicates focus on large scales, while positives indicates focus on small scales (e.g. edges).

Inception scores for the spaces and:

The code has some dependencies that can be easily installed

$ pip install https://github.com/adler-j/tensordata/archive/master.zip$ pip install https://github.com/adler-j/adler/archive/master.zip

You also need a recent version of tensorflow in order to use thetf.contrib.gan functionality.