/keras-relativistic-gan

Keras (TensorFlow) implementation of the Relativistic Average GAN.

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

Keras (TensorFlow) implementation of RaGAN.

This is an implementation of a Relativistic Average GAN, introduced in "The relativistic discriminator: a key element missing from standard GAN". The original Pytorch code is found here.

Usage

  1. Download the (Fashion)-MNIST *.csv dataset from kaggle, and place in ./data/fashion-mnist_train.csv .
  2. Create the Logdir: $ mkdir ./logs/ .
  3. Run $ python ragan.py to train using the default hyperparameters, or run $ python ragan.py -h for information about the tweakable parameters.

Example

Below is the Fashion-MNIST (DC)RaGAN loss at 100 epochs, along with some example outputs.

Generated clothing Loss