GitHub - jiamings/wgan: Tensorflow Implementation of Wasserstein GAN (and Improved version in wgan_v2)

Wasserstein GAN

Tensorflow implementation of Wasserstein GAN.

Two versions:

wgan.py: the original clipping method.
wgan_v2.py: the gradient penalty method. (Improved Training of Wasserstein GANs).

How to run (an example):

python wgan_v2.py --data mnist --model mlp --gpus 0

Name		Name	Last commit message	Last commit date
Latest commit History 9 Commits
lsun		lsun
mnist		mnist
.gitignore		.gitignore
README.md		README.md
layers.py		layers.py
visualize.py		visualize.py
wgan.py		wgan.py
wgan_v2.py		wgan_v2.py