SiyangJ/sobolev_gan

Pytorch implemention of Sobolev GAN (https://arxiv.org/abs/1711.04894)

Python

Pytorch implementation of Sobolev GAN https://arxiv.org/abs/1711.04894

Requires PyTorch 0.4+

Toy problem: Mixture of 8 Gaussians

python gan_toy.py [--options]

G and D are fully connected layers

CIFAR10

python gan_cifar10.py [--options]

G is a Resnet like the one in WGAN-GP paper https://arxiv.org/abs/1704.00028

Generated samples (300+ epochs)