Pytorch implementation of Sobolev GAN https://arxiv.org/abs/1711.04894
Requires PyTorch 0.4+
python gan_toy.py [--options]
G and D are fully connected layers
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)