igul222 / improved_wgan_training

Code for reproducing experiments in "Improved Training of Wasserstein GANs"
MIT License
2.35k stars 669 forks source link

I want to recurrent this Method with torch #7

Closed mathfinder closed 7 years ago

mathfinder commented 7 years ago

I want to recurrent this Method with torch.But I don't no how to compute the gradient of 'the penalty of gradient' to update D's parameters. Any suggestions?

igul222 commented 7 years ago

I don't believe there's currently a way of implementing it in torch unless they add functionality, unfortunately.