search

caogang / wgan-gp

A pytorch implementation of Paper "Improved Training of Wasserstein GANs"
MIT License
1.51k stars 346 forks source link

about wgan-gp #41

Open wtmlon opened 5 years ago

wtmlon commented 5 years ago

i rewrite caogang/wgangp code to generate 512*512 images , i add 2-3 conv2d layers in Dnet and Gnet, but when i run that code ,the loss explode to 30000000, and after i train the model 50k epochs with just one photo, the generated photo is still not good.so i want to know is the model or my learning rate(1e-4) cause this condition _20181216165115 _20181216165121

wtmlon commented 5 years ago

@caogang

haonanhe commented 4 years ago

I've been having the same trouble, either. Have you ever work it out?