explosion at 14000/500000 about began-pytorch HOT 7 OPEN

I'm working on this and other issues. Autoencoder which is D trained well but G doesn't. G seems to go to the right path only at the very early steps but when k_t goes near 0, it goes wrong.

from began-pytorch.

After I switched 'z_dim' from 1024 to 124, I guess explosion with default parameter is solved.

from began-pytorch.

Hi! Were you able to recreate the results given in paper?

from began-pytorch.

Not yet. But I will.

from began-pytorch.

https://github.com/rcalland/chainer-BEGAN
Here is another implementation which is done using chainer. It might help you.

from began-pytorch.

@malreddysid Yea. I saw it too but it doesn't have a learning rate schedule which, in my opinion, important to get out of mode collapsing. I guess it's hard to understand gradient flow of pytorch and I'm working on a tensorflow implementation which is better than this implementation and actually show some faces.

from began-pytorch.

Hi! Maybe you could change the L1loss() to tensor,reducemean(tensor.abs(D-X) ).
I have try by this way. The losses seems becoming similar to your tensorflow-version，but the image generated is still bad.

from began-pytorch.