Model collapse about bmsg-gan HOT 5 OPEN

Hi guys!
imho, 400 images is an extremely small dataset and mode collapse is expected. I'd suggest trying the data - augmentation suggested in https://arxiv.org/abs/2006.06676 rather than changing the network architecture. There also a very informative video about the paper here.
Hope this helps!

Cheers 🍻!

from bmsg-gan.

I faced the same problem even trying to change the GAN structure, add batch normalization,..
Did you find the answer yet?

from bmsg-gan.

sorry, not yet...................

from bmsg-gan.

I'm facing the same issue while trying to train my own dataset. What is the size of your dataset?

from bmsg-gan.

400 images from LFW dataset. But I'm using celaba dataset now.

from bmsg-gan.