Implementation of 'lightweight' GAN, proposed in ICLR 2021, in Pytorch. High resolution image generations that can be trained within a day or two
1.63k
stars
222
forks
source link
Regarding implementation of attn layers in discriminator #69
Closed
ferrophile closed 3 years ago
I noticed a slight difference in implementation of generator and discriminator.
Lines 423-424,
Generator.__init__
Lines 569-570,
Discriminator.__init__
Is the latter a typo?
resolution
is tied tolog2(image_size)
and does not change in the for loop.