-
```
#Hello author, I have some questions about wgangp. If I use grayscale images as input and change the number of channels to 1, the code will result in the following error：RuntimeError: Given group…
-
Hi, I can get good results with cyclegan in 'lsgan' mode on horse2zebra dataset, but when I change the gan mode to 'wgangp' and follow your codes:
self.loss_gp, gradients = cal_gradient_penalty()...
…
-
Traceback (most recent call last):
File "wgangp_64x64.py", line 444, in
train()
File "wgangp_64x64.py", line 340, in train
_x = inf_train_gen().next()
File "wgangp_64x64.py", l…
-
Hello,
I have a question regarding your implementation of WGAN-GP based pix2pix.
It seems you just used the WGAN loss
```
elif self.gan_mode == 'wgangp':
if target_is_real…
-
As refered in #93 , discriminator is the key to get sharp image. But in my experiments, sometime aeloss get negative values.
:
File "D:\Anaconda\envs\tensorflow\lib\site-packages\tensorflow\python\framework\common_shapes.py", line 686, in _cal…
-
HI, @A03ki
I want to train this f-anoGAN networks with my image datasets.
my directory structure is below,
├── f-AnoGAN
│ ├── your_own_dataset
│ │ ├── dataset
│ │ │ ├── SD_3D_…