Hi Junho,
I'm confused with the files you put in the DatasetAPI.
Did you mistakely put the wrong code files there?
ZernMern

I tried with latest and got this issue:

ValueError: Variable share_encoder/resblock_0/res1/conv2d/kernel already exists, disallowed. Did you mean to set reuse=True or reuse=tf.AUTO_REUSE in VarScope? Originally defined at:

UNIT-Tensorflow-master/UNIT.py", line 155, in generate_a2b
shared = self.share_encoder(out, self.is_training, reuse=True)

Hello,
thanks for this amazing work.

I have a doubt, I want to understand exactly the difference between cycle gan and UNIT in the loss function.
I read your documentation and mostly understand it, but I am still confused in the differences in the loss functions, especially the cycle loss (is it normal than that of the cycle gan) ?

Hi, Thank you for your share! I use your code to train on my dataset. After training, when I want to test the model using test images, I got this error. Could you please help solve it?

Total size of variables: 68329736
Total bytes of variables: 273318944
[] Reading checkpoints...
[] Success to read UNIT.model-101
[*] Load SUCCESS
Processing A image: ./dataset/face2depth/testA/1.png
Traceback (most recent call last):
File "main_multi_gpu.py", line 108, in
main()
File "main_multi_gpu.py", line 104, in main
gan.test()
File "/home/xxx/UNIT-Tensorflow/UNIT_multi_gpu.py", line 481, in test
save_images(fake_img, [1, 1], image_path)
File "/home/xxx/UNIT-Tensorflow/utils.py", line 103, in save_images
return imsave(inverse_transform(images), size, image_path)
File "/home/xxx/UNIT-Tensorflow/utils.py", line 106, in inverse_transform
return (images+1.) / 2
TypeError: can only concatenate tuple (not "float") to tuple

Hi, thank you for your share. When I run this code, I find that this code can not make use of gpu, even if i run the 'main_multi_gpu.py'

Excuse me, how can I get the dataset of cat species?

hello,when I try to run the code, it shows self.num_batches=0 ,but I cannot find the reason.
self.trainA:[0:0]=[ ]
len=0
What should I do?Thank you very much!

What is the function of gaussian_noise_layer？To add Gaussian noise？