UNSB
UNSB copied to clipboard
cyclomon
Official Repository of "Unpaired Image-to-Image Translation via Neural Schrödinger Bridge" (ICLR 2024)
Thank you for your wonderful work! I recently ran into some questions replicating your model on my own dataset and would like your answers. 1. Limited by the fact that...
Hello, Thanks for your excellent work. I have a question about the inference processing. When I inference, the size of the input images resized 256x256 at the result. I would...
Why does my model stop running after the sixth cycle and does not continue running? Deepseek's reply is as follows. How can I modify it?  
I intend to perform training using grayscale images, where both the input and output are grayscale. Is this model capable of doing that? Thanks!
https://github.com/cyclomon/UNSB/blob/d1f644f7777e19d5afe5aea3e5cb4bd3afd9b88b/models/ncsn_networks.py#L538 Is there a reason why it is `out = layer(x)` instead of `out = layer(out)`?
The work improves on the pytorch_cyclegan_and_pix2pix project, but why can't it also support multi-gpu training?
Hi, this work is very impressive! I was running the test.py script and ran into an issue that, when using my own dataset it could not find ```trainA``` in my...
Thanks for your excellent work. In your code, models/sb_model.py line 194 and line 200 z is a Gaussian noise, Can you help explain what the input z does for netG?...
hello It seems like multi-GPU training isn’t working as expected. Do I need to write separate code for it to function correctly?
Hi, thanks for making this great work. The result is very impressive. I'm trying to reproduce the result of Horse2Zebra and Map2Sat, but I cannot get similar FID values. Is...
