UGATIT-pytorch
UGATIT-pytorch copied to clipboard
znxlwm
Official PyTorch implementation of U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
Fix to [issue #23](https://github.com/znxlwm/UGATIT-pytorch/issues/23) (CUDA out of memory at 1000th epoch) as suggested by @07hyx06. Surrounded print output with torch.no_grad(). Tested by running on an internal large dataset. Before the...
I use unaligned dataset to run this project, but i found the g_loss so large. why? 
Can any one tell the difference between train and test cmds? In case want to process one image alone what are the steps?
According to the appendix section B.2 in the paper, the weights should be initialized from a zero-centered normal distribution with a standard deviation of 0.02. But I did not find...
Dear sir, I am a little confused with 2 CAM loss functions (5) and (6) in your paper. I don't know why one function uses BCE for generator, one uses...
When calling `python main.py --dataset maps --light True --phase test` the model seems to fail to load the checkpoint weights. Any Idea what could be the problem? Last output: ```bash...
Hi, I don't know how to calculate the KID index, can you share the relevant code? Thank you!
For line 169 in UGATIT.py https://github.com/znxlwm/UGATIT-pytorch/blob/b8c4251823673189999484d07e97fdcb9300e9e0/UGATIT.py#L169, why don't we use detach to stop gradient from G, as similar to implementation of pix2pix and cyclegan: https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/f13aab8148bd5f15b9eb47b690496df8dadbab0c/models/pix2pix_model.py#L94
