ProGamerGov

Related Issues: https://github.com/jzbontar/mc-cnn/issues/16 https://github.com/allenai/XNOR-Net/issues/22 https://github.com/soumith/dcgan.torch/issues/67 https://github.com/facebook/fb.resnet.torch/issues/153

After using `cudnn.verbose = true`, it seems that it may be a lack of memory issue after all: https://gist.github.com/ProGamerGov/9e5b367a90cd4be9cbd1ed023dafbb81 I thought I could go a lot higher in terms of...

So it turns out that I needed a content image in addition to the content mask: ``` python get_mask_hdf5.py --n_colors=5 --style_image=data/Monet2/style.png --style_mask=data/Monet2/style_mask_fix.png --target_mask=data/Monet2/monetblank_mask.png ``` I added an image that was...

That actually wasn't a fix. It's still broken.

My style and content masks have the exact same number of colors, and the exact same colors. Here at the input images that I'm trying to use: The Style Image:...

@DmitryUlyanov Do you have any ideas on what is causing this odd issue, and how to resolve it?

@Bird-NZ That's a very tricky issue I think, as you are exploring largely uncharted territory. Are you only training with one class? And are you visualization specific channels? I have...

@Bird-NZ I'm not sure what framework you would use to make it easier? PyTorch Lightning might be more user friendly, but idk how easy it would be to setup the...

@qomiter You need a GPU that supports ROCm and you need to install the ROCm supporting version of PyTorch: https://pytorch.org/blog/pytorch-for-amd-rocm-platform-now-available-as-python-package/

The summer model was accidentally trained with the wrong colorspace, and the others had a layer in the wrong place (both issues have been fixed since then). So I'm not...