WinodePino

I've installed the wavelet branch and added the configure wavelet loss from above to my config.yaml. But when I start the training I do not see anything back in the...

> @WinodePino During the startup process there will be a section that is like > > ``` > Wavelet Loss: > Level: 2 > Alpha: 1.0 > Transform: swt >...

Thanks for you help! At the moment version 2.7.1, but I also tried 2.4 and 2.5 I think. ` INFO Loaded Flux: flux_utils.py:137Traceback (most recent call last): File "B:\02_AI\sd-scripts-network-wavelet-loss\sd-scripts-network-wavelet-loss\flux_train_network.py", line...

> it's saying `torch.float8_e4m3fnuz` isn't a attribute which is likely a pytorch version issue. With the venv enabled do > > ``` > python -c "import torch; print(torch.__version__)" > ```...

Should wavelet_loss_type replace loss_type and should I remove it or not? My results so far are letting me generate on higher resolution and it looks nice, but I expected it...

> Found this PR finally. I hope you don’t mind if I share something relevant to the topic. > > A guy from an one forum made this wavelet implementation...

> @WinodePino > > > I am trying to test this, but somehow my caching latents gets stuck at 0%, any idea why this could be? > > I’m not...