City
City
Good luck! Just ask here if you need any info about how the interposer works, though that part is fairly simple. One thing which might cause issues is how latents...
Hi! The current version is just a quick proof-of-concept for testing. I will work on it more over the weekend when I have time again. Thank you for your patience.
Hello again. As promised, I've made some fixes to the code. Could you please update and test it again to see if the inconsistencies are still present?
It looks like I was missing an import in that file. Can you update and test again? Also, here is the workflow from my image if it helps. The normal...
Figured you'd appreciate being linked to, didn't really feel right naming my repo "advanced noise" when it isn't really, well, advanced. >I was visiting your repository, wondering if there would...
Actually, wait, it doesn't even need ESRGAN. I can literally get a working upscaler with a 2MB model lol. ```python module_list = [ nn.Conv2d(4, 64, kernel_size=5, padding=2), nn.ReLU(), nn.Upsample(scale_factor=2.0), nn.ReLU(),...
>WAIT WHAT?! How do you do that? I have yet to know how to do such things. I just changed the training code from my latent interposer to take v1...
>And it is always possible to try to loop back to get bigger sizes. 4x is pushing it, my training code/dataset is probably just awful. Though you do make a...
Greetings. I finished the models/repo. It's available here: https://github.com/city96/SD-Latent-Upscaler I have models for x1.25, x1.5 and x2.0, for both XL and v1.5. I teseted chaining two of the 2x ones...
No problem. These models are still relatively undertrained but from my testing they seems to work OK. Could've trained them longer but I just set them to a fixed epoch...