-
Notifications
You must be signed in to change notification settings - Fork 57
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Request] Custom ESRGAN models #54
Comments
This is already on my radar. Probably supporting custom Core ML models makes more sense because it can support ANE which runs way faster than GPU with only slight sacrifice in quality. |
I don't have any experince with Core ML, will transitioning those odels be relatively simple or will we need entire new models? |
Converting PyTorch (.pth) models to Core ML is relatively simple using coremltools. It requires basic understanding of ML frameworks though. |
Great, whenever the custom model support drops I'll start trying to convert some! Thanks a bunch |
Just gonna link the example documentation page here for anyone else that will look here: https://coremltools.readme.io/docs/pytorch-conversion |
The update to add custom models support is almost done, as well as a conversion script that supports most models from upscale.wiki. Would you like to help with beta testing? If so please send your Apple ID to support@waifu2x.app so that I can add you to a TestFlight beta version. Thanks in advance. |
Yes I would love to :) |
Close as the custom models feature has been released now. |
Close as the custom models feature has been released now. |
An option to choose a custom .pth model file for ESRGAN would be helpful, like being able to use the general Siax model or NMKD’s Jaywreck3 models.
Large collection of models: https://upscale.wiki/wiki/Model_Database
The text was updated successfully, but these errors were encountered: