Age | Commit message (Collapse) | Author |
|
This also handles type casting so that ROCm and MPS torch devices work correctly without --no-half. One cast is required for deepbooru in deepbooru_model.py, some explicit casting is required for img2img and inpainting. depth_model can't be converted to float16 or it won't work correctly on some systems (it's known to have issues on MPS) so in sd_models.py model.depth_model is removed for model.half().
|
|
about it mk2
|
|
about it
|
|
|
|
have relied on it existing
|
|
|
|
The loading of the model for approx nn live previews can change the internal state of PyTorch, resulting in a different image. This can be avoided by preloading the approx nn model in advance.
|
|
|
|
|
|
Add selector to interrogate categories
|
|
|
|
|
|
|
|
Extra network view style
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Fixes various button overflowing UI and compact checkbox
|
|
Fix dark mode
|
|
|
|
|
|
Fixes #7048
Co-Authored-By: J.J. Tolton <jjtolton@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
also fixes type for #tab_extensions in style.css
|
|
|
|
|
|
prevent previews from ruining hypernetwork training
|
|
Adds descriptions for merging methods in UI
|
|
|
|
|
|
Compile and serve js files via `src` instead of embedded inline scripts
|
|
|
|
|
|
|