- 26 1月, 2023 6 次提交
- 25 1月, 2023 16 次提交
-
-
由 AUTOMATIC1111 提交于
Fix extra network thumbs label color
-
由 AUTOMATIC 提交于
-
由 AUTOMATIC1111 提交于
Fix Approx NN previews changing first generation result
-
由 AUTOMATIC1111 提交于
Add instruct-pix2pix hijack
-
由 Kyle 提交于
Update requirement for transformers to version 4.25.1 to allow instruct-pix2pix demo code to work
-
由 Kyle 提交于
Allows loading instruct-pix2pix models via same method as inpainting models in sd_models.py and sd_hijack_ip2p.py Adds ddpm_edit.py necessary for instruct-pix2pix
-
由 AUTOMATIC1111 提交于
Adds X/Y/Z Grid Script
-
由 AUTOMATIC 提交于
-
由 brkirch 提交于
Adds "Upcast cross attention layer to float32" option in Stable Diffusion settings. This allows for generating images using SD 2.1 models without --no-half or xFormers. In order to make upcasting cross attention layer optimizations possible it is necessary to indent several sections of code in sd_hijack_optimizations.py so that a context manager can be used to disable autocast. Also, even though Stable Diffusion (and Diffusers) only upcast q and k, unfortunately my findings were that most of the cross attention layer optimizations could not function unless v is upcast also.
-
由 brkirch 提交于
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().
-
由 AUTOMATIC 提交于
-
由 AUTOMATIC 提交于
-
由 AUTOMATIC 提交于
add PIP_INSTALLER_LOCATION env var to install pip if it's not installed remove accidental call to accelerate when venv is disabled add another env var to skip venv - SKIP_VENV
-
由 AUTOMATIC 提交于
-
由 AUTOMATIC 提交于
-
由 AUTOMATIC 提交于
-
- 24 1月, 2023 18 次提交
-
-
由 EllangoK 提交于
-
由 EllangoK 提交于
-
由 EllangoK 提交于
-
由 EllangoK 提交于
this is mostly just so git can detect it properly
-
由 AUTOMATIC 提交于
also return the removed field to sdapi/v1/upscalers because someone might have relied on it existing
-
由 AUTOMATIC 提交于
-
由 AUTOMATIC1111 提交于
Add exception handling to API image decode
-
由 AUTOMATIC1111 提交于
Make SwinIR upscaler interruptible and skippable
-
由 AUTOMATIC1111 提交于
Set Linux xformers 0.0.16RC425
-
由 brkirch 提交于
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.
-
由 acncagua 提交于
-
由 catboxanon 提交于
-
由 catboxanon 提交于
-
由 Vladimir Mandic 提交于
-
由 Mykeehu 提交于
Added white color for labels.
-
由 AUTOMATIC 提交于
-
由 AUTOMATIC1111 提交于
Add selector to interrogate categories
-
由 AUTOMATIC 提交于
-