aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* use python importlib to load and execute extension modulesMax Audron2023-01-271-6/+4
| | | | | | | | | | | | | | previously module attributes like __file__ where not set correctly, leading to scripts getting the directory of the stable-diffusion repo location instead of their own script. This causes problem when loading user data from an external location using the --data-dir flag, as extensions would look for their own code in the stable-diffusion repo location instead of the data dir location. Using pythons importlib functions sets the modules specs correctly and executes them. But this will break extensions if they build paths based on the previously incorrect __file__ attribute.
* add data-dir flag and set all user data directories based on itMax Audron2023-01-2714-31/+39
|
* clarify the option to disable NaN check.AUTOMATIC2023-01-271-0/+2
|
* support detecting midas modelAUTOMATIC2023-01-273-8/+11
| | | | fix broken api for checkpoint list
* remove the need to place configs near modelsAUTOMATIC2023-01-2710-151/+360
|
* add an option to enable sections from extras tab in txt2img/img2imgAUTOMATIC2023-01-269-23/+133
| | | | fix some style inconsistenices
* Merge pull request #7234 from brkirch/fix-full-previewsAUTOMATIC11112023-01-262-5/+5
|\ | | | | Fix full previews and--no-half-vae to work correctly with --upcast-sampling
| * Fix full previews, --no-half-vaebrkirch2023-01-262-5/+5
|/
* fix prompt editing break after first batch in img2imgAUTOMATIC2023-01-251-1/+1
|
* add edit_image_conditioning from my earlier edits in case there's an attempt ↵AUTOMATIC2023-01-251-1/+9
| | | | | | to inegrate pix2pix properly this allows to use pix2pix model in img2img though it won't work well this way
* Merge pull request #7197 from mcmonkey4eva/fix-ti-symlinksAUTOMATIC11112023-01-251-1/+1
|\ | | | | allow symlinks in the textual inversion embeddings folder
| * allow symlinks in the textual inversion embeddings folderAlex "mcmonkey" Goodwin2023-01-251-1/+1
| |
* | Merge pull request #7201 from brkirch/update-macos-defaultsAUTOMATIC11112023-01-251-1/+1
|\ \ | | | | | | Update default Mac command line arguments to use --upcast-sampling instead of --no-half
| * | Update arguments to use --upcast-samplingbrkirch2023-01-251-1/+1
|/ /
* | fix for unet hijack breaking the train tabAUTOMATIC2023-01-251-2/+5
| |
* | make clicking extra networks button one more time close the extra networks UIAUTOMATIC2023-01-251-2/+7
|/
* add pix2pix creditsAUTOMATIC2023-01-251-1/+2
|
* Merge pull request #6510 from brkirch/unet16-upcast-precisionAUTOMATIC11112023-01-2510-71/+189
|\ | | | | Add upcast options, full precision sampling from float16 UNet and upcasting attention for inference using SD 2.1 models without --no-half
| * Add UI setting for upcasting attention to float32brkirch2023-01-255-64/+108
| | | | | | | | | | | | 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.
| * Add option for float32 sampling with float16 UNetbrkirch2023-01-258-8/+82
| | | | | | | | 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().
* | Merge pull request #7138 from mykeehu/patch-4AUTOMATIC11112023-01-251-0/+1
|\ \ | | | | | | Fix extra network thumbs label color
| * | Fix extra network thumbs label colorMykeehu2023-01-231-0/+1
| | | | | | | | | Added white color for labels.
* | | change to code for live preview fix on OSX to be bit more obviousAUTOMATIC2023-01-251-2/+2
| | |
* | | Merge pull request #7151 from brkirch/fix-approx-nnAUTOMATIC11112023-01-251-1/+5
|\ \ \ | | | | | | | | Fix Approx NN previews changing first generation result
| * | | Fix different first gen with Approx NN previewsbrkirch2023-01-241-1/+5
| |/ / | | | | | | | | | 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.
* | | Merge pull request #7195 from Klace/instruct-pix2pix_model_loadAUTOMATIC11112023-01-255-3/+1485
|\ \ \ | | | | | | | | Add instruct-pix2pix hijack
| * | | Update requirements transformers==4.25.1Kyle2023-01-252-2/+2
| | | | | | | | | | | | | | | | Update requirement for transformers to version 4.25.1 to allow instruct-pix2pix demo code to work
| * | | Add instruct-pix2pix hijackKyle2023-01-253-1/+1483
|/ / / | | | | | | | | | | | | | | | 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
* | | Merge pull request #7146 from EllangoK/masterAUTOMATIC11112023-01-252-45/+134
|\ \ \ | | | | | | | | Adds X/Y/Z Grid Script
| * | | swaps xyz axes internally if one costs moreEllangoK2023-01-241-12/+52
| | | |
| * | | handling sub grids and merging into oneEllangoK2023-01-242-12/+19
| | | |
| * | | implements most of xyz grid scriptEllangoK2023-01-241-36/+78
| | | |
| * | | renamed xy to xyz gridEllangoK2023-01-241-0/+0
| | | | | | | | | | | | | | | | this is mostly just so git can detect it properly
* | | | add an experimental option to apply loras to outputs rather than inputsAUTOMATIC2023-01-252-2/+10
| |_|/ |/| |
* | | remove the pip install stuff because it does not work as i hoped it wouldAUTOMATIC2023-01-243-8/+11
| | |
* | | add fastapi to requirementsAUTOMATIC2023-01-241-0/+1
| | |
* | | print progress when installing torchAUTOMATIC2023-01-242-6/+22
| | | | | | | | | | | | | | | | | | 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
* | | remove fairscale requirement, add fake fairscale to make BLIP not complain ↵AUTOMATIC2023-01-241-1/+1
| | | | | | | | | | | | about it mk2
* | | remove fairscale requirement, add fake fairscale to make BLIP not complain ↵AUTOMATIC2023-01-243-4/+9
| | | | | | | | | | | | about it
* | | make git commands not fail for extensions when you have spaces in webui ↵AUTOMATIC2023-01-241-4/+4
| | | | | | | | | | | | directory
* | | also return the removed field to sdapi/v1/upscalers because someone might ↵AUTOMATIC2023-01-242-0/+2
| | | | | | | | | | | | have relied on it existing
* | | repair sdapi/v1/upscalers returning bogus resultsAUTOMATIC2023-01-242-8/+10
| | |
* | | Merge pull request #7140 from vladmandic/api-decode-imageAUTOMATIC11112023-01-241-1/+5
|\ \ \ | | | | | | | | Add exception handling to API image decode
| * | | add image decod exception handlingVladimir Mandic2023-01-231-1/+5
| | |/ | |/|
* | | Merge pull request #7144 from catboxanon/swinir-interruptAUTOMATIC11112023-01-241-1/+7
|\ \ \ | | | | | | | | Make SwinIR upscaler interruptible and skippable
| * | | Also make SwinIR skippablecatboxanon2023-01-241-2/+2
| | | |
| * | | Make SwinIR interruptiblecatboxanon2023-01-241-1/+7
| |/ /
* | | Merge pull request #7148 from acncagua/improvement_launch.pyAUTOMATIC11112023-01-241-1/+1
|\ \ \ | |/ / |/| | Set Linux xformers 0.0.16RC425
| * | Set Linux xformers 0.0.16RC425acncagua2023-01-241-1/+1
|/ /
* | fix BLIP failing to import depending on configurationAUTOMATIC2023-01-232-1/+16
| |
generated by cgit v1.2.3 (git 2.25.1) at 2025-05-08 06:27:46 +0000