aboutsummaryrefslogtreecommitdiffstats
path: root/modules/processing.py
Commit message (Collapse)AuthorAgeFilesLines
* Extra networks loading fixinvincibledude2023-01-291-10/+3
|
* Extra networks loading fixinvincibledude2023-01-291-2/+2
|
* Extra networks loading fixinvincibledude2023-01-291-2/+14
|
* Merge branch 'master' into masterInvincibleDude2023-01-291-7/+26
|\
| * remove Batch size and Batch pos from textinfo (goodbye)AUTOMATIC2023-01-291-2/+0
| |
| * Merge pull request #7309 from brkirch/fix-embeddingsAUTOMATIC11112023-01-281-7/+8
| |\ | | | | | | Fix embeddings, upscalers, and refactor `--upcast-sampling`
| | * Refactor conditional casting, fix upscalersbrkirch2023-01-281-7/+8
| | |
| * | add data-dir flag and set all user data directories based on itMax Audron2023-01-271-1/+2
| | |
| * | add an option to enable sections from extras tab in txt2img/img2imgAUTOMATIC2023-01-261-1/+6
| |/ | | | | | | fix some style inconsistenices
| * Fix full previews, --no-half-vaebrkirch2023-01-261-4/+4
| |
| * 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 #6510 from brkirch/unet16-upcast-precisionAUTOMATIC11112023-01-251-7/+8
| |\ | | | | | | 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 float32brkirch2023-01-251-1/+1
| | | | | | | | | | | | | | | | | | 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 UNetbrkirch2023-01-251-7/+8
| | | | | | | | | | | | 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().
| * | change to code for live preview fix on OSX to be bit more obviousAUTOMATIC2023-01-251-2/+2
| | |
| * | 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 branch 'AUTOMATIC1111:master' into masterInvincibleDude2023-01-241-2/+6
|\|
| * enable compact view for train tabAUTOMATIC2023-01-211-2/+6
| | | | | | | | prevent previews from ruining hypernetwork training
* | Gen params paste improvementinvincibledude2023-01-221-2/+2
| |
* | Gen params paste improvementinvincibledude2023-01-221-2/+2
| |
* | UI and PNG info improvementsinvincibledude2023-01-221-2/+2
| |
* | UI and PNG info improvementsinvincibledude2023-01-221-0/+3
| |
* | hr conditioninginvincibledude2023-01-221-1/+1
| |
* | hr conditioninginvincibledude2023-01-221-2/+2
| |
* | hr conditioninginvincibledude2023-01-221-7/+12
| |
* | hr conditioninginvincibledude2023-01-221-4/+5
| |
* | hr conditioninginvincibledude2023-01-221-21/+13
| |
* | hr conditioninginvincibledude2023-01-221-26/+46
| |
* | Hr-fix separate prompt experimentationinvincibledude2023-01-221-21/+22
| |
* | Logging for debugginginvincibledude2023-01-221-0/+3
| |
* | Fixinvincibledude2023-01-221-1/+1
| |
* | Hr separate prompt testinvincibledude2023-01-221-1/+22
| |
* | PLMS edge-case handling fix 5invincibledude2023-01-211-2/+0
| |
* | PLMS edge-case handling fix 3invincibledude2023-01-211-2/+2
| |
* | PLMS edge-case handling fix 2invincibledude2023-01-211-2/+6
| |
* | PLMS edge-case handling fixinvincibledude2023-01-211-1/+1
| |
* | Type mismatch fixinvincibledude2023-01-211-2/+2
| |
* | First test of different sampler for hi-res fixinvincibledude2023-01-211-1/+6
|/
* extract extra network data from prompt earlierAUTOMATIC2023-01-211-2/+2
|
* make it so that extra networks are not removed from infotextAUTOMATIC2023-01-211-1/+3
|
* extra networks UIAUTOMATIC2023-01-211-11/+13
| | | | rework of hypernets: rather than via settings, hypernets are added directly to prompt as <hypernet:name:weight>
* Merge pull request #6854 from EllangoK/masterAUTOMATIC11112023-01-181-4/+4
|\ | | | | Saves Extra Generation Parameters to params.txt
| * Changed params.txt save to after manual init callEllangoK2023-01-171-4/+4
| |
* | use DDIM in hires fix is the sampler is PLMSAUTOMATIC2023-01-181-1/+2
|/
* make StableDiffusionProcessing class not hold a reference to shared.sd_model ↵AUTOMATIC2023-01-161-4/+5
| | | | object
* Add a check and explanation for tensor with all NaNs.AUTOMATIC2023-01-161-0/+3
|
* change hypernets to use sha256 hashesAUTOMATIC2023-01-141-1/+1
|
* Fix extension parameters not being saved to last used parametersspace-nuko2023-01-121-4/+4
|
* add an option to use old hiresfix width/height behaviorAUTOMATIC2023-01-091-2/+24
| | | | add a visual effect to inactive hires fix elements
* Merge branch 'AUTOMATIC1111:master' into img2img-api-scriptsnoodleanon2023-01-071-1/+0
|\
generated by cgit v1.2.3 (git 2.25.1) at 2025-05-06 04:13:56 +0000