Commit Graph

4794 Commits

Author SHA1 Message Date
AUTOMATIC1111 57e6d05a43 added tool for profiling code 2024-06-09 21:18:36 +03:00
AUTOMATIC1111 aafbb5b403 lint 2024-06-09 16:47:08 +03:00
AUTOMATIC1111 e368cd2810 stylistic changes for #15978 2024-06-09 16:46:08 +03:00
AUTOMATIC1111 981abbb1f2
Merge pull request #15978 from bluelovers/pr/pattern-001
feat: save pattern add `basename`
2024-06-09 16:45:06 +03:00
AUTOMATIC1111 6214aa7d2a performance: check for nans in unet only once, after all steps have been completed 2024-06-09 16:24:04 +03:00
bluelovers 6447ff49d3 feat: save pattern add `basename`
`grid` or `xyz_grid` or `img`

```py
'basename': lambda self: 'img' if self.basename == '' else self.basename,
```
2024-06-09 19:07:32 +08:00
huchenlei f89b5dbbd2 nit 2024-06-08 22:15:37 -04:00
huchenlei d875cda565 Fix sdxl inpaint 2024-06-08 22:11:11 -04:00
drhead d52a1e1a22
lint 2024-06-08 18:56:23 -04:00
drhead 39a6d5655f
patch k_diffusion to_d and strip device from schedulers 2024-06-08 18:55:07 -04:00
drhead 428975e1d3
Merge pull request #1 from AUTOMATIC1111/dev
Dev
2024-06-08 18:49:44 -04:00
AUTOMATIC1111 547778b10f possibly make NaN check cheaper 2024-06-08 12:41:38 +03:00
AUTOMATIC1111 194c2620d6
Merge pull request #15968 from AUTOMATIC1111/wsl-open
replace wsl-open with wslpath and explorer.exe
2024-06-08 12:30:11 +03:00
AUTOMATIC1111 5ecfc20d97
Merge pull request #15610 from pinanew/pinanew-patch-1
AVIF has quality setting too
2024-06-08 12:27:54 +03:00
AUTOMATIC1111 2dbc7aa688
Merge pull request #15627 from light-and-ray/more_extension_tag_filtering_options
more extension tag filtering options
2024-06-08 12:24:17 +03:00
AUTOMATIC1111 3ef9f2748d
Merge branch 'dev' into bgh-handle-metadata-issues-more-cleanly 2024-06-08 12:16:55 +03:00
AUTOMATIC1111 30461bef98
Merge pull request #15602 from AUTOMATIC1111/initial-model-download-integrity
Initial model download integrity
2024-06-08 12:13:24 +03:00
AUTOMATIC1111 569f17c6c6
Merge pull request #15654 from huchenlei/mime
Add correct mimetype for .mjs files
2024-06-08 12:10:42 +03:00
AUTOMATIC1111 9e5103124a
Merge pull request #15641 from AUTOMATIC1111/no-referrer
no-referrer
2024-06-08 11:56:53 +03:00
AUTOMATIC1111 a1130c26e5
Merge pull request #15664 from AUTOMATIC1111/fix-extra-batch-mode-P-Transparency
fix extra batch mode P Transparency
2024-06-08 11:53:08 +03:00
AUTOMATIC1111 b9dfc50a1b
Merge pull request #15705 from AUTOMATIC1111/use-script_path-for-webui-root-in-launch_utils
use script_path for webui root in launch_utils
2024-06-08 11:51:48 +03:00
AUTOMATIC1111 debc6dddeb
Merge pull request #15739 from LoganBooker/LoganBooker-AVIF-mimetype-patch
Add AVIF MIME type support to mimetype definitions
2024-06-08 11:36:21 +03:00
AUTOMATIC1111 1a7ffa2c76 remove extra local variable 2024-06-08 11:35:45 +03:00
AUTOMATIC1111 64783dd9cc
Merge pull request #15742 from MarcusNyne/m9-240508-model-dir
Added --models-dir option
2024-06-08 11:35:03 +03:00
AUTOMATIC1111 c1c4b3fb34
Merge pull request #15738 from JLipnerPitt/JLipnerPitt-patch-1
Fix AttributeError
2024-06-08 11:32:38 +03:00
AUTOMATIC1111 07cf95c76e update pickle safe filenames 2024-06-08 11:26:41 +03:00
AUTOMATIC1111 7b940e3879
Merge pull request #15797 from AUTOMATIC1111/fix-extention-update-when-not-on-main-branch
fix extention update when not on main branch
2024-06-08 11:19:26 +03:00
AUTOMATIC1111 b4723bb8c1
Merge pull request #15815 from AUTOMATIC1111/torch-float64-or-float32
fix soft inpainting on mps and xpu, torch_utils.float64
2024-06-08 11:07:29 +03:00
AUTOMATIC1111 6450d24afe
Merge pull request #15806 from huchenlei/inpaint_fix
[Performance 4/6] Precompute is_sdxl_inpaint flag
2024-06-08 11:06:39 +03:00
AUTOMATIC1111 64bf57b5ea
Merge pull request #15817 from light-and-ray/img2img_batch_upload
img2img batch upload method
2024-06-08 11:00:21 +03:00
AUTOMATIC1111 603509ec90 as per wfjsw's suggestion, revert changes for sd_hijack_checkpoint.py 2024-06-08 10:54:41 +03:00
AUTOMATIC1111 ad229fae43
Merge pull request #15803 from huchenlei/checkpoint_false
[Performance 1/6] use_checkpoint = False
2024-06-08 10:52:40 +03:00
w-e-w 510f025a01 replace wsl-open with wslpath and explorer.exe 2024-06-08 16:52:12 +09:00
AUTOMATIC1111 c3c90deec0
Merge pull request #15681 from AUTOMATIC1111/fix_p_invalid_sampler_and_scheduler
more old sampler scheduler compatibility
2024-06-08 10:42:42 +03:00
AUTOMATIC1111 ebfc9f6d09
Merge branch 'dev' into patch-4 2024-06-08 10:35:09 +03:00
AUTOMATIC1111 33b73c473c
Merge pull request #15820 from huchenlei/force_half
[Performance 6/6] Add --precision half option to avoid casting during inference
2024-06-08 10:26:23 +03:00
AUTOMATIC1111 ba54c747e2
Merge pull request #15656 from AUTOMATIC1111/api-old-sampler-names
Allow old sampler names in API
2024-06-08 10:18:20 +03:00
AUTOMATIC1111 15245d9d5e
Merge pull request #15600 from AUTOMATIC1111/fix-corrupt-model-loop
Fix corrupt model initial load loop
2024-06-08 10:12:45 +03:00
AUTOMATIC1111 5429e4cff5 add proper infotext support for #15607
fix settings override not working for NGMI, s_churn, etc...
2024-06-08 09:56:09 +03:00
AUTOMATIC1111 b150b3a3a4
Merge pull request #15607 from drhead/patch-1
add code for skipping CFG on early steps
2024-06-08 09:20:08 +03:00
AUTOMATIC1111 0edc04d126
Merge branch 'dev' into patch-2 2024-06-08 09:10:51 +03:00
AUTOMATIC1111 00f37ad73f
Merge pull request #15893 from alcacode/dev
Fix bug where file extension had an extra '.' under some circumstances
2024-06-08 09:06:03 +03:00
AUTOMATIC1111 0769aa318a integrated edits as recommended in the PR #15804 2024-06-08 09:05:35 +03:00
AUTOMATIC1111 de7f5cdc62
Merge pull request #15804 from huchenlei/rearrange_fix
[Performance 2/6] Replace einops.rearrange with torch native ops
2024-06-08 08:55:51 +03:00
AUTOMATIC1111 46bcfbe37c
Merge pull request #15751 from LoganBooker/LoganBooker-Add-AlignYourSteps-Scheduler
Add Align Your Steps to available schedulers
2024-06-08 08:43:02 +03:00
NouberNou 53f62674ae
Typo on edit
Edited in fix in Github editor and mistyped from local copy
2024-06-06 16:30:01 -07:00
NouberNou 25bbf31f57
Fix for grids without comprehensive infotexts
When generating grids, some scripts such as img2img loopback and ultimate SD upscale do not pass infotexts for each image since they are the same prompt.

If you attempt to save those images using the saved button in the UI it will fail because it will look for the selected image info text. This fixes those errors by replicating the infotext for as many images are passed into the image list if the infotext parameter is none.
2024-06-06 16:22:49 -07:00
alcacode 6dd53ce63d
Fix bug where file extension had an extra '.' under some circumstances
Fix bug where under some circumstances an extra "." was inserted between the file base name and the file extension.
The bug is triggered when the extension argument is one of "jpg", "jpeg", or "webp", and the image exceeds the format's dimension limit. Then the extension variable is set to ".png", resulting in the fullfn variable to evaluate to a string ending with "..png".
2024-05-26 15:36:55 +02:00
drhead feeb6802aa
fix case where first step skilled if skip early cond is 0 2024-05-18 01:22:31 -04:00
Logan 1d74482817 Default device for sigma tensor to CPU
* Consistent with implementations in k-diffusion.
* Makes this compatible with https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/15823
2024-05-18 09:09:57 +10:00
huchenlei b57a70f373 Proper fix of SD15 dtype 2024-05-17 13:34:04 -04:00
huchenlei dca9007ac7 Fix SD15 dtype 2024-05-17 13:23:12 -04:00
drhead cc9ca67664
Add transformer forward patch 2024-05-17 13:14:26 -04:00
drhead 53d67088ee
Patch timestep embedding to create tensor on-device 2024-05-17 12:12:57 -04:00
drhead 01491d303c
Keep sigmas on CPU 2024-05-17 10:36:08 -04:00
huchenlei 47f1d42a7e Fix for SD15 models 2024-05-16 20:06:04 -04:00
huchenlei 2a8a60c2c5 Add --precision half cmd option 2024-05-16 19:50:06 -04:00
huchenlei 58eec83a54 Fully prevent use_checkpoint 2024-05-16 16:39:02 -04:00
Andray 221ac0b9ab img2img batch upload method 2024-05-16 23:08:24 +04:00
w-e-w f015b94176 use torch_utils.float64 2024-05-16 23:19:06 +09:00
w-e-w 9c8075ba8e torch_utils.float64
return torch.float64 if device is not mps or xpu, else return torch.float32
2024-05-16 23:16:50 +09:00
huchenlei 3e20b36e8f Fix attr access 2024-05-15 17:27:01 -04:00
huchenlei 6a48476502 Fix flag check for SD15 2024-05-15 16:54:26 -04:00
huchenlei 9eb2f78631 Precompute is_sdxl_inpaint flag 2024-05-15 16:32:29 -04:00
huchenlei 0e98529365 Replace einops.rearrange with torch native 2024-05-15 15:46:53 -04:00
huchenlei 022d835565 use_checkpoint = False 2024-05-15 15:20:40 -04:00
w-e-w 5ab7d08a0a fix extention update when not on main branch 2024-05-15 17:27:05 +09:00
Logan d6b4444069 Use shared.sd_model.is_sdxl to determine base AYS sigmas 2024-05-10 18:05:45 +10:00
Logan 73d1caf8f2 Add Align Your Steps to available schedulers
* Include both SDXL and SD 1.5 variants (https://research.nvidia.com/labs/toronto-ai/AlignYourSteps/howto.html)
2024-05-10 12:38:10 +10:00
MarcusNyne 5fbac49791 Added --models-dir option
The --model-dir option overrides the location of the models directory for stable diffusion, so that models can be shared across multiple installations.  When --data-dir is specified alone, both the extensions and models folders are present in this folder.  --models-dir can be used independently, but when used with --data-dir, then the models folder is specified by --models-dir, and extensions are found in the --data-dir.
2024-05-08 16:48:10 -04:00
LoganBooker f7e349cea4
Add AVIF MIME type support to mimetype definitions
AVIF images will open, rather than download, as the default behaviour.
2024-05-08 21:23:18 +10:00
JLipnerPitt e736c3b36b
Add files via upload
Fixed an error (AttributeError: 'str' object has no attribute 'decode') coming from line 792 in images.py when trying to upscale certain images.
2024-05-08 05:22:12 -04:00
w-e-w f12886aefa use script_path for webui root in launch_utils 2024-05-04 23:42:37 +09:00
w-e-w 5d5224b322 fix_p_invalid_sampler_and_scheduler 2024-05-02 02:25:16 +09:00
w-e-w 9d39380705 fix extra batch mode P Transparency
red, green, blue = transparency TypeError: cannot unpack non-iterable int object
2024-04-30 19:17:53 +09:00
missionfloyd c8336c45b9 Use existing function for old sampler names 2024-04-30 01:53:41 -06:00
missionfloyd 579f1ef278 Allow old sampler names in API 2024-04-28 22:36:43 -06:00
huchenlei 3d3fc81f48 Add correct mimetype for .mjs files 2024-04-28 16:14:12 -04:00
drhead 3a215deff2
vectorize kl-optimal sigma calculation
Co-authored-by: mamei16 <marcel.1710@live.de>
2024-04-28 00:15:58 -04:00
w-e-w 9d964d3fc3 no-referrer 2024-04-27 19:21:34 +09:00
Brendan Hoar 60c0799958
Linter - except must not be bare. 2024-04-26 08:21:12 -04:00
Brendan Hoar 44afb48447
Linter fix - extraneous whitespace 2024-04-26 08:17:37 -04:00
Brendan Hoar c5ae225418
Better handling of embeddings with two rare, but not unusual, files in them
I have encountered pickled embeddings with a short byteorder file at the top-level, as well as a .data/serialization_id file.

Both load fine after allowing these files in the dataset.

I do not think it is likely adding them to the safe unpickle regular expression would be a security risk, but that's for the maintainers to decide.
2024-04-26 07:55:39 -04:00
Brendan Hoar c5b7559856
Better error handling when unable to extract contents of embedding/TI file 2024-04-26 06:57:32 -04:00
Brendan Hoar 8dc920228e
Better error handling when unable to read metadata from safetensors file 2024-04-26 06:52:21 -04:00
w-e-w d5f6fdb3c4 compact-checkbox-group 2024-04-26 18:47:04 +09:00
Andray e85e327ae0 more extension tag filtering options 2024-04-25 13:26:26 +04:00
w-e-w 1091e3a37e update jpeg_quality description 2024-04-24 02:54:26 +09:00
w-e-w 8fa3fa76c3 fix exif_bytes referenced before assignment 2024-04-24 02:41:31 +09:00
pinanew 50bb6e1179
AVIF has quality setting too 2024-04-23 18:45:42 +03:00
drhead 029adbe531
implement option to skip uncond on all steps below ngms 2024-04-23 03:15:56 -04:00
drhead 33cbbf9f8b
add s_min_uncond_all option 2024-04-23 03:15:00 -04:00
drhead 6e9b69a338
change skip_early_cond code to use float 2024-04-23 03:08:28 -04:00
drhead 83182d2799
change skip early cond option name and to float 2024-04-23 03:07:25 -04:00
drhead 83266205d0
Add KL Optimal scheduler 2024-04-23 00:09:43 -04:00
drhead 8016d78a4b
add option for early cfg skip 2024-04-22 23:42:24 -04:00
drhead a1aa0af8a4
add code for skipping CFG on early steps 2024-04-22 23:38:44 -04:00
w-e-w c69773d7e8 ensure integrity for initial sd model download 2024-04-23 03:09:45 +09:00
w-e-w 246c269af8 add option to check file hash after download
if the sha256 hash does not match it will be automatically deleted
2024-04-23 03:09:45 +09:00
w-e-w 2b717bb195 fix initial corrupt model loop
if for some reason the initial loading model at loading phase of webui  is corrupted
after entering this state the user will not be able to load even a good model is selected, due the the unload_model_weights  > send_model_to_cpu > m.lowvram attribute check will fail becaules m is None
webui will be stuck in the loop unable to recover without manual intervention
2024-04-23 02:35:25 +09:00