284fd8f415
Tweaked UI sliders and labels.
2023-11-28 23:03:50 -07:00
c5c7fa06aa
Added slider for detail preservation strength, removed largely needless offset parameter, changed labels in UI and for saving to/pasting data from PNG files.
2023-11-28 22:35:07 -07:00
debf836fcc
Added UI elements to control blending parameters.
2023-11-28 16:15:36 -07:00
a6e5846453
Nerfs the aggressive post-processing step of overlaying the original image.
2023-11-28 16:13:42 -07:00
e715e46b6a
Implements "scheduling" for blending of the original latents and a latent blending formula that preserves details in blend transition areas.
2023-11-28 16:10:22 -07:00
bbba133f05
Removed conflicting step that replaces the softly inpainted latents with a naive blend with the original latents.
2023-11-28 15:09:43 -07:00
dec791d35d
Removed code which forces the inpainting mask to be 0 or 1. Now fractional values (e.g. 0.5) are accepted.
2023-11-28 15:05:01 -07:00
81c00728b8
Fix the Ruff error about unused import
2023-11-29 04:59:35 +09:00
a0096c5897
Add FP32 fallback support on torch.nn.functional.interpolate
...
This tries to execute interpolate with FP32 if it failed.
Background is that
on some environment such as Mx chip MacOS devices, we get error as follows:
```
"torch/nn/functional.py", line 3931, in interpolate
return torch._C._nn.upsample_nearest2d(input, output_size, scale_factors)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: "upsample_nearest2d_channels_last" not implemented for 'Half'
```
In this case, ```--no-half``` doesn't help to solve. Therefore this commits add the FP32 fallback execution to solve it.
Note that the ```upsample_nearest2d``` is called from ```torch.nn.functional.interpolate```.
And the fallback for torch.nn.functional.interpolate is necessary at
```modules/sd_vae_approx.py``` 's ```VAEApprox.forward```
```repositories/stable-diffusion-stability-ai/ldm/modules/diffusionmodules/openaimodel.py``` 's ```Upsample.forward```
2023-11-29 04:45:04 +09:00
39eae9f009
Revert "Add FP32 fallback support on sd_vae_approx"
...
This reverts commit 58c19545c8https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/14046#issuecomment-1826731532 )
2023-11-29 04:07:48 +09:00
d608926f81
reformat file with uniform indentation
2023-11-28 12:12:27 +09:00
03ee297aa2
fix Auto focal point crop for opencv >= 4.8.x
...
autocrop.download_and_cache_models
in opencv >= 4.8 the face detection model was updated
download the base on opencv version
returns the model path or raise exception
2023-11-28 12:09:51 +09:00
3cd6e1d0a0
Update devices.py
...
fixes issue where "--use-cpu" all properly makes SD run on CPU but leaves ControlNet (and other extensions, I presume) pointed at GPU, causing a crash in ControlNet caused by a mismatch between devices between SD and CN
https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/14097
2023-11-27 19:21:43 -05:00
26a0c29587
Allow use of mutiple styles csv files
...
* https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/14122
Fix edge case where style text has multiple {prompt} placeholders
* https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/14005
2023-11-27 12:08:51 +00:00
8a6e4bda21
catch uncaught exception with ui creation scripts
...
prevent total webui crash
2023-11-27 14:18:17 +09:00
b30cc87b78
add Block component creation callback
2023-11-27 13:15:17 +09:00
1f6844eb7e
also consider extension url
2023-11-26 10:04:39 -06:00
f0f100e67b
add categories to settings
2023-11-26 17:56:22 +03:00
a15dd151ff
json.dump(ensure_ascii=False)
...
improve json readability
2023-11-26 21:56:21 +09:00
2a40d3c603
compact prompt layout: preserve scroll when switching between lora tabs
2023-11-26 14:58:56 +03:00
3d341ebc7d
Merge branch 'dev' into test-fp8
2023-11-26 17:32:52 +08:00
29f04149b6
update torch to 2.1.0
2023-11-26 12:07:33 +03:00
6955c210b7
Merge pull request #14059 from akx/upruff
...
Update Ruff to 0.1.6
2023-11-26 11:54:36 +03:00
c5a0c59a83
do not save HTML explanations from options page to config
2023-11-26 11:36:17 +03:00
f7f015e84b
Merge pull request #14084 from wfjsw/move-from-sysinfo-to-errors
...
Move exception_records related methods to errors.py
2023-11-26 11:29:27 +03:00
f85b74763d
Merge branch 'hypertile-in-sample' into dev
2023-11-26 11:18:49 +03:00
fd8674a4bc
Merge pull request #13948 from aria1th/hypertile-in-sample
...
support HyperTile optimization
2023-11-26 11:18:25 +03:00
d2e0c1ca13
rework hypertile into a built-in extension
2023-11-26 11:17:38 +03:00
3a9bf4ac10
move file
2023-11-26 08:29:12 +03:00
40ac134c55
Fix pre-fp8
2023-11-25 12:35:09 +08:00
5cedc8f9b2
remove traceback in sysinfo
2023-11-24 11:30:30 -06:00
86b99b1e98
Move exception_records related methods to errors.py
2023-11-24 11:28:54 -06:00
ac2a981c4f
use extension name for determining an extension is installed in the index
2023-11-22 22:40:24 -06:00
066afda2f6
Simplify restart_sampler (suggested by ruff)
2023-11-22 18:05:12 +02:00
f5d719d1f1
Add forced reload for fp16 cache
2023-11-22 01:45:56 +08:00
370a77f8e7
Option for using fp16 weight when apply lora
2023-11-21 19:59:34 +08:00
8aa51f682c
fix [Bug]: (Dev Branch) Placing "Dimensions" first in "ui_reorder_list" prevents start #14047
2023-11-21 08:32:07 +03:00
58c19545c8
Add FP32 fallback support on sd_vae_approx
...
This tries to execute interpolate with FP32 if it failed.
Background is that
on some environment such as Mx chip MacOS devices, we get error as follows:
```
"torch/nn/functional.py", line 3931, in interpolate
return torch._C._nn.upsample_nearest2d(input, output_size, scale_factors)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: "upsample_nearest2d_channels_last" not implemented for 'Half'
```
In this case, ```--no-half``` doesn't help to solve. Therefore this commits add the FP32 fallback execution to solve it.
Note that the submodule may require additional modifications. The following is the example modification on the other submodule.
```repositories/stable-diffusion-stability-ai/ldm/modules/diffusionmodules/openaimodel.py
class Upsample(nn.Module):
..snip..
def forward(self, x):
assert x.shape[1] == self.channels
if self.dims == 3:
x = F.interpolate(
x, (x.shape[2], x.shape[3] * 2, x.shape[4] * 2), mode="nearest"
)
else:
try:
x = F.interpolate(x, scale_factor=2, mode="nearest")
except:
x = F.interpolate(x.to(th.float32), scale_factor=2, mode="nearest").to(x.dtype)
if self.use_conv:
x = self.conv(x)
return x
..snip..
```
You can see the FP32 fallback execution as same as sd_vae_approx.py.
2023-11-21 01:13:53 +09:00
314ae1535e
added option for default behavior of dir buttons
2023-11-20 16:19:54 +01:00
5f36f6ab21
Merge pull request #14009 from AUTOMATIC1111/Option-to-show-batch-img2img-results-in-UI
...
Option to show batch img2img results in UI
2023-11-20 17:44:58 +03:00
1463cea949
Merge branch 'dag' into dev
2023-11-20 14:50:01 +03:00
73a0b4bba6
Merge pull request #13944 from wfjsw/dag
...
implementing script metadata and DAG sorting mechanism
2023-11-20 14:49:46 +03:00
9b471436b2
rework extensions metadata: use custom sorter that doesn't mess the order as much and ignores cyclic errors, use classes with named fields instead of dictionaries, eliminate some duplicated code
2023-11-20 14:47:09 +03:00
6d337bf23d
save sysinfo as .json
...
GitHub now allows uploading of .json files in issues
2023-11-20 01:38:31 +09:00
dea5e43c83
Option to show batch img2img results in UI
...
shared.opts.img2img_batch_show_results_limit
limit the number of images return to the UI for batch img2img
default limit 32
0 no images are shown
-1 unlimited, all images are shown
2023-11-19 17:37:32 +09:00
043d2edcf6
Better naming
2023-11-19 15:56:31 +08:00
f383af2729
update xformers/torch versions
2023-11-19 15:56:23 +08:00
890181e1d4
Update the xformers/torch versions
2023-11-19 15:54:39 +08:00
598da5cd49
Use options instead of cmd_args
2023-11-19 15:50:06 +08:00
b60e1088db
Merge branch 'dev' into test-fp8
2023-11-19 15:24:57 +08:00
bde439ef67
use metadata.ini for meta filename
2023-11-19 00:58:47 -06:00
fc83af4432
Merge pull request #13931 from AUTOMATIC1111/style-hotkeys
...
Enable prompt hotkeys in style editor
2023-11-19 09:11:49 +03:00
337bc4a2fb
Merge pull request #13014 from AUTOMATIC1111/thread-safe-extranetworks-list_items
...
thread safe extra network list_items
2023-11-19 09:09:21 +03:00
e4e875fffe
Merge pull request #13968 from kaalibro/extranetworks-path-sorting
...
Adds 'Path' sorting for Extra network cards
2023-11-19 09:00:05 +03:00
b945ba716b
Merge pull request #13977 from AUTOMATIC1111/hotfix-postprocessing-state-end
...
Hotfix: call shared.state.end() after postprocessing done
2023-11-19 08:59:32 +03:00
3a13b0e762
Merge pull request #13996 from Luxter77/patch-1
...
Adds tqdm handler to logging_config.py for progress bar integration
2023-11-19 08:57:14 +03:00
6429c3db11
Merge pull request #13826 from ezxzeng/ui_mobile_optimizations
...
added accordion settings options
2023-11-19 08:42:58 +03:00
5a9dc1c0ca
Merge pull request #14004 from storyicon/master
...
feat: fix randn found element of type float at pos 2
2023-11-19 08:40:29 +03:00
4f2a4a3615
feat: fix randn found element of type float at pos 2
...
Signed-off-by: storyicon <storyicon@foxmail.com>
2023-11-17 09:48:18 +00:00
97431f29fe
fix double gc and decoding with unet context
2023-11-17 10:05:28 +09:00
ffd0f8ddc3
set empty value for SD XL 3rd layer
2023-11-17 09:54:33 +09:00
c0725ba2d0
Fix inverted option issue
...
I'm pretty sure I was sleepy while implementing this
2023-11-17 09:34:50 +09:00
c40be2252a
Fix critical issue - unet apply
2023-11-17 09:22:27 +09:00
7021cdb1de
actually adds handler to logging_config.py
2023-11-16 17:53:57 -03:00
cdb60a690d
Take into account tqdm not being installed before first boot for logging
2023-11-16 16:49:59 -03:00
236eb82c3a
Adds tqdm handler to logging_config.py for progress bar integration
2023-11-16 13:20:33 -03:00
cd12256575
Merge branch 'dev' into test-fp8
2023-11-16 21:53:13 +08:00
472c22cc8a
fix ruff - add newline
2023-11-16 19:03:45 +09:00
bcfaf3979a
convert/add hypertile options
2023-11-16 18:43:16 +09:00
af45872fdb
copy LDM VAE key from XL
2023-11-15 15:15:14 +09:00
b29fc6d4de
Implement Hypertile
...
Co-Authored-By: Kieran Hunt <kph@hotmail.ca>
2023-11-15 15:13:39 +09:00
a292d2c47f
hotfix: call shared.state.end() after postprocessing done
2023-11-15 14:26:37 +09:00
c1c816006e
Adds 'Path' sorting for Extra network cards
2023-11-13 22:01:52 +06:00
8048f36072
Lint
2023-11-12 17:12:50 -07:00
f6762d2ad9
dir buttons start with / so only the correct dir will be shown and not dirs with a substrings as name from the dir
2023-11-12 14:14:16 +01:00
3bb32befe9
bug fix
2023-11-11 11:58:19 -06:00
48d6102b31
fix
2023-11-11 11:17:26 -06:00
520e52f846
allow comma and whitespace as separator
2023-11-11 10:58:26 -06:00
7af576e745
remove the assumption of same name
2023-11-11 10:46:47 -06:00
294f8a514f
add hyperTile
...
https://github.com/tfernd/HyperTile
2023-11-11 23:28:12 +09:00
bc1a450124
reverse the extension load order so builtin extensions load earlier natively
2023-11-11 04:08:45 -06:00
0d1924c48b
populate loaded_extensions from extension list instead
2023-11-11 04:03:55 -06:00
0fc7dc1c04
implementing script metadata and DAG sorting mechanism
2023-11-11 04:01:13 -06:00
3a4a6c43a4
ExitStack as alternative to suppress
2023-11-10 16:06:01 -05:00
5432d93013
fix added accordion settings options
2023-11-11 05:30:35 +09:00
7ff54005fe
Enable prompt hotkeys in style editor
2023-11-09 23:47:53 -07:00
ff2952f105
multiline with statement for readibility
2023-11-09 13:35:52 -05:00
9aa4d098f0
removed changes that weren't merged properly
2023-11-09 13:25:24 -05:00
a625a7bb81
moved nested with to single line to remove extra tabs
2023-11-09 13:15:06 -05:00
f9c14a8c8c
Merge branch 'dev' into ui_mobile_optimizations
2023-11-07 15:25:27 -05:00
5e80d9ee99
fix pix2pix producing bad results
2023-11-07 11:33:33 +03:00
9ba991cad8
Add option to set notification sound volume
2023-11-07 03:09:08 +01:00
9c1c0da026
fix exception related to the pix2pix
2023-11-06 11:17:36 +03:00
656437e0a5
fix img2img_tabs error
2023-11-06 10:32:21 +03:00
6ad666e479
more changes for #13865 : fix formatting, rename the function, add comment and add a readme entry
2023-11-05 19:46:20 +03:00
80d639a440
linter
2023-11-05 19:32:21 +03:00
ff805d8d0e
Merge branch 'dev' into master
2023-11-05 19:30:57 +03:00
c3699d4fd1
compact prompt option disabled by default
2023-11-05 19:23:48 +03:00
4d4a9e7332
added compact prompt option
2023-11-05 19:19:55 +03:00
44c5097375
Use devices.torch_gc() instead of empty_cache()
2023-11-05 20:31:57 +05:30
44db35fb1a
Added memory clearance after deletion
2023-11-05 19:15:38 +05:30
ff1609f91e
Add SSD-1B as a supported model
2023-11-05 19:13:49 +05:30
d9499f4301
properly apply sort order for extra network cards when selected from dropdown
...
allow selection of default sort order in settings
remove 'Default' sort order, replace with 'Name'
2023-11-05 10:12:50 +03:00
2b06cefe66
correct a typo
...
modify "defaul" to "default"
2023-11-05 11:37:23 +09:00
c3facab495
Merge branch 'dev' into test-fp8
2023-11-04 12:56:58 +08:00
452ab8fe72
Merge pull request #13718 from avantcontra/bugfix_gfpgan_custom_path
...
fix bug when using --gfpgan-models-path
2023-11-03 20:19:58 +03:00
21d561885e
Merge pull request #13762 from wkpark/nextjob
...
call state.jobnext() before postproces*()
2023-11-03 20:16:58 +03:00
73c74baa6a
Merge pull request #13797 from Meerkov/master
...
Fix #13796
2023-11-03 20:11:54 +03:00
759515316e
added accordion settings options
2023-11-02 21:54:48 -04:00
fbc5c531b9
Fix #13796
...
Fix comment error that makes understanding scheduling more confusing.
2023-10-29 15:37:08 -07:00
ddc2a3499b
Add MPS manual cast
2023-10-28 16:52:35 +08:00
d4d3134f6d
ManualCast for 10/16 series gpu
2023-10-28 15:24:26 +08:00
5121846d34
call state.jobnext() before postproces*()
2023-10-25 21:57:41 +09:00
0beb131c7f
change torch version
2023-10-25 20:07:37 +08:00
dda067f64d
ignore mps for fp8
2023-10-25 19:53:22 +08:00
bf5067f50c
Fix alphas cumprod
2023-10-25 12:54:28 +08:00
4830b25136
Fix alphas_cumprod dtype
2023-10-25 11:53:37 +08:00
1df6c8bfec
fp8 for TE
2023-10-25 11:36:43 +08:00
9c1eba2af3
Fix lint
2023-10-24 02:11:27 +08:00
eaa9f5162f
Add CPU fp8 support
...
Since norm layer need fp32, I only convert the linear operation layer(conv2d/linear)
And TE have some pytorch function not support bf16 amp in CPU. I add a condition to indicate if the autocast is for unet.
2023-10-24 01:49:05 +08:00
236dd55dbe
fix Blank line contains whitespace
2023-10-22 04:32:13 +08:00
443ca983ad
fix bug when using --gfpgan-models-path
2023-10-22 03:21:23 +08:00
384fab9627
rework some of changes for emphasis editing keys, force conversion of old-style emphasis
2023-10-21 08:45:51 +03:00
5f9ddfa46f
Add sdxl only arg
2023-10-19 23:57:22 +08:00
7c128bbdac
Add fp8 for sd unet
2023-10-19 13:56:17 +08:00
a3d9b011a3
Merge pull request #13533 from missionfloyd/edit-attention-fix
...
Edit-attention fixes
2023-10-15 10:08:52 +03:00
282903bb67
repair unload sd checkpoint button
2023-10-15 09:41:02 +03:00
0d65d0eabd
add an option to not print stack traces on ctrl+c.
2023-10-15 08:45:38 +03:00
4be7b620c2
Merge pull request #13568 from AUTOMATIC1111/lora_emb_bundle
...
Add lora-embedding bundle system
2023-10-14 12:18:55 +03:00
a8cbe50c9f
remove duplicated code
2023-10-14 12:17:59 +03:00
19f5795c27
Merge pull request #13463 from FluttyProger/patch-1
...
Ability for extensions to return custom data via api in response.images
2023-10-14 08:37:45 +03:00
eadef35512
Merge pull request #13567 from LeonZhao28/bugfix_key_error_in_processing
...
fix the key error exception when processing override_settings keys
2023-10-14 08:34:41 +03:00
0619df9835
use shallow copy for #13535
2023-10-14 08:01:04 +03:00
7cc96429f2
Merge pull request #13535 from chu8129/dev
...
fix: checkpoints_loaded:{checkpoint:state_dict}, model.load_state_dict issue in dict value empty
2023-10-14 08:00:04 +03:00
a109c7aeb8
more general case of adding an infotext when no images have been generated
2023-10-14 07:49:03 +03:00
27fdc26a74
Merge pull request #13630 from wkpark/indexerror-fix
...
fix IndexError
2023-10-14 07:46:34 +03:00
3a66c3c9e1
put notification.mp3 option at the end of the page
2023-10-14 07:35:06 +03:00
499543cf1d
Merge pull request #13631 from galekseev/master
...
added option to play notification sound or not
2023-10-14 07:30:31 +03:00
902afa6b4c
Merge pull request #13364 from superhero-7/master
...
Add altdiffusion-m18 support
2023-10-14 07:29:01 +03:00
fff1a0c74f
Make attention conversion optional
...
Fix square brackets multiplier
2023-10-13 17:18:02 -06:00
44d14bc32e
added option to play notification sound or not
2023-10-13 15:08:59 -03:00
fbc8d21354
fix IndexError: list index out of range error interrupted while postprocess
2023-10-14 02:45:09 +09:00
9821625a76
fix the key error exception when adding an overwriting key which is defined in the extensions
2023-10-09 18:36:48 +08:00
09a2da835e
Add brackets, vertical bar to default delimiters
2023-10-07 14:48:43 -06:00
770ee23f18
reverst
2023-10-07 15:38:50 +08:00
76010a51ef
up
2023-10-07 15:36:01 +08:00
c2279da522
fix re_param_code (regression bug PR #13458 )
2023-10-03 01:16:41 +09:00
dc2074c46d
Merge pull request #13466 from AUTOMATIC1111/denoising-none
...
Change denoising_strength default to None.
2023-10-02 13:05:27 +03:00
3f763d41e8
Change denoising_strength default to None.
2023-10-01 22:38:27 -06:00
f71e919ecb
Ability for extensions to return custom data via api in response.images
2023-10-01 18:06:48 +03:00
CodeHatchling