Commit Graph

3139 Commits

Author SHA1 Message Date
Kevin Turner
b6d37a70ca fix(inpainting model): threshold mask to avoid gray blurry seam 2023-01-28 13:34:22 -08:00
Kevin Turner
71b6ddf5fb fix(inpainting model): blank areas to be repainted in the masked image
Otherwise the model seems too reluctant to change these areas, even though the mask channel should allow it to.
2023-01-28 11:10:32 -08:00
mauwii
14de7ed925
remove requirements step from install manual 2023-01-28 00:58:32 +01:00
Kevin Turner
6556b200b5 remove experimental "blur" infill
It seems counterproductive for use with the inpainting model, and not especially useful otherwise.
2023-01-27 15:25:50 -08:00
Kevin Turner
d627cd1865 feat(inpaint): add simpler infill methods for use with inpainting model 2023-01-27 14:28:16 -08:00
Kevin Turner
09b6104bfd refactor(txt2img2img): factor out tensor shape 2023-01-27 12:04:12 -08:00
Kevin Turner
1bb5b4ab32 fix dimension errors when inpainting model is used with hires-fix 2023-01-27 11:52:05 -08:00
Lincoln Stein
c18db4e47b
removed defunct textual inversion script (#2433)
The original textual inversion script in scripts is now superseded. The
replacement can be found in ldm/invoke/textual_inversion.py and is a
merging of the command line and front end scripts. After running `pip
install -e .` there will be a `textual_inversion` command on your path.
You can activate the front end this way:

`textual_inversion -gui`
2023-01-27 10:44:34 -05:00
Jonathan
f9c92e3576
Merge branch 'main' into bugfix/remove-defunct-scripts 2023-01-27 08:32:15 -06:00
psychedelicious
1ceb7a60db
adds double-click to reset view to 100% (#2436)
Adds double-click to reset canvas view to 100%.

- Adds hook to manage single and double clicks
- Single Click `Reset Canvas View` --> scale to fit, no change to
current behaviour
- Double Click `Reset Canvas View` --> set scale to 1
2023-01-28 00:56:24 +11:00
psychedelicious
f509650ec5 adds double-click to reset view to 100% 2023-01-27 08:30:24 -05:00
psychedelicious
0d0f35a1e2
Fix download button styling (#2435)
fixes #2383
2023-01-28 00:29:34 +11:00
psychedelicious
6dbc42fc1a fixes download button styling 2023-01-27 20:23:12 +11:00
Lincoln Stein
f6018fe5aa removed defunct textual inversion script 2023-01-26 23:35:09 -05:00
blessedcoolant
e4cd66216e Frontend Build (diffusers-mm-fixes) 2023-01-27 17:23:25 +13:00
blessedcoolant
995fbc78c8 Diffusers Model Manager Fixes 2023-01-27 17:23:25 +13:00
blessedcoolant
3083f8313d Default Seam Steps to 30
Seems to be the temporary solution for the seams looking horrible with some diffuser models.
2023-01-27 17:23:25 +13:00
Lincoln Stein
c0614ac7f3
Improve configuration of initial Waifu models (#2426)
Testing suggests that the diffusers versions of Waifu-1.4 anything-v4.0
require the `sd-vae-ft-mse` to generate decent images, so the
appropriate arguments have been added to the initial model file.
2023-01-26 18:18:00 -05:00
Lincoln Stein
0186630514
Merge branch 'main' into install/better-initial-models 2023-01-26 17:42:10 -05:00
Lincoln Stein
d53df09203
[enhancement] Improve organization & behavior of model merging and textual inversion scripts (#2427)
- Model merging and textual inversion scripts have been moved into
`ldm/invoke`, which allows them to be installed properly by
pyproject.toml.
- As part of the pyproject install, the .py suffix is removed from the
command. I.e. use `invoke`, `configure_invokeai`, `merge_models` and
`textual_inversion`.
- GUI versions are activated by adding `--gui` to the command. Without
this, you get a classical argv-based command. Example: `merge_models
--gui`
- Fixed up the launcher scripts to accommodate new naming scheme.
- Keyboard behavior of the GUI front ends has been improved. You can now
use up and down arrow to move from field to field, in addition to <tab>
and ctrl-N/ctrl-P
2023-01-26 17:36:45 -05:00
Lincoln Stein
12a29bfbc0
Merge branch 'main' into install/change-script-locations 2023-01-26 17:10:33 -05:00
Lincoln Stein
f36114eb94
Fix Sliders unable to take typed input (#2407)
So far the slider component was unable to take typed input due to a
bunch of issues that were a pain to solve. This PR fixes it.

Things to test:

- Moving the slider also updates the value in the input text box.
- Input text box next to slider can be changed in two ways: If you type
a manual value, the slider will be updated when you lose focus from the
input box. If you use the stepper icons to update the values, the slider
should update immediately.
- Make sure the reset buttons next to the slider are updating correctly
and make sure this updates both the slider and the input box values.
- Brush Size slider -> make sure the hotkeys are updating the input box
too.
2023-01-26 17:10:16 -05:00
Lincoln Stein
c255481c11
Merge branch 'main' into slider-fix 2023-01-26 16:20:25 -05:00
Lincoln Stein
7f81105acf
dev: update to diffusers 0.12, transformers 4.26 (#2420)
Happy New Year!
2023-01-26 16:18:37 -05:00
Lincoln Stein
c8de679dc3
Merge branch 'main' into update-diffusers 2023-01-26 15:43:41 -05:00
Lincoln Stein
85b18fe9ee
Merge branch 'main' into install/better-initial-models 2023-01-26 15:42:13 -05:00
Lincoln Stein
e0d8c19da6 fix indentation problem 2023-01-26 15:39:59 -05:00
Lincoln Stein
5567808237 tweak documentation 2023-01-26 15:28:54 -05:00
Lincoln Stein
2817f8a428 update launcher shell scripts for new script names & paths 2023-01-26 15:26:38 -05:00
Lincoln Stein
8e4c044ca2 clean up tab/cursor behavior in textual inversion txt gui 2023-01-26 15:18:28 -05:00
Lincoln Stein
9dc3832b9b clean up merge_models 2023-01-26 15:10:16 -05:00
Lincoln Stein
046abb634e
Remove dependency on original clipseg library for text masking (#2425)
- This replaces the original clipseg library with the transformers
version from HuggingFace.
- This should make it possible to register InvokeAI at PyPi and do a
fully automated pip-based install.
- Minor regression: it is no longer possible to specify which device the
clipseg model will be loaded into, and it will reside in CPU. However,
performance is more than acceptable.
2023-01-26 12:14:13 -05:00
Lincoln Stein
d3a469d136 fix location of textual_inversion script 2023-01-26 11:56:23 -05:00
Lincoln Stein
e79f89b619 improve initial model configuration 2023-01-26 11:53:06 -05:00
Lincoln Stein
cbd967cbc4 add documentation caveat about location of HF cached models 2023-01-26 11:48:03 -05:00
damian
e090c0dc10 try without setting every time 2023-01-26 17:46:51 +01:00
damian
c381788ab9 don't restore None 2023-01-26 17:44:27 +01:00
damian
fb312f9ed3 use the correct value - whoops 2023-01-26 17:30:29 +01:00
damian
729752620b trying out JPPhoto's patch on vast.ai 2023-01-26 17:27:33 +01:00
damian
8ed8bf52d0 use 'auto' slice size 2023-01-26 17:04:22 +01:00
Lincoln Stein
a49d546125 simplified code a bit 2023-01-26 09:46:34 -05:00
Lincoln Stein
288e31fc60 remove dependency on original clipseg library
- This replaces the original clipseg library with the transformers
  version from HuggingFace.
- This should make it possible to register InvokeAI at PyPi and do
  a fully automated pip-based install.
- Minor regression: it is no longer possible to specify which device
  the clipseg model will be loaded into, and it will reside in CPU.
  However, performance is more than acceptable.
2023-01-26 09:35:16 -05:00
Lincoln Stein
7b2c0d12a3 add missing VAEs to initial diffuser models 2023-01-26 00:25:39 -05:00
Kevin Turner
2978c3eb8d
Merge branch 'main' into update-diffusers 2023-01-25 18:42:00 -08:00
Damian Stewart
5e7ed964d2 wip updating docs 2023-01-25 23:49:38 +01:00
Damian Stewart
93a24445dc Merge remote-tracking branch 'upstream/main' into diffusers_cross_attention_control_reimplementation 2023-01-25 23:05:39 +01:00
Damian Stewart
95d147c5df MPS support: negatory 2023-01-25 23:03:30 +01:00
Damian Stewart
41aed57449 wip tracking down MPS slicing support 2023-01-25 22:27:23 +01:00
Damian Stewart
34a3f4a820 cleanup 2023-01-25 21:47:17 +01:00
Damian Stewart
1f5ad1b05e sliced swap working 2023-01-25 21:38:27 +01:00