Commit Graph

443 Commits

Author SHA1 Message Date
hlky
dab958377e
#287 2022-08-29 23:52:49 +01:00
_nderscore
7ddbd6533a feat: improve integration with painterro image editor 2022-08-29 23:35:49 +01:00
hlky
9f8516b493
Fix for #287 2022-08-29 23:31:45 +01:00
hlky
4878a68b40
Update webui.py 2022-08-29 19:16:41 +01:00
hlky
3fb8e64a7d
remove default=int(os.environ.get('CUDA_VISIBLE_DEVICES', 0) 2022-08-29 19:14:24 +01:00
hlky
5eae4377ce f"cuda:{opt.gpu}" 2022-08-29 19:10:18 +01:00
hlky
0534b89ff5 --extra-models-gpu --gfpgan-gpu <id> --esrgan-gpu <id>
This will let you set the gpu id which gfgpan and esrgan load on. I don't have multiple gpus so i can't test it properly myself but the code is working without errors when I use --extra-models-gpu --gfpgan-gpu 0 --esrgan-gpu 0

Note: you will want to use this in combination with --gpu to ensure the main model loads of the gpu you want it to, otherwise the default is id 0
2022-08-29 19:10:18 +01:00
oobabooga
e253a27a0f Fix OOM issue
modelFS must be removed from GPU memory after each batch is generated, otherwise both modelFS and modelCS will be loaded into memory at once and the script will OOM
2022-08-29 17:51:49 +01:00
Oxlamon
10cf5e5317 Grid rows and batch count support. 2022-08-29 15:53:21 +01:00
jansteffen
59e1e8fe17 Better robustness of styling for prompt and submit 2022-08-29 15:20:53 +01:00
hlky
c1ad047a82
Revert "⚠️ note to everyone dont mess with line endings again ⚠️"
This reverts commit e6d06a2b24.
2022-08-29 13:44:40 +01:00
hlky
33ebc61204
Revert "⚠️ note to everyone dont mess with line endings again ⚠️"
This reverts commit 444117f3ba.
2022-08-29 13:44:39 +01:00
hlky
023e9a4038
Delete .gitattributes 2022-08-29 12:53:29 +01:00
hlky
444117f3ba
⚠️ note to everyone dont mess with line endings again ⚠️ 2022-08-29 12:48:17 +01:00
hlky
e6d06a2b24
⚠️ note to everyone dont mess with line endings again ⚠️ 2022-08-29 12:46:20 +01:00
Lex Darlog (DRL)
a90f49a56c fix *.sh line to be uniform with the others 2022-08-29 12:42:45 +01:00
Lex Darlog (DRL)
20a04a1ec3 .gitattributes from https://github.com/hlky/stable-diffusion/pull/91 2022-08-29 12:42:45 +01:00
Lex Darlog (DRL)
7698371d72 .gitignore from https://github.com/hlky/stable-diffusion/pull/55 + project-specific rules removed 2022-08-29 12:42:45 +01:00
hlky
2974abae30
Update README.md 2022-08-29 12:28:31 +01:00
_nderscore
0f28871a64 feat: add denoising strength to parameter output and pnginfo 2022-08-29 12:21:58 +01:00
hlky
2ae6be00ef Revert "Set auto prompt_matrix depending on prompt"
This reverts commit 6d5dd00981.
2022-08-29 12:20:40 +01:00
hlky
4d0572cfc1 Revert "Update matrix prompt generation code"
This reverts commit 2891456369.
2022-08-29 12:20:40 +01:00
hlky
d28ea40140 Revert "Bug fix"
This reverts commit 45db349689.
2022-08-29 12:20:40 +01:00
hlky
9156c8b834 Revert "Removed auto prompt_matrix check (revert)"
This reverts commit 0b2f5bf46a.
2022-08-29 12:20:40 +01:00
hlky
c0cd601d6a Revert "Removed [obsolete] text in check box"
This reverts commit ca9e20905f.
2022-08-29 12:20:40 +01:00
Oxlamon
ca9e20905f Removed [obsolete] text in check box 2022-08-29 12:20:14 +01:00
Oxlamon
0b2f5bf46a Removed auto prompt_matrix check (revert) 2022-08-29 12:20:14 +01:00
Oxlamon
45db349689 Bug fix 2022-08-29 12:20:14 +01:00
Oxlamon
2891456369 Update matrix prompt generation code 2022-08-29 12:20:14 +01:00
Oxlamon
6d5dd00981 Set auto prompt_matrix depending on prompt 2022-08-29 12:20:14 +01:00
Nelson Chen
4fac0f2558 Update frontend.py
typo fix small
2022-08-29 12:19:51 +01:00
jansteffen
899b811dd0 Keep the generate button (visually) inside the text box 2022-08-29 12:19:35 +01:00
_nderscore
5ff04a5a52 fix(css): prevent generate button from overlapping prompt, fix font size on multiline prompts 2022-08-29 12:19:35 +01:00
hlky
b4eb2c1d3a
Create LICENSE 2022-08-29 12:16:23 +01:00
hlky
083ffd9ad8
Update README.md 2022-08-29 00:59:36 +01:00
hlky
90d49a0c4a
Update README.md 2022-08-29 00:57:44 +01:00
hlky
45ad73e4f3 Update sync.yml 2022-08-29 00:41:17 +01:00
hlky
d88ca21491 Update sync.yml 2022-08-29 00:41:17 +01:00
Alex Volkov
923e437bcf Resize image was exported as well
Signed-off-by: Alex Volkov <alex.volkov@fundbox.com>
2022-08-29 00:41:17 +01:00
Alex Volkov
8662ea4623 Ugh I only added these to the playground! Added now 2022-08-29 00:41:17 +01:00
Alex Volkov
8781040614 Added gitignore 2022-08-29 00:41:17 +01:00
hlky
6b148f4858 Update sync.yml 2022-08-29 00:41:17 +01:00
hlky
783b7c6691 Update sync.yml 2022-08-29 00:41:17 +01:00
Alex Volkov
64ca78ca76 Refactored the UI, added a web_playground.py to be run without any models. 2022-08-29 00:41:17 +01:00
hlky
8d15f865e9
re-fix esrgan/gfpgan multiple images bug
Must have been undone by some pr

also, grid is now only appended to output images if more than 1 images is being generated
2022-08-28 22:12:59 +01:00
hlky
fbd4878551
Update README.md 2022-08-28 20:29:31 +01:00
_nderscore
ea4bc1aa53 fix: working loopback buttons on txt2img and img2img 2022-08-28 19:37:14 +01:00
jansteffen
244f945c40 Revamped output info area 2022-08-28 17:25:02 +01:00
hlky
1afcbe9fa0
Update README.md 2022-08-28 17:24:01 +01:00
hlky
7c6e04b9a8
Update README.md 2022-08-28 17:06:02 +01:00