Commit Graph

370 Commits

Author SHA1 Message Date
catboxanon
06ab10a1be Normalize cmd arg paths
In particular, this fixes an issue on Windows where some functions
will misbehave if forward slashes are provided rather than
double backslashes.
2024-02-15 14:22:13 -05:00
Kohaku-Blueleaf
90441294db
Add rescale mechanism
LyCORIS will support save oft_blocks instead of oft_diag in the near future (for both OFT and BOFT)

But this means we need to store the rescale if user enable it.
2024-02-12 14:25:09 +08:00
v0xie
eb6f2df826 Revert "fix: add butterfly_factor fn"
This reverts commit 81c16c965e.
2024-02-08 22:00:15 -08:00
v0xie
613b0d9548 doc: add boft comment 2024-02-08 21:58:59 -08:00
v0xie
325eaeb584 fix: get boft params from weight shape 2024-02-08 11:55:05 -08:00
v0xie
2f1073dc6e style: fix lint 2024-02-07 04:55:11 -08:00
v0xie
81c16c965e fix: add butterfly_factor fn 2024-02-07 04:54:14 -08:00
v0xie
a4668a16b6 fix: calculate butterfly factor 2024-02-07 04:51:22 -08:00
v0xie
9588721197 feat: support LyCORIS BOFT 2024-02-07 04:49:17 -08:00
AUTOMATIC1111
358e9e2847
Merge pull request #14726 from v0xie/fix-oft-device
Fix kohya-ss OFT network wrong device for eye and constraint
2024-01-23 22:34:24 +03:00
AUTOMATIC1111
569dc1919c
Merge pull request #14588 from Sj-Si/feature/extra-networks-tree-view
Feature: Extra Networks Tree View
2024-01-22 22:24:06 +03:00
v0xie
fd383140cf fix: wrong devices for eye and constraint 2024-01-22 02:52:34 -08:00
AUTOMATIC1111
0f2de4cfbe
Merge pull request #14655 from chi2nagisa/lora-alias-fix
[bug] fix using wrong model caused by alias
2024-01-20 10:27:54 +03:00
WebDev
30f31d23c6
Update hotkey_config.py 2024-01-19 16:39:33 -08:00
WebDev
bcdcc8be7d
Update hotkey_config.py 2024-01-19 16:39:07 -08:00
WebDev
4dde96109b
Update zoom.js 2024-01-19 16:38:34 -08:00
chi2nagisa
e280eb4055 fix using wrong model caused by alias 2024-01-16 03:45:19 +08:00
WebDev
39db1d09d1
Update zoom.js 2024-01-14 01:43:23 -08:00
WebDev
92032f2da8
Update hotkey_config.py 2024-01-14 01:42:03 -08:00
WebDev
541881e318
Adjust brush size with hotkeys. 2024-01-13 02:11:06 -08:00
Sj-Si
036500223d Merge changes from dev 2024-01-11 16:37:35 -05:00
Sj-Si
0726a6e12e Finish base layout. Fix bugs. Need to test for stability and clean up. 2024-01-11 15:06:57 -05:00
AUTOMATIC1111
8b6848c6db
Merge pull request #14546 from AUTOMATIC1111/fix-oft-dtype
Fix dtype casting in OFT module
2024-01-06 10:50:38 +03:00
Kohaku-Blueleaf
44744d6005 linting 2024-01-05 16:38:05 +08:00
Kohaku-Blueleaf
18ca987c92 Add general forward method for all modules. 2024-01-05 16:32:19 +08:00
Kohaku-Blueleaf
f8f38c7c28 Fix dtype casting for OFT module 2024-01-05 16:31:48 +08:00
Aarni Koskela
3d31d5c27b SwinIR: pass model.scale 2024-01-03 22:38:49 +02:00
Aarni Koskela
dfdc51246c SwinIR: use prefer_half 2024-01-03 22:38:13 +02:00
Aarni Koskela
cf14a6a7aa Refactor upscale_2 helper out of ScuNET/SwinIR; make sure devices are right 2024-01-02 08:57:12 +02:00
AUTOMATIC1111
5d7d1823af rename infotext.py again, this time to infotext_utils.py; I didn't realize infotext would be used for variable names in multiple places, which makes it awkward to import the module; also fix the bug I caused by this rename that breaks tests 2024-01-01 17:25:30 +03:00
AUTOMATIC1111
d859cec696 infotext.py: rename usages in the codebase 2024-01-01 13:53:12 +03:00
AUTOMATIC1111
f3af8c8d04
Merge pull request #14475 from Learwin/negative_prompt
Adding negative prompts to Loras in extra networks
2023-12-31 22:32:28 +03:00
Learwin
d4945f4422 Removed weight slider for negative prompts 2023-12-31 13:22:30 +01:00
Aarni Koskela
6f86b62a1b Deduplicate tiled inference code from SwinIR/ScuNET 2023-12-31 01:13:30 +02:00
Aarni Koskela
777af661a2 Be more clear about Spandrel model nomenclature 2023-12-31 00:22:58 +02:00
Learwin
a2f23f9d22 Code Style fixes 2023-12-30 22:16:51 +01:00
Learwin
bc5ae74c7d Added negative prompts to extra networks lora 2023-12-30 21:52:27 +01:00
Aarni Koskela
4ad0c0c0a8 Verify architecture for loaded Spandrel models 2023-12-30 16:37:03 +02:00
Aarni Koskela
b0f5934234 Use Spandrel for upscaling and face restoration architectures (aside from GFPGAN and LDSR) 2023-12-30 16:24:01 +02:00
w-e-w
59d060fd5e More lora not found warning 2023-12-30 17:11:03 +09:00
AUTOMATIC1111
93eae69895 move soft inpainting to a built-in extension 2023-12-16 11:00:42 +03:00
Kohaku-Blueleaf
0fb34b57b8 Merge branch 'dev' into test-fp8 2023-12-14 16:54:45 +08:00
AUTOMATIC1111
2be85f8fe0 Merge pull request #14237 from ReneKroon/dev
#13354 : solve lora loading issue
2023-12-14 10:15:36 +03:00
AUTOMATIC1111
b55f09c4e1 Merge pull request #14270 from kaalibro/extra-options-elem-id
Assign id for "extra_options". Replace numeric field with slider.
2023-12-14 10:14:46 +03:00
AUTOMATIC1111
6ef0ff39f2 Merge pull request #14300 from AUTOMATIC1111/oft_fixes
Fix wrong implementation in network_oft
2023-12-14 10:14:19 +03:00
AUTOMATIC1111
778a30a95e
Merge pull request #14237 from ReneKroon/dev
#13354 : solve lora loading issue
2023-12-14 10:08:03 +03:00
AUTOMATIC1111
c5631aa90d
Merge pull request #14270 from kaalibro/extra-options-elem-id
Assign id for "extra_options". Replace numeric field with slider.
2023-12-14 09:46:05 +03:00
Kohaku-Blueleaf
3772a82a70 better naming and correct order for device. 2023-12-14 01:47:13 +08:00
Kohaku-Blueleaf
8fc67f3851 remove debug print 2023-12-14 01:44:49 +08:00
Kohaku-Blueleaf
265bc26c21 Use self.scale instead of custom finalize 2023-12-14 01:43:24 +08:00
Kohaku-Blueleaf
735c9e8059 Fix network_oft 2023-12-14 01:38:32 +08:00
kaalibro
6b8143a84e
Number of columns slider: max count set to 20, add description info 2023-12-10 15:35:06 +06:00
kaalibro
1a79a5049b
Assign id for "extra_options". Replace numeric field with slider in Settings. 2023-12-09 22:35:31 +06:00
Rene Kroon
16bdcce92d #13354: solve lora loading issue 2023-12-08 21:20:55 +01:00
Kohaku-Blueleaf
294ec5ac37 Merge branch 'dev' into test-fp8 2023-12-06 15:16:49 +08:00
AUTOMATIC1111
368d66c9cc add hypertile infotext 2023-12-04 15:56:11 +03:00
AUTOMATIC1111
22e23dbf29 add hypertile infotext 2023-12-04 15:56:03 +03:00
Kohaku-Blueleaf
110485d5bb Merge branch 'dev' into test-fp8 2023-12-02 17:00:09 +08:00
AUTOMATIC1111
7547d7c791
Merge pull request #14126 from aria1th/hypertile-xyz
Support XYZ scripts / split hires path from unet
2023-12-02 09:48:40 +03:00
AUTOMATIC1111
e294e46d46 split UI settings page into many 2023-12-02 09:26:38 +03:00
aria1th
ec78354efa hypertile_xyz: we don't need isnumeric check for AxisOption 2023-11-27 22:25:28 +09:00
aria1th
524d6a4dba fix ruff - set comprehension 2023-11-27 22:13:18 +09:00
aria1th
f207eb7a0d fix ruff in hypertile_xyz.py 2023-11-27 22:11:28 +09:00
aria1th
601a7b4ce5 cache divisors / fix ruff 2023-11-27 22:10:31 +09:00
aria1th
23c36f59b4 Support XYZ scripts / split hires path from unet 2023-11-27 21:10:26 +09:00
Kohaku-Blueleaf
3d341ebc7d Merge branch 'dev' into test-fp8 2023-11-26 17:32:52 +08:00
AUTOMATIC1111
d1750e5eca fix linter errors 2023-11-26 11:37:12 +03:00
AUTOMATIC1111
f85b74763d Merge branch 'hypertile-in-sample' into dev 2023-11-26 11:18:49 +03:00
AUTOMATIC1111
d2e0c1ca13 rework hypertile into a built-in extension 2023-11-26 11:17:38 +03:00
AUTOMATIC1111
3a9bf4ac10 move file 2023-11-26 08:29:12 +03:00
Kohaku-Blueleaf
370a77f8e7 Option for using fp16 weight when apply lora 2023-11-21 19:59:34 +08:00
Kohaku-Blueleaf
b60e1088db Merge branch 'dev' into test-fp8 2023-11-19 15:24:57 +08:00
AUTOMATIC1111
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
AUTOMATIC1111
2207ef363a
Merge pull request #13692 from v0xie/network-oft
Support inference with OFT networks
2023-11-19 08:59:09 +03:00
Kohaku-Blueleaf
cd12256575 Merge branch 'dev' into test-fp8 2023-11-16 21:53:13 +08:00
v0xie
eb667e715a feat: LyCORIS/kohya OFT network support 2023-11-15 18:28:48 -08:00
v0xie
d6d0b22e66 fix: ignore calc_scale() for COFT which has very small alpha 2023-11-15 03:08:50 -08:00
AUTOMATIC1111
4d4a9e7332 added compact prompt option 2023-11-05 19:19:55 +03:00
v0xie
bbf00a96af refactor: remove unused function 2023-11-04 14:56:47 -07:00
v0xie
329c8bacce refactor: use same updown for both kohya OFT and LyCORIS diag-oft 2023-11-04 14:54:36 -07:00
v0xie
f6c8201e56 refactor: move factorization to lyco_helpers, separate calc_updown for kohya and kb 2023-11-03 19:35:15 -07:00
v0xie
fe1967a4c4 skip multihead attn for now 2023-11-03 17:52:55 -07:00
v0xie
d727ddfccd no idea what i'm doing, trying to support both type of OFT, kblueleaf diag_oft has MultiheadAttn which kohya's doesn't?, attempt create new module based off network_lora.py, errors about tensor dim mismatch 2023-11-02 00:13:11 -07:00
v0xie
65ccd6305f detect diag_oft type 2023-11-02 00:11:32 -07:00
v0xie
a2fad6ee05 test implementation based on kohaku diag-oft implementation 2023-11-01 22:34:27 -07:00
v0xie
6523edb8a4 style: conform style 2023-10-22 09:31:15 -07:00
v0xie
3b8515d2c9 fix: multiplier applied twice in finalize_updown 2023-10-22 09:27:48 -07:00
v0xie
4a50c9638c refactor: remove used OFT functions 2023-10-22 08:54:24 -07:00
v0xie
de8ee92ed8 fix: use merge_weight to cache value 2023-10-21 17:37:17 -07:00
v0xie
76f5abdbdb style: cleanup oft 2023-10-21 16:07:45 -07:00
v0xie
fce86ab7d7 fix: support multiplier, no forward pass hook 2023-10-21 16:03:54 -07:00
v0xie
7683547728 fix: return orig weights during updown, merge weights before forward 2023-10-21 14:42:24 -07:00
v0xie
2d8c894b27 refactor: use forward hook instead of custom forward 2023-10-21 13:43:31 -07:00
v0xie
0550659ce6 style: fix ambiguous variable name 2023-10-19 13:13:02 -07:00
v0xie
d10c4db57e style: formatting 2023-10-19 12:52:14 -07:00
v0xie
321680ccd0 refactor: fix constraint, re-use get_weight 2023-10-19 12:41:17 -07:00
Kohaku-Blueleaf
7c128bbdac Add fp8 for sd unet 2023-10-19 13:56:17 +08:00
v0xie
eb01d7f0e0 faster by calculating R in updown and using cached R in forward 2023-10-18 04:56:53 -07:00
v0xie
853e21d98e faster by using cached R in forward 2023-10-18 04:27:44 -07:00
v0xie
1c6efdbba7 inference working but SLOW 2023-10-18 04:16:01 -07:00