Commit Graph

63 Commits

Author SHA1 Message Date
Manuel Schmid
5b7ddf8b22
feat: advanced params refactoring + prevent users from skipping/stopping other users tasks in queue (#981)
* only make stop_button and skip_button interactive when rendering process starts

fix inconsistency in behaviour of stop_button and skip_button as it was possible to skip or stop other users processes while still being in queue

* use AsyncTask for last_stop handling instead of shared

* Revert "only make stop_button and skip_button interactive when rendering process starts"

This reverts commit d3f9156854.

* introduce state for task skipping/stopping

* fix return parameters of stop_clicked

* code cleanup, do not disable skip/stop on stop_clicked

* reset last_stop when skipping for further processing

* fix: replace fcbh with ldm_patched

* fix: use currentTask instead of ctrls after merging upstream

* feat: extract attribute disable_preview

* feat: extract attribute adm_scaler_positive

* feat: extract attribute adm_scaler_negative

* feat: extract attribute adm_scaler_end

* feat: extract attribute adaptive_cfg

* feat: extract attribute sampler_name

* feat: extract attribute scheduler_name

* feat: extract attribute generate_image_grid

* feat: extract attribute overwrite_step

* feat: extract attribute overwrite_switch

* feat: extract attribute overwrite_width

* feat: extract attribute overwrite_height

* feat: extract attribute overwrite_vary_strength

* feat: extract attribute overwrite_upscale_strength

* feat: extract attribute mixing_image_prompt_and_vary_upscale

* feat: extract attribute mixing_image_prompt_and_inpaint

* feat: extract attribute debugging_cn_preprocessor

* feat: extract attribute skipping_cn_preprocessor

* feat: extract attribute canny_low_threshold

* feat: extract attribute canny_high_threshold

* feat: extract attribute refiner_swap_method

* feat: extract freeu_ctrls attributes

freeu_enabled, freeu_b1, freeu_b2, freeu_s1, freeu_s2

* feat: extract inpaint_ctrls attributes

debugging_inpaint_preprocessor, inpaint_disable_initial_latent, inpaint_engine, inpaint_strength, inpaint_respective_field, inpaint_mask_upload_checkbox, invert_mask_checkbox, inpaint_erode_or_dilate

* wip: add TODOs

* chore: cleanup code

* feat: extract attribute controlnet_softness

* feat: extract remaining attributes, do not use globals in patch

* fix: resolve circular import, patch_all now in async_worker

* chore: cleanup pid code
2024-02-24 19:01:06 +01:00
lllyasviel
8e62a72a63
(requested) support AMD 8GB GPUs via Windows DirectML
this update is requested by users
2023-12-30 06:30:59 -08:00
lllyasviel
67808d5ee5
fix all precision issues
We fixed number precision issues again. Now 2.1.849 will give 100% exactly same results as 2.1.824.
2023-12-16 19:54:05 -08:00
lllyasviel
085bc3aeaa
revise caster 2023-12-16 16:58:46 -08:00
lllyasviel
ec5dd950a2
Fix many precision problems
Many users reported that image quality is different from 2.1.824. We reviewed all codes and fixed several precision problems in 2.1.846.
2023-12-16 15:55:53 -08:00
lllyasviel
179bcb2c4e
Fix inpaint device problem in --always-gpu mode. (#1420)
2.1.841
2023-12-14 19:15:16 -08:00
lllyasviel
323af5667a
2.1.840
* Fix some potential frozen after model mismatch
* Fix crash when cfg=1 when using anime preset
* Added some guidelines for troubleshoot the "CUDA kernel errors asynchronously" problem
2023-12-14 13:55:49 -08:00
lllyasviel
5b99e3a1e4
2.1.839 2023-12-13 21:14:50 -08:00
lllyasviel
df615d3781 fix some precision problems 2023-12-13 13:05:33 -08:00
lllyasviel
9c3fb90df3 try fix cpu tree 2023-12-12 14:14:30 -08:00
lllyasviel
e8d88d3e25 2.1.826 2023-12-12 11:38:05 -08:00
lllyasviel
3bc9ac88fd maintain 2023-11-23 13:46:50 -08:00
lllyasviel
dececbd060
[2.1.822] New Inpaint System
See related documents for more details.
2023-11-19 17:37:22 -08:00
lllyasviel
675805960a
2.1.821
* New UI for LoRAs.
* Improved preset system: normalized preset keys and file names.
* Improved session system: now multiple users can use one Fooocus at the same time without seeing others' results.
* Improved some computation related to model precision.
* Improved config loading system with user-friendly prints.
2023-11-17 11:25:39 -08:00
lllyasviel
09f70de40e fix math 2023-11-11 09:03:03 -08:00
lllyasviel
4fe08161a5 2.1.782
2.1.782
2023-11-11 01:43:01 -08:00
lllyasviel
49f47a6f5e try fix #849 2023-11-05 03:44:18 -08:00
lllyasviel
38e70cebcc Update Backend
Update Backend
2023-10-25 09:40:13 -07:00
lllyasviel
10a9f0fc9d try fix some mps problems 2023-10-23 20:19:24 -07:00
lllyasviel
504e5478b0 remove unused/unstable codes 2023-10-23 15:09:24 -07:00
lllyasviel
0b90fd9e8e Revert "fix autocast in less aggressive way"
This reverts commit 7a6775acdc.
2023-10-23 14:22:35 -07:00
lllyasviel
7a6775acdc fix autocast in less aggressive way 2023-10-23 14:15:20 -07:00
lllyasviel
47281e52c4 Fixed many autocast problems. 2023-10-23 13:07:55 -07:00
lllyasviel
6f3fa4b670 remove unused codes 2023-10-22 09:00:16 -07:00
lllyasviel
7d81eeed7e revise noise formulation
revise noise formulation
2023-10-22 06:02:35 -07:00
lllyasviel
e5f614c14e disable memory 2023-10-17 23:56:20 -07:00
lvmin
050a3bc23a remove unused files 2023-10-15 17:59:32 -07:00
lvmin
9ee64514c8 file corruption checking system 2023-10-15 17:51:42 -07:00
lvmin
a27ca7eabb file corruption checking system 2023-10-15 17:49:10 -07:00
lvmin
cdf22b8ccc revise code 2023-10-14 06:23:29 -07:00
lvmin
a60615de61 remove unused codes 2023-10-14 06:10:20 -07:00
lvmin
b3d662054f using global sigmas for better results 2023-10-14 06:06:34 -07:00
lvmin
e27383c07f finally remove torchsde warning 2023-10-14 04:50:13 -07:00
lvmin
10e2db67d6 log 2023-10-13 04:53:02 -07:00
lllyasviel
e61aac34ca
sync (#658) 2023-10-12 04:23:10 -07:00
lllyasviel
132afcc2a2
rework refiner
rework refiner
2023-10-11 23:44:40 -07:00
lllyasviel
a960d24825
better memory management (#629)
* better memory management

* better memory management
2023-10-10 13:44:01 -07:00
lllyasviel
1601af43be
better memory management (#628) 2023-10-10 13:27:31 -07:00
lllyasviel
53573fa073
fix auto_cast (#622) 2023-10-10 02:36:19 -07:00
lllyasviel
e00b647eee
improve result quality of control (#581) 2023-10-08 14:33:16 -07:00
lllyasviel
6faaac333b
i (#559) 2023-10-07 22:54:04 -07:00
lllyasviel
6189e482bd
improve human eyes again by using crop adm rather than interpolate adm (#544)
and add an debug slider
2023-10-04 21:42:43 -07:00
lllyasviel
ba58dbc8bc
speed up again (#527) 2023-10-03 17:14:38 -07:00
lllyasviel
480a7222c5
Improved the scheduling of ADM guidance and CFG mimicking for better visual quality in high frequency domain and small objects.
Improved the scheduling of ADM guidance and CFG mimicking for better visual quality in high frequency domain and small objects.
2023-10-03 14:05:14 -07:00
lllyasviel
bbae307ef2
2.0.80 (#520)
* Rework many patches and some UI details.
* Speed up processing.
* Move Colab to independent branch.
* Implemented CFG Scale and TSNR correction when CFG is bigger than 10.
* Implemented Developer Mode with more options to debug.
2023-10-03 10:36:42 -07:00
lllyasviel
cdf642437c
bfloat16 vae (#456)
* bfloat16 vae

* bfloat16 vae

* bfloat16 vae
2023-09-20 08:16:20 -07:00
lllyasviel
969a54e85b
fix some unknown clip problem (#439) 2023-09-19 13:14:25 -07:00
lllyasviel
0927445492
use fooocus inpaint control model (#429)
use fooocus inpaint control model (#429)
2023-09-19 04:52:22 -07:00
lllyasviel
b61642ecba
[Fooocus 2.0.60] Fooocus Inpaint or Outpaint (Midjourney Left/Right/Top/Bottom) (#402)
[Fooocus 2.0.60] Fooocus Inpaint or Outpaint (Midjourney Left/Right/Top/Bottom) (#402)
2023-09-18 01:16:07 -07:00
lllyasviel
7b73152fe3
use a1111 emphasizing (#399) 2023-09-16 15:01:34 -07:00