search

huchenlei / sd-forge-ic-light

A1111/SD Forge extension for IC-Light
Apache License 2.0
198 stars 7 forks source link

after updating 6.7 version, it doesn't work occasionally at the first inference right after launching webui #25

Closed greasebig closed 4 months ago

greasebig commented 5 months ago

seem like it didn't load iclight and directly text2img generate

greasebig commented 5 months ago
image image

it is kinda amusing :)

greasebig commented 5 months ago

i find out the reason: EVERYTIME restart webui, the first time it don't work，and it will work later. definitely bug

sbcdba commented 5 months ago

SD-forge work error.if use cpu start，is good，but use RTX 1070 is bad.if SD is all good. 微信图片_20240608145039

greasebig commented 5 months ago

i dont know why, but today i test again your extension work well

greasebig commented 5 months ago

today, i meet the problem again. but it seems to occur occasionally

hben35096 commented 4 months ago

For use in A1111, A1111 version: feee37d75f1b168768014e4634dcb156ee649c05

sd-forge-ic-light version: b8268e16a0b99ed7dffff930a18ad475372a49ae is working normally PixPin_2024-07-01_04-11-26

/root/autodl-tmp/stable-diffusion-webui
Python 3.10.8 (main, Nov 24 2022, 14:13:03) [GCC 11.2.0]
Version: v1.9.4
Commit hash: feee37d75f1b168768014e4634dcb156ee649c05
Launching Web UI with arguments: --port=6006 --xformers --theme=dark --enable-insecure-extension-access
[-] ADetailer initialized. version: 24.6.0, num models: 15
/root/miniconda3/lib/python3.10/site-packages/numba/np/ufunc/parallel.py:371: NumbaWarning: The TBB threading layer requires TBB version 2021 update 6 or later i.e., TBB_INTERFACE_VERSION >= 12060. Found TBB_INTERFACE_VERSION = 12050. The TBB threading layer is disabled.
  warnings.warn(problem)
ControlNet preprocessor location: /root/autodl-tmp/stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/downloads
2024-07-01 03:36:52,296 - ControlNet - INFO - ControlNet v1.1.449
Please 'pip install apex'
2024-07-01 03:36:52,681 - model_patcher_hook.py - INFO - __init__ hooks applied
INFO:model_patcher_hook.py:__init__ hooks applied
2024-07-01 03:36:52,681 - model_patcher_hook.py - INFO - close hooks applied
INFO:model_patcher_hook.py:close hooks applied
2024-07-01 03:36:52,681 - model_patcher_hook.py - INFO - sample hooks applied
INFO:model_patcher_hook.py:sample hooks applied
sd-webui-prompt-all-in-one background API service started successfully.
Loading weights [ec41bd2a82] from /root/autodl-tmp/stable-diffusion-webui/models/Stable-diffusion/photon_v1.safetensors
2024-07-01 03:36:53,168 - ControlNet - INFO - ControlNet UI callback registered.
Running on local URL:  http://127.0.0.1:6006/
Creating model from config: /root/autodl-tmp/stable-diffusion-webui/configs/v1-inference.yaml
Applying attention optimization: xformers... done.
2024-07-01 03:36:55,861 - model_patcher_hook.py - INFO - Init p.model_patcher.
INFO:model_patcher_hook.py:Init p.model_patcher.
2024-07-01 03:36:55,879 - model_patcher_hook.py - INFO - Init p.hr_model_patcher.
INFO:model_patcher_hook.py:Init p.hr_model_patcher.

To create a public link, set `share=True` in `launch()`.
IIB Database file has been successfully backed up to the backup folder.
Startup time: 17.0s (prepare environment: 2.4s, import torch: 3.6s, import gradio: 1.6s, setup paths: 1.9s, initialize shared: 0.2s, other imports: 0.5s, load scripts: 3.3s, create ui: 1.0s, gradio launch: 2.2s, app_started_callback: 0.2s).
Model loaded in 3.5s (load weights from disk: 1.1s, create model: 0.5s, apply weights to model: 0.9s, load textual inversion embeddings: 0.4s, calculate empty prompt: 0.5s).
2024-07-01 03:37:24,689 - model_patcher_hook.py - INFO - Init p.model_patcher.
INFO:model_patcher_hook.py:Init p.model_patcher.
2024-07-01 03:37:24,708 - model_patcher_hook.py - INFO - Init p.hr_model_patcher.
INFO:model_patcher_hook.py:Init p.hr_model_patcher.
2024-07-01 03:37:30,977 - model_patcher_hook.py - INFO - Patch model_patcher of StableDiffusionProcessingTxt2Img.
INFO:model_patcher_hook.py:Patch model_patcher of StableDiffusionProcessingTxt2Img.
  0%|                                                    | 0/20 [00:00<?, ?it/s]
  5%|██▏                                         | 1/20 [00:00<00:08,  2.18it/s]
 15%|██████▌                                     | 3/20 [00:00<00:03,  5.29it/s]
 25%|███████████                                 | 5/20 [00:00<00:02,  7.31it/s]
 35%|███████████████▍                            | 7/20 [00:00<00:01,  9.13it/s]
 45%|███████████████████▊                        | 9/20 [00:01<00:01, 10.46it/s]
 55%|███████████████████████▋                   | 11/20 [00:01<00:00, 11.42it/s]
 65%|███████████████████████████▉               | 13/20 [00:01<00:00, 12.11it/s]
 75%|████████████████████████████████▎          | 15/20 [00:01<00:00, 12.43it/s]
 85%|████████████████████████████████████▌      | 17/20 [00:01<00:00, 12.80it/s]
100%|███████████████████████████████████████████| 20/20 [00:01<00:00, 10.41it/s]
2024-07-01 03:37:32,913 - model_patcher_hook.py - INFO - Unpatch model_patcher of StableDiffusionProcessingTxt2Img.
INFO:model_patcher_hook.py:Unpatch model_patcher of StableDiffusionProcessingTxt2Img.
2024-07-01 03:37:33,206 - model_patcher_hook.py - INFO - Close p.hr_model_patcher.
INFO:model_patcher_hook.py:Close p.hr_model_patcher.
2024-07-01 03:37:33,207 - model_patcher_hook.py - INFO - Close p.model_patcher.
INFO:model_patcher_hook.py:Close p.model_patcher.

Total progress: 100%|███████████████████████████| 20/20 [00:01<00:00, 11.29it/s]

But since the sd-forge-ic-light version: 53d13bbf26383bbbfeea6ea49a1e3adff68e4e13 image

/root/autodl-tmp/stable-diffusion-webui
Python 3.10.8 (main, Nov 24 2022, 14:13:03) [GCC 11.2.0]
Version: v1.9.4
Commit hash: feee37d75f1b168768014e4634dcb156ee649c05
Launching Web UI with arguments: --port=6006 --xformers --theme=dark --enable-insecure-extension-access
[-] ADetailer initialized. version: 24.6.0, num models: 15
/root/miniconda3/lib/python3.10/site-packages/numba/np/ufunc/parallel.py:371: NumbaWarning: The TBB threading layer requires TBB version 2021 update 6 or later i.e., TBB_INTERFACE_VERSION >= 12060. Found TBB_INTERFACE_VERSION = 12050. The TBB threading layer is disabled.
  warnings.warn(problem)
ControlNet preprocessor location: /root/autodl-tmp/stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/downloads
2024-07-01 03:32:17,246 - ControlNet - INFO - ControlNet v1.1.449
Please 'pip install apex'
2024-07-01 03:32:17,608 - model_patcher_hook.py - INFO - __init__ hooks applied
INFO:model_patcher_hook.py:__init__ hooks applied
2024-07-01 03:32:17,608 - model_patcher_hook.py - INFO - close hooks applied
INFO:model_patcher_hook.py:close hooks applied
2024-07-01 03:32:17,608 - model_patcher_hook.py - INFO - sample hooks applied
INFO:model_patcher_hook.py:sample hooks applied
sd-webui-prompt-all-in-one background API service started successfully.
Loading weights [ec41bd2a82] from /root/autodl-tmp/stable-diffusion-webui/models/Stable-diffusion/photon_v1.safetensors
2024-07-01 03:32:18,076 - ControlNet - INFO - ControlNet UI callback registered.
Running on local URL:  http://127.0.0.1:6006/
Creating model from config: /root/autodl-tmp/stable-diffusion-webui/configs/v1-inference.yaml
Applying attention optimization: xformers... done.
2024-07-01 03:32:20,745 - model_patcher_hook.py - INFO - Init p.model_patcher.
INFO:model_patcher_hook.py:Init p.model_patcher.
2024-07-01 03:32:20,764 - model_patcher_hook.py - INFO - Init p.hr_model_patcher.
INFO:model_patcher_hook.py:Init p.hr_model_patcher.

To create a public link, set `share=True` in `launch()`.
IIB Database file has been successfully backed up to the backup folder.
Startup time: 17.0s (prepare environment: 2.4s, import torch: 3.6s, import gradio: 1.6s, setup paths: 1.9s, initialize shared: 0.2s, other imports: 0.5s, load scripts: 3.3s, create ui: 1.0s, gradio launch: 2.2s, app_started_callback: 0.2s).
Model loaded in 3.5s (load weights from disk: 1.1s, create model: 0.5s, apply weights to model: 0.9s, load textual inversion embeddings: 0.4s, calculate empty prompt: 0.5s).
2024-07-01 03:33:02,393 - model_patcher_hook.py - INFO - Init p.model_patcher.
INFO:model_patcher_hook.py:Init p.model_patcher.
2024-07-01 03:33:02,413 - model_patcher_hook.py - INFO - Init p.hr_model_patcher.
INFO:model_patcher_hook.py:Init p.hr_model_patcher.
2024-07-01 03:33:07,867 - model_patcher_hook.py - INFO - Patch model_patcher of StableDiffusionProcessingTxt2Img.
INFO:model_patcher_hook.py:Patch model_patcher of StableDiffusionProcessingTxt2Img.
  0%|                                                    | 0/20 [00:00<?, ?it/s]
  5%|██▏                                         | 1/20 [00:00<00:11,  1.72it/s]
 15%|██████▌                                     | 3/20 [00:00<00:03,  4.55it/s]
 25%|███████████                                 | 5/20 [00:00<00:02,  6.70it/s]
 35%|███████████████▍                            | 7/20 [00:01<00:01,  8.55it/s]
 45%|███████████████████▊                        | 9/20 [00:01<00:01,  9.98it/s]
 55%|███████████████████████▋                   | 11/20 [00:01<00:00, 11.02it/s]
 65%|███████████████████████████▉               | 13/20 [00:01<00:00, 11.78it/s]
 75%|████████████████████████████████▎          | 15/20 [00:01<00:00, 12.33it/s]
 85%|████████████████████████████████████▌      | 17/20 [00:01<00:00, 12.72it/s]
100%|███████████████████████████████████████████| 20/20 [00:02<00:00,  9.86it/s]
2024-07-01 03:33:09,897 - model_patcher_hook.py - INFO - Unpatch model_patcher of StableDiffusionProcessingTxt2Img.
INFO:model_patcher_hook.py:Unpatch model_patcher of StableDiffusionProcessingTxt2Img.
2024-07-01 03:33:10,142 - model_patcher_hook.py - INFO - Close p.hr_model_patcher.
INFO:model_patcher_hook.py:Close p.hr_model_patcher.
2024-07-01 03:33:10,142 - model_patcher_hook.py - INFO - Close p.model_patcher.
INFO:model_patcher_hook.py:Close p.model_patcher.

Total progress: 100%|███████████████████████████| 20/20 [00:01<00:00, 11.80it/s]
ivanoff13 commented 4 months ago

Снимок It stopped working today Automatic 1111

ToniVacaESAT commented 4 months ago

I have the same problem. I have reinstalled the latest version and it still ignores it. With the b8268e1, it works.

Haoming02 commented 4 months ago

The latest version now uses process_before_every_sampling, which is currently only in Forge, and will be added to Automatic1111 v1.10.

So for now, you can revert to the older version of IC-Light extension, or use the 1.10-RC branch of Automatic11111

greasebig commented 4 months ago

hi, any solution for my problem?

greasebig commented 4 months ago

it is so obvious that i can reproduce it with 50% Success rate