Comments (12)
from sd-webui-faceswaplab.
Unfortunately your mitigation that '
YOU NEED TO recreate ui-config.json (delete) or at least remove any faceswaplab reference to be able to use default settings again' doesn't work.
Having tried both deleting the ui-config.json file entirely so it is recreated AND editing the ui-config.json file to remove all lines referencing faceswaplab it will not remember any of the settings put into the txt2img tab under the FaceSwapLab V1.2.3 pull down in between sessions (i.e. if I go to the Setting tab, then the Defaults option, then click on View Changes, then Apply). In fact, it doesn't even pick up that there are any changes when View Changes is pressed. For example any changes to add a default Face Checkpoint (precedence over reference face) or Genger are not recognised and therefore forgotten.
I even deleted the entire FaceSwapLab extension, and the associated folders for FaceSwapLab in the Models and Extensions folders, and deleted the ui-config.json and once FaceSwapLab was reinstalled fresh still any changes applied in the txt2img tab are not remembered or even recognised when I have tried to set Defaults via the Settings tab.
This is a great extension. Please could you look at this again?
from sd-webui-faceswaplab.
Ok, reverting back would be easy but i need to be sure there is a problem first because the new way of handling parameters reduces load time drastically.
I have tested on linux on sdnext and a1111. I will try to test on windows. Edit : I did test on windows and it's working.
What is your configuration ?
The current version only store parameters that are set on SD setting tabs. Default checkpoint for instance is not possible. All otheer settings are lost. You also need to restart.
There are no settings stored in the ui-config.json as of now. Is that what you are looking for ?
from sd-webui-faceswaplab.
from sd-webui-faceswaplab.
Ok, just to be sure because it's puzzling me.
You are using these settings in sd settings :
And these settings are not reflected in txt2img parameters (after restart)?
Or you are expecting the configuration done in txt2img to be persistent (only when modifying it in txt2img tabs) ?
Where you using the settings before 1.2.3 ?
from sd-webui-faceswaplab.
from sd-webui-faceswaplab.
Ok thanks ! I got it :
If, for example, I change the Face Checkpoint to Face_3.safetensors it is impossible to make it persistent like all other changes in txt2img (e.g. Restore Faces or Seed)
I made settings do_not_save to be able to adjust settings by myself.
I expect the configurations done in txt2img to be persistent as they always are if you use the Settings/Default. This facility has been available for months in A1111 and worked perfectly in V1.2.2, but is now broken in V1.2.3. Have you never used or tested the Settings/Default facility in A1111?
I don't use this option. That's why. Settings cannot be managed by faceswaplab OR sd. I had to made a choice but but I applied this to all the changes.
I will try to think of a way to handle that properly. I hope it's not too critical in the meantime.
I can mark only the options I set as "do not save" and leave the others or i can add options in the settings. Or i can suppress the default settings in faceswaplab and let the users use the default interface (don't know if it is more user friendly)
Btw i don't see the screenshots, i think they are not uploaded when you reply by mail.
from sd-webui-faceswaplab.
I made a change to mark only fields that are not managed in faceswaplab settings. I think it is working. Would you have time to try and tell me if it is better for you ?
from sd-webui-faceswaplab.
from sd-webui-faceswaplab.
Don't worry about that, since all the tests are ok and the change is not critical, i will merge the version. You can tell me if it's better by just reinstalling the main branch.
from sd-webui-faceswaplab.
Reopening (the merge has closed the issue). Will close if you confirm it works properly.
I have opened a poll in discussion section to ask if faceswaplab should manage default settings : #69
I am starting to think it's a redundant functionality and error-prone. Thanks for making me discover the feature.
from sd-webui-faceswaplab.
from sd-webui-faceswaplab.
Related Issues (20)
- "Use improved segmented mask" not working
- Comfyui version
- Can't see faceswap in the Automatic1111 UI HOT 2
- Broken model download link
- Future warning 'rcond' parametr
- Pytorch and Torchvision incompatibility in Ubuntu
- It is not working HOT 3
- Cannot load faceswap model in webui-forge HOT 3
- Failed to swap face in postprocess method : apply_overlay() takes 3 positional arguments but 4 were given HOT 19
- Change only the target face in an existing image
- FaceSwapLab - ERROR - Conversion failed max() arg is an empty sequence
- ERROR - Failed to swap face in postprocess method : apply_overlay() takes 3 positional arguments but 4 were given HOT 20
- Please delete sorry
- failed to install in automatic1111 HOT 1
- img2img -> Inpaint -> dont work HOT 11
- How do I build a checkpoint with multiple angles of a face? HOT 2
- img2img not saving batch images into output directory HOT 3
- Does not work on Automatic1111 HOT 6
- Extension does not respect stable-diffusion-webui `--data-dir` option
- Documentation for API?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sd-webui-faceswaplab.