1580158
/
Hypernetwork-MonkeyPatch-Extension
mirror of https://github.com/aria1th/Hypernetwork-MonkeyPatch-Extension.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

introduce D-Adaptation

pull/51/head
AngelBottomless 2023-02-15 12:38:13 +09:00 committed by GitHub
parent e5d9668531
commit c9c107f2cf
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 8 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
README.md
View File

@ -62,11 +62,16 @@ Now you can select CLIP model, its difference is significant but whether its bet
The concept of ResNet, returning x + f(x) instead in layers, are available with option. Original webui does not support this, so you cannot load it without extension.
Unlike expanding type (1 -> 2 -> 1), shrinking type(1 -> 0.1 -> 1) network will lost information at initial phase. In this case, we need to additionally train transformation that compresses and decompresses it. This is currently only in code, its not offered in UI at default.
## D-Adaptation
Currently D-Adaptation is available for hypernetwork training. You can use this with enabling advanced AdamW parameter option and checking the checkbox.
Recommended LR is 1.0, only change it if its required. Other features are not tested with this feature.
The code references to this:
https://github.com/facebookresearch/dadaptation
### Planned features
Allow using general models with .safetensor save /loading
Training option loading and tuning for textual inversion
D-Adaptation for textual inversion
### Some personal researches