This is a style model that gives softer faces to characters, this was a mix of several artist that although not doing what I originally intended turned out very cute. Tends to work best at 0.6. This model was trained on NAI anime-full-pruned and generated in AOM3. Made with the Kohya tool for AI Training through google colab. Any suggestions to improve or recommended things for the data set would be appreciated.
Description
Now trained on 368 unique unflipped images from various artist this version brings a lot more outfits and customization without looking as repetitive. Bad images were removed from the dataset all images were hand edited to remove tremble effects and watermarks to prevent them from unnecessarily popping up. Tags of old photos have been updated.
credits:
- Meltryllis (Fate) Lora by ScottishMonstr
- Shizuka Marikawa - Highschool of the Dead by KuugaUltimate200
- Galko-chan | oshiete! galko-chan by mnemosynekotorin
FAQ
Comments (3)
Hi, could you please clear my curiosity? How is it that in some positive and negative prompts some words are enclosed in parentheses? These to be clear ( and sometimes even 2 (( or 3 ((( parentheses. Could you explain the reason? It's not enough to simply use the Comma? , XD
Oh just curious, parenthesis mean emphasis how much the ai should focus on one feature so if you do something like (hat) it means hat is more likely to show up and more parenthesis increases this so ((hat)) is stronger than (hat). Personally though rather than stacking parenthesis I like using the number system rather than increasing parenthesis so you can also do something like (hat:1.4) which is basically the same as multiple parenthesis but personally I think it distorts pictures a bit less that way.
You don't want to add too much emphasis though because then you can get a lot of noise and distortion as a result.
@rulles OK,thx a lot :)