Please contact me here on civitAi or on Twitter if you want to provide the LoRa in a generation service.
TLDR (Because I don't care)
Use mirabel madrigal <lora:mirabelMadrigal_v1:1> in positiv promt.
If you like the LoRa, please feel free to share your generated images here and leave a review (or at least a rating) ;D
About
Well, I watched Encanto yesterday and Mirabel, the main character, was the only one I didn't get annoyed with, even though she's not the "all pretty" type girl... so I made a LoRa for her :D If you like musicals, you'll really like this film, I myself mostly skipped the music parts (without missing any data I could use for training, of course).
Training
I managed to get 75 shots to work with... I was a bit disappointed that she only wore 2 different outfits in the whole film (mostly her, normal clothes, and in one scene a nightgown, (okay, in theory 3 outfits, but I did not use her childhood scenes for training)).
Usage
No weight adjustments required:
mirabel madrigal <lora:mirabelMadrigal_v1:1>
You are able to put on and of her glasses and her normal outfit was tagged with white blose, long green skirt so if the model does not behave when putting her in a different outfit simply put these tags into the negative prompt.
描述:
训练词语: mirabel madrigal,white blouse,green skirt,glasses
名称: MirabelMadrigal_v1.safetensors
大小 (KB): 147570
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
