這算是一個嘗試

結合純文字leco方式和圖像訓練產生的lora

1. 訓練一個leco 的lora

2. 訓練一個圖像的lora

3. 透過lora分層只對衣服的相關分層.利用超級融合器的功能把上面兩個功能融合起來

用到的相關工具

It's kind of an experiment.

Combining a purely textual leco approach with graphic training to produce a lora

1. Training a leco lora

2. Training a lora with an image

3. Layering by lora is only relevant for clothing. Use the Super Fusion function to combine the above two functions.

Related tools used

https://github.com/p1atdev/LECO

https://github.com/Akegarasu/lora-scripts

https://github.com/hako-mikan/sd-webui-lora-block-weight

https://github.com/hako-mikan/sd-webui-supermerger