
My first attempt at a Lora, made with the Civitai trainer and following this guide: https://civitai.com/articles/6503/how-to-easily-create-ponypdxl-lora-better-than-80percent-on-site-just-for-500buzz
The result is decent, but in full body the outfit barely comes out correctly, although the upper bodies come out more or less well. Maybe I need more full body images for the dataset. I'll see if I can improve it for a version 2.
Trigger words:
-
Face: 1girl, Nobume, blue hair, red eyes, long hair
-
Upper body: 1girl, Nobume, blue hair, red eyes, long hair, uniform
-
Full body: 1girl, Nobume, blue hair, red eyes, long hair, uniform, white boots
描述:
训练词语: Nobume, blue hair, red eyes, long hair,1girl, Nobume, blue hair, red eyes, long hair, uniform,1girl, Nobume, blue hair, red eyes, long hair, uniform, white boots
名称: 736041_training_data.zip
大小 (KB): 83881
类型: Training Data
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
名称: Nobume.safetensors
大小 (KB): 56084
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success