Train on a mix of random images with fine vibe for me. Some sharper outline, some blurred background to imitate focus on person etc. It works nice with basic Pony XLV6 while strength of lora set to 1. No trigger word needed.
It messes a little things like hands and feet, but there are some tweakable LoRAs to fix this issue, and they mostly work fine with it.
描述:
训练词语:
名称: 911341_training_data.zip
大小 (KB): 32413
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: Random_Style.safetensors
大小 (KB): 56086
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
原文链接：https://1111down.com/1114732.html，转载请注明出处
