THIS LORA IS ONLY FOR TEST, PLEASE DO NOT PUBLISH NSFW IMAGES TO THE COMMUNITY!!!
此LORA仅用于测试,请不要向社区发布NSFW图片!
训练的步数不多,分了4个批次,4 Epoch版本觉得眼睛会容易出问题,所以在用SDXL base模型的时候可以尝试降低权重,或者可以直接用3 Epoch的版本。
The training steps are not many and were divided into 4 batches. The 4 Epoch version feels that eye problems can occur easily, so when using the SDXL base model, you can try reducing the weight, or you can directly use the 3 Epoch version.
weight=0.6-1.3
本想炼点别的,也算是心血来潮,搜图片看到了汤唯,只用最基本的参数。
I originally wanted to refine something else, but it was also a whim. I searched the pictures and saw Tang Wei, using only the most basic parameters.
Disclaimer: This artificial intelligence model ("the Model") is developed solely for artistic expression and technological demonstration, not for creating or distributing illicit or inappropriate content. The creator expressly disapproves of such use. By downloading or using the Model, you ("the User") agree not to misuse it in any manner contrary to these intentions and accept all responsibility for compliance with these terms.