

这是做钢笔画风格的lora,训练图用了83张训练图,
底模用的是CounterfeitV30_v30.safetensors,尺寸是512x768,描述词简单描述就好了,
例如树, 凳, 单色, 草, 户外, 坐, 独奏, 帽子, 1girl, 传统媒体, 风景...
因为训练图大多都是用的是户外场景图,所以出图出户外场景效果比较好,
我试了其他偏2次元风格的大模型也合适,权重调看着调就好;
The following is machine translated
This is to do the pen drawing style lora, the training map used 83 training maps , the
The base model used is CounterfeitV30_v30.safetensors, the size is 512x768, the descriptors are simple to describe.
For example tree, stool, monochrome, grass, outdoor, sitting, solo, hat, 1girl, traditional media, landscape...
Since most of the training maps are of outdoor scenes, it's better to map out outdoor scenes.
I tried other big models that favored the 2-dimensional style that also fit, and the weights tuned to look at the tuning;
1. 재게시된 모델의 권리는 원 제작자에게 있습니다.
2. 모델 원작자가 모델을 인증받으려면 공식 채널을 통해 SeaArt.AI 직원에게 문의하세요. 저희는 모든 창작자의 권리를 보호하기 위해 노력합니다. 인증하러 이동
