
这个lora能够在保持整体结构和角色的情况下,提升画面的色彩表现,提高饱和度,
This Lora is able to enhance the color performance of the screen and increase saturation while maintaining the overall style and characters.
经测试跟各基础模型,角色Lora,画风Lora都能良好配合,
After testing with base model, character Lora , style Lora,It can work well together.
基于藤ちょこ(藤原)以及多位擅长高饱和度画风的画师作品训练,训练方法来自青龙的完美炼丹术,
Based on the works of 藤ちょこ and several artists skilled in high-saturation styles, the training method is based on the perfect alchemy technique of the 青龙 ,
经Lora分层,去除了大部分的污染只保留纯粹的上色风格,
After Lora block weight , most of the pollution has been removed and only pure coloring remains.
推荐/recommend 0~1.5
Lora block weight: 1,0,0,0,0,0,0,0,0,0,0.4,0.4,0.7,0.8,1,0.5,0.6
未经分层的完整LORA在colorful_V1_full.zip,可以尝试自己分层获取更好的效果,
The complete version of LORA blocks can be found in the colorful_V1_full.zip file, and you can improve the results by applying your own weights.
描述:
Lora block weight :1,0,0,0,0,0,0,0,0,0,0.4,0.4,0.7,0.8,1,0.5,0.6
训练词语:
名称: colorful_V1_full.zip
大小 (KB): 153030
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: colorful_V1_lbw.safetensors
大小 (KB): 168301
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success