原本模型中自带原神老芭数据，直接调用就行，但是估计模型训练的图里都是戴帽子的，疯狂加头饰类的负面提示词都没用，很难出没头饰的图，能不能只用提示词来降低帽子出现概率？
只有一张是我满意的（图1）其他都是图3这样的
我提示词取掉帽子披头散发用了不少，但是收效甚微
我用的提示词正面：score_9, score_8_up, score_7_up,source_anime,1 girl,shiny hair,clear skin,smooth skin,flat color,ultra detailed eyes,barbara_\(genshin_impact\),tired expression,in hospital,hospital gown:1.5, (no hat):1.5, (no headpiece):1.5, (no headwear):1.5, long flowing hair:1.5,natural hairstyle:1.5,detailed hair strands,no accessories
负面：low quality , worst quality, bad anatomy,bad proportions, extra legs, deformed anatomy, messy color, deformed fingers, bad, distracted, hyperrealistic,source_furry, source_pony, ,source_cartoon, extra arms,blurry, extra finger,zPDXLpg-neg,portrait,censored,headwear:1.5,hat:1.5,Bowtie:1.5,Headpiece:1.5,cap:1.5,crown:1.5,tiara:1.5
有没有办法在提示词阶段就取掉帽子呢？
只有一张是我满意的（图1）其他都是图3这样的
我提示词取掉帽子披头散发用了不少，但是收效甚微
我用的提示词正面：score_9, score_8_up, score_7_up,source_anime,1 girl,shiny hair,clear skin,smooth skin,flat color,ultra detailed eyes,barbara_\(genshin_impact\),tired expression,in hospital,hospital gown:1.5, (no hat):1.5, (no headpiece):1.5, (no headwear):1.5, long flowing hair:1.5,natural hairstyle:1.5,detailed hair strands,no accessories
负面：low quality , worst quality, bad anatomy,bad proportions, extra legs, deformed anatomy, messy color, deformed fingers, bad, distracted, hyperrealistic,source_furry, source_pony, ,source_cartoon, extra arms,blurry, extra finger,zPDXLpg-neg,portrait,censored,headwear:1.5,hat:1.5,Bowtie:1.5,Headpiece:1.5,cap:1.5,crown:1.5,tiara:1.5
有没有办法在提示词阶段就取掉帽子呢？
另外，你的括號應該括上1.5。（當然，這情況下也不重要了）
