何かをした後の表情のloraです。ZimageBase ver.固有のトリガーワードのみで、主に上半身の構図でトレーニングしました。あまり良い出来とは思えないので、改善策を知ってる方、教えてください。デフォルトのトリガーワードのみだと脱ぎやすいので、着せてあげてください。補助ワード: both hands up, peace sighn, tongue out,This is Lora's expression after doing something.I trained the model using only specific trigger words, focusing primarily on upper-body compositions.I don't think it went very well,so if anyone knows how to improve it, please let me know.The default trigger word makes it easy to take off the clothes, so please dress them properly.Supporting words: both hands up, peace sign, tongue outSD1.5ver.The model used to generate the sample is<a target="_blank" rel="ugc" href="https://huggingface.co/syaimu/7th_Layer">7th_anime_v3_C</a>, <a target="_blank" rel="ugc" href="https://civarchive.com/models/4384?modelVersionId=109123">dreamshaper7</a>.何かをした後の表情のloraです。Unetのみの学習で、トリガーワードは設定していません。下記プロンプトで再現性が上がります。rolling eyes, tongue out, upper body, double peace sighn, hand up,This is Lora's expression after doing something.trained using Unet only, and no trigger words were set.The following prompts will improve reproducibility:rolling eyes, tongue out, upper body,double peace sign, hand up,

Ahegao with Peace Sign

training: online tool,prodigy cosine, network dim8 alpha4,image size 512

Files with SHA256: e221e045065ea1c3985baf8413bf5c4f6217170bb840c02be33694089fae4266