タイプ
TextualInversion
投稿日時
2023-04-25
基本モデル
SD 1.5
トリガーワード
d381m4z4r
コピー




follow me Paranoid American Instagram, Youtube, Patreon and the website.
Suggested Weight: 0.7-1.2
photo of young d381m4z4r with a-line bob haircut and makeup as a [something], detailed face, skin texture, sharp focus, light bokeh
most of the training was from recent images, so to get a younger version add "young" to positive prompts, and try adding "old" and "wrinkles" and etc to the negative prompt
Works very well for closeup portraits without inpainting
trained on ~24 images
Learning Rate: 0.005:1000, 0.001:2000, 0.0001:5000, 0.00005
Total Steps: 15,000
Batch Size: 1
Gradient Steps: 1
1.転載モデルは学習・共有目的のみで使用し、著作権は原作者に帰属します
2.モデルの認証は公式チャンネルでご連絡ください。クリエイターの権利保護に努めています クリックして認証
