タイプ
LORA
投稿日時
2024-09-14
基本モデル
Flux.1 D
バージョン紹介
epoch 37; it was trained to 62 epochs but looks very over-trained and it loses coherency




Training data used for this particular LoRA were dated 1994 through 1999.
Dataset preparation using random segments of video were prepared using scripts from my github - https://github.com/akspa0/dataset-tools/blob/main/video/randomSampler/videosampler2x6.py
epoch 37; it was trained to 62 epochs but looks very over-trained and it loses coherency
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
