LORA trained on 588 images of artist 森倉円 (Morikura En). Doesn't require trigger words, recommended weight is 1. Training and dataset parameters: Images are scrapped from gelbooru, tagged by WD tagger; Trained using kohya-ss LORA training GUI, rank 32, resolution 768x768,other settings were default; Trained on anything-v4.5-pruned-fp16 model; 15 epochs, 1 repeat per image.
|