LoRA is generated entirely from anime images from Season 1.
-----------------------------------
Check samples for prompts :)
Notes:
I recommend against using the global activation tag 'Nagatoro'.
Includes all of her outfits from season 1
Most of her expressions included (except for uneven eyes, that one is hard to get right so I removed it from the dataset)
Frown, Grin, Smirk, Smug, Teeth, Fang, Half-closed eyes, Skin Fang, Teeth + Fang
------------
Crazy smile, crazy eyes <- doesn't seem to do much thoHer "cute" a bit more detailed style that appears from time to time:
NagatoroStyle, lens flare, evening, multicolored hair, black hair, purple hair (optional tag)Warning:
#1: purple hair tag seems to affect even her outfit so you might want to refrain from using it
#2: might not work at all... its somewhat randomSkin fang & fang are bleeding together with some checkpoints
Use ADetailer for better face/eyes
Recommended but optional negatives:
shaded face, wide-eyed, half-closed eyesIf you use hairclip, earclip she tends to generated more from the right side
Took 2 hours to train and much longer to prepare the dataset (derp face)
-----------------------------------
Leave a like, comment, review but first of all, ENJOY ;)