
This is why you don't teach someone like me machine learning, this is your fault
"Cash in cash out cash in cash out"
Ko-fi: https://ko-fi.com/keyennorman
The models I make are tested for resolutions under 512 (Not by choice), results may be different (Probably better) with higher resolutions.
The game was free on epic for a while, got it and now I can't stop playing. So because I'm angry I'll be posting two more characters from the game. This one is possible the weakest when it comes to dataset. Still can produce some good looking images though, check out the prerequisites
1. Activation tag is Adora, other tags to help include: blonde hair, purple eyes, tail, ponytail, high ponytail, shoulder armor, body fur, two-tone fur, cape, white armor, yellow fur, top knot, facial fur/facial mark, clothed, yellow skin, multicolored skin, adult, yellow hair,
2. Weight fluctuate quite a bit with this one, I'd start with 0.6 and try and tune it with whatever you may be looking for in the image you wish to generate (In which I have no clue what that may be with this LORA, get creative)
3. Model may generate the character in smaller proportions to make sure to include some tags to counter that if you want, check the ref images
On to my favorite hero monkey that just so happens to be the best out of the three I have worked on, Sauda
描述:
训练词语: Adora,blonde hair, purple eyes, tail, ponytail, high ponytail, shoulder armor, body fur, two-tone fur, cape, white armor, yellow fur, top knot, facial fur, clothed, yellow skin, multicolored skin, adult, yellow hair,
名称: AdoraLORA.safetensors
大小 (KB): 55414
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success