Honesty first as some may know this was the first lora I made and as you can tell by the skipping version numbers, it's always hated me. Though before beginning this one is a LoCon unlike my previous lora (seems to have helped a bit). Why it doesn't show up as such I don't know.
This is a model that's based on the monster girl feature of snake hair. Tends to work best at 0.6. This model was trained on NAI anime-full-pruned and generated in Anylora. Made with the Kohya tool for AI Training through google colab. Any suggestions to improve or recommended things for the data set would be appreciated.
Description
This model was trained on a single term trigger word due to suggestion. It's not a major increase from the previous version but makes less random snakes in the background.
This was a test to learn how to use the Kohya tools for AI Training. The model was trained with a mix of snake hair pictures as well as actual snakes. This model should not make characters with a mix of snake and real hair in favor of just snake hair. Any suggestions would be appreciated. Lora works best around 0.7. This model was trained on WarriorMama777's orange mix from huggingface.
FAQ
Comments (5)
Might want to use a vae with whatever base model you used with this lora. It could add some more vibrant colors if you want that
How did you make Lora apply only to the hairstyle?
This was made in the kohya tools, I haven't tried replicating it in google colab like my newer lora but if using kohya training tools there is a class token which tells the ai what its supposed to learn like a character or something I believe for its class I just put hair.
What art style did u got with? i found this lora on pixai and it uses a art style ive been looking for Would love to know how u got that
This is one of my old ones (due for an update at some point), no style models were applied it was just made on warriormamas orange mix WarriorMama777/OrangeMixs · Hugging Face to be specific i beleive it was aom2
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.