This is a model that fixes the multiple glasses problem. Now instead of mixing "sunglasses" and "eyewear on head" (which normally gets 2 sunglasses) you can now type "headshades". Tends to work best at 0.6. This model was trained on NAI anime-full-pruned and generated in CitrineDreamMix. 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 was done on requests. This was trained on 50+ images of various shape and color sunglasses. This is the third attempt. Version 1 I trained "eyewear on head" and "sunglasses" as a double trigger [didn't work well]. Version 2 I fused it to get "sunglasses on head" as a single trigger but would still occasionally duplicate. This version, I just made it 1 word 1 trigger and it works much better.
FAQ
Comments (6)
thank you for making this, its incredibly awesome :) I was able to make a rendition of Julia from Cowboy Bebop
Glad it worked.
@rulles thank you very much
@rulles looking forward to your other work :)
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.