This is a model trained on alleyways. As usual when generating display images I noticed that alleyways could use an improvement. May need to increase it starting at 0.6 for desired results. This model was trained on NAI anime-full-pruned and generated in anything5. 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
Trained on 35 images of varying alleyway. Tends to put characters center or up close to one of the walls.
Credits:
hakaan - atomic hips
FAQ
Comments (2)
What’s the base model?
For reference it's in the model description, it was trained on nai-anime-full pruned. The generations were anything5 both are on huggingface.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.