User LoRA
awawa
This is the Pony-edition of my Fantasy LORA, same dataset, different tags.The experiment with this one, is to get "better" results, due to exposing the dataset to noise.I put the images into Photoshop and add 4% to 6% gaussian noise