Meet jp-Karolina: My Latest Attempt in Textural Inversion Trainings

I'm thrilled to present jp-Karolina, another milestone in my exploration of textural inversion. Build on the Stable Diffusion 1.5 model.

Training Data:

• Built from 28 unique photographs taken by me, ensuring originality without directly replicating the subjects.

• Progressive learning rate implemented: 0.05:10, 0.02:20, 0.01:60, 0.005:200, 0.002:500, 0.001:3000, 0.0005.

• Enhanced textural depth with 10 vectors per token.

• Batch size of 1 and 1 gradient accumulation step.

I'm interested to see the diverse and imaginative uses of jp-Karolina. Your creative interpretations and shared artworks help keep me motivated. I invite you to experiment, share your results, and provide your valuable feedback.