akashsonowal/ddpo-pytorch

RLHF for Stable Diffusion

27
/ 100
Experimental

This project helps machine learning engineers refine text-to-image diffusion models, like Stable Diffusion, to generate images that are more aesthetically pleasing. You input an existing diffusion model and a dataset of images with aesthetic scores. The output is a fine-tuned diffusion model that produces higher quality, more visually appealing images, suitable for AI artists or content creators.

No commits in the last 6 months.

Use this if you want to improve the aesthetic quality of images generated by your diffusion models using reinforcement learning from human feedback (RLHF).

Not ideal if you need to fine-tune diffusion models based on complex, multi-faceted reward functions beyond simple aesthetic scoring.

AI art generation Diffusion model fine-tuning Image aesthetic improvement Generative AI Reinforcement learning for creatives
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 6 / 25

How are scores calculated?

Stars

14

Forks

1

Language

Python

License

Apache-2.0

Last pushed

Jul 09, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/akashsonowal/ddpo-pytorch"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.