TiankaiHang/Min-SNR-Diffusion-Training

[ICCV 2023] Efficient Diffusion Training via Min-SNR Weighting Strategy

25
/ 100
Experimental

This project helps machine learning researchers and practitioners train diffusion models for image generation more efficiently. It takes image datasets like ImageNet and applies a novel weighting strategy during the training process, resulting in significantly faster convergence. The outcome is a high-quality, pre-trained image generation model.

268 stars. No commits in the last 6 months.

Use this if you are developing or training diffusion models and need to reduce the time and computational resources required to achieve state-of-the-art image generation results.

Not ideal if you are looking for an out-of-the-box image generation tool without needing to engage with model training or fine-tuning.

deep-learning-research generative-ai image-synthesis model-optimization computer-vision
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 8 / 25
Community 7 / 25

How are scores calculated?

Stars

268

Forks

7

Language

Python

License

Last pushed

Dec 10, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/TiankaiHang/Min-SNR-Diffusion-Training"

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