yang-song/score_flow

Official code for "Maximum Likelihood Training of Score-Based Diffusion Models", NeurIPS 2021 (spotlight)

36
/ 100
Emerging

This project offers a method for training advanced image generation models that create highly realistic images from random noise. It takes training image datasets like CIFAR-10 or ImageNet 32x32 and produces a trained model capable of generating new, high-quality images, along with metrics to evaluate their realism. Machine learning researchers and generative AI practitioners would use this for developing state-of-the-art image synthesis techniques.

153 stars. No commits in the last 6 months.

Use this if you are a machine learning researcher focused on advancing image generation technologies and need to train highly efficient and accurate score-based diffusion models.

Not ideal if you are an application developer looking for a pre-built image generation API or if your primary goal is to deploy existing models for content creation rather than research into model training.

generative-AI image-synthesis machine-learning-research deep-learning-models
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 8 / 25
Community 18 / 25

How are scores calculated?

Stars

153

Forks

25

Language

Python

License

Last pushed

Dec 09, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/yang-song/score_flow"

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