ermongroup/ncsnv2

The official PyTorch implementation for NCSNv2 (NeurIPS 2020)

48
/ 100
Emerging

Score-based generative modeling that learns vector fields pointing toward high-density regions of data distributions, enabling high-resolution image synthesis without adversarial training. The architecture trains neural networks to estimate score functions across multiple noise scales, then generates samples through Langevin dynamics. Includes support for image interpolation, inpainting, and FID evaluation across PyTorch-based training pipelines configured via YAML files.

321 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

321

Forks

62

Language

Python

License

MIT

Last pushed

Jun 12, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/ermongroup/ncsnv2"

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