nedtaylor/athena

A Fortran-based neural network library for physics-based applications. Alongside standard neural network layer types, it also supports graph-based layers and physics informed neural networks.

38
/ 100
Emerging

This tool helps physicists and materials scientists analyze complex physical systems by building and training neural networks. You provide data like atomic structures or charge densities, and it helps you create models that can understand and predict physical properties or behaviors. It's designed for researchers working with physics-based simulations and data.

Use this if you are a physicist or materials scientist needing to apply neural networks, especially graph-based or physics-informed ones, to your complex data like atomic configurations or charge densities.

Not ideal if you are looking for a general-purpose machine learning tool for domains outside of physics-based applications, or if you don't use Fortran.

physics-modeling materials-science computational-physics data-analysis predictive-modeling
No Package No Dependents
Maintenance 6 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

31

Forks

3

Language

Fortran

License

MIT

Last pushed

Dec 10, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/nedtaylor/athena"

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