phydra-labs/phydrax

Modular Physics-ML Components in JAX

26
/ 100
Experimental

This tool helps scientists and engineers working with Partial Differential Equations (PDEs) to build and train physics-informed machine learning models. You define a physical domain, input data (like boundary conditions or observed measurements), and then define the governing physics equations. The tool outputs a trained model that satisfies both your data and the physical laws you've specified, allowing you to simulate or predict complex physical phenomena.

Use this if you need to solve complex physical problems by blending data-driven machine learning with fundamental physics equations, especially for systems modeled by Partial Differential Equations.

Not ideal if your primary goal is to perform general-purpose data analysis, standard supervised machine learning, or if you require a tool for commercial production use without a separate license.

physics-simulation computational-fluid-dynamics materials-science scientific-modeling engineering-analysis
No Package No Dependents
Maintenance 10 / 25
Adoption 5 / 25
Maturity 11 / 25
Community 0 / 25

How are scores calculated?

Stars

11

Forks

Language

Python

License

Last pushed

Feb 18, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/phydra-labs/phydrax"

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