mathLab/PINA

Physics-Informed Neural networks for Advanced modeling

57
/ 100
Established

This tool helps scientists and engineers build predictive models that adhere to known physical laws or integrate with existing data. You input your scientific problem, including any governing equations or experimental data, and it outputs a trained neural network that can simulate complex systems or make predictions while respecting physics. It's designed for researchers, computational scientists, and anyone working with scientific data and simulations.

719 stars.

Use this if you need to develop machine learning models for scientific or engineering problems where physical principles (like differential equations) are important, or you need to combine real-world data with scientific laws.

Not ideal if your problem is purely data-driven without any underlying physical laws to enforce, or if you only need standard machine learning models without scientific constraints.

computational-physics engineering-simulation mathematical-modeling scientific-machine-learning numerical-analysis
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 21 / 25

How are scores calculated?

Stars

719

Forks

95

Language

Python

License

MIT

Last pushed

Mar 05, 2026

Commits (30d)

0

Get this data via API

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

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