mir-group/allegro-pol

NequIP extension package that adapts the Allegro equivariant GNN architecture to predict the electric response of materials

39
/ 100
Emerging

This tool helps materials scientists and computational chemists predict the electrical behavior of materials. It takes structural information about a material (like atom positions) and outputs predictions for polarization, Born charges, and polarizability, along with energy and forces, using a single machine learning model. Researchers working on designing new materials with specific electrical properties would find this useful.

Use this if you need to simulate and predict the electric response (polarization, Born charges, polarizability) of materials alongside their energy and forces.

Not ideal if you are looking for a general-purpose machine learning library without a specific focus on materials science or electric response predictions.

materials-science computational-chemistry solid-state-physics ab-initio-simulations material-design
No Package No Dependents
Maintenance 10 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 6 / 25

How are scores calculated?

Stars

34

Forks

2

Language

Python

License

MIT

Last pushed

Mar 11, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mir-group/allegro-pol"

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