vldgroup/graph-pes

train and use graph-based ML models of potential energy surfaces

49
/ 100
Emerging

This toolkit helps materials scientists and computational chemists accurately model how atoms and molecules interact. You can input atomic structure data to train machine-learned models that predict energies and forces. These models then allow you to simulate material behavior for tasks like optimizing structures or running molecular dynamics simulations.

122 stars.

Use this if you are a researcher who needs to train and use machine-learned potential energy surface models for atomic structures, or if you are developing new methodologies for these types of models.

Not ideal if you are looking for a pre-built simulation tool without needing to train or fine-tune custom interaction models.

computational chemistry materials science molecular dynamics potential energy surfaces atomic simulations
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

122

Forks

12

Language

Python

License

MIT

Last pushed

Mar 09, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/vldgroup/graph-pes"

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