Skip to content
Marcus Wieder edited this page Nov 2, 2023 · 11 revisions

Welcome to the modelforge wiki!

This is the project overview page for the Choderalab's efforts towards differentiable modeling.

General roadmap

The objective of this project is to create an interoperable ecosystem compatible with various applications, primarily focused on differentiable physical modeling. Our approach emphasizes neural network potentials (NNPs) and Free Energy calculations. Our architecture is designed with clean APIs that provide developers with simplicity and minimal overhead requirements. All packages will be designed with interoperability in mind, ensuring that they can be used in conjunction with one another, as well as with other systems, to maximize versatility and functionality.

Modelforge package

This package is aimed at implementing, training, distributing, storing, and using neural network potentials for molecular simulations, as well as providing essential datasets. It can be used together with our automatic NNP testing package auditorium and our NNP-first molecular dynamics engine chiron.

Clone this wiki locally