MOLPIPx

MOLPIPx seamlessly integrates Permutationally Invariant Polynomials (PIPs) with modern machine learning frameworks. Built on JAX and EnzymeAD-Rust, it enables efficient, differentiable modeling of potential energy surfaces using linear models, neural networks, and Gaussian processes.

Capabilities at a Glance

MOLPIPx currently includes the following implementations:

Model Type	Energy	Forces
Linear Models	✓	✓
Neural Networks (Flax)	✓	✓
Gaussian Processes (GP)	✓	✓
Anisotropic Models	✓	✓

Check the Tutorials for more information.

Documentation for the source code can be found here. The full source code with examples and tests can be explored at github.