PLIP (Physical LassoLars Interactions Potential) is a hybrid python/C++ package for machine learning interatomic potentials for materials modelling. Physical Lasso-Lars Interaction Potential (PLIP) has been developed to address the lack of interpretability of MLIPs. PLIP uses a physically motivated mathematical formulation for the potential and a constrained linear regression. The MLIP is constructed as a linear combination of classical interactions, including two-body, three-body, and many-body interactions. These interactions are made up of simple functions such as Lennard-Jones, Gaussian, and Slatter-type orbitals, as well as explicit angular functions and embedded atom-like interactions.
The full documentation is available at plip.readthedocs.io.
Colab Python notebook tutorial: https://colab.research.google.com/github/LAM-GROUP/PLIP/blob/main/PLIPwithLAMMPS.ipynb
git clone https://github.com/LAM-GROUP/PLIP.git
cd PLIP
pip install -e .
- Python 3.8+
- Support for C++17
- Compiler
gcc6.0 and above - Python dependencies: see
pyproject.toml - LAMMPS
The PLIP package was developed by Julien Lam and Akshay Krishna AK
The PLIP package is distributed under terms of GNU GPLv3.