Polynomial Machine Learning Potential Repository

General purpose potentials

• Elemental systems
• Binary alloy systems
• Other systems
• Older distributions

On-the-fly trained potentials

• On-the-fly Machine Learning Potentials

DFT datasets

• Elemental systems
• Binary alloy systems
• Other systems

Related codes to use polynomial MLPs

pypolymlp

A generator of polynomial machine learning potentials.

Property calculations are possible using distributed polynomial MLPs. List of properties that can be calculated using pypolymlp:

• Energy, forces on atoms, and stress tensor

• Force constants

• Elastic constants

• Equation of states

• Structural features (Polynomial invariants)

• Local geometry optimization

• Phonon properties, Quasi-harmonic approximation

• Self-consistent phonon calculations

lammps-polymlp-package

A package to use polynomial MLPs in LAMMPS software

Theoretical background

• Structural features
• Potential energy models
• Parameters for developing MLPs

License and Citation

© Atsuto Seko (Licensed under CC BY 4.0)

If you use polynomial machine learning potentials and datasets in the repository for academic purposes, please cite the following article [1].

[1] A. Seko, “Systematic development of polynomial machine learning potentials for elemental and alloy systems”, J. Appl. Phys. 133, 011101 (2023).