MLFF

A collection of files related to machine learning force fields (MLFF) within our research group. Training data, scripts, and forcefields are included where appropriate.

Utilities

• ML_AB file manipulation: https://github.com/utf/pymlff
• Diffusion coefficient calculations: https://github.com/bjmorgan/kinisi
• Perovskite structure analysis: https://github.com/WMD-group/PDynA

Li₃N

Details: Gaussian Approximation Potential (GAP)-style MLFF trained using on-the-fly sampling of reference configurations in VASP.

Publication: Nature of the Superionic Phase Transition of Lithium Nitride from Machine Learning Force Fields (Chemistry of Materials, 2023)

Covalent Organic Frameworks

Details: Gaussian Approximation Potential (GAP)-style MLFF trained using on-the-fly sampling of reference configurations in VASP.

Publication: Room-temperature stacking disorder in layered covalent-organic frameworks from machine-learning force fields (Materials Horizons, 2023)