NEQUIP

References: Batzner2022

This section specifies the input parameters for NEQUIP potential type based on equivariant neural networks with message passing. Requires linking with libtorch library from https://pytorch.org/cppdocs/installing.html. [Edit on GitHub]

Keywords

ATOMS
PARM_FILE_NAME
UNIT_CELL
UNIT_COORDS
UNIT_ENERGY
UNIT_FORCES

Keyword descriptions

ATOMS: string

Usage: ATOMS {KIND 1} {KIND 2} .. {KIND N}

Defines the atomic kinds involved in the NEQUIP potential. Provide a list of each element, making sure that the mapping from the ATOMS list to NequIP atom types is correct. This mapping should also be consistent for the atomic coordinates as specified in the sections COORDS or TOPOLOGY. [Edit on GitHub]

PARM_FILE_NAME: string = model.pth 

Aliases: PARMFILE

Usage: PARM_FILE_NAME {FILENAME}

Specifies the filename that contains the NEQUIP model. [Edit on GitHub]

UNIT_CELL: string = angstrom 

Usage: UNIT angstrom

Units of the cell vectors in the NEQUIP model.pth file. [Edit on GitHub]

UNIT_COORDS: string = angstrom 

Usage: UNIT angstrom

Units of coordinates in the NEQUIP model.pth file. [Edit on GitHub]

UNIT_ENERGY: string = hartree 

Usage: UNIT hartree

Units of energy in the NEQUIP model.pth file. [Edit on GitHub]

UNIT_FORCES: string = hartree/bohr 

Usage: UNIT hartree/bohr

Units of the forces in the NEQUIP model.pth file. [Edit on GitHub]