gblearn.lammps

Functions for interacting with LAMMPS dump files.

Functions

make_lattice(box) Constructs a lattice array compatible with ASE using the box dimensions specified in a lammps format.

Classes

Dump(filepath[, stepfilter]) Represents a dump file that could potentially have more than one timestep.
Timestep(filepath[, index, openf, stepfilter]) Represents a single time step in a LAMMPS dump file.

API Documentation

Functions for interacting with LAMMPS dump files.

class gblearn.lammps.Dump(filepath, stepfilter=None)[source]

Represents a dump file that could potentially have more than one timestep.

Parameters:
  • filepath (str) – full path to the LAMMPS dump file to extract the time step from.
  • stepfilter (list) – of int timestep values that should be parsed. If the next timestep encountered is not in this list, it will be ignored.
steps

dict – keys are int timestep ids; values are Timestep.

dump(filename, mode=’w’, rebox=False)[source]

Dumps the specified structure to file in the LAMMPS format.

Parameters:
  • filename (str) – target path for saving the timestep.
  • mode (str) – file read/write mode; defaults to new file.
  • rebox (bool) – when True, the box is recalculated from the current set of atomic positions.
class gblearn.lammps.Timestep(filepath, index=0, openf=None, stepfilter=None)[source]

Represents a single time step in a LAMMPS dump file.

Parameters:
  • filepath (str) – full path to the LAMMPS dump file to extract the time step from.
  • index (int) – index of the time step in the dump file.
  • openf (file) – open file object for sequential instantiation of multiple time steps.
  • stepfilter (list) – of int timestep values that should be parsed. If the next timestep encountered is not in this list, it will be ignored.
types

numpy.ndarray – integer types of the atoms in the list.

ids

numpy.ndarray – integer atom ids; has same length as len(self).

xyz

numpy.ndarray – with shape (len(self), 3); float positions of the atoms in the time step.

periodic

list – of bool specifying whether the x, y, or z directions are periodic (for the box).

box

numpy.ndarray – with shape (3, 2) specifying the lo and hi bounds of the box in each direction. The box can also have shape (3, 3) to support non-orthogonal box vectors.

extras

list – of str indicating extra atomic parameters that are available in this time step.

dump(filename, mode=’a’, rebox=False)[source]

Dumps the specified structure to file in the LAMMPS format.

Parameters:
  • filename (str) – target path for saving the timestep.
  • mode (str) – file read/write mode; defaults to append.
  • rebox (bool) – when True, the box is recalculated from the current set of atomic positions.
gb(Z=None, method=’median’, pattr=’c_csd’, extras=True, soapargs={}, **kwargs)[source]

Returns the grain boundary for this time step.

Parameters:
  • Z (int or list) – element code(s) for the atomic species.
  • method (str) – one of [‘median’].
  • pattr (str) – name of an attribute in extras to pass as the selection parameter of the routine.
  • extras (bool) – when True, include extra attributes in the new GB structure.
  • soapargs (dict) – initialization parameters for the gblearn.soap.SOAPCalculator instance for the GB.
  • kwargs (dict) – additional arguments passed to the atom selection function. For median, see gblearn.selection.median() for the arguments. For cna* see gblearn.selection.cna_max().
Returns:

instance with only those atoms that appear

to be at the boundary.
Return type:

gblearn.gb.GrainBoundary
gbids(method=’median’, pattr=None, **kwargs)[source]

Returns the indices of the atoms that lie at the grain boundary.

Parameters:
  • method (str) – one of [‘median’, ‘cna’, ‘cna_z’].
  • pattr (str) – name of an attribute in extras to pass as the selection parameter of the routine.
  • kwargs (dict) – additional arguments passed to the atom selection function. For median, see gblearn.selection.median() for the arguments.
Returns:

of integer indices of atoms in this timestep that are

considered to lie on the boundary.
Return type:

numpy.ndarray

Examples

Retrieve the positions of the atoms that lie at the boundary using the median centro-symmetry parameter values.

>>> from gblearn.lammps import Timestep
>>> t0 = Timestep("lammps.dump")
>>> ids = t0.gbids()
>>> xyz = t0.xyz[ids,:]
gblearn.lammps.make_lattice(box)[source]

Constructs a lattice array compatible with ASE using the box dimensions specified in a lammps format. This was contributed by Jonathan Priedeman.

Parameters:box (numpy.ndarray) – box dimensions in cartesian directions in format lo hi. Shape (3, 2). Also supports tricilinic boxes when shape (3, 3) is specified.