gblearn.soap¶
Functions for generating the SOAP representation of a grain boundary.
Functions
S (a, b) |
Computes the SOAP similarity kernel between two SOAP vectors, \(d(a,b) = \sqrt{K(a,a)+K(b,b)-2*K(a,b)}\). |
Classes
SOAPCalculator ([rcut, nmax, lmax, sigma, …]) |
Represents a set of unique SOAP parameters for which SOAP vectors can be calculated. |
API Documentation¶
Functions for generating the SOAP representation of a grain boundary.
-
gblearn.soap.
S
(a, b)[source] Computes the SOAP similarity kernel between two SOAP vectors, \(d(a,b) = \sqrt{K(a,a)+K(b,b)-2*K(a,b)}\).
-
class
gblearn.soap.
SOAPCalculator
(rcut=5.0, nmax=12, lmax=12, sigma=0.5, trans_width=0.5)[source] Represents a set of unique SOAP parameters for which SOAP vectors can be calculated.
Parameters: - rcut (float) – local environment finite cutoff parameter.
- nmax (int) – bandwidth limits for the SOAP descriptor radial basis functions.
- lmax (int) – bandwidth limits for the SOAP descriptor spherical harmonics.
- sigma (float) – width parameter for the Gaussians on each atom.
- trans_width (float) – distance over which the coefficients in the radial functions are smoothly transitioned to zero.
-
rcut
¶ float – local environment finite cutoff parameter.
-
nmax
¶ int – bandwidth limits for the SOAP descriptor radial basis functions.
-
lmax
¶ int – bandwidth limits for the SOAP descriptor spherical harmonics.
-
sigma
¶ float – width parameter for the Gaussians on each atom.
-
trans_width
¶ float – distance over which the coefficients in the radial functions are smoothly transitioned to zero.
-
calc
(atoms, central, basis=None)[source] Calculates a SOAP vector for the specified species and atomic positions.
Parameters: - atoms (quippy.Atoms) – list of atoms to calculate the vector for.
- central (int) – integer element number to set as the central atom type for the SOAP calculation.
- basis (list) – of int defining which of the atoms in the conventional unit cell should be retained as a unique part of the basis.