occ::descriptors::PointwiseDistanceDistribution Class Reference

Pointwise Distance Distribution for crystals. More...

#include <pdd_amd.h>

Public Member Functions
	PointwiseDistanceDistribution (const crystal::Crystal &crystal, int k, const PointwiseDistanceDistributionConfig &config=PointwiseDistanceDistributionConfig{})
	Construct PDD from crystal structure.
const Vec &	weights () const
	Get the weights for each environment.
const Mat &	distances () const
	Get the distance matrix (environments as columns)
Vec	average_minimum_distance () const
	Calculate Average Minimum Distance from this PDD.
Mat	matrix () const
	Get the full PDD matrix (weights + distances) - assembled on demand.
size_t	size () const
	Number of unique chemical environments.
int	k () const
	Number of neighbors considered.
const Eigen::MatrixXi &	groups () const
	Get grouping information if available.

Detailed Description

Pointwise Distance Distribution for crystals.

Based on the work by Widdowson et al., providing geometry-based crystallographic descriptors independent of unit cell choice.

The PDD is a matrix where each row corresponds to a unique chemical environment and contains the k nearest neighbor distances along with a weight indicating the relative abundance of that environment.

References:

• Widdowson, D. & Kurlin, V. (2022). Average Minimum Distances of periodic point sets.
• https://github.com/dwiddo/average-minimum-distance

Constructor & Destructor Documentation

PointwiseDistanceDistribution()

occ::descriptors::PointwiseDistanceDistribution::PointwiseDistanceDistribution	(	const crystal::Crystal &	crystal,
		int	k,
		const PointwiseDistanceDistributionConfig &	config = PointwiseDistanceDistributionConfig{}
	)

Construct PDD from crystal structure.

Parameters

crystal	The crystal structure to analyze
k	Number of nearest neighbors to consider
config	Configuration options for the calculation

Member Function Documentation

average_minimum_distance()

Vec occ::descriptors::PointwiseDistanceDistribution::average_minimum_distance

(

)

const

Calculate Average Minimum Distance from this PDD.

Computes the weighted average of each distance column, providing a k-dimensional vector representing the average k-nearest neighbor distances across all chemical environments.

distances()

const Mat & occ::descriptors::PointwiseDistanceDistribution::distances ( ) const

inline

Get the distance matrix (environments as columns)

groups()

const Eigen::MatrixXi & occ::descriptors::PointwiseDistanceDistribution::groups ( ) const

inline

Get grouping information if available.

Returns a matrix where each column corresponds to a chemical environment and contains the indices of asymmetric unit atoms in that environment. Unused entries are filled with -1.

k()

int occ::descriptors::PointwiseDistanceDistribution::k ( ) const

inline

Number of neighbors considered.

matrix()

Mat occ::descriptors::PointwiseDistanceDistribution::matrix ( ) const

inline

Get the full PDD matrix (weights + distances) - assembled on demand.

size()

size_t occ::descriptors::PointwiseDistanceDistribution::size ( ) const

inline

Number of unique chemical environments.

weights()

const Vec & occ::descriptors::PointwiseDistanceDistribution::weights ( ) const

inline

Get the weights for each environment.

---

The documentation for this class was generated from the following file:

• /home/runner/work/occ/occ/include/occ/descriptors/pdd_amd.h