covMaterniso3.h File Reference

#include "Eigen/Dense"

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Functions
Eigen::MatrixXf	covMaterniso3 (const Eigen::MatrixXf &x, const Eigen::MatrixXf &z, double sf2, double ell, bool diag=false)
	Matern3 kernel. cov = sf2*(1+sqrt(3)/ell*d)*exp(-sqrt(3)/ell*d), in which d is the Euclidean distance of two points.
Eigen::MatrixXf	dist (const Eigen::MatrixXf &x, const Eigen::MatrixXf &z)
	Euclidean distance between two vectors. dist = [d(x1, z1) d(x1, z2) ... d(x1, zn) ...... d(xm, z1) d(xm, z2) ... d(xm, zn)].

---

Function Documentation

Eigen::MatrixXf covMaterniso3	(	const Eigen::MatrixXf &	x,
		const Eigen::MatrixXf &	z,
		double	sf2,
		double	ell,
		bool	diag = false
	)

Matern3 kernel. cov = sf2*(1+sqrt(3)/ell*d)*exp(-sqrt(3)/ell*d), in which d is the Euclidean distance of two points.

Parameters:

x	m by d(imension) input vector.
z	n by d(imension) input vector.
sf2	prior variance.
ell	lengthscale.
diag	if diag is true, only the diagonal of covariance matrix is returned.

Returns:

covariance matrix (or its diagonal).

Eigen::MatrixXf dist	(	const Eigen::MatrixXf &	x,
		const Eigen::MatrixXf &	z
	)

Euclidean distance between two vectors. dist = [d(x1, z1) d(x1, z2) ... d(x1, zn) ...... d(xm, z1) d(xm, z2) ... d(xm, zn)].

Parameters:

x	1 by m vector.
z	1 by n vector.

Returns:

m by n distance matrix.