Package jgpml.covariancefunctions

Interface CovarianceFunction

All Known Implementing Classes:

CovLINard, CovLINone, CovNNone, CovNNoneNoise, CovNoise, CovSEard, CovSEiso, CovSum

public interface CovarianceFunction

Method Summary

Modifier and Type	Method	Description
Jama.Matrix	compute(Jama.Matrix loghyper, Jama.Matrix X)	Compute covariance matrix of a dataset X
Jama.Matrix[]	compute(Jama.Matrix loghyper, Jama.Matrix X, Jama.Matrix Xstar)	Compute compute test set covariances
Jama.Matrix	computeDerivatives(Jama.Matrix loghyper, Jama.Matrix X, int index)	Coompute the derivatives of this CovarianceFunction with respect to the hyperparameter with index idx
int	numParameters()	Returns the number of hyperparameters of thisCovarianceFunction

Method Details

numParameters

int numParameters()

Returns the number of hyperparameters of thisCovarianceFunction

Returns:

number of hyperparameters

compute

Jama.Matrix compute(Jama.Matrix loghyper,
 Jama.Matrix X)

Compute covariance matrix of a dataset X

Parameters:

loghyper - column Matrix of hyperparameters

X - input dataset

Returns:

K covariance Matrix

compute

Jama.Matrix[] compute(Jama.Matrix loghyper,
 Jama.Matrix X,
 Jama.Matrix Xstar)

Compute compute test set covariances

Parameters:

loghyper - column Matrix of hyperparameters

X - input dataset

Xstar - test set

Returns:

[K(Xstar, Xstar) K(X,Xstar)]

computeDerivatives

Jama.Matrix computeDerivatives(Jama.Matrix loghyper,
 Jama.Matrix X,
 int index)

Coompute the derivatives of this CovarianceFunction with respect to the hyperparameter with index idx

Parameters:

loghyper - hyperparameters

X - input dataset

index - hyperparameter index

Returns:

Matrix of derivatives