kGauss: Gauss (Squared-Exponential) Kernel

In kergp: Gaussian Process Laboratory

Description

Gauss (or squared exponential) covariance function.

Usage

kGauss(d)

Arguments

d	Dimension.

Value

An object of class "covMan" with default parameters: 1 for ranges and variance values.

References

C.E. Rasmussen and C.K.I. Williams (2006), Gaussian Processes for Machine Learning, the MIT Press, http://www.GaussianProcess.org/gpml/

Examples

kGauss()  # default: d = 1, nu = 5/2
myGauss <- kGauss(d = 2)
coef(myGauss) <- c(range = c(2, 5), sigma2 = 0.1)
myGauss

Example output

Loading required package: Rcpp
Loading required package: testthat
Loading required package: nloptr
'User' covariance kernel
o Description: Gauss kernel 
o Dimension 'd' (nb of inputs): 1
o Parameters: "theta_1", "sigma2"
o Number of parameters: 2
o Accept matrix inputs.
o Analytical gradient is provided.
o Param. values: 
        Value Lower Upper
theta_1     1 1e-08   Inf
sigma2      1 1e-08   Inf
'User' covariance kernel
o Description: Gauss kernel 
o Dimension 'd' (nb of inputs): 2
o Parameters: "theta_1", "theta_2", "sigma2"
o Number of parameters: 3
o Accept matrix inputs.
o Analytical gradient is provided.
o Param. values: 
        Value Lower Upper
theta_1   2.0 1e-08   Inf
theta_2   5.0 1e-08   Inf
sigma2    0.1 1e-08   Inf

kergp documentation built on March 18, 2021, 5:06 p.m.