k_feature: Features isotropic kernel

In ick003/GPspt: Using Gaussian process models for spatio-temporal data

Description

Features isotropic kernel

Usage

k_feature(f, fp, param = list(f1 = 1, f2 = 1))

Arguments

f	Features for which the kernel is calculated
fp	Features for which the kernel is calculated
param	list of parameters specific to the kernel

Examples

f = matrix(rnorm(100,0,5), ncol = 5)
fd1 = f[rep(1:nrow(f), nrow(f)),]
fd2 = f[rep(1:nrow(f), each = nrow(f)),]
fdd = cbind(fd1,fd2)
param = list(s1 = 1, s2 = 10)
kern = apply(fdd,1, function(x) k_feature(x[1:ncol(f)],x[(ncol(f)+1):ncol(fdd)],param = param))
k = matrix(kern, ncol = nrow(f))

ick003/GPspt documentation built on March 16, 2020, 3:34 a.m.