grammpred: Gramm Matrix for Test dataset

In RMKL: Multiple Kernel Learning for Classification or Regression Problems

Description

This function creates gramm matrix for test dataset based upon several types of kernel.

Usage

grammpred(xtrain, xtest, kernel, sigma, degree, scale)

Arguments

xtrain	Matrix of predictors for the training set
xtest	Matrix of predictors for the test set
kernel	Type of kernel used to compute a gramm matrix
sigma	Hyperparameters for radial kernels
degree, scale	Hyperparameter for polynomial kernel

Value

Gramm matrix for test set

Examples

library(kernlab)
data(benchmark.data)
example.data=benchmark.data[[1]]
#Create split between training samples and test samples
training.samples=sample(1:dim(example.data)[1],floor(0.7*dim(example.data)[1]),replace=FALSE)
xtrain=example.data[training.samples,1:2]
xtest=example.data[-training.samples,1:2]
#Generate linear kernel
grammpred(xtrain,xtest,'linear',0,0,0)
#Generate radial kernels with different values for the hyperparameter.
grammpred(xtrain,xtest,'radial',2^seq(-3:0),0,0)

RMKL documentation built on May 2, 2019, 7:55 a.m.