R: Generate cross-validation partitions

generateCVRuns {TunePareto}

R Documentation

Generate cross-validation partitions

Description

This function generates a set of partitions for a cross-validation. It can be employed if the same cross-validation settings should be used in the objective functions of several experiments. The resulting fold list can be passed to the cross-validation objective functions (see predefinedObjectiveFunctions) and the internal cross-validation precalculation function crossValidation.

Usage

generateCVRuns(labels, 
               ntimes = 10, 
               nfold = 10, 
               leaveOneOut = FALSE, 
               stratified = FALSE)

Arguments

`labels`	A vector of class labels of the data set to be used for the cross-validation.
`nfold`	The number of groups of the cross-validation. Ignored if `leaveOneOut=TRUE`.
`ntimes`	The number of repeated runs of the cross-validation. Ignored if `leaveOneOut=TRUE`.
`leaveOneOut`	If this is true, a leave-one-out cross-validation is performed, i.e. each sample is left out once in the training phase and used as a test sample
`stratified`	If set to true, a stratified cross-validation is carried out. That is, the percentage of samples from different classes in the cross-validation folds corresponds to the class sizes in the complete data set. If set to false, the folds may be unbalanced.

Value

A list with ntimes elements, each representing a cross-validation run. Each of the runs is a list of nfold vectors specifying the indices of the samples to be left out in the folds.

Examples


# precalculate the cross-validation partitions
foldList <- generateCVRuns(labels = iris[, ncol(iris)],
                           ntimes = 10,
                           nfold = 10,
                           stratified=TRUE)

 # build a list of objective functions
objectiveFunctions <- list(cvError(foldList=foldList),
                           cvSensitivity(foldList=foldList,caseClass="setosa"))

# pass them to tunePareto
print(tunePareto(data = iris[, -ncol(iris)], 
                 labels = iris[, ncol(iris)],
                 classifier = tunePareto.knn(),
                 k = c(3,5,7,9),
                 objectiveFunctions = objectiveFunctions))