CV_partition {mppR}

Cross validation partition

Description

Partition the genotype indices into training and validation sets for cross-validation (CV).

Usage

CV_partition(cross.ind, k = 5)

Arguments

cross.ind

Character vector with the same length as the number of genotypes which specifies to which cross each genotype belongs.

k

Numeric value representing the number of subsets (fold) into which data are spread within cross. Default = 5.

Details

The genotype indices are randomly assigned within cross to k subsets (folds). Then each subset is used once as validation set, the remaining data go in the training set.

Value

Return:

fold

List of k lists (one for each fold). Each fold list contains two vectors with genotypes indices of the training ($train.set) and the validation set ($val.set).

Author(s)

Vincent Garin

See Also

mpp_CV

Examples

data(mppData)

part.cv <- CV_partition(cross.ind = mppData$cross.ind, k = 5)

part.cv[[1]]$train.set
part.cv[[1]]$val.set