cv_sets_method: Generate k cross-validation sets
In kimsche/MetaboMate: All you need for NMR-based metabolic profiling with R!

Description Usage Arguments Details Author(s) See Also

Generate k cross-validation sets

1 2	cv_sets_method(cv.k = 7, Y, type = c("R", "DA"), method = "k-fold_stratified")

`cv.k`	Number of cross validation sets (k-fold).
`Y`	Dependent variable in vector or dummy matrix format (see Details).
`type`	Indicating if a regression (`R`) or discriminant analysis (`DA`) will be performed.
`method`	Type of cross validation: k-fold or Monte Carlo-Cross Validation (MCCV), sampling can be performed totally random or group stratified: 'k-fold', 'k-fold_stratified', 'MC', 'MC_stratified'.

If input argument Y is a dummy marix then function centre_scale has been applied beforehand.The number of cross validation sets is related to the number of samples in each group. If in doubt, set k=nrow(Y) or k=length(Y) if Y is matrix or vector, respectively.

Torben Kimhofer tkimhofer@gmail.com

center_scale

kimsche/MetaboMate documentation built on Aug. 8, 2020, 1:14 a.m.