wrapper.fkt: Wrapper function for different classification methods

In MCRestimate: Misclassification error estimation with cross-validation

Description

Wrapper function for different classification methods used by MCRestimator. These functions are mainly used within the function MCRestimate

Usage

RF.wrap(x,y,...)
PAM.wrap(x,y,threshold,...)
PLR.wrap(x,y,kappa=0,eps=1e-4,...)
SVM.wrap(x,y,gamma = NULL, kernel = "radial", ...)
GPLS.wrap(x,y,...)

Arguments

x,y	x is a matrix where each row refers to a sample a each column refers to a gene; y is a factor which includes the class for each sample
threshold	the threshold for PAM
kappa	the penalty parameter for the penalised logistic regression
eps	precision of convergence
gamma	parameter for support vector machines
kernel	parameter for support vector machines
...	Further parameters

Value

Every function return a predict function which can be used to predict the classes for a new data set.

Author(s)

Markus Ruschhaupt mailto:[e-mail]

See Also

MCRestimate

Examples

library(golubEsets)
data(Golub_Train)

class.column <- "ALL.AML"
Preprocessingfunctions <- c("varSel.highest.var")
list.of.poss.parameter <- list(threshold = 6)

Preprocessingfunctions <- c("identity")
class.function <- "PAM.wrap"
plot.label <- "Samples"

cross.outer <- 10
cross.repeat <- 7
cross.inner <- 5

PAM.estimate <- MCRestimate(Golub_Train,
		class.column,
		classification.fun = class.function,
		thePreprocessingMethods = Preprocessingfunctions,
		poss.parameters = list.of.poss.parameter,
		cross.outer = cross.outer, cross.inner = cross.inner,
		cross.repeat = cross.repeat, plot.label = plot.label)

MCRestimate documentation built on Oct. 31, 2019, 10:29 a.m.