RF.wrap {MCRestimate} | R Documentation |
Wrapper function for different classification methods used
by MCRestimator
. These functions are mainly used within the
function MCRestimate
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,...)
x,y |
x is a matrix where each row refers to a sample a each colum 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 |
|
gamma |
parameter for support vector machines |
kernel |
parameter for support vector machines |
... |
Further parameters |
Every function return a predict function which can be used to predict the classes for a new data set.
Markus Ruschhaupt mailto:m.ruschhaupt@dkfz.de
library(MCRestimate) 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)