LedPred (version 1.6.0)

createModel: Create the model with the optimal features

Description

createModel function creates a SVM model from the training data set with the selected features.

Usage

createModel(data, cl = 1, kernel = "radial", cost = 1, gamma = 1, valid.times = 10, feature.ranking = NULL, feature.nb = NULL, file.prefix = NULL)

Arguments

data

data.frame containing the training set

cl

integer indicating the column number corresponding to the response vector that classify positive and negative regions (default = 1)

kernel

SVM kernel, a character string: "linear" or "radial". (default = "radial")

cost

The SVM cost parameter for both linear and radial kernels. If NULL (default), the function mcTune is run.

gamma

The SVM gamma parameter for radial kernel. If radial kernel and NULL (default), the function mcTune is run.

valid.times

Integer indicating how many times the training set will be split for the cross validation step (default = 10). This number must be smaller than positive and negative sets sizes.

feature.ranking

List of ordered features.

feature.nb

the optimal number of feature to use from the list of ordered features.

file.prefix

A character string that will be used as a prefix followed by "_model.RData" for the resulting model file, if it is NULL (default), no model is saved

Value

the best SVM model

Examples

    data(crm.features)
    cost <- 1
    gamma <- 1
    data(feature.ranking)
    feature.nb <- 70
#svm.model <- createModel(data.granges=crm.features, cost=cost, gamma=gamma,
#    feature.ranking=feature.ranking, feature.nb=feature.nb)
#feature.weights <- as.data.frame(t(t(svm.model$coefs) %*% svm.model$SV))