model.result.top.var: Train model Random Forest for the top-N variables N =...

In biocsuwb/EnsembleFS-package: Ensemble feature selection methods for analysis of molecular data

Train model Random Forest for the top-N variables N = 5,10,15,20,30,40,50,75,100

Description

Train model Random Forest for the top-N variables N = 5,10,15,20,30,40,50,75,100

Usage

model.result.top.var(x, y, list.selected.var, list.index.cross)

Arguments

x	input data where columns are variables and rows are observations (all numeric)
y	decision variable as a boolean vector of length equal to number of observations
list.selected.var	A list with selected variables in cross-validation
list.index.cross	A list with indexes obtained in cross-validation

Details

Train model Random Forest in cross-validation and variables selected in the cross-validation for the top-N variables N = 5,10,15,20,30,40,50,75,100

Value

A data.frame with metrics Accuracy, AUC, MCC for the top-N variables N = 5,10,15,20,30,40,50,75,100

Examples

## Not run: 

class <- data$class
data$class <- NULL

list.index.cross <- cross.val(x = data,
                             y = decisions,
                             method = 'kfoldcv',
                             params.cv = list(niter = 10, k = 3))

list.selected.var <- feature.selection(x = data,
                             y = class,
                             method = 'fs.utest',
                             list.index.cross = indexes,
                             params = list(adjust = 'holm'))

model.result <- model.result.top.var(x = data,
                                     y = class,
                                     list.selected.var = list.selected.var,
                                     list.index.cross = list.index.cross)


## End(Not run)

biocsuwb/EnsembleFS-package documentation built on Dec. 9, 2024, 5:32 p.m.