feature.selection.RFE: Recursive feature elimnation

In ProfessionalFarmer/loonR: Common function in bioinformatics analysis

Recursive feature elimnation

Description

Recursive feature elimnation

Usage

feature.selection.RFE(
  feature.df,
  group,
  functions = "lrFuncs",
  seed = 111,
  scale = TRUE,
  sizes = c(1:10),
  repeats = 5,
  number = 5,
  method = "cv",
  cores = 50
)

Arguments

feature.df	Row is sample, column is feature
functions	Default: lrFuncs. lrFuncs, rfFuncs http://topepo.github.io/caret/available-models.html There are a number of pre-defined sets of functions for several models, including: linear regression (in the object lmFuncs), random forests (rfFuncs), naive Bayes (nbFuncs), bagged trees (treebagFuncs) and functions that can be used with caret’s train function (caretFuncs).
seed	Default 111
scale	Deafult TRUE
sizes	Default c(1:5), The sizes determines the number of most important features the rfe should iterate.
repeats	For repeated k-fold cross-validation only: the number of complete sets of folds to compute
number	Either the number of folds or number of resampling iterations
method	The external resampling method: boot, repeatedcv, cv, LOOCV or LGOCV (for repeated training/test splits)
cores	cores for parallel

Examples

loonR::feature.selection.RFE(miR.df, group, functions="lrFuncs")

Recursive feature elimnation (rfe) offers a rigorous way to determine the important variables before you even feed them into a ML algo.

ProfessionalFarmer/loonR documentation built on Oct. 9, 2024, 9:56 p.m.