randomForest: Random forest

In jasenfinch/metabolyseR: Methods for Pre-Treatment, Data Mining and Correlation Analyses of Metabolomics Data

Random forest

Description

Perform random forest on an AnalysisData object

Usage

randomForest(
  x,
  cls = "class",
  rf = list(),
  reps = 1,
  binary = FALSE,
  comparisons = list(),
  perm = 0,
  returnModels = FALSE,
  seed = 1234
)

## S4 method for signature 'AnalysisData'
randomForest(
  x,
  cls = "class",
  rf = list(),
  reps = 1,
  binary = FALSE,
  comparisons = list(),
  perm = 0,
  returnModels = FALSE,
  seed = 1234
)

Arguments

x	S4 object of class AnalysisData
cls	vector of sample information columns to use for response variable information. Set to NULL for unsupervised.
rf	named list of arguments to pass to randomForest::randomForest
reps	number of repetitions to perform
binary	TRUE/FALSE should binary comparisons be performed. Ignored for unsupervised and regression. Ignored if comparisons specified.
comparisons	list of comparisons to perform. Ignored for unsupervised and regression. See details.
perm	number of permutations to perform. Ignored for unsupervised.
returnModels	TRUE/FALSE should model objects be returned.
seed	random number seed

Details

Specified class comparisons should be given as a list named according to cls. Comparisons should be given as class names separated by '~' (eg. '1~2~H').

Value

An S4 object of class RandomForest.

Examples

library(metaboData)

x <- analysisData(abr1$neg[,200:300],abr1$fact) %>%
       occupancyMaximum(cls = 'day') %>%
       transformTICnorm()
       
rf <- randomForest(x,cls = 'day')

plotMDS(rf,cls = 'day')

jasenfinch/metabolyseR documentation built on Sept. 18, 2023, 1:25 a.m.