permImportance: Evaluates interaction importance thhrough permutation

In sumbose/iRF: Iterative Random Forests

Description

Computes the prediction accuracy of a fitted Random Forest evaluated on data for which all columns have been permuted except for the specified interaction. For classification, accuracy is measured by AUROC. For regression, accuracy is measured by decrease in variance.

Usage

  permImportance(rfobj, x, y, ints,
                 n.perms=3,
                 varnames.group=NULL,
                 n.cores=1,
  )

Arguments

rfobj	Fitted randomForest object
x	numeric matrix of predictors
y	response vector
ints	a character vector specifying interactions, features separated by '_', as returned by iRF
n.perms	number of times to permute data matrix
varnames.grp	If features can be grouped based on their demographics or correlation patterns, use the group of features or “hyper-feature”s to conduct random intersection trees
n.cores	number of cores to parallelize over

Value

A numeric vector of the same length as ints giving the prediction accuracy with all other features permuted

See Also

randomForest

sumbose/iRF documentation built on March 12, 2021, 7:36 a.m.