RunFeatureImportanceCounts: Tabulate the unique feature combinations used in a single...

In rerf: Randomer Forest

Description

Computes feature importance of every unique feature used to make a split in a single tree.

Usage

RunFeatureImportanceCounts(tree, unique.projections)

Arguments

tree	a single tree from a trained RerF model with argument store.impurity = TRUE.
unique.projections	a list of all of the unique split projections used in the RerF model.

Value

feature.counts

Examples

library(rerf)
X <- iris[, -5]
Y <- iris[[5]]
store.impurity <- TRUE
FUN <- RandMatContinuous
forest <- RerF(X, Y, FUN = FUN, num.cores = 1L, store.impurity = store.impurity)
FeatureImportance(forest, num.cores = 1L, type = "C")

rerf documentation built on May 2, 2019, 8:16 a.m.