StrCorr: Compute tree strength and correlation

In rerf: Randomer Forest

Description

Computes estimates of tree strength and correlation according to the definitions in Breiman's 2001 Random Forests paper.

Usage

StrCorr(Yhats, Y)

Arguments

Yhats	predicted class labels for each tree in a forest.
Y	true class labels.

Value

scor

Examples

library(rerf)
trainIdx <- c(1:40, 51:90, 101:140)
X <- as.matrix(iris[, 1:4])
Y <- iris[[5]]
forest <- RerF(X[trainIdx, ], Y[trainIdx], num.cores = 1L)
predictions <- Predict(X[-trainIdx, ], forest, num.cores = 1L, aggregate.output = FALSE)
scor <- StrCorr(predictions, Y[-trainIdx])

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