randomForestCI (version 1.0.0)

randomForestInfJackMulticlass: The infinitesimal jackknife for random forests (multiclass target variable)

Description

The infinitesimal jackknife for random forests (multiclass target variable)

Usage

randomForestInfJackMulticlass(rf, newdata, calibrate = TRUE)

Arguments

rf

A random forest trained with replace = TRUE and keep.inbag = TRUE

newdata

A set of test points at which to evaluate standard errors

calibrate

whether to apply calibration to mitigate Monte Carlo noise warning: if calibrate = FALSE, some variance estimates may be negative due to Monte Carlo effects if the number of trees in rf is too small