varpro.strength: Obtain Strength Array and Other Values from a VarPro Object

In kogalur/varPro: Model-Independent Variable Selection via the Rule-Based Variable Priority (VarPro)

Obtain Strength Array and Other Values from a VarPro Object

Description

Used to parse values from a VarPro object.

Usage

varpro.strength(object,
                newdata,
                m.target = NULL,
                max.rules.tree = 150,
                max.tree = 150,
                stat = c("importance", "complement", "oob", "none"),
                membership = FALSE,
                neighbor = 5,
                seed = NULL,
                do.trace = FALSE, ...)

Arguments

object	rfsrc object.
newdata	Optional test data. If provided, returns branch and complementary branch membership of the training data corresponding to the test cases.
m.target	Character string specifying the target outcome for multivariate families. If unspecified, a default is selected automatically.
max.rules.tree	Maximum number of rules extracted per tree.
max.tree	Maximum number of trees used for rule extraction.
stat	Statistic to output. Options include "importance", "complement mean", and "oob mean".
membership	Return out-of-bag and complementary membership indices for each rule?
neighbor	Nearest neighbor parameter, used only when newdata is specified.
seed	Seed for reproducibility.
do.trace	Enable detailed trace output.
...	Additional arguments.

Details

Not intended for direct end-user use; primarily designed for internal package operations.

Examples

  

## ------------------------------------------------------------
## regression example: boston housing
## ------------------------------------------------------------

## load the data
data(BostonHousing, package = "mlbench")

o <- randomForestSRC::rfsrc(medv~., BostonHousing, ntree=100)

## call varpro.strength
varpro.strength(object = o, max.rules.tree = 10, max.tree = 15)

## call varpro.strength with test data
varpro.strength(object = o, newdata = BostonHousing[1:3,], max.rules.tree = 10, max.tree = 15)

kogalur/varPro documentation built on June 2, 2025, 6:24 a.m.