PrintTree: RerF Tree Printer

In rerf: Randomer Forest

Description

Prints the layout of a specified tree.

Usage

PrintTree(forest, numTree = 1, pretty = FALSE)

Arguments

forest	a rerf forest structure.
numTree	the tree number to print. (numTree=1)
pretty	boolean if TRUE the column of cut features are formatted nicely for viewing. (FALSE)

Value

a data.frame with the following information about the tree:

• nodeNum The node number

• LC The id of the left child of the node

• RC The id of the right child of the node

• CutValue The cut value of non-terminal nodes, otherwise NA.

• nodeClass The class vote of a terminal node when used for classification/prediction.

• CutFeatures a list of ordered pairs (d, w), where d is the original feature and w is the corresponding weight.

Examples

### Train RerF on numeric data ###
library(rerf)
numTree <- 1
forest <- RerF(as.matrix(iris[, 1:4]), iris[, 5], num.core = 1L)
forest.rmc <- RerF(as.matrix(iris[, 1:4]), iris[, 5], num.core = 1L, RandMatContinuous)
(out <- PrintTree(forest, numTree))
(out.rmc <- PrintTree(forest.rmc, numTree))

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