GrowUnsupervisedForest: Creates Urerf Tree.

Description Usage Arguments Value

View source: R/UrerfHelperFunctions.R

Description

Creates Urerf Tree.

Usage

1
2
3
4
GrowUnsupervisedForest(X, MinParent = 1, trees = 100, MaxDepth = Inf,
  bagging = 0.2, replacement = TRUE, FUN = makeAB, options = list(p
  = ncol(X), d = ceiling(ncol(X)^0.5), sparsity = 1/ncol(X)),
  Progress = TRUE, splitCrit = "twomeans", LinearCombo = TRUE)

Arguments

X

an Nxd matrix or Data frame of numeric values.

MinParent

the minimum splittable node size (MinParent=1).

trees

the number of trees to grow in a forest (trees=100).

MaxDepth

the maximum depth allowed in a forest (MaxDepth=Inf).

bagging

only used experimentally. Determines the hold out size if replacement=FALSE (bagging=.2).

replacement

method used to determine boot strap samples (replacement=TRUE).

FUN

the function to create the rotation matrix used to determine mtry features.

options

options provided to FUN.

Progress

logical that determines whether to show tree creation status (Progress=TRUE).

splitCrit

split based on twomeans(splitCrit="twomeans") or BIC test(splitCrit="bicfast")

LinearCombo

logical that determines whether to use linear combination of features. (LinearCombo=TRUE).

Value

tree


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