rxDForestUtils: Utility Functions for rxDForest

  • [アーティクル]

Description

Utility Functions for rxDForest.

Usage

  rxVarImpPlot(x, sort = TRUE, n.var = 30, main = deparse(substitute(x)),   ...  )
  rxLeafSize(x, use.weight = TRUE)
  rxTreeDepth(x)
  rxTreeSize(x, terminal = TRUE)
  rxVarUsed(x, by.tree = FALSE, count = TRUE)
  rxGetTree(x, k = 1)

Arguments

x

an object of class rxDForest or rxDTree.

sort

logical value. If TRUE, the variables will be sorted in decreasing importance.

n.var

an integer specifying the number of variables to show when sort=FALSE.

main

a character string specifying the main title for the plot.

...

other arguments to be passed on to dotchart.

use.weight

logical value. If TRUE, the leaf size is measured by the total weight of its observations instead of the total number of its observations.

terminal

logical value. If TRUE, only the terminal nodes will be counted.

by.tree

logical value. If TRUE, the list of variables used will be broken down by trees.

count

logical value. If TRUE, the frequencies that variables appear in trees will be returned.

k

an integer specifying the index of the tree to be extracted.

Value

  • rxVarImpPlot - plots a dotchart of the variable importance as measured by the decision forest.

  • rxLeafSize - returns the size of the terminal nodes in the decision forest.

  • rxTreeDepth - returns the depth of trees in the decision forest.

  • rxTreeSize - returns the size of trees in terms of the number of nodes in the decision forest.

  • rxVarUsed - finds out the variables actually used in the decision forest.

  • rxGetTree - extracts a single tree from the decision forest.

Author(s)

Microsoft Corporation Microsoft Technical Support

References

randomForest .

See Also

rxDForest, rxDTree, rxBTrees.

Examples


 set.seed(1234)

 # classification
 iris.sub <- c(sample(1:50, 25), sample(51:100, 25), sample(101:150, 25))
 iris.dforest <- rxDForest(Species ~ Sepal.Length + Sepal.Width + Petal.Length + Petal.Width, 
     data = iris[iris.sub, ], importance = TRUE)

 rxVarImpPlot(iris.dforest)
 rxTreeSize(iris.dforest)
 rxVarUsed(iris.dforest)
 rxGetTree(iris.dforest)