rfLSF {caretLSF}R Documentation

Parallel RandomForest

Description

RandomForest model building in parallel using LSF

Usage

rfLSF(x, y, workers = 10, control = lsf.ctrl(), ...)

Arguments

x a data frame or a matrix of predictor
y A response vector. If a factor, classification is assumed, otherwise regression is assumed. If omitted, randomForest will run in unsupervised mode.
workers the number of compute nodes. Note that the value of ntree will be used for each compute node.
control an optional control object for starting parallel jobs
... various options to pass to randomForest.default

Details

The function calls randomForest.default on several nodes and uses combine put all of the models back together.

Note that the call of the output object will mirror the values of x and y passed to rfLSF.

Value

a randomForest object

Author(s)

Max Kuhn

See Also

~~objects to See Also as help, ~~~

Examples

## Not run: 
  ## Classification:
  ##data(iris)
  set.seed(71)
  iris.rf <- rfLSF(
                   iris[, 1:4], 
                   iris$Species, 
                   importance=TRUE,
                   proximity=TRUE)

## End(Not run)

[Package caretLSF version 1.14 Index]