Joel Therrien
fdc708dad5
Add support for making predictions without specifying training data Add support for adding trees to an existing forest Add support for toggling displayProgress Also reduced the size of the package by removing some unused dependency classes.
34 lines
1.3 KiB
R
34 lines
1.3 KiB
R
% Generated by roxygen2: do not edit by hand
|
|
% Please edit documentation in R/connectToData.R
|
|
\name{connectToData}
|
|
\alias{connectToData}
|
|
\title{Connect To Data}
|
|
\usage{
|
|
connectToData(forest, responses, covariateData)
|
|
}
|
|
\arguments{
|
|
\item{forest}{The forest to connect data too}
|
|
|
|
\item{responses}{The responses in the data; aka the left hand side of the formula}
|
|
|
|
\item{covariateData}{A data.frame containing all of the covariates used in the training dataset}
|
|
}
|
|
\value{
|
|
The same forest, but connected to the training data.
|
|
}
|
|
\description{
|
|
When a trained forest is saved, the training dataset is not saved alongside
|
|
it. When it's loaded back up, it can be more convenient (and in some cases
|
|
necessary) to import the training dataset back into the Java environment so
|
|
that it's readily accessible. There are only two functions that look for the
|
|
training dataset: \code{predict}, where you can easily just specify an
|
|
alternative dataset, or \code{\link{addTrees}}, which requires the training
|
|
dataset be connected.
|
|
}
|
|
\examples{
|
|
data <- data.frame(x1=rnorm(1000), x2=rnorm(1000), y=rnorm(1000))
|
|
forest <- train(y~x1+x2, data, ntree=100, numberOfSplits=0, nodeSize=1, mtry=1)
|
|
forest$dataset <- NULL # what the forest looks like after being loaded
|
|
|
|
forest <- connectToData(forest, data$y, data)
|
|
}
|