bigrf (version 0.1-12)

merge-methods: Merge Two Random Forests

Description

Merge two random forests into one. This is useful, for example, for building random forests in parallel on multiple machines, then combining them into a single forest.

Usage

"merge"(x, y)

Arguments

x

A random forest of class "bigcforest".

y

A random forest of class "bigcforest", built using the same data sources and forest-building parameters as x.

Value

The merged forest of the same class as the input arguments.

Methods

signature(x = "bigcforest", y = "bigcforest")

Merges the classification random forests x and y.

Details

These methods copy all the trees from y into x, and calculates the error statistics and confusion matrices of the merged forest.

References

Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.

Breiman, L. & Cutler, A. (n.d.). Random Forests. Retrieved from http://www.stat.berkeley.edu/~breiman/RandomForests/cc_home.htm.

Examples

# Classify cars in the Cars93 data set by type (Compact, Large,
# Midsize, Small, Sporty, or Van).

# Load data.
data(Cars93, package="MASS")
x <- Cars93
y <- Cars93$Type

# Select variables with which to train model.
vars <- c(4:22)

# Run model, grow 15 trees.
forest <- bigrfc(x, y, ntree=15L, varselect=vars, cachepath=NULL)

# Build a second forest.
forest2 <- bigrfc(x, y, ntree=10L, varselect=vars, cachepath=NULL)

# Merge the two forests.
big.forest <- merge(forest, forest2)