predict.randomForest

From randomForest v1.0 by Andy Liaw

predict method for random forest objects

Prediction of test data using random forest.

Keywords: classif

Usage

## S3 method for class 'randomForest':
predict(object, newdata, type="class", norm.votes=TRUE, ...)

Arguments

object: an object of class randomForest, as that created by the function randomForest.
newdata: a data frame or matrix containing new data.
type: one of class, prob. or votes, indicating the type of output: predicted class, matrix of class probabilities, or matrix of vote counts.
norm.votes: Should the vote counts be normalized (i.e., expressed as fractions)?
...: not used currently.

Value

Depending on the argument type:
classpredicted class (the one with majority vote).
probmatrix of class probabilities (one column for each class and one row for each input).
votesmatrix of vote counts (one column for each class and one row for each new input); either in raw counts or in fractions (if norm.votes=TRUE).

References

Breiman, L., Random Forests, http://oz.berkeley.edu/users/breiman/randomforest2001.pdf.

Examples

data(iris)
set.seed(111)
n.iris <- nrow(iris)
iris.test <- sample(n.iris, round(n.iris/3), replace=FALSE)
iris.rf <- randomForest(Species ~ ., data=iris[-iris.test,])
iris.pred <- predict(iris.rf, iris[iris.test,])
table(observed = iris[iris.test,"Species"], predicted = iris.pred)

Documentation reproduced from package randomForest, version 1.0, License: GPL version 2 or later

Community examples

Looks like there are no examples yet.