predict.randomForest: predict method for random forest objects

Description

Prediction of test data using random forest.

Usage

# S3 method for randomForest
predict(object, newdata, type="response",
  norm.votes=TRUE, predict.all=FALSE, proximity=FALSE, nodes=FALSE,
  cutoff, ...)

Value

If object$type is regression, a vector of predicted values is returned. If predict.all=TRUE, then the returned object is a list of two components: aggregate, which is the vector of predicted values by the forest, and individual, which is a matrix where each column contains prediction by a tree in the forest.

If object$type is classification, the object returned depends on the argument type:

response: predicted classes (the classes with majority vote).
prob: matrix of class probabilities (one column for each class and one row for each input).
vote: matrix 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).

If predict.all=TRUE, then the individual component of the returned object is a character matrix where each column contains the predicted class by a tree in the forest.

If proximity=TRUE, the returned object is a list with two components: pred is the prediction (as described above) and proximity is the proximitry matrix. An error is issued if object$type is regression.

If nodes=TRUE, the returned object has a ``nodes'' attribute, which is an n by ntree matrix, each column containing the node number that the cases fall in for that tree.

NOTE: If the object inherits from randomForest.formula, then any data with NA are silently omitted from the prediction. The returned value will contain NA correspondingly in the aggregated and individual tree predictions (if requested), but not in the proximity or node matrices.

NOTE2: Any ties are broken at random, so if this is undesirable, avoid it by using odd number ntree in randomForest().

Arguments

object: an object of class randomForest, as that created by the function randomForest.
newdata: a data frame or matrix containing new data. (Note: If not given, the out-of-bag prediction in object is returned.
type: one of response, prob. or votes, indicating the type of output: predicted values, matrix of class probabilities, or matrix of vote counts. class is allowed, but automatically converted to "response", for backward compatibility.
norm.votes: Should the vote counts be normalized (i.e., expressed as fractions)? Ignored if object$type is regression.
predict.all: Should the predictions of all trees be kept?
proximity: Should proximity measures be computed? An error is issued if object$type is regression.
nodes: Should the terminal node indicators (an n by ntree matrix) be return? If so, it is in the ``nodes'' attribute of the returned object.
cutoff: (Classification only) A vector of length equal to number of classes. The `winning' class for an observation is the one with the maximum ratio of proportion of votes to cutoff. Default is taken from the forest$cutoff component of object (i.e., the setting used when running randomForest).
...: not used currently.

Author

Andy Liaw andy_liaw@merck.com and Matthew Wiener matthew_wiener@merck.com, based on original Fortran code by Leo Breiman and Adele Cutler.

References

Breiman, L. (2001), Random Forests, Machine Learning 45(1), 5-32.

Examples

Run this code

data(iris)
set.seed(111)
ind <- sample(2, nrow(iris), replace = TRUE, prob=c(0.8, 0.2))
iris.rf <- randomForest(Species ~ ., data=iris[ind == 1,])
iris.pred <- predict(iris.rf, iris[ind == 2,])
table(observed = iris[ind==2, "Species"], predicted = iris.pred)
## Get prediction for all trees.
predict(iris.rf, iris[ind == 2,], predict.all=TRUE)
## Proximities.
predict(iris.rf, iris[ind == 2,], proximity=TRUE)
## Nodes matrix.
str(attr(predict(iris.rf, iris[ind == 2,], nodes=TRUE), "nodes"))

Run the code above in your browser using DataLab