Prune each of the trees returned by bagging.
bagging
# S3 method for classbagg prune(tree, cp=0.01,...)
an object returned by bagging (calling this tree is needed by the generic function prune in package rpart).
tree
prune
rpart
complexity parameter, see prune.rpart.
prune.rpart
additional arguments to prune.rpart.
An object of the same class as tree with the trees pruned.
By default, bagging grows classification trees of maximal size. One may want to prune each tree, however, it is not clear whether or not this may decrease prediction error.
# NOT RUN { data("Glass", package = "mlbench") library("rpart") mod <- bagging(Type ~ ., data=Glass, nbagg=10, coob=TRUE) pmod <- prune(mod) print(pmod) # }
Run the code above in your browser using DataLab