ExtraTree

Package

weka.classifiers.trees

Synopsis

Class for generating a single Extra-Tree. Use with the RandomCommittee meta classifier to generate an Extra-Trees forest for classification or regression. This classifier requires all predictors to be numeric. Missing values are not allowed. Instance weights are taken into account. For more information, see

Pierre Geurts, Damien Ernst, Louis Wehenkel (2006). Extremely randomized trees. Machine Learning. 63(1):3-42.

This classifier is available in the extraTrees package for Weka >= 3.7.8.

Options

The table below describes the options available for ExtraTree.

Option	Description
debug	If set to true, classifier may output additional info to the console.
k	Number of attributes to randomly choose at a node. If values is -1, (m - 1) will be used for regression problems, and Math.rint(sqrt(m - 1)) for classification problems, where m is the number of predictors, as specified in Geurts et al.
nmin	The minimum number of instances required at a node for splitting to be considered. If value is -1, 5 will be used for regression problems and 2 for classification problems, as specified in Geurts et al.
seed	The random number seed to be used.

Capabilities

The table below describes the capabilities of ExtraTree.

Capability	Supported
Class	Numeric class, Nominal class, Binary class
Attributes	Date attributes, Numeric attributes
Min # of instances	1