TreeBagger class -

Bootstrap aggregation for ensemble of decision trees

Description

TreeBagger bags an ensemble of decision trees for either classification or regression. Bagging stands for bootstrap aggregation. Every tree in the ensemble is grown on an independently drawn bootstrap replica of input data. Observations not included in this replica are "out of bag" for this tree. To compute prediction of an ensemble of trees for unseen data, TreeBagger takes an average of predictions from individual trees. To estimate the prediction error of the bagged ensemble, you can compute predictions for each tree on its out-of-bag observations, average these predictions over the entire ensemble for each observation and then compare the predicted out-of-bag response with the true value at this observation.

TreeBagger relies on the classregtree functionality for growing individual trees. In particular, classregtree accepts the number of features selected at random for each decision split as an optional input argument.

The Compact property contains another class, CompactTreeBagger, with sufficient information to make predictions using new data. This information includes the tree ensemble, variable names, and class names (for classification). CompactTreeBagger requires less memory than TreeBagger, but only TreeBagger has methods for growing more trees for the ensemble. Once you grow an ensemble of trees using TreeBagger and no longer need access to the training data, you can opt to work with the compact version of the trained ensemble from then on.

Construction

TreeBagger

Create ensemble of bagged decision trees

Methods

append	Append new trees to ensemble
compact	Compact ensemble of decision trees
error	Error (misclassification probability or MSE)
fillProximities	Proximity matrix for training data
growTrees	Train additional trees and add to ensemble
margin	Classification margin
mdsProx	Multidimensional scaling of proximity matrix
meanMargin	Mean classification margin
oobError	Out-of-bag error
oobMargin	Out-of-bag margins
oobMeanMargin	Out-of-bag mean margins
oobPredict	Ensemble predictions for out-of-bag observations
predict	Predict response

Properties

ClassNames	Names of classes
ComputeOOBPrediction	Flag to compute out-of-bag predictions
ComputeOOBVarImp	Flag to compute out-of-bag variable importance
Cost	Misclassification costs
DefaultYfit	Default value returned by `predict` and `oobPredict`
DeltaCritDecisionSplit	Split criterion contributions for each predictor
FBoot	Fraction of in-bag observations
MergeLeaves	Flag to merge leaves that do not improve risk
Method	Method used by trees (classification or regression)
MinLeaf	Minimum number of observations per tree leaf
NTrees	Number of decision trees in ensemble
NVarToSample	Number of variables for random feature selection
OOBIndices	Indicator matrix for out-of-bag observations
OOBInstanceWeight	Count of out-of-bag trees for each observation
OOBPermutedVarCountRaiseMargin	Variable importance for raising margin
OOBPermutedVarDeltaError	Variable importance for classification error
OOBPermutedVarDeltaMeanMargin	Variable importance for classification margin
OutlierMeasure	Measure for determining outliers
Prior	Prior class probabilities
Proximity	Proximity matrix for observations
Prune	Flag to prune trees
SampleWithReplacement	Flag to sample with replacement
TreeArgs	Cell array of arguments for `classregtree`
Trees	Decision trees in ensemble
VarNames	Variable names
X	X data used to create ensemble
Y	Y data used to create ensemble

Copy Semantics

Value. To learn how this affects your use of the class, see Comparing Handle and Value Classes in the MATLAB Object-Oriented Programming documentation.