CompactTreeBagger

Compact ensemble of bagged decision trees

Description

CompactTreeBagger is a compact version of the TreeBagger ensemble. The compact ensemble does not contain the following: information about how the TreeBagger function grows the decision trees; the input data used for growing trees; or the training parameters (for example, minimal leaf size, number of variables sampled for each decision split at random, and so on). Use CompactTreeBagger for tasks such as predicting the response or class labels.

Creation

Create a CompactTreeBagger ensemble object from a full, trained TreeBagger ensemble by using compact.

Properties

ClassNames — Unique class names

Read-only: cell array of character vectors

This property is read-only.

Unique class names used in the training model, specified as a cell array of character vectors.

This property is empty ([]) for regression trees.

DefaultYfit — Default prediction value

Read-only: "" | "MostPopular" | numeric scalar

This property is read-only.

Default prediction value returned by predict, specified as "", "MostPopular", or a numeric scalar. This property controls the predicted value returned by the predict object function when no prediction is possible. You can set this property by using the setDefaultYfit function.

• For classification trees, you can set DefaultYfit to either "" or "MostPopular". If you specify "MostPopular" (default for classification), the property value is the name of the most probable class in the training data. If you specify "", the in-bag observations are excluded from computation of the out-of-bag error and margin.

• For regression trees, you can set DefaultYfit to any numeric scalar. The default value for regression is the mean of the response for the training data. If you set DefaultYfit to NaN, the in-bag observations are excluded from computation of the out-of-bag error and margin.

Example: CMdl = setDefaultYfit(CMdl,"MostPopular")

Data Types: single | double | char | string

DeltaCriterionDecisionSplit — Split criterion contributions for each predictor

Read-only: numeric vector

This property is read-only.

Split criterion contributions for each predictor, specified as a numeric vector. This property is a 1-by-Nvars vector, where Nvars is the number of changes in the split criterion. The software sums the changes in the split criterion over splits on each variable, then averages the sums across the entire ensemble of grown trees.

Data Types: single | double

Method — Type of ensemble

Read-only: "classification" | "regression"

This property is read-only.

Type of ensemble, specified as "classification" for classification ensembles or "regression" for regression ensembles.

NumPredictorSplit — Number of decision splits for each predictor

Read-only: numeric vector

This property is read-only.

Number of decision splits for each predictor, specified as a numeric vector. This property is a 1-by-Nvars vector, where Nvars is the number of predictor variables. Each element of NumPredictorSplit represents the number of splits on the predictor summed over all trees.

Data Types: single | double

NumTrees — Number of decision trees

Read-only: positive integer

This property is read-only.

Number of decision trees in the bagged ensemble, specified as a positive integer.

Data Types: single | double

PredictorNames — Predictor names

Read-only: cell array of character vectors

This property is read-only.

Predictor names, specified as a cell array of character vectors. The order of the elements in PredictorNames corresponds to the order in which the predictor names appear in the training data X.

SurrogateAssociation — Predictive measures of variable association

Read-only: numeric matrix

This property is read-only.

Predictive measures of variable association, specified as a numeric matrix. This property is an Nvars-by-Nvars matrix, where Nvars is the number of predictor variables. The property contains the predictive measures of variable association, averaged across the entire ensemble of grown trees.

• If you grow the ensemble with the Surrogate name-value argument set to "on", this matrix, for each tree, is filled with the predictive measures of association averaged over the surrogate splits.

• If you grow the ensemble with the Surrogate name-value argument set to "off", the SurrogateAssociation property is an identity matrix. By default, Surrogate is set to "off".

Data Types: single | double

Trees — Decision trees in ensemble

Read-only: cell array

This property is read-only.

Decision trees in the bagged ensemble, specified as a NumTrees-by-1 cell array. Each tree is a CompactClassificationTree or CompactRegressionTree object.

Object Functions

combine	Combine two ensembles
error	Error (misclassification probability or MSE)
margin	Classification margin
mdsprox	Multidimensional scaling of proximity matrix
meanMargin	Mean classification margin
outlierMeasure	Outlier measure for data in ensemble of decision trees
partialDependence	Compute partial dependence
plotPartialDependence	Create partial dependence plot (PDP) and individual conditional expectation (ICE) plots
predict	Predict responses using ensemble of bagged decision trees
proximity	Proximity matrix for data in ensemble of decision trees
setDefaultYfit	Set default value for predict

Examples

Reduce Size of Ensemble of Bagged Trees

Reduce the size of a full ensemble of bagged classification trees by removing the training data and parameters. Then, use the compact ensemble object to make predictions on new data. Using a compact ensemble improves memory efficiency.

Load the ionosphere data set.

load ionosphere

Set the random number generator to default for reproducibility.

rng("default")

Train an ensemble of 100 bagged classification trees using the entire data set. By default, TreeBagger grows deep trees.

Mdl = TreeBagger(100,X,Y,...
    Method="classification");

Mdl is a TreeBagger ensemble for classification trees.

Create a compact version of Mdl.

CMdl = compact(Mdl)

CMdl = 
  CompactTreeBagger
Ensemble with 100 bagged decision trees:
              Method:       classification
       NumPredictors:                   34
          ClassNames: 'b' 'g'

  Properties, Methods

CMdl is a CompactTreeBagger ensemble for classification trees.

Display the amount of memory used by each ensemble.

whos("Mdl","CMdl")

  Name      Size              Bytes  Class                Attributes

  CMdl      1x1              963112  CompactTreeBagger              
  Mdl       1x1             1107399  TreeBagger                     

Mdl takes up more space than CMdl.

The CMdl.Trees property is a 100-by-1 cell vector that contains the trained classification trees for the ensemble. Each tree is a CompactClassificationTree object. View the graphical display of the first trained classification tree.

view(CMdl.Trees{1},Mode="graph");

Predict the label of the mean of X by using the compact ensemble.

predMeanX = predict(CMdl,mean(X))

predMeanX = 1×1 cell array
    {'g'}

Tips

• For a CompactTreeBagger model CMdl, the Trees property contains a cell vector of CMdl.NumTrees CompactClassificationTree or CompactRegressionTree objects. View the graphical display of the t grown tree by entering:

  view(CMdl.Trees{t})

Version History

Introduced in R2009a