ClassificationTree
R2026bBinary decision tree for multiclass classification
Description
ClassificationTree is a decision tree with binary splits for
classification. Use the predict function of
ClassificationTree to predict labels for new data. Because the
ClassificationTree object contains the data used for training, you
can use the object to compute resubstitution predictions.
Creation
Create a ClassificationTree object by using fitctree.
Properties
Object Functions
compact | Reduce size of machine learning model |
compareHoldout | Compare accuracies of two classification models using new data |
crossval | Cross-validate machine learning model |
cvloss | Classification error by cross-validation for classification tree model |
edge | Classification edge for classification tree model |
gather | Gather properties of Statistics and Machine Learning Toolbox object from GPU |
lime | Local interpretable model-agnostic explanations (LIME) |
loss | Classification loss for classification tree model |
margin | Classification margins for classification tree model |
nodeVariableRange | Retrieve variable range of decision tree node |
partialDependence | Compute partial dependence |
plotPartialDependence | Create partial dependence plot (PDP) and individual conditional expectation (ICE) plots |
predict | Predict labels using classification tree model |
predictorImportance | Estimates of predictor importance for classification tree |
prune | Produce sequence of classification subtrees by pruning classification tree |
resubEdge | Resubstitution classification edge for classification tree model |
resubLoss | Resubstitution classification loss for classification tree model |
resubMargin | Resubstitution classification margins for classification tree model |
resubPredict | Classify observations in classification tree by resubstitution |
shapley | Shapley values |
surrogateAssociation | Mean predictive measure of association for surrogate splits in classification tree |
testckfold | Compare accuracies of two classification models by repeated cross-validation |
view | View classification tree |
Examples
Train a classification tree using the ionosphere data set.
load ionosphere
Mdl = fitctree(X,Y)Mdl =
ClassificationTree
ResponseName: 'Y'
CategoricalPredictors: []
ClassNames: {'b' 'g'}
ScoreTransform: 'none'
NumObservations: 351
Properties, Methods
Mdl is a ClassificationTree model object.
fitctree grows deep decision trees by default. You can grow shallower trees to reduce model complexity or computation time.
Load the ionosphere data set.
load ionosphereThe default values of the name-value arguments that control tree depth are:
n - 1forMaxNumSplits, wherenis the training sample size1forMinLeafSize10forMinParentSize
These default values tend to grow deep trees for large training sample sizes.
Train a classification tree using the default values. Cross-validate the model by using 10-fold cross-validation.
rng(1) % For reproducibility MdlDefault = fitctree(X,Y,CrossVal="on");
MdlDefault is a ClassificationPartitionedModel object that contains 10 tree models in its Trained property.
Plot a histogram of the number of imposed splits on the trees. Then, view one of the trees.
numBranches = @(x)sum(x.IsBranch); mdlDefaultNumSplits = cellfun(numBranches,MdlDefault.Trained); histogram(mdlDefaultNumSplits)
view(MdlDefault.Trained{1},Mode="graph")
The average number of splits is around 15.
Train another classification tree, but set the maximum number of splits to 7, which is about half the mean number of splits in the default classification tree. Cross-validate the model by using 10-fold cross-validation.
Mdl7 = fitctree(X,Y,MaxNumSplits=7,CrossVal="on"); view(Mdl7.Trained{1},Mode="graph")
Compare the cross-validation classification errors of the models.
classErrorDefault = kfoldLoss(MdlDefault)
classErrorDefault = 0.1168
classError7 = kfoldLoss(Mdl7)
classError7 = 0.1311
Mdl7 is much less complex than MdlDefault and performs only slightly worse than MdlDefault.
More About
References
[1] Breiman, L., J. Friedman, R. Olshen, and C. Stone. Classification and Regression Trees. Boca Raton, FL: CRC Press, 1984.
