HoeffdingTreeClassifier<T>

Models & Types · AiDotNet.Classification.Online

Implements the Hoeffding Tree (Very Fast Decision Tree) for online classification.

Inheritance: ClassifierBase<T> → HoeffdingTreeClassifier<T>

Implements: IOnlineClassifier<T>

For Beginners

Hoeffding Tree is a decision tree that can be built incrementally from streaming data. It uses the Hoeffding bound to determine when enough samples have been seen to make a confident split decision.

How It Works

How it works:

  • Start with a single leaf node
  • As samples arrive, update statistics at each leaf
  • When enough samples are seen, use Hoeffding bound to decide if the best split is significantly better
  • If yes, split the leaf into internal node with two children
  • Repeat for all leaves as new samples arrive

The Hoeffding bound: With probability (1 - δ), the true best attribute is within ε of the observed best, where ε = sqrt(R² * ln(1/δ) / (2n)) and n is the number of samples.

Advantages:

  • Constant memory per node (bounded statistics)
  • Processes each sample only once
  • Converges to batch decision tree with enough samples
  • Suitable for infinite data streams

Reference: Domingos & Hulten, “Mining High-Speed Data Streams” (2000)

Example

// Create Hoeffding tree for incremental stream classification
var options = new HoeffdingTreeOptions<double>();
            
// Prepare streaming data batch
var features = new Matrix<double>(new double[,] { { 1.0, 1.1 }, { 1.2, 0.9 }, { 0.8, 1.0 }, { 5.0, 5.1 }, { 5.2, 4.9 }, { 4.8, 5.0 } });
var labels = new Vector<double>(new double[] { 0, 0, 0, 1, 1, 1 });
            
// Train incrementally, splitting nodes when Hoeffding bound is satisfied
var result = new AiModelBuilder<double, Matrix<double>, Vector<double>>()
    .ConfigureModel(new HoeffdingTreeClassifier<double>(options))
    .Build(features, labels);
            
// Predict class for new sample
var newSample = new Matrix<double>(new double[,] { { 1.1, 1.0 } });
var prediction = result.Predict(newSample);
// Result is available in the returned value

Constructors

Constructor Summary
HoeffdingTreeClassifier(HoeffdingTreeOptions<T> options) Creates a new Hoeffding Tree classifier.

Properties

Property Summary
bool IsWarm { get; } Gets whether the model has seen at least one sample.
long SamplesSeen { get; set; } Gets the total number of samples the model has seen.

Methods

Method Summary
IFullModel<T, Matrix<T>, Vector<T>> WithParameters(Vector<T> parameters) Returns a fresh instance — tree structure cannot be reconstructed from flat parameters.
ModelOptions GetOptions()
Vector<T> Predict(Matrix<T> input)
void PartialFit(Matrix<T> features, Vector<T> labels) Updates the model with a batch of training samples.
void PartialFit(Vector<T> features, T label) Updates the model with a batch of training samples.
void Train(Matrix<T> x, Vector<T> y)