EnsembleClassifierBase<T>

Base Classes · AiDotNet.Classification.Ensemble

Base class for ensemble classification methods that combine multiple classifiers.

Inheritance: ClassifierBase<T> → ProbabilisticClassifierBase<T> → EnsembleClassifierBase<T>

Implements: IGeneratedParameterRegistrar<T>

Known derived types: AdaBoostClassifier<T>, DARTClassifier<T>, ExplainableBoostingClassifier<T>, ExtraTreesClassifier<T>, GradientBoostingClassifier<T>, NGBoostClassifier<T>, RandomForestClassifier<T>

For Beginners

Ensemble learning is like getting opinions from a group of experts instead of just one.

Imagine you want to predict if a movie will be successful. You could:

  1. Ask just one expert (single classifier)
  2. Ask many experts and combine their opinions (ensemble)

The second approach is usually more reliable because:

  • Individual experts may have blind spots that others don’t
  • Combining diverse opinions often leads to better decisions
  • Errors from one expert may be corrected by others

Common ensemble strategies:

  • Bagging: Train on different random subsets of data
  • Boosting: Train sequentially, focusing on mistakes
  • Voting: Let classifiers vote on the answer

How It Works

Ensemble methods combine multiple individual classifiers (base estimators) to produce a more robust and accurate prediction than any single classifier could achieve.

Properties

Property Summary
Vector<T> FeatureImportances { get; set; } Gets or sets the feature importances aggregated across all estimators.
int NEstimators { get; } The number of estimators in the ensemble.

Methods

Method Summary
Matrix<T> PredictProbabilities(Matrix<T> input) Aggregates predictions from all estimators in the ensemble.
ModelMetadata<T> GetModelMetadata()