QuadraticDiscriminantAnalysis<T>

Models & Types · AiDotNet.Classification.DiscriminantAnalysis

Quadratic Discriminant Analysis classifier.

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

Implements: IGeneratedParameterRegistrar<T>

For Beginners

Quadratic Discriminant Analysis (QDA) is a classification technique that:

  1. Models each class with its own Gaussian distribution
  2. Each class has its own covariance matrix (unlike LDA)
  3. Decision boundaries are quadratic (curved) instead of linear

When to use QDA over LDA:

  • When classes have different covariance structures
  • When you have enough samples per class to estimate covariance reliably
  • When decision boundaries are naturally curved

Trade-offs:

  • More flexible than LDA (can capture curved boundaries)
  • Needs more parameters (separate covariance per class)
  • More prone to overfitting with small datasets
  • Computationally more expensive than LDA

How It Works

QDA is similar to LDA but allows each class to have its own covariance matrix, which creates quadratic (curved) decision boundaries.

Example

// Create QDA classifier with per-class covariance matrices
var options = new DiscriminantAnalysisOptions<double>();
            
// Prepare training data with classes that have different covariance structures
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 to learn per-class Gaussian distributions with quadratic boundaries
var result = new AiModelBuilder<double, Matrix<double>, Vector<double>>()
    .ConfigureModel(new QuadraticDiscriminantAnalysis<double>(options))
    .Build(features, labels);
            
// Classify new sample using quadratic discriminant function
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
QuadraticDiscriminantAnalysis(DiscriminantAnalysisOptions<T> options, IRegularization<T, Matrix<T>, Vector<T>> regularization) Initializes a new instance of the QuadraticDiscriminantAnalysis class.

Properties

Property Summary
bool SupportsParameterInitialization { get; } QDA computes parameters from per-class covariance matrices during training.

Methods

Method Summary
IFullModel<T, Matrix<T>, Vector<T>> WithParameters(Vector<T> parameters)
Matrix<T> PredictLogProbabilities(Matrix<T> input)
Matrix<T> PredictProbabilities(Matrix<T> input)
ModelMetadata<T> GetModelMetadata()
Vector<T> Predict(Matrix<T> input)
void Train(Matrix<T> x, Vector<T> y) Returns the model type identifier for this classifier.