AutoMLEnsembleModel<T>

Models & Types ยท AiDotNet.AutoML

A simple tabular ensemble model used as a facade-safe AutoML final model.

Inheritance: ModelBase<T, Matrix<T>, Vector<T>> โ†’ AutoMLEnsembleModel<T>

Implements: IGeneratedParameterRegistrar<T>

For Beginners

Instead of trusting one model, an ensemble uses multiple models and combines their answers. This often improves stability and accuracy.

How It Works

This ensemble combines multiple IFullModel members by averaging (regression/binary) or voting (multi-class) over their predictions.

Example

var features = new Matrix<double>(new double[,] { { 1.0, 2.0 }, { 3.0, 4.0 }, { 5.0, 6.0 }, { 7.0, 8.0 } });
var targets = new Vector<double>(new double[] { 3.1, 7.2, 11.0, 15.1 });
            
// Any IFullModel over the same input and output types can join the ensemble.
var models = new List<IFullModel<double, Matrix<double>, Vector<double>>>
{
    new SimpleRegression<double>(),
    new SimpleRegression<double>(),
    new SimpleRegression<double>()
};
            
var result = new AiModelBuilder<double, Matrix<double>, Vector<double>>()
    .ConfigureModel(new AutoMLEnsembleModel<double>(models, PredictionType.Regression))
    .Build(features, targets);
            
var testData = new Matrix<double>(new double[,] { { 9.0, 10.0 } });
Vector<double> predictions = result.Predict(testData);

Constructors

Constructor Summary
AutoMLEnsembleModel() Initializes a new instance of the AutoMLEnsembleModel class.
AutoMLEnsembleModel(IEnumerable<IFullModel<T, Matrix<T>, Vector<T>>> members, PredictionType predictionType, IReadOnlyList<double> weights)

Properties

Property Summary
ILossFunction<T> DefaultLossFunction { get; }
IReadOnlyList<IFullModel<T, Matrix<T>, Vector<T>>> Members { get; } Gets the member models in the ensemble.
PredictionType PredictionType { get; set; } Gets or sets the prediction type used to combine outputs (regression vs classification).
double[] Weights { get; set; } Gets or sets the per-member weights used when combining predictions.

Methods

Method Summary
Dictionary<string, T> GetFeatureImportance()
IEnumerable<int> GetActiveFeatureIndices()
IFullModel<T, Matrix<T>, Vector<T>> WithParameters(Vector<T> parameters)
ModelMetadata<T> GetModelMetadata()
Vector<T> ComputeGradients(Matrix<T> input, Vector<T> target, ILossFunction<T> lossFunction)
Vector<T> Predict(Matrix<T> input)
bool IsFeatureUsed(int featureIndex)
void ApplyGradients(Vector<T> gradients, T learningRate)
void LoadModel(string filePath)
void LoadState(Stream stream)
void SaveModel(string filePath)
void SaveState(Stream stream)
void SetActiveFeatureIndices(IEnumerable<int> featureIndices)
void Train(Matrix<T> input, Vector<T> expectedOutput)