Class OptimizerHelper<T, TInput, TOutput>

Namespace

AiDotNet.Helpers

Assembly

AiDotNet.dll

public static class OptimizerHelper<T, TInput, TOutput>

Type Parameters

T

TInput

TOutput

Inheritance

object

OptimizerHelper<T, TInput, TOutput>

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Methods

CreateDatasetResult(TOutput, ErrorStats<T>?, BasicStats<T>?, BasicStats<T>?, PredictionStats<T>?, TInput, TOutput)

Creates a result object containing evaluation metrics for a specific dataset (training, validation, or test).

public static OptimizationResult<T, TInput, TOutput>.DatasetResult CreateDatasetResult(TOutput predictions, ErrorStats<T>? errorStats, BasicStats<T>? actualBasicStats, BasicStats<T>? predictedBasicStats, PredictionStats<T>? predictionStats, TInput features, TOutput y)

Parameters

predictions TOutput

The model's predictions for this dataset.

errorStats ErrorStats<T>

Statistics about prediction errors (like MSE, MAE).

actualBasicStats BasicStats<T>

Basic statistics about the actual values (min, max, mean, etc.).

predictedBasicStats BasicStats<T>

Basic statistics about the predicted values.

predictionStats PredictionStats<T>

Advanced statistics about prediction quality.

features TInput

The feature matrix (X) used for this dataset.

y TOutput

The target values for this dataset.

Returns

OptimizationResult<T, TInput, TOutput>.DatasetResult

A dataset result object containing all evaluation metrics.

CreateOptimizationInputData(TInput, TOutput, TInput, TOutput, TInput, TOutput)

Creates a data container for optimization algorithms with training, validation, and test datasets.

public static OptimizationInputData<T, TInput, TOutput> CreateOptimizationInputData(TInput xTrain, TOutput yTrain, TInput xValidation, TOutput yValidation, TInput xTest, TOutput yTest)

Parameters

xTrain TInput

Feature input for training the model.

yTrain TOutput

Target values for training the model.

xValidation TInput

Feature input for validating the model during training.

yValidation TOutput

Target values for validation.

xTest TInput

Feature input for final testing of the model.

yTest TOutput

Target values for testing.

Returns

OptimizationInputData<T, TInput, TOutput>

A structured container with all datasets needed for optimization.

CreateOptimizationResult(IFullModel<T, TInput, TOutput>, T, List<T>, List<Vector<T>>, DatasetResult, DatasetResult, DatasetResult, FitDetectorResult<T>, int)

Creates a result object containing all information about an optimization process.

public static OptimizationResult<T, TInput, TOutput> CreateOptimizationResult(IFullModel<T, TInput, TOutput> bestSolution, T bestFitness, List<T> fitnessHistory, List<Vector<T>> bestSelectedFeatures, OptimizationResult<T, TInput, TOutput>.DatasetResult trainingResult, OptimizationResult<T, TInput, TOutput>.DatasetResult validationResult, OptimizationResult<T, TInput, TOutput>.DatasetResult testResult, FitDetectorResult<T> bestFitDetectionResult, int iterationCount)

Parameters

bestSolution IFullModel<T, TInput, TOutput>

The best model found during optimization.

bestFitness T

The fitness score of the best solution (lower is better).

fitnessHistory List<T>

List of fitness scores throughout the optimization process.

bestSelectedFeatures List<Vector<T>>

List of feature vectors that were selected as most important.

trainingResult OptimizationResult<T, TInput, TOutput>.DatasetResult

Results from evaluating the model on training data.

validationResult OptimizationResult<T, TInput, TOutput>.DatasetResult

Results from evaluating the model on validation data.

testResult OptimizationResult<T, TInput, TOutput>.DatasetResult

Results from evaluating the model on test data.

bestFitDetectionResult FitDetectorResult<T>

Information about model fit quality (underfitting/overfitting).

iterationCount int

Number of iterations the optimization process ran for.

Returns

OptimizationResult<T, TInput, TOutput>

A complete optimization result object.

SelectFeatures(TInput, IEnumerable<Vector<T>>)

public static TInput SelectFeatures(TInput input, IEnumerable<Vector<T>> selectedFeatures)

Parameters

input TInput

selectedFeatures IEnumerable<Vector<T>>

Returns

TInput

SelectFeatures(TInput, List<Vector<T>>)

public static TInput SelectFeatures(TInput X, List<Vector<T>> selectedFeatures)

Parameters

X TInput

selectedFeatures List<Vector<T>>

Returns

TInput

SelectFeatures(TInput, List<int>)

public static TInput SelectFeatures(TInput X, List<int> selectedFeatures)

Parameters

X TInput

selectedFeatures List<int>

Returns

TInput