using System.Linq; using HeuristicLab.Common; using HeuristicLab.Core; using HeuristicLab.Data; using HeuristicLab.Parameters; using HeuristicLab.Persistence.Default.CompositeSerializers.Storable; namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Classification { [StorableClass] [Item("SymbolicClassificationPruningOperator", "An operator which prunes symbolic classificaton trees.")] public class SymbolicClassificationPruningOperator : SymbolicDataAnalysisExpressionPruningOperator { private const string ImpactValuesCalculatorParameterName = "ImpactValuesCalculator"; private const string ModelCreatorParameterName = "ModelCreator"; private const string ApplyLinearScalingParmameterName = "ApplyLinearScaling"; #region parameter properties public ILookupParameter ModelCreatorParameter { get { return (ILookupParameter)Parameters[ModelCreatorParameterName]; } } public ILookupParameter ApplyLinearScalingParameter { get { return (ILookupParameter)Parameters[ApplyLinearScalingParmameterName]; } } #endregion #region properties private ISymbolicClassificationModelCreator ModelCreator { get { return ModelCreatorParameter.ActualValue; } } private BoolValue ApplyLinearScaling { get { return ApplyLinearScalingParameter.ActualValue; } } #endregion protected SymbolicClassificationPruningOperator(SymbolicClassificationPruningOperator original, Cloner cloner) : base(original, cloner) { } public override IDeepCloneable Clone(Cloner cloner) { return new SymbolicClassificationPruningOperator(this, cloner); } [StorableConstructor] protected SymbolicClassificationPruningOperator(bool deserializing) : base(deserializing) { } public SymbolicClassificationPruningOperator() { Parameters.Add(new ValueParameter(ImpactValuesCalculatorParameterName, new SymbolicClassificationSolutionImpactValuesCalculator())); Parameters.Add(new LookupParameter(ModelCreatorParameterName)); } protected override ISymbolicDataAnalysisModel CreateModel() { var model = ModelCreator.CreateSymbolicClassificationModel(SymbolicExpressionTree, Interpreter, EstimationLimits.Lower, EstimationLimits.Upper); var rows = Enumerable.Range(FitnessCalculationPartition.Start, FitnessCalculationPartition.Size); var problemData = (IClassificationProblemData)ProblemData; model.RecalculateModelParameters(problemData, rows); return model; } protected override double Evaluate(IDataAnalysisModel model) { var classificationModel = (IClassificationModel)model; var classificationProblemData = (IClassificationProblemData)ProblemData; var trainingIndices = ProblemData.TrainingIndices.ToList(); var estimatedValues = classificationModel.GetEstimatedClassValues(ProblemData.Dataset, trainingIndices); var targetValues = ProblemData.Dataset.GetDoubleValues(classificationProblemData.TargetVariable, trainingIndices); OnlineCalculatorError errorState; var quality = OnlinePearsonsRSquaredCalculator.Calculate(targetValues, estimatedValues, out errorState); if (errorState != OnlineCalculatorError.None) return double.NaN; return quality; } } }