source: branches/3136_Structural_GP/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.cs @ 18095

Last change on this file since 18095 was 18095, checked in by dpiringe, 6 months ago

#3136

  • added a Evaluate method, which uses the static method Calculate and evaluates a ISymbolicExpressionTree without the need of an ExecutionContext
    • implemented this new method in all single objective SymReg evaluators
File size: 5.2 KB
RevLine 
[5500]1#region License Information
2/* HeuristicLab
[17180]3 * Copyright (C) Heuristic and Evolutionary Algorithms Laboratory (HEAL)
[5500]4 *
5 * This file is part of HeuristicLab.
6 *
7 * HeuristicLab is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation, either version 3 of the License, or
10 * (at your option) any later version.
11 *
12 * HeuristicLab is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
15 * GNU General Public License for more details.
16 *
17 * You should have received a copy of the GNU General Public License
18 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
19 */
20#endregion
21
22using System.Collections.Generic;
23using HeuristicLab.Common;
24using HeuristicLab.Core;
25using HeuristicLab.Data;
26using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
[16565]27using HEAL.Attic;
[5500]28
[5501]29namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Regression {
[5618]30  [Item("Pearson R² Evaluator", "Calculates the square of the pearson correlation coefficient (also known as coefficient of determination) of a symbolic regression solution.")]
[16565]31  [StorableType("6FAEC6C2-C747-452A-A60D-29AE37898A90")]
[5500]32  public class SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator : SymbolicRegressionSingleObjectiveEvaluator {
33    [StorableConstructor]
[16565]34    protected SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator(StorableConstructorFlag _) : base(_) { }
[5500]35    protected SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator(SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator original, Cloner cloner)
36      : base(original, cloner) {
37    }
38    public override IDeepCloneable Clone(Cloner cloner) {
39      return new SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator(this, cloner);
40    }
41
[5505]42    public SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator() : base() { }
43
[5514]44    public override bool Maximization { get { return true; } }
45
[10291]46    public override IOperation InstrumentedApply() {
[5851]47      var solution = SymbolicExpressionTreeParameter.ActualValue;
[5500]48      IEnumerable<int> rows = GenerateRowsToEvaluate();
[5851]49
[12977]50      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows, ApplyLinearScalingParameter.ActualValue.Value);
[5851]51      QualityParameter.ActualValue = new DoubleValue(quality);
52
[10291]53      return base.InstrumentedApply();
[5500]54    }
55
[8664]56    public static double Calculate(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree solution, double lowerEstimationLimit, double upperEstimationLimit, IRegressionProblemData problemData, IEnumerable<int> rows, bool applyLinearScaling) {
[5500]57      IEnumerable<double> estimatedValues = interpreter.GetSymbolicExpressionTreeValues(solution, problemData.Dataset, rows);
[8664]58      IEnumerable<double> targetValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, rows);
[5942]59      OnlineCalculatorError errorState;
[8664]60
[12641]61      double r;
[8664]62      if (applyLinearScaling) {
[12641]63        var rCalculator = new OnlinePearsonsRCalculator();
64        CalculateWithScaling(targetValues, estimatedValues, lowerEstimationLimit, upperEstimationLimit, rCalculator, problemData.Dataset.Rows);
65        errorState = rCalculator.ErrorState;
66        r = rCalculator.R;
[8664]67      } else {
68        IEnumerable<double> boundedEstimatedValues = estimatedValues.LimitToRange(lowerEstimationLimit, upperEstimationLimit);
[12641]69        r = OnlinePearsonsRCalculator.Calculate(targetValues, boundedEstimatedValues, out errorState);
[8664]70      }
71      if (errorState != OnlineCalculatorError.None) return double.NaN;
[14354]72      return r*r;
[5500]73    }
[5613]74
75    public override double Evaluate(IExecutionContext context, ISymbolicExpressionTree tree, IRegressionProblemData problemData, IEnumerable<int> rows) {
[5722]76      SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = context;
[5770]77      EstimationLimitsParameter.ExecutionContext = context;
[8664]78      ApplyLinearScalingParameter.ExecutionContext = context;
[5722]79
[8664]80      double r2 = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScalingParameter.ActualValue.Value);
[5722]81
82      SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = null;
[5770]83      EstimationLimitsParameter.ExecutionContext = null;
[8664]84      ApplyLinearScalingParameter.ExecutionContext = null;
[5722]85
86      return r2;
[5613]87    }
[18095]88
89    public override double Evaluate(IRegressionProblemData problemData,
90      ISymbolicExpressionTree solution,
91      ISymbolicDataAnalysisExpressionTreeInterpreter interpreter,
92      IEnumerable<int> rows = null,
93      bool applyLinearScaling = true,
94      double lowerEstimationLimit = double.MinValue,
95      double upperEstimationLimit = double.MaxValue) {
96      return Calculate(interpreter, solution, lowerEstimationLimit, upperEstimationLimit, problemData, rows ?? problemData.TrainingIndices, applyLinearScaling);
97    }
[5500]98  }
99}
Note: See TracBrowser for help on using the repository browser.