| 1 | #region License Information
|
|---|
| 2 | /* HeuristicLab
|
|---|
| 3 | * Copyright (C) 2002-2010 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
|
|---|
| 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 |
|
|---|
| 22 | using System;
|
|---|
| 23 | using HeuristicLab.Analysis;
|
|---|
| 24 | using HeuristicLab.Common;
|
|---|
| 25 | using HeuristicLab.Core;
|
|---|
| 26 | using HeuristicLab.Data;
|
|---|
| 27 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
|
|---|
| 28 | using HeuristicLab.Operators;
|
|---|
| 29 | using HeuristicLab.Optimization;
|
|---|
| 30 | using HeuristicLab.Optimization.Operators;
|
|---|
| 31 | using HeuristicLab.Parameters;
|
|---|
| 32 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
|---|
| 33 | using HeuristicLab.Problems.DataAnalysis.Symbolic;
|
|---|
| 34 | using System.Collections.Generic;
|
|---|
| 35 | using HeuristicLab.Problems.DataAnalysis.Symbolic.Symbols;
|
|---|
| 36 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding.Symbols;
|
|---|
| 37 |
|
|---|
| 38 | namespace HeuristicLab.Problems.DataAnalysis.Regression.Symbolic.Analyzers {
|
|---|
| 39 | /// <summary>
|
|---|
| 40 | /// An operator that analyzes the population diversity using fine grained structural tree similarity estimation.
|
|---|
| 41 | /// </summary>
|
|---|
| 42 | [Item("FineGrainedStructuralPopulationDiversityAnalyzer", "An operator that analyzes the population diversity using fine grained structural tree similarity estimation.")]
|
|---|
| 43 | [StorableClass]
|
|---|
| 44 | public sealed class FineGrainedStructuralPopulationDiversityAnalyzer : SymbolicRegressionPopulationDiversityAnalyzer {
|
|---|
| 45 |
|
|---|
| 46 | #region Properties and Parameters
|
|---|
| 47 |
|
|---|
| 48 | private const string FunctionTreeGrammarParameterName = "FunctionTreeGrammar";
|
|---|
| 49 | private const string MinimumLevelDeltaParameterName = "MinimumLevelDelta";
|
|---|
| 50 | private const string MaximumLevelDeltaParameterName = "MaximumLevelDelta";
|
|---|
| 51 | private const string PreventMultipleComparisonContributionParameterName = "PreventMultipleComparisonContribution";
|
|---|
| 52 | private const string MaximumExpressionDepthParameterName = "MaxExpressionDepth";
|
|---|
| 53 | private const string LevelDifferenceCoefficientParameterName = "LevelDifferenceCoefficient";
|
|---|
| 54 | private const string AncestorIndexCoefficientParameterName = "AncestorIndexCoefficient";
|
|---|
| 55 | private const string ConstantValueCoefficientParameterName = "ConstantValueCoefficient";
|
|---|
| 56 | private const string VariableWeightCoefficientParameterName = "VariableWeightCoefficient";
|
|---|
| 57 | private const string TimeOffsetCoefficientParameterName = "TimeOffsetCoefficient";
|
|---|
| 58 | private const string VariableIndexCoefficientParameterName = "VariableIndexCoefficient";
|
|---|
| 59 | private const string AdditiveSimilarityCalculationParameterName = "AdditiveSimilarityCalculation";
|
|---|
| 60 |
|
|---|
| 61 | public IValueLookupParameter<GlobalSymbolicExpressionGrammar> FunctionTreeGrammarParameter {
|
|---|
| 62 | get { return (IValueLookupParameter<GlobalSymbolicExpressionGrammar>)Parameters[FunctionTreeGrammarParameterName]; }
|
|---|
| 63 | }
|
|---|
| 64 | public GlobalSymbolicExpressionGrammar FunctionTreeGrammar {
|
|---|
| 65 | get { return FunctionTreeGrammarParameter.ActualValue; }
|
|---|
| 66 | }
|
|---|
| 67 | public IValueLookupParameter<IntValue> MaximumExpressionDepthParameter {
|
|---|
| 68 | get { return (IValueLookupParameter<IntValue>)Parameters[MaximumExpressionDepthParameterName]; }
|
|---|
| 69 | }
|
|---|
| 70 | public int MaximumExpressionDepth {
|
|---|
| 71 | get { return MaximumExpressionDepthParameter.ActualValue.Value; }
|
|---|
| 72 | }
|
|---|
| 73 |
|
|---|
| 74 | public IValueParameter<IntValue> MinimumLevelDeltaParameter {
|
|---|
| 75 | get { return (IValueParameter<IntValue>)Parameters[MinimumLevelDeltaParameterName]; }
|
|---|
| 76 | }
|
|---|
| 77 | public int MinimumLevelDelta {
|
|---|
| 78 | get { return MinimumLevelDeltaParameter.Value.Value; }
|
|---|
| 79 | }
|
|---|
| 80 | public IValueParameter<IntValue> MaximumLevelDeltaParameter {
|
|---|
| 81 | get { return (IValueParameter<IntValue>)Parameters[MaximumLevelDeltaParameterName]; }
|
|---|
| 82 | }
|
|---|
| 83 | public int MaximumLevelDelta {
|
|---|
| 84 | get { return MaximumLevelDeltaParameter.Value.Value; }
|
|---|
| 85 | }
|
|---|
| 86 | public IValueParameter<BoolValue> PreventMultipleComparisonContributionParameter {
|
|---|
| 87 | get { return (IValueParameter<BoolValue>)Parameters[PreventMultipleComparisonContributionParameterName]; }
|
|---|
| 88 | }
|
|---|
| 89 | public bool PreventMultipleComparisonContribution {
|
|---|
| 90 | get { return PreventMultipleComparisonContributionParameter.Value.Value; }
|
|---|
| 91 | }
|
|---|
| 92 |
|
|---|
| 93 | public IValueParameter<DoubleValue> LevelDifferenceCoefficientParameter {
|
|---|
| 94 | get { return (IValueParameter<DoubleValue>)Parameters[LevelDifferenceCoefficientParameterName]; }
|
|---|
| 95 | }
|
|---|
| 96 | public double LevelDifferenceCoefficient {
|
|---|
| 97 | get { return LevelDifferenceCoefficientParameter.Value.Value; }
|
|---|
| 98 | }
|
|---|
| 99 | public IValueParameter<DoubleValue> AncestorIndexCoefficientParameter {
|
|---|
| 100 | get { return (IValueParameter<DoubleValue>)Parameters[AncestorIndexCoefficientParameterName]; }
|
|---|
| 101 | }
|
|---|
| 102 | public double AncestorIndexCoefficient {
|
|---|
| 103 | get { return AncestorIndexCoefficientParameter.Value.Value; }
|
|---|
| 104 | }
|
|---|
| 105 | public IValueParameter<DoubleValue> ConstantValueCoefficientParameter {
|
|---|
| 106 | get { return (IValueParameter<DoubleValue>)Parameters[ConstantValueCoefficientParameterName]; }
|
|---|
| 107 | }
|
|---|
| 108 | public double ConstantValueCoefficient {
|
|---|
| 109 | get { return ConstantValueCoefficientParameter.Value.Value; }
|
|---|
| 110 | }
|
|---|
| 111 | public IValueParameter<DoubleValue> VariableWeightCoefficientParameter {
|
|---|
| 112 | get { return (IValueParameter<DoubleValue>)Parameters[VariableWeightCoefficientParameterName]; }
|
|---|
| 113 | }
|
|---|
| 114 | public double VariableWeightCoefficient {
|
|---|
| 115 | get { return VariableWeightCoefficientParameter.Value.Value; }
|
|---|
| 116 | }
|
|---|
| 117 | public IValueParameter<DoubleValue> TimeOffsetCoefficientParameter {
|
|---|
| 118 | get { return (IValueParameter<DoubleValue>)Parameters[TimeOffsetCoefficientParameterName]; }
|
|---|
| 119 | }
|
|---|
| 120 | public double TimeOffsetCoefficientCoefficient {
|
|---|
| 121 | get { return TimeOffsetCoefficientParameter.Value.Value; }
|
|---|
| 122 | }
|
|---|
| 123 | public IValueParameter<DoubleValue> VariableIndexCoefficientParameter {
|
|---|
| 124 | get { return (IValueParameter<DoubleValue>)Parameters[VariableIndexCoefficientParameterName]; }
|
|---|
| 125 | }
|
|---|
| 126 | public double VariableIndexCoefficient {
|
|---|
| 127 | get { return VariableIndexCoefficientParameter.Value.Value; }
|
|---|
| 128 | }
|
|---|
| 129 | public IValueParameter<BoolValue> AdditiveSimilarityCalculationParameter {
|
|---|
| 130 | get { return (IValueParameter<BoolValue>)Parameters[AdditiveSimilarityCalculationParameterName]; }
|
|---|
| 131 | }
|
|---|
| 132 | public bool AdditiveSimilarityCalculation {
|
|---|
| 133 | get { return AdditiveSimilarityCalculationParameter.Value.Value; }
|
|---|
| 134 | }
|
|---|
| 135 |
|
|---|
| 136 | #endregion
|
|---|
| 137 |
|
|---|
| 138 | [StorableConstructor]
|
|---|
| 139 | private FineGrainedStructuralPopulationDiversityAnalyzer(bool deserializing) : base(deserializing) { }
|
|---|
| 140 | private FineGrainedStructuralPopulationDiversityAnalyzer(FineGrainedStructuralPopulationDiversityAnalyzer original, Cloner cloner) : base(original, cloner) { }
|
|---|
| 141 | public FineGrainedStructuralPopulationDiversityAnalyzer() : base() {
|
|---|
| 142 | Parameters.Add(new ValueLookupParameter<GlobalSymbolicExpressionGrammar>(FunctionTreeGrammarParameterName, "The grammar that is used for symbolic regression models."));
|
|---|
| 143 | Parameters.Add(new ValueLookupParameter<IntValue>(MaximumExpressionDepthParameterName, "Maximal depth of the analyzed symbolic expressions."));
|
|---|
| 144 | Parameters.Add(new ValueParameter<IntValue>(MinimumLevelDeltaParameterName, "Minimum value for the level delta of the analyzed genetic information items.", new IntValue(0)));
|
|---|
| 145 | Parameters.Add(new ValueParameter<IntValue>(MaximumLevelDeltaParameterName, "Maximum value for the level delta of the analyzed genetic information items.", new IntValue(int.MaxValue)));
|
|---|
| 146 | Parameters.Add(new ValueParameter<BoolValue>(PreventMultipleComparisonContributionParameterName, "Flag that denotes whether genetic information items are hindered from contributing to the similarity function multiple times.", new BoolValue(false)));
|
|---|
| 147 | Parameters.Add(new ValueParameter<DoubleValue>(LevelDifferenceCoefficientParameterName, "Weighting coefficient for level differences.", new DoubleValue(0.2)));
|
|---|
| 148 | Parameters.Add(new ValueParameter<DoubleValue>(AncestorIndexCoefficientParameterName, "Weighting coefficient for ancestor index differences.", new DoubleValue(0.2)));
|
|---|
| 149 | Parameters.Add(new ValueParameter<DoubleValue>(ConstantValueCoefficientParameterName, "Weighting coefficient for constant value differences.", new DoubleValue(0.2)));
|
|---|
| 150 | Parameters.Add(new ValueParameter<DoubleValue>(VariableWeightCoefficientParameterName, "Weighting coefficient for variable weight differences.", new DoubleValue(0.2)));
|
|---|
| 151 | Parameters.Add(new ValueParameter<DoubleValue>(TimeOffsetCoefficientParameterName, "Weighting coefficient for time lag differences.", new DoubleValue(0.2)));
|
|---|
| 152 | Parameters.Add(new ValueParameter<DoubleValue>(VariableIndexCoefficientParameterName, "Weighting coefficient for variable index differences.", new DoubleValue(0.2)));
|
|---|
| 153 | Parameters.Add(new ValueParameter<BoolValue>(AdditiveSimilarityCalculationParameterName, "Flag that denotes whether the similarity of genetic information items shall be calculated using additive calculation.", new BoolValue(true)));
|
|---|
| 154 | }
|
|---|
| 155 |
|
|---|
| 156 | public override IDeepCloneable Clone(Cloner cloner) {
|
|---|
| 157 | return new FineGrainedStructuralPopulationDiversityAnalyzer(this, cloner);
|
|---|
| 158 | }
|
|---|
| 159 |
|
|---|
| 160 | protected override double[,] CalculateSimilarities(SymbolicExpressionTree[] solutions) {
|
|---|
| 161 | // collect information stored int the problem's parameters
|
|---|
| 162 | double variableWeightSigma = 0;
|
|---|
| 163 | double constantMinimumValue = 0;
|
|---|
| 164 | double constantMaximumValue = 0;
|
|---|
| 165 | int minimumTimeOffset = 0;
|
|---|
| 166 | int maximumTimeOffset = 0;
|
|---|
| 167 | foreach (Symbol symbol in FunctionTreeGrammar.Symbols) {
|
|---|
| 168 | Constant constant = symbol as Constant;
|
|---|
| 169 | if (constant !=null) {
|
|---|
| 170 | constantMinimumValue = constant.MinValue;
|
|---|
| 171 | constantMaximumValue = constant.MaxValue;
|
|---|
| 172 | }
|
|---|
| 173 | DataAnalysis.Symbolic.Symbols.Variable variable = symbol as DataAnalysis.Symbolic.Symbols.Variable;
|
|---|
| 174 | if (variable != null)
|
|---|
| 175 | variableWeightSigma = variable.WeightSigma;
|
|---|
| 176 | LaggedVariable laggedVariable = symbol as LaggedVariable;
|
|---|
| 177 | if (laggedVariable !=null) {
|
|---|
| 178 | minimumTimeOffset = laggedVariable.MinLag;
|
|---|
| 179 | maximumTimeOffset = laggedVariable.MaxLag;
|
|---|
| 180 | }
|
|---|
| 181 | }
|
|---|
| 182 | int n = solutions.Length;
|
|---|
| 183 | List<string> variableNames = new List<string>();
|
|---|
| 184 | foreach (StringValue variableName in ProblemData.InputVariables) {
|
|---|
| 185 | variableNames.Add(variableName.Value);
|
|---|
| 186 | }
|
|---|
| 187 | variableNames.Add(ProblemData.TargetVariable.Value);
|
|---|
| 188 | // collect genetic information item lists and store them also in dictionaries
|
|---|
| 189 | IList<GeneticInformationItem>[] geneticInformationItemsLists = new List<GeneticInformationItem>[n];
|
|---|
| 190 | IDictionary<string, IList<GeneticInformationItem>>[] geneticInformationItemsListsDictionaries = new IDictionary<string, IList<GeneticInformationItem>>[n];
|
|---|
| 191 | for (int i = 0; i < n; i++) {
|
|---|
| 192 | geneticInformationItemsLists[i] = GeneticInformationItem.GetGeneticInformationItems(solutions[i].Root, variableNames, MinimumLevelDelta, MaximumLevelDelta);
|
|---|
| 193 | geneticInformationItemsListsDictionaries[i] = GeneticInformationItem.GetDictionary(geneticInformationItemsLists[i]);
|
|---|
| 194 | }
|
|---|
| 195 | // calculate solution similarities
|
|---|
| 196 | double[,] similarities = new double[n, n];
|
|---|
| 197 | for (int i = 0; i < n; i++) {
|
|---|
| 198 | for (int j = 0; j < n; j++) {
|
|---|
| 199 | if (i == j)
|
|---|
| 200 | similarities[i, j] = 1;
|
|---|
| 201 | else {
|
|---|
| 202 | IList<GeneticInformationItem> solution1GeneticItems = geneticInformationItemsLists[i];
|
|---|
| 203 | IDictionary<string, IList<GeneticInformationItem>> solution2GeneticItemsDictionary = GeneticInformationItem.CopyDictionary(geneticInformationItemsListsDictionaries[j]);
|
|---|
| 204 | double similarity = 0;
|
|---|
| 205 | for (int k = 0; k < solution1GeneticItems.Count; k++) {
|
|---|
| 206 | double bestPendantSimilarity = 0;
|
|---|
| 207 | GeneticInformationItem item = solution1GeneticItems[k];
|
|---|
| 208 | GeneticInformationItem bestPendant = null;
|
|---|
| 209 | IList<GeneticInformationItem> geneticInformationItemsList = null;
|
|---|
| 210 | string key = GeneticInformationItem.GetKey(item);
|
|---|
| 211 | if (solution2GeneticItemsDictionary.ContainsKey(key)) {
|
|---|
| 212 | geneticInformationItemsList = solution2GeneticItemsDictionary[GeneticInformationItem.GetKey(item)];
|
|---|
| 213 | bestPendant = GeneticInformationItem.FindBestPendant(item, geneticInformationItemsList,
|
|---|
| 214 | constantMinimumValue, constantMaximumValue, variableWeightSigma,
|
|---|
| 215 | MaximumExpressionDepth, minimumTimeOffset, maximumTimeOffset,
|
|---|
| 216 | LevelDifferenceCoefficient, AncestorIndexCoefficient, ConstantValueCoefficient, VariableWeightCoefficient,
|
|---|
| 217 | TimeOffsetCoefficientCoefficient, VariableIndexCoefficient, AdditiveSimilarityCalculation,
|
|---|
| 218 | out bestPendantSimilarity);
|
|---|
| 219 | }
|
|---|
| 220 | if (bestPendant != null) {
|
|---|
| 221 | similarity += bestPendantSimilarity;
|
|---|
| 222 | if (PreventMultipleComparisonContribution)
|
|---|
| 223 | geneticInformationItemsList.Remove(bestPendant);
|
|---|
| 224 | }
|
|---|
| 225 | }
|
|---|
| 226 | similarities[i, j] = similarity / solution1GeneticItems.Count;
|
|---|
| 227 | }
|
|---|
| 228 | }
|
|---|
| 229 | }
|
|---|
| 230 | return similarities;
|
|---|
| 231 | }
|
|---|
| 232 |
|
|---|
| 233 | }
|
|---|
| 234 |
|
|---|
| 235 | }
|
|---|
