[12951] | 1 | #region License Information
|
---|
| 2 | /* HeuristicLab
|
---|
| 3 | * Copyright (C) 2002-2015 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;
|
---|
[13480] | 23 | using System.Collections.Generic;
|
---|
[12951] | 24 | using System.Linq;
|
---|
[12979] | 25 | using System.Threading.Tasks;
|
---|
[12951] | 26 | using HeuristicLab.Common;
|
---|
| 27 | using HeuristicLab.Core;
|
---|
| 28 | using HeuristicLab.Data;
|
---|
| 29 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
|
---|
| 30 | using HeuristicLab.EvolutionTracking;
|
---|
[13480] | 31 | using HeuristicLab.Optimization;
|
---|
[12951] | 32 | using HeuristicLab.Parameters;
|
---|
| 33 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
---|
| 34 | using HeuristicLab.Random;
|
---|
| 35 |
|
---|
[12958] | 36 | namespace HeuristicLab.Problems.DataAnalysis.Symbolic {
|
---|
[12951] | 37 | [Item("SchemaEvaluator", "An operator that builds schemas based on the heredity relationship in the genealogy graph.")]
|
---|
| 38 | [StorableClass]
|
---|
| 39 | public class SchemaEvaluator : EvolutionTrackingOperator<ISymbolicExpressionTree> {
|
---|
[12952] | 40 | #region parameter names
|
---|
[12951] | 41 | private const string MinimumSchemaFrequencyParameterName = "MinimumSchemaFrequency";
|
---|
| 42 | private const string MinimumPhenotypicSimilarityParameterName = "MinimumPhenotypicSimilarity";
|
---|
| 43 | private const string ReplacementRatioParameterName = "ReplacementRatio";
|
---|
| 44 | private const string SchemaParameterName = "Schema";
|
---|
| 45 | private const string PopulationSizeParameterName = "PopulationSize";
|
---|
| 46 | private const string RandomParameterName = "Random";
|
---|
| 47 | private const string EvaluatorParameterName = "Evaluator";
|
---|
| 48 | private const string ProblemDataParameterName = "ProblemData";
|
---|
| 49 | private const string InterpreterParameterName = "SymbolicExpressionTreeInterpreter";
|
---|
| 50 | private const string EstimationLimitsParameterName = "EstimationLimits";
|
---|
| 51 | private const string ApplyLinearScalingParameterName = "ApplyLinearScaling";
|
---|
| 52 | private const string MutatorParameterName = "Mutator";
|
---|
[13480] | 53 | private const string CrossoverParameterName = "Crossover";
|
---|
[12951] | 54 | private const string RandomReplacementParameterName = "RandomReplacement";
|
---|
[12988] | 55 | private const string NumberOfChangedTreesParameterName = "NumberOfChangedTrees";
|
---|
[12979] | 56 | private const string ExecuteInParallelParameterName = "ExecuteInParallel";
|
---|
| 57 | private const string MaxDegreeOfParalellismParameterName = "MaxDegreeOfParallelism";
|
---|
[12988] | 58 | private const string ExclusiveMatchingParameterName = "ExclusiveMatching";
|
---|
[13480] | 59 | private const string UseAdaptiveReplacementRatioParameterName = "UseAdaptiveReplacementRatio";
|
---|
[13496] | 60 | private const string StrictSchemaMatchingParameterName = "StrictSchemaMatching";
|
---|
[12952] | 61 | #endregion
|
---|
[12951] | 62 |
|
---|
| 63 | #region parameters
|
---|
[13480] | 64 | public ILookupParameter<BoolValue> UseAdaptiveReplacementRatioParameter {
|
---|
| 65 | get { return (ILookupParameter<BoolValue>)Parameters[UseAdaptiveReplacementRatioParameterName]; }
|
---|
| 66 | }
|
---|
[12988] | 67 | public ILookupParameter<BoolValue> ExclusiveMatchingParameter {
|
---|
| 68 | get { return (ILookupParameter<BoolValue>)Parameters[ExclusiveMatchingParameterName]; }
|
---|
| 69 | }
|
---|
[12979] | 70 | public ILookupParameter<BoolValue> ExecuteInParallelParameter {
|
---|
| 71 | get { return (ILookupParameter<BoolValue>)Parameters[ExecuteInParallelParameterName]; }
|
---|
| 72 | }
|
---|
| 73 | public ILookupParameter<IntValue> MaxDegreeOfParallelismParameter {
|
---|
| 74 | get { return (ILookupParameter<IntValue>)Parameters[MaxDegreeOfParalellismParameterName]; }
|
---|
| 75 | }
|
---|
[12951] | 76 | public ILookupParameter<ISymbolicDataAnalysisSingleObjectiveEvaluator<IRegressionProblemData>> EvaluatorParameter {
|
---|
| 77 | get { return (ILookupParameter<ISymbolicDataAnalysisSingleObjectiveEvaluator<IRegressionProblemData>>)Parameters[EvaluatorParameterName]; }
|
---|
| 78 | }
|
---|
| 79 | public ILookupParameter<IRegressionProblemData> ProblemDataParameter {
|
---|
| 80 | get { return (ILookupParameter<IRegressionProblemData>)Parameters[ProblemDataParameterName]; }
|
---|
| 81 | }
|
---|
| 82 | public ILookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter> InterpreterParameter {
|
---|
| 83 | get { return (ILookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter>)Parameters[InterpreterParameterName]; }
|
---|
| 84 | }
|
---|
| 85 | public ILookupParameter<DoubleLimit> EstimationLimitsParameter {
|
---|
| 86 | get { return (ILookupParameter<DoubleLimit>)Parameters[EstimationLimitsParameterName]; }
|
---|
| 87 | }
|
---|
| 88 | public ILookupParameter<BoolValue> ApplyLinearScalingParameter {
|
---|
| 89 | get { return (ILookupParameter<BoolValue>)Parameters[ApplyLinearScalingParameterName]; }
|
---|
| 90 | }
|
---|
| 91 | public ILookupParameter<BoolValue> RandomReplacementParameter {
|
---|
| 92 | get { return (ILookupParameter<BoolValue>)Parameters[RandomReplacementParameterName]; }
|
---|
| 93 | }
|
---|
[13480] | 94 | public ILookupParameter<ISymbolicExpressionTreeCrossover> CrossoverParameter {
|
---|
| 95 | get { return (ILookupParameter<ISymbolicExpressionTreeCrossover>)Parameters[CrossoverParameterName]; }
|
---|
| 96 | }
|
---|
[12951] | 97 | public ILookupParameter<IRandom> RandomParameter {
|
---|
| 98 | get { return (ILookupParameter<IRandom>)Parameters[RandomParameterName]; }
|
---|
| 99 | }
|
---|
| 100 | public ILookupParameter<IntValue> PopulationSizeParameter {
|
---|
| 101 | get { return (ILookupParameter<IntValue>)Parameters[PopulationSizeParameterName]; }
|
---|
| 102 | }
|
---|
| 103 | public ILookupParameter<ISymbolicExpressionTree> SchemaParameter {
|
---|
| 104 | get { return (ILookupParameter<ISymbolicExpressionTree>)Parameters[SchemaParameterName]; }
|
---|
| 105 | }
|
---|
| 106 | public ILookupParameter<PercentValue> MinimumSchemaFrequencyParameter {
|
---|
| 107 | get { return (ILookupParameter<PercentValue>)Parameters[MinimumSchemaFrequencyParameterName]; }
|
---|
| 108 | }
|
---|
| 109 | public ILookupParameter<PercentValue> ReplacementRatioParameter {
|
---|
| 110 | get { return (ILookupParameter<PercentValue>)Parameters[ReplacementRatioParameterName]; }
|
---|
| 111 | }
|
---|
| 112 | public ILookupParameter<PercentValue> MinimumPhenotypicSimilarityParameter {
|
---|
| 113 | get { return (ILookupParameter<PercentValue>)Parameters[MinimumPhenotypicSimilarityParameterName]; }
|
---|
| 114 | }
|
---|
[12988] | 115 | public LookupParameter<IntValue> NumberOfChangedTreesParameter {
|
---|
| 116 | get { return (LookupParameter<IntValue>)Parameters[NumberOfChangedTreesParameterName]; }
|
---|
[12952] | 117 | }
|
---|
[13496] | 118 | public LookupParameter<BoolValue> StrictSchemaMatchingParameter {
|
---|
| 119 | get { return (LookupParameter<BoolValue>)Parameters[StrictSchemaMatchingParameterName]; }
|
---|
| 120 | }
|
---|
[12951] | 121 | #endregion
|
---|
| 122 |
|
---|
| 123 | #region parameter properties
|
---|
[12979] | 124 | public PercentValue MinimumSchemaFrequency { get { return MinimumSchemaFrequencyParameter.ActualValue; } }
|
---|
| 125 | public PercentValue ReplacementRatio { get { return ReplacementRatioParameter.ActualValue; } }
|
---|
| 126 | public PercentValue MinimumPhenotypicSimilarity { get { return MinimumPhenotypicSimilarityParameter.ActualValue; } }
|
---|
| 127 | public BoolValue RandomReplacement { get { return RandomReplacementParameter.ActualValue; } }
|
---|
[12988] | 128 | public IntValue NumberOfChangedTrees { get { return NumberOfChangedTreesParameter.ActualValue; } }
|
---|
[12951] | 129 | #endregion
|
---|
| 130 |
|
---|
[13565] | 131 | private QueryMatch qm;
|
---|
[12952] | 132 |
|
---|
[13565] | 133 | [Storable]
|
---|
| 134 | private SymbolicExpressionTreePhenotypicSimilarityCalculator calculator;
|
---|
[13480] | 135 |
|
---|
[13565] | 136 | [Storable]
|
---|
| 137 | public string ReplacementRule { get; set; }
|
---|
[12952] | 138 |
|
---|
[13527] | 139 | [Storable]
|
---|
[13565] | 140 | private ISymbolicExpressionTreeNodeEqualityComparer comparer;
|
---|
| 141 |
|
---|
| 142 | [Storable]
|
---|
[12988] | 143 | public ISymbolicExpressionTree Schema { get; set; }
|
---|
| 144 |
|
---|
[12979] | 145 | [Storable]
|
---|
[13527] | 146 | private readonly UpdateQualityOperator updateQualityOperator;
|
---|
[12952] | 147 |
|
---|
[15482] | 148 | [Storable]
|
---|
| 149 | public ISymbolicExpressionTreeManipulator SchemaManipulator { get; set; }
|
---|
| 150 |
|
---|
[13496] | 151 | [StorableHook(HookType.AfterDeserialization)]
|
---|
| 152 | private void AfterDeserialization() {
|
---|
| 153 | if (!Parameters.ContainsKey(StrictSchemaMatchingParameterName))
|
---|
| 154 | Parameters.Add(new LookupParameter<BoolValue>(StrictSchemaMatchingParameterName));
|
---|
[13565] | 155 |
|
---|
| 156 | if (calculator == null)
|
---|
| 157 | calculator = new SymbolicExpressionTreePhenotypicSimilarityCalculator();
|
---|
| 158 |
|
---|
| 159 | if (comparer == null)
|
---|
| 160 | comparer = new SymbolicExpressionTreeNodeEqualityComparer { MatchVariableNames = true, MatchVariableWeights = true, MatchConstantValues = false };
|
---|
| 161 |
|
---|
| 162 | qm = new QueryMatch(comparer) { MatchParents = true };
|
---|
[13496] | 163 | }
|
---|
| 164 |
|
---|
[12951] | 165 | public SchemaEvaluator() {
|
---|
[13565] | 166 | calculator = new SymbolicExpressionTreePhenotypicSimilarityCalculator();
|
---|
| 167 | comparer = new SymbolicExpressionTreeNodeEqualityComparer { MatchVariableNames = true, MatchVariableWeights = true, MatchConstantValues = false };
|
---|
| 168 | qm = new QueryMatch(comparer) { MatchParents = true };
|
---|
| 169 | updateQualityOperator = new UpdateQualityOperator();
|
---|
[12988] | 170 | #region add parameters
|
---|
[12951] | 171 | Parameters.Add(new LookupParameter<ISymbolicExpressionTree>(SchemaParameterName, "The current schema to be evaluated"));
|
---|
| 172 | Parameters.Add(new LookupParameter<PercentValue>(MinimumSchemaFrequencyParameterName));
|
---|
| 173 | Parameters.Add(new LookupParameter<PercentValue>(ReplacementRatioParameterName));
|
---|
| 174 | Parameters.Add(new LookupParameter<PercentValue>(MinimumPhenotypicSimilarityParameterName));
|
---|
| 175 | Parameters.Add(new LookupParameter<ISymbolicExpressionTree>(PopulationSizeParameterName));
|
---|
| 176 | Parameters.Add(new LookupParameter<IRandom>(RandomParameterName));
|
---|
| 177 | Parameters.Add(new LookupParameter<ISymbolicDataAnalysisSingleObjectiveEvaluator<IRegressionProblemData>>(EvaluatorParameterName));
|
---|
| 178 | Parameters.Add(new LookupParameter<IRegressionProblemData>(ProblemDataParameterName));
|
---|
| 179 | Parameters.Add(new LookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter>(InterpreterParameterName));
|
---|
| 180 | Parameters.Add(new LookupParameter<DoubleLimit>(EstimationLimitsParameterName));
|
---|
| 181 | Parameters.Add(new LookupParameter<BoolValue>(ApplyLinearScalingParameterName));
|
---|
[13496] | 182 | Parameters.Add(new LookupParameter<BoolValue>(StrictSchemaMatchingParameterName));
|
---|
[12951] | 183 | Parameters.Add(new LookupParameter<ISymbolicExpressionTreeManipulator>(MutatorParameterName));
|
---|
[13480] | 184 | Parameters.Add(new LookupParameter<ISymbolicExpressionTreeCrossover>(CrossoverParameterName));
|
---|
[12951] | 185 | Parameters.Add(new LookupParameter<BoolValue>(RandomReplacementParameterName));
|
---|
[12988] | 186 | Parameters.Add(new LookupParameter<IntValue>(NumberOfChangedTreesParameterName));
|
---|
[12979] | 187 | Parameters.Add(new LookupParameter<BoolValue>(ExecuteInParallelParameterName));
|
---|
| 188 | Parameters.Add(new LookupParameter<IntValue>(MaxDegreeOfParalellismParameterName));
|
---|
[12988] | 189 | Parameters.Add(new LookupParameter<BoolValue>(ExclusiveMatchingParameterName));
|
---|
[13480] | 190 | Parameters.Add(new LookupParameter<BoolValue>(UseAdaptiveReplacementRatioParameterName));
|
---|
[12988] | 191 | #endregion
|
---|
[12951] | 192 | }
|
---|
| 193 |
|
---|
[12966] | 194 | [StorableConstructor]
|
---|
| 195 | protected SchemaEvaluator(bool deserializing) : base(deserializing) { }
|
---|
| 196 |
|
---|
[12951] | 197 | protected SchemaEvaluator(SchemaEvaluator original, Cloner cloner) : base(original, cloner) {
|
---|
[13565] | 198 | calculator = cloner.Clone(original.calculator);
|
---|
| 199 | comparer = cloner.Clone(original.comparer);
|
---|
| 200 | qm = new QueryMatch(comparer) { MatchParents = true };
|
---|
| 201 | updateQualityOperator = new UpdateQualityOperator();
|
---|
[13527] | 202 | Schema = original.Schema;
|
---|
[12951] | 203 | }
|
---|
| 204 |
|
---|
| 205 | public override IDeepCloneable Clone(Cloner cloner) {
|
---|
| 206 | return new SchemaEvaluator(this, cloner);
|
---|
| 207 | }
|
---|
| 208 |
|
---|
| 209 | public override IOperation Apply() {
|
---|
[13496] | 210 | var strictSchemaMatching = StrictSchemaMatchingParameter.ActualValue.Value;
|
---|
| 211 | if (strictSchemaMatching) {
|
---|
[13565] | 212 | comparer.MatchVariableWeights = true;
|
---|
| 213 | comparer.MatchConstantValues = true;
|
---|
[13496] | 214 | } else {
|
---|
[13565] | 215 | comparer.MatchVariableWeights = false;
|
---|
| 216 | comparer.MatchConstantValues = false;
|
---|
[13496] | 217 | }
|
---|
| 218 |
|
---|
[12951] | 219 | var individuals = ExecutionContext.Scope.SubScopes; // the scopes represent the individuals
|
---|
[13480] | 220 | var trees = individuals.Select(x => (ISymbolicExpressionTree)x.Variables["SymbolicExpressionTree"].Value).ToList();
|
---|
[12951] | 221 |
|
---|
| 222 | var random = RandomParameter.ActualValue;
|
---|
| 223 |
|
---|
[12988] | 224 | var s = Schema;
|
---|
[12979] | 225 | var sRoot = s.Root.GetSubtree(0).GetSubtree(0);
|
---|
| 226 | int countThreshold = (int)Math.Max(2, Math.Round(MinimumSchemaFrequency.Value * individuals.Count));
|
---|
| 227 |
|
---|
| 228 | // first apply the length and root equality checks in order to filter the individuals
|
---|
[12988] | 229 | var exclusiveMatching = ExclusiveMatchingParameter.ActualValue.Value;
|
---|
[13480] | 230 | var filtered = new List<int>();
|
---|
| 231 | for (int i = 0; i < individuals.Count; ++i) {
|
---|
| 232 | if (exclusiveMatching && individuals[i].Variables.ContainsKey("AlreadyMatched")) continue;
|
---|
| 233 | var t = trees[i];
|
---|
| 234 | var tRoot = t.Root.GetSubtree(0).GetSubtree(0);
|
---|
[14427] | 235 | if (t.Length < s.Length || !qm.EqualityComparer.Equals(tRoot, sRoot)) continue;
|
---|
[13480] | 236 | filtered.Add(i);
|
---|
| 237 | }
|
---|
[12979] | 238 |
|
---|
| 239 | // if we don't have enough filtered individuals, then we are done
|
---|
[13480] | 240 | // if the schema exceeds the minimum frequency, it gets processed further
|
---|
[12979] | 241 | if (filtered.Count < countThreshold) {
|
---|
| 242 | return base.Apply();
|
---|
| 243 | }
|
---|
| 244 |
|
---|
| 245 | // check if the filtered individuals match the schema
|
---|
| 246 | var sNodes = QueryMatch.InitializePostOrder(sRoot);
|
---|
[12958] | 247 | var matchingIndividuals = new ScopeList();
|
---|
[12979] | 248 | bool executeInParallel = ExecuteInParallelParameter.ActualValue.Value;
|
---|
| 249 | int maxDegreeOfParallelism = MaxDegreeOfParallelismParameter.ActualValue.Value;
|
---|
| 250 |
|
---|
| 251 | if (executeInParallel) {
|
---|
| 252 | var matching = new bool[filtered.Count];
|
---|
| 253 |
|
---|
| 254 | Parallel.For(0, filtered.Count, new ParallelOptions { MaxDegreeOfParallelism = maxDegreeOfParallelism },
|
---|
| 255 | i => {
|
---|
[13480] | 256 | var index = filtered[i];
|
---|
| 257 | var t = trees[index];
|
---|
[12979] | 258 | var tNodes = QueryMatch.InitializePostOrder(t.Root.GetSubtree(0).GetSubtree(0));
|
---|
| 259 | if (qm.Match(tNodes, sNodes)) {
|
---|
| 260 | matching[i] = true;
|
---|
| 261 | }
|
---|
| 262 | });
|
---|
| 263 |
|
---|
[13480] | 264 | matchingIndividuals.AddRange(filtered.Where((x, i) => matching[i]).Select(x => individuals[x]));
|
---|
[12979] | 265 | } else {
|
---|
| 266 | for (int i = 0; i < filtered.Count; ++i) {
|
---|
| 267 | // break early if it becomes impossible to reach the minimum threshold
|
---|
| 268 | if (matchingIndividuals.Count + filtered.Count - i < countThreshold)
|
---|
| 269 | break;
|
---|
| 270 |
|
---|
[13480] | 271 | var index = filtered[i];
|
---|
| 272 | var tRoot = trees[index].Root.GetSubtree(0).GetSubtree(0);
|
---|
| 273 | var tNodes = QueryMatch.InitializePostOrder(tRoot);
|
---|
[12979] | 274 | if (qm.Match(tNodes, sNodes))
|
---|
[13480] | 275 | matchingIndividuals.Add(individuals[index]);
|
---|
[12979] | 276 | }
|
---|
[12958] | 277 | }
|
---|
[12951] | 278 |
|
---|
[12979] | 279 | if (matchingIndividuals.Count < countThreshold) {
|
---|
[12951] | 280 | return base.Apply();
|
---|
[12952] | 281 | }
|
---|
[12951] | 282 |
|
---|
[13480] | 283 | var similarity = CalculateSimilarity(matchingIndividuals, calculator, executeInParallel, maxDegreeOfParallelism);
|
---|
[12952] | 284 | if (similarity < MinimumPhenotypicSimilarity.Value) {
|
---|
[12951] | 285 | return base.Apply();
|
---|
[12952] | 286 | }
|
---|
[12951] | 287 |
|
---|
[13480] | 288 | double replacementRatio;
|
---|
| 289 | var adaptiveReplacementRatio = UseAdaptiveReplacementRatioParameter.ActualValue.Value;
|
---|
| 290 |
|
---|
| 291 | if (adaptiveReplacementRatio) {
|
---|
| 292 | var r = (double)matchingIndividuals.Count / individuals.Count;
|
---|
[13565] | 293 | replacementRatio = CalculateReplacementRatio(r);
|
---|
[13480] | 294 | } else {
|
---|
| 295 | replacementRatio = ReplacementRatio.Value;
|
---|
| 296 | }
|
---|
| 297 |
|
---|
| 298 | int n = (int)Math.Floor(matchingIndividuals.Count * replacementRatio);
|
---|
[12952] | 299 | var individualsToReplace = RandomReplacement.Value ? matchingIndividuals.SampleRandomWithoutRepetition(random, n).ToList()
|
---|
| 300 | : matchingIndividuals.OrderBy(x => (DoubleValue)x.Variables["Quality"].Value).Take(n).ToList();
|
---|
[12979] | 301 | var mutationOc = new OperationCollection { Parallel = false }; // cannot be parallel due to the before/after operators which insert vertices in the genealogy graph
|
---|
| 302 | var updateEstimatedValues = new OperationCollection { Parallel = true }; // evaluation should be done in parallel when possible
|
---|
[12951] | 303 | foreach (var ind in individualsToReplace) {
|
---|
[15482] | 304 | var mutatorOp = ExecutionContext.CreateChildOperation(SchemaManipulator, ind);
|
---|
[13527] | 305 | var updateOp = ExecutionContext.CreateChildOperation(updateQualityOperator, ind);
|
---|
[12979] | 306 | mutationOc.Add(mutatorOp);
|
---|
| 307 | updateEstimatedValues.Add(updateOp);
|
---|
[12988] | 308 | if (exclusiveMatching)
|
---|
| 309 | ind.Variables.Add(new Core.Variable("AlreadyMatched"));
|
---|
[12951] | 310 | }
|
---|
[12988] | 311 |
|
---|
| 312 | NumberOfChangedTrees.Value += individualsToReplace.Count; // a lock is not necessary here because the SchemaEvaluators cannot be executed in parallel
|
---|
| 313 |
|
---|
[12979] | 314 | return new OperationCollection(mutationOc, updateEstimatedValues, base.Apply());
|
---|
[12951] | 315 | }
|
---|
[12979] | 316 |
|
---|
[13565] | 317 | private double CalculateReplacementRatio(double r) {
|
---|
| 318 | switch (ReplacementRule) {
|
---|
| 319 | case "f(x) = x": {
|
---|
| 320 | return r;
|
---|
| 321 | }
|
---|
| 322 | case "f(x) = tanh(x)": {
|
---|
| 323 | return Math.Tanh(r);
|
---|
| 324 | }
|
---|
| 325 | case "f(x) = tanh(2x)": {
|
---|
| 326 | return Math.Tanh(2 * r);
|
---|
| 327 | }
|
---|
| 328 | case "f(x) = tanh(3x)": {
|
---|
| 329 | return Math.Tanh(3 * r);
|
---|
| 330 | }
|
---|
| 331 | case "f(x) = tanh(4x)": {
|
---|
| 332 | return Math.Tanh(4 * r);
|
---|
| 333 | }
|
---|
| 334 | case "f(x) = 1-sqrt(1-x)": {
|
---|
| 335 | return 1 - Math.Sqrt(1 - r);
|
---|
| 336 | }
|
---|
| 337 | default:
|
---|
| 338 | throw new ArgumentException("Unknown replacement rule");
|
---|
| 339 | }
|
---|
| 340 | }
|
---|
| 341 |
|
---|
[13480] | 342 | public static double CalculateSimilarity(ScopeList individuals, ISolutionSimilarityCalculator calculator, bool parallel = false, int nThreads = -1) {
|
---|
[12979] | 343 | double similarity = 0;
|
---|
| 344 | int count = individuals.Count;
|
---|
[13480] | 345 | if (count < 2) return double.NaN;
|
---|
[12979] | 346 | if (parallel) {
|
---|
| 347 | var parallelOptions = new ParallelOptions { MaxDegreeOfParallelism = nThreads };
|
---|
| 348 | var simArray = new double[count - 1];
|
---|
| 349 | Parallel.For(0, count - 1, parallelOptions, i => {
|
---|
| 350 | double innerSim = 0;
|
---|
| 351 | for (int j = i + 1; j < count; ++j) {
|
---|
| 352 | innerSim += calculator.CalculateSolutionSimilarity(individuals[i], individuals[j]);
|
---|
| 353 | }
|
---|
| 354 | simArray[i] = innerSim;
|
---|
| 355 | });
|
---|
| 356 | similarity = simArray.Sum();
|
---|
| 357 | } else {
|
---|
| 358 | for (int i = 0; i < count - 1; ++i) {
|
---|
| 359 | for (int j = i + 1; j < count; ++j) {
|
---|
| 360 | similarity += calculator.CalculateSolutionSimilarity(individuals[i], individuals[j]);
|
---|
| 361 | }
|
---|
| 362 | }
|
---|
| 363 | }
|
---|
| 364 | return similarity / (count * (count - 1) / 2.0);
|
---|
| 365 | }
|
---|
[12951] | 366 | }
|
---|
| 367 | }
|
---|