Changeset 14776

Timestamp:

03/22/17 16:52:36 (8 years ago)

Author:

abeham

Message:

#2457: working on MemPR integration

Location:

branches/PerformanceComparison

Files:

• HeuristicLab.Algorithms.MemPR/3.3/Binary/SolutionModel/Univariate/BiasedModelTrainer.cs (modified) (3 diffs)
• HeuristicLab.Algorithms.MemPR/3.3/Binary/SolutionModel/Univariate/StaticAPI (deleted)
• HeuristicLab.Algorithms.MemPR/3.3/Binary/SolutionModel/Univariate/UnbiasedModelTrainer.cs (modified) (2 diffs)
• HeuristicLab.Algorithms.MemPR/3.3/HeuristicLab.Algorithms.MemPR-3.3.csproj (modified) (4 diffs)
• HeuristicLab.Algorithms.MemPR/3.3/Interfaces/Interfaces.cs (modified) (2 diffs)
• HeuristicLab.Algorithms.MemPR/3.3/LinearLinkage/SolutionModel/Univariate/StaticAPI/Trainer.cs (modified) (1 diff)
• HeuristicLab.Algorithms.MemPR/3.3/LinearLinkage/SolutionModel/Univariate/UnivariateSolutionModel.cs (modified) (1 diff)
• HeuristicLab.Algorithms.MemPR/3.3/Permutation/InversionPathRelinker.cs (added)
• HeuristicLab.Algorithms.MemPR/3.3/Permutation/LocalSearch/ExhaustiveInversionImprover.cs (added)
• HeuristicLab.Algorithms.MemPR/3.3/Permutation/SolutionModel/Univariate/StaticAPI/Trainer.cs (modified) (1 diff)
• HeuristicLab.Analysis (modified) (1 prop)
• HeuristicLab.Analysis.FitnessLandscape/3.3/CharacteristicCalculator/AdaptiveWalkCalculator.cs (modified) (1 diff)
• HeuristicLab.Analysis/3.3/DataVisualization/GanttData.cs (copied) (copied from trunk/sources/HeuristicLab.Analysis/3.3/DataVisualization/GanttData.cs)
• HeuristicLab.Analysis/3.3/HeuristicLab.Analysis-3.3.csproj (modified) (1 diff)
• HeuristicLab.Analysis/3.3/QualityAnalysis/ExpectedRuntimeHelper.cs (modified) (1 diff)
• HeuristicLab.Encodings.BinaryVectorEncoding (copied) (copied from trunk/sources/HeuristicLab.Encodings.BinaryVectorEncoding)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3 (modified) (1 prop)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/HeuristicLab.Encodings.BinaryVectorEncoding-3.3.csproj (modified) (2 diffs)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel (copied) (copied from branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Binary/SolutionModel)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/SolutionSamplingOperator.cs (added)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/Univariate/BiasedModelTrainer.cs (deleted)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/Univariate/StaticAPI (deleted)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/Univariate/UnbiasedModelTrainer.cs (deleted)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/Univariate/UnivariateModel.cs (moved) (moved from branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Binary/SolutionModel/Univariate/UnivariateSolutionModel.cs) (2 diffs)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/Univariate/UnivariateModelTrainer.cs (copied) (copied from branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Binary/SolutionModel/Univariate/StaticAPI/Trainer.cs) (1 diff)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/Univariate/UnivariateModelTrainingOperator.cs (added)
• HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/Univariate/UnivariateSolutionModel.cs (deleted)
• HeuristicLab.Encodings.LinearLinkageEncoding (copied) (copied from trunk/sources/HeuristicLab.Encodings.LinearLinkageEncoding)
• HeuristicLab.Encodings.LinearLinkageEncoding/3.4/HeuristicLab.Encodings.LinearLinkageEncoding-3.4.csproj (modified) (2 diffs)
• HeuristicLab.Optimization.Views (modified) (1 prop)
• HeuristicLab.Optimization.Views/3.3/RunCollectionViews/RunCollectionRLDView.cs (modified) (17 diffs)
• HeuristicLab.Optimization/3.3/HeuristicLab.Optimization-3.3.csproj (modified) (1 diff)
• HeuristicLab.Optimization/3.3/Interfaces/ISolutionModel.cs (added)
• HeuristicLab.OptimizationExpertSystem.Common/3.3/KnowledgeCenter.cs (modified) (2 diffs)
• PerformanceComparison.sln (modified) (2 diffs)
• ProblemInstanceIdentifier/InstanceExplorer.cs (modified) (1 diff)
• ProblemInstanceIdentifier/Program.cs (modified) (1 diff)

branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Binary/SolutionModel/Univariate/BiasedModelTrainer.cs

@@ -26 +26 @@
 using HeuristicLab.Data;
 using HeuristicLab.Encodings.BinaryVectorEncoding;
+using HeuristicLab.Encodings.BinaryVectorEncoding.SolutionModel;
 using HeuristicLab.Optimization;
 using HeuristicLab.Parameters;
@@ -31 +32 @@
 
 namespace HeuristicLab.Algorithms.MemPR.Binary.SolutionModel.Univariate {
+  public enum ModelBiasOptions { Rank, Fitness }
+
   [Item("Biased Univariate Model Trainer (binary)", "", ExcludeGenericTypeInfo = true)]
   [StorableClass]
@@ -60 +63 @@
 
     public void TrainModel(TContext context) {
-      context.Model = Trainer.TrainBiased(ModelBias, context.Random, context.Maximization, context.Population.Select(x => x.Solution), context.Population.Select(x => x.Fitness));
+      var biasType = modelBiasParameter.Value.Value;
+      switch (biasType) {
+        case ModelBiasOptions.Fitness:
+          context.Model = UnivariateModelTrainer.TrainWithFitnessBias(context.Random, context.Maximization,
+            context.Population.Select(x => x.Solution),
+            context.Population.Select(x => x.Fitness));
+          break;
+        case ModelBiasOptions.Rank:
+          context.Model = UnivariateModelTrainer.TrainWithRankBias(context.Random, context.Maximization,
+            context.Population.Select(x => x.Solution),
+            context.Population.Select(x => x.Fitness));
+          break;
+      }
     }
   }

branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Binary/SolutionModel/Univariate/UnbiasedModelTrainer.cs

@@ -25 +25 @@
 using HeuristicLab.Core;
 using HeuristicLab.Encodings.BinaryVectorEncoding;
+using HeuristicLab.Encodings.BinaryVectorEncoding.SolutionModel;
 using HeuristicLab.Optimization;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
@@ -49 +50 @@
 
     public void TrainModel(TContext context) {
-      context.Model = Trainer.TrainUnbiased(context.Random, context.Population.Select(x => x.Solution));
+      context.Model = UnivariateModelTrainer.TrainUnbiased(context.Random, context.Population.Select(x => x.Solution));
     }
   }

branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/HeuristicLab.Algorithms.MemPR-3.3.csproj

@@ -102 +102 @@
       <Private>False</Private>
     </Reference>
-    <Reference Include="HeuristicLab.Encodings.BinaryVectorEncoding-3.3">
-      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Encodings.BinaryVectorEncoding-3.3.dll</HintPath>
-      <Private>False</Private>
-    </Reference>
-    <Reference Include="HeuristicLab.Encodings.LinearLinkageEncoding-3.4">
-      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Encodings.LinearLinkageEncoding-3.4.dll</HintPath>
-      <Private>False</Private>
-    </Reference>
     <Reference Include="HeuristicLab.Operators-3.3">
       <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Operators-3.3.dll</HintPath>
@@ -165 +157 @@
     <Compile Include="Binary\LocalSearch\ExhaustiveBitflip.cs" />
     <Compile Include="Binary\LocalSearch\StaticAPI\ExhaustiveBitflip.cs" />
-    <Compile Include="Binary\SolutionModel\Univariate\StaticAPI\Trainer.cs" />
-    <Compile Include="Binary\SolutionModel\Univariate\UnivariateSolutionModel.cs" />
     <Compile Include="Binary\SolutionModel\Univariate\BiasedModelTrainer.cs" />
     <Compile Include="Binary\SolutionModel\Univariate\UnbiasedModelTrainer.cs" />
@@ -179 +169 @@
     <Compile Include="LinearLinkage\SolutionModel\Univariate\UnivariateSolutionModel.cs" />
     <Compile Include="MemPRAlgorithm.cs" />
+    <Compile Include="Permutation\InversionPathRelinker.cs" />
+    <Compile Include="Permutation\LocalSearch\ExhaustiveInversionImprover.cs" />
     <Compile Include="Permutation\PermutationMemPR.cs" />
     <Compile Include="Permutation\PermutationMemPRContext.cs" />
@@ -212 +204 @@
       <Private>False</Private>
     </ProjectReference>
+    <ProjectReference Include="..\..\HeuristicLab.Encodings.BinaryVectorEncoding\3.3\HeuristicLab.Encodings.BinaryVectorEncoding-3.3.csproj">
+      <Project>{66d249c3-a01d-42a8-82a2-919bc8ec3d83}</Project>
+      <Name>HeuristicLab.Encodings.BinaryVectorEncoding-3.3</Name>
+      <Private>False</Private>
+    </ProjectReference>
+    <ProjectReference Include="..\..\HeuristicLab.Encodings.LinearLinkageEncoding\3.4\HeuristicLab.Encodings.LinearLinkageEncoding-3.4.csproj">
+      <Project>{be698769-975a-429e-828c-72bb2b6182c8}</Project>
+      <Name>HeuristicLab.Encodings.LinearLinkageEncoding-3.4</Name>
+      <Private>False</Private>
+    </ProjectReference>
     <ProjectReference Include="..\..\HeuristicLab.Encodings.PermutationEncoding\3.3\HeuristicLab.Encodings.PermutationEncoding-3.3.csproj">
       <Project>{dbecb8b0-b166-4133-baf1-ed67c3fd7fca}</Project>

branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Interfaces/Interfaces.cs

@@ -37 +37 @@
    * ********************************************* *
    *************************************************/
-
-  public interface ISolutionModel<TSolution> : IItem {
-    TSolution Sample();
-  }
 
   public interface ISolutionSubspace<TSolution> : IItem { }
@@ -89 +85 @@
       where TProblem : class, ISingleObjectiveHeuristicOptimizationProblem {
     IEnumerable<ISingleObjectiveSolutionScope<TSolution>> Population { get; }
+    int PopulationCount { get; }
   }
 

branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/LinearLinkage/SolutionModel/Univariate/StaticAPI/Trainer.cs

@@ -21 +21 @@
 
 using System.Collections.Generic;
-using HeuristicLab.Algorithms.MemPR.Interfaces;
 using HeuristicLab.Core;
 using HeuristicLab.Encodings.LinearLinkageEncoding;
+using HeuristicLab.Optimization;
 
 namespace HeuristicLab.Algorithms.MemPR.Grouping.SolutionModel.Univariate {

branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/LinearLinkage/SolutionModel/Univariate/UnivariateSolutionModel.cs

@@ -22 +22 @@
 using System;
 using System.Collections.Generic;
-using HeuristicLab.Algorithms.MemPR.Interfaces;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
 using HeuristicLab.Data;
 using HeuristicLab.Encodings.LinearLinkageEncoding;
+using HeuristicLab.Optimization;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
 

branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Permutation/SolutionModel/Univariate/StaticAPI/Trainer.cs

@@ -22 +22 @@
 using System;
 using System.Collections.Generic;
-using HeuristicLab.Algorithms.MemPR.Interfaces;
 using HeuristicLab.Core;
 using HeuristicLab.Encodings.PermutationEncoding;
+using HeuristicLab.Optimization;
 
 namespace HeuristicLab.Algorithms.MemPR.Permutation.SolutionModel.Univariate {

branches/PerformanceComparison/HeuristicLab.Analysis.FitnessLandscape/3.3/CharacteristicCalculator/AdaptiveWalkCalculator.cs

@@ -58 +58 @@
       characteristics = cloner.Clone(original.characteristics);
     }
-    public AdaptiveWalkCalculator() {
+    public AdaptiveWalkCalculator() : this(new AdaptiveWalk()) { }
+    public AdaptiveWalkCalculator(AdaptiveWalk walker) {
       Name = ItemName;
       Description = ItemDescription;
-      walker = new AdaptiveWalk();
+      this.walker = walker;
       characteristics = new CheckedItemList<StringValue>(
         new[] { "AutoCorrelation1", "CorrelationLength", "InformationContent",
-        "PartialInformationContent", "DensityBasinInformation", "InformationStability", 
+        "PartialInformationContent", "DensityBasinInformation", "InformationStability",
         "Diversity", "Regularity", "TotalEntropy", "PeakInformationContent",
         "PeakDensityBasinInformation" }.Select(x => new StringValue(x)));

branches/PerformanceComparison/HeuristicLab.Analysis/3.3/HeuristicLab.Analysis-3.3.csproj

@@ -187 +187 @@
     <Compile Include="BestNScopesSolutionAnalyzer.cs" />
     <Compile Include="BestScopeSolutionAnalyzer.cs" />
+    <Compile Include="DataVisualization\GanttData.cs" />
     <Compile Include="Clustering\CkMeans1D.cs">
       <SubType>Code</SubType>

branches/PerformanceComparison/HeuristicLab.Analysis/3.3/QualityAnalysis/ExpectedRuntimeHelper.cs

@@ -30 +30 @@
       if (successful.Count > 0) {
         var succAvg = successful.Average();
-        var succDev = successful.StandardDeviation();
+        var succDev = successful.StandardDeviation() + 1e-7;
         successful.RemoveAll(x => x < succAvg - 2 * succDev);
         unsuccessful.RemoveAll(x => x < succAvg - 2 * succDev);

branches/PerformanceComparison/HeuristicLab.Encodings.BinaryVectorEncoding/3.3

@@ -5 +5 @@
 *.vs10x
 Plugin.cs
+*.DotSettings

@@ -5 +5 @@
 *.vs10x
 Plugin.cs
+*.DotSettings

branches/PerformanceComparison/HeuristicLab.Encodings.BinaryVectorEncoding/3.3/HeuristicLab.Encodings.BinaryVectorEncoding-3.3.csproj

@@ -99 +99 @@
   </PropertyGroup>
   <ItemGroup>
+    <Reference Include="HeuristicLab.Collections-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Collections-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Common-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Common-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Core-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Core-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Data-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Data-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Operators-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Operators-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Optimization.Operators-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Optimization.Operators-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Parameters-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Parameters-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Persistence-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Persistence-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.PluginInfrastructure-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.PluginInfrastructure-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
     <Reference Include="System" />
     <Reference Include="System.Core">
@@ -153 +189 @@
     <Compile Include="Properties\AssemblyInfo.cs" />
     <Compile Include="ShakingOperators\BinaryVectorShakingOperator.cs" />
-  </ItemGroup>
-  <ItemGroup>
-    <ProjectReference Include="..\..\HeuristicLab.Collections\3.3\HeuristicLab.Collections-3.3.csproj">
-      <Project>{958B43BC-CC5C-4FA2-8628-2B3B01D890B6}</Project>
-      <Name>HeuristicLab.Collections-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Common\3.3\HeuristicLab.Common-3.3.csproj">
-      <Project>{A9AD58B9-3EF9-4CC1-97E5-8D909039FF5C}</Project>
-      <Name>HeuristicLab.Common-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Core\3.3\HeuristicLab.Core-3.3.csproj">
-      <Project>{C36BD924-A541-4A00-AFA8-41701378DDC5}</Project>
-      <Name>HeuristicLab.Core-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Data\3.3\HeuristicLab.Data-3.3.csproj">
-      <Project>{BBAB9DF5-5EF3-4BA8-ADE9-B36E82114937}</Project>
-      <Name>HeuristicLab.Data-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Operators\3.3\HeuristicLab.Operators-3.3.csproj">
-      <Project>{23DA7FF4-D5B8-41B6-AA96-F0561D24F3EE}</Project>
-      <Name>HeuristicLab.Operators-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Optimization.Operators\3.3\HeuristicLab.Optimization.Operators-3.3.csproj">
-      <Project>{25087811-F74C-4128-BC86-8324271DA13E}</Project>
-      <Name>HeuristicLab.Optimization.Operators-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
+    <Compile Include="SolutionModel\Univariate\UnivariateModelTrainer.cs" />
+    <Compile Include="SolutionModel\Univariate\UnivariateModel.cs" />
+    <Compile Include="SolutionModel\SolutionSamplingOperator.cs" />
+    <Compile Include="SolutionModel\Univariate\UnivariateModelTrainingOperator.cs" />
+  </ItemGroup>
+  <ItemGroup>
     <ProjectReference Include="..\..\HeuristicLab.Optimization\3.3\HeuristicLab.Optimization-3.3.csproj">
       <Project>{14AB8D24-25BC-400C-A846-4627AA945192}</Project>
       <Name>HeuristicLab.Optimization-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Parameters\3.3\HeuristicLab.Parameters-3.3.csproj">
-      <Project>{56F9106A-079F-4C61-92F6-86A84C2D84B7}</Project>
-      <Name>HeuristicLab.Parameters-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Persistence\3.3\HeuristicLab.Persistence-3.3.csproj">
-      <Project>{102BC7D3-0EF9-439C-8F6D-96FF0FDB8E1B}</Project>
-      <Name>HeuristicLab.Persistence-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.PluginInfrastructure\3.3\HeuristicLab.PluginInfrastructure-3.3.csproj">
-      <Project>{94186A6A-5176-4402-AE83-886557B53CCA}</Project>
-      <Name>HeuristicLab.PluginInfrastructure-3.3</Name>
       <Private>False</Private>
     </ProjectReference>

branches/PerformanceComparison/HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/Univariate/UnivariateModel.cs

@@ -20 +20 @@
 #endregion
 
-using System;
-using System.Collections.Generic;
 using System.Linq;
-using HeuristicLab.Algorithms.MemPR.Interfaces;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
 using HeuristicLab.Data;
-using HeuristicLab.Encodings.BinaryVectorEncoding;
+using HeuristicLab.Optimization;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
-using HeuristicLab.Random;
 
-namespace HeuristicLab.Algorithms.MemPR.Binary.SolutionModel.Univariate {
+namespace HeuristicLab.Encodings.BinaryVectorEncoding.SolutionModel {
   [Item("Univariate solution model (binary)", "")]
   [StorableClass]
@@ -67 +63 @@
       return vec;
     }
-
-    public static ISolutionModel<BinaryVector> CreateWithoutBias(IRandom random, IEnumerable<BinaryVector> population) {
-      double[] model = null;
-      var popSize = 0;
-      foreach (var p in population) {
-        popSize++;
-        if (model == null) model = new double[p.Length];
-        for (var x = 0; x < model.Length; x++) {
-          if (p[x]) model[x]++;
-        }
-      }
-      if (model == null) throw new ArgumentException("Cannot train model from empty population.");
-      // normalize to [0;1]
-      var factor = 1.0 / popSize;
-      for (var x = 0; x < model.Length; x++) {
-        model[x] *= factor;
-      }
-      return new UnivariateModel(random, model);
-    }
-
-    public static ISolutionModel<BinaryVector> CreateWithRankBias(IRandom random, bool maximization, IEnumerable<BinaryVector> population, IEnumerable<double> qualities) {
-      var popSize = 0;
-
-      double[] model = null;
-      var pop = population.Zip(qualities, (b, q) => new { Solution = b, Fitness = q });
-      foreach (var ind in maximization ? pop.OrderBy(x => x.Fitness) : pop.OrderByDescending(x => x.Fitness)) {
-        // from worst to best, worst solution has 1 vote, best solution N votes
-        popSize++;
-        if (model == null) model = new double[ind.Solution.Length];
-        for (var x = 0; x < model.Length; x++) {
-          if (ind.Solution[x]) model[x] += popSize;
-        }
-      }
-      if (model == null) throw new ArgumentException("Cannot train model from empty population.");
-      // normalize to [0;1]
-      var factor = 2.0 / (popSize + 1);
-      for (var i = 0; i < model.Length; i++) {
-        model[i] *= factor / popSize;
-      }
-      return new UnivariateModel(random, model);
-    }
-
-    public static ISolutionModel<BinaryVector> CreateWithFitnessBias(IRandom random, bool maximization, IEnumerable<BinaryVector> population, IEnumerable<double> qualities) {
-      var proportions = Util.Auxiliary.PrepareProportional(qualities, true, !maximization);
-      var factor = 1.0 / proportions.Sum();
-      double[] model = null;
-      foreach (var ind in population.Zip(proportions, (p, q) => new { Solution = p, Proportion = q })) {
-        if (model == null) model = new double[ind.Solution.Length];
-        for (var x = 0; x < model.Length; x++) {
-          if (ind.Solution[x]) model[x] += ind.Proportion * factor;
-        }
-      }
-      if (model == null) throw new ArgumentException("Cannot train model from empty population.");
-      return new UnivariateModel(random, model);
-    }
   }
 }

branches/PerformanceComparison/HeuristicLab.Encodings.BinaryVectorEncoding/3.3/SolutionModel/Univariate/UnivariateModelTrainer.cs

@@ -22 +22 @@
 using System;
 using System.Collections.Generic;
-using HeuristicLab.Algorithms.MemPR.Interfaces;
+using System.Linq;
 using HeuristicLab.Core;
-using HeuristicLab.Encodings.BinaryVectorEncoding;
 
-namespace HeuristicLab.Algorithms.MemPR.Binary.SolutionModel.Univariate {
-  public enum ModelBiasOptions { Rank, Fitness }
+namespace HeuristicLab.Encodings.BinaryVectorEncoding.SolutionModel {
+  public static class UnivariateModelTrainer {
+    /// <summary>
+    /// Creates a univariate sampling model out of a ranked population.
+    /// The first solution in <paramref name="rankedPopulation"/> is the 
+    /// highest influential one, the last solution has the least influence.
+    /// </summary>
+    /// <param name="random">The model is stochastic and will use this RNG for sampling.</param>
+    /// <param name="maximization">Whether higher or lower qualities are better</param>
+    /// <param name="population">The population that is ranked from highest influential to least influential, e.g. best to worst.</param>
+    /// <param name="qualities">The solution quality of the respective solution.</param>
+    /// <returns>The sampling model which is created with the given population.</returns>
+    public static UnivariateModel TrainWithRankBias(IRandom random, bool maximization, IEnumerable<BinaryVector> population, IEnumerable<double> qualities) {
+      var popSize = 0;
+      double[] model = null;
+      var pop = population.Zip(qualities, (b, q) => new { Solution = b, Fitness = q });
+      foreach (var ind in maximization ? pop.OrderBy(x => x.Fitness) : pop.OrderByDescending(x => x.Fitness)) {
+        // from worst to best, worst solution has 1 vote, best solution N votes
+        popSize++;
+        if (model == null) model = new double[ind.Solution.Length];
+        for (var x = 0; x < model.Length; x++) {
+          if (ind.Solution[x]) model[x] += popSize;
+        }
+      }
+      if (model == null) throw new ArgumentException("Cannot train model from empty population.");
+      // normalize to [0;1]
+      var factor = 2.0 / (popSize + 1);
+      for (var i = 0; i < model.Length; i++) {
+        model[i] *= factor / popSize;
+      }
+      return new UnivariateModel(random, model);
+    }
 
-  public static class Trainer {
-    public static ISolutionModel<BinaryVector> TrainBiased(ModelBiasOptions modelBias, IRandom random, bool maximization, IEnumerable<BinaryVector> population, IEnumerable<double> qualities) {
-      switch (modelBias) {
-        case ModelBiasOptions.Rank:
-          return UnivariateModel.CreateWithRankBias(random, maximization, population, qualities);
-        case ModelBiasOptions.Fitness:
-          return UnivariateModel.CreateWithFitnessBias(random, maximization, population, qualities);
-        default:
-          throw new InvalidOperationException(string.Format("Unknown bias option {0}", modelBias));
+    /// <summary>
+    /// Creates a univariate sampling model out of solutions and their given fitness.
+    /// The best solution's influence is proportional to its fitness, while the worst
+    /// solution does not have an influence at all (except if the fitness of worst and
+    /// best are equal).
+    /// </summary>
+    /// <param name="random">The model is stochastic and makes use of this RNG instance for sampling.</param>
+    /// <param name="maximization">Whether higher fitness values are better or lower ones.</param>
+    /// <param name="population">The solutions that will be used to create the model.</param>
+    /// <param name="qualities">The solutions' associated qualities.</param>
+    /// <returns>The sampling model which is created with the given population.</returns>
+    public static UnivariateModel TrainWithFitnessBias(IRandom random, bool maximization, IEnumerable<BinaryVector> population, IEnumerable<double> qualities) {
+      var proportions = PrepareProportional(qualities, true, !maximization);
+      var factor = 1.0 / proportions.Sum();
+      double[] model = null;
+      foreach (var ind in population.Zip(proportions, (p, q) => new { Solution = p, Proportion = q })) {
+        if (model == null) model = new double[ind.Solution.Length];
+        for (var x = 0; x < model.Length; x++) {
+          if (ind.Solution[x]) model[x] += ind.Proportion * factor;
+        }
+      }
+      if (model == null) throw new ArgumentException("Cannot train model from empty population.");
+      return new UnivariateModel(random, model);
+    }
+
+    /// <summary>
+    /// Creates a univariate sampling model out of solutions. Each of the solutions
+    /// has the same influence on the model.
+    /// </summary>
+    /// <param name="random">The model is stochastic and will make use of this RNG instance.</param>
+    /// <param name="population">The solutions that are used to create the model.</param>
+    /// <returns>The model created from the population.</returns>
+    public static UnivariateModel TrainUnbiased(IRandom random, IEnumerable<BinaryVector> population) {
+      double[] model = null;
+      var popSize = 0;
+      foreach (var p in population) {
+        popSize++;
+        if (model == null) model = new double[p.Length];
+        for (var x = 0; x < model.Length; x++) {
+          if (p[x]) model[x]++;
+        }
+      }
+      if (model == null) throw new ArgumentException("Cannot train model from empty population.");
+      // normalize to [0;1]
+      var factor = 1.0 / popSize;
+      for (var x = 0; x < model.Length; x++) {
+        model[x] *= factor;
+      }
+      return new UnivariateModel(random, model);
+    }
+
+    // TODO: make PrepareProportional public in EnumerableExtensions
+    private static double[] PrepareProportional(IEnumerable<double> weights, bool windowing, bool inverseProportional) {
+      double maxValue = double.MinValue, minValue = double.MaxValue;
+      double[] valueArray = weights.ToArray();
+
+      for (int i = 0; i < valueArray.Length; i++) {
+        if (valueArray[i] > maxValue) maxValue = valueArray[i];
+        if (valueArray[i] < minValue) minValue = valueArray[i];
+      }
+      if (minValue == maxValue) {  // all values are equal
+        for (int i = 0; i < valueArray.Length; i++) {
+          valueArray[i] = 1.0;
+        }
+      } else {
+        if (windowing) {
+          if (inverseProportional) InverseProportionalScale(valueArray, maxValue);
+          else ProportionalScale(valueArray, minValue);
+        } else {
+          if (minValue < 0.0) throw new InvalidOperationException("Proportional selection without windowing does not work with values < 0.");
+          if (inverseProportional) InverseProportionalScale(valueArray, 2 * maxValue);
+        }
+      }
+      return valueArray;
+    }
+    private static void ProportionalScale(double[] values, double minValue) {
+      for (int i = 0; i < values.Length; i++) {
+        values[i] = values[i] - minValue;
       }
     }
-
-    public static ISolutionModel<BinaryVector> TrainUnbiased(IRandom random, IEnumerable<BinaryVector> population) {
-      return UnivariateModel.CreateWithoutBias(random, population);
+    private static void InverseProportionalScale(double[] values, double maxValue) {
+      for (int i = 0; i < values.Length; i++) {
+        values[i] = maxValue - values[i];
+      }
     }
   }

branches/PerformanceComparison/HeuristicLab.Encodings.LinearLinkageEncoding/3.4/HeuristicLab.Encodings.LinearLinkageEncoding-3.4.csproj

@@ -88 +88 @@
   </PropertyGroup>
   <ItemGroup>
-    <ProjectReference Include="..\..\HeuristicLab.Collections\3.3\HeuristicLab.Collections-3.3.csproj">
-      <Project>{958B43BC-CC5C-4FA2-8628-2B3B01D890B6}</Project>
-      <Name>HeuristicLab.Collections-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Common\3.3\HeuristicLab.Common-3.3.csproj">
-      <Project>{A9AD58B9-3EF9-4CC1-97E5-8D909039FF5C}</Project>
-      <Name>HeuristicLab.Common-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Core\3.3\HeuristicLab.Core-3.3.csproj">
-      <Project>{C36BD924-A541-4A00-AFA8-41701378DDC5}</Project>
-      <Name>HeuristicLab.Core-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Data\3.3\HeuristicLab.Data-3.3.csproj">
-      <Project>{BBAB9DF5-5EF3-4BA8-ADE9-B36E82114937}</Project>
-      <Name>HeuristicLab.Data-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Operators\3.3\HeuristicLab.Operators-3.3.csproj">
-      <Project>{23DA7FF4-D5B8-41B6-AA96-F0561D24F3EE}</Project>
-      <Name>HeuristicLab.Operators-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Optimization.Operators\3.3\HeuristicLab.Optimization.Operators-3.3.csproj">
-      <Project>{25087811-F74C-4128-BC86-8324271DA13E}</Project>
-      <Name>HeuristicLab.Optimization.Operators-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
     <ProjectReference Include="..\..\HeuristicLab.Optimization\3.3\HeuristicLab.Optimization-3.3.csproj">
       <Project>{14AB8D24-25BC-400C-A846-4627AA945192}</Project>
@@ -123 +93 @@
       <Private>False</Private>
     </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Parameters\3.3\HeuristicLab.Parameters-3.3.csproj">
-      <Project>{56F9106A-079F-4C61-92F6-86A84C2D84B7}</Project>
-      <Name>HeuristicLab.Parameters-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Persistence\3.3\HeuristicLab.Persistence-3.3.csproj">
-      <Project>{102BC7D3-0EF9-439C-8F6D-96FF0FDB8E1B}</Project>
-      <Name>HeuristicLab.Persistence-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.PluginInfrastructure\3.3\HeuristicLab.PluginInfrastructure-3.3.csproj">
-      <Project>{94186A6A-5176-4402-AE83-886557B53CCA}</Project>
-      <Name>HeuristicLab.PluginInfrastructure-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
-    <ProjectReference Include="..\..\HeuristicLab.Random\3.3\HeuristicLab.Random-3.3.csproj">
-      <Project>{f4539fb6-4708-40c9-be64-0a1390aea197}</Project>
-      <Name>HeuristicLab.Random-3.3</Name>
-      <Private>False</Private>
-    </ProjectReference>
   </ItemGroup>
   <ItemGroup>
+    <Reference Include="HeuristicLab.Collections-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Collections-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Common-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Common-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Core-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Core-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Data-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Data-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Operators-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Operators-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Optimization.Operators-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Optimization.Operators-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Parameters-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Parameters-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Persistence-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Persistence-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.PluginInfrastructure-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.PluginInfrastructure-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
+    <Reference Include="HeuristicLab.Random-3.3">
+      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Random-3.3.dll</HintPath>
+      <Private>False</Private>
+    </Reference>
     <Reference Include="System" />
     <Reference Include="System.Core" />

branches/PerformanceComparison/HeuristicLab.Optimization.Views/3.3/RunCollectionViews/RunCollectionRLDView.cs

@@ -1 +1 @@
 #region License Information
 /* HeuristicLab
- * Copyright (C) 2002-2015 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
+ * Copyright (C) 2002-2017 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
  *
  * This file is part of HeuristicLab.
@@ -42 +42 @@
     private List<Series> invisibleTargetSeries;
 
-    private const string AllRuns = "All Runs";
+    private const string AllInstances = "All Instances";
 
     private static readonly Color[] colors = new[] {
@@ -223 +223 @@
       var groupings = Content.ParameterNames.OrderBy(x => x).ToArray();
       groupComboBox.Items.Clear();
-      groupComboBox.Items.Add(AllRuns);
+      groupComboBox.Items.Add(AllInstances);
       groupComboBox.Items.AddRange(groupings);
       if (selectedGroupItem != null && groupComboBox.Items.Contains(selectedGroupItem)) {
@@ -241 +241 @@
 
       var selectedProblemItem = (ProblemInstance)problemComboBox.SelectedItem;
+      problemComboBox.DataSource = null;
+      problemComboBox.Items.Clear();
       problems.Clear();
       problems.Add(ProblemInstance.MatchAll);
+      problemComboBox.DataSource = new BindingSource() { DataSource = problems };
+      problemComboBox.DataBindings.DefaultDataSourceUpdateMode = DataSourceUpdateMode.OnPropertyChanged;
       if (problems[0].Equals(selectedProblemItem)) problemComboBox.SelectedItem = problems[0];
       foreach (var p in problemDict.ToList()) {
@@ -282 +286 @@
     }
 
-    private Dictionary<string, Dictionary<ProblemInstance, List<IRun>>> GroupRuns() {
-      var groupedRuns = new Dictionary<string, Dictionary<ProblemInstance, List<IRun>>>();
-
+    private IEnumerable<AlgorithmInstance> GroupRuns() {
       var table = (string)dataTableComboBox.SelectedItem;
-      if (string.IsNullOrEmpty(table)) return groupedRuns;
+      if (string.IsNullOrEmpty(table)) yield break;
 
       var selectedGroup = (string)groupComboBox.SelectedItem;
-      if (string.IsNullOrEmpty(selectedGroup)) return groupedRuns;
+      if (string.IsNullOrEmpty(selectedGroup)) yield break;
 
       var selectedProblem = (ProblemInstance)problemComboBox.SelectedItem;
-      if (selectedProblem == null) return groupedRuns;
+      if (selectedProblem == null) yield break;
       
       foreach (var x in (from r in Content
-                         where (selectedGroup == AllRuns || r.Parameters.ContainsKey(selectedGroup))
+                         where (selectedGroup == AllInstances || r.Parameters.ContainsKey(selectedGroup))
                            && selectedProblem.Match(r)
                            && r.Results.ContainsKey(table)
                            && r.Visible
-                         let key = selectedGroup == AllRuns ? AllRuns : r.Parameters[selectedGroup].ToString()
+                         let key = selectedGroup == AllInstances ? AllInstances : r.Parameters[selectedGroup].ToString()
                          group r by key into g
-                         select Tuple.Create(g.Key, g.ToList()))) {
-        var pDict = problems.ToDictionary(r => r, _ => new List<IRun>());
-        foreach (var y in x.Item2) {
-          var pd = new ProblemInstance(y);
-          List<IRun> l;
-          if (pDict.TryGetValue(pd, out l)) l.Add(y);
-        }
-        groupedRuns[x.Item1] = pDict.Where(a => a.Value.Count > 0).ToDictionary(a => a.Key, a => a.Value);
-      }
-
-      return groupedRuns;
+                         select new AlgorithmInstance(g.Key, g, problems))) {
+        yield return x;
+      }
     }
 
@@ -326 +320 @@
       if (targets == null) GenerateDefaultTargets();
 
-      var groupedRuns = GroupRuns();
-      if (groupedRuns.Count == 0) return;
+      var algInstances = GroupRuns().ToList();
+      if (algInstances.Count == 0) return;
 
       var xAxisTitles = new HashSet<string>();
@@ -337 +331 @@
       // how many targets have been left open at the point when the run ended
       var misses = new Dictionary<string, SortedList<double, int>>();
+      var totalRuns = new Dictionary<string, int>();
 
       var aggregate = aggregateTargetsCheckBox.Checked;
       double minEff = double.MaxValue, maxEff = double.MinValue;
-      var noRuns = 0;
-      foreach (var group in groupedRuns) {
+      foreach (var alg in algInstances) {
+        var noRuns = 0;
         SortedList<double, int> epdfHits = null, epdfMisses = null;
         if (aggregate) {
-          hits[group.Key] = epdfHits = new SortedList<double, int>();
-          misses[group.Key] = epdfMisses = new SortedList<double, int>();
-        }
-        foreach (var problem in group.Value) {
-          var max = problem.Key.IsMaximization();
-          var absTargets = GetAbsoluteTargets(problem.Key).ToArray();
-          foreach (var run in problem.Value) {
+          hits[alg.Name] = epdfHits = new SortedList<double, int>();
+          misses[alg.Name] = epdfMisses = new SortedList<double, int>();
+        }
+        foreach (var problem in alg.GetProblemInstances()) {
+          var max = problem.IsMaximization();
+          var absTargets = GetAbsoluteTargets(problem).ToArray();
+          foreach (var run in alg.GetRuns(problem)) {
             noRuns++;
             var resultsTable = (IndexedDataTable<double>)run.Results[table];
@@ -361 +356 @@
               var e = efforts[idx];
               if (!aggregate) {
-                var key = group.Key + "@" + (targetsAreRelative
-                  ? (targets[idx] * 100).ToString(CultureInfo.CurrentCulture.NumberFormat) + "%"
-                  : targets[idx].ToString(CultureInfo.CurrentCulture.NumberFormat));
+                var key = alg.Name + "@" + (targetsAreRelative
+                            ? (targets[idx] * 100).ToString(CultureInfo.CurrentCulture.NumberFormat) + "%"
+                            : targets[idx].ToString(CultureInfo.CurrentCulture.NumberFormat));
                 if (!hits.TryGetValue(key, out epdfHits))
                   hits[key] = epdfHits = new SortedList<double, int>();
                 if (!misses.TryGetValue(key, out epdfMisses))
                   misses[key] = epdfMisses = new SortedList<double, int>();
-              }
+                totalRuns[key] = noRuns;
+              };
               var list = e.Item1 ? epdfHits : epdfMisses;
               int v;
@@ -377 +373 @@
           }
         }
+        if (aggregate) totalRuns[alg.Name] = noRuns;
       }
 
       UpdateTargetChartAxisXBounds(minEff, maxEff);
 
-      DrawTargetsEcdf(hits, misses, noRuns);
+      DrawTargetsEcdf(hits, misses, totalRuns);
 
       if (targets.Length == 1) {
@@ -392 +389 @@
       targetChart.ChartAreas[0].CursorY.Interval = 0.05;
 
-      UpdateErtTables(groupedRuns);
-    }
-
-    private void DrawTargetsEcdf(Dictionary<string, SortedList<double, int>> hits, Dictionary<string, SortedList<double, int>> misses, int noRuns) {
+      UpdateErtTables(algInstances);
+    }
+
+    private void DrawTargetsEcdf(Dictionary<string, SortedList<double, int>> hits, Dictionary<string, SortedList<double, int>> misses, Dictionary<string, int> noRuns) {
       var colorCount = 0;
       var lineStyleCount = 0;
@@ -418 +415 @@
         var iter = misses[list.Key].GetEnumerator();
         var moreMisses = iter.MoveNext();
-        var totalTargets = noRuns;
+        var totalTargets = noRuns[list.Key];
         if (aggregateTargetsCheckBox.Checked) totalTargets *= targets.Length;
         var movingTargets = totalTargets;
@@ -594 +591 @@
     }
 
-    private void UpdateErtTables(Dictionary<string, Dictionary<ProblemInstance, List<IRun>>> groupedRuns) {
+    private void UpdateErtTables(List<AlgorithmInstance> algorithmInstances) {
       ertTableView.Content = null;
       var columns = 1 + targets.Length + 1;
-      var matrix = new string[groupedRuns.Count * groupedRuns.Max(x => x.Value.Count) + groupedRuns.Max(x => x.Value.Count), columns];
+      var matrix = new string[algorithmInstances.Count * algorithmInstances.Max(x => x.GetNumberOfProblemInstances()) + algorithmInstances.Max(x => x.GetNumberOfProblemInstances()), columns];
       var rowCount = 0;
 
@@ -603 +600 @@
       if (string.IsNullOrEmpty(tableName)) return;
       
-      var problems = groupedRuns.SelectMany(x => x.Value.Keys).Distinct().ToList();
+      var problems = algorithmInstances.SelectMany(x => x.GetProblemInstances()).Distinct().ToList();
 
       foreach (var problem in problems) {
@@ -615 +612 @@
         rowCount++;
 
-        foreach (var group in groupedRuns) {
-          matrix[rowCount, 0] = group.Key;
-          if (!group.Value.ContainsKey(problem)) {
+        foreach (var alg in algorithmInstances) {
+          matrix[rowCount, 0] = alg.Name;
+          var runs = alg.GetRuns(problem).ToList();
+          if (runs.Count == 0) {
             matrix[rowCount, columns - 1] = "N/A";
             rowCount++;
             continue;
           }
-          var runs = group.Value[problem];
           var result = default(ErtCalculationResult);
           for (var i = 0; i < absTargets.Length; i++) {
@@ -648 +645 @@
       if (budgets == null) GenerateDefaultBudgets(table);
 
-      var groupedRuns = GroupRuns();
-      if (groupedRuns.Count == 0) return;
+      var algInstances = GroupRuns().ToList();
+      if (algInstances.Count == 0) return;
 
       var colorCount = 0;
       var lineStyleCount = 0;
       
-      foreach (var group in groupedRuns) {
+      foreach (var alg in algInstances) {
         var hits = new Dictionary<string, SortedList<double, double>>();
 
-        foreach (var problem in group.Value) {
-          foreach (var run in problem.Value) {
+        foreach (var problem in alg.GetProblemInstances()) {
+          foreach (var run in alg.GetRuns(problem)) {
             var resultsTable = (IndexedDataTable<double>)run.Results[table];
 
             if (aggregateBudgetsCheckBox.Checked) {
-              CalculateHitsForAllBudgets(hits, resultsTable.Rows.First(), group.Value.Count, problem.Key, group.Key, problem.Value.Count);
+              CalculateHitsForAllBudgets(hits, resultsTable.Rows.First(), alg.GetNumberOfProblemInstances(), problem, alg.Name, alg.GetNumberOfRuns(problem));
             } else {
-              CalculateHitsForEachBudget(hits, resultsTable.Rows.First(), group.Value.Count, problem.Key, group.Key, problem.Value.Count);
+              CalculateHitsForEachBudget(hits, resultsTable.Rows.First(), alg.GetNumberOfProblemInstances(), problem, alg.Name, alg.GetNumberOfRuns(problem));
             }
           }
@@ -697 +694 @@
 
     private void GenerateDefaultBudgets(string table) {
-      var runs = GroupRuns().SelectMany(x => x.Value.Values).SelectMany(x => x).ToList();
+      var runs = Content;
       var min = runs.Select(x => ((IndexedDataTable<double>)x.Results[table]).Rows.First().Values.Select(y => y.Item1).Min()).Min();
       var max = runs.Select(x => ((IndexedDataTable<double>)x.Results[table]).Rows.First().Values.Select(y => y.Item1).Max()).Max();
@@ -1172 +1169 @@
     }
 
+    private class AlgorithmInstance : INotifyPropertyChanged {
+      private string name;
+      public string Name {
+        get { return name; }
+        set {
+          if (name == value) return;
+          name = value;
+          OnPropertyChanged("Name");
+        }
+      }
+
+      private Dictionary<ProblemInstance, List<IRun>> performanceData;
+
+      public int GetNumberOfProblemInstances() {
+        return performanceData.Count;
+      }
+
+      public IEnumerable<ProblemInstance> GetProblemInstances() {
+        return performanceData.Keys;
+      }
+
+      public int GetNumberOfRuns(ProblemInstance p) {
+        if (p == ProblemInstance.MatchAll) return performanceData.Select(x => x.Value.Count).Sum();
+        List<IRun> runs;
+        if (performanceData.TryGetValue(p, out runs))
+          return runs.Count;
+
+        return 0;
+      }
+
+      public IEnumerable<IRun> GetRuns(ProblemInstance p) {
+        if (p == ProblemInstance.MatchAll) return performanceData.SelectMany(x => x.Value);
+        List<IRun> runs;
+        if (performanceData.TryGetValue(p, out runs))
+          return runs;
+
+        return Enumerable.Empty<IRun>();
+      }
+
+      public AlgorithmInstance(string name, IEnumerable<IRun> runs, IEnumerable<ProblemInstance> problems) {
+        this.name = name;
+
+        var pDict = problems.ToDictionary(r => r, _ => new List<IRun>());
+        foreach (var y in runs) {
+          var pd = new ProblemInstance(y);
+          List<IRun> l;
+          if (pDict.TryGetValue(pd, out l)) l.Add(y);
+        }
+        performanceData = pDict.Where(x => x.Value.Count > 0).ToDictionary(x => x.Key, x => x.Value);
+      }
+
+      public override bool Equals(object obj) {
+        var other = obj as AlgorithmInstance;
+        if (other == null) return false;
+        return name == other.name;
+      }
+
+      public override int GetHashCode() {
+        return name.GetHashCode();
+      }
+
+      public event PropertyChangedEventHandler PropertyChanged;
+      protected virtual void OnPropertyChanged(string propertyName = null) {
+        var handler = PropertyChanged;
+        if (handler != null) handler(this, new PropertyChangedEventArgs(propertyName));
+      }
+    }
+
     private class ProblemInstance : INotifyPropertyChanged {
       private readonly bool matchAll;

branches/PerformanceComparison/HeuristicLab.Optimization/3.3/HeuristicLab.Optimization-3.3.csproj

@@ -202 +202 @@
     <Compile Include="Interfaces\IMultiObjectiveOperator.cs" />
     <Compile Include="Interfaces\IPreexistingSolutionCreator.cs" />
+    <Compile Include="Interfaces\ISolutionModel.cs" />
     <Compile Include="Results\IResultParameter.cs" />
     <Compile Include="Interfaces\ISingleObjectiveOperator.cs" />

branches/PerformanceComparison/HeuristicLab.OptimizationExpertSystem.Common/3.3/KnowledgeCenter.cs

@@ -202 +202 @@
         key2Idx.Add(kvp.Key, kvp.Index);
 
+      Func<double[], double[], double> euclid = (a, b) => Math.Sqrt(a.Zip(b, (x, y) => (x - y)).Sum(x => x * x));
+      Func<DoubleArray, DoubleArray, double> euclidDArray = (a, b) => Math.Sqrt(a.Zip(b, (x, y) => (x - y)).Sum(x => x * x));
       #region MDS
-      Func<double[], double[], double> euclid = (a, b) => Math.Sqrt(a.Zip(b, (x, y) => (x - y)).Sum(x => x * x));
       var num = instances.Count;
       var matrix = new DoubleMatrix(num, num);
@@ -243 +244 @@
         tsneFeatures[key2Idx.GetByFirst(instance.Key)] = new DoubleArray(instance.Value);
       }
-      var tsneCoords = TSNE<DoubleArray>.Run(tsneFeatures, 2, Math.Min((instances.Count - 1) / 3, 50), 0.1, (a, b) => Math.Sqrt(a.Zip(b, (c, d) => (c - d) * (c - d)).Sum()), new FastRandom(42));
+      var tsne = new TSNE<DoubleArray>(new EuclidianDistance(), new FastRandom(42), stopLyingIter: 0, momSwitchIter: 0, momentum: 0, finalMomentum: 0, eta: 10);
+      var tsneCoords = tsne.Run(tsneFeatures, 2, Math.Min((instances.Count - 1) / 4, 50), 0);
+      //var tsneCoords = TSNE<DoubleArray>.Run(tsneFeatures, 2, Math.Min((instances.Count - 1) / 4, 50), 0, euclidDArray, new FastRandom(42));
       #endregion
 

branches/PerformanceComparison/PerformanceComparison.sln

@@ -27 +27 @@
 EndProject
 Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "HeuristicLab.Problems.Instances.QAPLIB-3.3", "HeuristicLab.Problems.Instances.QAPLIB\3.3\HeuristicLab.Problems.Instances.QAPLIB-3.3.csproj", "{73F29D43-5714-4069-8FAB-0D18FEB5F175}"
+EndProject
+Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "HeuristicLab.Encodings.BinaryVectorEncoding-3.3", "HeuristicLab.Encodings.BinaryVectorEncoding\3.3\HeuristicLab.Encodings.BinaryVectorEncoding-3.3.csproj", "{66D249C3-A01D-42A8-82A2-919BC8EC3D83}"
+EndProject
+Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "HeuristicLab.Encodings.LinearLinkageEncoding-3.4", "HeuristicLab.Encodings.LinearLinkageEncoding\3.4\HeuristicLab.Encodings.LinearLinkageEncoding-3.4.csproj", "{BE698769-975A-429E-828C-72BB2B6182C8}"
 EndProject
 Global
@@ -182 +186 @@
     {73F29D43-5714-4069-8FAB-0D18FEB5F175}.Release|x86.ActiveCfg = Release|x86
     {73F29D43-5714-4069-8FAB-0D18FEB5F175}.Release|x86.Build.0 = Release|x86
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Debug|Any CPU.Build.0 = Debug|Any CPU
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Debug|x64.ActiveCfg = Debug|x64
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Debug|x64.Build.0 = Debug|x64
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Debug|x86.ActiveCfg = Debug|x86
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Debug|x86.Build.0 = Debug|x86
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Release|Any CPU.ActiveCfg = Release|Any CPU
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Release|Any CPU.Build.0 = Release|Any CPU
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Release|x64.ActiveCfg = Release|x64
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Release|x64.Build.0 = Release|x64
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Release|x86.ActiveCfg = Release|x86
+    {66D249C3-A01D-42A8-82A2-919BC8EC3D83}.Release|x86.Build.0 = Release|x86
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Debug|Any CPU.Build.0 = Debug|Any CPU
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Debug|x64.ActiveCfg = Debug|x64
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Debug|x64.Build.0 = Debug|x64
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Debug|x86.ActiveCfg = Debug|x86
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Debug|x86.Build.0 = Debug|x86
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Release|Any CPU.ActiveCfg = Release|Any CPU
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Release|Any CPU.Build.0 = Release|Any CPU
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Release|x64.ActiveCfg = Release|x64
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Release|x64.Build.0 = Release|x64
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Release|x86.ActiveCfg = Release|x86
+    {BE698769-975A-429E-828C-72BB2B6182C8}.Release|x86.Build.0 = Release|x86
   EndGlobalSection
   GlobalSection(SolutionProperties) = preSolution

branches/PerformanceComparison/ProblemInstanceIdentifier/InstanceExplorer.cs

@@ -74 +74 @@
   }
 
+  public class AdaptiveWalkExplorer : InstanceExplorer {
+    public int Iterations { get; set; }
+
+    public override string Name { get { return "Adaptive-Walk Explorer"; } }
+    public override int Effort { get { return Iterations; } }
+
+    public AdaptiveWalkExplorer() {
+
+    }
+
+    public override InstanceDescriptor Explore(QuadraticAssignmentProblem problem, int? seed = null) {
+      var walk = new AdaptiveWalk() {
+        SeedParameter = { Value = { Value = seed ?? 0 } },
+        SetSeedRandomlyParameter = { Value = { Value = seed.HasValue } },
+        MaximumIterationsParameter = { Value = { Value = Iterations } },
+        RepetitionsParameter = { Value = { Value = 1 } }
+      };
+      walk.Problem = problem;
+      walk.Engine = new SequentialEngine();
+      walk.MutatorParameter.Value = walk.MutatorParameter.ValidValues.First(x => x is Swap2Manipulator);
+      var calculator = new AdaptiveWalkCalculator(walk) { Problem = problem };
+      var features = calculator.Calculate().ToDictionary(x => x.Name, x => x.Value);
+
+      return new InstanceDescriptor(problem.Name, InstanceDescriptor.GetClass(problem.Name), problem.Weights.Rows,
+        features.Keys.ToArray(), features.Values.Select(x => ((DoubleValue)x).Value).ToArray());
+    }
+  }
+
   public class MemPRExplorer : InstanceExplorer {
     public int Seconds { get; set; }

branches/PerformanceComparison/ProblemInstanceIdentifier/Program.cs

@@ -4 +4 @@
 using System.Linq;
 using System.Threading.Tasks;
-using HeuristicLab.Algorithms.MemPR.Permutation;
 using HeuristicLab.PluginInfrastructure;
 using HeuristicLab.Problems.Instances;

@@ -5 +5 @@
 *.vs10x
 Plugin.cs
+*.DotSettings