Free cookie consent management tool by TermsFeed Policy Generator

Ignore:
Timestamp:
02/11/17 01:06:37 (8 years ago)
Author:
abeham
Message:

#2457: copied MemPR algorithm from its branch to this branch

Location:
branches/PerformanceComparison
Files:
1 added
7 edited
1 copied

Legend:

Unmodified
Added
Removed
  • branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Binary/SolutionModel/Univariate/UnivariateSolutionModel.cs

    r14450 r14666  
    110110
    111111    public static ISolutionModel<BinaryVector> CreateWithFitnessBias(IRandom random, bool maximization, IEnumerable<BinaryVector> population, IEnumerable<double> qualities) {
    112       var proportions = RandomEnumerable.PrepareProportional(qualities, true, !maximization);
     112      var proportions = Util.Auxiliary.PrepareProportional(qualities, true, !maximization);
    113113      var factor = 1.0 / proportions.Sum();
    114114      double[] model = null;
  • branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/HeuristicLab.Algorithms.MemPR-3.3.csproj

    r14552 r14666  
    1818    <DebugType>full</DebugType>
    1919    <Optimize>false</Optimize>
    20     <OutputPath>..\..\bin\</OutputPath>
     20    <OutputPath>..\..\..\..\trunk\sources\bin\</OutputPath>
    2121    <DefineConstants>DEBUG;TRACE</DefineConstants>
    2222    <ErrorReport>prompt</ErrorReport>
     
    2626    <DebugType>pdbonly</DebugType>
    2727    <Optimize>true</Optimize>
    28     <OutputPath>..\..\bin\</OutputPath>
     28    <OutputPath>..\..\..\..\trunk\sources\bin\</OutputPath>
    2929    <DefineConstants>TRACE</DefineConstants>
    3030    <ErrorReport>prompt</ErrorReport>
     
    3939  <PropertyGroup Condition="'$(Configuration)|$(Platform)' == 'Debug|x64'">
    4040    <DebugSymbols>true</DebugSymbols>
    41     <OutputPath>..\..\bin\</OutputPath>
     41    <OutputPath>..\..\..\..\trunk\sources\bin\</OutputPath>
    4242    <DefineConstants>DEBUG;TRACE</DefineConstants>
    4343    <DebugType>full</DebugType>
     
    4747  </PropertyGroup>
    4848  <PropertyGroup Condition="'$(Configuration)|$(Platform)' == 'Release|x64'">
    49     <OutputPath>..\..\bin\</OutputPath>
     49    <OutputPath>..\..\..\..\trunk\sources\bin\</OutputPath>
    5050    <DefineConstants>TRACE</DefineConstants>
    5151    <Optimize>true</Optimize>
     
    5757  <PropertyGroup Condition="'$(Configuration)|$(Platform)' == 'Debug|x86'">
    5858    <DebugSymbols>true</DebugSymbols>
    59     <OutputPath>..\..\bin\</OutputPath>
     59    <OutputPath>..\..\..\..\trunk\sources\bin\</OutputPath>
    6060    <DefineConstants>DEBUG;TRACE</DefineConstants>
    6161    <DebugType>full</DebugType>
     
    6565  </PropertyGroup>
    6666  <PropertyGroup Condition="'$(Configuration)|$(Platform)' == 'Release|x86'">
    67     <OutputPath>..\..\bin\</OutputPath>
     67    <OutputPath>..\..\..\..\trunk\sources\bin\</OutputPath>
    6868    <DefineConstants>TRACE</DefineConstants>
    6969    <Optimize>true</Optimize>
     
    7474  </PropertyGroup>
    7575  <ItemGroup>
    76     <Reference Include="ALGLIB-3.7.0, Version=3.7.0.0, Culture=neutral, PublicKeyToken=ba48961d6f65dcec, processorArchitecture=MSIL">
    77       <SpecificVersion>False</SpecificVersion>
    78       <HintPath>..\..\bin\ALGLIB-3.7.0.dll</HintPath>
     76    <Reference Include="ALGLIB-3.7.0">
     77      <HintPath>..\..\..\..\trunk\sources\bin\ALGLIB-3.7.0.dll</HintPath>
     78      <Private>False</Private>
     79    </Reference>
     80    <Reference Include="HeuristicLab.Algorithms.DataAnalysis-3.4">
     81      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Algorithms.DataAnalysis-3.4.dll</HintPath>
     82      <Private>False</Private>
     83    </Reference>
     84    <Reference Include="HeuristicLab.Collections-3.3">
     85      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Collections-3.3.dll</HintPath>
     86      <Private>False</Private>
     87    </Reference>
     88    <Reference Include="HeuristicLab.Common-3.3">
     89      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Common-3.3.dll</HintPath>
     90      <Private>False</Private>
     91    </Reference>
     92    <Reference Include="HeuristicLab.Common.Resources-3.3">
     93      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Common.Resources-3.3.dll</HintPath>
     94      <Private>False</Private>
     95    </Reference>
     96    <Reference Include="HeuristicLab.Core-3.3">
     97      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Core-3.3.dll</HintPath>
     98      <Private>False</Private>
     99    </Reference>
     100    <Reference Include="HeuristicLab.Data-3.3">
     101      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Data-3.3.dll</HintPath>
     102      <Private>False</Private>
     103    </Reference>
     104    <Reference Include="HeuristicLab.Encodings.BinaryVectorEncoding-3.3">
     105      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Encodings.BinaryVectorEncoding-3.3.dll</HintPath>
     106      <Private>False</Private>
     107    </Reference>
     108    <Reference Include="HeuristicLab.Encodings.LinearLinkageEncoding-3.4">
     109      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Encodings.LinearLinkageEncoding-3.4.dll</HintPath>
     110      <Private>False</Private>
     111    </Reference>
     112    <Reference Include="HeuristicLab.Operators-3.3">
     113      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Operators-3.3.dll</HintPath>
     114      <Private>False</Private>
     115    </Reference>
     116    <Reference Include="HeuristicLab.Optimization.Operators-3.3">
     117      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Optimization.Operators-3.3.dll</HintPath>
     118      <Private>False</Private>
     119    </Reference>
     120    <Reference Include="HeuristicLab.Parameters-3.3">
     121      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Parameters-3.3.dll</HintPath>
     122      <Private>False</Private>
     123    </Reference>
     124    <Reference Include="HeuristicLab.Persistence-3.3">
     125      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Persistence-3.3.dll</HintPath>
     126      <Private>False</Private>
     127    </Reference>
     128    <Reference Include="HeuristicLab.PluginInfrastructure-3.3">
     129      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.PluginInfrastructure-3.3.dll</HintPath>
     130      <Private>False</Private>
     131    </Reference>
     132    <Reference Include="HeuristicLab.Problems.DataAnalysis-3.4">
     133      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Problems.DataAnalysis-3.4.dll</HintPath>
     134      <Private>False</Private>
     135    </Reference>
     136    <Reference Include="HeuristicLab.Random-3.3">
     137      <HintPath>..\..\..\..\trunk\sources\bin\HeuristicLab.Random-3.3.dll</HintPath>
    79138      <Private>False</Private>
    80139    </Reference>
     
    128187    <Compile Include="Plugin.cs" />
    129188    <Compile Include="Properties\AssemblyInfo.cs" />
     189    <Compile Include="Util\Auxiliary.cs" />
    130190    <Compile Include="Util\CkMeans1D.cs" />
    131191    <Compile Include="Util\FitnessComparer.cs" />
    132192  </ItemGroup>
    133193  <ItemGroup>
    134     <ProjectReference Include="..\..\HeuristicLab.Algorithms.DataAnalysis\3.4\HeuristicLab.Algorithms.DataAnalysis-3.4.csproj">
    135       <Project>{2e782078-fa81-4b70-b56f-74ce38dac6c8}</Project>
    136       <Name>HeuristicLab.Algorithms.DataAnalysis-3.4</Name>
    137       <Private>False</Private>
    138     </ProjectReference>
    139194    <ProjectReference Include="..\..\HeuristicLab.Analysis\3.3\HeuristicLab.Analysis-3.3.csproj">
    140195      <Project>{887425b4-4348-49ed-a457-b7d2c26ddbf9}</Project>
    141196      <Name>HeuristicLab.Analysis-3.3</Name>
    142       <Private>False</Private>
    143     </ProjectReference>
    144     <ProjectReference Include="..\..\HeuristicLab.Collections\3.3\HeuristicLab.Collections-3.3.csproj">
    145       <Project>{958b43bc-cc5c-4fa2-8628-2b3b01d890b6}</Project>
    146       <Name>HeuristicLab.Collections-3.3</Name>
    147       <Private>False</Private>
    148     </ProjectReference>
    149     <ProjectReference Include="..\..\HeuristicLab.Common.Resources\3.3\HeuristicLab.Common.Resources-3.3.csproj">
    150       <Project>{0e27a536-1c4a-4624-a65e-dc4f4f23e3e1}</Project>
    151       <Name>HeuristicLab.Common.Resources-3.3</Name>
    152       <Private>False</Private>
    153     </ProjectReference>
    154     <ProjectReference Include="..\..\HeuristicLab.Common\3.3\HeuristicLab.Common-3.3.csproj">
    155       <Project>{a9ad58b9-3ef9-4cc1-97e5-8d909039ff5c}</Project>
    156       <Name>HeuristicLab.Common-3.3</Name>
    157       <Private>False</Private>
    158     </ProjectReference>
    159     <ProjectReference Include="..\..\HeuristicLab.Core\3.3\HeuristicLab.Core-3.3.csproj">
    160       <Project>{c36bd924-a541-4a00-afa8-41701378ddc5}</Project>
    161       <Name>HeuristicLab.Core-3.3</Name>
    162       <Private>False</Private>
    163     </ProjectReference>
    164     <ProjectReference Include="..\..\HeuristicLab.Data\3.3\HeuristicLab.Data-3.3.csproj">
    165       <Project>{bbab9df5-5ef3-4ba8-ade9-b36e82114937}</Project>
    166       <Name>HeuristicLab.Data-3.3</Name>
    167       <Private>False</Private>
    168     </ProjectReference>
    169     <ProjectReference Include="..\..\HeuristicLab.Encodings.BinaryVectorEncoding\3.3\HeuristicLab.Encodings.BinaryVectorEncoding-3.3.csproj">
    170       <Project>{66d249c3-a01d-42a8-82a2-919bc8ec3d83}</Project>
    171       <Name>HeuristicLab.Encodings.BinaryVectorEncoding-3.3</Name>
    172       <Private>False</Private>
    173     </ProjectReference>
    174     <ProjectReference Include="..\..\HeuristicLab.Encodings.LinearLinkageEncoding\3.4\HeuristicLab.Encodings.LinearLinkageEncoding-3.4.csproj">
    175       <Project>{166507c9-ef26-4370-bb80-699742a29d4f}</Project>
    176       <Name>HeuristicLab.Encodings.LinearLinkageEncoding-3.4</Name>
    177197      <Private>False</Private>
    178198    </ProjectReference>
     
    182202      <Private>False</Private>
    183203    </ProjectReference>
    184     <ProjectReference Include="..\..\HeuristicLab.Operators\3.3\HeuristicLab.Operators-3.3.csproj">
    185       <Project>{23da7ff4-d5b8-41b6-aa96-f0561d24f3ee}</Project>
    186       <Name>HeuristicLab.Operators-3.3</Name>
    187       <Private>False</Private>
    188     </ProjectReference>
    189     <ProjectReference Include="..\..\HeuristicLab.Optimization.Operators\3.3\HeuristicLab.Optimization.Operators-3.3.csproj">
    190       <Project>{25087811-F74C-4128-BC86-8324271DA13E}</Project>
    191       <Name>HeuristicLab.Optimization.Operators-3.3</Name>
    192       <Private>False</Private>
    193     </ProjectReference>
    194204    <ProjectReference Include="..\..\HeuristicLab.Optimization\3.3\HeuristicLab.Optimization-3.3.csproj">
    195205      <Project>{14ab8d24-25bc-400c-a846-4627aa945192}</Project>
    196206      <Name>HeuristicLab.Optimization-3.3</Name>
    197       <Private>False</Private>
    198     </ProjectReference>
    199     <ProjectReference Include="..\..\HeuristicLab.Parameters\3.3\HeuristicLab.Parameters-3.3.csproj">
    200       <Project>{56f9106a-079f-4c61-92f6-86a84c2d84b7}</Project>
    201       <Name>HeuristicLab.Parameters-3.3</Name>
    202       <Private>False</Private>
    203     </ProjectReference>
    204     <ProjectReference Include="..\..\HeuristicLab.Persistence\3.3\HeuristicLab.Persistence-3.3.csproj">
    205       <Project>{102bc7d3-0ef9-439c-8f6d-96ff0fdb8e1b}</Project>
    206       <Name>HeuristicLab.Persistence-3.3</Name>
    207       <Private>False</Private>
    208     </ProjectReference>
    209     <ProjectReference Include="..\..\HeuristicLab.PluginInfrastructure\3.3\HeuristicLab.PluginInfrastructure-3.3.csproj">
    210       <Project>{94186a6a-5176-4402-ae83-886557b53cca}</Project>
    211       <Name>HeuristicLab.PluginInfrastructure-3.3</Name>
    212       <Private>False</Private>
    213     </ProjectReference>
    214     <ProjectReference Include="..\..\HeuristicLab.Problems.DataAnalysis\3.4\HeuristicLab.Problems.DataAnalysis-3.4.csproj">
    215       <Project>{df87c13e-a889-46ff-8153-66dcaa8c5674}</Project>
    216       <Name>HeuristicLab.Problems.DataAnalysis-3.4</Name>
    217     </ProjectReference>
    218     <ProjectReference Include="..\..\HeuristicLab.Random\3.3\HeuristicLab.Random-3.3.csproj">
    219       <Project>{f4539fb6-4708-40c9-be64-0a1390aea197}</Project>
    220       <Name>HeuristicLab.Random-3.3</Name>
    221207      <Private>False</Private>
    222208    </ProjectReference>
  • branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/MemPRAlgorithm.cs

    r14573 r14666  
    248248          Context.AddToPopulation(child);
    249249          Context.BestQuality = child.Fitness;
    250           Analyze(token);
     250          Analyze(CancellationToken.None);
    251251          token.ThrowIfCancellationRequested();
    252252          if (Terminate()) return;
     
    398398      } else res.Value = sp;
    399399
    400       if (Context.BreedingPerformanceModel != null) {
    401         var sol = Context.GetSolution(Context.BreedingPerformanceModel, Context.BreedingStat);
    402         if (!Results.TryGetValue("Breeding Performance", out res)) {
    403           Results.Add(new Result("Breeding Performance", sol));
    404         } else res.Value = sol;
    405       }
    406       if (Context.RelinkingPerformanceModel != null) {
    407         var sol = Context.GetSolution(Context.RelinkingPerformanceModel, Context.RelinkingStat);
    408         if (!Results.TryGetValue("Relinking Performance", out res)) {
    409           Results.Add(new Result("Relinking Performance", sol));
    410         } else res.Value = sol;
    411       }
    412       if (Context.DelinkingPerformanceModel != null) {
    413         var sol = Context.GetSolution(Context.DelinkingPerformanceModel, Context.DelinkingStat);
    414         if (!Results.TryGetValue("Delinking Performance", out res)) {
    415           Results.Add(new Result("Delinking Performance", sol));
    416         } else res.Value = sol;
    417       }
    418       if (Context.SamplingPerformanceModel != null) {
    419         var sol = Context.GetSolution(Context.SamplingPerformanceModel, Context.SamplingStat);
    420         if (!Results.TryGetValue("Sampling Performance", out res)) {
    421           Results.Add(new Result("Sampling Performance", sol));
    422         } else res.Value = sol;
    423       }
    424       if (Context.HillclimbingPerformanceModel != null) {
    425         var sol = Context.GetSolution(Context.HillclimbingPerformanceModel, Context.HillclimbingStat);
    426         if (!Results.TryGetValue("Hillclimbing Performance", out res)) {
    427           Results.Add(new Result("Hillclimbing Performance", sol));
    428         } else res.Value = sol;
    429       }
    430       if (Context.AdaptiveWalkPerformanceModel != null) {
    431         var sol = Context.GetSolution(Context.AdaptiveWalkPerformanceModel, Context.AdaptivewalkingStat);
    432         if (!Results.TryGetValue("Adaptivewalk Performance", out res)) {
    433           Results.Add(new Result("Adaptivewalk Performance", sol));
    434         } else res.Value = sol;
    435       }
    436 
    437400      Context.RunOperator(Analyzer, Context.Scope, token);
    438401    }
  • branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/MemPRContext.cs

    r14573 r14666  
    191191      get { return scope.SubScopes.Count; }
    192192    }
    193 
    194     [Storable]
    195     private IConfidenceRegressionModel breedingPerformanceModel;
    196     public IConfidenceRegressionModel BreedingPerformanceModel {
    197       get { return breedingPerformanceModel; }
    198     }
     193   
    199194    [Storable]
    200195    private List<Tuple<double, double, double, double>> breedingStat;
     
    203198    }
    204199    [Storable]
    205     private IConfidenceRegressionModel relinkingPerformanceModel;
    206     public IConfidenceRegressionModel RelinkingPerformanceModel {
    207       get { return relinkingPerformanceModel; }
    208     }
    209     [Storable]
    210200    private List<Tuple<double, double, double, double>> relinkingStat;
    211201    public IEnumerable<Tuple<double, double, double, double>> RelinkingStat {
     
    213203    }
    214204    [Storable]
    215     private IConfidenceRegressionModel delinkingPerformanceModel;
    216     public IConfidenceRegressionModel DelinkingPerformanceModel {
    217       get { return delinkingPerformanceModel; }
    218     }
    219     [Storable]
    220205    private List<Tuple<double, double, double, double>> delinkingStat;
    221206    public IEnumerable<Tuple<double, double, double, double>> DelinkingStat {
     
    223208    }
    224209    [Storable]
    225     private IConfidenceRegressionModel samplingPerformanceModel;
    226     public IConfidenceRegressionModel SamplingPerformanceModel {
    227       get { return samplingPerformanceModel; }
    228     }
    229     [Storable]
    230210    private List<Tuple<double, double>> samplingStat;
    231211    public IEnumerable<Tuple<double, double>> SamplingStat {
     
    233213    }
    234214    [Storable]
    235     private IConfidenceRegressionModel hillclimbingPerformanceModel;
    236     public IConfidenceRegressionModel HillclimbingPerformanceModel {
    237       get { return hillclimbingPerformanceModel; }
    238     }
    239     [Storable]
    240215    private List<Tuple<double, double>> hillclimbingStat;
    241216    public IEnumerable<Tuple<double, double>> HillclimbingStat {
    242217      get { return hillclimbingStat; }
    243     }
    244     [Storable]
    245     private IConfidenceRegressionModel adaptiveWalkPerformanceModel;
    246     public IConfidenceRegressionModel AdaptiveWalkPerformanceModel {
    247       get { return adaptiveWalkPerformanceModel; }
    248218    }
    249219    [Storable]
     
    276246      byAdaptivewalking = cloner.Clone(original.byAdaptivewalking);
    277247      random = cloner.Clone(original.random);
    278       breedingPerformanceModel = cloner.Clone(original.breedingPerformanceModel);
    279248      breedingStat = original.breedingStat.Select(x => Tuple.Create(x.Item1, x.Item2, x.Item3, x.Item4)).ToList();
    280       relinkingPerformanceModel = cloner.Clone(original.relinkingPerformanceModel);
    281249      relinkingStat = original.relinkingStat.Select(x => Tuple.Create(x.Item1, x.Item2, x.Item3, x.Item4)).ToList();
    282       delinkingPerformanceModel = cloner.Clone(original.delinkingPerformanceModel);
    283250      delinkingStat = original.delinkingStat.Select(x => Tuple.Create(x.Item1, x.Item2, x.Item3, x.Item4)).ToList();
    284       samplingPerformanceModel = cloner.Clone(original.samplingPerformanceModel);
    285251      samplingStat = original.samplingStat.Select(x => Tuple.Create(x.Item1, x.Item2)).ToList();
    286       hillclimbingPerformanceModel = cloner.Clone(original.hillclimbingPerformanceModel);
    287252      hillclimbingStat = original.hillclimbingStat.Select(x => Tuple.Create(x.Item1, x.Item2)).ToList();
    288       adaptiveWalkPerformanceModel = cloner.Clone(original.adaptiveWalkPerformanceModel);
    289253      adaptivewalkingStat = original.adaptivewalkingStat.Select(x => Tuple.Create(x.Item1, x.Item2)).ToList();
    290254     
     
    349313        breedingStat.Add(Tuple.Create(a.Fitness, b.Fitness, parentDist, child.Fitness));
    350314      else breedingStat.Add(Tuple.Create(b.Fitness, a.Fitness, parentDist, child.Fitness));
    351       if (breedingStat.Count % 10 == 0) RelearnBreedingPerformanceModel();
    352     }
    353     public void RelearnBreedingPerformanceModel() {
    354       breedingPerformanceModel = RunRegression(PrepareRegression(ToListRow(breedingStat)), breedingPerformanceModel).Model;
    355315    }
    356316    public bool BreedingSuited(ISingleObjectiveSolutionScope<TSolution> p1, ISingleObjectiveSolutionScope<TSolution> p2, double dist) {
    357       if (breedingPerformanceModel == null) return true;
    358       double minI1 = double.MaxValue, minI2 = double.MaxValue, maxI1 = double.MinValue, maxI2 = double.MinValue;
    359       foreach (var d in BreedingStat) {
    360         if (d.Item1 < minI1) minI1 = d.Item1;
    361         if (d.Item1 > maxI1) maxI1 = d.Item1;
    362         if (d.Item2 < minI2) minI2 = d.Item2;
    363         if (d.Item2 > maxI2) maxI2 = d.Item2;
    364       }
    365       if (p1.Fitness < minI1 || p1.Fitness > maxI1 || p2.Fitness < minI2 || p2.Fitness > maxI2)
    366         return true;
    367      
    368       return Random.NextDouble() < ProbabilityAcceptAbsolutePerformanceModel(new List<double> { p1.Fitness, p2.Fitness, dist }, breedingPerformanceModel);
     317      return true;
    369318    }
    370319    #endregion
     
    375324        relinkingStat.Add(Tuple.Create(a.Fitness, b.Fitness, parentDist, Maximization ? child.Fitness - a.Fitness : a.Fitness - child.Fitness));
    376325      else relinkingStat.Add(Tuple.Create(a.Fitness, b.Fitness, parentDist, Maximization ? child.Fitness - b.Fitness : b.Fitness - child.Fitness));
    377       if (relinkingStat.Count % 10 == 0) RelearnRelinkingPerformanceModel();
    378     }
    379     public void RelearnRelinkingPerformanceModel() {
    380       relinkingPerformanceModel = RunRegression(PrepareRegression(ToListRow(relinkingStat)), relinkingPerformanceModel).Model;
    381326    }
    382327    public bool RelinkSuited(ISingleObjectiveSolutionScope<TSolution> p1, ISingleObjectiveSolutionScope<TSolution> p2, double dist) {
    383       if (relinkingPerformanceModel == null) return true;
    384       double minI1 = double.MaxValue, minI2 = double.MaxValue, maxI1 = double.MinValue, maxI2 = double.MinValue;
    385       foreach (var d in RelinkingStat) {
    386         if (d.Item1 < minI1) minI1 = d.Item1;
    387         if (d.Item1 > maxI1) maxI1 = d.Item1;
    388         if (d.Item2 < minI2) minI2 = d.Item2;
    389         if (d.Item2 > maxI2) maxI2 = d.Item2;
    390       }
    391       if (p1.Fitness < minI1 || p1.Fitness > maxI1 || p2.Fitness < minI2 || p2.Fitness > maxI2)
    392         return true;
    393 
    394       if (IsBetter(p1, p2)) {
    395         return Random.NextDouble() < ProbabilityAcceptRelativePerformanceModel(p1.Fitness, new List<double> { p1.Fitness, p2.Fitness, dist }, relinkingPerformanceModel);
    396       }
    397       return Random.NextDouble() < ProbabilityAcceptRelativePerformanceModel(p2.Fitness, new List<double> { p1.Fitness, p2.Fitness, dist }, relinkingPerformanceModel);
     328      return true;
    398329    }
    399330    #endregion
     
    404335        delinkingStat.Add(Tuple.Create(a.Fitness, b.Fitness, parentDist, Maximization ? child.Fitness - a.Fitness : a.Fitness - child.Fitness));
    405336      else delinkingStat.Add(Tuple.Create(a.Fitness, b.Fitness, parentDist, Maximization ? child.Fitness - b.Fitness : b.Fitness - child.Fitness));
    406       if (delinkingStat.Count % 10 == 0) RelearnDelinkingPerformanceModel();
    407     }
    408     public void RelearnDelinkingPerformanceModel() {
    409       delinkingPerformanceModel = RunRegression(PrepareRegression(ToListRow(delinkingStat)), delinkingPerformanceModel).Model;
    410337    }
    411338    public bool DelinkSuited(ISingleObjectiveSolutionScope<TSolution> p1, ISingleObjectiveSolutionScope<TSolution> p2, double dist) {
    412       if (delinkingPerformanceModel == null) return true;
    413       double minI1 = double.MaxValue, minI2 = double.MaxValue, maxI1 = double.MinValue, maxI2 = double.MinValue;
    414       foreach (var d in DelinkingStat) {
    415         if (d.Item1 < minI1) minI1 = d.Item1;
    416         if (d.Item1 > maxI1) maxI1 = d.Item1;
    417         if (d.Item2 < minI2) minI2 = d.Item2;
    418         if (d.Item2 > maxI2) maxI2 = d.Item2;
    419       }
    420       if (p1.Fitness < minI1 || p1.Fitness > maxI1 || p2.Fitness < minI2 || p2.Fitness > maxI2)
    421         return true;
    422       if (IsBetter(p1, p2)) {
    423         return Random.NextDouble() < ProbabilityAcceptRelativePerformanceModel(p1.Fitness, new List<double> { p1.Fitness, p2.Fitness, dist }, delinkingPerformanceModel);
    424       }
    425       return Random.NextDouble() < ProbabilityAcceptRelativePerformanceModel(p2.Fitness, new List<double> { p1.Fitness, p2.Fitness, dist }, delinkingPerformanceModel);
     339      return true;
    426340    }
    427341    #endregion
     
    430344    public void AddSamplingResult(ISingleObjectiveSolutionScope<TSolution> sample, double avgDist) {
    431345      samplingStat.Add(Tuple.Create(avgDist, sample.Fitness));
    432       if (samplingStat.Count % 10 == 0) RelearnSamplingPerformanceModel();
    433     }
    434     public void RelearnSamplingPerformanceModel() {
    435       samplingPerformanceModel = RunRegression(PrepareRegression(ToListRow(samplingStat)), samplingPerformanceModel).Model;
    436346    }
    437347    public bool SamplingSuited(double avgDist) {
    438       if (samplingPerformanceModel == null) return true;
    439       if (avgDist < samplingStat.Min(x => x.Item1) || avgDist > samplingStat.Max(x => x.Item1)) return true;
    440       return Random.NextDouble() < ProbabilityAcceptAbsolutePerformanceModel(new List<double> { avgDist }, samplingPerformanceModel);
     348      return true;
    441349    }
    442350    #endregion
     
    445353    public void AddHillclimbingResult(ISingleObjectiveSolutionScope<TSolution> input, ISingleObjectiveSolutionScope<TSolution> outcome) {
    446354      hillclimbingStat.Add(Tuple.Create(input.Fitness, Maximization ? outcome.Fitness - input.Fitness : input.Fitness - outcome.Fitness));
    447       if (hillclimbingStat.Count % 10 == 0) RelearnHillclimbingPerformanceModel();
    448     }
    449     public void RelearnHillclimbingPerformanceModel() {
    450       hillclimbingPerformanceModel = RunRegression(PrepareRegression(ToListRow(hillclimbingStat)), hillclimbingPerformanceModel).Model;
    451355    }
    452356    public bool HillclimbingSuited(double startingFitness) {
    453       if (hillclimbingPerformanceModel == null) return true;
    454       if (startingFitness < HillclimbingStat.Min(x => x.Item1) || startingFitness > HillclimbingStat.Max(x => x.Item1))
    455         return true;
    456       return Random.NextDouble() < ProbabilityAcceptRelativePerformanceModel(startingFitness, new List<double> { startingFitness }, hillclimbingPerformanceModel);
     357      return true;
    457358    }
    458359    #endregion
     
    461362    public void AddAdaptivewalkingResult(ISingleObjectiveSolutionScope<TSolution> input, ISingleObjectiveSolutionScope<TSolution> outcome) {
    462363      adaptivewalkingStat.Add(Tuple.Create(input.Fitness, Maximization ? outcome.Fitness - input.Fitness : input.Fitness - outcome.Fitness));
    463       if (adaptivewalkingStat.Count % 10 == 0) RelearnAdaptiveWalkPerformanceModel();
    464     }
    465     public void RelearnAdaptiveWalkPerformanceModel() {
    466       adaptiveWalkPerformanceModel = RunRegression(PrepareRegression(ToListRow(adaptivewalkingStat)), adaptiveWalkPerformanceModel).Model;
    467364    }
    468365    public bool AdaptivewalkingSuited(double startingFitness) {
    469       if (adaptiveWalkPerformanceModel == null) return true;
    470       if (startingFitness < AdaptivewalkingStat.Min(x => x.Item1) || startingFitness > AdaptivewalkingStat.Max(x => x.Item1))
    471         return true;
    472       return Random.NextDouble() < ProbabilityAcceptRelativePerformanceModel(startingFitness, new List<double> { startingFitness }, adaptiveWalkPerformanceModel);
    473     }
    474     #endregion
    475 
    476     public IConfidenceRegressionSolution GetSolution(IConfidenceRegressionModel model, IEnumerable<Tuple<double, double>> data) {
    477       return new ConfidenceRegressionSolution(model, PrepareRegression(ToListRow(data.ToList())));
    478     }
    479     public IConfidenceRegressionSolution GetSolution(IConfidenceRegressionModel model, IEnumerable<Tuple<double, double, double>> data) {
    480       return new ConfidenceRegressionSolution(model, PrepareRegression(ToListRow(data.ToList())));
    481     }
    482     public IConfidenceRegressionSolution GetSolution(IConfidenceRegressionModel model, IEnumerable<Tuple<double, double, double, double>> data) {
    483       return new ConfidenceRegressionSolution(model, PrepareRegression(ToListRow(data.ToList())));
    484     }
    485 
    486     protected RegressionProblemData PrepareRegression(List<List<double>> data) {
    487       var columns = data.First().Select(y => new List<double>()).ToList();
    488       foreach (var next in data.Shuffle(Random)) {
    489         for (var i = 0; i < next.Count; i++) {
    490           columns[i].Add(next[i]);
    491         }
    492       }
    493       var ds = new Dataset(columns.Select((v, i) => i < columns.Count - 1 ? "in" + i : "out").ToList(), columns);
    494       var regPrb = new RegressionProblemData(ds, Enumerable.Range(0, columns.Count - 1).Select(x => "in" + x), "out") {
    495         TrainingPartition = { Start = 0, End = Math.Min(50, data.Count) },
    496         TestPartition = { Start = Math.Min(50, data.Count), End = data.Count }
    497       };
    498       return regPrb;
    499     }
    500 
    501     protected static IConfidenceRegressionSolution RunRegression(RegressionProblemData trainingData, IConfidenceRegressionModel baseLineModel = null) {
    502       var targetValues = trainingData.Dataset.GetDoubleValues(trainingData.TargetVariable, trainingData.TrainingIndices).ToList();
    503       var baseline = baseLineModel != null ? new ConfidenceRegressionSolution(baseLineModel, trainingData) : null;
    504       var constantSolution = new ConfidenceRegressionSolution(new ConfidenceConstantModel(targetValues.Average(), targetValues.Variance(), trainingData.TargetVariable), trainingData);
    505       var gpr = new GaussianProcessRegression { Problem = { ProblemData = trainingData } };
    506       if (trainingData.InputVariables.CheckedItems.Any(x => alglib.pearsoncorr2(trainingData.Dataset.GetDoubleValues(x.Value.Value).ToArray(), trainingData.TargetVariableValues.ToArray()) > 0.8)) {
    507         gpr.MeanFunction = new MeanZero();
    508         var cov1 = new CovarianceSum();
    509         cov1.Terms.Add(new CovarianceLinearArd());
    510         cov1.Terms.Add(new CovarianceConst());
    511         gpr.CovarianceFunction = cov1;
    512       }
    513       IConfidenceRegressionSolution solution = null;
    514       var cnt = 0;
    515       do {
    516         ExecuteAlgorithm(gpr);
    517         solution = (IConfidenceRegressionSolution)gpr.Results["Solution"].Value;
    518         cnt++;
    519       } while (cnt < 10 && (solution == null || solution.TrainingRSquared.IsAlmost(0)));
    520 
    521       return GetBestRegressionSolution(constantSolution, baseline, solution);
    522     }
    523 
    524     private static IConfidenceRegressionSolution GetBestRegressionSolution(IConfidenceRegressionSolution constant, IConfidenceRegressionSolution baseline, IConfidenceRegressionSolution solution) {
    525       if (baseline == null)
    526         return constant.TrainingMeanAbsoluteError < solution.TrainingMeanAbsoluteError ? constant : solution;
    527 
    528       double a, b, c;
    529       if (constant.ProblemData.Dataset.Rows < 60) {
    530         c = constant.TrainingMeanAbsoluteError;
    531         b = baseline.TrainingMeanAbsoluteError;
    532         a = solution.TrainingMeanAbsoluteError;
    533       } else {
    534         c = constant.TestMeanAbsoluteError;
    535         b = baseline.TestMeanAbsoluteError;
    536         a = solution.TestMeanAbsoluteError;
    537       }
    538       if (c < b && (c < a || b < a)) return constant;
    539       if (b < c && (b < a || c < a)) return baseline;
    540       return solution;
    541     }
    542 
    543     protected static void ExecuteAlgorithm(IAlgorithm algorithm) {
    544       using (var evt = new AutoResetEvent(false)) {
    545         EventHandler exeStateChanged = (o, args) => {
    546           if (algorithm.ExecutionState != ExecutionState.Started)
    547             evt.Set();
    548         };
    549         algorithm.ExecutionStateChanged += exeStateChanged;
    550         if (algorithm.ExecutionState != ExecutionState.Prepared) {
    551           algorithm.Prepare(true);
    552           evt.WaitOne();
    553         }
    554         algorithm.Start();
    555         evt.WaitOne();
    556         algorithm.ExecutionStateChanged -= exeStateChanged;
    557       }
    558     }
    559 
    560     private double ProbabilityAcceptAbsolutePerformanceModel(List<double> inputs, IConfidenceRegressionModel model) {
    561       var inputVariables = inputs.Select((v, i) => "in" + i);
    562       var ds = new Dataset(inputVariables.Concat( new [] { "out" }), inputs.Select(x => new List<double> { x }).Concat(new [] { new List<double> { double.NaN } }));
    563       var mean = model.GetEstimatedValues(ds, new[] { 0 }).Single();
    564       var sdev = Math.Sqrt(model.GetEstimatedVariances(ds, new[] { 0 }).Single());
    565 
    566       // calculate the fitness goal
    567       var goal = Maximization ? Population.Min(x => x.Fitness) : Population.Max(x => x.Fitness);
    568       var z = (goal - mean) / sdev;
    569       // return the probability of achieving or surpassing that goal
    570       var y = alglib.invnormaldistribution(z);
    571       return Maximization ? 1.0 - y /* P(X >= z) */ : y; // P(X <= z)
    572     }
    573 
    574     private double ProbabilityAcceptRelativePerformanceModel(double basePerformance, List<double> inputs, IConfidenceRegressionModel model) {
    575       var inputVariables = inputs.Select((v, i) => "in" + i);
    576       var ds = new Dataset(inputVariables.Concat(new[] { "out" }), inputs.Select(x => new List<double> { x }).Concat(new[] { new List<double> { double.NaN } }));
    577       var mean = model.GetEstimatedValues(ds, new[] { 0 }).Single();
    578       var sdev = Math.Sqrt(model.GetEstimatedVariances(ds, new[] { 0 }).Single());
    579 
    580       // calculate the improvement goal
    581       var goal = Maximization ? Population.Min(x => x.Fitness) - basePerformance : basePerformance - Population.Max(x => x.Fitness);
    582       var z = (goal - mean) / sdev;
    583       // return the probability of achieving or surpassing that goal
    584       return 1.0 - alglib.invnormaldistribution(z); /* P(X >= z) */
    585     }
    586 
    587     private static List<List<double>> ToListRow(List<Tuple<double, double>> rows) {
    588       return rows.Select(x => new List<double> { x.Item1, x.Item2 }).ToList();
    589     }
    590     private static List<List<double>> ToListRow(List<Tuple<double, double, double>> rows) {
    591       return rows.Select(x => new List<double> { x.Item1, x.Item2, x.Item3 }).ToList();
    592     }
    593     private static List<List<double>> ToListRow(List<Tuple<double, double, double, double>> rows) {
    594       return rows.Select(x => new List<double> { x.Item1, x.Item2, x.Item3, x.Item4 }).ToList();
    595     }
     366      return true;
     367    }
     368    #endregion
    596369
    597370    [MethodImpl(MethodImplOptions.AggressiveInlining)]
  • branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Permutation/SolutionModel/Univariate/UnivariateAbsoluteModel.cs

    r14496 r14666  
    121121
    122122    public static UnivariateAbsoluteModel CreateWithFitnessBias(IRandom random, bool maximization, IList<Encodings.PermutationEncoding.Permutation> population, IEnumerable<double> qualities, int N) {
    123       var proportions = RandomEnumerable.PrepareProportional(qualities, true, !maximization);
     123      var proportions = Util.Auxiliary.PrepareProportional(qualities, true, !maximization);
    124124      var factor = 1.0 / proportions.Sum();
    125125      var model = new double[N, N];
  • branches/PerformanceComparison/HeuristicLab.Algorithms.MemPR/3.3/Permutation/SolutionModel/Univariate/UnivariateRelativeModel.cs

    r14496 r14666  
    126126
    127127    public static UnivariateRelativeModel CreateDirectedWithFitnessBias(IRandom random, bool maximization, IList<Encodings.PermutationEncoding.Permutation> population, IEnumerable<double> qualities, int N) {
    128       var proportions = RandomEnumerable.PrepareProportional(qualities, true, !maximization);
     128      var proportions = Util.Auxiliary.PrepareProportional(qualities, true, !maximization);
    129129      var factor = 1.0 / proportions.Sum();
    130130      var model = new double[N, N];
     
    180180
    181181    public static UnivariateRelativeModel CreateUndirectedWithFitnessBias(IRandom random, bool maximization, IList<Encodings.PermutationEncoding.Permutation> population, IEnumerable<double> qualities, int N) {
    182       var proportions = RandomEnumerable.PrepareProportional(qualities, true, !maximization);
     182      var proportions = Util.Auxiliary.PrepareProportional(qualities, true, !maximization);
    183183      var factor = 1.0 / proportions.Sum();
    184184      var model = new double[N, N];
  • branches/PerformanceComparison/PerformanceComparison.sln

    r14059 r14666  
    11
    22Microsoft Visual Studio Solution File, Format Version 12.00
    3 # Visual Studio 2013
    4 VisualStudioVersion = 12.0.40629.0
     3# Visual Studio 14
     4VisualStudioVersion = 14.0.25420.1
    55MinimumVisualStudioVersion = 10.0.40219.1
    66Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "HeuristicLab.Analysis-3.3", "HeuristicLab.Analysis\3.3\HeuristicLab.Analysis-3.3.csproj", "{887425B4-4348-49ED-A457-B7D2C26DDBF9}"
     
    1919EndProject
    2020Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "HeuristicLab.Encodings.PermutationEncoding-3.3", "HeuristicLab.Encodings.PermutationEncoding\3.3\HeuristicLab.Encodings.PermutationEncoding-3.3.csproj", "{DBECB8B0-B166-4133-BAF1-ED67C3FD7FCA}"
     21EndProject
     22Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "HeuristicLab.Algorithms.MemPR-3.3", "HeuristicLab.Algorithms.MemPR\3.3\HeuristicLab.Algorithms.MemPR-3.3.csproj", "{9D274421-6332-4FBC-AAE4-467ACE27C368}"
    2123EndProject
    2224Global
     
    126128    {DBECB8B0-B166-4133-BAF1-ED67C3FD7FCA}.Release|x86.ActiveCfg = Release|x86
    127129    {DBECB8B0-B166-4133-BAF1-ED67C3FD7FCA}.Release|x86.Build.0 = Release|x86
     130    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
     131    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Debug|Any CPU.Build.0 = Debug|Any CPU
     132    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Debug|x64.ActiveCfg = Debug|x64
     133    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Debug|x64.Build.0 = Debug|x64
     134    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Debug|x86.ActiveCfg = Debug|x86
     135    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Debug|x86.Build.0 = Debug|x86
     136    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Release|Any CPU.ActiveCfg = Release|Any CPU
     137    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Release|Any CPU.Build.0 = Release|Any CPU
     138    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Release|x64.ActiveCfg = Release|x64
     139    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Release|x64.Build.0 = Release|x64
     140    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Release|x86.ActiveCfg = Release|x86
     141    {9D274421-6332-4FBC-AAE4-467ACE27C368}.Release|x86.Build.0 = Release|x86
    128142  EndGlobalSection
    129143  GlobalSection(SolutionProperties) = preSolution
Note: See TracChangeset for help on using the changeset viewer.