Changeset 6440

Timestamp:

06/17/11 15:41:41 (13 years ago)

Author:

gkronber

Message:

#1553: implemented unit test to create and run the symbolic classification sample (mammographic masses)

Location:

trunk/sources

Files:

• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/ClassificationProblemData.cs (modified) (1 diff)
• HeuristicLab/3.3/Tests/GeneticProgrammingSamplesTest.cs (modified) (4 diffs)
• HeuristicLab/3.3/Tests/HeuristicLab-3.3.Tests.csproj (modified) (1 diff)
• HeuristicLab/3.3/Tests/Properties/Resources.Designer.cs (modified) (1 diff)
• HeuristicLab/3.3/Tests/Properties/Resources.resx (modified) (1 diff)
• HeuristicLab/3.3/Tests/mammographic_masses.txt (added)

trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Classification/ClassificationProblemData.cs

@@ -185 +185 @@
 
     #region parameter properties
-    public IValueParameter<StringValue> TargetVariableParameter {
-      get { return (IValueParameter<StringValue>)Parameters[TargetVariableParameterName]; }
+    public ConstrainedValueParameter<StringValue> TargetVariableParameter {
+      get { return (ConstrainedValueParameter<StringValue>)Parameters[TargetVariableParameterName]; }
     }
     public IFixedValueParameter<StringMatrix> ClassNamesParameter {

trunk/sources/HeuristicLab/3.3/Tests/GeneticProgrammingSamplesTest.cs

@@ -17 +17 @@
 using HeuristicLab.Problems.DataAnalysis.Symbolic;
 using System.IO;
+using HeuristicLab.Problems.DataAnalysis.Symbolic.Classification;
 
 namespace HeuristicLab_33.Tests {
@@ -72 +73 @@
       #region problem configuration
       SymbolicRegressionSingleObjectiveProblem symbRegProblem = new SymbolicRegressionSingleObjectiveProblem();
+      symbRegProblem.Name = "Tower Symbolic Regression Problem";
+      symbRegProblem.Description = "Tower Dataset (downloaded from: http://vanillamodeling.com/realproblems.html)";
       // import and configure problem data
       string filename = Path.GetTempFileName();
@@ -139 +142 @@
       symbRegProblem.ValidationPartition.End = 4000;
       symbRegProblem.RelativeNumberOfEvaluatedSamples.Value = 0.3;
+      symbRegProblem.MaximumSymbolicExpressionTreeLength.Value = 150;
+      symbRegProblem.MaximumSymbolicExpressionTreeDepth.Value = 12;
+      symbRegProblem.MaximumFunctionDefinitions.Value = 0;
+      symbRegProblem.MaximumFunctionArguments.Value = 0;
+
       symbRegProblem.EvaluatorParameter.Value = new SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator();
       #endregion
@@ -175 +183 @@
     }
 
+    [TestMethod]
+    public void CreateSymbolicClassificationSample() {
+      GeneticAlgorithm ga = new GeneticAlgorithm();
+      #region problem configuration
+      SymbolicClassificationSingleObjectiveProblem symbClassProblem = new SymbolicClassificationSingleObjectiveProblem();
+      symbClassProblem.Name = "Mammography Classification Problem";
+      symbClassProblem.Description = "Mammography dataset imported from the UCI machine learning repository (http://archive.ics.uci.edu/ml/datasets/Mammographic+Mass)";
+      // import and configure problem data
+      string filename = Path.GetTempFileName();
+      using (var writer = File.CreateText(filename)) {
+        writer.Write(HeuristicLab_33.Tests.Properties.Resources.MammographicMasses);
+      }
+      var mammoData = ClassificationProblemData.ImportFromFile(filename);
+      mammoData.TargetVariableParameter.Value = mammoData.TargetVariableParameter.ValidValues
+        .First(v => v.Value == "Severity");
+      mammoData.InputVariables.SetItemCheckedState(
+        mammoData.InputVariables.Single(x => x.Value == "BI-RADS"), false);
+      mammoData.InputVariables.SetItemCheckedState(
+        mammoData.InputVariables.Single(x => x.Value == "Age"), true);
+      mammoData.InputVariables.SetItemCheckedState(
+        mammoData.InputVariables.Single(x => x.Value == "Shape"), true);
+      mammoData.InputVariables.SetItemCheckedState(
+        mammoData.InputVariables.Single(x => x.Value == "Margin"), true);
+      mammoData.InputVariables.SetItemCheckedState(
+        mammoData.InputVariables.Single(x => x.Value == "Density"), true);
+      mammoData.InputVariables.SetItemCheckedState(
+        mammoData.InputVariables.Single(x => x.Value == "Severity"), false);
+      mammoData.TrainingPartition.Start = 0;
+      mammoData.TrainingPartition.End = 800;
+      mammoData.TestPartition.Start = 800;
+      mammoData.TestPartition.End = 961;
+      mammoData.Name = "Data imported from mammographic_masses.csv";
+      mammoData.Description = "Original dataset: http://archive.ics.uci.edu/ml/datasets/Mammographic+Mass, missing values have been replaced with median values.";
+      symbClassProblem.ProblemData = mammoData;
+
+      // configure grammar
+      var grammar = new TypeCoherentExpressionGrammar();
+      grammar.Symbols.OfType<Sine>().Single().InitialFrequency = 0.0;
+      grammar.Symbols.OfType<Cosine>().Single().InitialFrequency = 0.0;
+      grammar.Symbols.OfType<Tangent>().Single().InitialFrequency = 0.0;
+      grammar.Symbols.OfType<Power>().Single().InitialFrequency = 0.0;
+      grammar.Symbols.OfType<Root>().Single().InitialFrequency = 0.0;
+      grammar.Symbols.OfType<TimeLag>().Single().InitialFrequency = 0.0;
+      grammar.Symbols.OfType<Integral>().Single().InitialFrequency = 0.0;
+      grammar.Symbols.OfType<Derivative>().Single().InitialFrequency = 0.0;
+      grammar.Symbols.OfType<LaggedVariable>().Single().InitialFrequency = 0.0;
+      grammar.Symbols.OfType<VariableCondition>().Single().InitialFrequency = 0.0;
+      var varSymbol = grammar.Symbols.OfType<Variable>().Where(x => !(x is LaggedVariable)).Single();
+      varSymbol.WeightMu = 1.0;
+      varSymbol.WeightSigma = 1.0;
+      varSymbol.WeightManipulatorMu = 0.0;
+      varSymbol.WeightManipulatorSigma = 0.05;
+      varSymbol.MultiplicativeWeightManipulatorSigma = 0.03;
+      var constSymbol = grammar.Symbols.OfType<Constant>().Single();
+      constSymbol.MaxValue = 20;
+      constSymbol.MinValue = -20;
+      constSymbol.ManipulatorMu = 0.0;
+      constSymbol.ManipulatorSigma = 1;
+      constSymbol.MultiplicativeManipulatorSigma = 0.03;
+      symbClassProblem.SymbolicExpressionTreeGrammar = grammar;
+
+      // configure remaining problem parameters
+      symbClassProblem.BestKnownQuality.Value = 0.0;
+      symbClassProblem.FitnessCalculationPartition.Start = 0;
+      symbClassProblem.FitnessCalculationPartition.End = 400;
+      symbClassProblem.ValidationPartition.Start = 400;
+      symbClassProblem.ValidationPartition.End = 800;
+      symbClassProblem.RelativeNumberOfEvaluatedSamples.Value = 1;
+      symbClassProblem.MaximumSymbolicExpressionTreeLength.Value = 100;
+      symbClassProblem.MaximumSymbolicExpressionTreeDepth.Value = 10;
+      symbClassProblem.MaximumFunctionDefinitions.Value = 0;
+      symbClassProblem.MaximumFunctionArguments.Value = 0;
+      symbClassProblem.EvaluatorParameter.Value = new SymbolicClassificationSingleObjectiveMeanSquaredErrorEvaluator();
+      #endregion
+      #region algorithm configuration
+      ga.Problem = symbClassProblem;
+      ga.Name = "Genetic Programming - Symbolic Classification";
+      ga.Description = "A standard genetic programming algorithm to solve a classification problem (Mammographic+Mass dataset)";
+      ga.Elites.Value = 1;
+      ga.MaximumGenerations.Value = 100;
+      ga.MutationProbability.Value = 0.15;
+      ga.PopulationSize.Value = 1000;
+      ga.Seed.Value = 0;
+      ga.SetSeedRandomly.Value = false;
+      var tSelector = ga.SelectorParameter.ValidValues
+        .OfType<TournamentSelector>()
+        .Single();
+      tSelector.GroupSizeParameter.Value.Value = 5;
+      ga.Selector = tSelector;
+      var mutator = ga.MutatorParameter.ValidValues
+        .OfType<MultiSymbolicExpressionTreeManipulator>()
+        .Single();
+      mutator.Operators.OfType<FullTreeShaker>().Single().ShakingFactor = 0.1;
+      mutator.Operators.OfType<OnePointShaker>().Single().ShakingFactor = 1.0;
+      ga.Mutator = mutator;
+
+      ga.Analyzer.Operators.SetItemCheckedState(
+        ga.Analyzer.Operators
+        .OfType<SymbolicClassificationSingleObjectiveOverfittingAnalyzer>()
+        .Single(), false);
+      ga.Engine = new ParallelEngine();
+      #endregion
+
+      XmlGenerator.Serialize(ga, "../../SGP_SymbClass.hl");
+
+      RunAlgorithm(ga);
+    }
+
 
     private void RunAlgorithm(IAlgorithm a) {

trunk/sources/HeuristicLab/3.3/Tests/HeuristicLab-3.3.Tests.csproj

@@ -556 +556 @@
   </ItemGroup>
   <ItemGroup>
+    <Content Include="mammographic_masses.txt">
+      <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
+    </Content>
     <Content Include="towerData.txt">
       <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>

trunk/sources/HeuristicLab/3.3/Tests/Properties/Resources.Designer.cs

@@ -62 +62 @@
         
         /// <summary>
+        ///   Looks up a localized string similar to BI-RADS; Age; Shape; Margin; Density; Severity
+        ///5;67;3;5;3;1
+        ///4;43;1;1;3;1
+        ///5;58;4;5;3;1
+        ///4;28;1;1;3;0
+        ///5;74;1;5;3;1
+        ///4;65;1;3;3;0
+        ///4;70;3;3;3;0
+        ///5;42;1;3;3;0
+        ///5;57;1;5;3;1
+        ///5;60;3;5;1;1
+        ///5;76;1;4;3;1
+        ///3;42;2;1;3;1
+        ///4;64;1;3;3;0
+        ///4;36;3;1;2;0
+        ///4;60;2;1;2;0
+        ///4;54;1;1;3;0
+        ///3;52;3;4;3;0
+        ///4;59;2;1;3;1
+        ///4;54;1;1;3;1
+        ///4;40;1;3;3;0
+        ///4;66;3;3;1;1
+        ///5;56;4;3;1;1
+        ///4;43;1;3;3;0
+        ///5;42;4;4;3;1
+        ///4;59;2;4;3;1
+        ///5;75;4;5;3;1
+        ///2;66;1;1;3;0
+        ///5;63;3;3;3;0
+        ///5;45;4;5;3;1
+        ///5;55;4;4;3;0
+        ///4;46;1;5;2;0
+        ///5;54;4;4;3;1
+        ///5;57;4;4;3;1
+        ///4; [rest of string was truncated]&quot;;.
+        /// </summary>
+        internal static string MammographicMasses {
+            get {
+                return ResourceManager.GetString("MammographicMasses", resourceCulture);
+            }
+        }
+        
+        /// <summary>
         ///   Looks up a localized string similar to x1;x2;x3;x4;x5;x6;x7;x8;x9;x10;x11;x12;x13;x14;x15;x16;x17;x18;x19;x20;x21;x22;x23;x24;x25;towerResponse
         ///28.0784464518229;13.939023844401;87.6339436848958;20.0777689615885;63.0026733398437;70.0181884765625;59.5551432291667;8.80629475911458;94.8240641276042;1.50430590311686;473.995963541667;165.433089192708;163.806526692708;161.395735677083;140.468115234375;129.394759114583;173.246484375;29.12998046875;7.02796274820964;124.702001953125;122.926342773437;51.7155436197917;90.4672119140625;129.585375976562;8.80 [rest of string was truncated]&quot;;.

trunk/sources/HeuristicLab/3.3/Tests/Properties/Resources.resx

@@ -119 +119 @@
   </resheader>
   <assembly alias="System.Windows.Forms" name="System.Windows.Forms, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089" />
+  <data name="MammographicMasses" type="System.Resources.ResXFileRef, System.Windows.Forms">
+    <value>..\mammographic_masses.txt;System.String, mscorlib, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089;Windows-1252</value>
+  </data>
   <data name="TowerData" type="System.Resources.ResXFileRef, System.Windows.Forms">
     <value>..\towerdata.txt;System.String, mscorlib, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089;Windows-1252</value>