Changeset 14022 for stable/HeuristicLab.Problems.DataAnalysis

Timestamp:

07/08/16 12:25:13 (8 years ago)

Author:

mkommend

Message:

#2595: Merged r13766, r13942, r13985, r13986, r13987 into stable.

Location:

stable

Files:

• . (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis/3.4/HeuristicLab.Problems.DataAnalysis-3.4.csproj (modified) (2 diffs)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/DataAnalysisProblemData.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionProblemData.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionSolutionVariableImpactsCalculator.cs (copied) (copied from trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionSolutionVariableImpactsCalculator.cs) (3 diffs)
• HeuristicLab.Problems.DataAnalysis/3.4/Interfaces/IDataAnalysisProblemData.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis/3.4/Interfaces/Regression/IRegressionProblemData.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis/3.4/Plugin.cs.frame (modified) (1 diff)

stable/HeuristicLab.Problems.DataAnalysis/3.4/HeuristicLab.Problems.DataAnalysis-3.4.csproj

@@ -143 +143 @@
     </Compile>
     <Compile Include="Implementation\Regression\RegressionEnsembleSolution.cs" />
+    <Compile Include="Implementation\Regression\RegressionSolutionVariableImpactsCalculator.cs" />
     <Compile Include="Implementation\TimeSeriesPrognosis\Models\ConstantTimeSeriesPrognosisModel.cs" />
     <Compile Include="Implementation\TimeSeriesPrognosis\Models\TimeSeriesPrognosisAutoRegressiveModel.cs" />
@@ -313 +314 @@
       <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>
+    </ProjectReference>
   </ItemGroup>
   <ItemGroup />

stable/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/DataAnalysisProblemData.cs

@@ -80 +80 @@
     }
 
+    public virtual IEnumerable<int> AllIndices {
+      get { return Enumerable.Range(0, Dataset.Rows); }
+    }
     public virtual IEnumerable<int> TrainingIndices {
       get {

stable/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionProblemData.cs

@@ -110 +110 @@
     }
 
+    public IEnumerable<double> TargetVariableValues {
+      get { return Dataset.GetDoubleValues(TargetVariable); }
+    }
+    public IEnumerable<double> TargetVariableTrainingValues {
+      get { return Dataset.GetDoubleValues(TargetVariable, TrainingIndices); }
+    }
+    public IEnumerable<double> TargetVariableTestValues {
+      get { return Dataset.GetDoubleValues(TargetVariable, TestIndices); }
+    }
+
+
     [StorableConstructor]
     protected RegressionProblemData(bool deserializing) : base(deserializing) { }

stable/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionSolutionVariableImpactsCalculator.cs

@@ -30 +30 @@
 using HeuristicLab.Parameters;
 using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
+using HeuristicLab.Random;
 
 namespace HeuristicLab.Problems.DataAnalysis {
   [StorableClass]
-  [Item("RegressionSolution Impacts Calculator", "Calculation of the impacts of input variables for a concrete ")]
+  [Item("RegressionSolution Impacts Calculator", "Calculation of the impacts of input variables for any regression solution")]
   public sealed class RegressionSolutionVariableImpactsCalculator : ParameterizedNamedItem {
     public enum ReplacementMethodEnum {
       Median,
-      Average
+      Average,
+      Shuffle,
+      Noise
     }
 
@@ -77 +80 @@
       : base() {
       Parameters.Add(new FixedValueParameter<EnumValue<ReplacementMethodEnum>>(ReplacementParameterName, "The replacement method for variables during impact calculation.", new EnumValue<ReplacementMethodEnum>(ReplacementMethodEnum.Median)));
-      Parameters.Add(new FixedValueParameter<EnumValue<DataPartitionEnum>>(ReplacementParameterName, "The data partition on which the impacts are calculated.", new EnumValue<DataPartitionEnum>(DataPartitionEnum.Training)));
+      Parameters.Add(new FixedValueParameter<EnumValue<DataPartitionEnum>>(DataPartitionParameterName, "The data partition on which the impacts are calculated.", new EnumValue<DataPartitionEnum>(DataPartitionEnum.Training)));
     }
 
@@ -138 +141 @@
       var originalValues = dataset.GetReadOnlyDoubleValues(variable).ToList();
       double replacementValue;
+      List<double> replacementValues;
+      IRandom rand;
 
       switch (replacement) {
         case ReplacementMethodEnum.Median:
           replacementValue = rows.Select(r => originalValues[r]).Median();
+          replacementValues = Enumerable.Repeat(replacementValue, dataset.Rows).ToList();
           break;
         case ReplacementMethodEnum.Average:
           replacementValue = rows.Select(r => originalValues[r]).Average();
+          replacementValues = Enumerable.Repeat(replacementValue, dataset.Rows).ToList();
           break;
+        case ReplacementMethodEnum.Shuffle:
+          // new var has same empirical distribution but the relation to y is broken
+          rand = new FastRandom(31415);
+          replacementValues = rows.Select(r => originalValues[r]).Shuffle(rand).ToList();
+          break;
+        case ReplacementMethodEnum.Noise:
+          var avg = rows.Select(r => originalValues[r]).Average();
+          var stdDev = rows.Select(r => originalValues[r]).StandardDeviation();
+          rand = new FastRandom(31415);
+          replacementValues = rows.Select(_ => NormalDistributedRandom.NextDouble(rand, avg, stdDev)).ToList();
+          break;
+
         default:
           throw new ArgumentException(string.Format("ReplacementMethod {0} cannot be handled.", replacement));
       }
 
-      dataset.ReplaceVariable(variable, Enumerable.Repeat(replacementValue, dataset.Rows).ToList());
+      dataset.ReplaceVariable(variable, replacementValues);
       //mkommend: ToList is used on purpose to avoid lazy evaluation that could result in wrong estimates due to variable replacements
       var estimates = model.GetEstimatedValues(dataset, rows).ToList();

stable/HeuristicLab.Problems.DataAnalysis/3.4/Interfaces/IDataAnalysisProblemData.cs

@@ -36 +36 @@
     IntRange TestPartition { get; }
 
+    IEnumerable<int> AllIndices { get; }
     IEnumerable<int> TrainingIndices { get; }
     IEnumerable<int> TestIndices { get; }

stable/HeuristicLab.Problems.DataAnalysis/3.4/Interfaces/Regression/IRegressionProblemData.cs

@@ -20 +20 @@
 #endregion
 
+using System.Collections.Generic;
+
 namespace HeuristicLab.Problems.DataAnalysis {
   public interface IRegressionProblemData : IDataAnalysisProblemData {
     string TargetVariable { get; set; }
+
+    IEnumerable<double> TargetVariableValues { get; }
+    IEnumerable<double> TargetVariableTrainingValues { get; }
+    IEnumerable<double> TargetVariableTestValues { get; }
   }
 }

stable/HeuristicLab.Problems.DataAnalysis/3.4/Plugin.cs.frame

@@ -38 +38 @@
   [PluginDependency("HeuristicLab.Persistence", "3.3")]
   [PluginDependency("HeuristicLab.Problems.Instances", "3.3")]  
+  [PluginDependency("HeuristicLab.Random", "3.3")]
   public class HeuristicLabProblemsDataAnalysisPlugin : PluginBase {
   }