Changeset 8508 for branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression

Timestamp:

08/20/12 17:24:14 (12 years ago)

Author:

sforsten

Message:

#1776: merged r7864:8507 trunk into branch

Location:

branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis

Files:

• . (modified) (1 prop)
• 3.4/Implementation/Regression/RegressionEnsembleSolution.cs (modified) (11 diffs)
• 3.4/Implementation/Regression/RegressionProblemData.cs (modified) (1 diff)
• 3.4/Implementation/Regression/RegressionSolution.cs (modified) (1 diff)
• 3.4/Implementation/Regression/RegressionSolutionBase.cs (modified) (5 diffs)

branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionEnsembleSolution.cs

@@ -37 +37 @@
   [Creatable("Data Analysis - Ensembles")]
   public sealed class RegressionEnsembleSolution : RegressionSolution, IRegressionEnsembleSolution {
+    private readonly Dictionary<int, double> trainingEvaluationCache = new Dictionary<int, double>();
+    private readonly Dictionary<int, double> testEvaluationCache = new Dictionary<int, double>();
+
     public new IRegressionEnsembleModel Model {
       get { return (IRegressionEnsembleModel)base.Model; }
@@ -52 +55 @@
 
     [Storable]
-    private Dictionary<IRegressionModel, IntRange> trainingPartitions;
+    private readonly Dictionary<IRegressionModel, IntRange> trainingPartitions;
     [Storable]
-    private Dictionary<IRegressionModel, IntRange> testPartitions;
+    private readonly Dictionary<IRegressionModel, IntRange> testPartitions;
 
     [StorableConstructor]
@@ -86 +89 @@
       }
 
+      trainingEvaluationCache = new Dictionary<int, double>(original.ProblemData.TrainingIndices.Count());
+      testEvaluationCache = new Dictionary<int, double>(original.ProblemData.TestIndices.Count());
+
       regressionSolutions = cloner.Clone(original.regressionSolutions);
       RegisterRegressionSolutionsEventHandler();
@@ -133 +139 @@
       }
 
+      trainingEvaluationCache = new Dictionary<int, double>(problemData.TrainingIndices.Count());
+      testEvaluationCache = new Dictionary<int, double>(problemData.TestIndices.Count());
+
       RegisterRegressionSolutionsEventHandler();
       regressionSolutions.AddRange(solutions);
@@ -153 +162 @@
     public override IEnumerable<double> EstimatedTrainingValues {
       get {
-        var rows = ProblemData.TrainingIndizes;
-        var estimatedValuesEnumerators = (from model in Model.Models
-                                          select new { Model = model, EstimatedValuesEnumerator = model.GetEstimatedValues(ProblemData.Dataset, rows).GetEnumerator() })
-                                         .ToList();
-        var rowsEnumerator = rows.GetEnumerator();
-        // aggregate to make sure that MoveNext is called for all enumerators 
-        while (rowsEnumerator.MoveNext() & estimatedValuesEnumerators.Select(en => en.EstimatedValuesEnumerator.MoveNext()).Aggregate(true, (acc, b) => acc & b)) {
-          int currentRow = rowsEnumerator.Current;
-
-          var selectedEnumerators = from pair in estimatedValuesEnumerators
-                                    where RowIsTrainingForModel(currentRow, pair.Model) && !RowIsTestForModel(currentRow, pair.Model)
-                                    select pair.EstimatedValuesEnumerator;
-          yield return AggregateEstimatedValues(selectedEnumerators.Select(x => x.Current));
+        var rows = ProblemData.TrainingIndices;
+        var rowsToEvaluate = rows.Except(trainingEvaluationCache.Keys);
+        var rowsEnumerator = rowsToEvaluate.GetEnumerator();
+        var valuesEnumerator = GetEstimatedValues(rowsToEvaluate, (r, m) => RowIsTrainingForModel(r, m) && !RowIsTestForModel(r, m)).GetEnumerator();
+
+        while (rowsEnumerator.MoveNext() & valuesEnumerator.MoveNext()) {
+          trainingEvaluationCache.Add(rowsEnumerator.Current, valuesEnumerator.Current);
         }
+
+        return rows.Select(row => trainingEvaluationCache[row]);
       }
     }
@@ -172 +177 @@
     public override IEnumerable<double> EstimatedTestValues {
       get {
-        var rows = ProblemData.TestIndizes;
-        var estimatedValuesEnumerators = (from model in Model.Models
-                                          select new { Model = model, EstimatedValuesEnumerator = model.GetEstimatedValues(ProblemData.Dataset, rows).GetEnumerator() })
-                                         .ToList();
-        var rowsEnumerator = ProblemData.TestIndizes.GetEnumerator();
-        // aggregate to make sure that MoveNext is called for all enumerators 
-        while (rowsEnumerator.MoveNext() & estimatedValuesEnumerators.Select(en => en.EstimatedValuesEnumerator.MoveNext()).Aggregate(true, (acc, b) => acc & b)) {
-          int currentRow = rowsEnumerator.Current;
-
-          var selectedEnumerators = from pair in estimatedValuesEnumerators
-                                    where RowIsTestForModel(currentRow, pair.Model)
-                                    select pair.EstimatedValuesEnumerator;
-
-          yield return AggregateEstimatedValues(selectedEnumerators.Select(x => x.Current));
+        var rows = ProblemData.TestIndices;
+        var rowsToEvaluate = rows.Except(testEvaluationCache.Keys);
+        var rowsEnumerator = rowsToEvaluate.GetEnumerator();
+        var valuesEnumerator = GetEstimatedValues(rowsToEvaluate, RowIsTestForModel).GetEnumerator();
+
+        while (rowsEnumerator.MoveNext() & valuesEnumerator.MoveNext()) {
+          testEvaluationCache.Add(rowsEnumerator.Current, valuesEnumerator.Current);
         }
+
+        return rows.Select(row => testEvaluationCache[row]);
+      }
+    }
+
+    private IEnumerable<double> GetEstimatedValues(IEnumerable<int> rows, Func<int, IRegressionModel, bool> modelSelectionPredicate) {
+      var estimatedValuesEnumerators = (from model in Model.Models
+                                        select new { Model = model, EstimatedValuesEnumerator = model.GetEstimatedValues(ProblemData.Dataset, rows).GetEnumerator() })
+                                       .ToList();
+      var rowsEnumerator = rows.GetEnumerator();
+      // aggregate to make sure that MoveNext is called for all enumerators 
+      while (rowsEnumerator.MoveNext() & estimatedValuesEnumerators.Select(en => en.EstimatedValuesEnumerator.MoveNext()).Aggregate(true, (acc, b) => acc & b)) {
+        int currentRow = rowsEnumerator.Current;
+
+        var selectedEnumerators = from pair in estimatedValuesEnumerators
+                                  where modelSelectionPredicate(currentRow, pair.Model)
+                                  select pair.EstimatedValuesEnumerator;
+
+        yield return AggregateEstimatedValues(selectedEnumerators.Select(x => x.Current));
       }
     }
@@ -201 +218 @@
 
     public override IEnumerable<double> GetEstimatedValues(IEnumerable<int> rows) {
-      return from xs in GetEstimatedValueVectors(ProblemData.Dataset, rows)
-             select AggregateEstimatedValues(xs);
+      var rowsToEvaluate = rows.Except(evaluationCache.Keys);
+      var rowsEnumerator = rowsToEvaluate.GetEnumerator();
+      var valuesEnumerator = (from xs in GetEstimatedValueVectors(ProblemData.Dataset, rowsToEvaluate)
+                              select AggregateEstimatedValues(xs))
+                             .GetEnumerator();
+
+      while (rowsEnumerator.MoveNext() & valuesEnumerator.MoveNext()) {
+        evaluationCache.Add(rowsEnumerator.Current, valuesEnumerator.Current);
+      }
+
+      return rows.Select(row => evaluationCache[row]);
     }
 
@@ -223 +249 @@
 
     protected override void OnProblemDataChanged() {
+      trainingEvaluationCache.Clear();
+      testEvaluationCache.Clear();
+      evaluationCache.Clear();
       IRegressionProblemData problemData = new RegressionProblemData(ProblemData.Dataset,
                                                                      ProblemData.AllowedInputVariables,
@@ -251 +280 @@
     public void AddRegressionSolutions(IEnumerable<IRegressionSolution> solutions) {
       regressionSolutions.AddRange(solutions);
+
+      trainingEvaluationCache.Clear();
+      testEvaluationCache.Clear();
+      evaluationCache.Clear();
     }
     public void RemoveRegressionSolutions(IEnumerable<IRegressionSolution> solutions) {
       regressionSolutions.RemoveRange(solutions);
+
+      trainingEvaluationCache.Clear();
+      testEvaluationCache.Clear();
+      evaluationCache.Clear();
     }
 
@@ -275 +312 @@
       trainingPartitions[solution.Model] = solution.ProblemData.TrainingPartition;
       testPartitions[solution.Model] = solution.ProblemData.TestPartition;
+
+      trainingEvaluationCache.Clear();
+      testEvaluationCache.Clear();
+      evaluationCache.Clear();
     }
 
@@ -282 +323 @@
       trainingPartitions.Remove(solution.Model);
       testPartitions.Remove(solution.Model);
+
+      trainingEvaluationCache.Clear();
+      testEvaluationCache.Clear();
+      evaluationCache.Clear();
     }
   }

branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionProblemData.cs

@@ -95 +95 @@
     #endregion
 
-    public ConstrainedValueParameter<StringValue> TargetVariableParameter {
-      get { return (ConstrainedValueParameter<StringValue>)Parameters[TargetVariableParameterName]; }
+    public IConstrainedValueParameter<StringValue> TargetVariableParameter {
+      get { return (IConstrainedValueParameter<StringValue>)Parameters[TargetVariableParameterName]; }
     }
     public string TargetVariable {

branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionSolution.cs

@@ -55 +55 @@
     }
     public override IEnumerable<double> EstimatedTrainingValues {
-      get { return GetEstimatedValues(ProblemData.TrainingIndizes); }
+      get { return GetEstimatedValues(ProblemData.TrainingIndices); }
     }
     public override IEnumerable<double> EstimatedTestValues {
-      get { return GetEstimatedValues(ProblemData.TestIndizes); }
+      get { return GetEstimatedValues(ProblemData.TestIndices); }
     }
 

branches/ClassificationEnsembleVoting/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionSolutionBase.cs

@@ -138 +138 @@
         OnlineCalculatorError errorState;
         Add(new Result(TrainingMeanAbsoluteErrorResultName, "Mean of absolute errors of the model on the training partition", new DoubleValue()));
-        double trainingMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(EstimatedTrainingValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndizes), out errorState);
+        double trainingMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(EstimatedTrainingValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndices), out errorState);
         TrainingMeanAbsoluteError = errorState == OnlineCalculatorError.None ? trainingMAE : double.NaN;
       }
@@ -145 +145 @@
         OnlineCalculatorError errorState;
         Add(new Result(TestMeanAbsoluteErrorResultName, "Mean of absolute errors of the model on the test partition", new DoubleValue()));
-        double testMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(EstimatedTestValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndizes), out errorState);
+        double testMAE = OnlineMeanAbsoluteErrorCalculator.Calculate(EstimatedTestValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndices), out errorState);
         TestMeanAbsoluteError = errorState == OnlineCalculatorError.None ? testMAE : double.NaN;
       }
@@ -152 +152 @@
         OnlineCalculatorError errorState;
         Add(new Result(TrainingMeanErrorResultName, "Mean of errors of the model on the training partition", new DoubleValue()));
-        double trainingME = OnlineMeanErrorCalculator.Calculate(EstimatedTrainingValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndizes), out errorState);
+        double trainingME = OnlineMeanErrorCalculator.Calculate(EstimatedTrainingValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndices), out errorState);
         TrainingMeanError = errorState == OnlineCalculatorError.None ? trainingME : double.NaN;
       }
@@ -158 +158 @@
         OnlineCalculatorError errorState;
         Add(new Result(TestMeanErrorResultName, "Mean of errors of the model on the test partition", new DoubleValue()));
-        double testME = OnlineMeanErrorCalculator.Calculate(EstimatedTestValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndizes), out errorState);
+        double testME = OnlineMeanErrorCalculator.Calculate(EstimatedTestValues, ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndices), out errorState);
         TestMeanError = errorState == OnlineCalculatorError.None ? testME : double.NaN;
       }
@@ -166 +166 @@
     protected void CalculateResults() {
       IEnumerable<double> estimatedTrainingValues = EstimatedTrainingValues; // cache values
-      IEnumerable<double> originalTrainingValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndizes);
+      IEnumerable<double> originalTrainingValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TrainingIndices);
       IEnumerable<double> estimatedTestValues = EstimatedTestValues; // cache values
-      IEnumerable<double> originalTestValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndizes);
+      IEnumerable<double> originalTestValues = ProblemData.Dataset.GetDoubleValues(ProblemData.TargetVariable, ProblemData.TestIndices);
 
       OnlineCalculatorError errorState;