Changeset 2038 for trunk/sources

Timestamp:

06/10/09 10:46:56 (15 years ago)

Author:

gkronber

Message:

• Renamed VariableImpactCalculator to VariableQualityImpactCalculator and calculate the ratio of new quality value to old quality value to get an idea how the quality value is influenced by each variable.
• Changes in Dataset to improve the speed of SetValue (only set a dirty flag instead of reallocating the caches)

#644 (Variable impact of CEDMA models should be calculated and stored in the result DB)

Location:

trunk/sources

Files:

• HeuristicLab.DataAnalysis/3.2/Dataset.cs (modified) (10 diffs)
• HeuristicLab.GP.StructureIdentification/3.3/Evaluators/GPEvaluatorBase.cs (modified) (2 diffs)
• HeuristicLab.Modeling/3.2/HeuristicLab.Modeling-3.2.csproj (modified) (2 diffs)
• HeuristicLab.Modeling/3.2/VariableQualityImpactCalculator.cs (moved) (moved from trunk/sources/HeuristicLab.Modeling/3.2/VariableImpactCalculator.cs) (5 diffs)

trunk/sources/HeuristicLab.DataAnalysis/3.2/Dataset.cs

@@ -39 +39 @@
     private double[] scalingFactor;
     private double[] scalingOffset;
+    private bool cachedValuesInvalidated = true;
+
+    private bool fireChangeEvents = true;
+    public bool FireChangeEvents {
+      get { return fireChangeEvents; }
+      set { fireChangeEvents = value; }
+    }
 
     public string Name {
@@ -74 +81 @@
       if (v != samples[columns * i + j]) {
         samples[columns * i + j] = v;
-        CreateDictionaries();
-        FireChanged();
+        cachedValuesInvalidated = true;
+        if (fireChangeEvents) FireChanged();
       }
     }
@@ -89 +96 @@
         }
         samples = value;
-        CreateDictionaries();
-        FireChanged();
+        cachedValuesInvalidated = true;
+        if (fireChangeEvents) FireChanged();
       }
     }
@@ -104 +111 @@
       scalingOffset = new double[] { 0.0 };
       scalingFactor = new double[] { 1.0 };
-    }
-
-    private void CreateDictionaries() {
-      // keep a means and ranges dictionary for each column (possible target variable) of the dataset.
-      cachedMeans = new Dictionary<int, Dictionary<int, double>>[columns];
-      cachedRanges = new Dictionary<int, Dictionary<int, double>>[columns];
-      for (int i = 0; i < columns; i++) {
-        cachedMeans[i] = new Dictionary<int, Dictionary<int, double>>();
-        cachedRanges[i] = new Dictionary<int, Dictionary<int, double>>();
-      }
+      cachedValuesInvalidated = true;
+      fireChangeEvents = true;
     }
 
@@ -206 +205 @@
         }
       }
-      CreateDictionaries();
     }
 
@@ -270 +268 @@
 
     public double GetMean(int column, int from, int to) {
+      if (cachedValuesInvalidated) CreateDictionaries();
       if (!cachedMeans[column].ContainsKey(from) || !cachedMeans[column][from].ContainsKey(to)) {
         double[] values = new double[to - from];
@@ -289 +288 @@
 
     public double GetRange(int column, int from, int to) {
+      if (cachedValuesInvalidated) CreateDictionaries();
       if (!cachedRanges[column].ContainsKey(from) || !cachedRanges[column][from].ContainsKey(to)) {
         double[] values = new double[to - from];
@@ -329 +329 @@
         else ScaleVariable(column, 1.0 / range, -min);
       }
-      CreateDictionaries();
-      FireChanged();
+      cachedValuesInvalidated = true;
+      if (fireChangeEvents) FireChanged();
     }
 
@@ -340 +340 @@
         samples[i * columns + column] = (origValue + offset) * factor;
       }
-      CreateDictionaries();
-      FireChanged();
+      cachedValuesInvalidated = true;
+      if (fireChangeEvents) FireChanged();
     }
 
@@ -353 +353 @@
         scalingOffset[column] = 0.0;
       }
+      cachedValuesInvalidated = true;
+      if (fireChangeEvents) FireChanged();
+    }
+
+    private void CreateDictionaries() {
+      // keep a means and ranges dictionary for each column (possible target variable) of the dataset.
+      cachedMeans = new Dictionary<int, Dictionary<int, double>>[columns];
+      cachedRanges = new Dictionary<int, Dictionary<int, double>>[columns];
+      for (int i = 0; i < columns; i++) {
+        cachedMeans[i] = new Dictionary<int, Dictionary<int, double>>();
+        cachedRanges[i] = new Dictionary<int, Dictionary<int, double>>();
+      }
     }
   }

trunk/sources/HeuristicLab.GP.StructureIdentification/3.3/Evaluators/GPEvaluatorBase.cs

@@ -72 +72 @@
         }
       }
+      dataset.FireChangeEvents = false;
 
       Evaluate(scope, evaluator, dataset, targetVariable, start, end, useEstimatedValues);
@@ -81 +82 @@
         }
       }
+      dataset.FireChangeEvents = true;
+      dataset.FireChanged();
 
       // update the value of total evaluated nodes

trunk/sources/HeuristicLab.Modeling/3.2/HeuristicLab.Modeling-3.2.csproj

@@ -83 +83 @@
   <ItemGroup>
     <Compile Include="ClassificationProblemInjector.cs" />
-    <Compile Include="VariableImpactCalculator.cs" />
     <Compile Include="Model.cs" />
     <Compile Include="IModel.cs" />
@@ -106 +105 @@
     <Compile Include="TimeSeriesProblemInjector.cs" />
     <Compile Include="SimpleVarianceAccountedForEvaluator.cs" />
+    <Compile Include="VariableQualityImpactCalculator.cs" />
   </ItemGroup>
   <ItemGroup>

trunk/sources/HeuristicLab.Modeling/3.2/VariableQualityImpactCalculator.cs

@@ -30 +30 @@
 
 namespace HeuristicLab.Modeling {
-  public class VariableImpactCalculator : OperatorBase {
+  public class VariableQualityImpactCalculator : OperatorBase {
     public override string Description {
       get { return @"Calculates the impact of all allowed input variables on the quality of the model using evaluator supplied as suboperator."; }
     }
 
-    public VariableImpactCalculator()
+    public VariableQualityImpactCalculator()
       : base() {
       AddVariableInfo(new VariableInfo("Dataset", "Dataset", typeof(Dataset), VariableKind.In));
@@ -42 +42 @@
       AddVariableInfo(new VariableInfo("TrainingSamplesStart", "TrainingSamplesStart", typeof(IntData), VariableKind.In));
       AddVariableInfo(new VariableInfo("TrainingSamplesEnd", "TrainingSamplesEnd", typeof(IntData), VariableKind.In));
-      AddVariableInfo(new VariableInfo("VariableImpacts", "Variable impacts", typeof(ItemList), VariableKind.New));
+      AddVariableInfo(new VariableInfo("VariableQualityImpacts", "Effect on quality of model (percentage of original quality) if variable is replaced by its mean.", typeof(ItemList), VariableKind.New));
     }
 
@@ -53 +53 @@
       int end = GetVariableValue<IntData>("TrainingSamplesEnd", scope, true).Data;
 
-      if (SubOperators.Count < 1) throw new InvalidOperationException("VariableImpactCalculator needs a suboperator to evaluate the model");
+      if (SubOperators.Count < 1) throw new InvalidOperationException("VariableQualityImpactCalculator needs a suboperator to evaluate the model");
       IOperator evaluationOperator = this.SubOperators[0];
-
-      ItemList variableImpacts = new ItemList();
+      ItemList variableQualityImpacts = new ItemList();
 
       // calculateReferenceQuality
@@ -65 +64 @@
         var oldValues = ReplaceVariableValues(dirtyDataset, currentVariable , CalculateNewValues(dirtyDataset, currentVariable, start, end), start, end);
         double newQuality = CalculateQuality(scope, dirtyDataset, evaluationOperator);
-        double ratio = referenceQuality / newQuality;
-        double impact = ratio < 1.0 ? 1.0 - ratio : 1.0 - 1.0 / ratio;
+        double ratio = newQuality / referenceQuality;
         ItemList row = new ItemList();
         row.Add(new StringData(dataset.GetVariableName(currentVariable)));
-        row.Add(new DoubleData(impact));
-        variableImpacts.Add(row);
+        row.Add(new DoubleData(ratio));
+        variableQualityImpacts.Add(row);
         ReplaceVariableValues(dirtyDataset, currentVariable, oldValues, start, end);
       }
-      scope.AddVariable(new Variable(scope.TranslateName("VariableImpacts"), variableImpacts));
+      scope.AddVariable(new Variable(scope.TranslateName("VariableQualityImpacts"), variableQualityImpacts));
       return null;
     }
@@ -93 +91 @@
 
       int index = start;
+      ds.FireChangeEvents = false;
       foreach(double v in newValues) {
         ds.SetValue(index++, variableIndex, v);
       }
+      ds.FireChangeEvents = true;
+      ds.FireChanged();
       return oldValues;
     }