Changeset 8599 for trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis

Timestamp:

09/07/12 16:01:26 (12 years ago)

Author:

sforsten

Message:

#1942: Training and test partition can be defined (with a TrackBar in percent), when importing a csv file for data analysis problems.

Location:

trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3

Files:

• Classification/CSV/ClassifiactionCSVInstanceProvider.cs (modified) (3 diffs)
• Clustering/CSV/ClusteringCSVInstanceProvider.cs (modified) (3 diffs)
• Regression/CSV/RegressionCSVInstanceProvider.cs (modified) (3 diffs)

trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Classification/CSV/ClassifiactionCSVInstanceProvider.cs

@@ -70 +70 @@
       var trainingIndizes = Enumerable.Range(0, (csvFileParser.Rows * 2) / 3);
       foreach (var variableName in dataset.DoubleVariables) {
-        if (dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
+        if (trainingIndizes.Count() >= 2 && dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
           variableName != targetVar)
           allowedInputVars.Add(variableName);
@@ -102 +102 @@
       // turn of input variables that are constant in the training partition
       var allowedInputVars = new List<string>();
-      var trainingIndizes = Enumerable.Range(0, (csvFileParser.Rows * 2) / 3);
+      int trainingPartEnd = (csvFileParser.Rows * type.Training) / 100;
+      var trainingIndizes = Enumerable.Range(0, trainingPartEnd);
       foreach (var variableName in dataset.DoubleVariables) {
-        if (dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
+        if (trainingIndizes.Count() >= 2 && dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
           variableName != targetVar)
           allowedInputVars.Add(variableName);
@@ -111 +112 @@
       ClassificationProblemData classificationData = new ClassificationProblemData(dataset, allowedInputVars, targetVar);
 
-      int trainingPartEnd = trainingIndizes.Last();
-      classificationData.TrainingPartition.Start = trainingIndizes.First();
+      classificationData.TrainingPartition.Start = 0;
       classificationData.TrainingPartition.End = trainingPartEnd;
       classificationData.TestPartition.Start = trainingPartEnd;

trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Clustering/CSV/ClusteringCSVInstanceProvider.cs

@@ -69 +69 @@
       var trainingIndizes = Enumerable.Range(0, (csvFileParser.Rows * 2) / 3);
       foreach (var variableName in dataset.DoubleVariables) {
-        if (dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
+        if (trainingIndizes.Count() >= 2 && dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
           variableName != targetVar)
           allowedInputVars.Add(variableName);
@@ -101 +101 @@
       // turn of input variables that are constant in the training partition
       var allowedInputVars = new List<string>();
-      var trainingIndizes = Enumerable.Range(0, (csvFileParser.Rows * 2) / 3);
+      int trainingPartEnd = (csvFileParser.Rows * type.Training) / 100;
+      var trainingIndizes = Enumerable.Range(0, trainingPartEnd);
       foreach (var variableName in dataset.DoubleVariables) {
-        if (dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
+        if (trainingIndizes.Count() >= 2 && dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
           variableName != targetVar)
           allowedInputVars.Add(variableName);
@@ -110 +111 @@
       ClusteringProblemData clusteringData = new ClusteringProblemData(dataset, allowedInputVars);
 
-      int trainingPartEnd = trainingIndizes.Last();
-      clusteringData.TrainingPartition.Start = trainingIndizes.First();
+      clusteringData.TrainingPartition.Start = 0;
       clusteringData.TrainingPartition.End = trainingPartEnd;
       clusteringData.TestPartition.Start = trainingPartEnd;

trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/CSV/RegressionCSVInstanceProvider.cs

@@ -68 +68 @@
       var trainingIndizes = Enumerable.Range(0, (csvFileParser.Rows * 2) / 3);
       foreach (var variableName in dataset.DoubleVariables) {
-        if (dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
+        if (trainingIndizes.Count() >= 2 && dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
           variableName != targetVar)
           allowedInputVars.Add(variableName);
@@ -99 +99 @@
       // turn of input variables that are constant in the training partition
       var allowedInputVars = new List<string>();
-      var trainingIndizes = Enumerable.Range(0, (csvFileParser.Rows * 2) / 3);
+      int trainingPartEnd = (csvFileParser.Rows * type.Training) / 100;
+      var trainingIndizes = Enumerable.Range(0, trainingPartEnd);
       foreach (var variableName in dataset.DoubleVariables) {
-        if (dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
+        if (trainingIndizes.Count() >= 2 && dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
           variableName != targetVar)
           allowedInputVars.Add(variableName);
@@ -108 +109 @@
       RegressionProblemData regressionData = new RegressionProblemData(dataset, allowedInputVars, targetVar);
 
-      int trainingPartEnd = trainingIndizes.Last();
-      regressionData.TrainingPartition.Start = trainingIndizes.First();
+      regressionData.TrainingPartition.Start = 0;
       regressionData.TrainingPartition.End = trainingPartEnd;
       regressionData.TestPartition.Start = trainingPartEnd;