Changeset 7851 for trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Classification/ResourceClassificationInstanceProvider.cs

Timestamp:

05/21/12 11:13:32 (12 years ago)

Author:

sforsten

Message:

#1784: changed the TableFileParser, so that you don't have to determine the file format by yourself. Comments have been added for the different Parse methods.

Location:

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

Files:

• . (modified) (1 prop)
• Classification/ResourceClassificationInstanceProvider.cs (modified) (1 diff)

trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Classification/ResourceClassificationInstanceProvider.cs

@@ -64 +64 @@
         }
 
-        IClassificationProblemData claData;
+        TableFileParser csvFileParser = new TableFileParser();
         using (Stream stream = instancesZipFile.GetInputStream(entry)) {
-          claData = LoadData(stream, numberFormat, dateFormat, separator);
+          csvFileParser.Parse(stream, numberFormat, dateFormat, separator);
         }
+
+        Dataset dataset = new Dataset(csvFileParser.VariableNames, csvFileParser.Values);
+        string targetVar = csvFileParser.VariableNames.Last();
+        IEnumerable<string> allowedInputVars = csvFileParser.VariableNames.Where(x => !x.Equals(targetVar));
+
+        ClassificationProblemData claData = new ClassificationProblemData(dataset, allowedInputVars, targetVar);
+
+        int trainingPartEnd = csvFileParser.Rows * 2 / 3;
+        claData.TrainingPartition.Start = 0;
+        claData.TrainingPartition.End = trainingPartEnd;
+        claData.TestPartition.Start = trainingPartEnd;
+        claData.TestPartition.End = csvFileParser.Rows;
+
         claData.Name = descriptor.Name;
         claData.Description = descriptor.Description;