Changeset 8331 for branches/ScatterSearch (trunk integration)/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Clustering/CSV/ClusteringCSVInstanceProvider.cs

Timestamp:

07/26/12 09:51:13 (12 years ago)

Author:

jkarder

Message:

#1331: merged r8086:8330 from trunk

Location:

branches/ScatterSearch (trunk integration)

Files:

• . (modified) (2 props)
• HeuristicLab.Problems.Instances.DataAnalysis/3.3/Clustering/CSV/ClusteringCSVInstanceProvider.cs (modified) (3 diffs)

branches/ScatterSearch (trunk integration)

@@ -21 +21 @@
 protoc.exe
 _ReSharper.HeuristicLab 3.3 Tests
+Google.ProtocolBuffers-2.4.1.473.dll

@@ -21 +21 @@
 protoc.exe
 _ReSharper.HeuristicLab 3.3 Tests
+Google.ProtocolBuffers-2.4.1.473.dll

branches/ScatterSearch (trunk integration)/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Clustering/CSV/ClusteringCSVInstanceProvider.cs

@@ -22 +22 @@
 using System;
 using System.Collections.Generic;
+using System.IO;
+using System.Text;
 using HeuristicLab.Problems.DataAnalysis;
 
@@ -27 +29 @@
   public class ClusteringCSVInstanceProvider : ClusteringInstanceProvider {
     public override string Name {
-      get { return "CSV Problem Provider"; }
+      get { return "CSV File"; }
     }
     public override string Description {
@@ -48 +50 @@
       throw new NotImplementedException();
     }
+
+    public override bool CanImportData {
+      get { return true; }
+    }
+    public override IClusteringProblemData ImportData(string path) {
+      var csvFileParser = new TableFileParser();
+
+      csvFileParser.Parse(path);
+
+      var dataset = new Dataset(csvFileParser.VariableNames, csvFileParser.Values);
+      var claData = new ClusteringProblemData(dataset, dataset.DoubleVariables);
+
+      int trainingPartEnd = csvFileParser.Rows * 2 / 3;
+      claData.TrainingPartition.Start = 0;
+      claData.TrainingPartition.End = trainingPartEnd;
+      claData.TestPartition.Start = trainingPartEnd;
+      claData.TestPartition.End = csvFileParser.Rows;
+      int pos = path.LastIndexOf('\\');
+      if (pos < 0)
+        claData.Name = path;
+      else {
+        pos++;
+        claData.Name = path.Substring(pos, path.Length - pos);
+      }
+
+      return claData;
+    }
+
+    public override bool CanExportData {
+      get { return true; }
+    }
+    public override void ExportData(IClusteringProblemData instance, string path) {
+      var strBuilder = new StringBuilder();
+
+      foreach (var variable in instance.InputVariables) {
+        strBuilder.Append(variable + ";");
+      }
+      strBuilder.Remove(strBuilder.Length - 1, 1);
+      strBuilder.AppendLine();
+
+      var dataset = instance.Dataset;
+
+      for (int i = 0; i < dataset.Rows; i++) {
+        for (int j = 0; j < dataset.Columns; j++) {
+          strBuilder.Append(dataset.GetValue(i, j) + ";");
+        }
+        strBuilder.Remove(strBuilder.Length - 1, 1);
+        strBuilder.AppendLine();
+      }
+
+      using (var writer = new StreamWriter(path)) {
+        writer.Write(strBuilder);
+      }
+    }
   }
 }

@@ -21 +21 @@
 protoc.exe
 _ReSharper.HeuristicLab 3.3 Tests
+Google.ProtocolBuffers-2.4.1.473.dll