source: trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/CSV/RegressionCSVInstanceProvider.cs @ 8598

Last change on this file since 8598 was 8598, checked in by sforsten, 7 years ago

#1942: csv files for data analysis problems can be shuffled when imported

File size: 5.6 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2012 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
4 *
5 * This file is part of HeuristicLab.
6 *
7 * HeuristicLab is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation, either version 3 of the License, or
10 * (at your option) any later version.
11 *
12 * HeuristicLab is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
15 * GNU General Public License for more details.
16 *
17 * You should have received a copy of the GNU General Public License
18 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
19 */
20#endregion
21
22using System;
23using System.Collections;
24using System.Collections.Generic;
25using System.Globalization;
26using System.IO;
27using System.Linq;
28using System.Text;
29using HeuristicLab.Common;
30using HeuristicLab.Problems.DataAnalysis;
31
32namespace HeuristicLab.Problems.Instances.DataAnalysis {
33  public class RegressionCSVInstanceProvider : RegressionInstanceProvider {
34    public override string Name {
35      get { return "CSV File"; }
36    }
37    public override string Description {
38      get {
39        return "";
40      }
41    }
42    public override Uri WebLink {
43      get { return new Uri("http://dev.heuristiclab.com/trac/hl/core/wiki/UsersFAQ#DataAnalysisImportFileFormat"); }
44    }
45    public override string ReferencePublication {
46      get { return ""; }
47    }
48
49    public override IEnumerable<IDataDescriptor> GetDataDescriptors() {
50      return new List<IDataDescriptor>();
51    }
52    public override IRegressionProblemData LoadData(IDataDescriptor descriptor) {
53      throw new NotImplementedException();
54    }
55
56    public override bool CanImportData {
57      get { return true; }
58    }
59    public override IRegressionProblemData ImportData(string path) {
60      TableFileParser csvFileParser = new TableFileParser();
61      csvFileParser.Parse(path);
62
63      Dataset dataset = new Dataset(csvFileParser.VariableNames, csvFileParser.Values);
64      string targetVar = dataset.DoubleVariables.Last();
65
66      // turn of input variables that are constant in the training partition
67      var allowedInputVars = new List<string>();
68      var trainingIndizes = Enumerable.Range(0, (csvFileParser.Rows * 2) / 3);
69      foreach (var variableName in dataset.DoubleVariables) {
70        if (dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
71          variableName != targetVar)
72          allowedInputVars.Add(variableName);
73      }
74
75      IRegressionProblemData regressionData = new RegressionProblemData(dataset, allowedInputVars, targetVar);
76
77      var trainingPartEnd = trainingIndizes.Last();
78      regressionData.TrainingPartition.Start = trainingIndizes.First();
79      regressionData.TrainingPartition.End = trainingPartEnd;
80      regressionData.TestPartition.Start = trainingPartEnd;
81      regressionData.TestPartition.End = csvFileParser.Rows;
82
83      regressionData.Name = Path.GetFileName(path);
84
85      return regressionData;
86    }
87
88    public override IRegressionProblemData ImportData(string path, DataAnalysisImportType type) {
89      TableFileParser csvFileParser = new TableFileParser();
90      csvFileParser.Parse(path);
91
92      List<IList> values = csvFileParser.Values;
93      if (type.Shuffle) {
94        values = Shuffle(values);
95      }
96      Dataset dataset = new Dataset(csvFileParser.VariableNames, values);
97      string targetVar = dataset.DoubleVariables.Last();
98
99      // turn of input variables that are constant in the training partition
100      var allowedInputVars = new List<string>();
101      var trainingIndizes = Enumerable.Range(0, (csvFileParser.Rows * 2) / 3);
102      foreach (var variableName in dataset.DoubleVariables) {
103        if (dataset.GetDoubleValues(variableName, trainingIndizes).Range() > 0 &&
104          variableName != targetVar)
105          allowedInputVars.Add(variableName);
106      }
107
108      RegressionProblemData regressionData = new RegressionProblemData(dataset, allowedInputVars, targetVar);
109
110      int trainingPartEnd = trainingIndizes.Last();
111      regressionData.TrainingPartition.Start = trainingIndizes.First();
112      regressionData.TrainingPartition.End = trainingPartEnd;
113      regressionData.TestPartition.Start = trainingPartEnd;
114      regressionData.TestPartition.End = csvFileParser.Rows;
115
116      regressionData.Name = Path.GetFileName(path);
117
118      return regressionData;
119    }
120
121    public override bool CanExportData {
122      get { return true; }
123    }
124    public override void ExportData(IRegressionProblemData instance, string path) {
125      var strBuilder = new StringBuilder();
126
127      foreach (var variable in instance.InputVariables) {
128        strBuilder.Append(variable + CultureInfo.CurrentCulture.TextInfo.ListSeparator);
129      }
130      strBuilder.Remove(strBuilder.Length - CultureInfo.CurrentCulture.TextInfo.ListSeparator.Length, CultureInfo.CurrentCulture.TextInfo.ListSeparator.Length);
131      strBuilder.AppendLine();
132
133      var dataset = instance.Dataset;
134
135      for (int i = 0; i < dataset.Rows; i++) {
136        for (int j = 0; j < dataset.Columns; j++) {
137          if (j > 0) strBuilder.Append(CultureInfo.CurrentCulture.TextInfo.ListSeparator);
138          strBuilder.Append(dataset.GetValue(i, j));
139        }
140        strBuilder.AppendLine();
141      }
142
143      using (var writer = new StreamWriter(path)) {
144        writer.Write(strBuilder);
145      }
146    }
147  }
148}
Note: See TracBrowser for help on using the repository browser.