Free cookie consent management tool by TermsFeed Policy Generator

source: trunk/sources/HeuristicLab.Problems.DataAnalysis/3.4/Dataset.cs @ 12641

Last change on this file since 12641 was 12509, checked in by mkommend, 9 years ago

#2276: Reintegrated branch for dataset refactoring.

File size: 10.4 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2015 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.Collections.ObjectModel;
26using System.Linq;
27using HeuristicLab.Common;
28using HeuristicLab.Core;
29using HeuristicLab.Data;
30using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
31
32namespace HeuristicLab.Problems.DataAnalysis {
33  [Item("Dataset", "Represents a dataset containing data that should be analyzed.")]
34  [StorableClass]
35  public class Dataset : NamedItem, IDataset {
36    [StorableConstructor]
37    protected Dataset(bool deserializing) : base(deserializing) { }
38    protected Dataset(Dataset original, Cloner cloner)
39      : base(original, cloner) {
40      variableValues = new Dictionary<string, IList>(original.variableValues);
41      variableNames = new List<string>(original.variableNames);
42      rows = original.rows;
43    }
44    public override IDeepCloneable Clone(Cloner cloner) { return new Dataset(this, cloner); }
45
46    public Dataset()
47      : base() {
48      Name = "-";
49      VariableNames = Enumerable.Empty<string>();
50      variableValues = new Dictionary<string, IList>();
51      rows = 0;
52    }
53
54    public Dataset(IEnumerable<string> variableNames, IEnumerable<IList> variableValues)
55      : base() {
56      Name = "-";
57      if (!variableNames.Any()) {
58        this.variableNames = Enumerable.Range(0, variableValues.Count()).Select(x => "Column " + x).ToList();
59      } else if (variableNames.Count() != variableValues.Count()) {
60        throw new ArgumentException("Number of variable names doesn't match the number of columns of variableValues");
61      } else if (!variableValues.All(list => list.Count == variableValues.First().Count)) {
62        throw new ArgumentException("The number of values must be equal for every variable");
63      } else if (variableNames.Distinct().Count() != variableNames.Count()) {
64        var duplicateVariableNames =
65          variableNames.GroupBy(v => v).Where(g => g.Count() > 1).Select(g => g.Key).ToList();
66        string message = "The dataset cannot contain duplicate variables names: " + Environment.NewLine;
67        foreach (var duplicateVariableName in duplicateVariableNames)
68          message += duplicateVariableName + Environment.NewLine;
69        throw new ArgumentException(message);
70      }
71
72      rows = variableValues.First().Count;
73      this.variableNames = new List<string>(variableNames);
74      this.variableValues = new Dictionary<string, IList>(this.variableNames.Count);
75      for (int i = 0; i < this.variableNames.Count; i++) {
76        var values = variableValues.ElementAt(i);
77        IList clonedValues = null;
78        if (values is IList<double>)
79          clonedValues = new List<double>(values.Cast<double>());
80        else if (values is IList<string>)
81          clonedValues = new List<string>(values.Cast<string>());
82        else if (values is IList<DateTime>)
83          clonedValues = new List<DateTime>(values.Cast<DateTime>());
84        else {
85          this.variableNames = new List<string>();
86          this.variableValues = new Dictionary<string, IList>();
87          throw new ArgumentException("The variable values must be of type IList<double>, IList<string> or IList<DateTime>");
88        }
89        this.variableValues.Add(this.variableNames[i], clonedValues);
90      }
91    }
92
93    public Dataset(IEnumerable<string> variableNames, double[,] variableValues) {
94      Name = "-";
95      if (variableNames.Count() != variableValues.GetLength(1)) {
96        throw new ArgumentException("Number of variable names doesn't match the number of columns of variableValues");
97      }
98      if (variableNames.Distinct().Count() != variableNames.Count()) {
99        var duplicateVariableNames = variableNames.GroupBy(v => v).Where(g => g.Count() > 1).Select(g => g.Key).ToList();
100        string message = "The dataset cannot contain duplicate variables names: " + Environment.NewLine;
101        foreach (var duplicateVariableName in duplicateVariableNames)
102          message += duplicateVariableName + Environment.NewLine;
103        throw new ArgumentException(message);
104      }
105
106      rows = variableValues.GetLength(0);
107      this.variableNames = new List<string>(variableNames);
108
109      this.variableValues = new Dictionary<string, IList>(variableValues.GetLength(1));
110      for (int col = 0; col < variableValues.GetLength(1); col++) {
111        string columName = this.variableNames[col];
112        var values = new List<double>(variableValues.GetLength(0));
113        for (int row = 0; row < variableValues.GetLength(0); row++) {
114          values.Add(variableValues[row, col]);
115        }
116        this.variableValues.Add(columName, values);
117      }
118    }
119
120    protected Dataset(Dataset dataset) : this(dataset.variableNames, dataset.variableValues.Values) { }
121
122    #region Backwards compatible code, remove with 3.5
123    private double[,] storableData;
124    //name alias used to suppport backwards compatibility
125    [Storable(Name = "data", AllowOneWay = true)]
126    private double[,] StorableData { set { storableData = value; } }
127
128    [StorableHook(HookType.AfterDeserialization)]
129    private void AfterDeserialization() {
130      if (variableValues == null) {
131        rows = storableData.GetLength(0);
132        variableValues = new Dictionary<string, IList>();
133        for (int col = 0; col < storableData.GetLength(1); col++) {
134          string columName = variableNames[col];
135          var values = new List<double>(rows);
136          for (int row = 0; row < rows; row++) {
137            values.Add(storableData[row, col]);
138          }
139          variableValues.Add(columName, values);
140        }
141        storableData = null;
142      }
143    }
144    #endregion
145
146    [Storable(Name = "VariableValues")]
147    protected Dictionary<string, IList> variableValues;
148
149    protected List<string> variableNames;
150    [Storable]
151    public IEnumerable<string> VariableNames {
152      get { return variableNames; }
153      protected set {
154        if (variableNames != null) throw new InvalidOperationException();
155        variableNames = new List<string>(value);
156      }
157    }
158    public IEnumerable<string> DoubleVariables {
159      get { return variableValues.Where(p => p.Value is List<double>).Select(p => p.Key); }
160    }
161    public IEnumerable<double> GetDoubleValues(string variableName) {
162      return GetValues<double>(variableName);
163    }
164    public IEnumerable<string> GetStringValues(string variableName) {
165      return GetValues<string>(variableName);
166    }
167    public IEnumerable<DateTime> GetDateTimeValues(string variableName) {
168      return GetValues<DateTime>(variableName);
169    }
170
171    public ReadOnlyCollection<double> GetReadOnlyDoubleValues(string variableName) {
172      var values = GetValues<double>(variableName);
173      return values.AsReadOnly();
174    }
175    public double GetDoubleValue(string variableName, int row) {
176      var values = GetValues<double>(variableName);
177      return values[row];
178    }
179    public IEnumerable<double> GetDoubleValues(string variableName, IEnumerable<int> rows) {
180      return GetValues<double>(variableName, rows);
181    }
182    private IEnumerable<T> GetValues<T>(string variableName, IEnumerable<int> rows) {
183      var values = GetValues<T>(variableName);
184      return rows.Select(x => values[x]);
185    }
186    private List<T> GetValues<T>(string variableName) {
187      IList list;
188      if (!variableValues.TryGetValue(variableName, out list))
189        throw new ArgumentException("The variable " + variableName + " does not exist in the dataset.");
190      List<T> values = list as List<T>;
191      if (values == null) throw new ArgumentException("The variable " + variableName + " is not a " + typeof(T) + " variable.");
192      return values;
193    }
194    public bool VariableHasType<T>(string variableName) {
195      return variableValues[variableName] is IList<T>;
196    }
197
198    #region IStringConvertibleMatrix Members
199    [Storable]
200    protected int rows;
201    public int Rows {
202      get { return rows; }
203      set { throw new NotSupportedException(); }
204    }
205    public int Columns {
206      get { return variableNames.Count; }
207      set { throw new NotSupportedException(); }
208    }
209    public bool SortableView {
210      get { return false; }
211      set { throw new NotSupportedException(); }
212    }
213    public bool ReadOnly {
214      get { return true; }
215    }
216    IEnumerable<string> IStringConvertibleMatrix.ColumnNames {
217      get { return this.VariableNames; }
218      set { throw new NotSupportedException(); }
219    }
220    IEnumerable<string> IStringConvertibleMatrix.RowNames {
221      get { return Enumerable.Empty<string>(); }
222      set { throw new NotSupportedException(); }
223    }
224    public string GetValue(int rowIndex, int columnIndex) {
225      return variableValues[variableNames[columnIndex]][rowIndex].ToString();
226    }
227    bool IStringConvertibleMatrix.SetValue(string value, int rowIndex, int columnIndex) {
228      throw new NotSupportedException();
229    }
230    bool IStringConvertibleMatrix.Validate(string value, out string errorMessage) {
231      throw new NotSupportedException();
232    }
233
234    public virtual event EventHandler ColumnsChanged { add { } remove { } }
235    public virtual event EventHandler RowsChanged { add { } remove { } }
236    public virtual event EventHandler ColumnNamesChanged { add { } remove { } }
237    public virtual event EventHandler RowNamesChanged { add { } remove { } }
238    public virtual event EventHandler SortableViewChanged { add { } remove { } }
239    public virtual event EventHandler<EventArgs<int, int>> ItemChanged { add { } remove { } }
240    public virtual event EventHandler Reset { add { } remove { } }
241    #endregion
242  }
243}
Note: See TracBrowser for help on using the repository browser.