source: trunk/sources/HeuristicLab.DataPreprocessing/3.4/Implementations/SearchLogic.cs @ 12012

Last change on this file since 12012 was 12012, checked in by ascheibe, 5 years ago

#2212 merged r12008, r12009, r12010 back into trunk

File size: 5.9 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.Linq;
26
27namespace HeuristicLab.DataPreprocessing {
28  public class SearchLogic : ISearchLogic {
29    private readonly ITransactionalPreprocessingData preprocessingData;
30
31    private Dictionary<int, IList<int>> MissingValueIndicies { get; set; }
32    private Dictionary<int, IList> ValuesWithoutNaN { get; set; }
33
34    public IEnumerable<string> VariableNames {
35      get { return preprocessingData.VariableNames; }
36    }
37
38    public int Columns {
39      get { return preprocessingData.Columns; }
40    }
41
42    public int Rows {
43      get { return preprocessingData.Rows; }
44    }
45
46    public SearchLogic(ITransactionalPreprocessingData thePreprocessingData) {
47      preprocessingData = thePreprocessingData;
48
49      MissingValueIndicies = new Dictionary<int, IList<int>>();
50      ValuesWithoutNaN = new Dictionary<int, IList>();
51
52      preprocessingData.Changed += preprocessingData_Changed;
53    }
54
55    void preprocessingData_Changed(object sender, DataPreprocessingChangedEventArgs e)
56    {
57      switch (e.Type) {
58        case DataPreprocessingChangedEventType.DeleteColumn:
59        case DataPreprocessingChangedEventType.ChangeColumn:
60          MissingValueIndicies.Remove(e.Column);
61          ValuesWithoutNaN.Remove(e.Column);
62          break;
63        case DataPreprocessingChangedEventType.AddColumn:
64          //cache does not need to be updated, will be calculated the first time it is requested
65          break;
66        case DataPreprocessingChangedEventType.DeleteRow:
67        case DataPreprocessingChangedEventType.AddRow:
68        case DataPreprocessingChangedEventType.ChangeItem:
69        case DataPreprocessingChangedEventType.Any:
70        case DataPreprocessingChangedEventType.Transformation:
71        default:
72          MissingValueIndicies = new Dictionary<int, IList<int>>();
73          ValuesWithoutNaN = new Dictionary<int, IList>();
74          break;
75      } 
76    }
77
78    public IDictionary<int, IList<int>> GetMissingValueIndices() {
79      var dic = new Dictionary<int, IList<int>>();
80      for (int i = 0; i < preprocessingData.Columns; ++i) {
81        dic.Add(i, GetMissingValueIndices(i));
82      }
83      return dic;
84    }
85
86    public bool IsMissingValue(int columnIndex, int rowIndex) {
87      if (preprocessingData.VariableHasType<double>(columnIndex)) {
88        return double.IsNaN(preprocessingData.GetCell<double>(columnIndex, rowIndex));
89      } else if (preprocessingData.VariableHasType<string>(columnIndex)) {
90        return string.IsNullOrEmpty(preprocessingData.GetCell<string>(columnIndex, rowIndex));
91      } else if (preprocessingData.VariableHasType<DateTime>(columnIndex)) {
92        return preprocessingData.GetCell<DateTime>(columnIndex, rowIndex).Equals(DateTime.MinValue);
93      } else {
94        throw new ArgumentException("cell in column " + columnIndex + " and row index " + rowIndex + " contains a non supported type.");
95      }
96    }
97
98    public IList<int> GetMissingValueIndices(int columnIndex) {
99      if (!MissingValueIndicies.ContainsKey(columnIndex)){       
100          if (preprocessingData.VariableHasType<double>(columnIndex)) {
101            MissingValueIndicies[columnIndex] = GetMissingValueIndices<double>(columnIndex);
102          } else if (preprocessingData.VariableHasType<string>(columnIndex)) {
103            MissingValueIndicies[columnIndex] = GetMissingValueIndices<string>(columnIndex);
104          } else if (preprocessingData.VariableHasType<DateTime>(columnIndex)) {
105            MissingValueIndicies[columnIndex] = GetMissingValueIndices<DateTime>(columnIndex);
106          } else {
107            throw new ArgumentException("column " + columnIndex + " contains a non supported type.");
108          }
109      } 
110
111      return MissingValueIndicies[columnIndex];
112   }
113    private IList<int> GetMissingValueIndices<T>(int columnIndex) {
114      List<int> missingIndices = new List<int>();
115     
116      for(int row = 0; row < preprocessingData.Rows; ++row) {
117        if (IsMissingValue(columnIndex, row)) {
118          missingIndices.Add(row);
119        }
120      }
121
122      return missingIndices;
123    }
124
125    public IEnumerable<T> GetValuesWithoutNaN<T>(int columnIndex, bool considerSelection)
126    {
127      if (considerSelection) {     
128        var selectedRows =  preprocessingData.Selection[columnIndex];
129       
130        List<T> values = new List<T>();
131        foreach (var rowIdx in selectedRows) {
132          if (!IsMissingValue(columnIndex, rowIdx)) {
133            values.Add(preprocessingData.GetCell<T>(columnIndex, rowIdx));
134          }
135        }
136        return values;
137      } else {
138        if (!ValuesWithoutNaN.ContainsKey(columnIndex)) {
139          List<T> values = new List<T>();
140
141          for (int row = 0; row < preprocessingData.Rows; ++row) {
142            if (!IsMissingValue(columnIndex, row)) {
143              values.Add(preprocessingData.GetCell<T>(columnIndex, row));
144            }
145          }
146
147          ValuesWithoutNaN[columnIndex] = values;
148        }
149        return (IEnumerable<T>)ValuesWithoutNaN[columnIndex];
150      }
151    }
152  }
153}
Note: See TracBrowser for help on using the repository browser.